Integrated charge and sealassessment in the Monagas foldand thrust belt of VenezuelaMartin Neumaier Ralf Littke Thomas HantschelLaurent Maerten Jean-Pierre Joonnekindtand Peter Kukla

ABSTRACT

Conventional basin and petroleum systems modeling uses the ver-tical backstripping approach to describe the structural evolution ofa basin In structurally complex regions this is not sufficient Iflateral rock movement and faulting are inputs the basin and petro-leum systems modeling should be performed using structurallyrestored models This requires a specific methodology to simulaterock stress pore pressure and compaction followed by the mod-eling of the thermal history and the petroleum systems We dem-onstrate the strength of this approach in a case study from theMonagas fold and thrust belt (Eastern Venezuela Basin) The dif-ferent petroleum systems have been evaluated through geologictime within a pressure and temperature framework Particularemphasis has been given to investigating structural dependenciesof the petroleum systems such as the relationship between thrust-ing and hydrocarbon generation dynamic structure-related migra-tion pathways and the general impact of deformation We alsofocus on seal integrity through geologic time by using two inde-pendent methods forward rock stress simulation and fault activityanalysis We describe the uncertainty that is introduced by replac-ing backstripped paleogeometry with structural restoration anddiscuss decompaction adequacy We have built two end-memberscenarios using structural restoration one assuming hydrostaticdecompaction and one neglecting it We have quantified theimpact through geologic time of both scenarios by analyzingimportant parameters such as rock matrix mass balance sourcerock burial depth temperature and transformation ratio

Copyright copy2014 The American Association of Petroleum Geologists All rights reserved

Manuscript received September 12 2012 provisional acceptance February 02 2013 revised manuscriptreceived October 01 2013 final acceptance January 13 2013DOI 10130601131412157

AUTHORS

Martin Neumaier sim Institute of Geologyand Geochemistry of Petroleum and CoalEnergy and Mineral Resources (EMR)RWTH Aachen University Lochnerstr 4-2052056 Aachen Germany SchlumbergerRitterstr 23 52072 Aachen Germanymneumaierslbcom

Martin Neumaier received his masterrsquosdegree in geological reservoirs at theUniversity of Montpellier (France) in 2008He joined Schlumberger in 2008 as ageologist and is currently preparing a PhDat RWTH Aachen University Martinrsquosbackground and research interests arecentered on geodynamics and basinevolution regional and structural geologypetroleum systems modeling and analysishydrocarbon exploration and play toprospect resource evaluation

Ralf Littke sim Institute of Geology andGeochemistry of Petroleum and CoalEnergy and Mineral Resources (EMR)RWTH Aachen University Lochnerstr 4-2052056 Aachen Germany ralflittkeemrrwth-aachende

Ralf Littke received his diploma in geologyand PhD at Ruhr-University Bochum(Germany) He is professor of geology andgeochemistry of petroleum and coal andcurrently dean of the faculty ofGeoresources and Material Science at RWTHAachen University His research interestsinclude sedimentary basin dynamicspetroleum and source rock geochemistryporosity and permeability of low-permeability rocks natural gasgeochemistry organic petrology petroleumsystem modeling and petroleum geology

Thomas Hantschel sim SchlumbergerRitterstr 23 52072 Aachen Germanythantschelslbcom

Thomas Hantschel received his PhD innumerical modeling in 1987 and worked forIES (Integrated Exploration Systems) as themain developer of the numerical simulatorof the PetroMod (Mark of Schlumberger)Basin and Petroleum Systems Modelingsoftware In 2002 he became the managingdirector of IES which was acquired by

AAPG Bulletin v 98 no 7 (July 2014) pp 1325ndash1350 1325

INTRODUCTION

Basin and petroleum systems modeling (BPSM) is essential topetroleum exploration because it provides quantitative estimatesof the hydrocarbon charge history (hydrocarbon generationmigration accumulation and respective timing) and enableshydrocarbon volume and property predictions (Hantschel andKauerauf 2009) Basin and petroleum systems modeling is alsocommonly used for pore pressure prediction as an alternative toseismic velocityndashderived pore pressure prediction especiallywhere the seismic data quality is poor for example in subsaltand subthrust settings Standard BPSM is however limited forcomplex basin geometries especially in areas with thrusting andsalt movement In these areas structural restoration whichaccounts for lateral rock movement caused by faulting and fold-ing should be applied to describe the geometrical evolution of abasin Recently structural restoration methods incorporate geo-mechanics during the restoration process while keeping a recordof the evolving strain and stress distributions (Maerten andMaerten 2006) The structurally restored paleogeometries (struc-tural geometry of the basin at a given geologic time) can then beused in BPSM replacing the conventional backstripping methodThe integration of structural restoration into BPSM provides a bet-ter understanding of the petroleum systemrsquos evolution throughgeologic time in complex tectonic settings This workflow hasbeen successfully applied to several case studies with both two-dimensional (2D) and three-dimensional (3D) models (eg Bauret al 2009)

The traditional stressndashstrain model has also undergone recentimprovements To more accurately predict a compaction-relateddecrease in porosity lateral compressional boundary conditionscan be applied resulting in higher horizontal stresses and porepressures (Hantschel et al 2012) This advanced method replacesthe simple approach using vertical lithostatic pressure (Terzaghi1923) which is simply equal to overburden load The evolutionof geomechanical properties calculated by a combined approachof structural restoration and BPSM analysis can assist in predict-ing fracturing andor seal integrity through geologic time

In this paper we describe the theoretical background of theconcept as well as its applicability and limitations Dynel softwarefor structural restoration and PetroMod (Mark of Schlumberger)software for petroleum system modeling were applied in a com-bined 2D basin and petroleum systems analysis workflow of thewestern Monagas fold and thrust belt (Venezuela) This workflowenabled us to describe the compaction pressure and temperaturehistory and evolution of the different petroleum systems throughgeologic time Structural controls on the petroleum systems such

Schlumberger in 2008 Thomasrsquo researchinterests are centered on heat flow porepressure and fluid flow rock stressgeochemical models unconventionalpetroleum systems domains in which hehas contributed many publications He iscurrently center manager of the Aachen-based Technology Center for PetroleumSystems Modeling and Exploration Geology

Laurent Maerten sim Schlumberger 340 rueLouis Pasteur Grabels France lmaertenslbcom

Laurent Maerten received his PhD ingeology and environmental sciences in 1999from Stanford University (USA) incollaboration with Norsk Hydro He workedat the French Petroleum Institute in Paris(France) as a research engineer in structuralgeology for two years In 2004 Laurentfounded the structural geology companyIGEOSS of which he was the CEO andprincipal consultant He joinedSchlumberger in 2010 with its acquisition ofIGEOSS His main interests are structuralgeology and geomechanics with a strongfocus on rock and fracture mechanicsstructural restoration and applications tounconventional reservoirs

Jean-Pierre Joonnekindt sim Schlumberger340 rue Louis Pasteur Grabels Francejjoonnekindtslbcom

Jean-Pierre Joonnekindt received hismasterrsquos degree in structural geology in1994 from the Earth Sciences MontpellierUniversity (France) in collaboration withJean-Pierre Petit and a masterrsquos degree incomputer sciences in 2000 from the JosephFourier Grenoble University (France) Hejoined IGEOSS in 2008 as a structuralgeology consultant and Schlumberger in2010 with its acquisition of IGEOSS Hisinterests focus on geological structures 2Dand 3D restoration geomechanics fracturemechanics and the development ofinnovative methods to improve hydrocarbonproduction in fractured reservoirs

Peter Kukla sim Geological Institute Energyand Mineral Resources (EMR) RWTHAachen University Wuellnerstr 2 52062Aachen Germany kuklageolrwth-aachende

1326 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

as the effects of thrusting on pressure and hydrocarbon generationdynamic structure-related migration pathways and the generalimpact of deformation are described Additionally we evaluatedseal integrity by analyzing simulated stresses through geo-logic time

GEOLOGIC SETTING

The Eastern Venezuela Basin (Figure 1A) is bounded by theSerrania del Interior and the El Pilar fault to the north by a dextralstrike-slip fault system at the interface of the Caribbean and SouthAmerican tectonic plates and by the Orinoco River and thePrecambrian Guyana shield to the south (Hedberg 1950Passalacqua et al 1995 Pindell and Tabbutt 1995) The northernparts of the basin the Monagas fold and thrust belt override a lessdeformed foreland part of the basin to the south

The basin initially formed as a passive margin on the SouthAmerican shield following late Jurassic to early Cretaceous rift-ing During the transcollision of the Caribbean plate with theSouth American plate in the late Cenozoic subsidence becameflexural and extensive folding and thrusting occurred in thenorthern part of the basin (Figure 1B) Roure et al (2003)described both the geodynamic history and the stratigraphy(Figure 1C) Cenozoic platformal and basinal facies of the passivemargin overlie the Mesozoic sediments which include the preriftand synrift sequences The most recent sediments belong to thesynorogenic north-sourced depositional sequence The principalsource rocks in the area are within the Querecual and SanAntonio Formations of the Guayuta Group (Talukdar et al1988 Gallango et al 1992 Alberdi and Lafargue 1993) Thesemudstones are characterized by type II kerogen with 2 to 6 wttotal organic carbon (TOC) for partly highly mature samplesHydrogen index (HI) values are as high as 500 mg HCg TOCSumma et al (2003) suggested an initial TOC as high as 12 forpresent-day values of up to 8 with an HI as high as 700 mgHCg TOC Minor source rocks exist in the Carapita Formationparticularly in the basal part with mixed type IIIII kerogenSumma et al (2003) estimated TOC up to 45 and an HI of350 mg HCg TOC They also discussed the presence of Jurassicand Albian source rocks in the Serrania del Interior

Surface oil seeps and asphalt lakes are very common in thisarea and hydrocarbons have been found since the early 20th cen-tury (Young 1978 Carnevali 1988 Krause and James 1989Aymard et al 1990 James 1990 2000a 2000b Erlich andBarrett 1992 Prieto and Valdes 1992) In the south heavy oiland tar sands (Oronico Oil Belt see Figure 1A) and conventionaloil accumulations are abundant In the thrust area the Furrial trend

Peter Kukla is a professor of geology anddirector of the Geological Institute at RWTHAachen University since 2000 He workedfor Shell International in the Netherlands(1991ndash1996) and in Australia (seconded toWoodside Energy in Perth 1996ndash2000)Peter graduated with degrees in geologyfrom Wuerzburg University Germany(MSc 1986) and Witwatersrand UniversitySouth Africa (PhD 1991) His researchinterests include basin analysis clastic andcarbonate reservoir characterizationunconventional gas geodynamics of saltbasins and pore pressures

ACKNOWLEDGMENTS

We thank Francois Roure (Institut Francaisdu Petrole) Sandro Serra (SerraGeoConsulting) and an anonymousreviewer for their comments andsuggestions Ian Bryant Martin RohdeAdrian Kleine Alan Monahan Will StapleyArmin Kauerauf and Thomas Fuchs(Schlumberger) also provided very helpfultechnical contribution discussions andinternal reviews

The AAPG Editor thanks the followingreviewers for their work on this paperFrancois M Roure Sandro Serra Carl KSteffensen and an anonymous reviewer fortheir work on this paper

NEUMAIER ET AL 1327

contains some of the worldrsquos largest oil reserves forbasins with this type of tectonics Many shallow oilfields in the Eastern Venezuela Basin are located inNeogene sandstone reservoir rocks (Las Piedras LaPica Morichito Chapapotal Freites and Oficinafields) Other reservoir formations are the synoro-genic Miocene Naricual Formation and theOligocene Merecure Formation which are part of aproximal passive-margin sequence that shales outtoward the north The latter is the main reservoir inthe thrusted Furrial trend In the deeper passive-margin sequence there are potential reservoirs in theMesozoic (eg San Juan Formation) Because of its

shaly lithology large thickness and lateral continu-ity the synorogenic Carapita Formation is probablythe most effective seal However several other for-mations have good sealing capacities even thoughnot well compacted as numerous fields at shallowdepths prove

The first publications on the thermal history ofthe Eastern Venezuela Basin were based on one-dimensional (1D) models Because of the com-plexities leading to multiple sequences caused byoverthrusting modeling was done either in theunthrusted area or if within the thrusted area onlyfor geologic time intervals preceding thrusting

Figure 1 (A) Map of the Eastern Venezuela Basin showing structural elements and hydrocarbon occurrence (modified after Parnaudet al 1995) Solid line shows location of two-dimensional (2D) line modeled in this study (B) Generalized northndashsouth geodynamiccross section of the study area (modified after Chevalier 1987) (C) Synthetic stratigraphy of the southern foreland basin of theEastern Venezuela Basin (modified after Roure et al 2003)

1328 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

(Talukdar et al 1988) Summa et al (2003) pub-lished multiple 1D models over a large area andextrapolated the results onto regional maturity mapsfor the Guayuta Group and Cenozoic source rocksThese maps include a hanging-wall cutoff and there-fore do not represent the higher maturities found inthe thrust wedges Elsewhere the authors suggestedthat most source rocks have been in the oil windowin the past A present-day drainage area analysis ofthe top Cretaceous structure map was used to describethe southward charge from the deeper kitchen into theforeland basin This regional migration modelexplains oil provenance and quality Important timingaspects such as the relationship between the struc-tural history and source rock maturation and expul-sion are described using a regional structuralrestoration in conjunction with the 1D models Parraet al (2011) included the thrusting effect by using a15D module mimicking lateral rock transportationthat is due to thrusting Gallango and Parnaud(1995) performed 2D modeling including hydrocar-bon migration in the unthrusted southern area Theyalso described a second model within the thrustedarea that simulates the basin history prior to compres-sion (from the Mesozoic to early Miocene) Thismodel proposes long-distance petroleum migrationsouthward before the transpressional event whichresulted in the observed structural complexity Inother papers Roure et al (2003) Schneider (2003)and Schneider et al (2004) described fluid flow porefluid history and diagenetic processes in a 2D modelalong a cross section parallel to the location of themodel presented here

MODELING METHODS

To address the charge and seal history of the SantaBarbara transect (Figure 1) in the western Monagasfold and thrust belt we applied a combined approachusing BPSM and structural restoration A BPSM is adynamic model of the subbasinrsquos physical processessuch as pore pressure and temperature evolution andthe development of the petroleum systems includinghydrocarbon generation expulsion migration accu-mulation and preservation (Hantschel and Kauerauf2009) The BPSM is performed in 1D (on a well orpseudowell) in 2D (usually along a seismic section)

and in 3D The BPSM can also provide predictionsabout seal capacity and integrity especially if fielddata are available for calibration

In contrast to a static model which consistsonly of geometry and properties for present day(as observed) a dynamic earth model includesthe evolution of geometry and properties Thepaleogeometrymdashor structural geometry of the modelat a given geologic age (or simulation time step)mdashisstrongly dependent on the paleowater depth (geomet-rical boundary condition) faulting activity the lithol-ogy of the various layers and their associatedcompaction behavior Approaching accurate paleoge-ometry is crucial for achieving consistent resultsespecially for fluid flow simulations In standardBPSM the paleogeometries are automatically gener-ated using a simplified restoration approach knownas backstripping (Figure 2A B)

Geomechanically based structural restorationuses the present-day interpreted and depth-convertedgeometry of a sedimentary basin or subbasin for therestoration of the basin structure backward in geo-logic time (Figure 2C D) This is done by sequen-tially removing the syntectonic sedimentary layersfrom the youngest to the oldest (the reverse of deposi-tion) allowing unfolding and unfaulting during therestoration (the reverse of deformation) In contrastto the simple backstripping approach structural resto-ration allows much more control through additionalgeometrical boundary conditions such as lateralmovement of rocks along faults or bedding slipRecent development has resulted in geomechanicalrestoration algorithms extending the pure geometri-cal approach by applying physical laws of rockdeformation (Maerten and Maerten 2006) The paleo-geometries obtained by structural restoration are thenused to describe basin geometry through time in apetroleum systems model

The backstripping approach is applied to tectoni-cally undisturbed areas and to areas where faultingoccurred much earlier than the main phase of hydro-carbon generation (typically a sag basin or the postriftsequence of a passive margin Figure 2A) In basinswhere extensional faulting occurred and where faultslip has an important horizontal component (eg lis-tric faults) backstripped paleogeometries showldquospikyrdquo artifacts (Figure 2B) these paleogeometries

NEUMAIER ET AL 1329

are generally not good enough for predicting prefault-ing hydrocarbon migration but might be acceptablefor hydrocarbon generation modeling Here structur-ally restored paleogeometries are important for thelayer shape in the proximity of faults and whencross-fault connectivity (like sandndashsand overlaps)matters (Figure 2C) Alternatively manual correctionof paleothickness can be done In halokinetic settingsbackstripping is generally superimposed by simplesalt reconstruction workflows (area balancing andextrapolation) Structural restoration should also beconsidered if salt movement results in geometries thatare more complex In any case thermal modeling inthe vicinity of salt is difficult due to its extremelyhigh thermal conductivity In areas with overthrustingor overturned folds (Figure 2D) with multiple strati-graphic sequences typically a fold and thrust beltbackstripping fails completely because lateral rockmovements are missing In the latter case paleo-geometries need to be generated by means other thanbackstripping (from simple conceptual drawing totrue structural restoration)

Paleogeometries generated from structuralrestorations can be loaded into BPSM software as analternative to performing a calculation from back-stripping Meshing is done section per section (timestep per time step) and tracking of cells from one sec-tion to the other is following rules described inHantschel and Kauerauf (2009) For the ages corre-sponding to the paleosection the basin geometry is

fixed it is not a function of compaction-controlledforward modeling as is the case when using back-stripping The BPSM simulation is restricted to thecalculation of properties such as porosity overpres-sure temperature vitrinite reflectance (Ro) andsource rock maturity which populate the fixed paleo-geometries as well as hydrocarbon migration Thesecompaction-controlled results are not used to adjustthe depositional amounts and to correct the paleoge-ometry as done in the backstripping approach andconsequently geometry optimization cannot beperformed The possible difference between thecompaction-law-controlled porosity change and thepredefined geometry change is indirectly compen-sated by the corresponding adjustment of the solidgrain mass

When paleogeometries are used for basin analy-sis and BPSM some factors become very importantRestoration must focus toward a best estimate paleo-water depth because changes in slope directions andtopography can largely affect fluid flow Erosionshould be estimated and reconstructed and manygeologic events need to be taken into account toincrease the resolution through time for dynamicmodeling The sediment thickness must be correctedin geologic time for decompaction All these factorswhich are common sources of uncertainty are criticalfor a geologically meaningful restoration

Because of the optimization method justdescribed adequately decompacted paleogeometries

Figure 2 Steps for (A B) backstripping and (C D) structural restoration of geometries of different structural complexity For detailssee text

1330 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

are required from structural restoration Inadequatedecompaction can produce large errors resulting inan unbalanced mass of solid rock matrix underesti-mated source rock paleoburial depth and deviatedpaleomigration paths related to differential compac-tion However it is difficult to define adequatedecompaction As a rule of thumb in normally pres-sured clastic basins where the main compactiondriver is linearly increasing effective stress simplehydrostatic decompaction laws such as thosedescribed by Athy (1930) can be used for restorationIn cases of abnormal pressure (overpressure) hydro-static decompaction overestimates the effectivestress or underestimates porosity and might fail tothe same degree as restoration neglecting decompac-tion One might be scaling normal decompactioncurves wherever regional overpressure occurs andhas occurred In basins with complex erosion pres-sure and diagenetic histories optimization loops(using BPSM simulated properties for optimizeddecompaction in second restoration) might benecessary Any sort of decompaction in a carbonate-dominated environment where compaction is gener-ally less understood because of the effect of mineraltransformations and cementation is subject to muchlarger uncertainty The question of whether to usehydrostatically decompacted or nondecompactedstructurally restored paleogeometries is discussed inthe section on uncertainty analysis

MODELING

Prior to structural restoration we performed several1D simulations These generated erosion and paleoal-titude estimations for structural restoration andprovided a calibration for the 2D model One-dimensional modeling also gave the first informationon hydrocarbon generation and timing to identifythe period of first source rock maturation (ldquocriticalmomentrdquo as defined by Magoon and Dow 1994)This is very important because prior to structuralrestoration its identification indicates when addi-tional time steps are needed to increase time resolu-tion for critical time intervals so that the activepetroleum system can be dynamically modeled in anadequate way However for simulation regular time

steps are needed not only during the critical momentbut also for the entire basin history

Structural Restoration

The following section describes our structural resto-ration results shown in Figure 3 of the post-riftstructural evolution of the Monagas fold and thrustbelt It can be divided into three main steps(1) From the Mesozoic to Oligocene the Monagasarea formed parts of the northern passive margin ofthe South American shield (Figure 3A) The sub-sidence driver was mostly thermal cooling and iso-static load with only a few normal faults Thesediments were sourced by the craton in the south(2) In the early Miocene the margin experiencedcompression from the transcollision with theCaribbean plate (Figure 3B) Two major normalfaults that formed during rifting were inverted in thecourse of the collision generating the regionalFurrial and Pirital thrusts Synorogenic deposits werecovered by the thrusted allochthon and intramontanepiggyback basins were formed With increasing upliftand erosion (middle Miocene Figure 3C) their sedi-ment fills were redeposited to form the important syn-tectonic Oficina and Carapita Formations south of thethrusts (3) From late Miocene on the late compres-sional phase started Figure 3D shows the present-day structure of the Monagas fold and thrust belt(2D model crossing the Santa Barbara field)According to our restoration the total shortening onthe section accommodated by the thrusts during thecompression is around 40 km (2486 mi) which cor-responds to 35 It is important to note that the Piritalthrust was active until late Miocene as indicated bythe fault displacement analysis (Figure 4) These latetectonic movements might have an important effecton the seal integrity of prospects that are in the vicin-ity of the thrust

Model Input Data

Table 1 shows geologic ages lithologies and petro-physical parameters that were assigned to our 2Dmodel For the main source rocks (Querecual andSan Antonio Formations) a type II B kinetic wasassigned (Pepper and Corvi 1995) As a base case

NEUMAIER ET AL 1331

we assumed a constant initial TOC of 6 wt and aninitial HI of 500 mg HCg TOC However becauseinitial TOC and HI were estimated from present-daymeasured values these initial values are highly uncer-tain In addition there is both lateral and vertical vari-ability within the layer even though locally very highsource rock potential is likely to occur a bulk TOCand HI need to be assigned to the model cells Forthese reasons several models with different initialTOC have been run and their sensitivity to the mod-eled charge is discussed below Faults were set open

to fluid flow in the course of their main kinematicactivity which is estimated by fault displacementanalysis of the structural restoration models Aftertheir main activity phase faults are assumed to beclosed Fault permeability was also a varying param-eter in the sensitivity analysis discussed belowThermal boundary conditions are given in Table 2Surface temperature ranges between 23 and 30degC(734 and 86degF) through geologic time with apresent-day average temperature of 28degC (824degF)These temperatures at sea level have been corrected

Figure 3 Structural evolution of the Monagas fold and thrust belt demonstrated by structural restoration (A) Passive margin stage(until 23 Ma) (B) Early compression stage (21 Ma) (C) Late compression stage (12 Ma) (D) Present day For details see text

1332 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

Figure 4 Fault displacement from lateMiocene to present day for all faults onrestored cross sections (A) and for the Piritalthrust only (B) Black and white shading andsize of points show magnitude of displacementin meters

NEUMAIER ET AL 1333

Table1

Basin

andPetro

leum

Syste

mModeling(BPSM)Input

ParametersGeologicAges

andLithologies

Compressib

ility

(GPa

-1)

Perm

eability

(logmD)

Thermal

conductivity

(Wm

K)

Heatcapacity

(kcalkgK)

Age

(Ma)

GeologicTime

Facies

Density

(kg∕m

3 )Initial

Porosity(

)max

min

at1

por

at25por

at20degC

at100degC

at20degC

at100degC

0Pleisto

cene

LasP

iedras

sandsto

ne2720

412747

115

minus18

3395

338

02

024

LasPiedrasshale

2700

4040327

403

minus852

minus3

164

169

021

024

26

Pliocene

LasP

iedras

sandsto

ne2720

412747

115

minus18

3395

338

02

024

LasPiedrasshale

2700

4040327

403

minus852

minus3

164

169

021

024

53

upperM

iocene

2La

Pica

sandsto

ne2720

412747

115

minus18

3395

338

02

024

LaPica

shale

2700

4040327

403

minus852

minus3

164

169

021

024

Morich

itoshale

2700

4040327

403

minus852

minus3

164

169

021

024

8upperM

iocene

1La

Pica

sandsto

ne2720

412747

115

minus18

3395

338

02

024

LaPica

shale

2700

4040327

403

minus852

minus3

164

169

021

024

Morich

itobasal

sandsto

ne2715

4812142

187

minus348

15

317

284

02

024

105

middleMiocene

2Carapitasandsto

ne2715

4812142

187

minus348

15

317

284

02

024

Carapitashale

2705

6330932

331

minus684

minus15

204

201

02

024

12middleMiocene

1Carapitashale

2705

6330932

331

minus684

minus15

204

201

02

024

15lower

Miocene

2Carapitashale

2705

6330932

331

minus684

minus15

204

201

02

024

19lower

Miocene

1lower

Carapitashale

2700

7040327

403

minus852

minus3

164

169

021

024

23Oligocene2

Merecureplatform

sandsto

ne2715

4812142

187

minus348

15

317

284

02

024

27Oligocene1

Merecureplatform

sandsto

ne2715

4812142

187

minus348

15

317

284

02

024

1334 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

339

PaleoEocene

Vidono

sandsto

ne2720

412747

115

minus18

3395

338

02

024

Caratasbasin

alshale

2700

7040327

403

minus852

minus3

164

169

021

024

655

upperC

retaceous4

SanJuan

sandsto

ne2720

412747

115

minus18

3395

338

02

024

74upperC

retaceous3

SanAntoniosiltstone

(SOUR

CERO

CK)

2710

5510422

212

minus628

minus1

201

196

023

026

825

upperC

retaceous2

SanAntoniosiltstone

(SOUR

CERO

CK)

2710

5510422

212

minus628

minus1

201

196

023

026

91upperC

retaceous1

Querecualmudsto

ne(SOUR

CERO

CK)

2700

7040327

403

minus852

minus3

164

169

021

024

996

lower

Cretaceous

2undifflower

Cretaceous

(30

Lst50

Sst

20Sh)

2722

459445

141

minus334

174

305

275

02

023

122

lower

Cretaceous

1undifflower

Cretaceous

(30

Lst50

Sst

20Sh)

2722

459445

141

minus334

174

305

275

02

023

1455

sub-Cretaceous

2undiffsub-Cretaceous

(50

Sh50Sst)

2710

5521537

259

minus516

0255

239

02

024

251

NEUMAIER ET AL 1335

by the influence of the paleowater depth using theapproach of Wygrala (1989) Because of the hightopographic elevation of approximately 3500 m(11500 ft) the maximum altitude during the pastcompressional phase we adjusted the model takinginto account a temperature decrease of 6degCkm (33degF1000 ft) related to the paleoaltitude This resultsin a difference of up to 20degC (36degF) compared to sea-level temperature We then assigned the adjustedsedimentwater interface temperatures to the modelas the upper thermal boundary condition At the baseof the model we assigned a heat flow ranging from42 to 80 mW∕m2 through geologic time The highestheat flow is assumed during the rifting event beforethe basin cooled as a passive margin A thermal eventin the Eocene detected by fission track analysis(Locke and Garver 2005 Perez de Armas 2005)has been addressed by a second peak before a secondcooling phase For uncertainty analysis basal paleo-heat flow has been varied within geological meaning-ful ranges and the thermal calibration data has beenhonored as much as possible

Modeling Results

Basin and petroleum system modeling can be dividedinto simulation of the physical parameters within abasin such as pressure and temperature (basin model-ing) and the subsequent evolution of the source rocksand their derived hydrocarbon compounds (petro-leum systems modeling) In the following sectionswe use this subdivision to describe the BPSM resultscompaction modeling thermal modeling and chargehistory It is important to state that these are predic-tions based on forward modeling as just describedThe results are compared and calibrated to actualmeasurements and observations where available

Compaction Modeling

Figure 5 presents an extraction of the modeled resultsat the location of Well Alpha in the Furrial Trend Inthe absence of direct calibration data for pore pres-sure we undertook a sensitivity analysis to test thedifferent scenarios Using the model we tested howpore pressure in the main reservoir of the FurrialTrend reacts to changes in fault and seal-rock per-meability In a first step fault permeability has beenevaluated Faults were assumed open during the mainfault activity phase and closed afterward In additionwe performed several variations from that base casescenario Figure 5A shows the effect of two end-member scenarios on pore pressure within the mainreservoir of the Furrial Trend (Santa Barbara field) inwhich faults have been modeled to be either open orclosed since their formation A third scenario consistsof thrust faults that are open during the compressionalphase (from 15 Ma on) and closed afterward (from105 Ma on) This scenario has been chosen as theldquomaster scenariordquo because it results in overpressureof 23 MPa which is in the range of pressure reportedin similar structures of the Furrial Trend by Schneideret al (2004) In addition this pressure scenariomatches the calibration data for porosity (Figure 5B)

The 2D distribution of modeled overpressure atpresent day is shown in Figure 6A Apart fromimportant overpressure in the Querecual Formationthe allochthon is almost normally pressured Thethrust wedge is overpressured especially in thelow-permeability Carapita Formation Although

Table 2 Basin and Petroleum System Modeling (BPSM) InputParameters Thermal Boundary Conditions

Age (Ma)

SedimentWaterInterface

Temperature (degC)

Age (Ma)

Basal HeatFlow

(mW∕m2)min max

0 28 28 0 4226 28 28 30 6053 76 23 34 708 184 231 50 50105 62 233 107 6012 31 238 1455 8015 5 242 260 6019 5 24723 234 2527 78 235339 94 236655 146 2831455 148 287255 5 239

1336 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

the autochthon is highly overpressured close tonormal hydrostatic pressures are simulated in theforeland basin The discussed model assumption ofclosed faults in the postcompressional phase resultsin a regional pressure compartmentalization

The simulated pressure evolution through geo-logic time can be seen in Figure 7A A time extractionof a source rock cell in the thrust wedge shows a sud-den increase of both lithostatic and hydrostatic pres-sures that is a consequence of the tectonic burial

load associated with the thrusting In addition themodeled pore pressure buildup even though delayedin time can be directly linked to the thrust event Itcan be clearly seen that the timing of fault permeabil-ity directly impacts changes in the present-day porepressure The earlier the faults close the more over-pressure generated by the massive syntectonic sedi-mentation is kept until present day given adequatesealing lithologies are present The present-day over-pressure is modeled as14 MPa when the faults are

Figure 5 Plots of one-dimensional model output and available calibration data for Well Alpha (Table 3) (A) Dotted lines are hydro-static and lithostatic pressures continuous lines are pore pressure the master scenario is in black (a1 faults closing at 105 Ma) andadditional scenarios in gray (a2 open faults a3 closed faults) (B) modeled porosity (C) borehole temperature and (D) Ro The blackline shows the master scenario and the gray line the alternative high-heat-flow scenario Bold lines in the stratigraphic column arethrusts (T) and unconformities (U)

Table 3 Calibration Data for Well Alpha (Porosity Temperature Ro)

TVD (m) Porosity () TVD (m) Corrected Temperature (degC) TVD (m) Vitrinite Reflectance (Ro)

4350 15 600 40 400 0264400 15 1100 53 1200 0484800 10 1800 68 1485 068

4400 131 1700 0394800 133 2900 048

3100 0493400 0494000 054300 0524800 06

NEUMAIER ET AL 1337

open since 8 Ma as 23 MPa for 105 Ma and as37 MPa for 12 Ma Compared to this order of magni-tude changes in the permeability of the overlyingCarapita shale are negligible for the pore pressureTherefore the controlling parameter for pore pressurein the main reservoir of the Furrial trend is assumedfault permeability and resulting lateral connectivityrather than the permeability of the seal rock

Thermal Modeling

Modeled temperature and thermal maturity(Figure 5C D) have been calibrated with availablewell data Some discontinuities in the modeled Ro

trend are noted (Figure 5D) The upper kinks (egbase Pliocene) are related to unconformities withassociated erosion and the lower one (base UpperCretaceous) is due to the well penetrating a reverse

Figure 6 Two-dimensional model at present day with (A) modeled overpressure and (B) modeled thermal maturity and isothermsSource rocks marked by lithology pattern

1338 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

fault displacing mature rocks upon less mature ones(see Figure 6B for 2D view) There is a mismatchbetween the modeled Ro trend and the data in thedeeper part of the well In the Oligocene rocks themodeled temperature is too low (2degC or 36degF)whereas the modeled Ro is too high (up to 02 Ro)Any attempt to reduce the modeled Ro (eg bychanging thermal conductivity of the MioceneCarapita Formation) also results in a lower modeledpresent-day temperature Therefore the thermal sce-nario used for this calibration is a trade-off betweenmatching present-day temperature and Ro (using thekinetics of Sweeney and Burnham 1990) The basalheat flow assumed in the master scenario (Table 2)results in a modeled present-day surface heat flowranging from 37 mW∕m2 in the southern part of thethrusts to 52 mW∕m2 in parts of the uplifted areaAt the same depth the allochthonous northern partis thermally more mature than the southern autochthon(Figure 6B) because of the recent uplift Locallyinverted trends around the thrusts can be observedfor example at the location of Well Alpha (seeFigure 5D) In the autochthon the Upper Cretaceoussource rocks are entirely in the oil or even gas windowat present day because of the Neogene tectonic loadingand synorogenic deposition In the northern alloch-thon most of the Upper Cretaceous source rocks havebeen in the oil window in the past The modeled ther-mal maturity is in good agreement with the regionalmaturity suggested by Summa et al (2003)

Similar to the pressures the thrusting also affectsthe modeled temperature and Ro In the thrust wedgea direct relationship exists between the thrusting andthe increase of temperature and maturity The thrust-ing is the trigger for source rocks entering the oil win-dow within the thrust wedge (Figure 7B) and thepresent-day thermal maturity has been reached sincethe last 10 Ma Although this recent thermal historycan be considered as relatively well constrained bythe measured Ro in the calibration well this doesnot apply to the past In this thrusted area any ther-mal maturity reached prior to the thrusting onset hasbeen ldquooverprintedrdquo and therefore no thermal calibra-tion is possible for this period However theCretaceous source rocks were not buried deeply dur-ing the passive margin phase and even very hot sce-narios (basal heat flow of 100 mW∕m2) do not

significantly change their maturity (Figure 7B) Inthe uplifted allochthon several wells show low(03) to moderate (1) measured Ro (Parra et al2011) Because the allochthon experienced its maxi-mum burial and the highest temperatures during thepassive margin phase these maturity levels can beassumed as representative of the maturity prior tothe compression In contrast to the thrust wedge theconsequence of the thrusting is inverse for the alloch-thon (Figure 7C) as the hanging wall is uplifted andthe accompanying erosion removes the overburdenThe highest temperatures are simulated to have beenreached just before the thrusting event Then temper-ature decreased and Ro stagnated leaving the sourcerocks in the early mature zone until present day Theassumption of a regionally relatively homogeneousthermal field during the cooling phase of the passivemargin in combination with the relatively low matur-ities measured in the uplifted allochthon makes thehot heat-flow scenario unlikely However a laterallyvarying heat flow cannot be excluded and the conse-quently hotter paleoheat flow would have led to muchhigher temperatures prior to compression resultingalso in a higher present-day maturity level (14)despite the uplift (Figure 7C) Any variations of basalheat flow during the rift period (Late Jurassic to EarlyCretaceous) occurred too early to affect the maturityof the source rocks deposited during the lateCretaceous

Charge History

After calibration of the basinrsquos physical environ-ment the evolution of the petroleum systems wasmodeled Hydrocarbon generation was simulatedthrough geologic time using the activation energydistribution of the assigned source rock kineticsFrom the moment of hydrocarbon expulsiononward secondary migration was modeled usingthe invasion percolation method (Wilkinson 1984)It takes into account the geometry overpressure gra-dient and capillary entry pressures of the rockswhich are opposed to the buoyancy of the migratingoil and gas Figure 8 shows the time evolution of thetotal study area and Figure 9 focuses on the thrustwedge As already mentioned the described chargehistory is based on forward modeling It is consistent

NEUMAIER ET AL 1339

Figure 7 Time extractions (see Figure 6 for location) (A) Pressure development through geologic time for a source rock in the ver-tical continuity of Well Alpha Dotted lines are hydrostatic and lithostatic pressures continuous lines are pore pressure the master sce-nario is in black (a1 faults closing at 105 Ma) and additional scenarios in gray (a2 open faults a3 closed faults a4 faults closing at12 Ma a5 faults closing at 8 Ma) (B) Temperature (continuous lines) and Ro (dotted lines) through geologic time for a source rockin the vertical continuity of Well Alpha the black line shows the master scenario and the gray line the alternative high-heat-flow sce-nario (C) Temperature (continuous lines) and Ro (dotted lines) through geologic time for a source rock in the uplifted allochthonthe black line shows the master scenario and the gray line the alternative high heat flow scenario

1340 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

with the modeled compaction and thermal historyand therefore shows one scenario that explains thecharge of the Santa Barbara field and the forelandbasin Simplifications are inherent to any such work-flow and shall be considered when analyzing themodeling results

Passive Margin PhasemdashMesozoic toOligocene

Modeling results indicate that already during the LateCretaceous any Jurassic source rocks potentiallydeposited in the northern part of the section (Summaet al 2003) have reached temperatures of more than200degC (392degF) and corresponding high levels of matu-rity However the temperature at the base of theQuerecual Formation (Upper Cretaceous) was lessthan 100degC (212degF) leaving the kerogen immatureThe source rock became effective only during lateEocene with temperatures of up to 120degC (248degF)and a (reactive kerogen to hydrocarbon) transforma-tion ratio (TR) of more than 20 in the deeper distalpart of the passive margin Oil was quickly expelledmigrated vertically and seeped to the surface becauseof the general lack of seals in the overburden Duringthe Oligocene the time of deposition of the Merecurereservoir sandstones in the southern platform this sit-uation did not change apart from increased transforma-tion in the deeper basinal part of the passive margin

First oil accumulations are modeled in the Vidontildeosandstones in the basinal part in the late Oligocene(Figure 8A) in which the base of the QuerecualFormation shows TRs of up to 60 at around 130degC(266degF) In the proximal part there is no hydrocarbongeneration Although there is a general northwardtilting slope no southward migration from the distalmargin is modeled this has been checked usingdifferent migration methods including Darcy flowwith scenarios of continuous intrasource carrier bedsThis is in contrast to the studies from Gallango andParnaud (1995) and Schneider (2003) who suggestedlateral migration at that stage The difference couldbe explained by the presence of a regional top seal intheir models However our model suggests that theshallow rocks do not have sufficient sealing capacityto allow lateral hydrocarbon migration During thepassive-margin phase the basin was normally

pressured (Figure 7A) and simulated stresses havenot yet resulted in any fracturing at that stage

Main Compressional PhasemdashEarly to MiddleMiocene

With the onset of compression during early Miocene(Figure 8B) both the Pirital and Furrial thrusts wereformed and the northern distal part of the passivemargin was rapidly uplifted Important overpressuresas well as mechanical rock failure both the result ofthe rapid uplift are simulated in the allochthonDespite this uplift the kerogen-to-hydrocarbon trans-formation continued and the rate even increasedlocally close to the footwall cutoffs and where syn-orogenic deposition provided rapid overburden load-ing Oil accumulations are modeled in the freshlycreated anticlines

The tectonicndashsedimentary wedge between thetwo main thrusts experienced very rapid burial Infact the Upper Cretaceous source rocks were broughtinto depths where temperatures were as high as185degC (365degF) at 19 Ma and 225degC (437degF) at15 Ma Especially close to the cutoff of the Piritalthrust they very quickly reached a high maturity at19 Ma and reached full transformation in most placesat 12 Ma The expelled hydrocarbons migrated intothe sandstones of the Merecure Formation which weresealed by the Carapita Formation The charge toward aseries of recent anticlines is modeled by the fill-to-spillmechanism from north to south (Figure 9C D) Afterthe first oil fill the northern prospect quickly receiveda local gas charge the central structure containedmainly oil Note the presence of a lateral seal that isdue to sandndashshale juxtaposition The southernmoststructure which forms the Santa Barbara field todaywas charged by oil only at 15 Ma (Figure 9D)

The generated hydrocarbons in the southernautochthon were leaking to the surface until theCarapita Formation was compacted enough to holdsome of them (Figure 8C) At 12 Ma the wedgesource rocks were mostly spent but expulsion wasstill ongoing Important rock failure is modeled tohave occurred in the thrusted footwall breaking theformerly sealing Carapita Formation (Figure 10)This is in good agreement with cemented hydraulicfractures observed in a few cores in the Carapita

NEUMAIER ET AL 1341

Figure 8 Two-dimensional model through geologic time (selected time steps only) with modeled transformation ratio for the sourcerock layers isotherms and liquid and vapor migration and accumulation

1342 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

shales from the Furrial Trend (Roure et al 2005)This important fracturing event resulted in thedestruction of the modeled accumulation (Figure 9E)Hydrocarbon generation and expulsion were alsohappening in the southern foreland basin at transfor-mation ratios of up to 30

Late Compressional and PostcompressionalPhasemdashLate Miocene to Present Day

The late compressional phase was marked by the par-tial destruction of the allochthonous petroleum systemsby erosion (Figure 8E F) Many traps disappeared and

Figure 8 Continued

NEUMAIER ET AL 1343

the hydrocarbon generation within the source rocksstopped The erosion triggered an important sedimenttransfer toward the south contributing to further burialand increased temperatures in the thrust wedge and theforeland basin During late Miocene the petroleumsystem in the wedge matured An important northwardtilting was triggered by flexural tectonic loading andthe loading of syntectonic sediments (Summa et al2003) In combination with the presence of an effec-tive regional top seal the compacted CarapitaFormation allowed a change from dominantly verticalmigration into lateral migration First accumulationsare modeled in the Neogene synorogenic sequence

reservoirs in the south and important southwardmigration toward the foreland basin started(Figure 8G) These trends continued through Plioceneand Pleistocene times to present day filling theOnado field (Figure 8H) and other known fields inthe south (Figure 1A) of the Eastern Venezuela Basin

Prospectivity

We modeled the charge of the Santa Barbara field inthe Furrial Trend and of the Onado field in the fore-land basin (Figure 8H) Other prospects in the thrustwedge were also investigated The structure called

Figure 9 Detail of two-dimensionalmodel through geologic time (seeFigure 8) showing modeled transforma-tion ratio for the source rock layers iso-therms and liquid and vapor migrationand accumulation The Santa Barbara fieldand the prospect are shown by ellipses

1344 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

ldquoprospectrdquo (Figure 9G) looks similar to that of theSanta Barbara field because both are anticlines andthe reservoir and seal formations are the sameHowever they differ in several aspects First theprospect is in the vicinity of the Pirital thrust whichpresents a high risk of impacting the seal integrityThe prospect location is directly below the area wherewe presume the maximum of late tectonic activityoccurred (Figure 4) As modeled the CarapitaFormation the seal of the prospect was already brit-tle when this deformation occurred Calculatedstresses also predict rock failure from 12 Ma onward(Figure 10) Even though charge continued ourmodel suggests continuous leakage of the structure(Figure 9EndashG) The second major difference is thatthe source rock in the fault block of the prospect ismodeled to have already started generating and expel-ling hydrocarbons during the Oligocene and earlyMiocene (Figure 9A) before the trap formed and the

seal became effective (Figures 9C 10) Even thoughgeneration expulsion and migration continued aftertrap formation the prospect was not charged withthe earlier expelled oil In addition the present-daymaturity of the source rocks within the prospectrsquosfault block is higher than in the source rocks withinthe tectonic block of the Santa Barbara field poten-tially indicating expulsion of more gas A southndashnorth trend can be seen on the modeled accumulationcomposition with more gas toward the allochthon(Figure 9) The modeled composition of the prospectprior to seal failure was much richer in gas than inthe neighboring Santa Barbara field (Figure 9D) Infact simulation scenarios with unbreached sealsresulted in accumulation of predominantly gas Thecombination of both seal integrity and charge compo-sition concerns makes this a risky prospect In addi-tion potential diagenetic fluids related to the faultfluid flow might have altered the reservoir quality

Figure 10 Two-dimensional model (at 12 Ma) with modeled stress to failure Mohr circles (with σ standing for normal stress and τfor shear stress) and petroleum system events charts for the Santa Barbara field (left) and the prospect (right) Model predicts rock fail-ure in the prospect from about 12 Ma For further details see text

NEUMAIER ET AL 1345

Apart from the Onado oil field only minor accu-mulations are modeled in the southern Neogene syn-orogenic sequence The proportion of gas is higherthan in the Furrial trend The accumulations at thesouthern model edge are a modeling artifact repre-senting hydrocarbons (mainly oil) that in reality arethought to have charged the southern foreland basin(from mid to late Miocene times to present day)However parts of the migrated oil and gas could havebeen trapped in the Upper Cretaceous San JuanFormation south of the horst structure

The potential for conventional hydrocarbon accu-mulations in the allochthon is very limited by theimportant fracturing that occurred during the upliftperiod The basal sandstones of the Morochito piggy-back basin north of the Pirital thrust are not chargedin our 2D model because of the absence of structuraltraps Nevertheless there is a potential for strati-graphic traps to have been charged as migrationthrough the basin occurred However seal strengthmight be critical at this low-compaction stage Thesame applies for the Las Piedras Formation We thinkthere might be potential for shale oil because theQuerecual Formation is at shallow depths but showsmaturity within the oil window From the thermalmaturity point of view any potential Jurassic sourcerocks as discussed by Summa et al (2003) mightbe a gas source or shale gas target

UNCERTAINTY ANALYSIS

There is often significant uncertainty in basin geom-etry and depth of stratigraphic horizons based onuncertain seismic interpretation and velocity models(Kukla et al 2000 Brandes et al 2008 Baur et al2010) Restoring of present-day geometries to obtainpaleogeometries adds further uncertainty Additionaluncertainty is related to thermal boundary conditionssuch as paleoheat flow (Beha et al 2008) and togeochemical properties such as source rock character-istics and petroleum generation kinetics and calcula-tion of biodegradation and petroleum density(Blumenstein et al 2008) In addition to the porepressure and thermal sensitivity analysis justdescribed we varied initial TOC within a range of2 to 10 A variation of initial TOC obviously

directly scales the amount of initial kerogen availablefor transformation into hydrocarbons Neither thecharge volume nor composition of the Santa Barbarafield nor the wedge prospects are significantly sensi-tive to variation of initial TOC these traps are filledto spill in each scenario However the amount ofhydrocarbons charging the foreland basin (eg theOnado field) is reduced in the case of low initialTOC A high initial TOC only increased the modeledamount of surface seepage The sensitivity of the ini-tial HI is analogous to the sensitivity of the ini-tial TOC

In addition to the more classical uncertaintyand sensitivity analysis described we focused onquantitative analysis of the decompaction uncer-tainty when performing structural restoration Thisuncertainty is particular to the applied workflowof using structurally restored paleogeometries inBPSM Note that this uncertainty is significantlylower when using the backstripping method withthe optimization algorithms discussed One methodto analyze the adequacy of the decompactionmethod is comparison of the output rock matrixmass balance In the backstripping approach therock matrix mass is kept constant of a model cellwith a simulated decrease in porosity controlledby compaction laws and the corresponding reduc-tion of the layer thickness When using fixed pale-ogeometries from structural restoration the cellthickness is imposed However the decrease inporosity is calculated and the mass is adapted tofit the imposed cell thickness That means that therock matrix mass is constant only if ideal decom-paction has been applied

Figure 11A shows the percentage variation fromthe present-day matrix mass (100) of a given sourcerock cell through geologic time The dotted line rep-resents the ideal case with constant rock matrix massWe used two end-member decompaction scenarios todescribe the decompaction-related uncertainty that isintroduced when fixed paleogeometries replace back-stripping idealized hydrostatic decompaction assum-ing no overpressure and idealized lithostaticdecompaction assuming no effective stress (nodecompaction resulting in constant cell thicknessthrough time) In the latter case with decreasingporosity during compaction the rock matrix mass

1346 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

increases until it reaches the present-day mass Inother words rock mass matrix has been highly under-estimated in the past (in this particular case by morethan 20) This is a source rock so the kerogen massdecreases by the same factor This yields an underes-timation of more than 20 in TOC Note the rockmatrix mass is constant for time steps prior to23 Ma in which backstripping optimization couldbe performed toward the earliest paleogeometry Forthe second end-member scenario of idealized hydro-static decompaction the rock matrix mass is nearlyconstant which is a good decompaction adequacy

indicator for that given cell in the model Howeverthe so-measured adequacy changes throughout themodel In some areas the rock matrix mass has beenunderestimated by about 50 for nondecompactedpaleogeometries however in the case of hydrostaticdecompaction values generally range from 80 to120 of the present-day rock matrix mass Thatmeans in this particular case that hydrostatic decom-paction is more adequate (or less wrong) than nodecompaction at all The decompaction adequacycan be roughly correlated with pressure changesthrough geologic time

Figure 11 Time extractions for comparison of two end-member decompaction scenarios (see Figure 6 for location) (A) Rock matrixmass conservation through geologic time (B) Source rock burial depth through geologic time (C) Temperature and transformation ratiothrough geologic time

NEUMAIER ET AL 1347

Inadequate decompaction results in wrong depthfor stratigraphic horizons in the paleogeometriesFigure 11B shows the burial depth of the same inves-tigated source rock cell We can see that during thepassive margin phase there is a slight difference inpaleoburial depth for the two end-member scenariosDuring the compressional phase the differencechanges dramatically reaching more than 1000 m(3281 ft) in some places Toward present day bothcurves meet again because the input to the two resto-ration scenarios was the same (identical present-dayburial depth) This difference in compaction calcula-tion has severe implications on the modeled tempera-ture history and the kerogen-to-hydrocarbontransformation (Figure 11C) For an identical thermalframework in terms of modeling input parameterstemperature differences of up to 25degC (45degF) resultin TR differences of more than 20 in the early com-pressional phase before TR values meet again for thepresent-day situation

CONCLUSIONS

Structurally complex areas are attracting increasinginterest in hydrocarbon exploration Exploration riskcan be substantially narrowed by analyzing the his-tory of the basin and its petroleum systems based onstructural evolution We presented a further progressin technology that allows the integration of structuralrestoration and BPSM This combined analysis isseen as a way forward to improve modeling in thrustbelts with stacked stratigraphic sequences Despitelimitations of nonoptimized decompaction whichare added to the general uncertainty related to theindividual methods we modeled the charge and sealhistory of the Monagas fold and thrust belt inVenezuela Calibrated porosity temperature andthermal maturity as well as basin-scale stress pro-vided the regional framework through geologic timein which the petroleum systems evolved such thatthe charge of the known Santa Barbara and Onadofields and other trends could be understood We dis-cussed seal integrity of a potential exploration pros-pect by analyzing structural restoration-derived faultdisplacement through geologic time and an advancedstress forward simulation The uncertainty of using

decompacted or nondecompacted structural restora-tion has been assessed It can have important influ-ence on rock matrix mass balance through geologictime paleogeometry and paleoburial depth andtherefore affect paleotemperatures and paleomatur-ities We conclude that decompaction assuminghydrostatic pressure is the best option in that particu-lar case

REFERENCES CITED

Alberdi M and E Lafargue 1993 Vertical variations of organicmatter content in Guayuta Group (Upper Cretaceous)interior mountain belt eastern Venezuela OrganicGeochemistry v 20 p 425ndash436 doi 1010160146-6380(93)90091-O

Athy L F 1930 Density porosity and compaction of sedimen-tary rocks AAPG Bulletin v 14 p 1ndash24

Aymard R L Pimentel P Eitz P Lopez A ChaouchJ Navarro J Mijares and J G Pereira 1990 Geologic inte-gration and evaluation of Northern Monagas EasternVenezuelan Basin in J Brooks ed Classic petroleum prov-inces The Geologic Society (London) Special Publicationv 50 p 37ndash53

Baur F M Di Benedetto T Fuchs C Lampe andS Sciamanna 2009 Integrating structural geology andpetroleum systems modelingmdashA pilot project from Boliviarsquosfold and thrust belt Marine and Petroleum Geology v 26no 4 p 573ndash579 doi 101016jmarpetgeo200901004

Baur F R Littke H Wielens C Lampe and T Fuchs 2010Basin modeling meets rift analysismdashA numerical modelingstudy from the Jeanne drsquoArc basin offshore NewfoundlandCanada Marine and Petroleum Geology v 27 p 585ndash599doi 101016jmarpetgeo200906003

Beha A R O Thomsen and R Littke 2008 Thermal historyhydrocarbon generation and migration in the Horn Grabenin the Danish North Sea A 2D basin modelling studyInternational Journal of Earth Sciences v 97 p 1087ndash1100doi 101007s00531-007-0247-2

Blumenstein I O B M Krooss R di Primio W RottkeE Muumlller C Westerlage and R Littke 2008Biodegradation in numerical basin modelling A case studyfrom the Gifhorn Trough N-Germany International Journalof Earth Sciences v 97 p 1115ndash1129 doi 101007s00531-007-0272-1

Brandes C A Astorga R Littke and J Winsemann 2008Basin modelling of the Limoacuten back-arc basin (Costa Rica)Burial history and temperature evolution of an island arc-related basin-system Basin Research v 20 p 119ndash142doi 101111bre200820issue-1

Carnevali J O 1988 El Furrial oil field northeasternVenezuela First giant in foreland fold and thrust belts ofwestern hemisphere AAPG Bulletin v 72 68 p

Chevalier Y 1987 Contribution a lrsquoetude geacuteologique de lafrontiereacute Sud-Est de la plaque Caraibe Les zones internesde la chaine Sud-Caraibe sur le transect ile de Margarita-

1348 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

INTRODUCTION

Basin and petroleum systems modeling (BPSM) is essential topetroleum exploration because it provides quantitative estimatesof the hydrocarbon charge history (hydrocarbon generationmigration accumulation and respective timing) and enableshydrocarbon volume and property predictions (Hantschel andKauerauf 2009) Basin and petroleum systems modeling is alsocommonly used for pore pressure prediction as an alternative toseismic velocityndashderived pore pressure prediction especiallywhere the seismic data quality is poor for example in subsaltand subthrust settings Standard BPSM is however limited forcomplex basin geometries especially in areas with thrusting andsalt movement In these areas structural restoration whichaccounts for lateral rock movement caused by faulting and fold-ing should be applied to describe the geometrical evolution of abasin Recently structural restoration methods incorporate geo-mechanics during the restoration process while keeping a recordof the evolving strain and stress distributions (Maerten andMaerten 2006) The structurally restored paleogeometries (struc-tural geometry of the basin at a given geologic time) can then beused in BPSM replacing the conventional backstripping methodThe integration of structural restoration into BPSM provides a bet-ter understanding of the petroleum systemrsquos evolution throughgeologic time in complex tectonic settings This workflow hasbeen successfully applied to several case studies with both two-dimensional (2D) and three-dimensional (3D) models (eg Bauret al 2009)

The traditional stressndashstrain model has also undergone recentimprovements To more accurately predict a compaction-relateddecrease in porosity lateral compressional boundary conditionscan be applied resulting in higher horizontal stresses and porepressures (Hantschel et al 2012) This advanced method replacesthe simple approach using vertical lithostatic pressure (Terzaghi1923) which is simply equal to overburden load The evolutionof geomechanical properties calculated by a combined approachof structural restoration and BPSM analysis can assist in predict-ing fracturing andor seal integrity through geologic time

In this paper we describe the theoretical background of theconcept as well as its applicability and limitations Dynel softwarefor structural restoration and PetroMod (Mark of Schlumberger)software for petroleum system modeling were applied in a com-bined 2D basin and petroleum systems analysis workflow of thewestern Monagas fold and thrust belt (Venezuela) This workflowenabled us to describe the compaction pressure and temperaturehistory and evolution of the different petroleum systems throughgeologic time Structural controls on the petroleum systems such

Schlumberger in 2008 Thomasrsquo researchinterests are centered on heat flow porepressure and fluid flow rock stressgeochemical models unconventionalpetroleum systems domains in which hehas contributed many publications He iscurrently center manager of the Aachen-based Technology Center for PetroleumSystems Modeling and Exploration Geology

Laurent Maerten sim Schlumberger 340 rueLouis Pasteur Grabels France lmaertenslbcom

Laurent Maerten received his PhD ingeology and environmental sciences in 1999from Stanford University (USA) incollaboration with Norsk Hydro He workedat the French Petroleum Institute in Paris(France) as a research engineer in structuralgeology for two years In 2004 Laurentfounded the structural geology companyIGEOSS of which he was the CEO andprincipal consultant He joinedSchlumberger in 2010 with its acquisition ofIGEOSS His main interests are structuralgeology and geomechanics with a strongfocus on rock and fracture mechanicsstructural restoration and applications tounconventional reservoirs

Jean-Pierre Joonnekindt sim Schlumberger340 rue Louis Pasteur Grabels Francejjoonnekindtslbcom

Jean-Pierre Joonnekindt received hismasterrsquos degree in structural geology in1994 from the Earth Sciences MontpellierUniversity (France) in collaboration withJean-Pierre Petit and a masterrsquos degree incomputer sciences in 2000 from the JosephFourier Grenoble University (France) Hejoined IGEOSS in 2008 as a structuralgeology consultant and Schlumberger in2010 with its acquisition of IGEOSS Hisinterests focus on geological structures 2Dand 3D restoration geomechanics fracturemechanics and the development ofinnovative methods to improve hydrocarbonproduction in fractured reservoirs

Peter Kukla sim Geological Institute Energyand Mineral Resources (EMR) RWTHAachen University Wuellnerstr 2 52062Aachen Germany kuklageolrwth-aachende

1326 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

as the effects of thrusting on pressure and hydrocarbon generationdynamic structure-related migration pathways and the generalimpact of deformation are described Additionally we evaluatedseal integrity by analyzing simulated stresses through geo-logic time

GEOLOGIC SETTING

The Eastern Venezuela Basin (Figure 1A) is bounded by theSerrania del Interior and the El Pilar fault to the north by a dextralstrike-slip fault system at the interface of the Caribbean and SouthAmerican tectonic plates and by the Orinoco River and thePrecambrian Guyana shield to the south (Hedberg 1950Passalacqua et al 1995 Pindell and Tabbutt 1995) The northernparts of the basin the Monagas fold and thrust belt override a lessdeformed foreland part of the basin to the south

The basin initially formed as a passive margin on the SouthAmerican shield following late Jurassic to early Cretaceous rift-ing During the transcollision of the Caribbean plate with theSouth American plate in the late Cenozoic subsidence becameflexural and extensive folding and thrusting occurred in thenorthern part of the basin (Figure 1B) Roure et al (2003)described both the geodynamic history and the stratigraphy(Figure 1C) Cenozoic platformal and basinal facies of the passivemargin overlie the Mesozoic sediments which include the preriftand synrift sequences The most recent sediments belong to thesynorogenic north-sourced depositional sequence The principalsource rocks in the area are within the Querecual and SanAntonio Formations of the Guayuta Group (Talukdar et al1988 Gallango et al 1992 Alberdi and Lafargue 1993) Thesemudstones are characterized by type II kerogen with 2 to 6 wttotal organic carbon (TOC) for partly highly mature samplesHydrogen index (HI) values are as high as 500 mg HCg TOCSumma et al (2003) suggested an initial TOC as high as 12 forpresent-day values of up to 8 with an HI as high as 700 mgHCg TOC Minor source rocks exist in the Carapita Formationparticularly in the basal part with mixed type IIIII kerogenSumma et al (2003) estimated TOC up to 45 and an HI of350 mg HCg TOC They also discussed the presence of Jurassicand Albian source rocks in the Serrania del Interior

Surface oil seeps and asphalt lakes are very common in thisarea and hydrocarbons have been found since the early 20th cen-tury (Young 1978 Carnevali 1988 Krause and James 1989Aymard et al 1990 James 1990 2000a 2000b Erlich andBarrett 1992 Prieto and Valdes 1992) In the south heavy oiland tar sands (Oronico Oil Belt see Figure 1A) and conventionaloil accumulations are abundant In the thrust area the Furrial trend

Peter Kukla is a professor of geology anddirector of the Geological Institute at RWTHAachen University since 2000 He workedfor Shell International in the Netherlands(1991ndash1996) and in Australia (seconded toWoodside Energy in Perth 1996ndash2000)Peter graduated with degrees in geologyfrom Wuerzburg University Germany(MSc 1986) and Witwatersrand UniversitySouth Africa (PhD 1991) His researchinterests include basin analysis clastic andcarbonate reservoir characterizationunconventional gas geodynamics of saltbasins and pore pressures

ACKNOWLEDGMENTS

We thank Francois Roure (Institut Francaisdu Petrole) Sandro Serra (SerraGeoConsulting) and an anonymousreviewer for their comments andsuggestions Ian Bryant Martin RohdeAdrian Kleine Alan Monahan Will StapleyArmin Kauerauf and Thomas Fuchs(Schlumberger) also provided very helpfultechnical contribution discussions andinternal reviews

The AAPG Editor thanks the followingreviewers for their work on this paperFrancois M Roure Sandro Serra Carl KSteffensen and an anonymous reviewer fortheir work on this paper

NEUMAIER ET AL 1327

contains some of the worldrsquos largest oil reserves forbasins with this type of tectonics Many shallow oilfields in the Eastern Venezuela Basin are located inNeogene sandstone reservoir rocks (Las Piedras LaPica Morichito Chapapotal Freites and Oficinafields) Other reservoir formations are the synoro-genic Miocene Naricual Formation and theOligocene Merecure Formation which are part of aproximal passive-margin sequence that shales outtoward the north The latter is the main reservoir inthe thrusted Furrial trend In the deeper passive-margin sequence there are potential reservoirs in theMesozoic (eg San Juan Formation) Because of its

shaly lithology large thickness and lateral continu-ity the synorogenic Carapita Formation is probablythe most effective seal However several other for-mations have good sealing capacities even thoughnot well compacted as numerous fields at shallowdepths prove

The first publications on the thermal history ofthe Eastern Venezuela Basin were based on one-dimensional (1D) models Because of the com-plexities leading to multiple sequences caused byoverthrusting modeling was done either in theunthrusted area or if within the thrusted area onlyfor geologic time intervals preceding thrusting

Figure 1 (A) Map of the Eastern Venezuela Basin showing structural elements and hydrocarbon occurrence (modified after Parnaudet al 1995) Solid line shows location of two-dimensional (2D) line modeled in this study (B) Generalized northndashsouth geodynamiccross section of the study area (modified after Chevalier 1987) (C) Synthetic stratigraphy of the southern foreland basin of theEastern Venezuela Basin (modified after Roure et al 2003)

1328 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

(Talukdar et al 1988) Summa et al (2003) pub-lished multiple 1D models over a large area andextrapolated the results onto regional maturity mapsfor the Guayuta Group and Cenozoic source rocksThese maps include a hanging-wall cutoff and there-fore do not represent the higher maturities found inthe thrust wedges Elsewhere the authors suggestedthat most source rocks have been in the oil windowin the past A present-day drainage area analysis ofthe top Cretaceous structure map was used to describethe southward charge from the deeper kitchen into theforeland basin This regional migration modelexplains oil provenance and quality Important timingaspects such as the relationship between the struc-tural history and source rock maturation and expul-sion are described using a regional structuralrestoration in conjunction with the 1D models Parraet al (2011) included the thrusting effect by using a15D module mimicking lateral rock transportationthat is due to thrusting Gallango and Parnaud(1995) performed 2D modeling including hydrocar-bon migration in the unthrusted southern area Theyalso described a second model within the thrustedarea that simulates the basin history prior to compres-sion (from the Mesozoic to early Miocene) Thismodel proposes long-distance petroleum migrationsouthward before the transpressional event whichresulted in the observed structural complexity Inother papers Roure et al (2003) Schneider (2003)and Schneider et al (2004) described fluid flow porefluid history and diagenetic processes in a 2D modelalong a cross section parallel to the location of themodel presented here

MODELING METHODS

To address the charge and seal history of the SantaBarbara transect (Figure 1) in the western Monagasfold and thrust belt we applied a combined approachusing BPSM and structural restoration A BPSM is adynamic model of the subbasinrsquos physical processessuch as pore pressure and temperature evolution andthe development of the petroleum systems includinghydrocarbon generation expulsion migration accu-mulation and preservation (Hantschel and Kauerauf2009) The BPSM is performed in 1D (on a well orpseudowell) in 2D (usually along a seismic section)

and in 3D The BPSM can also provide predictionsabout seal capacity and integrity especially if fielddata are available for calibration

In contrast to a static model which consistsonly of geometry and properties for present day(as observed) a dynamic earth model includesthe evolution of geometry and properties Thepaleogeometrymdashor structural geometry of the modelat a given geologic age (or simulation time step)mdashisstrongly dependent on the paleowater depth (geomet-rical boundary condition) faulting activity the lithol-ogy of the various layers and their associatedcompaction behavior Approaching accurate paleoge-ometry is crucial for achieving consistent resultsespecially for fluid flow simulations In standardBPSM the paleogeometries are automatically gener-ated using a simplified restoration approach knownas backstripping (Figure 2A B)

Geomechanically based structural restorationuses the present-day interpreted and depth-convertedgeometry of a sedimentary basin or subbasin for therestoration of the basin structure backward in geo-logic time (Figure 2C D) This is done by sequen-tially removing the syntectonic sedimentary layersfrom the youngest to the oldest (the reverse of deposi-tion) allowing unfolding and unfaulting during therestoration (the reverse of deformation) In contrastto the simple backstripping approach structural resto-ration allows much more control through additionalgeometrical boundary conditions such as lateralmovement of rocks along faults or bedding slipRecent development has resulted in geomechanicalrestoration algorithms extending the pure geometri-cal approach by applying physical laws of rockdeformation (Maerten and Maerten 2006) The paleo-geometries obtained by structural restoration are thenused to describe basin geometry through time in apetroleum systems model

The backstripping approach is applied to tectoni-cally undisturbed areas and to areas where faultingoccurred much earlier than the main phase of hydro-carbon generation (typically a sag basin or the postriftsequence of a passive margin Figure 2A) In basinswhere extensional faulting occurred and where faultslip has an important horizontal component (eg lis-tric faults) backstripped paleogeometries showldquospikyrdquo artifacts (Figure 2B) these paleogeometries

NEUMAIER ET AL 1329

are generally not good enough for predicting prefault-ing hydrocarbon migration but might be acceptablefor hydrocarbon generation modeling Here structur-ally restored paleogeometries are important for thelayer shape in the proximity of faults and whencross-fault connectivity (like sandndashsand overlaps)matters (Figure 2C) Alternatively manual correctionof paleothickness can be done In halokinetic settingsbackstripping is generally superimposed by simplesalt reconstruction workflows (area balancing andextrapolation) Structural restoration should also beconsidered if salt movement results in geometries thatare more complex In any case thermal modeling inthe vicinity of salt is difficult due to its extremelyhigh thermal conductivity In areas with overthrustingor overturned folds (Figure 2D) with multiple strati-graphic sequences typically a fold and thrust beltbackstripping fails completely because lateral rockmovements are missing In the latter case paleo-geometries need to be generated by means other thanbackstripping (from simple conceptual drawing totrue structural restoration)

Paleogeometries generated from structuralrestorations can be loaded into BPSM software as analternative to performing a calculation from back-stripping Meshing is done section per section (timestep per time step) and tracking of cells from one sec-tion to the other is following rules described inHantschel and Kauerauf (2009) For the ages corre-sponding to the paleosection the basin geometry is

fixed it is not a function of compaction-controlledforward modeling as is the case when using back-stripping The BPSM simulation is restricted to thecalculation of properties such as porosity overpres-sure temperature vitrinite reflectance (Ro) andsource rock maturity which populate the fixed paleo-geometries as well as hydrocarbon migration Thesecompaction-controlled results are not used to adjustthe depositional amounts and to correct the paleoge-ometry as done in the backstripping approach andconsequently geometry optimization cannot beperformed The possible difference between thecompaction-law-controlled porosity change and thepredefined geometry change is indirectly compen-sated by the corresponding adjustment of the solidgrain mass

When paleogeometries are used for basin analy-sis and BPSM some factors become very importantRestoration must focus toward a best estimate paleo-water depth because changes in slope directions andtopography can largely affect fluid flow Erosionshould be estimated and reconstructed and manygeologic events need to be taken into account toincrease the resolution through time for dynamicmodeling The sediment thickness must be correctedin geologic time for decompaction All these factorswhich are common sources of uncertainty are criticalfor a geologically meaningful restoration

Because of the optimization method justdescribed adequately decompacted paleogeometries

Figure 2 Steps for (A B) backstripping and (C D) structural restoration of geometries of different structural complexity For detailssee text

1330 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

are required from structural restoration Inadequatedecompaction can produce large errors resulting inan unbalanced mass of solid rock matrix underesti-mated source rock paleoburial depth and deviatedpaleomigration paths related to differential compac-tion However it is difficult to define adequatedecompaction As a rule of thumb in normally pres-sured clastic basins where the main compactiondriver is linearly increasing effective stress simplehydrostatic decompaction laws such as thosedescribed by Athy (1930) can be used for restorationIn cases of abnormal pressure (overpressure) hydro-static decompaction overestimates the effectivestress or underestimates porosity and might fail tothe same degree as restoration neglecting decompac-tion One might be scaling normal decompactioncurves wherever regional overpressure occurs andhas occurred In basins with complex erosion pres-sure and diagenetic histories optimization loops(using BPSM simulated properties for optimizeddecompaction in second restoration) might benecessary Any sort of decompaction in a carbonate-dominated environment where compaction is gener-ally less understood because of the effect of mineraltransformations and cementation is subject to muchlarger uncertainty The question of whether to usehydrostatically decompacted or nondecompactedstructurally restored paleogeometries is discussed inthe section on uncertainty analysis

MODELING

Prior to structural restoration we performed several1D simulations These generated erosion and paleoal-titude estimations for structural restoration andprovided a calibration for the 2D model One-dimensional modeling also gave the first informationon hydrocarbon generation and timing to identifythe period of first source rock maturation (ldquocriticalmomentrdquo as defined by Magoon and Dow 1994)This is very important because prior to structuralrestoration its identification indicates when addi-tional time steps are needed to increase time resolu-tion for critical time intervals so that the activepetroleum system can be dynamically modeled in anadequate way However for simulation regular time

steps are needed not only during the critical momentbut also for the entire basin history

Structural Restoration

The following section describes our structural resto-ration results shown in Figure 3 of the post-riftstructural evolution of the Monagas fold and thrustbelt It can be divided into three main steps(1) From the Mesozoic to Oligocene the Monagasarea formed parts of the northern passive margin ofthe South American shield (Figure 3A) The sub-sidence driver was mostly thermal cooling and iso-static load with only a few normal faults Thesediments were sourced by the craton in the south(2) In the early Miocene the margin experiencedcompression from the transcollision with theCaribbean plate (Figure 3B) Two major normalfaults that formed during rifting were inverted in thecourse of the collision generating the regionalFurrial and Pirital thrusts Synorogenic deposits werecovered by the thrusted allochthon and intramontanepiggyback basins were formed With increasing upliftand erosion (middle Miocene Figure 3C) their sedi-ment fills were redeposited to form the important syn-tectonic Oficina and Carapita Formations south of thethrusts (3) From late Miocene on the late compres-sional phase started Figure 3D shows the present-day structure of the Monagas fold and thrust belt(2D model crossing the Santa Barbara field)According to our restoration the total shortening onthe section accommodated by the thrusts during thecompression is around 40 km (2486 mi) which cor-responds to 35 It is important to note that the Piritalthrust was active until late Miocene as indicated bythe fault displacement analysis (Figure 4) These latetectonic movements might have an important effecton the seal integrity of prospects that are in the vicin-ity of the thrust

Model Input Data

Table 1 shows geologic ages lithologies and petro-physical parameters that were assigned to our 2Dmodel For the main source rocks (Querecual andSan Antonio Formations) a type II B kinetic wasassigned (Pepper and Corvi 1995) As a base case

NEUMAIER ET AL 1331

we assumed a constant initial TOC of 6 wt and aninitial HI of 500 mg HCg TOC However becauseinitial TOC and HI were estimated from present-daymeasured values these initial values are highly uncer-tain In addition there is both lateral and vertical vari-ability within the layer even though locally very highsource rock potential is likely to occur a bulk TOCand HI need to be assigned to the model cells Forthese reasons several models with different initialTOC have been run and their sensitivity to the mod-eled charge is discussed below Faults were set open

to fluid flow in the course of their main kinematicactivity which is estimated by fault displacementanalysis of the structural restoration models Aftertheir main activity phase faults are assumed to beclosed Fault permeability was also a varying param-eter in the sensitivity analysis discussed belowThermal boundary conditions are given in Table 2Surface temperature ranges between 23 and 30degC(734 and 86degF) through geologic time with apresent-day average temperature of 28degC (824degF)These temperatures at sea level have been corrected

Figure 3 Structural evolution of the Monagas fold and thrust belt demonstrated by structural restoration (A) Passive margin stage(until 23 Ma) (B) Early compression stage (21 Ma) (C) Late compression stage (12 Ma) (D) Present day For details see text

1332 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

Figure 4 Fault displacement from lateMiocene to present day for all faults onrestored cross sections (A) and for the Piritalthrust only (B) Black and white shading andsize of points show magnitude of displacementin meters

NEUMAIER ET AL 1333

Table1

Basin

andPetro

leum

Syste

mModeling(BPSM)Input

ParametersGeologicAges

andLithologies

Compressib

ility

(GPa

-1)

Perm

eability

(logmD)

Thermal

conductivity

(Wm

K)

Heatcapacity

(kcalkgK)

Age

(Ma)

GeologicTime

Facies

Density

(kg∕m

3 )Initial

Porosity(

)max

min

at1

por

at25por

at20degC

at100degC

at20degC

at100degC

0Pleisto

cene

LasP

iedras

sandsto

ne2720

412747

115

minus18

3395

338

02

024

LasPiedrasshale

2700

4040327

403

minus852

minus3

164

169

021

024

26

Pliocene

LasP

iedras

sandsto

ne2720

412747

115

minus18

3395

338

02

024

LasPiedrasshale

2700

4040327

403

minus852

minus3

164

169

021

024

53

upperM

iocene

2La

Pica

sandsto

ne2720

412747

115

minus18

3395

338

02

024

LaPica

shale

2700

4040327

403

minus852

minus3

164

169

021

024

Morich

itoshale

2700

4040327

403

minus852

minus3

164

169

021

024

8upperM

iocene

1La

Pica

sandsto

ne2720

412747

115

minus18

3395

338

02

024

LaPica

shale

2700

4040327

403

minus852

minus3

164

169

021

024

Morich

itobasal

sandsto

ne2715

4812142

187

minus348

15

317

284

02

024

105

middleMiocene

2Carapitasandsto

ne2715

4812142

187

minus348

15

317

284

02

024

Carapitashale

2705

6330932

331

minus684

minus15

204

201

02

024

12middleMiocene

1Carapitashale

2705

6330932

331

minus684

minus15

204

201

02

024

15lower

Miocene

2Carapitashale

2705

6330932

331

minus684

minus15

204

201

02

024

19lower

Miocene

1lower

Carapitashale

2700

7040327

403

minus852

minus3

164

169

021

024

23Oligocene2

Merecureplatform

sandsto

ne2715

4812142

187

minus348

15

317

284

02

024

27Oligocene1

Merecureplatform

sandsto

ne2715

4812142

187

minus348

15

317

284

02

024

1334 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

339

PaleoEocene

Vidono

sandsto

ne2720

412747

115

minus18

3395

338

02

024

Caratasbasin

alshale

2700

7040327

403

minus852

minus3

164

169

021

024

655

upperC

retaceous4

SanJuan

sandsto

ne2720

412747

115

minus18

3395

338

02

024

74upperC

retaceous3

SanAntoniosiltstone

(SOUR

CERO

CK)

2710

5510422

212

minus628

minus1

201

196

023

026

825

upperC

retaceous2

SanAntoniosiltstone

(SOUR

CERO

CK)

2710

5510422

212

minus628

minus1

201

196

023

026

91upperC

retaceous1

Querecualmudsto

ne(SOUR

CERO

CK)

2700

7040327

403

minus852

minus3

164

169

021

024

996

lower

Cretaceous

2undifflower

Cretaceous

(30

Lst50

Sst

20Sh)

2722

459445

141

minus334

174

305

275

02

023

122

lower

Cretaceous

1undifflower

Cretaceous

(30

Lst50

Sst

20Sh)

2722

459445

141

minus334

174

305

275

02

023

1455

sub-Cretaceous

2undiffsub-Cretaceous

(50

Sh50Sst)

2710

5521537

259

minus516

0255

239

02

024

251

NEUMAIER ET AL 1335

by the influence of the paleowater depth using theapproach of Wygrala (1989) Because of the hightopographic elevation of approximately 3500 m(11500 ft) the maximum altitude during the pastcompressional phase we adjusted the model takinginto account a temperature decrease of 6degCkm (33degF1000 ft) related to the paleoaltitude This resultsin a difference of up to 20degC (36degF) compared to sea-level temperature We then assigned the adjustedsedimentwater interface temperatures to the modelas the upper thermal boundary condition At the baseof the model we assigned a heat flow ranging from42 to 80 mW∕m2 through geologic time The highestheat flow is assumed during the rifting event beforethe basin cooled as a passive margin A thermal eventin the Eocene detected by fission track analysis(Locke and Garver 2005 Perez de Armas 2005)has been addressed by a second peak before a secondcooling phase For uncertainty analysis basal paleo-heat flow has been varied within geological meaning-ful ranges and the thermal calibration data has beenhonored as much as possible

Modeling Results

Basin and petroleum system modeling can be dividedinto simulation of the physical parameters within abasin such as pressure and temperature (basin model-ing) and the subsequent evolution of the source rocksand their derived hydrocarbon compounds (petro-leum systems modeling) In the following sectionswe use this subdivision to describe the BPSM resultscompaction modeling thermal modeling and chargehistory It is important to state that these are predic-tions based on forward modeling as just describedThe results are compared and calibrated to actualmeasurements and observations where available

Compaction Modeling

Figure 5 presents an extraction of the modeled resultsat the location of Well Alpha in the Furrial Trend Inthe absence of direct calibration data for pore pres-sure we undertook a sensitivity analysis to test thedifferent scenarios Using the model we tested howpore pressure in the main reservoir of the FurrialTrend reacts to changes in fault and seal-rock per-meability In a first step fault permeability has beenevaluated Faults were assumed open during the mainfault activity phase and closed afterward In additionwe performed several variations from that base casescenario Figure 5A shows the effect of two end-member scenarios on pore pressure within the mainreservoir of the Furrial Trend (Santa Barbara field) inwhich faults have been modeled to be either open orclosed since their formation A third scenario consistsof thrust faults that are open during the compressionalphase (from 15 Ma on) and closed afterward (from105 Ma on) This scenario has been chosen as theldquomaster scenariordquo because it results in overpressureof 23 MPa which is in the range of pressure reportedin similar structures of the Furrial Trend by Schneideret al (2004) In addition this pressure scenariomatches the calibration data for porosity (Figure 5B)

The 2D distribution of modeled overpressure atpresent day is shown in Figure 6A Apart fromimportant overpressure in the Querecual Formationthe allochthon is almost normally pressured Thethrust wedge is overpressured especially in thelow-permeability Carapita Formation Although

Table 2 Basin and Petroleum System Modeling (BPSM) InputParameters Thermal Boundary Conditions

Age (Ma)

SedimentWaterInterface

Temperature (degC)

Age (Ma)

Basal HeatFlow

(mW∕m2)min max

0 28 28 0 4226 28 28 30 6053 76 23 34 708 184 231 50 50105 62 233 107 6012 31 238 1455 8015 5 242 260 6019 5 24723 234 2527 78 235339 94 236655 146 2831455 148 287255 5 239

1336 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

the autochthon is highly overpressured close tonormal hydrostatic pressures are simulated in theforeland basin The discussed model assumption ofclosed faults in the postcompressional phase resultsin a regional pressure compartmentalization

The simulated pressure evolution through geo-logic time can be seen in Figure 7A A time extractionof a source rock cell in the thrust wedge shows a sud-den increase of both lithostatic and hydrostatic pres-sures that is a consequence of the tectonic burial

load associated with the thrusting In addition themodeled pore pressure buildup even though delayedin time can be directly linked to the thrust event Itcan be clearly seen that the timing of fault permeabil-ity directly impacts changes in the present-day porepressure The earlier the faults close the more over-pressure generated by the massive syntectonic sedi-mentation is kept until present day given adequatesealing lithologies are present The present-day over-pressure is modeled as14 MPa when the faults are

Figure 5 Plots of one-dimensional model output and available calibration data for Well Alpha (Table 3) (A) Dotted lines are hydro-static and lithostatic pressures continuous lines are pore pressure the master scenario is in black (a1 faults closing at 105 Ma) andadditional scenarios in gray (a2 open faults a3 closed faults) (B) modeled porosity (C) borehole temperature and (D) Ro The blackline shows the master scenario and the gray line the alternative high-heat-flow scenario Bold lines in the stratigraphic column arethrusts (T) and unconformities (U)

Table 3 Calibration Data for Well Alpha (Porosity Temperature Ro)

TVD (m) Porosity () TVD (m) Corrected Temperature (degC) TVD (m) Vitrinite Reflectance (Ro)

4350 15 600 40 400 0264400 15 1100 53 1200 0484800 10 1800 68 1485 068

4400 131 1700 0394800 133 2900 048

3100 0493400 0494000 054300 0524800 06

NEUMAIER ET AL 1337

open since 8 Ma as 23 MPa for 105 Ma and as37 MPa for 12 Ma Compared to this order of magni-tude changes in the permeability of the overlyingCarapita shale are negligible for the pore pressureTherefore the controlling parameter for pore pressurein the main reservoir of the Furrial trend is assumedfault permeability and resulting lateral connectivityrather than the permeability of the seal rock

Thermal Modeling

Modeled temperature and thermal maturity(Figure 5C D) have been calibrated with availablewell data Some discontinuities in the modeled Ro

trend are noted (Figure 5D) The upper kinks (egbase Pliocene) are related to unconformities withassociated erosion and the lower one (base UpperCretaceous) is due to the well penetrating a reverse

Figure 6 Two-dimensional model at present day with (A) modeled overpressure and (B) modeled thermal maturity and isothermsSource rocks marked by lithology pattern

1338 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

fault displacing mature rocks upon less mature ones(see Figure 6B for 2D view) There is a mismatchbetween the modeled Ro trend and the data in thedeeper part of the well In the Oligocene rocks themodeled temperature is too low (2degC or 36degF)whereas the modeled Ro is too high (up to 02 Ro)Any attempt to reduce the modeled Ro (eg bychanging thermal conductivity of the MioceneCarapita Formation) also results in a lower modeledpresent-day temperature Therefore the thermal sce-nario used for this calibration is a trade-off betweenmatching present-day temperature and Ro (using thekinetics of Sweeney and Burnham 1990) The basalheat flow assumed in the master scenario (Table 2)results in a modeled present-day surface heat flowranging from 37 mW∕m2 in the southern part of thethrusts to 52 mW∕m2 in parts of the uplifted areaAt the same depth the allochthonous northern partis thermally more mature than the southern autochthon(Figure 6B) because of the recent uplift Locallyinverted trends around the thrusts can be observedfor example at the location of Well Alpha (seeFigure 5D) In the autochthon the Upper Cretaceoussource rocks are entirely in the oil or even gas windowat present day because of the Neogene tectonic loadingand synorogenic deposition In the northern alloch-thon most of the Upper Cretaceous source rocks havebeen in the oil window in the past The modeled ther-mal maturity is in good agreement with the regionalmaturity suggested by Summa et al (2003)

Similar to the pressures the thrusting also affectsthe modeled temperature and Ro In the thrust wedgea direct relationship exists between the thrusting andthe increase of temperature and maturity The thrust-ing is the trigger for source rocks entering the oil win-dow within the thrust wedge (Figure 7B) and thepresent-day thermal maturity has been reached sincethe last 10 Ma Although this recent thermal historycan be considered as relatively well constrained bythe measured Ro in the calibration well this doesnot apply to the past In this thrusted area any ther-mal maturity reached prior to the thrusting onset hasbeen ldquooverprintedrdquo and therefore no thermal calibra-tion is possible for this period However theCretaceous source rocks were not buried deeply dur-ing the passive margin phase and even very hot sce-narios (basal heat flow of 100 mW∕m2) do not

significantly change their maturity (Figure 7B) Inthe uplifted allochthon several wells show low(03) to moderate (1) measured Ro (Parra et al2011) Because the allochthon experienced its maxi-mum burial and the highest temperatures during thepassive margin phase these maturity levels can beassumed as representative of the maturity prior tothe compression In contrast to the thrust wedge theconsequence of the thrusting is inverse for the alloch-thon (Figure 7C) as the hanging wall is uplifted andthe accompanying erosion removes the overburdenThe highest temperatures are simulated to have beenreached just before the thrusting event Then temper-ature decreased and Ro stagnated leaving the sourcerocks in the early mature zone until present day Theassumption of a regionally relatively homogeneousthermal field during the cooling phase of the passivemargin in combination with the relatively low matur-ities measured in the uplifted allochthon makes thehot heat-flow scenario unlikely However a laterallyvarying heat flow cannot be excluded and the conse-quently hotter paleoheat flow would have led to muchhigher temperatures prior to compression resultingalso in a higher present-day maturity level (14)despite the uplift (Figure 7C) Any variations of basalheat flow during the rift period (Late Jurassic to EarlyCretaceous) occurred too early to affect the maturityof the source rocks deposited during the lateCretaceous

Charge History

After calibration of the basinrsquos physical environ-ment the evolution of the petroleum systems wasmodeled Hydrocarbon generation was simulatedthrough geologic time using the activation energydistribution of the assigned source rock kineticsFrom the moment of hydrocarbon expulsiononward secondary migration was modeled usingthe invasion percolation method (Wilkinson 1984)It takes into account the geometry overpressure gra-dient and capillary entry pressures of the rockswhich are opposed to the buoyancy of the migratingoil and gas Figure 8 shows the time evolution of thetotal study area and Figure 9 focuses on the thrustwedge As already mentioned the described chargehistory is based on forward modeling It is consistent

NEUMAIER ET AL 1339

Figure 7 Time extractions (see Figure 6 for location) (A) Pressure development through geologic time for a source rock in the ver-tical continuity of Well Alpha Dotted lines are hydrostatic and lithostatic pressures continuous lines are pore pressure the master sce-nario is in black (a1 faults closing at 105 Ma) and additional scenarios in gray (a2 open faults a3 closed faults a4 faults closing at12 Ma a5 faults closing at 8 Ma) (B) Temperature (continuous lines) and Ro (dotted lines) through geologic time for a source rockin the vertical continuity of Well Alpha the black line shows the master scenario and the gray line the alternative high-heat-flow sce-nario (C) Temperature (continuous lines) and Ro (dotted lines) through geologic time for a source rock in the uplifted allochthonthe black line shows the master scenario and the gray line the alternative high heat flow scenario

1340 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

with the modeled compaction and thermal historyand therefore shows one scenario that explains thecharge of the Santa Barbara field and the forelandbasin Simplifications are inherent to any such work-flow and shall be considered when analyzing themodeling results

Passive Margin PhasemdashMesozoic toOligocene

Modeling results indicate that already during the LateCretaceous any Jurassic source rocks potentiallydeposited in the northern part of the section (Summaet al 2003) have reached temperatures of more than200degC (392degF) and corresponding high levels of matu-rity However the temperature at the base of theQuerecual Formation (Upper Cretaceous) was lessthan 100degC (212degF) leaving the kerogen immatureThe source rock became effective only during lateEocene with temperatures of up to 120degC (248degF)and a (reactive kerogen to hydrocarbon) transforma-tion ratio (TR) of more than 20 in the deeper distalpart of the passive margin Oil was quickly expelledmigrated vertically and seeped to the surface becauseof the general lack of seals in the overburden Duringthe Oligocene the time of deposition of the Merecurereservoir sandstones in the southern platform this sit-uation did not change apart from increased transforma-tion in the deeper basinal part of the passive margin

First oil accumulations are modeled in the Vidontildeosandstones in the basinal part in the late Oligocene(Figure 8A) in which the base of the QuerecualFormation shows TRs of up to 60 at around 130degC(266degF) In the proximal part there is no hydrocarbongeneration Although there is a general northwardtilting slope no southward migration from the distalmargin is modeled this has been checked usingdifferent migration methods including Darcy flowwith scenarios of continuous intrasource carrier bedsThis is in contrast to the studies from Gallango andParnaud (1995) and Schneider (2003) who suggestedlateral migration at that stage The difference couldbe explained by the presence of a regional top seal intheir models However our model suggests that theshallow rocks do not have sufficient sealing capacityto allow lateral hydrocarbon migration During thepassive-margin phase the basin was normally

pressured (Figure 7A) and simulated stresses havenot yet resulted in any fracturing at that stage

Main Compressional PhasemdashEarly to MiddleMiocene

With the onset of compression during early Miocene(Figure 8B) both the Pirital and Furrial thrusts wereformed and the northern distal part of the passivemargin was rapidly uplifted Important overpressuresas well as mechanical rock failure both the result ofthe rapid uplift are simulated in the allochthonDespite this uplift the kerogen-to-hydrocarbon trans-formation continued and the rate even increasedlocally close to the footwall cutoffs and where syn-orogenic deposition provided rapid overburden load-ing Oil accumulations are modeled in the freshlycreated anticlines

The tectonicndashsedimentary wedge between thetwo main thrusts experienced very rapid burial Infact the Upper Cretaceous source rocks were broughtinto depths where temperatures were as high as185degC (365degF) at 19 Ma and 225degC (437degF) at15 Ma Especially close to the cutoff of the Piritalthrust they very quickly reached a high maturity at19 Ma and reached full transformation in most placesat 12 Ma The expelled hydrocarbons migrated intothe sandstones of the Merecure Formation which weresealed by the Carapita Formation The charge toward aseries of recent anticlines is modeled by the fill-to-spillmechanism from north to south (Figure 9C D) Afterthe first oil fill the northern prospect quickly receiveda local gas charge the central structure containedmainly oil Note the presence of a lateral seal that isdue to sandndashshale juxtaposition The southernmoststructure which forms the Santa Barbara field todaywas charged by oil only at 15 Ma (Figure 9D)

The generated hydrocarbons in the southernautochthon were leaking to the surface until theCarapita Formation was compacted enough to holdsome of them (Figure 8C) At 12 Ma the wedgesource rocks were mostly spent but expulsion wasstill ongoing Important rock failure is modeled tohave occurred in the thrusted footwall breaking theformerly sealing Carapita Formation (Figure 10)This is in good agreement with cemented hydraulicfractures observed in a few cores in the Carapita

NEUMAIER ET AL 1341

Figure 8 Two-dimensional model through geologic time (selected time steps only) with modeled transformation ratio for the sourcerock layers isotherms and liquid and vapor migration and accumulation

1342 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

shales from the Furrial Trend (Roure et al 2005)This important fracturing event resulted in thedestruction of the modeled accumulation (Figure 9E)Hydrocarbon generation and expulsion were alsohappening in the southern foreland basin at transfor-mation ratios of up to 30

Late Compressional and PostcompressionalPhasemdashLate Miocene to Present Day

The late compressional phase was marked by the par-tial destruction of the allochthonous petroleum systemsby erosion (Figure 8E F) Many traps disappeared and

Figure 8 Continued

NEUMAIER ET AL 1343

the hydrocarbon generation within the source rocksstopped The erosion triggered an important sedimenttransfer toward the south contributing to further burialand increased temperatures in the thrust wedge and theforeland basin During late Miocene the petroleumsystem in the wedge matured An important northwardtilting was triggered by flexural tectonic loading andthe loading of syntectonic sediments (Summa et al2003) In combination with the presence of an effec-tive regional top seal the compacted CarapitaFormation allowed a change from dominantly verticalmigration into lateral migration First accumulationsare modeled in the Neogene synorogenic sequence

reservoirs in the south and important southwardmigration toward the foreland basin started(Figure 8G) These trends continued through Plioceneand Pleistocene times to present day filling theOnado field (Figure 8H) and other known fields inthe south (Figure 1A) of the Eastern Venezuela Basin

Prospectivity

We modeled the charge of the Santa Barbara field inthe Furrial Trend and of the Onado field in the fore-land basin (Figure 8H) Other prospects in the thrustwedge were also investigated The structure called

Figure 9 Detail of two-dimensionalmodel through geologic time (seeFigure 8) showing modeled transforma-tion ratio for the source rock layers iso-therms and liquid and vapor migrationand accumulation The Santa Barbara fieldand the prospect are shown by ellipses

1344 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

ldquoprospectrdquo (Figure 9G) looks similar to that of theSanta Barbara field because both are anticlines andthe reservoir and seal formations are the sameHowever they differ in several aspects First theprospect is in the vicinity of the Pirital thrust whichpresents a high risk of impacting the seal integrityThe prospect location is directly below the area wherewe presume the maximum of late tectonic activityoccurred (Figure 4) As modeled the CarapitaFormation the seal of the prospect was already brit-tle when this deformation occurred Calculatedstresses also predict rock failure from 12 Ma onward(Figure 10) Even though charge continued ourmodel suggests continuous leakage of the structure(Figure 9EndashG) The second major difference is thatthe source rock in the fault block of the prospect ismodeled to have already started generating and expel-ling hydrocarbons during the Oligocene and earlyMiocene (Figure 9A) before the trap formed and the

seal became effective (Figures 9C 10) Even thoughgeneration expulsion and migration continued aftertrap formation the prospect was not charged withthe earlier expelled oil In addition the present-daymaturity of the source rocks within the prospectrsquosfault block is higher than in the source rocks withinthe tectonic block of the Santa Barbara field poten-tially indicating expulsion of more gas A southndashnorth trend can be seen on the modeled accumulationcomposition with more gas toward the allochthon(Figure 9) The modeled composition of the prospectprior to seal failure was much richer in gas than inthe neighboring Santa Barbara field (Figure 9D) Infact simulation scenarios with unbreached sealsresulted in accumulation of predominantly gas Thecombination of both seal integrity and charge compo-sition concerns makes this a risky prospect In addi-tion potential diagenetic fluids related to the faultfluid flow might have altered the reservoir quality

Figure 10 Two-dimensional model (at 12 Ma) with modeled stress to failure Mohr circles (with σ standing for normal stress and τfor shear stress) and petroleum system events charts for the Santa Barbara field (left) and the prospect (right) Model predicts rock fail-ure in the prospect from about 12 Ma For further details see text

NEUMAIER ET AL 1345

Apart from the Onado oil field only minor accu-mulations are modeled in the southern Neogene syn-orogenic sequence The proportion of gas is higherthan in the Furrial trend The accumulations at thesouthern model edge are a modeling artifact repre-senting hydrocarbons (mainly oil) that in reality arethought to have charged the southern foreland basin(from mid to late Miocene times to present day)However parts of the migrated oil and gas could havebeen trapped in the Upper Cretaceous San JuanFormation south of the horst structure

The potential for conventional hydrocarbon accu-mulations in the allochthon is very limited by theimportant fracturing that occurred during the upliftperiod The basal sandstones of the Morochito piggy-back basin north of the Pirital thrust are not chargedin our 2D model because of the absence of structuraltraps Nevertheless there is a potential for strati-graphic traps to have been charged as migrationthrough the basin occurred However seal strengthmight be critical at this low-compaction stage Thesame applies for the Las Piedras Formation We thinkthere might be potential for shale oil because theQuerecual Formation is at shallow depths but showsmaturity within the oil window From the thermalmaturity point of view any potential Jurassic sourcerocks as discussed by Summa et al (2003) mightbe a gas source or shale gas target

UNCERTAINTY ANALYSIS

There is often significant uncertainty in basin geom-etry and depth of stratigraphic horizons based onuncertain seismic interpretation and velocity models(Kukla et al 2000 Brandes et al 2008 Baur et al2010) Restoring of present-day geometries to obtainpaleogeometries adds further uncertainty Additionaluncertainty is related to thermal boundary conditionssuch as paleoheat flow (Beha et al 2008) and togeochemical properties such as source rock character-istics and petroleum generation kinetics and calcula-tion of biodegradation and petroleum density(Blumenstein et al 2008) In addition to the porepressure and thermal sensitivity analysis justdescribed we varied initial TOC within a range of2 to 10 A variation of initial TOC obviously

directly scales the amount of initial kerogen availablefor transformation into hydrocarbons Neither thecharge volume nor composition of the Santa Barbarafield nor the wedge prospects are significantly sensi-tive to variation of initial TOC these traps are filledto spill in each scenario However the amount ofhydrocarbons charging the foreland basin (eg theOnado field) is reduced in the case of low initialTOC A high initial TOC only increased the modeledamount of surface seepage The sensitivity of the ini-tial HI is analogous to the sensitivity of the ini-tial TOC

In addition to the more classical uncertaintyand sensitivity analysis described we focused onquantitative analysis of the decompaction uncer-tainty when performing structural restoration Thisuncertainty is particular to the applied workflowof using structurally restored paleogeometries inBPSM Note that this uncertainty is significantlylower when using the backstripping method withthe optimization algorithms discussed One methodto analyze the adequacy of the decompactionmethod is comparison of the output rock matrixmass balance In the backstripping approach therock matrix mass is kept constant of a model cellwith a simulated decrease in porosity controlledby compaction laws and the corresponding reduc-tion of the layer thickness When using fixed pale-ogeometries from structural restoration the cellthickness is imposed However the decrease inporosity is calculated and the mass is adapted tofit the imposed cell thickness That means that therock matrix mass is constant only if ideal decom-paction has been applied

Figure 11A shows the percentage variation fromthe present-day matrix mass (100) of a given sourcerock cell through geologic time The dotted line rep-resents the ideal case with constant rock matrix massWe used two end-member decompaction scenarios todescribe the decompaction-related uncertainty that isintroduced when fixed paleogeometries replace back-stripping idealized hydrostatic decompaction assum-ing no overpressure and idealized lithostaticdecompaction assuming no effective stress (nodecompaction resulting in constant cell thicknessthrough time) In the latter case with decreasingporosity during compaction the rock matrix mass

1346 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

increases until it reaches the present-day mass Inother words rock mass matrix has been highly under-estimated in the past (in this particular case by morethan 20) This is a source rock so the kerogen massdecreases by the same factor This yields an underes-timation of more than 20 in TOC Note the rockmatrix mass is constant for time steps prior to23 Ma in which backstripping optimization couldbe performed toward the earliest paleogeometry Forthe second end-member scenario of idealized hydro-static decompaction the rock matrix mass is nearlyconstant which is a good decompaction adequacy

indicator for that given cell in the model Howeverthe so-measured adequacy changes throughout themodel In some areas the rock matrix mass has beenunderestimated by about 50 for nondecompactedpaleogeometries however in the case of hydrostaticdecompaction values generally range from 80 to120 of the present-day rock matrix mass Thatmeans in this particular case that hydrostatic decom-paction is more adequate (or less wrong) than nodecompaction at all The decompaction adequacycan be roughly correlated with pressure changesthrough geologic time

Figure 11 Time extractions for comparison of two end-member decompaction scenarios (see Figure 6 for location) (A) Rock matrixmass conservation through geologic time (B) Source rock burial depth through geologic time (C) Temperature and transformation ratiothrough geologic time

NEUMAIER ET AL 1347

Inadequate decompaction results in wrong depthfor stratigraphic horizons in the paleogeometriesFigure 11B shows the burial depth of the same inves-tigated source rock cell We can see that during thepassive margin phase there is a slight difference inpaleoburial depth for the two end-member scenariosDuring the compressional phase the differencechanges dramatically reaching more than 1000 m(3281 ft) in some places Toward present day bothcurves meet again because the input to the two resto-ration scenarios was the same (identical present-dayburial depth) This difference in compaction calcula-tion has severe implications on the modeled tempera-ture history and the kerogen-to-hydrocarbontransformation (Figure 11C) For an identical thermalframework in terms of modeling input parameterstemperature differences of up to 25degC (45degF) resultin TR differences of more than 20 in the early com-pressional phase before TR values meet again for thepresent-day situation

CONCLUSIONS

Structurally complex areas are attracting increasinginterest in hydrocarbon exploration Exploration riskcan be substantially narrowed by analyzing the his-tory of the basin and its petroleum systems based onstructural evolution We presented a further progressin technology that allows the integration of structuralrestoration and BPSM This combined analysis isseen as a way forward to improve modeling in thrustbelts with stacked stratigraphic sequences Despitelimitations of nonoptimized decompaction whichare added to the general uncertainty related to theindividual methods we modeled the charge and sealhistory of the Monagas fold and thrust belt inVenezuela Calibrated porosity temperature andthermal maturity as well as basin-scale stress pro-vided the regional framework through geologic timein which the petroleum systems evolved such thatthe charge of the known Santa Barbara and Onadofields and other trends could be understood We dis-cussed seal integrity of a potential exploration pros-pect by analyzing structural restoration-derived faultdisplacement through geologic time and an advancedstress forward simulation The uncertainty of using

decompacted or nondecompacted structural restora-tion has been assessed It can have important influ-ence on rock matrix mass balance through geologictime paleogeometry and paleoburial depth andtherefore affect paleotemperatures and paleomatur-ities We conclude that decompaction assuminghydrostatic pressure is the best option in that particu-lar case

REFERENCES CITED

Alberdi M and E Lafargue 1993 Vertical variations of organicmatter content in Guayuta Group (Upper Cretaceous)interior mountain belt eastern Venezuela OrganicGeochemistry v 20 p 425ndash436 doi 1010160146-6380(93)90091-O

Athy L F 1930 Density porosity and compaction of sedimen-tary rocks AAPG Bulletin v 14 p 1ndash24

Aymard R L Pimentel P Eitz P Lopez A ChaouchJ Navarro J Mijares and J G Pereira 1990 Geologic inte-gration and evaluation of Northern Monagas EasternVenezuelan Basin in J Brooks ed Classic petroleum prov-inces The Geologic Society (London) Special Publicationv 50 p 37ndash53

Baur F M Di Benedetto T Fuchs C Lampe andS Sciamanna 2009 Integrating structural geology andpetroleum systems modelingmdashA pilot project from Boliviarsquosfold and thrust belt Marine and Petroleum Geology v 26no 4 p 573ndash579 doi 101016jmarpetgeo200901004

Baur F R Littke H Wielens C Lampe and T Fuchs 2010Basin modeling meets rift analysismdashA numerical modelingstudy from the Jeanne drsquoArc basin offshore NewfoundlandCanada Marine and Petroleum Geology v 27 p 585ndash599doi 101016jmarpetgeo200906003

Beha A R O Thomsen and R Littke 2008 Thermal historyhydrocarbon generation and migration in the Horn Grabenin the Danish North Sea A 2D basin modelling studyInternational Journal of Earth Sciences v 97 p 1087ndash1100doi 101007s00531-007-0247-2

Blumenstein I O B M Krooss R di Primio W RottkeE Muumlller C Westerlage and R Littke 2008Biodegradation in numerical basin modelling A case studyfrom the Gifhorn Trough N-Germany International Journalof Earth Sciences v 97 p 1115ndash1129 doi 101007s00531-007-0272-1

Brandes C A Astorga R Littke and J Winsemann 2008Basin modelling of the Limoacuten back-arc basin (Costa Rica)Burial history and temperature evolution of an island arc-related basin-system Basin Research v 20 p 119ndash142doi 101111bre200820issue-1

Carnevali J O 1988 El Furrial oil field northeasternVenezuela First giant in foreland fold and thrust belts ofwestern hemisphere AAPG Bulletin v 72 68 p

Chevalier Y 1987 Contribution a lrsquoetude geacuteologique de lafrontiereacute Sud-Est de la plaque Caraibe Les zones internesde la chaine Sud-Caraibe sur le transect ile de Margarita-

1348 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

as the effects of thrusting on pressure and hydrocarbon generationdynamic structure-related migration pathways and the generalimpact of deformation are described Additionally we evaluatedseal integrity by analyzing simulated stresses through geo-logic time

GEOLOGIC SETTING

The Eastern Venezuela Basin (Figure 1A) is bounded by theSerrania del Interior and the El Pilar fault to the north by a dextralstrike-slip fault system at the interface of the Caribbean and SouthAmerican tectonic plates and by the Orinoco River and thePrecambrian Guyana shield to the south (Hedberg 1950Passalacqua et al 1995 Pindell and Tabbutt 1995) The northernparts of the basin the Monagas fold and thrust belt override a lessdeformed foreland part of the basin to the south

The basin initially formed as a passive margin on the SouthAmerican shield following late Jurassic to early Cretaceous rift-ing During the transcollision of the Caribbean plate with theSouth American plate in the late Cenozoic subsidence becameflexural and extensive folding and thrusting occurred in thenorthern part of the basin (Figure 1B) Roure et al (2003)described both the geodynamic history and the stratigraphy(Figure 1C) Cenozoic platformal and basinal facies of the passivemargin overlie the Mesozoic sediments which include the preriftand synrift sequences The most recent sediments belong to thesynorogenic north-sourced depositional sequence The principalsource rocks in the area are within the Querecual and SanAntonio Formations of the Guayuta Group (Talukdar et al1988 Gallango et al 1992 Alberdi and Lafargue 1993) Thesemudstones are characterized by type II kerogen with 2 to 6 wttotal organic carbon (TOC) for partly highly mature samplesHydrogen index (HI) values are as high as 500 mg HCg TOCSumma et al (2003) suggested an initial TOC as high as 12 forpresent-day values of up to 8 with an HI as high as 700 mgHCg TOC Minor source rocks exist in the Carapita Formationparticularly in the basal part with mixed type IIIII kerogenSumma et al (2003) estimated TOC up to 45 and an HI of350 mg HCg TOC They also discussed the presence of Jurassicand Albian source rocks in the Serrania del Interior

Surface oil seeps and asphalt lakes are very common in thisarea and hydrocarbons have been found since the early 20th cen-tury (Young 1978 Carnevali 1988 Krause and James 1989Aymard et al 1990 James 1990 2000a 2000b Erlich andBarrett 1992 Prieto and Valdes 1992) In the south heavy oiland tar sands (Oronico Oil Belt see Figure 1A) and conventionaloil accumulations are abundant In the thrust area the Furrial trend

Peter Kukla is a professor of geology anddirector of the Geological Institute at RWTHAachen University since 2000 He workedfor Shell International in the Netherlands(1991ndash1996) and in Australia (seconded toWoodside Energy in Perth 1996ndash2000)Peter graduated with degrees in geologyfrom Wuerzburg University Germany(MSc 1986) and Witwatersrand UniversitySouth Africa (PhD 1991) His researchinterests include basin analysis clastic andcarbonate reservoir characterizationunconventional gas geodynamics of saltbasins and pore pressures

ACKNOWLEDGMENTS

We thank Francois Roure (Institut Francaisdu Petrole) Sandro Serra (SerraGeoConsulting) and an anonymousreviewer for their comments andsuggestions Ian Bryant Martin RohdeAdrian Kleine Alan Monahan Will StapleyArmin Kauerauf and Thomas Fuchs(Schlumberger) also provided very helpfultechnical contribution discussions andinternal reviews

The AAPG Editor thanks the followingreviewers for their work on this paperFrancois M Roure Sandro Serra Carl KSteffensen and an anonymous reviewer fortheir work on this paper

NEUMAIER ET AL 1327

contains some of the worldrsquos largest oil reserves forbasins with this type of tectonics Many shallow oilfields in the Eastern Venezuela Basin are located inNeogene sandstone reservoir rocks (Las Piedras LaPica Morichito Chapapotal Freites and Oficinafields) Other reservoir formations are the synoro-genic Miocene Naricual Formation and theOligocene Merecure Formation which are part of aproximal passive-margin sequence that shales outtoward the north The latter is the main reservoir inthe thrusted Furrial trend In the deeper passive-margin sequence there are potential reservoirs in theMesozoic (eg San Juan Formation) Because of its

shaly lithology large thickness and lateral continu-ity the synorogenic Carapita Formation is probablythe most effective seal However several other for-mations have good sealing capacities even thoughnot well compacted as numerous fields at shallowdepths prove

The first publications on the thermal history ofthe Eastern Venezuela Basin were based on one-dimensional (1D) models Because of the com-plexities leading to multiple sequences caused byoverthrusting modeling was done either in theunthrusted area or if within the thrusted area onlyfor geologic time intervals preceding thrusting

Figure 1 (A) Map of the Eastern Venezuela Basin showing structural elements and hydrocarbon occurrence (modified after Parnaudet al 1995) Solid line shows location of two-dimensional (2D) line modeled in this study (B) Generalized northndashsouth geodynamiccross section of the study area (modified after Chevalier 1987) (C) Synthetic stratigraphy of the southern foreland basin of theEastern Venezuela Basin (modified after Roure et al 2003)

1328 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

(Talukdar et al 1988) Summa et al (2003) pub-lished multiple 1D models over a large area andextrapolated the results onto regional maturity mapsfor the Guayuta Group and Cenozoic source rocksThese maps include a hanging-wall cutoff and there-fore do not represent the higher maturities found inthe thrust wedges Elsewhere the authors suggestedthat most source rocks have been in the oil windowin the past A present-day drainage area analysis ofthe top Cretaceous structure map was used to describethe southward charge from the deeper kitchen into theforeland basin This regional migration modelexplains oil provenance and quality Important timingaspects such as the relationship between the struc-tural history and source rock maturation and expul-sion are described using a regional structuralrestoration in conjunction with the 1D models Parraet al (2011) included the thrusting effect by using a15D module mimicking lateral rock transportationthat is due to thrusting Gallango and Parnaud(1995) performed 2D modeling including hydrocar-bon migration in the unthrusted southern area Theyalso described a second model within the thrustedarea that simulates the basin history prior to compres-sion (from the Mesozoic to early Miocene) Thismodel proposes long-distance petroleum migrationsouthward before the transpressional event whichresulted in the observed structural complexity Inother papers Roure et al (2003) Schneider (2003)and Schneider et al (2004) described fluid flow porefluid history and diagenetic processes in a 2D modelalong a cross section parallel to the location of themodel presented here

MODELING METHODS

To address the charge and seal history of the SantaBarbara transect (Figure 1) in the western Monagasfold and thrust belt we applied a combined approachusing BPSM and structural restoration A BPSM is adynamic model of the subbasinrsquos physical processessuch as pore pressure and temperature evolution andthe development of the petroleum systems includinghydrocarbon generation expulsion migration accu-mulation and preservation (Hantschel and Kauerauf2009) The BPSM is performed in 1D (on a well orpseudowell) in 2D (usually along a seismic section)

and in 3D The BPSM can also provide predictionsabout seal capacity and integrity especially if fielddata are available for calibration

In contrast to a static model which consistsonly of geometry and properties for present day(as observed) a dynamic earth model includesthe evolution of geometry and properties Thepaleogeometrymdashor structural geometry of the modelat a given geologic age (or simulation time step)mdashisstrongly dependent on the paleowater depth (geomet-rical boundary condition) faulting activity the lithol-ogy of the various layers and their associatedcompaction behavior Approaching accurate paleoge-ometry is crucial for achieving consistent resultsespecially for fluid flow simulations In standardBPSM the paleogeometries are automatically gener-ated using a simplified restoration approach knownas backstripping (Figure 2A B)

Geomechanically based structural restorationuses the present-day interpreted and depth-convertedgeometry of a sedimentary basin or subbasin for therestoration of the basin structure backward in geo-logic time (Figure 2C D) This is done by sequen-tially removing the syntectonic sedimentary layersfrom the youngest to the oldest (the reverse of deposi-tion) allowing unfolding and unfaulting during therestoration (the reverse of deformation) In contrastto the simple backstripping approach structural resto-ration allows much more control through additionalgeometrical boundary conditions such as lateralmovement of rocks along faults or bedding slipRecent development has resulted in geomechanicalrestoration algorithms extending the pure geometri-cal approach by applying physical laws of rockdeformation (Maerten and Maerten 2006) The paleo-geometries obtained by structural restoration are thenused to describe basin geometry through time in apetroleum systems model

The backstripping approach is applied to tectoni-cally undisturbed areas and to areas where faultingoccurred much earlier than the main phase of hydro-carbon generation (typically a sag basin or the postriftsequence of a passive margin Figure 2A) In basinswhere extensional faulting occurred and where faultslip has an important horizontal component (eg lis-tric faults) backstripped paleogeometries showldquospikyrdquo artifacts (Figure 2B) these paleogeometries

NEUMAIER ET AL 1329

are generally not good enough for predicting prefault-ing hydrocarbon migration but might be acceptablefor hydrocarbon generation modeling Here structur-ally restored paleogeometries are important for thelayer shape in the proximity of faults and whencross-fault connectivity (like sandndashsand overlaps)matters (Figure 2C) Alternatively manual correctionof paleothickness can be done In halokinetic settingsbackstripping is generally superimposed by simplesalt reconstruction workflows (area balancing andextrapolation) Structural restoration should also beconsidered if salt movement results in geometries thatare more complex In any case thermal modeling inthe vicinity of salt is difficult due to its extremelyhigh thermal conductivity In areas with overthrustingor overturned folds (Figure 2D) with multiple strati-graphic sequences typically a fold and thrust beltbackstripping fails completely because lateral rockmovements are missing In the latter case paleo-geometries need to be generated by means other thanbackstripping (from simple conceptual drawing totrue structural restoration)

Paleogeometries generated from structuralrestorations can be loaded into BPSM software as analternative to performing a calculation from back-stripping Meshing is done section per section (timestep per time step) and tracking of cells from one sec-tion to the other is following rules described inHantschel and Kauerauf (2009) For the ages corre-sponding to the paleosection the basin geometry is

fixed it is not a function of compaction-controlledforward modeling as is the case when using back-stripping The BPSM simulation is restricted to thecalculation of properties such as porosity overpres-sure temperature vitrinite reflectance (Ro) andsource rock maturity which populate the fixed paleo-geometries as well as hydrocarbon migration Thesecompaction-controlled results are not used to adjustthe depositional amounts and to correct the paleoge-ometry as done in the backstripping approach andconsequently geometry optimization cannot beperformed The possible difference between thecompaction-law-controlled porosity change and thepredefined geometry change is indirectly compen-sated by the corresponding adjustment of the solidgrain mass

When paleogeometries are used for basin analy-sis and BPSM some factors become very importantRestoration must focus toward a best estimate paleo-water depth because changes in slope directions andtopography can largely affect fluid flow Erosionshould be estimated and reconstructed and manygeologic events need to be taken into account toincrease the resolution through time for dynamicmodeling The sediment thickness must be correctedin geologic time for decompaction All these factorswhich are common sources of uncertainty are criticalfor a geologically meaningful restoration

Because of the optimization method justdescribed adequately decompacted paleogeometries

Figure 2 Steps for (A B) backstripping and (C D) structural restoration of geometries of different structural complexity For detailssee text

1330 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

are required from structural restoration Inadequatedecompaction can produce large errors resulting inan unbalanced mass of solid rock matrix underesti-mated source rock paleoburial depth and deviatedpaleomigration paths related to differential compac-tion However it is difficult to define adequatedecompaction As a rule of thumb in normally pres-sured clastic basins where the main compactiondriver is linearly increasing effective stress simplehydrostatic decompaction laws such as thosedescribed by Athy (1930) can be used for restorationIn cases of abnormal pressure (overpressure) hydro-static decompaction overestimates the effectivestress or underestimates porosity and might fail tothe same degree as restoration neglecting decompac-tion One might be scaling normal decompactioncurves wherever regional overpressure occurs andhas occurred In basins with complex erosion pres-sure and diagenetic histories optimization loops(using BPSM simulated properties for optimizeddecompaction in second restoration) might benecessary Any sort of decompaction in a carbonate-dominated environment where compaction is gener-ally less understood because of the effect of mineraltransformations and cementation is subject to muchlarger uncertainty The question of whether to usehydrostatically decompacted or nondecompactedstructurally restored paleogeometries is discussed inthe section on uncertainty analysis

MODELING

Prior to structural restoration we performed several1D simulations These generated erosion and paleoal-titude estimations for structural restoration andprovided a calibration for the 2D model One-dimensional modeling also gave the first informationon hydrocarbon generation and timing to identifythe period of first source rock maturation (ldquocriticalmomentrdquo as defined by Magoon and Dow 1994)This is very important because prior to structuralrestoration its identification indicates when addi-tional time steps are needed to increase time resolu-tion for critical time intervals so that the activepetroleum system can be dynamically modeled in anadequate way However for simulation regular time

steps are needed not only during the critical momentbut also for the entire basin history

Structural Restoration

The following section describes our structural resto-ration results shown in Figure 3 of the post-riftstructural evolution of the Monagas fold and thrustbelt It can be divided into three main steps(1) From the Mesozoic to Oligocene the Monagasarea formed parts of the northern passive margin ofthe South American shield (Figure 3A) The sub-sidence driver was mostly thermal cooling and iso-static load with only a few normal faults Thesediments were sourced by the craton in the south(2) In the early Miocene the margin experiencedcompression from the transcollision with theCaribbean plate (Figure 3B) Two major normalfaults that formed during rifting were inverted in thecourse of the collision generating the regionalFurrial and Pirital thrusts Synorogenic deposits werecovered by the thrusted allochthon and intramontanepiggyback basins were formed With increasing upliftand erosion (middle Miocene Figure 3C) their sedi-ment fills were redeposited to form the important syn-tectonic Oficina and Carapita Formations south of thethrusts (3) From late Miocene on the late compres-sional phase started Figure 3D shows the present-day structure of the Monagas fold and thrust belt(2D model crossing the Santa Barbara field)According to our restoration the total shortening onthe section accommodated by the thrusts during thecompression is around 40 km (2486 mi) which cor-responds to 35 It is important to note that the Piritalthrust was active until late Miocene as indicated bythe fault displacement analysis (Figure 4) These latetectonic movements might have an important effecton the seal integrity of prospects that are in the vicin-ity of the thrust

Model Input Data

Table 1 shows geologic ages lithologies and petro-physical parameters that were assigned to our 2Dmodel For the main source rocks (Querecual andSan Antonio Formations) a type II B kinetic wasassigned (Pepper and Corvi 1995) As a base case

NEUMAIER ET AL 1331

we assumed a constant initial TOC of 6 wt and aninitial HI of 500 mg HCg TOC However becauseinitial TOC and HI were estimated from present-daymeasured values these initial values are highly uncer-tain In addition there is both lateral and vertical vari-ability within the layer even though locally very highsource rock potential is likely to occur a bulk TOCand HI need to be assigned to the model cells Forthese reasons several models with different initialTOC have been run and their sensitivity to the mod-eled charge is discussed below Faults were set open

to fluid flow in the course of their main kinematicactivity which is estimated by fault displacementanalysis of the structural restoration models Aftertheir main activity phase faults are assumed to beclosed Fault permeability was also a varying param-eter in the sensitivity analysis discussed belowThermal boundary conditions are given in Table 2Surface temperature ranges between 23 and 30degC(734 and 86degF) through geologic time with apresent-day average temperature of 28degC (824degF)These temperatures at sea level have been corrected

Figure 3 Structural evolution of the Monagas fold and thrust belt demonstrated by structural restoration (A) Passive margin stage(until 23 Ma) (B) Early compression stage (21 Ma) (C) Late compression stage (12 Ma) (D) Present day For details see text

1332 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

Figure 4 Fault displacement from lateMiocene to present day for all faults onrestored cross sections (A) and for the Piritalthrust only (B) Black and white shading andsize of points show magnitude of displacementin meters

NEUMAIER ET AL 1333

Table1

Basin

andPetro

leum

Syste

mModeling(BPSM)Input

ParametersGeologicAges

andLithologies

Compressib

ility

(GPa

-1)

Perm

eability

(logmD)

Thermal

conductivity

(Wm

K)

Heatcapacity

(kcalkgK)

Age

(Ma)

GeologicTime

Facies

Density

(kg∕m

3 )Initial

Porosity(

)max

min

at1

por

at25por

at20degC

at100degC

at20degC

at100degC

0Pleisto

cene

LasP

iedras

sandsto

ne2720

412747

115

minus18

3395

338

02

024

LasPiedrasshale

2700

4040327

403

minus852

minus3

164

169

021

024

26

Pliocene

LasP

iedras

sandsto

ne2720

412747

115

minus18

3395

338

02

024

LasPiedrasshale

2700

4040327

403

minus852

minus3

164

169

021

024

53

upperM

iocene

2La

Pica

sandsto

ne2720

412747

115

minus18

3395

338

02

024

LaPica

shale

2700

4040327

403

minus852

minus3

164

169

021

024

Morich

itoshale

2700

4040327

403

minus852

minus3

164

169

021

024

8upperM

iocene

1La

Pica

sandsto

ne2720

412747

115

minus18

3395

338

02

024

LaPica

shale

2700

4040327

403

minus852

minus3

164

169

021

024

Morich

itobasal

sandsto

ne2715

4812142

187

minus348

15

317

284

02

024

105

middleMiocene

2Carapitasandsto

ne2715

4812142

187

minus348

15

317

284

02

024

Carapitashale

2705

6330932

331

minus684

minus15

204

201

02

024

12middleMiocene

1Carapitashale

2705

6330932

331

minus684

minus15

204

201

02

024

15lower

Miocene

2Carapitashale

2705

6330932

331

minus684

minus15

204

201

02

024

19lower

Miocene

1lower

Carapitashale

2700

7040327

403

minus852

minus3

164

169

021

024

23Oligocene2

Merecureplatform

sandsto

ne2715

4812142

187

minus348

15

317

284

02

024

27Oligocene1

Merecureplatform

sandsto

ne2715

4812142

187

minus348

15

317

284

02

024

1334 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

339

PaleoEocene

Vidono

sandsto

ne2720

412747

115

minus18

3395

338

02

024

Caratasbasin

alshale

2700

7040327

403

minus852

minus3

164

169

021

024

655

upperC

retaceous4

SanJuan

sandsto

ne2720

412747

115

minus18

3395

338

02

024

74upperC

retaceous3

SanAntoniosiltstone

(SOUR

CERO

CK)

2710

5510422

212

minus628

minus1

201

196

023

026

825

upperC

retaceous2

SanAntoniosiltstone

(SOUR

CERO

CK)

2710

5510422

212

minus628

minus1

201

196

023

026

91upperC

retaceous1

Querecualmudsto

ne(SOUR

CERO

CK)

2700

7040327

403

minus852

minus3

164

169

021

024

996

lower

Cretaceous

2undifflower

Cretaceous

(30

Lst50

Sst

20Sh)

2722

459445

141

minus334

174

305

275

02

023

122

lower

Cretaceous

1undifflower

Cretaceous

(30

Lst50

Sst

20Sh)

2722

459445

141

minus334

174

305

275

02

023

1455

sub-Cretaceous

2undiffsub-Cretaceous

(50

Sh50Sst)

2710

5521537

259

minus516

0255

239

02

024

251

NEUMAIER ET AL 1335

by the influence of the paleowater depth using theapproach of Wygrala (1989) Because of the hightopographic elevation of approximately 3500 m(11500 ft) the maximum altitude during the pastcompressional phase we adjusted the model takinginto account a temperature decrease of 6degCkm (33degF1000 ft) related to the paleoaltitude This resultsin a difference of up to 20degC (36degF) compared to sea-level temperature We then assigned the adjustedsedimentwater interface temperatures to the modelas the upper thermal boundary condition At the baseof the model we assigned a heat flow ranging from42 to 80 mW∕m2 through geologic time The highestheat flow is assumed during the rifting event beforethe basin cooled as a passive margin A thermal eventin the Eocene detected by fission track analysis(Locke and Garver 2005 Perez de Armas 2005)has been addressed by a second peak before a secondcooling phase For uncertainty analysis basal paleo-heat flow has been varied within geological meaning-ful ranges and the thermal calibration data has beenhonored as much as possible

Modeling Results

Basin and petroleum system modeling can be dividedinto simulation of the physical parameters within abasin such as pressure and temperature (basin model-ing) and the subsequent evolution of the source rocksand their derived hydrocarbon compounds (petro-leum systems modeling) In the following sectionswe use this subdivision to describe the BPSM resultscompaction modeling thermal modeling and chargehistory It is important to state that these are predic-tions based on forward modeling as just describedThe results are compared and calibrated to actualmeasurements and observations where available

Compaction Modeling

Figure 5 presents an extraction of the modeled resultsat the location of Well Alpha in the Furrial Trend Inthe absence of direct calibration data for pore pres-sure we undertook a sensitivity analysis to test thedifferent scenarios Using the model we tested howpore pressure in the main reservoir of the FurrialTrend reacts to changes in fault and seal-rock per-meability In a first step fault permeability has beenevaluated Faults were assumed open during the mainfault activity phase and closed afterward In additionwe performed several variations from that base casescenario Figure 5A shows the effect of two end-member scenarios on pore pressure within the mainreservoir of the Furrial Trend (Santa Barbara field) inwhich faults have been modeled to be either open orclosed since their formation A third scenario consistsof thrust faults that are open during the compressionalphase (from 15 Ma on) and closed afterward (from105 Ma on) This scenario has been chosen as theldquomaster scenariordquo because it results in overpressureof 23 MPa which is in the range of pressure reportedin similar structures of the Furrial Trend by Schneideret al (2004) In addition this pressure scenariomatches the calibration data for porosity (Figure 5B)

The 2D distribution of modeled overpressure atpresent day is shown in Figure 6A Apart fromimportant overpressure in the Querecual Formationthe allochthon is almost normally pressured Thethrust wedge is overpressured especially in thelow-permeability Carapita Formation Although

Table 2 Basin and Petroleum System Modeling (BPSM) InputParameters Thermal Boundary Conditions

Age (Ma)

SedimentWaterInterface

Temperature (degC)

Age (Ma)

Basal HeatFlow

(mW∕m2)min max

0 28 28 0 4226 28 28 30 6053 76 23 34 708 184 231 50 50105 62 233 107 6012 31 238 1455 8015 5 242 260 6019 5 24723 234 2527 78 235339 94 236655 146 2831455 148 287255 5 239

1336 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

the autochthon is highly overpressured close tonormal hydrostatic pressures are simulated in theforeland basin The discussed model assumption ofclosed faults in the postcompressional phase resultsin a regional pressure compartmentalization

The simulated pressure evolution through geo-logic time can be seen in Figure 7A A time extractionof a source rock cell in the thrust wedge shows a sud-den increase of both lithostatic and hydrostatic pres-sures that is a consequence of the tectonic burial

load associated with the thrusting In addition themodeled pore pressure buildup even though delayedin time can be directly linked to the thrust event Itcan be clearly seen that the timing of fault permeabil-ity directly impacts changes in the present-day porepressure The earlier the faults close the more over-pressure generated by the massive syntectonic sedi-mentation is kept until present day given adequatesealing lithologies are present The present-day over-pressure is modeled as14 MPa when the faults are

Figure 5 Plots of one-dimensional model output and available calibration data for Well Alpha (Table 3) (A) Dotted lines are hydro-static and lithostatic pressures continuous lines are pore pressure the master scenario is in black (a1 faults closing at 105 Ma) andadditional scenarios in gray (a2 open faults a3 closed faults) (B) modeled porosity (C) borehole temperature and (D) Ro The blackline shows the master scenario and the gray line the alternative high-heat-flow scenario Bold lines in the stratigraphic column arethrusts (T) and unconformities (U)

Table 3 Calibration Data for Well Alpha (Porosity Temperature Ro)

TVD (m) Porosity () TVD (m) Corrected Temperature (degC) TVD (m) Vitrinite Reflectance (Ro)

4350 15 600 40 400 0264400 15 1100 53 1200 0484800 10 1800 68 1485 068

4400 131 1700 0394800 133 2900 048

3100 0493400 0494000 054300 0524800 06

NEUMAIER ET AL 1337

open since 8 Ma as 23 MPa for 105 Ma and as37 MPa for 12 Ma Compared to this order of magni-tude changes in the permeability of the overlyingCarapita shale are negligible for the pore pressureTherefore the controlling parameter for pore pressurein the main reservoir of the Furrial trend is assumedfault permeability and resulting lateral connectivityrather than the permeability of the seal rock

Thermal Modeling

Modeled temperature and thermal maturity(Figure 5C D) have been calibrated with availablewell data Some discontinuities in the modeled Ro

trend are noted (Figure 5D) The upper kinks (egbase Pliocene) are related to unconformities withassociated erosion and the lower one (base UpperCretaceous) is due to the well penetrating a reverse

Figure 6 Two-dimensional model at present day with (A) modeled overpressure and (B) modeled thermal maturity and isothermsSource rocks marked by lithology pattern

1338 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

fault displacing mature rocks upon less mature ones(see Figure 6B for 2D view) There is a mismatchbetween the modeled Ro trend and the data in thedeeper part of the well In the Oligocene rocks themodeled temperature is too low (2degC or 36degF)whereas the modeled Ro is too high (up to 02 Ro)Any attempt to reduce the modeled Ro (eg bychanging thermal conductivity of the MioceneCarapita Formation) also results in a lower modeledpresent-day temperature Therefore the thermal sce-nario used for this calibration is a trade-off betweenmatching present-day temperature and Ro (using thekinetics of Sweeney and Burnham 1990) The basalheat flow assumed in the master scenario (Table 2)results in a modeled present-day surface heat flowranging from 37 mW∕m2 in the southern part of thethrusts to 52 mW∕m2 in parts of the uplifted areaAt the same depth the allochthonous northern partis thermally more mature than the southern autochthon(Figure 6B) because of the recent uplift Locallyinverted trends around the thrusts can be observedfor example at the location of Well Alpha (seeFigure 5D) In the autochthon the Upper Cretaceoussource rocks are entirely in the oil or even gas windowat present day because of the Neogene tectonic loadingand synorogenic deposition In the northern alloch-thon most of the Upper Cretaceous source rocks havebeen in the oil window in the past The modeled ther-mal maturity is in good agreement with the regionalmaturity suggested by Summa et al (2003)

Similar to the pressures the thrusting also affectsthe modeled temperature and Ro In the thrust wedgea direct relationship exists between the thrusting andthe increase of temperature and maturity The thrust-ing is the trigger for source rocks entering the oil win-dow within the thrust wedge (Figure 7B) and thepresent-day thermal maturity has been reached sincethe last 10 Ma Although this recent thermal historycan be considered as relatively well constrained bythe measured Ro in the calibration well this doesnot apply to the past In this thrusted area any ther-mal maturity reached prior to the thrusting onset hasbeen ldquooverprintedrdquo and therefore no thermal calibra-tion is possible for this period However theCretaceous source rocks were not buried deeply dur-ing the passive margin phase and even very hot sce-narios (basal heat flow of 100 mW∕m2) do not

significantly change their maturity (Figure 7B) Inthe uplifted allochthon several wells show low(03) to moderate (1) measured Ro (Parra et al2011) Because the allochthon experienced its maxi-mum burial and the highest temperatures during thepassive margin phase these maturity levels can beassumed as representative of the maturity prior tothe compression In contrast to the thrust wedge theconsequence of the thrusting is inverse for the alloch-thon (Figure 7C) as the hanging wall is uplifted andthe accompanying erosion removes the overburdenThe highest temperatures are simulated to have beenreached just before the thrusting event Then temper-ature decreased and Ro stagnated leaving the sourcerocks in the early mature zone until present day Theassumption of a regionally relatively homogeneousthermal field during the cooling phase of the passivemargin in combination with the relatively low matur-ities measured in the uplifted allochthon makes thehot heat-flow scenario unlikely However a laterallyvarying heat flow cannot be excluded and the conse-quently hotter paleoheat flow would have led to muchhigher temperatures prior to compression resultingalso in a higher present-day maturity level (14)despite the uplift (Figure 7C) Any variations of basalheat flow during the rift period (Late Jurassic to EarlyCretaceous) occurred too early to affect the maturityof the source rocks deposited during the lateCretaceous

Charge History

After calibration of the basinrsquos physical environ-ment the evolution of the petroleum systems wasmodeled Hydrocarbon generation was simulatedthrough geologic time using the activation energydistribution of the assigned source rock kineticsFrom the moment of hydrocarbon expulsiononward secondary migration was modeled usingthe invasion percolation method (Wilkinson 1984)It takes into account the geometry overpressure gra-dient and capillary entry pressures of the rockswhich are opposed to the buoyancy of the migratingoil and gas Figure 8 shows the time evolution of thetotal study area and Figure 9 focuses on the thrustwedge As already mentioned the described chargehistory is based on forward modeling It is consistent

NEUMAIER ET AL 1339

Figure 7 Time extractions (see Figure 6 for location) (A) Pressure development through geologic time for a source rock in the ver-tical continuity of Well Alpha Dotted lines are hydrostatic and lithostatic pressures continuous lines are pore pressure the master sce-nario is in black (a1 faults closing at 105 Ma) and additional scenarios in gray (a2 open faults a3 closed faults a4 faults closing at12 Ma a5 faults closing at 8 Ma) (B) Temperature (continuous lines) and Ro (dotted lines) through geologic time for a source rockin the vertical continuity of Well Alpha the black line shows the master scenario and the gray line the alternative high-heat-flow sce-nario (C) Temperature (continuous lines) and Ro (dotted lines) through geologic time for a source rock in the uplifted allochthonthe black line shows the master scenario and the gray line the alternative high heat flow scenario

1340 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

with the modeled compaction and thermal historyand therefore shows one scenario that explains thecharge of the Santa Barbara field and the forelandbasin Simplifications are inherent to any such work-flow and shall be considered when analyzing themodeling results

Passive Margin PhasemdashMesozoic toOligocene

Modeling results indicate that already during the LateCretaceous any Jurassic source rocks potentiallydeposited in the northern part of the section (Summaet al 2003) have reached temperatures of more than200degC (392degF) and corresponding high levels of matu-rity However the temperature at the base of theQuerecual Formation (Upper Cretaceous) was lessthan 100degC (212degF) leaving the kerogen immatureThe source rock became effective only during lateEocene with temperatures of up to 120degC (248degF)and a (reactive kerogen to hydrocarbon) transforma-tion ratio (TR) of more than 20 in the deeper distalpart of the passive margin Oil was quickly expelledmigrated vertically and seeped to the surface becauseof the general lack of seals in the overburden Duringthe Oligocene the time of deposition of the Merecurereservoir sandstones in the southern platform this sit-uation did not change apart from increased transforma-tion in the deeper basinal part of the passive margin

First oil accumulations are modeled in the Vidontildeosandstones in the basinal part in the late Oligocene(Figure 8A) in which the base of the QuerecualFormation shows TRs of up to 60 at around 130degC(266degF) In the proximal part there is no hydrocarbongeneration Although there is a general northwardtilting slope no southward migration from the distalmargin is modeled this has been checked usingdifferent migration methods including Darcy flowwith scenarios of continuous intrasource carrier bedsThis is in contrast to the studies from Gallango andParnaud (1995) and Schneider (2003) who suggestedlateral migration at that stage The difference couldbe explained by the presence of a regional top seal intheir models However our model suggests that theshallow rocks do not have sufficient sealing capacityto allow lateral hydrocarbon migration During thepassive-margin phase the basin was normally

pressured (Figure 7A) and simulated stresses havenot yet resulted in any fracturing at that stage

Main Compressional PhasemdashEarly to MiddleMiocene

With the onset of compression during early Miocene(Figure 8B) both the Pirital and Furrial thrusts wereformed and the northern distal part of the passivemargin was rapidly uplifted Important overpressuresas well as mechanical rock failure both the result ofthe rapid uplift are simulated in the allochthonDespite this uplift the kerogen-to-hydrocarbon trans-formation continued and the rate even increasedlocally close to the footwall cutoffs and where syn-orogenic deposition provided rapid overburden load-ing Oil accumulations are modeled in the freshlycreated anticlines

The tectonicndashsedimentary wedge between thetwo main thrusts experienced very rapid burial Infact the Upper Cretaceous source rocks were broughtinto depths where temperatures were as high as185degC (365degF) at 19 Ma and 225degC (437degF) at15 Ma Especially close to the cutoff of the Piritalthrust they very quickly reached a high maturity at19 Ma and reached full transformation in most placesat 12 Ma The expelled hydrocarbons migrated intothe sandstones of the Merecure Formation which weresealed by the Carapita Formation The charge toward aseries of recent anticlines is modeled by the fill-to-spillmechanism from north to south (Figure 9C D) Afterthe first oil fill the northern prospect quickly receiveda local gas charge the central structure containedmainly oil Note the presence of a lateral seal that isdue to sandndashshale juxtaposition The southernmoststructure which forms the Santa Barbara field todaywas charged by oil only at 15 Ma (Figure 9D)

The generated hydrocarbons in the southernautochthon were leaking to the surface until theCarapita Formation was compacted enough to holdsome of them (Figure 8C) At 12 Ma the wedgesource rocks were mostly spent but expulsion wasstill ongoing Important rock failure is modeled tohave occurred in the thrusted footwall breaking theformerly sealing Carapita Formation (Figure 10)This is in good agreement with cemented hydraulicfractures observed in a few cores in the Carapita

NEUMAIER ET AL 1341

Figure 8 Two-dimensional model through geologic time (selected time steps only) with modeled transformation ratio for the sourcerock layers isotherms and liquid and vapor migration and accumulation

1342 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

shales from the Furrial Trend (Roure et al 2005)This important fracturing event resulted in thedestruction of the modeled accumulation (Figure 9E)Hydrocarbon generation and expulsion were alsohappening in the southern foreland basin at transfor-mation ratios of up to 30

Late Compressional and PostcompressionalPhasemdashLate Miocene to Present Day

The late compressional phase was marked by the par-tial destruction of the allochthonous petroleum systemsby erosion (Figure 8E F) Many traps disappeared and

Figure 8 Continued

NEUMAIER ET AL 1343

the hydrocarbon generation within the source rocksstopped The erosion triggered an important sedimenttransfer toward the south contributing to further burialand increased temperatures in the thrust wedge and theforeland basin During late Miocene the petroleumsystem in the wedge matured An important northwardtilting was triggered by flexural tectonic loading andthe loading of syntectonic sediments (Summa et al2003) In combination with the presence of an effec-tive regional top seal the compacted CarapitaFormation allowed a change from dominantly verticalmigration into lateral migration First accumulationsare modeled in the Neogene synorogenic sequence

reservoirs in the south and important southwardmigration toward the foreland basin started(Figure 8G) These trends continued through Plioceneand Pleistocene times to present day filling theOnado field (Figure 8H) and other known fields inthe south (Figure 1A) of the Eastern Venezuela Basin

Prospectivity

We modeled the charge of the Santa Barbara field inthe Furrial Trend and of the Onado field in the fore-land basin (Figure 8H) Other prospects in the thrustwedge were also investigated The structure called

Figure 9 Detail of two-dimensionalmodel through geologic time (seeFigure 8) showing modeled transforma-tion ratio for the source rock layers iso-therms and liquid and vapor migrationand accumulation The Santa Barbara fieldand the prospect are shown by ellipses

1344 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

ldquoprospectrdquo (Figure 9G) looks similar to that of theSanta Barbara field because both are anticlines andthe reservoir and seal formations are the sameHowever they differ in several aspects First theprospect is in the vicinity of the Pirital thrust whichpresents a high risk of impacting the seal integrityThe prospect location is directly below the area wherewe presume the maximum of late tectonic activityoccurred (Figure 4) As modeled the CarapitaFormation the seal of the prospect was already brit-tle when this deformation occurred Calculatedstresses also predict rock failure from 12 Ma onward(Figure 10) Even though charge continued ourmodel suggests continuous leakage of the structure(Figure 9EndashG) The second major difference is thatthe source rock in the fault block of the prospect ismodeled to have already started generating and expel-ling hydrocarbons during the Oligocene and earlyMiocene (Figure 9A) before the trap formed and the

seal became effective (Figures 9C 10) Even thoughgeneration expulsion and migration continued aftertrap formation the prospect was not charged withthe earlier expelled oil In addition the present-daymaturity of the source rocks within the prospectrsquosfault block is higher than in the source rocks withinthe tectonic block of the Santa Barbara field poten-tially indicating expulsion of more gas A southndashnorth trend can be seen on the modeled accumulationcomposition with more gas toward the allochthon(Figure 9) The modeled composition of the prospectprior to seal failure was much richer in gas than inthe neighboring Santa Barbara field (Figure 9D) Infact simulation scenarios with unbreached sealsresulted in accumulation of predominantly gas Thecombination of both seal integrity and charge compo-sition concerns makes this a risky prospect In addi-tion potential diagenetic fluids related to the faultfluid flow might have altered the reservoir quality

Figure 10 Two-dimensional model (at 12 Ma) with modeled stress to failure Mohr circles (with σ standing for normal stress and τfor shear stress) and petroleum system events charts for the Santa Barbara field (left) and the prospect (right) Model predicts rock fail-ure in the prospect from about 12 Ma For further details see text

NEUMAIER ET AL 1345

Apart from the Onado oil field only minor accu-mulations are modeled in the southern Neogene syn-orogenic sequence The proportion of gas is higherthan in the Furrial trend The accumulations at thesouthern model edge are a modeling artifact repre-senting hydrocarbons (mainly oil) that in reality arethought to have charged the southern foreland basin(from mid to late Miocene times to present day)However parts of the migrated oil and gas could havebeen trapped in the Upper Cretaceous San JuanFormation south of the horst structure

The potential for conventional hydrocarbon accu-mulations in the allochthon is very limited by theimportant fracturing that occurred during the upliftperiod The basal sandstones of the Morochito piggy-back basin north of the Pirital thrust are not chargedin our 2D model because of the absence of structuraltraps Nevertheless there is a potential for strati-graphic traps to have been charged as migrationthrough the basin occurred However seal strengthmight be critical at this low-compaction stage Thesame applies for the Las Piedras Formation We thinkthere might be potential for shale oil because theQuerecual Formation is at shallow depths but showsmaturity within the oil window From the thermalmaturity point of view any potential Jurassic sourcerocks as discussed by Summa et al (2003) mightbe a gas source or shale gas target

UNCERTAINTY ANALYSIS

There is often significant uncertainty in basin geom-etry and depth of stratigraphic horizons based onuncertain seismic interpretation and velocity models(Kukla et al 2000 Brandes et al 2008 Baur et al2010) Restoring of present-day geometries to obtainpaleogeometries adds further uncertainty Additionaluncertainty is related to thermal boundary conditionssuch as paleoheat flow (Beha et al 2008) and togeochemical properties such as source rock character-istics and petroleum generation kinetics and calcula-tion of biodegradation and petroleum density(Blumenstein et al 2008) In addition to the porepressure and thermal sensitivity analysis justdescribed we varied initial TOC within a range of2 to 10 A variation of initial TOC obviously

directly scales the amount of initial kerogen availablefor transformation into hydrocarbons Neither thecharge volume nor composition of the Santa Barbarafield nor the wedge prospects are significantly sensi-tive to variation of initial TOC these traps are filledto spill in each scenario However the amount ofhydrocarbons charging the foreland basin (eg theOnado field) is reduced in the case of low initialTOC A high initial TOC only increased the modeledamount of surface seepage The sensitivity of the ini-tial HI is analogous to the sensitivity of the ini-tial TOC

In addition to the more classical uncertaintyand sensitivity analysis described we focused onquantitative analysis of the decompaction uncer-tainty when performing structural restoration Thisuncertainty is particular to the applied workflowof using structurally restored paleogeometries inBPSM Note that this uncertainty is significantlylower when using the backstripping method withthe optimization algorithms discussed One methodto analyze the adequacy of the decompactionmethod is comparison of the output rock matrixmass balance In the backstripping approach therock matrix mass is kept constant of a model cellwith a simulated decrease in porosity controlledby compaction laws and the corresponding reduc-tion of the layer thickness When using fixed pale-ogeometries from structural restoration the cellthickness is imposed However the decrease inporosity is calculated and the mass is adapted tofit the imposed cell thickness That means that therock matrix mass is constant only if ideal decom-paction has been applied

Figure 11A shows the percentage variation fromthe present-day matrix mass (100) of a given sourcerock cell through geologic time The dotted line rep-resents the ideal case with constant rock matrix massWe used two end-member decompaction scenarios todescribe the decompaction-related uncertainty that isintroduced when fixed paleogeometries replace back-stripping idealized hydrostatic decompaction assum-ing no overpressure and idealized lithostaticdecompaction assuming no effective stress (nodecompaction resulting in constant cell thicknessthrough time) In the latter case with decreasingporosity during compaction the rock matrix mass

1346 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

increases until it reaches the present-day mass Inother words rock mass matrix has been highly under-estimated in the past (in this particular case by morethan 20) This is a source rock so the kerogen massdecreases by the same factor This yields an underes-timation of more than 20 in TOC Note the rockmatrix mass is constant for time steps prior to23 Ma in which backstripping optimization couldbe performed toward the earliest paleogeometry Forthe second end-member scenario of idealized hydro-static decompaction the rock matrix mass is nearlyconstant which is a good decompaction adequacy

indicator for that given cell in the model Howeverthe so-measured adequacy changes throughout themodel In some areas the rock matrix mass has beenunderestimated by about 50 for nondecompactedpaleogeometries however in the case of hydrostaticdecompaction values generally range from 80 to120 of the present-day rock matrix mass Thatmeans in this particular case that hydrostatic decom-paction is more adequate (or less wrong) than nodecompaction at all The decompaction adequacycan be roughly correlated with pressure changesthrough geologic time

Figure 11 Time extractions for comparison of two end-member decompaction scenarios (see Figure 6 for location) (A) Rock matrixmass conservation through geologic time (B) Source rock burial depth through geologic time (C) Temperature and transformation ratiothrough geologic time

NEUMAIER ET AL 1347

Inadequate decompaction results in wrong depthfor stratigraphic horizons in the paleogeometriesFigure 11B shows the burial depth of the same inves-tigated source rock cell We can see that during thepassive margin phase there is a slight difference inpaleoburial depth for the two end-member scenariosDuring the compressional phase the differencechanges dramatically reaching more than 1000 m(3281 ft) in some places Toward present day bothcurves meet again because the input to the two resto-ration scenarios was the same (identical present-dayburial depth) This difference in compaction calcula-tion has severe implications on the modeled tempera-ture history and the kerogen-to-hydrocarbontransformation (Figure 11C) For an identical thermalframework in terms of modeling input parameterstemperature differences of up to 25degC (45degF) resultin TR differences of more than 20 in the early com-pressional phase before TR values meet again for thepresent-day situation

CONCLUSIONS

Structurally complex areas are attracting increasinginterest in hydrocarbon exploration Exploration riskcan be substantially narrowed by analyzing the his-tory of the basin and its petroleum systems based onstructural evolution We presented a further progressin technology that allows the integration of structuralrestoration and BPSM This combined analysis isseen as a way forward to improve modeling in thrustbelts with stacked stratigraphic sequences Despitelimitations of nonoptimized decompaction whichare added to the general uncertainty related to theindividual methods we modeled the charge and sealhistory of the Monagas fold and thrust belt inVenezuela Calibrated porosity temperature andthermal maturity as well as basin-scale stress pro-vided the regional framework through geologic timein which the petroleum systems evolved such thatthe charge of the known Santa Barbara and Onadofields and other trends could be understood We dis-cussed seal integrity of a potential exploration pros-pect by analyzing structural restoration-derived faultdisplacement through geologic time and an advancedstress forward simulation The uncertainty of using

decompacted or nondecompacted structural restora-tion has been assessed It can have important influ-ence on rock matrix mass balance through geologictime paleogeometry and paleoburial depth andtherefore affect paleotemperatures and paleomatur-ities We conclude that decompaction assuminghydrostatic pressure is the best option in that particu-lar case

REFERENCES CITED

Alberdi M and E Lafargue 1993 Vertical variations of organicmatter content in Guayuta Group (Upper Cretaceous)interior mountain belt eastern Venezuela OrganicGeochemistry v 20 p 425ndash436 doi 1010160146-6380(93)90091-O

Athy L F 1930 Density porosity and compaction of sedimen-tary rocks AAPG Bulletin v 14 p 1ndash24

Aymard R L Pimentel P Eitz P Lopez A ChaouchJ Navarro J Mijares and J G Pereira 1990 Geologic inte-gration and evaluation of Northern Monagas EasternVenezuelan Basin in J Brooks ed Classic petroleum prov-inces The Geologic Society (London) Special Publicationv 50 p 37ndash53

Baur F M Di Benedetto T Fuchs C Lampe andS Sciamanna 2009 Integrating structural geology andpetroleum systems modelingmdashA pilot project from Boliviarsquosfold and thrust belt Marine and Petroleum Geology v 26no 4 p 573ndash579 doi 101016jmarpetgeo200901004

Baur F R Littke H Wielens C Lampe and T Fuchs 2010Basin modeling meets rift analysismdashA numerical modelingstudy from the Jeanne drsquoArc basin offshore NewfoundlandCanada Marine and Petroleum Geology v 27 p 585ndash599doi 101016jmarpetgeo200906003

Beha A R O Thomsen and R Littke 2008 Thermal historyhydrocarbon generation and migration in the Horn Grabenin the Danish North Sea A 2D basin modelling studyInternational Journal of Earth Sciences v 97 p 1087ndash1100doi 101007s00531-007-0247-2

Blumenstein I O B M Krooss R di Primio W RottkeE Muumlller C Westerlage and R Littke 2008Biodegradation in numerical basin modelling A case studyfrom the Gifhorn Trough N-Germany International Journalof Earth Sciences v 97 p 1115ndash1129 doi 101007s00531-007-0272-1

Brandes C A Astorga R Littke and J Winsemann 2008Basin modelling of the Limoacuten back-arc basin (Costa Rica)Burial history and temperature evolution of an island arc-related basin-system Basin Research v 20 p 119ndash142doi 101111bre200820issue-1

Carnevali J O 1988 El Furrial oil field northeasternVenezuela First giant in foreland fold and thrust belts ofwestern hemisphere AAPG Bulletin v 72 68 p

Chevalier Y 1987 Contribution a lrsquoetude geacuteologique de lafrontiereacute Sud-Est de la plaque Caraibe Les zones internesde la chaine Sud-Caraibe sur le transect ile de Margarita-

1348 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

contains some of the worldrsquos largest oil reserves forbasins with this type of tectonics Many shallow oilfields in the Eastern Venezuela Basin are located inNeogene sandstone reservoir rocks (Las Piedras LaPica Morichito Chapapotal Freites and Oficinafields) Other reservoir formations are the synoro-genic Miocene Naricual Formation and theOligocene Merecure Formation which are part of aproximal passive-margin sequence that shales outtoward the north The latter is the main reservoir inthe thrusted Furrial trend In the deeper passive-margin sequence there are potential reservoirs in theMesozoic (eg San Juan Formation) Because of its

shaly lithology large thickness and lateral continu-ity the synorogenic Carapita Formation is probablythe most effective seal However several other for-mations have good sealing capacities even thoughnot well compacted as numerous fields at shallowdepths prove

The first publications on the thermal history ofthe Eastern Venezuela Basin were based on one-dimensional (1D) models Because of the com-plexities leading to multiple sequences caused byoverthrusting modeling was done either in theunthrusted area or if within the thrusted area onlyfor geologic time intervals preceding thrusting

Figure 1 (A) Map of the Eastern Venezuela Basin showing structural elements and hydrocarbon occurrence (modified after Parnaudet al 1995) Solid line shows location of two-dimensional (2D) line modeled in this study (B) Generalized northndashsouth geodynamiccross section of the study area (modified after Chevalier 1987) (C) Synthetic stratigraphy of the southern foreland basin of theEastern Venezuela Basin (modified after Roure et al 2003)

1328 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

(Talukdar et al 1988) Summa et al (2003) pub-lished multiple 1D models over a large area andextrapolated the results onto regional maturity mapsfor the Guayuta Group and Cenozoic source rocksThese maps include a hanging-wall cutoff and there-fore do not represent the higher maturities found inthe thrust wedges Elsewhere the authors suggestedthat most source rocks have been in the oil windowin the past A present-day drainage area analysis ofthe top Cretaceous structure map was used to describethe southward charge from the deeper kitchen into theforeland basin This regional migration modelexplains oil provenance and quality Important timingaspects such as the relationship between the struc-tural history and source rock maturation and expul-sion are described using a regional structuralrestoration in conjunction with the 1D models Parraet al (2011) included the thrusting effect by using a15D module mimicking lateral rock transportationthat is due to thrusting Gallango and Parnaud(1995) performed 2D modeling including hydrocar-bon migration in the unthrusted southern area Theyalso described a second model within the thrustedarea that simulates the basin history prior to compres-sion (from the Mesozoic to early Miocene) Thismodel proposes long-distance petroleum migrationsouthward before the transpressional event whichresulted in the observed structural complexity Inother papers Roure et al (2003) Schneider (2003)and Schneider et al (2004) described fluid flow porefluid history and diagenetic processes in a 2D modelalong a cross section parallel to the location of themodel presented here

MODELING METHODS

To address the charge and seal history of the SantaBarbara transect (Figure 1) in the western Monagasfold and thrust belt we applied a combined approachusing BPSM and structural restoration A BPSM is adynamic model of the subbasinrsquos physical processessuch as pore pressure and temperature evolution andthe development of the petroleum systems includinghydrocarbon generation expulsion migration accu-mulation and preservation (Hantschel and Kauerauf2009) The BPSM is performed in 1D (on a well orpseudowell) in 2D (usually along a seismic section)

and in 3D The BPSM can also provide predictionsabout seal capacity and integrity especially if fielddata are available for calibration

In contrast to a static model which consistsonly of geometry and properties for present day(as observed) a dynamic earth model includesthe evolution of geometry and properties Thepaleogeometrymdashor structural geometry of the modelat a given geologic age (or simulation time step)mdashisstrongly dependent on the paleowater depth (geomet-rical boundary condition) faulting activity the lithol-ogy of the various layers and their associatedcompaction behavior Approaching accurate paleoge-ometry is crucial for achieving consistent resultsespecially for fluid flow simulations In standardBPSM the paleogeometries are automatically gener-ated using a simplified restoration approach knownas backstripping (Figure 2A B)

Geomechanically based structural restorationuses the present-day interpreted and depth-convertedgeometry of a sedimentary basin or subbasin for therestoration of the basin structure backward in geo-logic time (Figure 2C D) This is done by sequen-tially removing the syntectonic sedimentary layersfrom the youngest to the oldest (the reverse of deposi-tion) allowing unfolding and unfaulting during therestoration (the reverse of deformation) In contrastto the simple backstripping approach structural resto-ration allows much more control through additionalgeometrical boundary conditions such as lateralmovement of rocks along faults or bedding slipRecent development has resulted in geomechanicalrestoration algorithms extending the pure geometri-cal approach by applying physical laws of rockdeformation (Maerten and Maerten 2006) The paleo-geometries obtained by structural restoration are thenused to describe basin geometry through time in apetroleum systems model

The backstripping approach is applied to tectoni-cally undisturbed areas and to areas where faultingoccurred much earlier than the main phase of hydro-carbon generation (typically a sag basin or the postriftsequence of a passive margin Figure 2A) In basinswhere extensional faulting occurred and where faultslip has an important horizontal component (eg lis-tric faults) backstripped paleogeometries showldquospikyrdquo artifacts (Figure 2B) these paleogeometries

NEUMAIER ET AL 1329

are generally not good enough for predicting prefault-ing hydrocarbon migration but might be acceptablefor hydrocarbon generation modeling Here structur-ally restored paleogeometries are important for thelayer shape in the proximity of faults and whencross-fault connectivity (like sandndashsand overlaps)matters (Figure 2C) Alternatively manual correctionof paleothickness can be done In halokinetic settingsbackstripping is generally superimposed by simplesalt reconstruction workflows (area balancing andextrapolation) Structural restoration should also beconsidered if salt movement results in geometries thatare more complex In any case thermal modeling inthe vicinity of salt is difficult due to its extremelyhigh thermal conductivity In areas with overthrustingor overturned folds (Figure 2D) with multiple strati-graphic sequences typically a fold and thrust beltbackstripping fails completely because lateral rockmovements are missing In the latter case paleo-geometries need to be generated by means other thanbackstripping (from simple conceptual drawing totrue structural restoration)

Paleogeometries generated from structuralrestorations can be loaded into BPSM software as analternative to performing a calculation from back-stripping Meshing is done section per section (timestep per time step) and tracking of cells from one sec-tion to the other is following rules described inHantschel and Kauerauf (2009) For the ages corre-sponding to the paleosection the basin geometry is

fixed it is not a function of compaction-controlledforward modeling as is the case when using back-stripping The BPSM simulation is restricted to thecalculation of properties such as porosity overpres-sure temperature vitrinite reflectance (Ro) andsource rock maturity which populate the fixed paleo-geometries as well as hydrocarbon migration Thesecompaction-controlled results are not used to adjustthe depositional amounts and to correct the paleoge-ometry as done in the backstripping approach andconsequently geometry optimization cannot beperformed The possible difference between thecompaction-law-controlled porosity change and thepredefined geometry change is indirectly compen-sated by the corresponding adjustment of the solidgrain mass

When paleogeometries are used for basin analy-sis and BPSM some factors become very importantRestoration must focus toward a best estimate paleo-water depth because changes in slope directions andtopography can largely affect fluid flow Erosionshould be estimated and reconstructed and manygeologic events need to be taken into account toincrease the resolution through time for dynamicmodeling The sediment thickness must be correctedin geologic time for decompaction All these factorswhich are common sources of uncertainty are criticalfor a geologically meaningful restoration

Because of the optimization method justdescribed adequately decompacted paleogeometries

Figure 2 Steps for (A B) backstripping and (C D) structural restoration of geometries of different structural complexity For detailssee text

1330 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

are required from structural restoration Inadequatedecompaction can produce large errors resulting inan unbalanced mass of solid rock matrix underesti-mated source rock paleoburial depth and deviatedpaleomigration paths related to differential compac-tion However it is difficult to define adequatedecompaction As a rule of thumb in normally pres-sured clastic basins where the main compactiondriver is linearly increasing effective stress simplehydrostatic decompaction laws such as thosedescribed by Athy (1930) can be used for restorationIn cases of abnormal pressure (overpressure) hydro-static decompaction overestimates the effectivestress or underestimates porosity and might fail tothe same degree as restoration neglecting decompac-tion One might be scaling normal decompactioncurves wherever regional overpressure occurs andhas occurred In basins with complex erosion pres-sure and diagenetic histories optimization loops(using BPSM simulated properties for optimizeddecompaction in second restoration) might benecessary Any sort of decompaction in a carbonate-dominated environment where compaction is gener-ally less understood because of the effect of mineraltransformations and cementation is subject to muchlarger uncertainty The question of whether to usehydrostatically decompacted or nondecompactedstructurally restored paleogeometries is discussed inthe section on uncertainty analysis

MODELING

Prior to structural restoration we performed several1D simulations These generated erosion and paleoal-titude estimations for structural restoration andprovided a calibration for the 2D model One-dimensional modeling also gave the first informationon hydrocarbon generation and timing to identifythe period of first source rock maturation (ldquocriticalmomentrdquo as defined by Magoon and Dow 1994)This is very important because prior to structuralrestoration its identification indicates when addi-tional time steps are needed to increase time resolu-tion for critical time intervals so that the activepetroleum system can be dynamically modeled in anadequate way However for simulation regular time

steps are needed not only during the critical momentbut also for the entire basin history

Structural Restoration

The following section describes our structural resto-ration results shown in Figure 3 of the post-riftstructural evolution of the Monagas fold and thrustbelt It can be divided into three main steps(1) From the Mesozoic to Oligocene the Monagasarea formed parts of the northern passive margin ofthe South American shield (Figure 3A) The sub-sidence driver was mostly thermal cooling and iso-static load with only a few normal faults Thesediments were sourced by the craton in the south(2) In the early Miocene the margin experiencedcompression from the transcollision with theCaribbean plate (Figure 3B) Two major normalfaults that formed during rifting were inverted in thecourse of the collision generating the regionalFurrial and Pirital thrusts Synorogenic deposits werecovered by the thrusted allochthon and intramontanepiggyback basins were formed With increasing upliftand erosion (middle Miocene Figure 3C) their sedi-ment fills were redeposited to form the important syn-tectonic Oficina and Carapita Formations south of thethrusts (3) From late Miocene on the late compres-sional phase started Figure 3D shows the present-day structure of the Monagas fold and thrust belt(2D model crossing the Santa Barbara field)According to our restoration the total shortening onthe section accommodated by the thrusts during thecompression is around 40 km (2486 mi) which cor-responds to 35 It is important to note that the Piritalthrust was active until late Miocene as indicated bythe fault displacement analysis (Figure 4) These latetectonic movements might have an important effecton the seal integrity of prospects that are in the vicin-ity of the thrust

Model Input Data

Table 1 shows geologic ages lithologies and petro-physical parameters that were assigned to our 2Dmodel For the main source rocks (Querecual andSan Antonio Formations) a type II B kinetic wasassigned (Pepper and Corvi 1995) As a base case

NEUMAIER ET AL 1331

we assumed a constant initial TOC of 6 wt and aninitial HI of 500 mg HCg TOC However becauseinitial TOC and HI were estimated from present-daymeasured values these initial values are highly uncer-tain In addition there is both lateral and vertical vari-ability within the layer even though locally very highsource rock potential is likely to occur a bulk TOCand HI need to be assigned to the model cells Forthese reasons several models with different initialTOC have been run and their sensitivity to the mod-eled charge is discussed below Faults were set open

to fluid flow in the course of their main kinematicactivity which is estimated by fault displacementanalysis of the structural restoration models Aftertheir main activity phase faults are assumed to beclosed Fault permeability was also a varying param-eter in the sensitivity analysis discussed belowThermal boundary conditions are given in Table 2Surface temperature ranges between 23 and 30degC(734 and 86degF) through geologic time with apresent-day average temperature of 28degC (824degF)These temperatures at sea level have been corrected

Figure 3 Structural evolution of the Monagas fold and thrust belt demonstrated by structural restoration (A) Passive margin stage(until 23 Ma) (B) Early compression stage (21 Ma) (C) Late compression stage (12 Ma) (D) Present day For details see text

1332 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

Figure 4 Fault displacement from lateMiocene to present day for all faults onrestored cross sections (A) and for the Piritalthrust only (B) Black and white shading andsize of points show magnitude of displacementin meters

NEUMAIER ET AL 1333

Table1

Basin

andPetro

leum

Syste

mModeling(BPSM)Input

ParametersGeologicAges

andLithologies

Compressib

ility

(GPa

-1)

Perm

eability

(logmD)

Thermal

conductivity

(Wm

K)

Heatcapacity

(kcalkgK)

Age

(Ma)

GeologicTime

Facies

Density

(kg∕m

3 )Initial

Porosity(

)max

min

at1

por

at25por

at20degC

at100degC

at20degC

at100degC

0Pleisto

cene

LasP

iedras

sandsto

ne2720

412747

115

minus18

3395

338

02

024

LasPiedrasshale

2700

4040327

403

minus852

minus3

164

169

021

024

26

Pliocene

LasP

iedras

sandsto

ne2720

412747

115

minus18

3395

338

02

024

LasPiedrasshale

2700

4040327

403

minus852

minus3

164

169

021

024

53

upperM

iocene

2La

Pica

sandsto

ne2720

412747

115

minus18

3395

338

02

024

LaPica

shale

2700

4040327

403

minus852

minus3

164

169

021

024

Morich

itoshale

2700

4040327

403

minus852

minus3

164

169

021

024

8upperM

iocene

1La

Pica

sandsto

ne2720

412747

115

minus18

3395

338

02

024

LaPica

shale

2700

4040327

403

minus852

minus3

164

169

021

024

Morich

itobasal

sandsto

ne2715

4812142

187

minus348

15

317

284

02

024

105

middleMiocene

2Carapitasandsto

ne2715

4812142

187

minus348

15

317

284

02

024

Carapitashale

2705

6330932

331

minus684

minus15

204

201

02

024

12middleMiocene

1Carapitashale

2705

6330932

331

minus684

minus15

204

201

02

024

15lower

Miocene

2Carapitashale

2705

6330932

331

minus684

minus15

204

201

02

024

19lower

Miocene

1lower

Carapitashale

2700

7040327

403

minus852

minus3

164

169

021

024

23Oligocene2

Merecureplatform

sandsto

ne2715

4812142

187

minus348

15

317

284

02

024

27Oligocene1

Merecureplatform

sandsto

ne2715

4812142

187

minus348

15

317

284

02

024

1334 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

339

PaleoEocene

Vidono

sandsto

ne2720

412747

115

minus18

3395

338

02

024

Caratasbasin

alshale

2700

7040327

403

minus852

minus3

164

169

021

024

655

upperC

retaceous4

SanJuan

sandsto

ne2720

412747

115

minus18

3395

338

02

024

74upperC

retaceous3

SanAntoniosiltstone

(SOUR

CERO

CK)

2710

5510422

212

minus628

minus1

201

196

023

026

825

upperC

retaceous2

SanAntoniosiltstone

(SOUR

CERO

CK)

2710

5510422

212

minus628

minus1

201

196

023

026

91upperC

retaceous1

Querecualmudsto

ne(SOUR

CERO

CK)

2700

7040327

403

minus852

minus3

164

169

021

024

996

lower

Cretaceous

2undifflower

Cretaceous

(30

Lst50

Sst

20Sh)

2722

459445

141

minus334

174

305

275

02

023

122

lower

Cretaceous

1undifflower

Cretaceous

(30

Lst50

Sst

20Sh)

2722

459445

141

minus334

174

305

275

02

023

1455

sub-Cretaceous

2undiffsub-Cretaceous

(50

Sh50Sst)

2710

5521537

259

minus516

0255

239

02

024

251

NEUMAIER ET AL 1335

by the influence of the paleowater depth using theapproach of Wygrala (1989) Because of the hightopographic elevation of approximately 3500 m(11500 ft) the maximum altitude during the pastcompressional phase we adjusted the model takinginto account a temperature decrease of 6degCkm (33degF1000 ft) related to the paleoaltitude This resultsin a difference of up to 20degC (36degF) compared to sea-level temperature We then assigned the adjustedsedimentwater interface temperatures to the modelas the upper thermal boundary condition At the baseof the model we assigned a heat flow ranging from42 to 80 mW∕m2 through geologic time The highestheat flow is assumed during the rifting event beforethe basin cooled as a passive margin A thermal eventin the Eocene detected by fission track analysis(Locke and Garver 2005 Perez de Armas 2005)has been addressed by a second peak before a secondcooling phase For uncertainty analysis basal paleo-heat flow has been varied within geological meaning-ful ranges and the thermal calibration data has beenhonored as much as possible

Modeling Results

Basin and petroleum system modeling can be dividedinto simulation of the physical parameters within abasin such as pressure and temperature (basin model-ing) and the subsequent evolution of the source rocksand their derived hydrocarbon compounds (petro-leum systems modeling) In the following sectionswe use this subdivision to describe the BPSM resultscompaction modeling thermal modeling and chargehistory It is important to state that these are predic-tions based on forward modeling as just describedThe results are compared and calibrated to actualmeasurements and observations where available

Compaction Modeling

Figure 5 presents an extraction of the modeled resultsat the location of Well Alpha in the Furrial Trend Inthe absence of direct calibration data for pore pres-sure we undertook a sensitivity analysis to test thedifferent scenarios Using the model we tested howpore pressure in the main reservoir of the FurrialTrend reacts to changes in fault and seal-rock per-meability In a first step fault permeability has beenevaluated Faults were assumed open during the mainfault activity phase and closed afterward In additionwe performed several variations from that base casescenario Figure 5A shows the effect of two end-member scenarios on pore pressure within the mainreservoir of the Furrial Trend (Santa Barbara field) inwhich faults have been modeled to be either open orclosed since their formation A third scenario consistsof thrust faults that are open during the compressionalphase (from 15 Ma on) and closed afterward (from105 Ma on) This scenario has been chosen as theldquomaster scenariordquo because it results in overpressureof 23 MPa which is in the range of pressure reportedin similar structures of the Furrial Trend by Schneideret al (2004) In addition this pressure scenariomatches the calibration data for porosity (Figure 5B)

The 2D distribution of modeled overpressure atpresent day is shown in Figure 6A Apart fromimportant overpressure in the Querecual Formationthe allochthon is almost normally pressured Thethrust wedge is overpressured especially in thelow-permeability Carapita Formation Although

Table 2 Basin and Petroleum System Modeling (BPSM) InputParameters Thermal Boundary Conditions

Age (Ma)

SedimentWaterInterface

Temperature (degC)

Age (Ma)

Basal HeatFlow

(mW∕m2)min max

0 28 28 0 4226 28 28 30 6053 76 23 34 708 184 231 50 50105 62 233 107 6012 31 238 1455 8015 5 242 260 6019 5 24723 234 2527 78 235339 94 236655 146 2831455 148 287255 5 239

1336 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

the autochthon is highly overpressured close tonormal hydrostatic pressures are simulated in theforeland basin The discussed model assumption ofclosed faults in the postcompressional phase resultsin a regional pressure compartmentalization

The simulated pressure evolution through geo-logic time can be seen in Figure 7A A time extractionof a source rock cell in the thrust wedge shows a sud-den increase of both lithostatic and hydrostatic pres-sures that is a consequence of the tectonic burial

load associated with the thrusting In addition themodeled pore pressure buildup even though delayedin time can be directly linked to the thrust event Itcan be clearly seen that the timing of fault permeabil-ity directly impacts changes in the present-day porepressure The earlier the faults close the more over-pressure generated by the massive syntectonic sedi-mentation is kept until present day given adequatesealing lithologies are present The present-day over-pressure is modeled as14 MPa when the faults are

Figure 5 Plots of one-dimensional model output and available calibration data for Well Alpha (Table 3) (A) Dotted lines are hydro-static and lithostatic pressures continuous lines are pore pressure the master scenario is in black (a1 faults closing at 105 Ma) andadditional scenarios in gray (a2 open faults a3 closed faults) (B) modeled porosity (C) borehole temperature and (D) Ro The blackline shows the master scenario and the gray line the alternative high-heat-flow scenario Bold lines in the stratigraphic column arethrusts (T) and unconformities (U)

Table 3 Calibration Data for Well Alpha (Porosity Temperature Ro)

TVD (m) Porosity () TVD (m) Corrected Temperature (degC) TVD (m) Vitrinite Reflectance (Ro)

4350 15 600 40 400 0264400 15 1100 53 1200 0484800 10 1800 68 1485 068

4400 131 1700 0394800 133 2900 048

3100 0493400 0494000 054300 0524800 06

NEUMAIER ET AL 1337

open since 8 Ma as 23 MPa for 105 Ma and as37 MPa for 12 Ma Compared to this order of magni-tude changes in the permeability of the overlyingCarapita shale are negligible for the pore pressureTherefore the controlling parameter for pore pressurein the main reservoir of the Furrial trend is assumedfault permeability and resulting lateral connectivityrather than the permeability of the seal rock

Thermal Modeling

Modeled temperature and thermal maturity(Figure 5C D) have been calibrated with availablewell data Some discontinuities in the modeled Ro

trend are noted (Figure 5D) The upper kinks (egbase Pliocene) are related to unconformities withassociated erosion and the lower one (base UpperCretaceous) is due to the well penetrating a reverse

Figure 6 Two-dimensional model at present day with (A) modeled overpressure and (B) modeled thermal maturity and isothermsSource rocks marked by lithology pattern

1338 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

fault displacing mature rocks upon less mature ones(see Figure 6B for 2D view) There is a mismatchbetween the modeled Ro trend and the data in thedeeper part of the well In the Oligocene rocks themodeled temperature is too low (2degC or 36degF)whereas the modeled Ro is too high (up to 02 Ro)Any attempt to reduce the modeled Ro (eg bychanging thermal conductivity of the MioceneCarapita Formation) also results in a lower modeledpresent-day temperature Therefore the thermal sce-nario used for this calibration is a trade-off betweenmatching present-day temperature and Ro (using thekinetics of Sweeney and Burnham 1990) The basalheat flow assumed in the master scenario (Table 2)results in a modeled present-day surface heat flowranging from 37 mW∕m2 in the southern part of thethrusts to 52 mW∕m2 in parts of the uplifted areaAt the same depth the allochthonous northern partis thermally more mature than the southern autochthon(Figure 6B) because of the recent uplift Locallyinverted trends around the thrusts can be observedfor example at the location of Well Alpha (seeFigure 5D) In the autochthon the Upper Cretaceoussource rocks are entirely in the oil or even gas windowat present day because of the Neogene tectonic loadingand synorogenic deposition In the northern alloch-thon most of the Upper Cretaceous source rocks havebeen in the oil window in the past The modeled ther-mal maturity is in good agreement with the regionalmaturity suggested by Summa et al (2003)

Similar to the pressures the thrusting also affectsthe modeled temperature and Ro In the thrust wedgea direct relationship exists between the thrusting andthe increase of temperature and maturity The thrust-ing is the trigger for source rocks entering the oil win-dow within the thrust wedge (Figure 7B) and thepresent-day thermal maturity has been reached sincethe last 10 Ma Although this recent thermal historycan be considered as relatively well constrained bythe measured Ro in the calibration well this doesnot apply to the past In this thrusted area any ther-mal maturity reached prior to the thrusting onset hasbeen ldquooverprintedrdquo and therefore no thermal calibra-tion is possible for this period However theCretaceous source rocks were not buried deeply dur-ing the passive margin phase and even very hot sce-narios (basal heat flow of 100 mW∕m2) do not

significantly change their maturity (Figure 7B) Inthe uplifted allochthon several wells show low(03) to moderate (1) measured Ro (Parra et al2011) Because the allochthon experienced its maxi-mum burial and the highest temperatures during thepassive margin phase these maturity levels can beassumed as representative of the maturity prior tothe compression In contrast to the thrust wedge theconsequence of the thrusting is inverse for the alloch-thon (Figure 7C) as the hanging wall is uplifted andthe accompanying erosion removes the overburdenThe highest temperatures are simulated to have beenreached just before the thrusting event Then temper-ature decreased and Ro stagnated leaving the sourcerocks in the early mature zone until present day Theassumption of a regionally relatively homogeneousthermal field during the cooling phase of the passivemargin in combination with the relatively low matur-ities measured in the uplifted allochthon makes thehot heat-flow scenario unlikely However a laterallyvarying heat flow cannot be excluded and the conse-quently hotter paleoheat flow would have led to muchhigher temperatures prior to compression resultingalso in a higher present-day maturity level (14)despite the uplift (Figure 7C) Any variations of basalheat flow during the rift period (Late Jurassic to EarlyCretaceous) occurred too early to affect the maturityof the source rocks deposited during the lateCretaceous

Charge History

After calibration of the basinrsquos physical environ-ment the evolution of the petroleum systems wasmodeled Hydrocarbon generation was simulatedthrough geologic time using the activation energydistribution of the assigned source rock kineticsFrom the moment of hydrocarbon expulsiononward secondary migration was modeled usingthe invasion percolation method (Wilkinson 1984)It takes into account the geometry overpressure gra-dient and capillary entry pressures of the rockswhich are opposed to the buoyancy of the migratingoil and gas Figure 8 shows the time evolution of thetotal study area and Figure 9 focuses on the thrustwedge As already mentioned the described chargehistory is based on forward modeling It is consistent

NEUMAIER ET AL 1339

Figure 7 Time extractions (see Figure 6 for location) (A) Pressure development through geologic time for a source rock in the ver-tical continuity of Well Alpha Dotted lines are hydrostatic and lithostatic pressures continuous lines are pore pressure the master sce-nario is in black (a1 faults closing at 105 Ma) and additional scenarios in gray (a2 open faults a3 closed faults a4 faults closing at12 Ma a5 faults closing at 8 Ma) (B) Temperature (continuous lines) and Ro (dotted lines) through geologic time for a source rockin the vertical continuity of Well Alpha the black line shows the master scenario and the gray line the alternative high-heat-flow sce-nario (C) Temperature (continuous lines) and Ro (dotted lines) through geologic time for a source rock in the uplifted allochthonthe black line shows the master scenario and the gray line the alternative high heat flow scenario

1340 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

with the modeled compaction and thermal historyand therefore shows one scenario that explains thecharge of the Santa Barbara field and the forelandbasin Simplifications are inherent to any such work-flow and shall be considered when analyzing themodeling results

Passive Margin PhasemdashMesozoic toOligocene

Modeling results indicate that already during the LateCretaceous any Jurassic source rocks potentiallydeposited in the northern part of the section (Summaet al 2003) have reached temperatures of more than200degC (392degF) and corresponding high levels of matu-rity However the temperature at the base of theQuerecual Formation (Upper Cretaceous) was lessthan 100degC (212degF) leaving the kerogen immatureThe source rock became effective only during lateEocene with temperatures of up to 120degC (248degF)and a (reactive kerogen to hydrocarbon) transforma-tion ratio (TR) of more than 20 in the deeper distalpart of the passive margin Oil was quickly expelledmigrated vertically and seeped to the surface becauseof the general lack of seals in the overburden Duringthe Oligocene the time of deposition of the Merecurereservoir sandstones in the southern platform this sit-uation did not change apart from increased transforma-tion in the deeper basinal part of the passive margin

First oil accumulations are modeled in the Vidontildeosandstones in the basinal part in the late Oligocene(Figure 8A) in which the base of the QuerecualFormation shows TRs of up to 60 at around 130degC(266degF) In the proximal part there is no hydrocarbongeneration Although there is a general northwardtilting slope no southward migration from the distalmargin is modeled this has been checked usingdifferent migration methods including Darcy flowwith scenarios of continuous intrasource carrier bedsThis is in contrast to the studies from Gallango andParnaud (1995) and Schneider (2003) who suggestedlateral migration at that stage The difference couldbe explained by the presence of a regional top seal intheir models However our model suggests that theshallow rocks do not have sufficient sealing capacityto allow lateral hydrocarbon migration During thepassive-margin phase the basin was normally

pressured (Figure 7A) and simulated stresses havenot yet resulted in any fracturing at that stage

Main Compressional PhasemdashEarly to MiddleMiocene

With the onset of compression during early Miocene(Figure 8B) both the Pirital and Furrial thrusts wereformed and the northern distal part of the passivemargin was rapidly uplifted Important overpressuresas well as mechanical rock failure both the result ofthe rapid uplift are simulated in the allochthonDespite this uplift the kerogen-to-hydrocarbon trans-formation continued and the rate even increasedlocally close to the footwall cutoffs and where syn-orogenic deposition provided rapid overburden load-ing Oil accumulations are modeled in the freshlycreated anticlines

The tectonicndashsedimentary wedge between thetwo main thrusts experienced very rapid burial Infact the Upper Cretaceous source rocks were broughtinto depths where temperatures were as high as185degC (365degF) at 19 Ma and 225degC (437degF) at15 Ma Especially close to the cutoff of the Piritalthrust they very quickly reached a high maturity at19 Ma and reached full transformation in most placesat 12 Ma The expelled hydrocarbons migrated intothe sandstones of the Merecure Formation which weresealed by the Carapita Formation The charge toward aseries of recent anticlines is modeled by the fill-to-spillmechanism from north to south (Figure 9C D) Afterthe first oil fill the northern prospect quickly receiveda local gas charge the central structure containedmainly oil Note the presence of a lateral seal that isdue to sandndashshale juxtaposition The southernmoststructure which forms the Santa Barbara field todaywas charged by oil only at 15 Ma (Figure 9D)

The generated hydrocarbons in the southernautochthon were leaking to the surface until theCarapita Formation was compacted enough to holdsome of them (Figure 8C) At 12 Ma the wedgesource rocks were mostly spent but expulsion wasstill ongoing Important rock failure is modeled tohave occurred in the thrusted footwall breaking theformerly sealing Carapita Formation (Figure 10)This is in good agreement with cemented hydraulicfractures observed in a few cores in the Carapita

NEUMAIER ET AL 1341

Figure 8 Two-dimensional model through geologic time (selected time steps only) with modeled transformation ratio for the sourcerock layers isotherms and liquid and vapor migration and accumulation

1342 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

shales from the Furrial Trend (Roure et al 2005)This important fracturing event resulted in thedestruction of the modeled accumulation (Figure 9E)Hydrocarbon generation and expulsion were alsohappening in the southern foreland basin at transfor-mation ratios of up to 30

Late Compressional and PostcompressionalPhasemdashLate Miocene to Present Day

The late compressional phase was marked by the par-tial destruction of the allochthonous petroleum systemsby erosion (Figure 8E F) Many traps disappeared and

Figure 8 Continued

NEUMAIER ET AL 1343

the hydrocarbon generation within the source rocksstopped The erosion triggered an important sedimenttransfer toward the south contributing to further burialand increased temperatures in the thrust wedge and theforeland basin During late Miocene the petroleumsystem in the wedge matured An important northwardtilting was triggered by flexural tectonic loading andthe loading of syntectonic sediments (Summa et al2003) In combination with the presence of an effec-tive regional top seal the compacted CarapitaFormation allowed a change from dominantly verticalmigration into lateral migration First accumulationsare modeled in the Neogene synorogenic sequence

reservoirs in the south and important southwardmigration toward the foreland basin started(Figure 8G) These trends continued through Plioceneand Pleistocene times to present day filling theOnado field (Figure 8H) and other known fields inthe south (Figure 1A) of the Eastern Venezuela Basin

Prospectivity

We modeled the charge of the Santa Barbara field inthe Furrial Trend and of the Onado field in the fore-land basin (Figure 8H) Other prospects in the thrustwedge were also investigated The structure called

Figure 9 Detail of two-dimensionalmodel through geologic time (seeFigure 8) showing modeled transforma-tion ratio for the source rock layers iso-therms and liquid and vapor migrationand accumulation The Santa Barbara fieldand the prospect are shown by ellipses

1344 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

ldquoprospectrdquo (Figure 9G) looks similar to that of theSanta Barbara field because both are anticlines andthe reservoir and seal formations are the sameHowever they differ in several aspects First theprospect is in the vicinity of the Pirital thrust whichpresents a high risk of impacting the seal integrityThe prospect location is directly below the area wherewe presume the maximum of late tectonic activityoccurred (Figure 4) As modeled the CarapitaFormation the seal of the prospect was already brit-tle when this deformation occurred Calculatedstresses also predict rock failure from 12 Ma onward(Figure 10) Even though charge continued ourmodel suggests continuous leakage of the structure(Figure 9EndashG) The second major difference is thatthe source rock in the fault block of the prospect ismodeled to have already started generating and expel-ling hydrocarbons during the Oligocene and earlyMiocene (Figure 9A) before the trap formed and the

seal became effective (Figures 9C 10) Even thoughgeneration expulsion and migration continued aftertrap formation the prospect was not charged withthe earlier expelled oil In addition the present-daymaturity of the source rocks within the prospectrsquosfault block is higher than in the source rocks withinthe tectonic block of the Santa Barbara field poten-tially indicating expulsion of more gas A southndashnorth trend can be seen on the modeled accumulationcomposition with more gas toward the allochthon(Figure 9) The modeled composition of the prospectprior to seal failure was much richer in gas than inthe neighboring Santa Barbara field (Figure 9D) Infact simulation scenarios with unbreached sealsresulted in accumulation of predominantly gas Thecombination of both seal integrity and charge compo-sition concerns makes this a risky prospect In addi-tion potential diagenetic fluids related to the faultfluid flow might have altered the reservoir quality

Figure 10 Two-dimensional model (at 12 Ma) with modeled stress to failure Mohr circles (with σ standing for normal stress and τfor shear stress) and petroleum system events charts for the Santa Barbara field (left) and the prospect (right) Model predicts rock fail-ure in the prospect from about 12 Ma For further details see text

NEUMAIER ET AL 1345

Apart from the Onado oil field only minor accu-mulations are modeled in the southern Neogene syn-orogenic sequence The proportion of gas is higherthan in the Furrial trend The accumulations at thesouthern model edge are a modeling artifact repre-senting hydrocarbons (mainly oil) that in reality arethought to have charged the southern foreland basin(from mid to late Miocene times to present day)However parts of the migrated oil and gas could havebeen trapped in the Upper Cretaceous San JuanFormation south of the horst structure

The potential for conventional hydrocarbon accu-mulations in the allochthon is very limited by theimportant fracturing that occurred during the upliftperiod The basal sandstones of the Morochito piggy-back basin north of the Pirital thrust are not chargedin our 2D model because of the absence of structuraltraps Nevertheless there is a potential for strati-graphic traps to have been charged as migrationthrough the basin occurred However seal strengthmight be critical at this low-compaction stage Thesame applies for the Las Piedras Formation We thinkthere might be potential for shale oil because theQuerecual Formation is at shallow depths but showsmaturity within the oil window From the thermalmaturity point of view any potential Jurassic sourcerocks as discussed by Summa et al (2003) mightbe a gas source or shale gas target

UNCERTAINTY ANALYSIS

There is often significant uncertainty in basin geom-etry and depth of stratigraphic horizons based onuncertain seismic interpretation and velocity models(Kukla et al 2000 Brandes et al 2008 Baur et al2010) Restoring of present-day geometries to obtainpaleogeometries adds further uncertainty Additionaluncertainty is related to thermal boundary conditionssuch as paleoheat flow (Beha et al 2008) and togeochemical properties such as source rock character-istics and petroleum generation kinetics and calcula-tion of biodegradation and petroleum density(Blumenstein et al 2008) In addition to the porepressure and thermal sensitivity analysis justdescribed we varied initial TOC within a range of2 to 10 A variation of initial TOC obviously

directly scales the amount of initial kerogen availablefor transformation into hydrocarbons Neither thecharge volume nor composition of the Santa Barbarafield nor the wedge prospects are significantly sensi-tive to variation of initial TOC these traps are filledto spill in each scenario However the amount ofhydrocarbons charging the foreland basin (eg theOnado field) is reduced in the case of low initialTOC A high initial TOC only increased the modeledamount of surface seepage The sensitivity of the ini-tial HI is analogous to the sensitivity of the ini-tial TOC

In addition to the more classical uncertaintyand sensitivity analysis described we focused onquantitative analysis of the decompaction uncer-tainty when performing structural restoration Thisuncertainty is particular to the applied workflowof using structurally restored paleogeometries inBPSM Note that this uncertainty is significantlylower when using the backstripping method withthe optimization algorithms discussed One methodto analyze the adequacy of the decompactionmethod is comparison of the output rock matrixmass balance In the backstripping approach therock matrix mass is kept constant of a model cellwith a simulated decrease in porosity controlledby compaction laws and the corresponding reduc-tion of the layer thickness When using fixed pale-ogeometries from structural restoration the cellthickness is imposed However the decrease inporosity is calculated and the mass is adapted tofit the imposed cell thickness That means that therock matrix mass is constant only if ideal decom-paction has been applied

Figure 11A shows the percentage variation fromthe present-day matrix mass (100) of a given sourcerock cell through geologic time The dotted line rep-resents the ideal case with constant rock matrix massWe used two end-member decompaction scenarios todescribe the decompaction-related uncertainty that isintroduced when fixed paleogeometries replace back-stripping idealized hydrostatic decompaction assum-ing no overpressure and idealized lithostaticdecompaction assuming no effective stress (nodecompaction resulting in constant cell thicknessthrough time) In the latter case with decreasingporosity during compaction the rock matrix mass

1346 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

increases until it reaches the present-day mass Inother words rock mass matrix has been highly under-estimated in the past (in this particular case by morethan 20) This is a source rock so the kerogen massdecreases by the same factor This yields an underes-timation of more than 20 in TOC Note the rockmatrix mass is constant for time steps prior to23 Ma in which backstripping optimization couldbe performed toward the earliest paleogeometry Forthe second end-member scenario of idealized hydro-static decompaction the rock matrix mass is nearlyconstant which is a good decompaction adequacy

indicator for that given cell in the model Howeverthe so-measured adequacy changes throughout themodel In some areas the rock matrix mass has beenunderestimated by about 50 for nondecompactedpaleogeometries however in the case of hydrostaticdecompaction values generally range from 80 to120 of the present-day rock matrix mass Thatmeans in this particular case that hydrostatic decom-paction is more adequate (or less wrong) than nodecompaction at all The decompaction adequacycan be roughly correlated with pressure changesthrough geologic time

Figure 11 Time extractions for comparison of two end-member decompaction scenarios (see Figure 6 for location) (A) Rock matrixmass conservation through geologic time (B) Source rock burial depth through geologic time (C) Temperature and transformation ratiothrough geologic time

NEUMAIER ET AL 1347

Inadequate decompaction results in wrong depthfor stratigraphic horizons in the paleogeometriesFigure 11B shows the burial depth of the same inves-tigated source rock cell We can see that during thepassive margin phase there is a slight difference inpaleoburial depth for the two end-member scenariosDuring the compressional phase the differencechanges dramatically reaching more than 1000 m(3281 ft) in some places Toward present day bothcurves meet again because the input to the two resto-ration scenarios was the same (identical present-dayburial depth) This difference in compaction calcula-tion has severe implications on the modeled tempera-ture history and the kerogen-to-hydrocarbontransformation (Figure 11C) For an identical thermalframework in terms of modeling input parameterstemperature differences of up to 25degC (45degF) resultin TR differences of more than 20 in the early com-pressional phase before TR values meet again for thepresent-day situation

CONCLUSIONS

Structurally complex areas are attracting increasinginterest in hydrocarbon exploration Exploration riskcan be substantially narrowed by analyzing the his-tory of the basin and its petroleum systems based onstructural evolution We presented a further progressin technology that allows the integration of structuralrestoration and BPSM This combined analysis isseen as a way forward to improve modeling in thrustbelts with stacked stratigraphic sequences Despitelimitations of nonoptimized decompaction whichare added to the general uncertainty related to theindividual methods we modeled the charge and sealhistory of the Monagas fold and thrust belt inVenezuela Calibrated porosity temperature andthermal maturity as well as basin-scale stress pro-vided the regional framework through geologic timein which the petroleum systems evolved such thatthe charge of the known Santa Barbara and Onadofields and other trends could be understood We dis-cussed seal integrity of a potential exploration pros-pect by analyzing structural restoration-derived faultdisplacement through geologic time and an advancedstress forward simulation The uncertainty of using

decompacted or nondecompacted structural restora-tion has been assessed It can have important influ-ence on rock matrix mass balance through geologictime paleogeometry and paleoburial depth andtherefore affect paleotemperatures and paleomatur-ities We conclude that decompaction assuminghydrostatic pressure is the best option in that particu-lar case

REFERENCES CITED

Alberdi M and E Lafargue 1993 Vertical variations of organicmatter content in Guayuta Group (Upper Cretaceous)interior mountain belt eastern Venezuela OrganicGeochemistry v 20 p 425ndash436 doi 1010160146-6380(93)90091-O

Athy L F 1930 Density porosity and compaction of sedimen-tary rocks AAPG Bulletin v 14 p 1ndash24

Aymard R L Pimentel P Eitz P Lopez A ChaouchJ Navarro J Mijares and J G Pereira 1990 Geologic inte-gration and evaluation of Northern Monagas EasternVenezuelan Basin in J Brooks ed Classic petroleum prov-inces The Geologic Society (London) Special Publicationv 50 p 37ndash53

Baur F M Di Benedetto T Fuchs C Lampe andS Sciamanna 2009 Integrating structural geology andpetroleum systems modelingmdashA pilot project from Boliviarsquosfold and thrust belt Marine and Petroleum Geology v 26no 4 p 573ndash579 doi 101016jmarpetgeo200901004

Baur F R Littke H Wielens C Lampe and T Fuchs 2010Basin modeling meets rift analysismdashA numerical modelingstudy from the Jeanne drsquoArc basin offshore NewfoundlandCanada Marine and Petroleum Geology v 27 p 585ndash599doi 101016jmarpetgeo200906003

Beha A R O Thomsen and R Littke 2008 Thermal historyhydrocarbon generation and migration in the Horn Grabenin the Danish North Sea A 2D basin modelling studyInternational Journal of Earth Sciences v 97 p 1087ndash1100doi 101007s00531-007-0247-2

Blumenstein I O B M Krooss R di Primio W RottkeE Muumlller C Westerlage and R Littke 2008Biodegradation in numerical basin modelling A case studyfrom the Gifhorn Trough N-Germany International Journalof Earth Sciences v 97 p 1115ndash1129 doi 101007s00531-007-0272-1

Brandes C A Astorga R Littke and J Winsemann 2008Basin modelling of the Limoacuten back-arc basin (Costa Rica)Burial history and temperature evolution of an island arc-related basin-system Basin Research v 20 p 119ndash142doi 101111bre200820issue-1

Carnevali J O 1988 El Furrial oil field northeasternVenezuela First giant in foreland fold and thrust belts ofwestern hemisphere AAPG Bulletin v 72 68 p

Chevalier Y 1987 Contribution a lrsquoetude geacuteologique de lafrontiereacute Sud-Est de la plaque Caraibe Les zones internesde la chaine Sud-Caraibe sur le transect ile de Margarita-

1348 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

(Talukdar et al 1988) Summa et al (2003) pub-lished multiple 1D models over a large area andextrapolated the results onto regional maturity mapsfor the Guayuta Group and Cenozoic source rocksThese maps include a hanging-wall cutoff and there-fore do not represent the higher maturities found inthe thrust wedges Elsewhere the authors suggestedthat most source rocks have been in the oil windowin the past A present-day drainage area analysis ofthe top Cretaceous structure map was used to describethe southward charge from the deeper kitchen into theforeland basin This regional migration modelexplains oil provenance and quality Important timingaspects such as the relationship between the struc-tural history and source rock maturation and expul-sion are described using a regional structuralrestoration in conjunction with the 1D models Parraet al (2011) included the thrusting effect by using a15D module mimicking lateral rock transportationthat is due to thrusting Gallango and Parnaud(1995) performed 2D modeling including hydrocar-bon migration in the unthrusted southern area Theyalso described a second model within the thrustedarea that simulates the basin history prior to compres-sion (from the Mesozoic to early Miocene) Thismodel proposes long-distance petroleum migrationsouthward before the transpressional event whichresulted in the observed structural complexity Inother papers Roure et al (2003) Schneider (2003)and Schneider et al (2004) described fluid flow porefluid history and diagenetic processes in a 2D modelalong a cross section parallel to the location of themodel presented here

MODELING METHODS

To address the charge and seal history of the SantaBarbara transect (Figure 1) in the western Monagasfold and thrust belt we applied a combined approachusing BPSM and structural restoration A BPSM is adynamic model of the subbasinrsquos physical processessuch as pore pressure and temperature evolution andthe development of the petroleum systems includinghydrocarbon generation expulsion migration accu-mulation and preservation (Hantschel and Kauerauf2009) The BPSM is performed in 1D (on a well orpseudowell) in 2D (usually along a seismic section)

and in 3D The BPSM can also provide predictionsabout seal capacity and integrity especially if fielddata are available for calibration

In contrast to a static model which consistsonly of geometry and properties for present day(as observed) a dynamic earth model includesthe evolution of geometry and properties Thepaleogeometrymdashor structural geometry of the modelat a given geologic age (or simulation time step)mdashisstrongly dependent on the paleowater depth (geomet-rical boundary condition) faulting activity the lithol-ogy of the various layers and their associatedcompaction behavior Approaching accurate paleoge-ometry is crucial for achieving consistent resultsespecially for fluid flow simulations In standardBPSM the paleogeometries are automatically gener-ated using a simplified restoration approach knownas backstripping (Figure 2A B)

Geomechanically based structural restorationuses the present-day interpreted and depth-convertedgeometry of a sedimentary basin or subbasin for therestoration of the basin structure backward in geo-logic time (Figure 2C D) This is done by sequen-tially removing the syntectonic sedimentary layersfrom the youngest to the oldest (the reverse of deposi-tion) allowing unfolding and unfaulting during therestoration (the reverse of deformation) In contrastto the simple backstripping approach structural resto-ration allows much more control through additionalgeometrical boundary conditions such as lateralmovement of rocks along faults or bedding slipRecent development has resulted in geomechanicalrestoration algorithms extending the pure geometri-cal approach by applying physical laws of rockdeformation (Maerten and Maerten 2006) The paleo-geometries obtained by structural restoration are thenused to describe basin geometry through time in apetroleum systems model

The backstripping approach is applied to tectoni-cally undisturbed areas and to areas where faultingoccurred much earlier than the main phase of hydro-carbon generation (typically a sag basin or the postriftsequence of a passive margin Figure 2A) In basinswhere extensional faulting occurred and where faultslip has an important horizontal component (eg lis-tric faults) backstripped paleogeometries showldquospikyrdquo artifacts (Figure 2B) these paleogeometries

NEUMAIER ET AL 1329

are generally not good enough for predicting prefault-ing hydrocarbon migration but might be acceptablefor hydrocarbon generation modeling Here structur-ally restored paleogeometries are important for thelayer shape in the proximity of faults and whencross-fault connectivity (like sandndashsand overlaps)matters (Figure 2C) Alternatively manual correctionof paleothickness can be done In halokinetic settingsbackstripping is generally superimposed by simplesalt reconstruction workflows (area balancing andextrapolation) Structural restoration should also beconsidered if salt movement results in geometries thatare more complex In any case thermal modeling inthe vicinity of salt is difficult due to its extremelyhigh thermal conductivity In areas with overthrustingor overturned folds (Figure 2D) with multiple strati-graphic sequences typically a fold and thrust beltbackstripping fails completely because lateral rockmovements are missing In the latter case paleo-geometries need to be generated by means other thanbackstripping (from simple conceptual drawing totrue structural restoration)

Paleogeometries generated from structuralrestorations can be loaded into BPSM software as analternative to performing a calculation from back-stripping Meshing is done section per section (timestep per time step) and tracking of cells from one sec-tion to the other is following rules described inHantschel and Kauerauf (2009) For the ages corre-sponding to the paleosection the basin geometry is

fixed it is not a function of compaction-controlledforward modeling as is the case when using back-stripping The BPSM simulation is restricted to thecalculation of properties such as porosity overpres-sure temperature vitrinite reflectance (Ro) andsource rock maturity which populate the fixed paleo-geometries as well as hydrocarbon migration Thesecompaction-controlled results are not used to adjustthe depositional amounts and to correct the paleoge-ometry as done in the backstripping approach andconsequently geometry optimization cannot beperformed The possible difference between thecompaction-law-controlled porosity change and thepredefined geometry change is indirectly compen-sated by the corresponding adjustment of the solidgrain mass

When paleogeometries are used for basin analy-sis and BPSM some factors become very importantRestoration must focus toward a best estimate paleo-water depth because changes in slope directions andtopography can largely affect fluid flow Erosionshould be estimated and reconstructed and manygeologic events need to be taken into account toincrease the resolution through time for dynamicmodeling The sediment thickness must be correctedin geologic time for decompaction All these factorswhich are common sources of uncertainty are criticalfor a geologically meaningful restoration

Because of the optimization method justdescribed adequately decompacted paleogeometries

Figure 2 Steps for (A B) backstripping and (C D) structural restoration of geometries of different structural complexity For detailssee text

1330 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

are required from structural restoration Inadequatedecompaction can produce large errors resulting inan unbalanced mass of solid rock matrix underesti-mated source rock paleoburial depth and deviatedpaleomigration paths related to differential compac-tion However it is difficult to define adequatedecompaction As a rule of thumb in normally pres-sured clastic basins where the main compactiondriver is linearly increasing effective stress simplehydrostatic decompaction laws such as thosedescribed by Athy (1930) can be used for restorationIn cases of abnormal pressure (overpressure) hydro-static decompaction overestimates the effectivestress or underestimates porosity and might fail tothe same degree as restoration neglecting decompac-tion One might be scaling normal decompactioncurves wherever regional overpressure occurs andhas occurred In basins with complex erosion pres-sure and diagenetic histories optimization loops(using BPSM simulated properties for optimizeddecompaction in second restoration) might benecessary Any sort of decompaction in a carbonate-dominated environment where compaction is gener-ally less understood because of the effect of mineraltransformations and cementation is subject to muchlarger uncertainty The question of whether to usehydrostatically decompacted or nondecompactedstructurally restored paleogeometries is discussed inthe section on uncertainty analysis

MODELING

Prior to structural restoration we performed several1D simulations These generated erosion and paleoal-titude estimations for structural restoration andprovided a calibration for the 2D model One-dimensional modeling also gave the first informationon hydrocarbon generation and timing to identifythe period of first source rock maturation (ldquocriticalmomentrdquo as defined by Magoon and Dow 1994)This is very important because prior to structuralrestoration its identification indicates when addi-tional time steps are needed to increase time resolu-tion for critical time intervals so that the activepetroleum system can be dynamically modeled in anadequate way However for simulation regular time

steps are needed not only during the critical momentbut also for the entire basin history

Structural Restoration

The following section describes our structural resto-ration results shown in Figure 3 of the post-riftstructural evolution of the Monagas fold and thrustbelt It can be divided into three main steps(1) From the Mesozoic to Oligocene the Monagasarea formed parts of the northern passive margin ofthe South American shield (Figure 3A) The sub-sidence driver was mostly thermal cooling and iso-static load with only a few normal faults Thesediments were sourced by the craton in the south(2) In the early Miocene the margin experiencedcompression from the transcollision with theCaribbean plate (Figure 3B) Two major normalfaults that formed during rifting were inverted in thecourse of the collision generating the regionalFurrial and Pirital thrusts Synorogenic deposits werecovered by the thrusted allochthon and intramontanepiggyback basins were formed With increasing upliftand erosion (middle Miocene Figure 3C) their sedi-ment fills were redeposited to form the important syn-tectonic Oficina and Carapita Formations south of thethrusts (3) From late Miocene on the late compres-sional phase started Figure 3D shows the present-day structure of the Monagas fold and thrust belt(2D model crossing the Santa Barbara field)According to our restoration the total shortening onthe section accommodated by the thrusts during thecompression is around 40 km (2486 mi) which cor-responds to 35 It is important to note that the Piritalthrust was active until late Miocene as indicated bythe fault displacement analysis (Figure 4) These latetectonic movements might have an important effecton the seal integrity of prospects that are in the vicin-ity of the thrust

Model Input Data

Table 1 shows geologic ages lithologies and petro-physical parameters that were assigned to our 2Dmodel For the main source rocks (Querecual andSan Antonio Formations) a type II B kinetic wasassigned (Pepper and Corvi 1995) As a base case

NEUMAIER ET AL 1331

we assumed a constant initial TOC of 6 wt and aninitial HI of 500 mg HCg TOC However becauseinitial TOC and HI were estimated from present-daymeasured values these initial values are highly uncer-tain In addition there is both lateral and vertical vari-ability within the layer even though locally very highsource rock potential is likely to occur a bulk TOCand HI need to be assigned to the model cells Forthese reasons several models with different initialTOC have been run and their sensitivity to the mod-eled charge is discussed below Faults were set open

to fluid flow in the course of their main kinematicactivity which is estimated by fault displacementanalysis of the structural restoration models Aftertheir main activity phase faults are assumed to beclosed Fault permeability was also a varying param-eter in the sensitivity analysis discussed belowThermal boundary conditions are given in Table 2Surface temperature ranges between 23 and 30degC(734 and 86degF) through geologic time with apresent-day average temperature of 28degC (824degF)These temperatures at sea level have been corrected

Figure 3 Structural evolution of the Monagas fold and thrust belt demonstrated by structural restoration (A) Passive margin stage(until 23 Ma) (B) Early compression stage (21 Ma) (C) Late compression stage (12 Ma) (D) Present day For details see text

1332 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

Figure 4 Fault displacement from lateMiocene to present day for all faults onrestored cross sections (A) and for the Piritalthrust only (B) Black and white shading andsize of points show magnitude of displacementin meters

NEUMAIER ET AL 1333

Table1

Basin

andPetro

leum

Syste

mModeling(BPSM)Input

ParametersGeologicAges

andLithologies

Compressib

ility

(GPa

-1)

Perm

eability

(logmD)

Thermal

conductivity

(Wm

K)

Heatcapacity

(kcalkgK)

Age

(Ma)

GeologicTime

Facies

Density

(kg∕m

3 )Initial

Porosity(

)max

min

at1

por

at25por

at20degC

at100degC

at20degC

at100degC

0Pleisto

cene

LasP

iedras

sandsto

ne2720

412747

115

minus18

3395

338

02

024

LasPiedrasshale

2700

4040327

403

minus852

minus3

164

169

021

024

26

Pliocene

LasP

iedras

sandsto

ne2720

412747

115

minus18

3395

338

02

024

LasPiedrasshale

2700

4040327

403

minus852

minus3

164

169

021

024

53

upperM

iocene

2La

Pica

sandsto

ne2720

412747

115

minus18

3395

338

02

024

LaPica

shale

2700

4040327

403

minus852

minus3

164

169

021

024

Morich

itoshale

2700

4040327

403

minus852

minus3

164

169

021

024

8upperM

iocene

1La

Pica

sandsto

ne2720

412747

115

minus18

3395

338

02

024

LaPica

shale

2700

4040327

403

minus852

minus3

164

169

021

024

Morich

itobasal

sandsto

ne2715

4812142

187

minus348

15

317

284

02

024

105

middleMiocene

2Carapitasandsto

ne2715

4812142

187

minus348

15

317

284

02

024

Carapitashale

2705

6330932

331

minus684

minus15

204

201

02

024

12middleMiocene

1Carapitashale

2705

6330932

331

minus684

minus15

204

201

02

024

15lower

Miocene

2Carapitashale

2705

6330932

331

minus684

minus15

204

201

02

024

19lower

Miocene

1lower

Carapitashale

2700

7040327

403

minus852

minus3

164

169

021

024

23Oligocene2

Merecureplatform

sandsto

ne2715

4812142

187

minus348

15

317

284

02

024

27Oligocene1

Merecureplatform

sandsto

ne2715

4812142

187

minus348

15

317

284

02

024

1334 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

339

PaleoEocene

Vidono

sandsto

ne2720

412747

115

minus18

3395

338

02

024

Caratasbasin

alshale

2700

7040327

403

minus852

minus3

164

169

021

024

655

upperC

retaceous4

SanJuan

sandsto

ne2720

412747

115

minus18

3395

338

02

024

74upperC

retaceous3

SanAntoniosiltstone

(SOUR

CERO

CK)

2710

5510422

212

minus628

minus1

201

196

023

026

825

upperC

retaceous2

SanAntoniosiltstone

(SOUR

CERO

CK)

2710

5510422

212

minus628

minus1

201

196

023

026

91upperC

retaceous1

Querecualmudsto

ne(SOUR

CERO

CK)

2700

7040327

403

minus852

minus3

164

169

021

024

996

lower

Cretaceous

2undifflower

Cretaceous

(30

Lst50

Sst

20Sh)

2722

459445

141

minus334

174

305

275

02

023

122

lower

Cretaceous

1undifflower

Cretaceous

(30

Lst50

Sst

20Sh)

2722

459445

141

minus334

174

305

275

02

023

1455

sub-Cretaceous

2undiffsub-Cretaceous

(50

Sh50Sst)

2710

5521537

259

minus516

0255

239

02

024

251

NEUMAIER ET AL 1335

by the influence of the paleowater depth using theapproach of Wygrala (1989) Because of the hightopographic elevation of approximately 3500 m(11500 ft) the maximum altitude during the pastcompressional phase we adjusted the model takinginto account a temperature decrease of 6degCkm (33degF1000 ft) related to the paleoaltitude This resultsin a difference of up to 20degC (36degF) compared to sea-level temperature We then assigned the adjustedsedimentwater interface temperatures to the modelas the upper thermal boundary condition At the baseof the model we assigned a heat flow ranging from42 to 80 mW∕m2 through geologic time The highestheat flow is assumed during the rifting event beforethe basin cooled as a passive margin A thermal eventin the Eocene detected by fission track analysis(Locke and Garver 2005 Perez de Armas 2005)has been addressed by a second peak before a secondcooling phase For uncertainty analysis basal paleo-heat flow has been varied within geological meaning-ful ranges and the thermal calibration data has beenhonored as much as possible

Modeling Results

Basin and petroleum system modeling can be dividedinto simulation of the physical parameters within abasin such as pressure and temperature (basin model-ing) and the subsequent evolution of the source rocksand their derived hydrocarbon compounds (petro-leum systems modeling) In the following sectionswe use this subdivision to describe the BPSM resultscompaction modeling thermal modeling and chargehistory It is important to state that these are predic-tions based on forward modeling as just describedThe results are compared and calibrated to actualmeasurements and observations where available

Compaction Modeling

Figure 5 presents an extraction of the modeled resultsat the location of Well Alpha in the Furrial Trend Inthe absence of direct calibration data for pore pres-sure we undertook a sensitivity analysis to test thedifferent scenarios Using the model we tested howpore pressure in the main reservoir of the FurrialTrend reacts to changes in fault and seal-rock per-meability In a first step fault permeability has beenevaluated Faults were assumed open during the mainfault activity phase and closed afterward In additionwe performed several variations from that base casescenario Figure 5A shows the effect of two end-member scenarios on pore pressure within the mainreservoir of the Furrial Trend (Santa Barbara field) inwhich faults have been modeled to be either open orclosed since their formation A third scenario consistsof thrust faults that are open during the compressionalphase (from 15 Ma on) and closed afterward (from105 Ma on) This scenario has been chosen as theldquomaster scenariordquo because it results in overpressureof 23 MPa which is in the range of pressure reportedin similar structures of the Furrial Trend by Schneideret al (2004) In addition this pressure scenariomatches the calibration data for porosity (Figure 5B)

The 2D distribution of modeled overpressure atpresent day is shown in Figure 6A Apart fromimportant overpressure in the Querecual Formationthe allochthon is almost normally pressured Thethrust wedge is overpressured especially in thelow-permeability Carapita Formation Although

Table 2 Basin and Petroleum System Modeling (BPSM) InputParameters Thermal Boundary Conditions

Age (Ma)

SedimentWaterInterface

Temperature (degC)

Age (Ma)

Basal HeatFlow

(mW∕m2)min max

0 28 28 0 4226 28 28 30 6053 76 23 34 708 184 231 50 50105 62 233 107 6012 31 238 1455 8015 5 242 260 6019 5 24723 234 2527 78 235339 94 236655 146 2831455 148 287255 5 239

1336 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

the autochthon is highly overpressured close tonormal hydrostatic pressures are simulated in theforeland basin The discussed model assumption ofclosed faults in the postcompressional phase resultsin a regional pressure compartmentalization

The simulated pressure evolution through geo-logic time can be seen in Figure 7A A time extractionof a source rock cell in the thrust wedge shows a sud-den increase of both lithostatic and hydrostatic pres-sures that is a consequence of the tectonic burial

load associated with the thrusting In addition themodeled pore pressure buildup even though delayedin time can be directly linked to the thrust event Itcan be clearly seen that the timing of fault permeabil-ity directly impacts changes in the present-day porepressure The earlier the faults close the more over-pressure generated by the massive syntectonic sedi-mentation is kept until present day given adequatesealing lithologies are present The present-day over-pressure is modeled as14 MPa when the faults are

Figure 5 Plots of one-dimensional model output and available calibration data for Well Alpha (Table 3) (A) Dotted lines are hydro-static and lithostatic pressures continuous lines are pore pressure the master scenario is in black (a1 faults closing at 105 Ma) andadditional scenarios in gray (a2 open faults a3 closed faults) (B) modeled porosity (C) borehole temperature and (D) Ro The blackline shows the master scenario and the gray line the alternative high-heat-flow scenario Bold lines in the stratigraphic column arethrusts (T) and unconformities (U)

Table 3 Calibration Data for Well Alpha (Porosity Temperature Ro)

TVD (m) Porosity () TVD (m) Corrected Temperature (degC) TVD (m) Vitrinite Reflectance (Ro)

4350 15 600 40 400 0264400 15 1100 53 1200 0484800 10 1800 68 1485 068

4400 131 1700 0394800 133 2900 048

3100 0493400 0494000 054300 0524800 06

NEUMAIER ET AL 1337

open since 8 Ma as 23 MPa for 105 Ma and as37 MPa for 12 Ma Compared to this order of magni-tude changes in the permeability of the overlyingCarapita shale are negligible for the pore pressureTherefore the controlling parameter for pore pressurein the main reservoir of the Furrial trend is assumedfault permeability and resulting lateral connectivityrather than the permeability of the seal rock

Thermal Modeling

Modeled temperature and thermal maturity(Figure 5C D) have been calibrated with availablewell data Some discontinuities in the modeled Ro

trend are noted (Figure 5D) The upper kinks (egbase Pliocene) are related to unconformities withassociated erosion and the lower one (base UpperCretaceous) is due to the well penetrating a reverse

Figure 6 Two-dimensional model at present day with (A) modeled overpressure and (B) modeled thermal maturity and isothermsSource rocks marked by lithology pattern

1338 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

fault displacing mature rocks upon less mature ones(see Figure 6B for 2D view) There is a mismatchbetween the modeled Ro trend and the data in thedeeper part of the well In the Oligocene rocks themodeled temperature is too low (2degC or 36degF)whereas the modeled Ro is too high (up to 02 Ro)Any attempt to reduce the modeled Ro (eg bychanging thermal conductivity of the MioceneCarapita Formation) also results in a lower modeledpresent-day temperature Therefore the thermal sce-nario used for this calibration is a trade-off betweenmatching present-day temperature and Ro (using thekinetics of Sweeney and Burnham 1990) The basalheat flow assumed in the master scenario (Table 2)results in a modeled present-day surface heat flowranging from 37 mW∕m2 in the southern part of thethrusts to 52 mW∕m2 in parts of the uplifted areaAt the same depth the allochthonous northern partis thermally more mature than the southern autochthon(Figure 6B) because of the recent uplift Locallyinverted trends around the thrusts can be observedfor example at the location of Well Alpha (seeFigure 5D) In the autochthon the Upper Cretaceoussource rocks are entirely in the oil or even gas windowat present day because of the Neogene tectonic loadingand synorogenic deposition In the northern alloch-thon most of the Upper Cretaceous source rocks havebeen in the oil window in the past The modeled ther-mal maturity is in good agreement with the regionalmaturity suggested by Summa et al (2003)

Similar to the pressures the thrusting also affectsthe modeled temperature and Ro In the thrust wedgea direct relationship exists between the thrusting andthe increase of temperature and maturity The thrust-ing is the trigger for source rocks entering the oil win-dow within the thrust wedge (Figure 7B) and thepresent-day thermal maturity has been reached sincethe last 10 Ma Although this recent thermal historycan be considered as relatively well constrained bythe measured Ro in the calibration well this doesnot apply to the past In this thrusted area any ther-mal maturity reached prior to the thrusting onset hasbeen ldquooverprintedrdquo and therefore no thermal calibra-tion is possible for this period However theCretaceous source rocks were not buried deeply dur-ing the passive margin phase and even very hot sce-narios (basal heat flow of 100 mW∕m2) do not

significantly change their maturity (Figure 7B) Inthe uplifted allochthon several wells show low(03) to moderate (1) measured Ro (Parra et al2011) Because the allochthon experienced its maxi-mum burial and the highest temperatures during thepassive margin phase these maturity levels can beassumed as representative of the maturity prior tothe compression In contrast to the thrust wedge theconsequence of the thrusting is inverse for the alloch-thon (Figure 7C) as the hanging wall is uplifted andthe accompanying erosion removes the overburdenThe highest temperatures are simulated to have beenreached just before the thrusting event Then temper-ature decreased and Ro stagnated leaving the sourcerocks in the early mature zone until present day Theassumption of a regionally relatively homogeneousthermal field during the cooling phase of the passivemargin in combination with the relatively low matur-ities measured in the uplifted allochthon makes thehot heat-flow scenario unlikely However a laterallyvarying heat flow cannot be excluded and the conse-quently hotter paleoheat flow would have led to muchhigher temperatures prior to compression resultingalso in a higher present-day maturity level (14)despite the uplift (Figure 7C) Any variations of basalheat flow during the rift period (Late Jurassic to EarlyCretaceous) occurred too early to affect the maturityof the source rocks deposited during the lateCretaceous

Charge History

After calibration of the basinrsquos physical environ-ment the evolution of the petroleum systems wasmodeled Hydrocarbon generation was simulatedthrough geologic time using the activation energydistribution of the assigned source rock kineticsFrom the moment of hydrocarbon expulsiononward secondary migration was modeled usingthe invasion percolation method (Wilkinson 1984)It takes into account the geometry overpressure gra-dient and capillary entry pressures of the rockswhich are opposed to the buoyancy of the migratingoil and gas Figure 8 shows the time evolution of thetotal study area and Figure 9 focuses on the thrustwedge As already mentioned the described chargehistory is based on forward modeling It is consistent

NEUMAIER ET AL 1339

Figure 7 Time extractions (see Figure 6 for location) (A) Pressure development through geologic time for a source rock in the ver-tical continuity of Well Alpha Dotted lines are hydrostatic and lithostatic pressures continuous lines are pore pressure the master sce-nario is in black (a1 faults closing at 105 Ma) and additional scenarios in gray (a2 open faults a3 closed faults a4 faults closing at12 Ma a5 faults closing at 8 Ma) (B) Temperature (continuous lines) and Ro (dotted lines) through geologic time for a source rockin the vertical continuity of Well Alpha the black line shows the master scenario and the gray line the alternative high-heat-flow sce-nario (C) Temperature (continuous lines) and Ro (dotted lines) through geologic time for a source rock in the uplifted allochthonthe black line shows the master scenario and the gray line the alternative high heat flow scenario

1340 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

with the modeled compaction and thermal historyand therefore shows one scenario that explains thecharge of the Santa Barbara field and the forelandbasin Simplifications are inherent to any such work-flow and shall be considered when analyzing themodeling results

Passive Margin PhasemdashMesozoic toOligocene

Modeling results indicate that already during the LateCretaceous any Jurassic source rocks potentiallydeposited in the northern part of the section (Summaet al 2003) have reached temperatures of more than200degC (392degF) and corresponding high levels of matu-rity However the temperature at the base of theQuerecual Formation (Upper Cretaceous) was lessthan 100degC (212degF) leaving the kerogen immatureThe source rock became effective only during lateEocene with temperatures of up to 120degC (248degF)and a (reactive kerogen to hydrocarbon) transforma-tion ratio (TR) of more than 20 in the deeper distalpart of the passive margin Oil was quickly expelledmigrated vertically and seeped to the surface becauseof the general lack of seals in the overburden Duringthe Oligocene the time of deposition of the Merecurereservoir sandstones in the southern platform this sit-uation did not change apart from increased transforma-tion in the deeper basinal part of the passive margin

First oil accumulations are modeled in the Vidontildeosandstones in the basinal part in the late Oligocene(Figure 8A) in which the base of the QuerecualFormation shows TRs of up to 60 at around 130degC(266degF) In the proximal part there is no hydrocarbongeneration Although there is a general northwardtilting slope no southward migration from the distalmargin is modeled this has been checked usingdifferent migration methods including Darcy flowwith scenarios of continuous intrasource carrier bedsThis is in contrast to the studies from Gallango andParnaud (1995) and Schneider (2003) who suggestedlateral migration at that stage The difference couldbe explained by the presence of a regional top seal intheir models However our model suggests that theshallow rocks do not have sufficient sealing capacityto allow lateral hydrocarbon migration During thepassive-margin phase the basin was normally

pressured (Figure 7A) and simulated stresses havenot yet resulted in any fracturing at that stage

Main Compressional PhasemdashEarly to MiddleMiocene

With the onset of compression during early Miocene(Figure 8B) both the Pirital and Furrial thrusts wereformed and the northern distal part of the passivemargin was rapidly uplifted Important overpressuresas well as mechanical rock failure both the result ofthe rapid uplift are simulated in the allochthonDespite this uplift the kerogen-to-hydrocarbon trans-formation continued and the rate even increasedlocally close to the footwall cutoffs and where syn-orogenic deposition provided rapid overburden load-ing Oil accumulations are modeled in the freshlycreated anticlines

The tectonicndashsedimentary wedge between thetwo main thrusts experienced very rapid burial Infact the Upper Cretaceous source rocks were broughtinto depths where temperatures were as high as185degC (365degF) at 19 Ma and 225degC (437degF) at15 Ma Especially close to the cutoff of the Piritalthrust they very quickly reached a high maturity at19 Ma and reached full transformation in most placesat 12 Ma The expelled hydrocarbons migrated intothe sandstones of the Merecure Formation which weresealed by the Carapita Formation The charge toward aseries of recent anticlines is modeled by the fill-to-spillmechanism from north to south (Figure 9C D) Afterthe first oil fill the northern prospect quickly receiveda local gas charge the central structure containedmainly oil Note the presence of a lateral seal that isdue to sandndashshale juxtaposition The southernmoststructure which forms the Santa Barbara field todaywas charged by oil only at 15 Ma (Figure 9D)

The generated hydrocarbons in the southernautochthon were leaking to the surface until theCarapita Formation was compacted enough to holdsome of them (Figure 8C) At 12 Ma the wedgesource rocks were mostly spent but expulsion wasstill ongoing Important rock failure is modeled tohave occurred in the thrusted footwall breaking theformerly sealing Carapita Formation (Figure 10)This is in good agreement with cemented hydraulicfractures observed in a few cores in the Carapita

NEUMAIER ET AL 1341

Figure 8 Two-dimensional model through geologic time (selected time steps only) with modeled transformation ratio for the sourcerock layers isotherms and liquid and vapor migration and accumulation

1342 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

shales from the Furrial Trend (Roure et al 2005)This important fracturing event resulted in thedestruction of the modeled accumulation (Figure 9E)Hydrocarbon generation and expulsion were alsohappening in the southern foreland basin at transfor-mation ratios of up to 30

Late Compressional and PostcompressionalPhasemdashLate Miocene to Present Day

The late compressional phase was marked by the par-tial destruction of the allochthonous petroleum systemsby erosion (Figure 8E F) Many traps disappeared and

Figure 8 Continued

NEUMAIER ET AL 1343

the hydrocarbon generation within the source rocksstopped The erosion triggered an important sedimenttransfer toward the south contributing to further burialand increased temperatures in the thrust wedge and theforeland basin During late Miocene the petroleumsystem in the wedge matured An important northwardtilting was triggered by flexural tectonic loading andthe loading of syntectonic sediments (Summa et al2003) In combination with the presence of an effec-tive regional top seal the compacted CarapitaFormation allowed a change from dominantly verticalmigration into lateral migration First accumulationsare modeled in the Neogene synorogenic sequence

reservoirs in the south and important southwardmigration toward the foreland basin started(Figure 8G) These trends continued through Plioceneand Pleistocene times to present day filling theOnado field (Figure 8H) and other known fields inthe south (Figure 1A) of the Eastern Venezuela Basin

Prospectivity

We modeled the charge of the Santa Barbara field inthe Furrial Trend and of the Onado field in the fore-land basin (Figure 8H) Other prospects in the thrustwedge were also investigated The structure called

Figure 9 Detail of two-dimensionalmodel through geologic time (seeFigure 8) showing modeled transforma-tion ratio for the source rock layers iso-therms and liquid and vapor migrationand accumulation The Santa Barbara fieldand the prospect are shown by ellipses

1344 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

ldquoprospectrdquo (Figure 9G) looks similar to that of theSanta Barbara field because both are anticlines andthe reservoir and seal formations are the sameHowever they differ in several aspects First theprospect is in the vicinity of the Pirital thrust whichpresents a high risk of impacting the seal integrityThe prospect location is directly below the area wherewe presume the maximum of late tectonic activityoccurred (Figure 4) As modeled the CarapitaFormation the seal of the prospect was already brit-tle when this deformation occurred Calculatedstresses also predict rock failure from 12 Ma onward(Figure 10) Even though charge continued ourmodel suggests continuous leakage of the structure(Figure 9EndashG) The second major difference is thatthe source rock in the fault block of the prospect ismodeled to have already started generating and expel-ling hydrocarbons during the Oligocene and earlyMiocene (Figure 9A) before the trap formed and the

seal became effective (Figures 9C 10) Even thoughgeneration expulsion and migration continued aftertrap formation the prospect was not charged withthe earlier expelled oil In addition the present-daymaturity of the source rocks within the prospectrsquosfault block is higher than in the source rocks withinthe tectonic block of the Santa Barbara field poten-tially indicating expulsion of more gas A southndashnorth trend can be seen on the modeled accumulationcomposition with more gas toward the allochthon(Figure 9) The modeled composition of the prospectprior to seal failure was much richer in gas than inthe neighboring Santa Barbara field (Figure 9D) Infact simulation scenarios with unbreached sealsresulted in accumulation of predominantly gas Thecombination of both seal integrity and charge compo-sition concerns makes this a risky prospect In addi-tion potential diagenetic fluids related to the faultfluid flow might have altered the reservoir quality

Figure 10 Two-dimensional model (at 12 Ma) with modeled stress to failure Mohr circles (with σ standing for normal stress and τfor shear stress) and petroleum system events charts for the Santa Barbara field (left) and the prospect (right) Model predicts rock fail-ure in the prospect from about 12 Ma For further details see text

NEUMAIER ET AL 1345

Apart from the Onado oil field only minor accu-mulations are modeled in the southern Neogene syn-orogenic sequence The proportion of gas is higherthan in the Furrial trend The accumulations at thesouthern model edge are a modeling artifact repre-senting hydrocarbons (mainly oil) that in reality arethought to have charged the southern foreland basin(from mid to late Miocene times to present day)However parts of the migrated oil and gas could havebeen trapped in the Upper Cretaceous San JuanFormation south of the horst structure

The potential for conventional hydrocarbon accu-mulations in the allochthon is very limited by theimportant fracturing that occurred during the upliftperiod The basal sandstones of the Morochito piggy-back basin north of the Pirital thrust are not chargedin our 2D model because of the absence of structuraltraps Nevertheless there is a potential for strati-graphic traps to have been charged as migrationthrough the basin occurred However seal strengthmight be critical at this low-compaction stage Thesame applies for the Las Piedras Formation We thinkthere might be potential for shale oil because theQuerecual Formation is at shallow depths but showsmaturity within the oil window From the thermalmaturity point of view any potential Jurassic sourcerocks as discussed by Summa et al (2003) mightbe a gas source or shale gas target

UNCERTAINTY ANALYSIS

There is often significant uncertainty in basin geom-etry and depth of stratigraphic horizons based onuncertain seismic interpretation and velocity models(Kukla et al 2000 Brandes et al 2008 Baur et al2010) Restoring of present-day geometries to obtainpaleogeometries adds further uncertainty Additionaluncertainty is related to thermal boundary conditionssuch as paleoheat flow (Beha et al 2008) and togeochemical properties such as source rock character-istics and petroleum generation kinetics and calcula-tion of biodegradation and petroleum density(Blumenstein et al 2008) In addition to the porepressure and thermal sensitivity analysis justdescribed we varied initial TOC within a range of2 to 10 A variation of initial TOC obviously

directly scales the amount of initial kerogen availablefor transformation into hydrocarbons Neither thecharge volume nor composition of the Santa Barbarafield nor the wedge prospects are significantly sensi-tive to variation of initial TOC these traps are filledto spill in each scenario However the amount ofhydrocarbons charging the foreland basin (eg theOnado field) is reduced in the case of low initialTOC A high initial TOC only increased the modeledamount of surface seepage The sensitivity of the ini-tial HI is analogous to the sensitivity of the ini-tial TOC

In addition to the more classical uncertaintyand sensitivity analysis described we focused onquantitative analysis of the decompaction uncer-tainty when performing structural restoration Thisuncertainty is particular to the applied workflowof using structurally restored paleogeometries inBPSM Note that this uncertainty is significantlylower when using the backstripping method withthe optimization algorithms discussed One methodto analyze the adequacy of the decompactionmethod is comparison of the output rock matrixmass balance In the backstripping approach therock matrix mass is kept constant of a model cellwith a simulated decrease in porosity controlledby compaction laws and the corresponding reduc-tion of the layer thickness When using fixed pale-ogeometries from structural restoration the cellthickness is imposed However the decrease inporosity is calculated and the mass is adapted tofit the imposed cell thickness That means that therock matrix mass is constant only if ideal decom-paction has been applied

Figure 11A shows the percentage variation fromthe present-day matrix mass (100) of a given sourcerock cell through geologic time The dotted line rep-resents the ideal case with constant rock matrix massWe used two end-member decompaction scenarios todescribe the decompaction-related uncertainty that isintroduced when fixed paleogeometries replace back-stripping idealized hydrostatic decompaction assum-ing no overpressure and idealized lithostaticdecompaction assuming no effective stress (nodecompaction resulting in constant cell thicknessthrough time) In the latter case with decreasingporosity during compaction the rock matrix mass

1346 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

increases until it reaches the present-day mass Inother words rock mass matrix has been highly under-estimated in the past (in this particular case by morethan 20) This is a source rock so the kerogen massdecreases by the same factor This yields an underes-timation of more than 20 in TOC Note the rockmatrix mass is constant for time steps prior to23 Ma in which backstripping optimization couldbe performed toward the earliest paleogeometry Forthe second end-member scenario of idealized hydro-static decompaction the rock matrix mass is nearlyconstant which is a good decompaction adequacy

indicator for that given cell in the model Howeverthe so-measured adequacy changes throughout themodel In some areas the rock matrix mass has beenunderestimated by about 50 for nondecompactedpaleogeometries however in the case of hydrostaticdecompaction values generally range from 80 to120 of the present-day rock matrix mass Thatmeans in this particular case that hydrostatic decom-paction is more adequate (or less wrong) than nodecompaction at all The decompaction adequacycan be roughly correlated with pressure changesthrough geologic time

Figure 11 Time extractions for comparison of two end-member decompaction scenarios (see Figure 6 for location) (A) Rock matrixmass conservation through geologic time (B) Source rock burial depth through geologic time (C) Temperature and transformation ratiothrough geologic time

NEUMAIER ET AL 1347

Inadequate decompaction results in wrong depthfor stratigraphic horizons in the paleogeometriesFigure 11B shows the burial depth of the same inves-tigated source rock cell We can see that during thepassive margin phase there is a slight difference inpaleoburial depth for the two end-member scenariosDuring the compressional phase the differencechanges dramatically reaching more than 1000 m(3281 ft) in some places Toward present day bothcurves meet again because the input to the two resto-ration scenarios was the same (identical present-dayburial depth) This difference in compaction calcula-tion has severe implications on the modeled tempera-ture history and the kerogen-to-hydrocarbontransformation (Figure 11C) For an identical thermalframework in terms of modeling input parameterstemperature differences of up to 25degC (45degF) resultin TR differences of more than 20 in the early com-pressional phase before TR values meet again for thepresent-day situation

CONCLUSIONS

Structurally complex areas are attracting increasinginterest in hydrocarbon exploration Exploration riskcan be substantially narrowed by analyzing the his-tory of the basin and its petroleum systems based onstructural evolution We presented a further progressin technology that allows the integration of structuralrestoration and BPSM This combined analysis isseen as a way forward to improve modeling in thrustbelts with stacked stratigraphic sequences Despitelimitations of nonoptimized decompaction whichare added to the general uncertainty related to theindividual methods we modeled the charge and sealhistory of the Monagas fold and thrust belt inVenezuela Calibrated porosity temperature andthermal maturity as well as basin-scale stress pro-vided the regional framework through geologic timein which the petroleum systems evolved such thatthe charge of the known Santa Barbara and Onadofields and other trends could be understood We dis-cussed seal integrity of a potential exploration pros-pect by analyzing structural restoration-derived faultdisplacement through geologic time and an advancedstress forward simulation The uncertainty of using

decompacted or nondecompacted structural restora-tion has been assessed It can have important influ-ence on rock matrix mass balance through geologictime paleogeometry and paleoburial depth andtherefore affect paleotemperatures and paleomatur-ities We conclude that decompaction assuminghydrostatic pressure is the best option in that particu-lar case

REFERENCES CITED

Alberdi M and E Lafargue 1993 Vertical variations of organicmatter content in Guayuta Group (Upper Cretaceous)interior mountain belt eastern Venezuela OrganicGeochemistry v 20 p 425ndash436 doi 1010160146-6380(93)90091-O

Athy L F 1930 Density porosity and compaction of sedimen-tary rocks AAPG Bulletin v 14 p 1ndash24

Aymard R L Pimentel P Eitz P Lopez A ChaouchJ Navarro J Mijares and J G Pereira 1990 Geologic inte-gration and evaluation of Northern Monagas EasternVenezuelan Basin in J Brooks ed Classic petroleum prov-inces The Geologic Society (London) Special Publicationv 50 p 37ndash53

Baur F M Di Benedetto T Fuchs C Lampe andS Sciamanna 2009 Integrating structural geology andpetroleum systems modelingmdashA pilot project from Boliviarsquosfold and thrust belt Marine and Petroleum Geology v 26no 4 p 573ndash579 doi 101016jmarpetgeo200901004

Baur F R Littke H Wielens C Lampe and T Fuchs 2010Basin modeling meets rift analysismdashA numerical modelingstudy from the Jeanne drsquoArc basin offshore NewfoundlandCanada Marine and Petroleum Geology v 27 p 585ndash599doi 101016jmarpetgeo200906003

Beha A R O Thomsen and R Littke 2008 Thermal historyhydrocarbon generation and migration in the Horn Grabenin the Danish North Sea A 2D basin modelling studyInternational Journal of Earth Sciences v 97 p 1087ndash1100doi 101007s00531-007-0247-2

Blumenstein I O B M Krooss R di Primio W RottkeE Muumlller C Westerlage and R Littke 2008Biodegradation in numerical basin modelling A case studyfrom the Gifhorn Trough N-Germany International Journalof Earth Sciences v 97 p 1115ndash1129 doi 101007s00531-007-0272-1

Brandes C A Astorga R Littke and J Winsemann 2008Basin modelling of the Limoacuten back-arc basin (Costa Rica)Burial history and temperature evolution of an island arc-related basin-system Basin Research v 20 p 119ndash142doi 101111bre200820issue-1

Carnevali J O 1988 El Furrial oil field northeasternVenezuela First giant in foreland fold and thrust belts ofwestern hemisphere AAPG Bulletin v 72 68 p

Chevalier Y 1987 Contribution a lrsquoetude geacuteologique de lafrontiereacute Sud-Est de la plaque Caraibe Les zones internesde la chaine Sud-Caraibe sur le transect ile de Margarita-

1348 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

are generally not good enough for predicting prefault-ing hydrocarbon migration but might be acceptablefor hydrocarbon generation modeling Here structur-ally restored paleogeometries are important for thelayer shape in the proximity of faults and whencross-fault connectivity (like sandndashsand overlaps)matters (Figure 2C) Alternatively manual correctionof paleothickness can be done In halokinetic settingsbackstripping is generally superimposed by simplesalt reconstruction workflows (area balancing andextrapolation) Structural restoration should also beconsidered if salt movement results in geometries thatare more complex In any case thermal modeling inthe vicinity of salt is difficult due to its extremelyhigh thermal conductivity In areas with overthrustingor overturned folds (Figure 2D) with multiple strati-graphic sequences typically a fold and thrust beltbackstripping fails completely because lateral rockmovements are missing In the latter case paleo-geometries need to be generated by means other thanbackstripping (from simple conceptual drawing totrue structural restoration)

Paleogeometries generated from structuralrestorations can be loaded into BPSM software as analternative to performing a calculation from back-stripping Meshing is done section per section (timestep per time step) and tracking of cells from one sec-tion to the other is following rules described inHantschel and Kauerauf (2009) For the ages corre-sponding to the paleosection the basin geometry is

fixed it is not a function of compaction-controlledforward modeling as is the case when using back-stripping The BPSM simulation is restricted to thecalculation of properties such as porosity overpres-sure temperature vitrinite reflectance (Ro) andsource rock maturity which populate the fixed paleo-geometries as well as hydrocarbon migration Thesecompaction-controlled results are not used to adjustthe depositional amounts and to correct the paleoge-ometry as done in the backstripping approach andconsequently geometry optimization cannot beperformed The possible difference between thecompaction-law-controlled porosity change and thepredefined geometry change is indirectly compen-sated by the corresponding adjustment of the solidgrain mass

When paleogeometries are used for basin analy-sis and BPSM some factors become very importantRestoration must focus toward a best estimate paleo-water depth because changes in slope directions andtopography can largely affect fluid flow Erosionshould be estimated and reconstructed and manygeologic events need to be taken into account toincrease the resolution through time for dynamicmodeling The sediment thickness must be correctedin geologic time for decompaction All these factorswhich are common sources of uncertainty are criticalfor a geologically meaningful restoration

Because of the optimization method justdescribed adequately decompacted paleogeometries

Figure 2 Steps for (A B) backstripping and (C D) structural restoration of geometries of different structural complexity For detailssee text

1330 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

are required from structural restoration Inadequatedecompaction can produce large errors resulting inan unbalanced mass of solid rock matrix underesti-mated source rock paleoburial depth and deviatedpaleomigration paths related to differential compac-tion However it is difficult to define adequatedecompaction As a rule of thumb in normally pres-sured clastic basins where the main compactiondriver is linearly increasing effective stress simplehydrostatic decompaction laws such as thosedescribed by Athy (1930) can be used for restorationIn cases of abnormal pressure (overpressure) hydro-static decompaction overestimates the effectivestress or underestimates porosity and might fail tothe same degree as restoration neglecting decompac-tion One might be scaling normal decompactioncurves wherever regional overpressure occurs andhas occurred In basins with complex erosion pres-sure and diagenetic histories optimization loops(using BPSM simulated properties for optimizeddecompaction in second restoration) might benecessary Any sort of decompaction in a carbonate-dominated environment where compaction is gener-ally less understood because of the effect of mineraltransformations and cementation is subject to muchlarger uncertainty The question of whether to usehydrostatically decompacted or nondecompactedstructurally restored paleogeometries is discussed inthe section on uncertainty analysis

MODELING

Prior to structural restoration we performed several1D simulations These generated erosion and paleoal-titude estimations for structural restoration andprovided a calibration for the 2D model One-dimensional modeling also gave the first informationon hydrocarbon generation and timing to identifythe period of first source rock maturation (ldquocriticalmomentrdquo as defined by Magoon and Dow 1994)This is very important because prior to structuralrestoration its identification indicates when addi-tional time steps are needed to increase time resolu-tion for critical time intervals so that the activepetroleum system can be dynamically modeled in anadequate way However for simulation regular time

steps are needed not only during the critical momentbut also for the entire basin history

Structural Restoration

The following section describes our structural resto-ration results shown in Figure 3 of the post-riftstructural evolution of the Monagas fold and thrustbelt It can be divided into three main steps(1) From the Mesozoic to Oligocene the Monagasarea formed parts of the northern passive margin ofthe South American shield (Figure 3A) The sub-sidence driver was mostly thermal cooling and iso-static load with only a few normal faults Thesediments were sourced by the craton in the south(2) In the early Miocene the margin experiencedcompression from the transcollision with theCaribbean plate (Figure 3B) Two major normalfaults that formed during rifting were inverted in thecourse of the collision generating the regionalFurrial and Pirital thrusts Synorogenic deposits werecovered by the thrusted allochthon and intramontanepiggyback basins were formed With increasing upliftand erosion (middle Miocene Figure 3C) their sedi-ment fills were redeposited to form the important syn-tectonic Oficina and Carapita Formations south of thethrusts (3) From late Miocene on the late compres-sional phase started Figure 3D shows the present-day structure of the Monagas fold and thrust belt(2D model crossing the Santa Barbara field)According to our restoration the total shortening onthe section accommodated by the thrusts during thecompression is around 40 km (2486 mi) which cor-responds to 35 It is important to note that the Piritalthrust was active until late Miocene as indicated bythe fault displacement analysis (Figure 4) These latetectonic movements might have an important effecton the seal integrity of prospects that are in the vicin-ity of the thrust

Model Input Data

Table 1 shows geologic ages lithologies and petro-physical parameters that were assigned to our 2Dmodel For the main source rocks (Querecual andSan Antonio Formations) a type II B kinetic wasassigned (Pepper and Corvi 1995) As a base case

NEUMAIER ET AL 1331

we assumed a constant initial TOC of 6 wt and aninitial HI of 500 mg HCg TOC However becauseinitial TOC and HI were estimated from present-daymeasured values these initial values are highly uncer-tain In addition there is both lateral and vertical vari-ability within the layer even though locally very highsource rock potential is likely to occur a bulk TOCand HI need to be assigned to the model cells Forthese reasons several models with different initialTOC have been run and their sensitivity to the mod-eled charge is discussed below Faults were set open

to fluid flow in the course of their main kinematicactivity which is estimated by fault displacementanalysis of the structural restoration models Aftertheir main activity phase faults are assumed to beclosed Fault permeability was also a varying param-eter in the sensitivity analysis discussed belowThermal boundary conditions are given in Table 2Surface temperature ranges between 23 and 30degC(734 and 86degF) through geologic time with apresent-day average temperature of 28degC (824degF)These temperatures at sea level have been corrected

Figure 3 Structural evolution of the Monagas fold and thrust belt demonstrated by structural restoration (A) Passive margin stage(until 23 Ma) (B) Early compression stage (21 Ma) (C) Late compression stage (12 Ma) (D) Present day For details see text

1332 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

Figure 4 Fault displacement from lateMiocene to present day for all faults onrestored cross sections (A) and for the Piritalthrust only (B) Black and white shading andsize of points show magnitude of displacementin meters

NEUMAIER ET AL 1333

Table1

Basin

andPetro

leum

Syste

mModeling(BPSM)Input

ParametersGeologicAges

andLithologies

Compressib

ility

(GPa

-1)

Perm

eability

(logmD)

Thermal

conductivity

(Wm

K)

Heatcapacity

(kcalkgK)

Age

(Ma)

GeologicTime

Facies

Density

(kg∕m

3 )Initial

Porosity(

)max

min

at1

por

at25por

at20degC

at100degC

at20degC

at100degC

0Pleisto

cene

LasP

iedras

sandsto

ne2720

412747

115

minus18

3395

338

02

024

LasPiedrasshale

2700

4040327

403

minus852

minus3

164

169

021

024

26

Pliocene

LasP

iedras

sandsto

ne2720

412747

115

minus18

3395

338

02

024

LasPiedrasshale

2700

4040327

403

minus852

minus3

164

169

021

024

53

upperM

iocene

2La

Pica

sandsto

ne2720

412747

115

minus18

3395

338

02

024

LaPica

shale

2700

4040327

403

minus852

minus3

164

169

021

024

Morich

itoshale

2700

4040327

403

minus852

minus3

164

169

021

024

8upperM

iocene

1La

Pica

sandsto

ne2720

412747

115

minus18

3395

338

02

024

LaPica

shale

2700

4040327

403

minus852

minus3

164

169

021

024

Morich

itobasal

sandsto

ne2715

4812142

187

minus348

15

317

284

02

024

105

middleMiocene

2Carapitasandsto

ne2715

4812142

187

minus348

15

317

284

02

024

Carapitashale

2705

6330932

331

minus684

minus15

204

201

02

024

12middleMiocene

1Carapitashale

2705

6330932

331

minus684

minus15

204

201

02

024

15lower

Miocene

2Carapitashale

2705

6330932

331

minus684

minus15

204

201

02

024

19lower

Miocene

1lower

Carapitashale

2700

7040327

403

minus852

minus3

164

169

021

024

23Oligocene2

Merecureplatform

sandsto

ne2715

4812142

187

minus348

15

317

284

02

024

27Oligocene1

Merecureplatform

sandsto

ne2715

4812142

187

minus348

15

317

284

02

024

1334 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

339

PaleoEocene

Vidono

sandsto

ne2720

412747

115

minus18

3395

338

02

024

Caratasbasin

alshale

2700

7040327

403

minus852

minus3

164

169

021

024

655

upperC

retaceous4

SanJuan

sandsto

ne2720

412747

115

minus18

3395

338

02

024

74upperC

retaceous3

SanAntoniosiltstone

(SOUR

CERO

CK)

2710

5510422

212

minus628

minus1

201

196

023

026

825

upperC

retaceous2

SanAntoniosiltstone

(SOUR

CERO

CK)

2710

5510422

212

minus628

minus1

201

196

023

026

91upperC

retaceous1

Querecualmudsto

ne(SOUR

CERO

CK)

2700

7040327

403

minus852

minus3

164

169

021

024

996

lower

Cretaceous

2undifflower

Cretaceous

(30

Lst50

Sst

20Sh)

2722

459445

141

minus334

174

305

275

02

023

122

lower

Cretaceous

1undifflower

Cretaceous

(30

Lst50

Sst

20Sh)

2722

459445

141

minus334

174

305

275

02

023

1455

sub-Cretaceous

2undiffsub-Cretaceous

(50

Sh50Sst)

2710

5521537

259

minus516

0255

239

02

024

251

NEUMAIER ET AL 1335

by the influence of the paleowater depth using theapproach of Wygrala (1989) Because of the hightopographic elevation of approximately 3500 m(11500 ft) the maximum altitude during the pastcompressional phase we adjusted the model takinginto account a temperature decrease of 6degCkm (33degF1000 ft) related to the paleoaltitude This resultsin a difference of up to 20degC (36degF) compared to sea-level temperature We then assigned the adjustedsedimentwater interface temperatures to the modelas the upper thermal boundary condition At the baseof the model we assigned a heat flow ranging from42 to 80 mW∕m2 through geologic time The highestheat flow is assumed during the rifting event beforethe basin cooled as a passive margin A thermal eventin the Eocene detected by fission track analysis(Locke and Garver 2005 Perez de Armas 2005)has been addressed by a second peak before a secondcooling phase For uncertainty analysis basal paleo-heat flow has been varied within geological meaning-ful ranges and the thermal calibration data has beenhonored as much as possible

Modeling Results

Basin and petroleum system modeling can be dividedinto simulation of the physical parameters within abasin such as pressure and temperature (basin model-ing) and the subsequent evolution of the source rocksand their derived hydrocarbon compounds (petro-leum systems modeling) In the following sectionswe use this subdivision to describe the BPSM resultscompaction modeling thermal modeling and chargehistory It is important to state that these are predic-tions based on forward modeling as just describedThe results are compared and calibrated to actualmeasurements and observations where available

Compaction Modeling

Figure 5 presents an extraction of the modeled resultsat the location of Well Alpha in the Furrial Trend Inthe absence of direct calibration data for pore pres-sure we undertook a sensitivity analysis to test thedifferent scenarios Using the model we tested howpore pressure in the main reservoir of the FurrialTrend reacts to changes in fault and seal-rock per-meability In a first step fault permeability has beenevaluated Faults were assumed open during the mainfault activity phase and closed afterward In additionwe performed several variations from that base casescenario Figure 5A shows the effect of two end-member scenarios on pore pressure within the mainreservoir of the Furrial Trend (Santa Barbara field) inwhich faults have been modeled to be either open orclosed since their formation A third scenario consistsof thrust faults that are open during the compressionalphase (from 15 Ma on) and closed afterward (from105 Ma on) This scenario has been chosen as theldquomaster scenariordquo because it results in overpressureof 23 MPa which is in the range of pressure reportedin similar structures of the Furrial Trend by Schneideret al (2004) In addition this pressure scenariomatches the calibration data for porosity (Figure 5B)

The 2D distribution of modeled overpressure atpresent day is shown in Figure 6A Apart fromimportant overpressure in the Querecual Formationthe allochthon is almost normally pressured Thethrust wedge is overpressured especially in thelow-permeability Carapita Formation Although

Table 2 Basin and Petroleum System Modeling (BPSM) InputParameters Thermal Boundary Conditions

Age (Ma)

SedimentWaterInterface

Temperature (degC)

Age (Ma)

Basal HeatFlow

(mW∕m2)min max

0 28 28 0 4226 28 28 30 6053 76 23 34 708 184 231 50 50105 62 233 107 6012 31 238 1455 8015 5 242 260 6019 5 24723 234 2527 78 235339 94 236655 146 2831455 148 287255 5 239

1336 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

the autochthon is highly overpressured close tonormal hydrostatic pressures are simulated in theforeland basin The discussed model assumption ofclosed faults in the postcompressional phase resultsin a regional pressure compartmentalization

The simulated pressure evolution through geo-logic time can be seen in Figure 7A A time extractionof a source rock cell in the thrust wedge shows a sud-den increase of both lithostatic and hydrostatic pres-sures that is a consequence of the tectonic burial

load associated with the thrusting In addition themodeled pore pressure buildup even though delayedin time can be directly linked to the thrust event Itcan be clearly seen that the timing of fault permeabil-ity directly impacts changes in the present-day porepressure The earlier the faults close the more over-pressure generated by the massive syntectonic sedi-mentation is kept until present day given adequatesealing lithologies are present The present-day over-pressure is modeled as14 MPa when the faults are

Figure 5 Plots of one-dimensional model output and available calibration data for Well Alpha (Table 3) (A) Dotted lines are hydro-static and lithostatic pressures continuous lines are pore pressure the master scenario is in black (a1 faults closing at 105 Ma) andadditional scenarios in gray (a2 open faults a3 closed faults) (B) modeled porosity (C) borehole temperature and (D) Ro The blackline shows the master scenario and the gray line the alternative high-heat-flow scenario Bold lines in the stratigraphic column arethrusts (T) and unconformities (U)

Table 3 Calibration Data for Well Alpha (Porosity Temperature Ro)

TVD (m) Porosity () TVD (m) Corrected Temperature (degC) TVD (m) Vitrinite Reflectance (Ro)

4350 15 600 40 400 0264400 15 1100 53 1200 0484800 10 1800 68 1485 068

4400 131 1700 0394800 133 2900 048

3100 0493400 0494000 054300 0524800 06

NEUMAIER ET AL 1337

open since 8 Ma as 23 MPa for 105 Ma and as37 MPa for 12 Ma Compared to this order of magni-tude changes in the permeability of the overlyingCarapita shale are negligible for the pore pressureTherefore the controlling parameter for pore pressurein the main reservoir of the Furrial trend is assumedfault permeability and resulting lateral connectivityrather than the permeability of the seal rock

Thermal Modeling

Modeled temperature and thermal maturity(Figure 5C D) have been calibrated with availablewell data Some discontinuities in the modeled Ro

trend are noted (Figure 5D) The upper kinks (egbase Pliocene) are related to unconformities withassociated erosion and the lower one (base UpperCretaceous) is due to the well penetrating a reverse

Figure 6 Two-dimensional model at present day with (A) modeled overpressure and (B) modeled thermal maturity and isothermsSource rocks marked by lithology pattern

1338 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

fault displacing mature rocks upon less mature ones(see Figure 6B for 2D view) There is a mismatchbetween the modeled Ro trend and the data in thedeeper part of the well In the Oligocene rocks themodeled temperature is too low (2degC or 36degF)whereas the modeled Ro is too high (up to 02 Ro)Any attempt to reduce the modeled Ro (eg bychanging thermal conductivity of the MioceneCarapita Formation) also results in a lower modeledpresent-day temperature Therefore the thermal sce-nario used for this calibration is a trade-off betweenmatching present-day temperature and Ro (using thekinetics of Sweeney and Burnham 1990) The basalheat flow assumed in the master scenario (Table 2)results in a modeled present-day surface heat flowranging from 37 mW∕m2 in the southern part of thethrusts to 52 mW∕m2 in parts of the uplifted areaAt the same depth the allochthonous northern partis thermally more mature than the southern autochthon(Figure 6B) because of the recent uplift Locallyinverted trends around the thrusts can be observedfor example at the location of Well Alpha (seeFigure 5D) In the autochthon the Upper Cretaceoussource rocks are entirely in the oil or even gas windowat present day because of the Neogene tectonic loadingand synorogenic deposition In the northern alloch-thon most of the Upper Cretaceous source rocks havebeen in the oil window in the past The modeled ther-mal maturity is in good agreement with the regionalmaturity suggested by Summa et al (2003)

Similar to the pressures the thrusting also affectsthe modeled temperature and Ro In the thrust wedgea direct relationship exists between the thrusting andthe increase of temperature and maturity The thrust-ing is the trigger for source rocks entering the oil win-dow within the thrust wedge (Figure 7B) and thepresent-day thermal maturity has been reached sincethe last 10 Ma Although this recent thermal historycan be considered as relatively well constrained bythe measured Ro in the calibration well this doesnot apply to the past In this thrusted area any ther-mal maturity reached prior to the thrusting onset hasbeen ldquooverprintedrdquo and therefore no thermal calibra-tion is possible for this period However theCretaceous source rocks were not buried deeply dur-ing the passive margin phase and even very hot sce-narios (basal heat flow of 100 mW∕m2) do not

significantly change their maturity (Figure 7B) Inthe uplifted allochthon several wells show low(03) to moderate (1) measured Ro (Parra et al2011) Because the allochthon experienced its maxi-mum burial and the highest temperatures during thepassive margin phase these maturity levels can beassumed as representative of the maturity prior tothe compression In contrast to the thrust wedge theconsequence of the thrusting is inverse for the alloch-thon (Figure 7C) as the hanging wall is uplifted andthe accompanying erosion removes the overburdenThe highest temperatures are simulated to have beenreached just before the thrusting event Then temper-ature decreased and Ro stagnated leaving the sourcerocks in the early mature zone until present day Theassumption of a regionally relatively homogeneousthermal field during the cooling phase of the passivemargin in combination with the relatively low matur-ities measured in the uplifted allochthon makes thehot heat-flow scenario unlikely However a laterallyvarying heat flow cannot be excluded and the conse-quently hotter paleoheat flow would have led to muchhigher temperatures prior to compression resultingalso in a higher present-day maturity level (14)despite the uplift (Figure 7C) Any variations of basalheat flow during the rift period (Late Jurassic to EarlyCretaceous) occurred too early to affect the maturityof the source rocks deposited during the lateCretaceous

Charge History

After calibration of the basinrsquos physical environ-ment the evolution of the petroleum systems wasmodeled Hydrocarbon generation was simulatedthrough geologic time using the activation energydistribution of the assigned source rock kineticsFrom the moment of hydrocarbon expulsiononward secondary migration was modeled usingthe invasion percolation method (Wilkinson 1984)It takes into account the geometry overpressure gra-dient and capillary entry pressures of the rockswhich are opposed to the buoyancy of the migratingoil and gas Figure 8 shows the time evolution of thetotal study area and Figure 9 focuses on the thrustwedge As already mentioned the described chargehistory is based on forward modeling It is consistent

NEUMAIER ET AL 1339

Figure 7 Time extractions (see Figure 6 for location) (A) Pressure development through geologic time for a source rock in the ver-tical continuity of Well Alpha Dotted lines are hydrostatic and lithostatic pressures continuous lines are pore pressure the master sce-nario is in black (a1 faults closing at 105 Ma) and additional scenarios in gray (a2 open faults a3 closed faults a4 faults closing at12 Ma a5 faults closing at 8 Ma) (B) Temperature (continuous lines) and Ro (dotted lines) through geologic time for a source rockin the vertical continuity of Well Alpha the black line shows the master scenario and the gray line the alternative high-heat-flow sce-nario (C) Temperature (continuous lines) and Ro (dotted lines) through geologic time for a source rock in the uplifted allochthonthe black line shows the master scenario and the gray line the alternative high heat flow scenario

1340 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

with the modeled compaction and thermal historyand therefore shows one scenario that explains thecharge of the Santa Barbara field and the forelandbasin Simplifications are inherent to any such work-flow and shall be considered when analyzing themodeling results

Passive Margin PhasemdashMesozoic toOligocene

Modeling results indicate that already during the LateCretaceous any Jurassic source rocks potentiallydeposited in the northern part of the section (Summaet al 2003) have reached temperatures of more than200degC (392degF) and corresponding high levels of matu-rity However the temperature at the base of theQuerecual Formation (Upper Cretaceous) was lessthan 100degC (212degF) leaving the kerogen immatureThe source rock became effective only during lateEocene with temperatures of up to 120degC (248degF)and a (reactive kerogen to hydrocarbon) transforma-tion ratio (TR) of more than 20 in the deeper distalpart of the passive margin Oil was quickly expelledmigrated vertically and seeped to the surface becauseof the general lack of seals in the overburden Duringthe Oligocene the time of deposition of the Merecurereservoir sandstones in the southern platform this sit-uation did not change apart from increased transforma-tion in the deeper basinal part of the passive margin

First oil accumulations are modeled in the Vidontildeosandstones in the basinal part in the late Oligocene(Figure 8A) in which the base of the QuerecualFormation shows TRs of up to 60 at around 130degC(266degF) In the proximal part there is no hydrocarbongeneration Although there is a general northwardtilting slope no southward migration from the distalmargin is modeled this has been checked usingdifferent migration methods including Darcy flowwith scenarios of continuous intrasource carrier bedsThis is in contrast to the studies from Gallango andParnaud (1995) and Schneider (2003) who suggestedlateral migration at that stage The difference couldbe explained by the presence of a regional top seal intheir models However our model suggests that theshallow rocks do not have sufficient sealing capacityto allow lateral hydrocarbon migration During thepassive-margin phase the basin was normally

pressured (Figure 7A) and simulated stresses havenot yet resulted in any fracturing at that stage

Main Compressional PhasemdashEarly to MiddleMiocene

With the onset of compression during early Miocene(Figure 8B) both the Pirital and Furrial thrusts wereformed and the northern distal part of the passivemargin was rapidly uplifted Important overpressuresas well as mechanical rock failure both the result ofthe rapid uplift are simulated in the allochthonDespite this uplift the kerogen-to-hydrocarbon trans-formation continued and the rate even increasedlocally close to the footwall cutoffs and where syn-orogenic deposition provided rapid overburden load-ing Oil accumulations are modeled in the freshlycreated anticlines

The tectonicndashsedimentary wedge between thetwo main thrusts experienced very rapid burial Infact the Upper Cretaceous source rocks were broughtinto depths where temperatures were as high as185degC (365degF) at 19 Ma and 225degC (437degF) at15 Ma Especially close to the cutoff of the Piritalthrust they very quickly reached a high maturity at19 Ma and reached full transformation in most placesat 12 Ma The expelled hydrocarbons migrated intothe sandstones of the Merecure Formation which weresealed by the Carapita Formation The charge toward aseries of recent anticlines is modeled by the fill-to-spillmechanism from north to south (Figure 9C D) Afterthe first oil fill the northern prospect quickly receiveda local gas charge the central structure containedmainly oil Note the presence of a lateral seal that isdue to sandndashshale juxtaposition The southernmoststructure which forms the Santa Barbara field todaywas charged by oil only at 15 Ma (Figure 9D)

The generated hydrocarbons in the southernautochthon were leaking to the surface until theCarapita Formation was compacted enough to holdsome of them (Figure 8C) At 12 Ma the wedgesource rocks were mostly spent but expulsion wasstill ongoing Important rock failure is modeled tohave occurred in the thrusted footwall breaking theformerly sealing Carapita Formation (Figure 10)This is in good agreement with cemented hydraulicfractures observed in a few cores in the Carapita

NEUMAIER ET AL 1341

Figure 8 Two-dimensional model through geologic time (selected time steps only) with modeled transformation ratio for the sourcerock layers isotherms and liquid and vapor migration and accumulation

1342 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

shales from the Furrial Trend (Roure et al 2005)This important fracturing event resulted in thedestruction of the modeled accumulation (Figure 9E)Hydrocarbon generation and expulsion were alsohappening in the southern foreland basin at transfor-mation ratios of up to 30

Late Compressional and PostcompressionalPhasemdashLate Miocene to Present Day

The late compressional phase was marked by the par-tial destruction of the allochthonous petroleum systemsby erosion (Figure 8E F) Many traps disappeared and

Figure 8 Continued

NEUMAIER ET AL 1343

the hydrocarbon generation within the source rocksstopped The erosion triggered an important sedimenttransfer toward the south contributing to further burialand increased temperatures in the thrust wedge and theforeland basin During late Miocene the petroleumsystem in the wedge matured An important northwardtilting was triggered by flexural tectonic loading andthe loading of syntectonic sediments (Summa et al2003) In combination with the presence of an effec-tive regional top seal the compacted CarapitaFormation allowed a change from dominantly verticalmigration into lateral migration First accumulationsare modeled in the Neogene synorogenic sequence

reservoirs in the south and important southwardmigration toward the foreland basin started(Figure 8G) These trends continued through Plioceneand Pleistocene times to present day filling theOnado field (Figure 8H) and other known fields inthe south (Figure 1A) of the Eastern Venezuela Basin

Prospectivity

We modeled the charge of the Santa Barbara field inthe Furrial Trend and of the Onado field in the fore-land basin (Figure 8H) Other prospects in the thrustwedge were also investigated The structure called

Figure 9 Detail of two-dimensionalmodel through geologic time (seeFigure 8) showing modeled transforma-tion ratio for the source rock layers iso-therms and liquid and vapor migrationand accumulation The Santa Barbara fieldand the prospect are shown by ellipses

1344 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

ldquoprospectrdquo (Figure 9G) looks similar to that of theSanta Barbara field because both are anticlines andthe reservoir and seal formations are the sameHowever they differ in several aspects First theprospect is in the vicinity of the Pirital thrust whichpresents a high risk of impacting the seal integrityThe prospect location is directly below the area wherewe presume the maximum of late tectonic activityoccurred (Figure 4) As modeled the CarapitaFormation the seal of the prospect was already brit-tle when this deformation occurred Calculatedstresses also predict rock failure from 12 Ma onward(Figure 10) Even though charge continued ourmodel suggests continuous leakage of the structure(Figure 9EndashG) The second major difference is thatthe source rock in the fault block of the prospect ismodeled to have already started generating and expel-ling hydrocarbons during the Oligocene and earlyMiocene (Figure 9A) before the trap formed and the

seal became effective (Figures 9C 10) Even thoughgeneration expulsion and migration continued aftertrap formation the prospect was not charged withthe earlier expelled oil In addition the present-daymaturity of the source rocks within the prospectrsquosfault block is higher than in the source rocks withinthe tectonic block of the Santa Barbara field poten-tially indicating expulsion of more gas A southndashnorth trend can be seen on the modeled accumulationcomposition with more gas toward the allochthon(Figure 9) The modeled composition of the prospectprior to seal failure was much richer in gas than inthe neighboring Santa Barbara field (Figure 9D) Infact simulation scenarios with unbreached sealsresulted in accumulation of predominantly gas Thecombination of both seal integrity and charge compo-sition concerns makes this a risky prospect In addi-tion potential diagenetic fluids related to the faultfluid flow might have altered the reservoir quality

Figure 10 Two-dimensional model (at 12 Ma) with modeled stress to failure Mohr circles (with σ standing for normal stress and τfor shear stress) and petroleum system events charts for the Santa Barbara field (left) and the prospect (right) Model predicts rock fail-ure in the prospect from about 12 Ma For further details see text

NEUMAIER ET AL 1345

Apart from the Onado oil field only minor accu-mulations are modeled in the southern Neogene syn-orogenic sequence The proportion of gas is higherthan in the Furrial trend The accumulations at thesouthern model edge are a modeling artifact repre-senting hydrocarbons (mainly oil) that in reality arethought to have charged the southern foreland basin(from mid to late Miocene times to present day)However parts of the migrated oil and gas could havebeen trapped in the Upper Cretaceous San JuanFormation south of the horst structure

The potential for conventional hydrocarbon accu-mulations in the allochthon is very limited by theimportant fracturing that occurred during the upliftperiod The basal sandstones of the Morochito piggy-back basin north of the Pirital thrust are not chargedin our 2D model because of the absence of structuraltraps Nevertheless there is a potential for strati-graphic traps to have been charged as migrationthrough the basin occurred However seal strengthmight be critical at this low-compaction stage Thesame applies for the Las Piedras Formation We thinkthere might be potential for shale oil because theQuerecual Formation is at shallow depths but showsmaturity within the oil window From the thermalmaturity point of view any potential Jurassic sourcerocks as discussed by Summa et al (2003) mightbe a gas source or shale gas target

UNCERTAINTY ANALYSIS

There is often significant uncertainty in basin geom-etry and depth of stratigraphic horizons based onuncertain seismic interpretation and velocity models(Kukla et al 2000 Brandes et al 2008 Baur et al2010) Restoring of present-day geometries to obtainpaleogeometries adds further uncertainty Additionaluncertainty is related to thermal boundary conditionssuch as paleoheat flow (Beha et al 2008) and togeochemical properties such as source rock character-istics and petroleum generation kinetics and calcula-tion of biodegradation and petroleum density(Blumenstein et al 2008) In addition to the porepressure and thermal sensitivity analysis justdescribed we varied initial TOC within a range of2 to 10 A variation of initial TOC obviously

directly scales the amount of initial kerogen availablefor transformation into hydrocarbons Neither thecharge volume nor composition of the Santa Barbarafield nor the wedge prospects are significantly sensi-tive to variation of initial TOC these traps are filledto spill in each scenario However the amount ofhydrocarbons charging the foreland basin (eg theOnado field) is reduced in the case of low initialTOC A high initial TOC only increased the modeledamount of surface seepage The sensitivity of the ini-tial HI is analogous to the sensitivity of the ini-tial TOC

In addition to the more classical uncertaintyand sensitivity analysis described we focused onquantitative analysis of the decompaction uncer-tainty when performing structural restoration Thisuncertainty is particular to the applied workflowof using structurally restored paleogeometries inBPSM Note that this uncertainty is significantlylower when using the backstripping method withthe optimization algorithms discussed One methodto analyze the adequacy of the decompactionmethod is comparison of the output rock matrixmass balance In the backstripping approach therock matrix mass is kept constant of a model cellwith a simulated decrease in porosity controlledby compaction laws and the corresponding reduc-tion of the layer thickness When using fixed pale-ogeometries from structural restoration the cellthickness is imposed However the decrease inporosity is calculated and the mass is adapted tofit the imposed cell thickness That means that therock matrix mass is constant only if ideal decom-paction has been applied

Figure 11A shows the percentage variation fromthe present-day matrix mass (100) of a given sourcerock cell through geologic time The dotted line rep-resents the ideal case with constant rock matrix massWe used two end-member decompaction scenarios todescribe the decompaction-related uncertainty that isintroduced when fixed paleogeometries replace back-stripping idealized hydrostatic decompaction assum-ing no overpressure and idealized lithostaticdecompaction assuming no effective stress (nodecompaction resulting in constant cell thicknessthrough time) In the latter case with decreasingporosity during compaction the rock matrix mass

1346 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

increases until it reaches the present-day mass Inother words rock mass matrix has been highly under-estimated in the past (in this particular case by morethan 20) This is a source rock so the kerogen massdecreases by the same factor This yields an underes-timation of more than 20 in TOC Note the rockmatrix mass is constant for time steps prior to23 Ma in which backstripping optimization couldbe performed toward the earliest paleogeometry Forthe second end-member scenario of idealized hydro-static decompaction the rock matrix mass is nearlyconstant which is a good decompaction adequacy

indicator for that given cell in the model Howeverthe so-measured adequacy changes throughout themodel In some areas the rock matrix mass has beenunderestimated by about 50 for nondecompactedpaleogeometries however in the case of hydrostaticdecompaction values generally range from 80 to120 of the present-day rock matrix mass Thatmeans in this particular case that hydrostatic decom-paction is more adequate (or less wrong) than nodecompaction at all The decompaction adequacycan be roughly correlated with pressure changesthrough geologic time

Figure 11 Time extractions for comparison of two end-member decompaction scenarios (see Figure 6 for location) (A) Rock matrixmass conservation through geologic time (B) Source rock burial depth through geologic time (C) Temperature and transformation ratiothrough geologic time

NEUMAIER ET AL 1347

Inadequate decompaction results in wrong depthfor stratigraphic horizons in the paleogeometriesFigure 11B shows the burial depth of the same inves-tigated source rock cell We can see that during thepassive margin phase there is a slight difference inpaleoburial depth for the two end-member scenariosDuring the compressional phase the differencechanges dramatically reaching more than 1000 m(3281 ft) in some places Toward present day bothcurves meet again because the input to the two resto-ration scenarios was the same (identical present-dayburial depth) This difference in compaction calcula-tion has severe implications on the modeled tempera-ture history and the kerogen-to-hydrocarbontransformation (Figure 11C) For an identical thermalframework in terms of modeling input parameterstemperature differences of up to 25degC (45degF) resultin TR differences of more than 20 in the early com-pressional phase before TR values meet again for thepresent-day situation

CONCLUSIONS

Structurally complex areas are attracting increasinginterest in hydrocarbon exploration Exploration riskcan be substantially narrowed by analyzing the his-tory of the basin and its petroleum systems based onstructural evolution We presented a further progressin technology that allows the integration of structuralrestoration and BPSM This combined analysis isseen as a way forward to improve modeling in thrustbelts with stacked stratigraphic sequences Despitelimitations of nonoptimized decompaction whichare added to the general uncertainty related to theindividual methods we modeled the charge and sealhistory of the Monagas fold and thrust belt inVenezuela Calibrated porosity temperature andthermal maturity as well as basin-scale stress pro-vided the regional framework through geologic timein which the petroleum systems evolved such thatthe charge of the known Santa Barbara and Onadofields and other trends could be understood We dis-cussed seal integrity of a potential exploration pros-pect by analyzing structural restoration-derived faultdisplacement through geologic time and an advancedstress forward simulation The uncertainty of using

decompacted or nondecompacted structural restora-tion has been assessed It can have important influ-ence on rock matrix mass balance through geologictime paleogeometry and paleoburial depth andtherefore affect paleotemperatures and paleomatur-ities We conclude that decompaction assuminghydrostatic pressure is the best option in that particu-lar case

REFERENCES CITED

Alberdi M and E Lafargue 1993 Vertical variations of organicmatter content in Guayuta Group (Upper Cretaceous)interior mountain belt eastern Venezuela OrganicGeochemistry v 20 p 425ndash436 doi 1010160146-6380(93)90091-O

Athy L F 1930 Density porosity and compaction of sedimen-tary rocks AAPG Bulletin v 14 p 1ndash24

Aymard R L Pimentel P Eitz P Lopez A ChaouchJ Navarro J Mijares and J G Pereira 1990 Geologic inte-gration and evaluation of Northern Monagas EasternVenezuelan Basin in J Brooks ed Classic petroleum prov-inces The Geologic Society (London) Special Publicationv 50 p 37ndash53

Baur F M Di Benedetto T Fuchs C Lampe andS Sciamanna 2009 Integrating structural geology andpetroleum systems modelingmdashA pilot project from Boliviarsquosfold and thrust belt Marine and Petroleum Geology v 26no 4 p 573ndash579 doi 101016jmarpetgeo200901004

Baur F R Littke H Wielens C Lampe and T Fuchs 2010Basin modeling meets rift analysismdashA numerical modelingstudy from the Jeanne drsquoArc basin offshore NewfoundlandCanada Marine and Petroleum Geology v 27 p 585ndash599doi 101016jmarpetgeo200906003

Beha A R O Thomsen and R Littke 2008 Thermal historyhydrocarbon generation and migration in the Horn Grabenin the Danish North Sea A 2D basin modelling studyInternational Journal of Earth Sciences v 97 p 1087ndash1100doi 101007s00531-007-0247-2

Blumenstein I O B M Krooss R di Primio W RottkeE Muumlller C Westerlage and R Littke 2008Biodegradation in numerical basin modelling A case studyfrom the Gifhorn Trough N-Germany International Journalof Earth Sciences v 97 p 1115ndash1129 doi 101007s00531-007-0272-1

Brandes C A Astorga R Littke and J Winsemann 2008Basin modelling of the Limoacuten back-arc basin (Costa Rica)Burial history and temperature evolution of an island arc-related basin-system Basin Research v 20 p 119ndash142doi 101111bre200820issue-1

Carnevali J O 1988 El Furrial oil field northeasternVenezuela First giant in foreland fold and thrust belts ofwestern hemisphere AAPG Bulletin v 72 68 p

Chevalier Y 1987 Contribution a lrsquoetude geacuteologique de lafrontiereacute Sud-Est de la plaque Caraibe Les zones internesde la chaine Sud-Caraibe sur le transect ile de Margarita-

1348 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

are required from structural restoration Inadequatedecompaction can produce large errors resulting inan unbalanced mass of solid rock matrix underesti-mated source rock paleoburial depth and deviatedpaleomigration paths related to differential compac-tion However it is difficult to define adequatedecompaction As a rule of thumb in normally pres-sured clastic basins where the main compactiondriver is linearly increasing effective stress simplehydrostatic decompaction laws such as thosedescribed by Athy (1930) can be used for restorationIn cases of abnormal pressure (overpressure) hydro-static decompaction overestimates the effectivestress or underestimates porosity and might fail tothe same degree as restoration neglecting decompac-tion One might be scaling normal decompactioncurves wherever regional overpressure occurs andhas occurred In basins with complex erosion pres-sure and diagenetic histories optimization loops(using BPSM simulated properties for optimizeddecompaction in second restoration) might benecessary Any sort of decompaction in a carbonate-dominated environment where compaction is gener-ally less understood because of the effect of mineraltransformations and cementation is subject to muchlarger uncertainty The question of whether to usehydrostatically decompacted or nondecompactedstructurally restored paleogeometries is discussed inthe section on uncertainty analysis

MODELING

Prior to structural restoration we performed several1D simulations These generated erosion and paleoal-titude estimations for structural restoration andprovided a calibration for the 2D model One-dimensional modeling also gave the first informationon hydrocarbon generation and timing to identifythe period of first source rock maturation (ldquocriticalmomentrdquo as defined by Magoon and Dow 1994)This is very important because prior to structuralrestoration its identification indicates when addi-tional time steps are needed to increase time resolu-tion for critical time intervals so that the activepetroleum system can be dynamically modeled in anadequate way However for simulation regular time

steps are needed not only during the critical momentbut also for the entire basin history

Structural Restoration

The following section describes our structural resto-ration results shown in Figure 3 of the post-riftstructural evolution of the Monagas fold and thrustbelt It can be divided into three main steps(1) From the Mesozoic to Oligocene the Monagasarea formed parts of the northern passive margin ofthe South American shield (Figure 3A) The sub-sidence driver was mostly thermal cooling and iso-static load with only a few normal faults Thesediments were sourced by the craton in the south(2) In the early Miocene the margin experiencedcompression from the transcollision with theCaribbean plate (Figure 3B) Two major normalfaults that formed during rifting were inverted in thecourse of the collision generating the regionalFurrial and Pirital thrusts Synorogenic deposits werecovered by the thrusted allochthon and intramontanepiggyback basins were formed With increasing upliftand erosion (middle Miocene Figure 3C) their sedi-ment fills were redeposited to form the important syn-tectonic Oficina and Carapita Formations south of thethrusts (3) From late Miocene on the late compres-sional phase started Figure 3D shows the present-day structure of the Monagas fold and thrust belt(2D model crossing the Santa Barbara field)According to our restoration the total shortening onthe section accommodated by the thrusts during thecompression is around 40 km (2486 mi) which cor-responds to 35 It is important to note that the Piritalthrust was active until late Miocene as indicated bythe fault displacement analysis (Figure 4) These latetectonic movements might have an important effecton the seal integrity of prospects that are in the vicin-ity of the thrust

Model Input Data

Table 1 shows geologic ages lithologies and petro-physical parameters that were assigned to our 2Dmodel For the main source rocks (Querecual andSan Antonio Formations) a type II B kinetic wasassigned (Pepper and Corvi 1995) As a base case

NEUMAIER ET AL 1331

we assumed a constant initial TOC of 6 wt and aninitial HI of 500 mg HCg TOC However becauseinitial TOC and HI were estimated from present-daymeasured values these initial values are highly uncer-tain In addition there is both lateral and vertical vari-ability within the layer even though locally very highsource rock potential is likely to occur a bulk TOCand HI need to be assigned to the model cells Forthese reasons several models with different initialTOC have been run and their sensitivity to the mod-eled charge is discussed below Faults were set open

to fluid flow in the course of their main kinematicactivity which is estimated by fault displacementanalysis of the structural restoration models Aftertheir main activity phase faults are assumed to beclosed Fault permeability was also a varying param-eter in the sensitivity analysis discussed belowThermal boundary conditions are given in Table 2Surface temperature ranges between 23 and 30degC(734 and 86degF) through geologic time with apresent-day average temperature of 28degC (824degF)These temperatures at sea level have been corrected

Figure 3 Structural evolution of the Monagas fold and thrust belt demonstrated by structural restoration (A) Passive margin stage(until 23 Ma) (B) Early compression stage (21 Ma) (C) Late compression stage (12 Ma) (D) Present day For details see text

1332 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

Figure 4 Fault displacement from lateMiocene to present day for all faults onrestored cross sections (A) and for the Piritalthrust only (B) Black and white shading andsize of points show magnitude of displacementin meters

NEUMAIER ET AL 1333

Table1

Basin

andPetro

leum

Syste

mModeling(BPSM)Input

ParametersGeologicAges

andLithologies

Compressib

ility

(GPa

-1)

Perm

eability

(logmD)

Thermal

conductivity

(Wm

K)

Heatcapacity

(kcalkgK)

Age

(Ma)

GeologicTime

Facies

Density

(kg∕m

3 )Initial

Porosity(

)max

min

at1

por

at25por

at20degC

at100degC

at20degC

at100degC

0Pleisto

cene

LasP

iedras

sandsto

ne2720

412747

115

minus18

3395

338

02

024

LasPiedrasshale

2700

4040327

403

minus852

minus3

164

169

021

024

26

Pliocene

LasP

iedras

sandsto

ne2720

412747

115

minus18

3395

338

02

024

LasPiedrasshale

2700

4040327

403

minus852

minus3

164

169

021

024

53

upperM

iocene

2La

Pica

sandsto

ne2720

412747

115

minus18

3395

338

02

024

LaPica

shale

2700

4040327

403

minus852

minus3

164

169

021

024

Morich

itoshale

2700

4040327

403

minus852

minus3

164

169

021

024

8upperM

iocene

1La

Pica

sandsto

ne2720

412747

115

minus18

3395

338

02

024

LaPica

shale

2700

4040327

403

minus852

minus3

164

169

021

024

Morich

itobasal

sandsto

ne2715

4812142

187

minus348

15

317

284

02

024

105

middleMiocene

2Carapitasandsto

ne2715

4812142

187

minus348

15

317

284

02

024

Carapitashale

2705

6330932

331

minus684

minus15

204

201

02

024

12middleMiocene

1Carapitashale

2705

6330932

331

minus684

minus15

204

201

02

024

15lower

Miocene

2Carapitashale

2705

6330932

331

minus684

minus15

204

201

02

024

19lower

Miocene

1lower

Carapitashale

2700

7040327

403

minus852

minus3

164

169

021

024

23Oligocene2

Merecureplatform

sandsto

ne2715

4812142

187

minus348

15

317

284

02

024

27Oligocene1

Merecureplatform

sandsto

ne2715

4812142

187

minus348

15

317

284

02

024

1334 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

339

PaleoEocene

Vidono

sandsto

ne2720

412747

115

minus18

3395

338

02

024

Caratasbasin

alshale

2700

7040327

403

minus852

minus3

164

169

021

024

655

upperC

retaceous4

SanJuan

sandsto

ne2720

412747

115

minus18

3395

338

02

024

74upperC

retaceous3

SanAntoniosiltstone

(SOUR

CERO

CK)

2710

5510422

212

minus628

minus1

201

196

023

026

825

upperC

retaceous2

SanAntoniosiltstone

(SOUR

CERO

CK)

2710

5510422

212

minus628

minus1

201

196

023

026

91upperC

retaceous1

Querecualmudsto

ne(SOUR

CERO

CK)

2700

7040327

403

minus852

minus3

164

169

021

024

996

lower

Cretaceous

2undifflower

Cretaceous

(30

Lst50

Sst

20Sh)

2722

459445

141

minus334

174

305

275

02

023

122

lower

Cretaceous

1undifflower

Cretaceous

(30

Lst50

Sst

20Sh)

2722

459445

141

minus334

174

305

275

02

023

1455

sub-Cretaceous

2undiffsub-Cretaceous

(50

Sh50Sst)

2710

5521537

259

minus516

0255

239

02

024

251

NEUMAIER ET AL 1335

by the influence of the paleowater depth using theapproach of Wygrala (1989) Because of the hightopographic elevation of approximately 3500 m(11500 ft) the maximum altitude during the pastcompressional phase we adjusted the model takinginto account a temperature decrease of 6degCkm (33degF1000 ft) related to the paleoaltitude This resultsin a difference of up to 20degC (36degF) compared to sea-level temperature We then assigned the adjustedsedimentwater interface temperatures to the modelas the upper thermal boundary condition At the baseof the model we assigned a heat flow ranging from42 to 80 mW∕m2 through geologic time The highestheat flow is assumed during the rifting event beforethe basin cooled as a passive margin A thermal eventin the Eocene detected by fission track analysis(Locke and Garver 2005 Perez de Armas 2005)has been addressed by a second peak before a secondcooling phase For uncertainty analysis basal paleo-heat flow has been varied within geological meaning-ful ranges and the thermal calibration data has beenhonored as much as possible

Modeling Results

Basin and petroleum system modeling can be dividedinto simulation of the physical parameters within abasin such as pressure and temperature (basin model-ing) and the subsequent evolution of the source rocksand their derived hydrocarbon compounds (petro-leum systems modeling) In the following sectionswe use this subdivision to describe the BPSM resultscompaction modeling thermal modeling and chargehistory It is important to state that these are predic-tions based on forward modeling as just describedThe results are compared and calibrated to actualmeasurements and observations where available

Compaction Modeling

Figure 5 presents an extraction of the modeled resultsat the location of Well Alpha in the Furrial Trend Inthe absence of direct calibration data for pore pres-sure we undertook a sensitivity analysis to test thedifferent scenarios Using the model we tested howpore pressure in the main reservoir of the FurrialTrend reacts to changes in fault and seal-rock per-meability In a first step fault permeability has beenevaluated Faults were assumed open during the mainfault activity phase and closed afterward In additionwe performed several variations from that base casescenario Figure 5A shows the effect of two end-member scenarios on pore pressure within the mainreservoir of the Furrial Trend (Santa Barbara field) inwhich faults have been modeled to be either open orclosed since their formation A third scenario consistsof thrust faults that are open during the compressionalphase (from 15 Ma on) and closed afterward (from105 Ma on) This scenario has been chosen as theldquomaster scenariordquo because it results in overpressureof 23 MPa which is in the range of pressure reportedin similar structures of the Furrial Trend by Schneideret al (2004) In addition this pressure scenariomatches the calibration data for porosity (Figure 5B)

The 2D distribution of modeled overpressure atpresent day is shown in Figure 6A Apart fromimportant overpressure in the Querecual Formationthe allochthon is almost normally pressured Thethrust wedge is overpressured especially in thelow-permeability Carapita Formation Although

Table 2 Basin and Petroleum System Modeling (BPSM) InputParameters Thermal Boundary Conditions

Age (Ma)

SedimentWaterInterface

Temperature (degC)

Age (Ma)

Basal HeatFlow

(mW∕m2)min max

0 28 28 0 4226 28 28 30 6053 76 23 34 708 184 231 50 50105 62 233 107 6012 31 238 1455 8015 5 242 260 6019 5 24723 234 2527 78 235339 94 236655 146 2831455 148 287255 5 239

1336 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

the autochthon is highly overpressured close tonormal hydrostatic pressures are simulated in theforeland basin The discussed model assumption ofclosed faults in the postcompressional phase resultsin a regional pressure compartmentalization

The simulated pressure evolution through geo-logic time can be seen in Figure 7A A time extractionof a source rock cell in the thrust wedge shows a sud-den increase of both lithostatic and hydrostatic pres-sures that is a consequence of the tectonic burial

load associated with the thrusting In addition themodeled pore pressure buildup even though delayedin time can be directly linked to the thrust event Itcan be clearly seen that the timing of fault permeabil-ity directly impacts changes in the present-day porepressure The earlier the faults close the more over-pressure generated by the massive syntectonic sedi-mentation is kept until present day given adequatesealing lithologies are present The present-day over-pressure is modeled as14 MPa when the faults are

Figure 5 Plots of one-dimensional model output and available calibration data for Well Alpha (Table 3) (A) Dotted lines are hydro-static and lithostatic pressures continuous lines are pore pressure the master scenario is in black (a1 faults closing at 105 Ma) andadditional scenarios in gray (a2 open faults a3 closed faults) (B) modeled porosity (C) borehole temperature and (D) Ro The blackline shows the master scenario and the gray line the alternative high-heat-flow scenario Bold lines in the stratigraphic column arethrusts (T) and unconformities (U)

Table 3 Calibration Data for Well Alpha (Porosity Temperature Ro)

TVD (m) Porosity () TVD (m) Corrected Temperature (degC) TVD (m) Vitrinite Reflectance (Ro)

4350 15 600 40 400 0264400 15 1100 53 1200 0484800 10 1800 68 1485 068

4400 131 1700 0394800 133 2900 048

3100 0493400 0494000 054300 0524800 06

NEUMAIER ET AL 1337

open since 8 Ma as 23 MPa for 105 Ma and as37 MPa for 12 Ma Compared to this order of magni-tude changes in the permeability of the overlyingCarapita shale are negligible for the pore pressureTherefore the controlling parameter for pore pressurein the main reservoir of the Furrial trend is assumedfault permeability and resulting lateral connectivityrather than the permeability of the seal rock

Thermal Modeling

Modeled temperature and thermal maturity(Figure 5C D) have been calibrated with availablewell data Some discontinuities in the modeled Ro

trend are noted (Figure 5D) The upper kinks (egbase Pliocene) are related to unconformities withassociated erosion and the lower one (base UpperCretaceous) is due to the well penetrating a reverse

Figure 6 Two-dimensional model at present day with (A) modeled overpressure and (B) modeled thermal maturity and isothermsSource rocks marked by lithology pattern

1338 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

fault displacing mature rocks upon less mature ones(see Figure 6B for 2D view) There is a mismatchbetween the modeled Ro trend and the data in thedeeper part of the well In the Oligocene rocks themodeled temperature is too low (2degC or 36degF)whereas the modeled Ro is too high (up to 02 Ro)Any attempt to reduce the modeled Ro (eg bychanging thermal conductivity of the MioceneCarapita Formation) also results in a lower modeledpresent-day temperature Therefore the thermal sce-nario used for this calibration is a trade-off betweenmatching present-day temperature and Ro (using thekinetics of Sweeney and Burnham 1990) The basalheat flow assumed in the master scenario (Table 2)results in a modeled present-day surface heat flowranging from 37 mW∕m2 in the southern part of thethrusts to 52 mW∕m2 in parts of the uplifted areaAt the same depth the allochthonous northern partis thermally more mature than the southern autochthon(Figure 6B) because of the recent uplift Locallyinverted trends around the thrusts can be observedfor example at the location of Well Alpha (seeFigure 5D) In the autochthon the Upper Cretaceoussource rocks are entirely in the oil or even gas windowat present day because of the Neogene tectonic loadingand synorogenic deposition In the northern alloch-thon most of the Upper Cretaceous source rocks havebeen in the oil window in the past The modeled ther-mal maturity is in good agreement with the regionalmaturity suggested by Summa et al (2003)

Similar to the pressures the thrusting also affectsthe modeled temperature and Ro In the thrust wedgea direct relationship exists between the thrusting andthe increase of temperature and maturity The thrust-ing is the trigger for source rocks entering the oil win-dow within the thrust wedge (Figure 7B) and thepresent-day thermal maturity has been reached sincethe last 10 Ma Although this recent thermal historycan be considered as relatively well constrained bythe measured Ro in the calibration well this doesnot apply to the past In this thrusted area any ther-mal maturity reached prior to the thrusting onset hasbeen ldquooverprintedrdquo and therefore no thermal calibra-tion is possible for this period However theCretaceous source rocks were not buried deeply dur-ing the passive margin phase and even very hot sce-narios (basal heat flow of 100 mW∕m2) do not

significantly change their maturity (Figure 7B) Inthe uplifted allochthon several wells show low(03) to moderate (1) measured Ro (Parra et al2011) Because the allochthon experienced its maxi-mum burial and the highest temperatures during thepassive margin phase these maturity levels can beassumed as representative of the maturity prior tothe compression In contrast to the thrust wedge theconsequence of the thrusting is inverse for the alloch-thon (Figure 7C) as the hanging wall is uplifted andthe accompanying erosion removes the overburdenThe highest temperatures are simulated to have beenreached just before the thrusting event Then temper-ature decreased and Ro stagnated leaving the sourcerocks in the early mature zone until present day Theassumption of a regionally relatively homogeneousthermal field during the cooling phase of the passivemargin in combination with the relatively low matur-ities measured in the uplifted allochthon makes thehot heat-flow scenario unlikely However a laterallyvarying heat flow cannot be excluded and the conse-quently hotter paleoheat flow would have led to muchhigher temperatures prior to compression resultingalso in a higher present-day maturity level (14)despite the uplift (Figure 7C) Any variations of basalheat flow during the rift period (Late Jurassic to EarlyCretaceous) occurred too early to affect the maturityof the source rocks deposited during the lateCretaceous

Charge History

After calibration of the basinrsquos physical environ-ment the evolution of the petroleum systems wasmodeled Hydrocarbon generation was simulatedthrough geologic time using the activation energydistribution of the assigned source rock kineticsFrom the moment of hydrocarbon expulsiononward secondary migration was modeled usingthe invasion percolation method (Wilkinson 1984)It takes into account the geometry overpressure gra-dient and capillary entry pressures of the rockswhich are opposed to the buoyancy of the migratingoil and gas Figure 8 shows the time evolution of thetotal study area and Figure 9 focuses on the thrustwedge As already mentioned the described chargehistory is based on forward modeling It is consistent

NEUMAIER ET AL 1339

Figure 7 Time extractions (see Figure 6 for location) (A) Pressure development through geologic time for a source rock in the ver-tical continuity of Well Alpha Dotted lines are hydrostatic and lithostatic pressures continuous lines are pore pressure the master sce-nario is in black (a1 faults closing at 105 Ma) and additional scenarios in gray (a2 open faults a3 closed faults a4 faults closing at12 Ma a5 faults closing at 8 Ma) (B) Temperature (continuous lines) and Ro (dotted lines) through geologic time for a source rockin the vertical continuity of Well Alpha the black line shows the master scenario and the gray line the alternative high-heat-flow sce-nario (C) Temperature (continuous lines) and Ro (dotted lines) through geologic time for a source rock in the uplifted allochthonthe black line shows the master scenario and the gray line the alternative high heat flow scenario

1340 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

with the modeled compaction and thermal historyand therefore shows one scenario that explains thecharge of the Santa Barbara field and the forelandbasin Simplifications are inherent to any such work-flow and shall be considered when analyzing themodeling results

Passive Margin PhasemdashMesozoic toOligocene

Modeling results indicate that already during the LateCretaceous any Jurassic source rocks potentiallydeposited in the northern part of the section (Summaet al 2003) have reached temperatures of more than200degC (392degF) and corresponding high levels of matu-rity However the temperature at the base of theQuerecual Formation (Upper Cretaceous) was lessthan 100degC (212degF) leaving the kerogen immatureThe source rock became effective only during lateEocene with temperatures of up to 120degC (248degF)and a (reactive kerogen to hydrocarbon) transforma-tion ratio (TR) of more than 20 in the deeper distalpart of the passive margin Oil was quickly expelledmigrated vertically and seeped to the surface becauseof the general lack of seals in the overburden Duringthe Oligocene the time of deposition of the Merecurereservoir sandstones in the southern platform this sit-uation did not change apart from increased transforma-tion in the deeper basinal part of the passive margin

First oil accumulations are modeled in the Vidontildeosandstones in the basinal part in the late Oligocene(Figure 8A) in which the base of the QuerecualFormation shows TRs of up to 60 at around 130degC(266degF) In the proximal part there is no hydrocarbongeneration Although there is a general northwardtilting slope no southward migration from the distalmargin is modeled this has been checked usingdifferent migration methods including Darcy flowwith scenarios of continuous intrasource carrier bedsThis is in contrast to the studies from Gallango andParnaud (1995) and Schneider (2003) who suggestedlateral migration at that stage The difference couldbe explained by the presence of a regional top seal intheir models However our model suggests that theshallow rocks do not have sufficient sealing capacityto allow lateral hydrocarbon migration During thepassive-margin phase the basin was normally

pressured (Figure 7A) and simulated stresses havenot yet resulted in any fracturing at that stage

Main Compressional PhasemdashEarly to MiddleMiocene

With the onset of compression during early Miocene(Figure 8B) both the Pirital and Furrial thrusts wereformed and the northern distal part of the passivemargin was rapidly uplifted Important overpressuresas well as mechanical rock failure both the result ofthe rapid uplift are simulated in the allochthonDespite this uplift the kerogen-to-hydrocarbon trans-formation continued and the rate even increasedlocally close to the footwall cutoffs and where syn-orogenic deposition provided rapid overburden load-ing Oil accumulations are modeled in the freshlycreated anticlines

The tectonicndashsedimentary wedge between thetwo main thrusts experienced very rapid burial Infact the Upper Cretaceous source rocks were broughtinto depths where temperatures were as high as185degC (365degF) at 19 Ma and 225degC (437degF) at15 Ma Especially close to the cutoff of the Piritalthrust they very quickly reached a high maturity at19 Ma and reached full transformation in most placesat 12 Ma The expelled hydrocarbons migrated intothe sandstones of the Merecure Formation which weresealed by the Carapita Formation The charge toward aseries of recent anticlines is modeled by the fill-to-spillmechanism from north to south (Figure 9C D) Afterthe first oil fill the northern prospect quickly receiveda local gas charge the central structure containedmainly oil Note the presence of a lateral seal that isdue to sandndashshale juxtaposition The southernmoststructure which forms the Santa Barbara field todaywas charged by oil only at 15 Ma (Figure 9D)

The generated hydrocarbons in the southernautochthon were leaking to the surface until theCarapita Formation was compacted enough to holdsome of them (Figure 8C) At 12 Ma the wedgesource rocks were mostly spent but expulsion wasstill ongoing Important rock failure is modeled tohave occurred in the thrusted footwall breaking theformerly sealing Carapita Formation (Figure 10)This is in good agreement with cemented hydraulicfractures observed in a few cores in the Carapita

NEUMAIER ET AL 1341

Figure 8 Two-dimensional model through geologic time (selected time steps only) with modeled transformation ratio for the sourcerock layers isotherms and liquid and vapor migration and accumulation

1342 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

shales from the Furrial Trend (Roure et al 2005)This important fracturing event resulted in thedestruction of the modeled accumulation (Figure 9E)Hydrocarbon generation and expulsion were alsohappening in the southern foreland basin at transfor-mation ratios of up to 30

Late Compressional and PostcompressionalPhasemdashLate Miocene to Present Day

The late compressional phase was marked by the par-tial destruction of the allochthonous petroleum systemsby erosion (Figure 8E F) Many traps disappeared and

Figure 8 Continued

NEUMAIER ET AL 1343

the hydrocarbon generation within the source rocksstopped The erosion triggered an important sedimenttransfer toward the south contributing to further burialand increased temperatures in the thrust wedge and theforeland basin During late Miocene the petroleumsystem in the wedge matured An important northwardtilting was triggered by flexural tectonic loading andthe loading of syntectonic sediments (Summa et al2003) In combination with the presence of an effec-tive regional top seal the compacted CarapitaFormation allowed a change from dominantly verticalmigration into lateral migration First accumulationsare modeled in the Neogene synorogenic sequence

reservoirs in the south and important southwardmigration toward the foreland basin started(Figure 8G) These trends continued through Plioceneand Pleistocene times to present day filling theOnado field (Figure 8H) and other known fields inthe south (Figure 1A) of the Eastern Venezuela Basin

Prospectivity

We modeled the charge of the Santa Barbara field inthe Furrial Trend and of the Onado field in the fore-land basin (Figure 8H) Other prospects in the thrustwedge were also investigated The structure called

Figure 9 Detail of two-dimensionalmodel through geologic time (seeFigure 8) showing modeled transforma-tion ratio for the source rock layers iso-therms and liquid and vapor migrationand accumulation The Santa Barbara fieldand the prospect are shown by ellipses

1344 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

ldquoprospectrdquo (Figure 9G) looks similar to that of theSanta Barbara field because both are anticlines andthe reservoir and seal formations are the sameHowever they differ in several aspects First theprospect is in the vicinity of the Pirital thrust whichpresents a high risk of impacting the seal integrityThe prospect location is directly below the area wherewe presume the maximum of late tectonic activityoccurred (Figure 4) As modeled the CarapitaFormation the seal of the prospect was already brit-tle when this deformation occurred Calculatedstresses also predict rock failure from 12 Ma onward(Figure 10) Even though charge continued ourmodel suggests continuous leakage of the structure(Figure 9EndashG) The second major difference is thatthe source rock in the fault block of the prospect ismodeled to have already started generating and expel-ling hydrocarbons during the Oligocene and earlyMiocene (Figure 9A) before the trap formed and the

seal became effective (Figures 9C 10) Even thoughgeneration expulsion and migration continued aftertrap formation the prospect was not charged withthe earlier expelled oil In addition the present-daymaturity of the source rocks within the prospectrsquosfault block is higher than in the source rocks withinthe tectonic block of the Santa Barbara field poten-tially indicating expulsion of more gas A southndashnorth trend can be seen on the modeled accumulationcomposition with more gas toward the allochthon(Figure 9) The modeled composition of the prospectprior to seal failure was much richer in gas than inthe neighboring Santa Barbara field (Figure 9D) Infact simulation scenarios with unbreached sealsresulted in accumulation of predominantly gas Thecombination of both seal integrity and charge compo-sition concerns makes this a risky prospect In addi-tion potential diagenetic fluids related to the faultfluid flow might have altered the reservoir quality

Figure 10 Two-dimensional model (at 12 Ma) with modeled stress to failure Mohr circles (with σ standing for normal stress and τfor shear stress) and petroleum system events charts for the Santa Barbara field (left) and the prospect (right) Model predicts rock fail-ure in the prospect from about 12 Ma For further details see text

NEUMAIER ET AL 1345

Apart from the Onado oil field only minor accu-mulations are modeled in the southern Neogene syn-orogenic sequence The proportion of gas is higherthan in the Furrial trend The accumulations at thesouthern model edge are a modeling artifact repre-senting hydrocarbons (mainly oil) that in reality arethought to have charged the southern foreland basin(from mid to late Miocene times to present day)However parts of the migrated oil and gas could havebeen trapped in the Upper Cretaceous San JuanFormation south of the horst structure

The potential for conventional hydrocarbon accu-mulations in the allochthon is very limited by theimportant fracturing that occurred during the upliftperiod The basal sandstones of the Morochito piggy-back basin north of the Pirital thrust are not chargedin our 2D model because of the absence of structuraltraps Nevertheless there is a potential for strati-graphic traps to have been charged as migrationthrough the basin occurred However seal strengthmight be critical at this low-compaction stage Thesame applies for the Las Piedras Formation We thinkthere might be potential for shale oil because theQuerecual Formation is at shallow depths but showsmaturity within the oil window From the thermalmaturity point of view any potential Jurassic sourcerocks as discussed by Summa et al (2003) mightbe a gas source or shale gas target

UNCERTAINTY ANALYSIS

There is often significant uncertainty in basin geom-etry and depth of stratigraphic horizons based onuncertain seismic interpretation and velocity models(Kukla et al 2000 Brandes et al 2008 Baur et al2010) Restoring of present-day geometries to obtainpaleogeometries adds further uncertainty Additionaluncertainty is related to thermal boundary conditionssuch as paleoheat flow (Beha et al 2008) and togeochemical properties such as source rock character-istics and petroleum generation kinetics and calcula-tion of biodegradation and petroleum density(Blumenstein et al 2008) In addition to the porepressure and thermal sensitivity analysis justdescribed we varied initial TOC within a range of2 to 10 A variation of initial TOC obviously

directly scales the amount of initial kerogen availablefor transformation into hydrocarbons Neither thecharge volume nor composition of the Santa Barbarafield nor the wedge prospects are significantly sensi-tive to variation of initial TOC these traps are filledto spill in each scenario However the amount ofhydrocarbons charging the foreland basin (eg theOnado field) is reduced in the case of low initialTOC A high initial TOC only increased the modeledamount of surface seepage The sensitivity of the ini-tial HI is analogous to the sensitivity of the ini-tial TOC

In addition to the more classical uncertaintyand sensitivity analysis described we focused onquantitative analysis of the decompaction uncer-tainty when performing structural restoration Thisuncertainty is particular to the applied workflowof using structurally restored paleogeometries inBPSM Note that this uncertainty is significantlylower when using the backstripping method withthe optimization algorithms discussed One methodto analyze the adequacy of the decompactionmethod is comparison of the output rock matrixmass balance In the backstripping approach therock matrix mass is kept constant of a model cellwith a simulated decrease in porosity controlledby compaction laws and the corresponding reduc-tion of the layer thickness When using fixed pale-ogeometries from structural restoration the cellthickness is imposed However the decrease inporosity is calculated and the mass is adapted tofit the imposed cell thickness That means that therock matrix mass is constant only if ideal decom-paction has been applied

Figure 11A shows the percentage variation fromthe present-day matrix mass (100) of a given sourcerock cell through geologic time The dotted line rep-resents the ideal case with constant rock matrix massWe used two end-member decompaction scenarios todescribe the decompaction-related uncertainty that isintroduced when fixed paleogeometries replace back-stripping idealized hydrostatic decompaction assum-ing no overpressure and idealized lithostaticdecompaction assuming no effective stress (nodecompaction resulting in constant cell thicknessthrough time) In the latter case with decreasingporosity during compaction the rock matrix mass

1346 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

increases until it reaches the present-day mass Inother words rock mass matrix has been highly under-estimated in the past (in this particular case by morethan 20) This is a source rock so the kerogen massdecreases by the same factor This yields an underes-timation of more than 20 in TOC Note the rockmatrix mass is constant for time steps prior to23 Ma in which backstripping optimization couldbe performed toward the earliest paleogeometry Forthe second end-member scenario of idealized hydro-static decompaction the rock matrix mass is nearlyconstant which is a good decompaction adequacy

indicator for that given cell in the model Howeverthe so-measured adequacy changes throughout themodel In some areas the rock matrix mass has beenunderestimated by about 50 for nondecompactedpaleogeometries however in the case of hydrostaticdecompaction values generally range from 80 to120 of the present-day rock matrix mass Thatmeans in this particular case that hydrostatic decom-paction is more adequate (or less wrong) than nodecompaction at all The decompaction adequacycan be roughly correlated with pressure changesthrough geologic time

Figure 11 Time extractions for comparison of two end-member decompaction scenarios (see Figure 6 for location) (A) Rock matrixmass conservation through geologic time (B) Source rock burial depth through geologic time (C) Temperature and transformation ratiothrough geologic time

NEUMAIER ET AL 1347

Inadequate decompaction results in wrong depthfor stratigraphic horizons in the paleogeometriesFigure 11B shows the burial depth of the same inves-tigated source rock cell We can see that during thepassive margin phase there is a slight difference inpaleoburial depth for the two end-member scenariosDuring the compressional phase the differencechanges dramatically reaching more than 1000 m(3281 ft) in some places Toward present day bothcurves meet again because the input to the two resto-ration scenarios was the same (identical present-dayburial depth) This difference in compaction calcula-tion has severe implications on the modeled tempera-ture history and the kerogen-to-hydrocarbontransformation (Figure 11C) For an identical thermalframework in terms of modeling input parameterstemperature differences of up to 25degC (45degF) resultin TR differences of more than 20 in the early com-pressional phase before TR values meet again for thepresent-day situation

CONCLUSIONS

Structurally complex areas are attracting increasinginterest in hydrocarbon exploration Exploration riskcan be substantially narrowed by analyzing the his-tory of the basin and its petroleum systems based onstructural evolution We presented a further progressin technology that allows the integration of structuralrestoration and BPSM This combined analysis isseen as a way forward to improve modeling in thrustbelts with stacked stratigraphic sequences Despitelimitations of nonoptimized decompaction whichare added to the general uncertainty related to theindividual methods we modeled the charge and sealhistory of the Monagas fold and thrust belt inVenezuela Calibrated porosity temperature andthermal maturity as well as basin-scale stress pro-vided the regional framework through geologic timein which the petroleum systems evolved such thatthe charge of the known Santa Barbara and Onadofields and other trends could be understood We dis-cussed seal integrity of a potential exploration pros-pect by analyzing structural restoration-derived faultdisplacement through geologic time and an advancedstress forward simulation The uncertainty of using

decompacted or nondecompacted structural restora-tion has been assessed It can have important influ-ence on rock matrix mass balance through geologictime paleogeometry and paleoburial depth andtherefore affect paleotemperatures and paleomatur-ities We conclude that decompaction assuminghydrostatic pressure is the best option in that particu-lar case

REFERENCES CITED

Alberdi M and E Lafargue 1993 Vertical variations of organicmatter content in Guayuta Group (Upper Cretaceous)interior mountain belt eastern Venezuela OrganicGeochemistry v 20 p 425ndash436 doi 1010160146-6380(93)90091-O

Athy L F 1930 Density porosity and compaction of sedimen-tary rocks AAPG Bulletin v 14 p 1ndash24

Aymard R L Pimentel P Eitz P Lopez A ChaouchJ Navarro J Mijares and J G Pereira 1990 Geologic inte-gration and evaluation of Northern Monagas EasternVenezuelan Basin in J Brooks ed Classic petroleum prov-inces The Geologic Society (London) Special Publicationv 50 p 37ndash53

Baur F M Di Benedetto T Fuchs C Lampe andS Sciamanna 2009 Integrating structural geology andpetroleum systems modelingmdashA pilot project from Boliviarsquosfold and thrust belt Marine and Petroleum Geology v 26no 4 p 573ndash579 doi 101016jmarpetgeo200901004

Baur F R Littke H Wielens C Lampe and T Fuchs 2010Basin modeling meets rift analysismdashA numerical modelingstudy from the Jeanne drsquoArc basin offshore NewfoundlandCanada Marine and Petroleum Geology v 27 p 585ndash599doi 101016jmarpetgeo200906003

Beha A R O Thomsen and R Littke 2008 Thermal historyhydrocarbon generation and migration in the Horn Grabenin the Danish North Sea A 2D basin modelling studyInternational Journal of Earth Sciences v 97 p 1087ndash1100doi 101007s00531-007-0247-2

Blumenstein I O B M Krooss R di Primio W RottkeE Muumlller C Westerlage and R Littke 2008Biodegradation in numerical basin modelling A case studyfrom the Gifhorn Trough N-Germany International Journalof Earth Sciences v 97 p 1115ndash1129 doi 101007s00531-007-0272-1

Brandes C A Astorga R Littke and J Winsemann 2008Basin modelling of the Limoacuten back-arc basin (Costa Rica)Burial history and temperature evolution of an island arc-related basin-system Basin Research v 20 p 119ndash142doi 101111bre200820issue-1

Carnevali J O 1988 El Furrial oil field northeasternVenezuela First giant in foreland fold and thrust belts ofwestern hemisphere AAPG Bulletin v 72 68 p

Chevalier Y 1987 Contribution a lrsquoetude geacuteologique de lafrontiereacute Sud-Est de la plaque Caraibe Les zones internesde la chaine Sud-Caraibe sur le transect ile de Margarita-

1348 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

we assumed a constant initial TOC of 6 wt and aninitial HI of 500 mg HCg TOC However becauseinitial TOC and HI were estimated from present-daymeasured values these initial values are highly uncer-tain In addition there is both lateral and vertical vari-ability within the layer even though locally very highsource rock potential is likely to occur a bulk TOCand HI need to be assigned to the model cells Forthese reasons several models with different initialTOC have been run and their sensitivity to the mod-eled charge is discussed below Faults were set open

to fluid flow in the course of their main kinematicactivity which is estimated by fault displacementanalysis of the structural restoration models Aftertheir main activity phase faults are assumed to beclosed Fault permeability was also a varying param-eter in the sensitivity analysis discussed belowThermal boundary conditions are given in Table 2Surface temperature ranges between 23 and 30degC(734 and 86degF) through geologic time with apresent-day average temperature of 28degC (824degF)These temperatures at sea level have been corrected

Figure 3 Structural evolution of the Monagas fold and thrust belt demonstrated by structural restoration (A) Passive margin stage(until 23 Ma) (B) Early compression stage (21 Ma) (C) Late compression stage (12 Ma) (D) Present day For details see text

1332 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

Figure 4 Fault displacement from lateMiocene to present day for all faults onrestored cross sections (A) and for the Piritalthrust only (B) Black and white shading andsize of points show magnitude of displacementin meters

NEUMAIER ET AL 1333

Table1

Basin

andPetro

leum

Syste

mModeling(BPSM)Input

ParametersGeologicAges

andLithologies

Compressib

ility

(GPa

-1)

Perm

eability

(logmD)

Thermal

conductivity

(Wm

K)

Heatcapacity

(kcalkgK)

Age

(Ma)

GeologicTime

Facies

Density

(kg∕m

3 )Initial

Porosity(

)max

min

at1

por

at25por

at20degC

at100degC

at20degC

at100degC

0Pleisto

cene

LasP

iedras

sandsto

ne2720

412747

115

minus18

3395

338

02

024

LasPiedrasshale

2700

4040327

403

minus852

minus3

164

169

021

024

26

Pliocene

LasP

iedras

sandsto

ne2720

412747

115

minus18

3395

338

02

024

LasPiedrasshale

2700

4040327

403

minus852

minus3

164

169

021

024

53

upperM

iocene

2La

Pica

sandsto

ne2720

412747

115

minus18

3395

338

02

024

LaPica

shale

2700

4040327

403

minus852

minus3

164

169

021

024

Morich

itoshale

2700

4040327

403

minus852

minus3

164

169

021

024

8upperM

iocene

1La

Pica

sandsto

ne2720

412747

115

minus18

3395

338

02

024

LaPica

shale

2700

4040327

403

minus852

minus3

164

169

021

024

Morich

itobasal

sandsto

ne2715

4812142

187

minus348

15

317

284

02

024

105

middleMiocene

2Carapitasandsto

ne2715

4812142

187

minus348

15

317

284

02

024

Carapitashale

2705

6330932

331

minus684

minus15

204

201

02

024

12middleMiocene

1Carapitashale

2705

6330932

331

minus684

minus15

204

201

02

024

15lower

Miocene

2Carapitashale

2705

6330932

331

minus684

minus15

204

201

02

024

19lower

Miocene

1lower

Carapitashale

2700

7040327

403

minus852

minus3

164

169

021

024

23Oligocene2

Merecureplatform

sandsto

ne2715

4812142

187

minus348

15

317

284

02

024

27Oligocene1

Merecureplatform

sandsto

ne2715

4812142

187

minus348

15

317

284

02

024

1334 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

339

PaleoEocene

Vidono

sandsto

ne2720

412747

115

minus18

3395

338

02

024

Caratasbasin

alshale

2700

7040327

403

minus852

minus3

164

169

021

024

655

upperC

retaceous4

SanJuan

sandsto

ne2720

412747

115

minus18

3395

338

02

024

74upperC

retaceous3

SanAntoniosiltstone

(SOUR

CERO

CK)

2710

5510422

212

minus628

minus1

201

196

023

026

825

upperC

retaceous2

SanAntoniosiltstone

(SOUR

CERO

CK)

2710

5510422

212

minus628

minus1

201

196

023

026

91upperC

retaceous1

Querecualmudsto

ne(SOUR

CERO

CK)

2700

7040327

403

minus852

minus3

164

169

021

024

996

lower

Cretaceous

2undifflower

Cretaceous

(30

Lst50

Sst

20Sh)

2722

459445

141

minus334

174

305

275

02

023

122

lower

Cretaceous

1undifflower

Cretaceous

(30

Lst50

Sst

20Sh)

2722

459445

141

minus334

174

305

275

02

023

1455

sub-Cretaceous

2undiffsub-Cretaceous

(50

Sh50Sst)

2710

5521537

259

minus516

0255

239

02

024

251

NEUMAIER ET AL 1335

by the influence of the paleowater depth using theapproach of Wygrala (1989) Because of the hightopographic elevation of approximately 3500 m(11500 ft) the maximum altitude during the pastcompressional phase we adjusted the model takinginto account a temperature decrease of 6degCkm (33degF1000 ft) related to the paleoaltitude This resultsin a difference of up to 20degC (36degF) compared to sea-level temperature We then assigned the adjustedsedimentwater interface temperatures to the modelas the upper thermal boundary condition At the baseof the model we assigned a heat flow ranging from42 to 80 mW∕m2 through geologic time The highestheat flow is assumed during the rifting event beforethe basin cooled as a passive margin A thermal eventin the Eocene detected by fission track analysis(Locke and Garver 2005 Perez de Armas 2005)has been addressed by a second peak before a secondcooling phase For uncertainty analysis basal paleo-heat flow has been varied within geological meaning-ful ranges and the thermal calibration data has beenhonored as much as possible

Modeling Results

Basin and petroleum system modeling can be dividedinto simulation of the physical parameters within abasin such as pressure and temperature (basin model-ing) and the subsequent evolution of the source rocksand their derived hydrocarbon compounds (petro-leum systems modeling) In the following sectionswe use this subdivision to describe the BPSM resultscompaction modeling thermal modeling and chargehistory It is important to state that these are predic-tions based on forward modeling as just describedThe results are compared and calibrated to actualmeasurements and observations where available

Compaction Modeling

Figure 5 presents an extraction of the modeled resultsat the location of Well Alpha in the Furrial Trend Inthe absence of direct calibration data for pore pres-sure we undertook a sensitivity analysis to test thedifferent scenarios Using the model we tested howpore pressure in the main reservoir of the FurrialTrend reacts to changes in fault and seal-rock per-meability In a first step fault permeability has beenevaluated Faults were assumed open during the mainfault activity phase and closed afterward In additionwe performed several variations from that base casescenario Figure 5A shows the effect of two end-member scenarios on pore pressure within the mainreservoir of the Furrial Trend (Santa Barbara field) inwhich faults have been modeled to be either open orclosed since their formation A third scenario consistsof thrust faults that are open during the compressionalphase (from 15 Ma on) and closed afterward (from105 Ma on) This scenario has been chosen as theldquomaster scenariordquo because it results in overpressureof 23 MPa which is in the range of pressure reportedin similar structures of the Furrial Trend by Schneideret al (2004) In addition this pressure scenariomatches the calibration data for porosity (Figure 5B)

The 2D distribution of modeled overpressure atpresent day is shown in Figure 6A Apart fromimportant overpressure in the Querecual Formationthe allochthon is almost normally pressured Thethrust wedge is overpressured especially in thelow-permeability Carapita Formation Although

Table 2 Basin and Petroleum System Modeling (BPSM) InputParameters Thermal Boundary Conditions

Age (Ma)

SedimentWaterInterface

Temperature (degC)

Age (Ma)

Basal HeatFlow

(mW∕m2)min max

0 28 28 0 4226 28 28 30 6053 76 23 34 708 184 231 50 50105 62 233 107 6012 31 238 1455 8015 5 242 260 6019 5 24723 234 2527 78 235339 94 236655 146 2831455 148 287255 5 239

1336 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

the autochthon is highly overpressured close tonormal hydrostatic pressures are simulated in theforeland basin The discussed model assumption ofclosed faults in the postcompressional phase resultsin a regional pressure compartmentalization

The simulated pressure evolution through geo-logic time can be seen in Figure 7A A time extractionof a source rock cell in the thrust wedge shows a sud-den increase of both lithostatic and hydrostatic pres-sures that is a consequence of the tectonic burial

load associated with the thrusting In addition themodeled pore pressure buildup even though delayedin time can be directly linked to the thrust event Itcan be clearly seen that the timing of fault permeabil-ity directly impacts changes in the present-day porepressure The earlier the faults close the more over-pressure generated by the massive syntectonic sedi-mentation is kept until present day given adequatesealing lithologies are present The present-day over-pressure is modeled as14 MPa when the faults are

Figure 5 Plots of one-dimensional model output and available calibration data for Well Alpha (Table 3) (A) Dotted lines are hydro-static and lithostatic pressures continuous lines are pore pressure the master scenario is in black (a1 faults closing at 105 Ma) andadditional scenarios in gray (a2 open faults a3 closed faults) (B) modeled porosity (C) borehole temperature and (D) Ro The blackline shows the master scenario and the gray line the alternative high-heat-flow scenario Bold lines in the stratigraphic column arethrusts (T) and unconformities (U)

Table 3 Calibration Data for Well Alpha (Porosity Temperature Ro)

TVD (m) Porosity () TVD (m) Corrected Temperature (degC) TVD (m) Vitrinite Reflectance (Ro)

4350 15 600 40 400 0264400 15 1100 53 1200 0484800 10 1800 68 1485 068

4400 131 1700 0394800 133 2900 048

3100 0493400 0494000 054300 0524800 06

NEUMAIER ET AL 1337

open since 8 Ma as 23 MPa for 105 Ma and as37 MPa for 12 Ma Compared to this order of magni-tude changes in the permeability of the overlyingCarapita shale are negligible for the pore pressureTherefore the controlling parameter for pore pressurein the main reservoir of the Furrial trend is assumedfault permeability and resulting lateral connectivityrather than the permeability of the seal rock

Thermal Modeling

Modeled temperature and thermal maturity(Figure 5C D) have been calibrated with availablewell data Some discontinuities in the modeled Ro

trend are noted (Figure 5D) The upper kinks (egbase Pliocene) are related to unconformities withassociated erosion and the lower one (base UpperCretaceous) is due to the well penetrating a reverse

Figure 6 Two-dimensional model at present day with (A) modeled overpressure and (B) modeled thermal maturity and isothermsSource rocks marked by lithology pattern

1338 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

fault displacing mature rocks upon less mature ones(see Figure 6B for 2D view) There is a mismatchbetween the modeled Ro trend and the data in thedeeper part of the well In the Oligocene rocks themodeled temperature is too low (2degC or 36degF)whereas the modeled Ro is too high (up to 02 Ro)Any attempt to reduce the modeled Ro (eg bychanging thermal conductivity of the MioceneCarapita Formation) also results in a lower modeledpresent-day temperature Therefore the thermal sce-nario used for this calibration is a trade-off betweenmatching present-day temperature and Ro (using thekinetics of Sweeney and Burnham 1990) The basalheat flow assumed in the master scenario (Table 2)results in a modeled present-day surface heat flowranging from 37 mW∕m2 in the southern part of thethrusts to 52 mW∕m2 in parts of the uplifted areaAt the same depth the allochthonous northern partis thermally more mature than the southern autochthon(Figure 6B) because of the recent uplift Locallyinverted trends around the thrusts can be observedfor example at the location of Well Alpha (seeFigure 5D) In the autochthon the Upper Cretaceoussource rocks are entirely in the oil or even gas windowat present day because of the Neogene tectonic loadingand synorogenic deposition In the northern alloch-thon most of the Upper Cretaceous source rocks havebeen in the oil window in the past The modeled ther-mal maturity is in good agreement with the regionalmaturity suggested by Summa et al (2003)

Similar to the pressures the thrusting also affectsthe modeled temperature and Ro In the thrust wedgea direct relationship exists between the thrusting andthe increase of temperature and maturity The thrust-ing is the trigger for source rocks entering the oil win-dow within the thrust wedge (Figure 7B) and thepresent-day thermal maturity has been reached sincethe last 10 Ma Although this recent thermal historycan be considered as relatively well constrained bythe measured Ro in the calibration well this doesnot apply to the past In this thrusted area any ther-mal maturity reached prior to the thrusting onset hasbeen ldquooverprintedrdquo and therefore no thermal calibra-tion is possible for this period However theCretaceous source rocks were not buried deeply dur-ing the passive margin phase and even very hot sce-narios (basal heat flow of 100 mW∕m2) do not

significantly change their maturity (Figure 7B) Inthe uplifted allochthon several wells show low(03) to moderate (1) measured Ro (Parra et al2011) Because the allochthon experienced its maxi-mum burial and the highest temperatures during thepassive margin phase these maturity levels can beassumed as representative of the maturity prior tothe compression In contrast to the thrust wedge theconsequence of the thrusting is inverse for the alloch-thon (Figure 7C) as the hanging wall is uplifted andthe accompanying erosion removes the overburdenThe highest temperatures are simulated to have beenreached just before the thrusting event Then temper-ature decreased and Ro stagnated leaving the sourcerocks in the early mature zone until present day Theassumption of a regionally relatively homogeneousthermal field during the cooling phase of the passivemargin in combination with the relatively low matur-ities measured in the uplifted allochthon makes thehot heat-flow scenario unlikely However a laterallyvarying heat flow cannot be excluded and the conse-quently hotter paleoheat flow would have led to muchhigher temperatures prior to compression resultingalso in a higher present-day maturity level (14)despite the uplift (Figure 7C) Any variations of basalheat flow during the rift period (Late Jurassic to EarlyCretaceous) occurred too early to affect the maturityof the source rocks deposited during the lateCretaceous

Charge History

After calibration of the basinrsquos physical environ-ment the evolution of the petroleum systems wasmodeled Hydrocarbon generation was simulatedthrough geologic time using the activation energydistribution of the assigned source rock kineticsFrom the moment of hydrocarbon expulsiononward secondary migration was modeled usingthe invasion percolation method (Wilkinson 1984)It takes into account the geometry overpressure gra-dient and capillary entry pressures of the rockswhich are opposed to the buoyancy of the migratingoil and gas Figure 8 shows the time evolution of thetotal study area and Figure 9 focuses on the thrustwedge As already mentioned the described chargehistory is based on forward modeling It is consistent

NEUMAIER ET AL 1339

Figure 7 Time extractions (see Figure 6 for location) (A) Pressure development through geologic time for a source rock in the ver-tical continuity of Well Alpha Dotted lines are hydrostatic and lithostatic pressures continuous lines are pore pressure the master sce-nario is in black (a1 faults closing at 105 Ma) and additional scenarios in gray (a2 open faults a3 closed faults a4 faults closing at12 Ma a5 faults closing at 8 Ma) (B) Temperature (continuous lines) and Ro (dotted lines) through geologic time for a source rockin the vertical continuity of Well Alpha the black line shows the master scenario and the gray line the alternative high-heat-flow sce-nario (C) Temperature (continuous lines) and Ro (dotted lines) through geologic time for a source rock in the uplifted allochthonthe black line shows the master scenario and the gray line the alternative high heat flow scenario

1340 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

with the modeled compaction and thermal historyand therefore shows one scenario that explains thecharge of the Santa Barbara field and the forelandbasin Simplifications are inherent to any such work-flow and shall be considered when analyzing themodeling results

Passive Margin PhasemdashMesozoic toOligocene

Modeling results indicate that already during the LateCretaceous any Jurassic source rocks potentiallydeposited in the northern part of the section (Summaet al 2003) have reached temperatures of more than200degC (392degF) and corresponding high levels of matu-rity However the temperature at the base of theQuerecual Formation (Upper Cretaceous) was lessthan 100degC (212degF) leaving the kerogen immatureThe source rock became effective only during lateEocene with temperatures of up to 120degC (248degF)and a (reactive kerogen to hydrocarbon) transforma-tion ratio (TR) of more than 20 in the deeper distalpart of the passive margin Oil was quickly expelledmigrated vertically and seeped to the surface becauseof the general lack of seals in the overburden Duringthe Oligocene the time of deposition of the Merecurereservoir sandstones in the southern platform this sit-uation did not change apart from increased transforma-tion in the deeper basinal part of the passive margin

First oil accumulations are modeled in the Vidontildeosandstones in the basinal part in the late Oligocene(Figure 8A) in which the base of the QuerecualFormation shows TRs of up to 60 at around 130degC(266degF) In the proximal part there is no hydrocarbongeneration Although there is a general northwardtilting slope no southward migration from the distalmargin is modeled this has been checked usingdifferent migration methods including Darcy flowwith scenarios of continuous intrasource carrier bedsThis is in contrast to the studies from Gallango andParnaud (1995) and Schneider (2003) who suggestedlateral migration at that stage The difference couldbe explained by the presence of a regional top seal intheir models However our model suggests that theshallow rocks do not have sufficient sealing capacityto allow lateral hydrocarbon migration During thepassive-margin phase the basin was normally

pressured (Figure 7A) and simulated stresses havenot yet resulted in any fracturing at that stage

Main Compressional PhasemdashEarly to MiddleMiocene

With the onset of compression during early Miocene(Figure 8B) both the Pirital and Furrial thrusts wereformed and the northern distal part of the passivemargin was rapidly uplifted Important overpressuresas well as mechanical rock failure both the result ofthe rapid uplift are simulated in the allochthonDespite this uplift the kerogen-to-hydrocarbon trans-formation continued and the rate even increasedlocally close to the footwall cutoffs and where syn-orogenic deposition provided rapid overburden load-ing Oil accumulations are modeled in the freshlycreated anticlines

The tectonicndashsedimentary wedge between thetwo main thrusts experienced very rapid burial Infact the Upper Cretaceous source rocks were broughtinto depths where temperatures were as high as185degC (365degF) at 19 Ma and 225degC (437degF) at15 Ma Especially close to the cutoff of the Piritalthrust they very quickly reached a high maturity at19 Ma and reached full transformation in most placesat 12 Ma The expelled hydrocarbons migrated intothe sandstones of the Merecure Formation which weresealed by the Carapita Formation The charge toward aseries of recent anticlines is modeled by the fill-to-spillmechanism from north to south (Figure 9C D) Afterthe first oil fill the northern prospect quickly receiveda local gas charge the central structure containedmainly oil Note the presence of a lateral seal that isdue to sandndashshale juxtaposition The southernmoststructure which forms the Santa Barbara field todaywas charged by oil only at 15 Ma (Figure 9D)

The generated hydrocarbons in the southernautochthon were leaking to the surface until theCarapita Formation was compacted enough to holdsome of them (Figure 8C) At 12 Ma the wedgesource rocks were mostly spent but expulsion wasstill ongoing Important rock failure is modeled tohave occurred in the thrusted footwall breaking theformerly sealing Carapita Formation (Figure 10)This is in good agreement with cemented hydraulicfractures observed in a few cores in the Carapita

NEUMAIER ET AL 1341

Figure 8 Two-dimensional model through geologic time (selected time steps only) with modeled transformation ratio for the sourcerock layers isotherms and liquid and vapor migration and accumulation

1342 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

shales from the Furrial Trend (Roure et al 2005)This important fracturing event resulted in thedestruction of the modeled accumulation (Figure 9E)Hydrocarbon generation and expulsion were alsohappening in the southern foreland basin at transfor-mation ratios of up to 30

Late Compressional and PostcompressionalPhasemdashLate Miocene to Present Day

The late compressional phase was marked by the par-tial destruction of the allochthonous petroleum systemsby erosion (Figure 8E F) Many traps disappeared and

Figure 8 Continued

NEUMAIER ET AL 1343

the hydrocarbon generation within the source rocksstopped The erosion triggered an important sedimenttransfer toward the south contributing to further burialand increased temperatures in the thrust wedge and theforeland basin During late Miocene the petroleumsystem in the wedge matured An important northwardtilting was triggered by flexural tectonic loading andthe loading of syntectonic sediments (Summa et al2003) In combination with the presence of an effec-tive regional top seal the compacted CarapitaFormation allowed a change from dominantly verticalmigration into lateral migration First accumulationsare modeled in the Neogene synorogenic sequence

reservoirs in the south and important southwardmigration toward the foreland basin started(Figure 8G) These trends continued through Plioceneand Pleistocene times to present day filling theOnado field (Figure 8H) and other known fields inthe south (Figure 1A) of the Eastern Venezuela Basin

Prospectivity

We modeled the charge of the Santa Barbara field inthe Furrial Trend and of the Onado field in the fore-land basin (Figure 8H) Other prospects in the thrustwedge were also investigated The structure called

Figure 9 Detail of two-dimensionalmodel through geologic time (seeFigure 8) showing modeled transforma-tion ratio for the source rock layers iso-therms and liquid and vapor migrationand accumulation The Santa Barbara fieldand the prospect are shown by ellipses

1344 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

ldquoprospectrdquo (Figure 9G) looks similar to that of theSanta Barbara field because both are anticlines andthe reservoir and seal formations are the sameHowever they differ in several aspects First theprospect is in the vicinity of the Pirital thrust whichpresents a high risk of impacting the seal integrityThe prospect location is directly below the area wherewe presume the maximum of late tectonic activityoccurred (Figure 4) As modeled the CarapitaFormation the seal of the prospect was already brit-tle when this deformation occurred Calculatedstresses also predict rock failure from 12 Ma onward(Figure 10) Even though charge continued ourmodel suggests continuous leakage of the structure(Figure 9EndashG) The second major difference is thatthe source rock in the fault block of the prospect ismodeled to have already started generating and expel-ling hydrocarbons during the Oligocene and earlyMiocene (Figure 9A) before the trap formed and the

seal became effective (Figures 9C 10) Even thoughgeneration expulsion and migration continued aftertrap formation the prospect was not charged withthe earlier expelled oil In addition the present-daymaturity of the source rocks within the prospectrsquosfault block is higher than in the source rocks withinthe tectonic block of the Santa Barbara field poten-tially indicating expulsion of more gas A southndashnorth trend can be seen on the modeled accumulationcomposition with more gas toward the allochthon(Figure 9) The modeled composition of the prospectprior to seal failure was much richer in gas than inthe neighboring Santa Barbara field (Figure 9D) Infact simulation scenarios with unbreached sealsresulted in accumulation of predominantly gas Thecombination of both seal integrity and charge compo-sition concerns makes this a risky prospect In addi-tion potential diagenetic fluids related to the faultfluid flow might have altered the reservoir quality

Figure 10 Two-dimensional model (at 12 Ma) with modeled stress to failure Mohr circles (with σ standing for normal stress and τfor shear stress) and petroleum system events charts for the Santa Barbara field (left) and the prospect (right) Model predicts rock fail-ure in the prospect from about 12 Ma For further details see text

NEUMAIER ET AL 1345

Apart from the Onado oil field only minor accu-mulations are modeled in the southern Neogene syn-orogenic sequence The proportion of gas is higherthan in the Furrial trend The accumulations at thesouthern model edge are a modeling artifact repre-senting hydrocarbons (mainly oil) that in reality arethought to have charged the southern foreland basin(from mid to late Miocene times to present day)However parts of the migrated oil and gas could havebeen trapped in the Upper Cretaceous San JuanFormation south of the horst structure

The potential for conventional hydrocarbon accu-mulations in the allochthon is very limited by theimportant fracturing that occurred during the upliftperiod The basal sandstones of the Morochito piggy-back basin north of the Pirital thrust are not chargedin our 2D model because of the absence of structuraltraps Nevertheless there is a potential for strati-graphic traps to have been charged as migrationthrough the basin occurred However seal strengthmight be critical at this low-compaction stage Thesame applies for the Las Piedras Formation We thinkthere might be potential for shale oil because theQuerecual Formation is at shallow depths but showsmaturity within the oil window From the thermalmaturity point of view any potential Jurassic sourcerocks as discussed by Summa et al (2003) mightbe a gas source or shale gas target

UNCERTAINTY ANALYSIS

There is often significant uncertainty in basin geom-etry and depth of stratigraphic horizons based onuncertain seismic interpretation and velocity models(Kukla et al 2000 Brandes et al 2008 Baur et al2010) Restoring of present-day geometries to obtainpaleogeometries adds further uncertainty Additionaluncertainty is related to thermal boundary conditionssuch as paleoheat flow (Beha et al 2008) and togeochemical properties such as source rock character-istics and petroleum generation kinetics and calcula-tion of biodegradation and petroleum density(Blumenstein et al 2008) In addition to the porepressure and thermal sensitivity analysis justdescribed we varied initial TOC within a range of2 to 10 A variation of initial TOC obviously

directly scales the amount of initial kerogen availablefor transformation into hydrocarbons Neither thecharge volume nor composition of the Santa Barbarafield nor the wedge prospects are significantly sensi-tive to variation of initial TOC these traps are filledto spill in each scenario However the amount ofhydrocarbons charging the foreland basin (eg theOnado field) is reduced in the case of low initialTOC A high initial TOC only increased the modeledamount of surface seepage The sensitivity of the ini-tial HI is analogous to the sensitivity of the ini-tial TOC

In addition to the more classical uncertaintyand sensitivity analysis described we focused onquantitative analysis of the decompaction uncer-tainty when performing structural restoration Thisuncertainty is particular to the applied workflowof using structurally restored paleogeometries inBPSM Note that this uncertainty is significantlylower when using the backstripping method withthe optimization algorithms discussed One methodto analyze the adequacy of the decompactionmethod is comparison of the output rock matrixmass balance In the backstripping approach therock matrix mass is kept constant of a model cellwith a simulated decrease in porosity controlledby compaction laws and the corresponding reduc-tion of the layer thickness When using fixed pale-ogeometries from structural restoration the cellthickness is imposed However the decrease inporosity is calculated and the mass is adapted tofit the imposed cell thickness That means that therock matrix mass is constant only if ideal decom-paction has been applied

Figure 11A shows the percentage variation fromthe present-day matrix mass (100) of a given sourcerock cell through geologic time The dotted line rep-resents the ideal case with constant rock matrix massWe used two end-member decompaction scenarios todescribe the decompaction-related uncertainty that isintroduced when fixed paleogeometries replace back-stripping idealized hydrostatic decompaction assum-ing no overpressure and idealized lithostaticdecompaction assuming no effective stress (nodecompaction resulting in constant cell thicknessthrough time) In the latter case with decreasingporosity during compaction the rock matrix mass

1346 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

increases until it reaches the present-day mass Inother words rock mass matrix has been highly under-estimated in the past (in this particular case by morethan 20) This is a source rock so the kerogen massdecreases by the same factor This yields an underes-timation of more than 20 in TOC Note the rockmatrix mass is constant for time steps prior to23 Ma in which backstripping optimization couldbe performed toward the earliest paleogeometry Forthe second end-member scenario of idealized hydro-static decompaction the rock matrix mass is nearlyconstant which is a good decompaction adequacy

indicator for that given cell in the model Howeverthe so-measured adequacy changes throughout themodel In some areas the rock matrix mass has beenunderestimated by about 50 for nondecompactedpaleogeometries however in the case of hydrostaticdecompaction values generally range from 80 to120 of the present-day rock matrix mass Thatmeans in this particular case that hydrostatic decom-paction is more adequate (or less wrong) than nodecompaction at all The decompaction adequacycan be roughly correlated with pressure changesthrough geologic time

Figure 11 Time extractions for comparison of two end-member decompaction scenarios (see Figure 6 for location) (A) Rock matrixmass conservation through geologic time (B) Source rock burial depth through geologic time (C) Temperature and transformation ratiothrough geologic time

NEUMAIER ET AL 1347

Inadequate decompaction results in wrong depthfor stratigraphic horizons in the paleogeometriesFigure 11B shows the burial depth of the same inves-tigated source rock cell We can see that during thepassive margin phase there is a slight difference inpaleoburial depth for the two end-member scenariosDuring the compressional phase the differencechanges dramatically reaching more than 1000 m(3281 ft) in some places Toward present day bothcurves meet again because the input to the two resto-ration scenarios was the same (identical present-dayburial depth) This difference in compaction calcula-tion has severe implications on the modeled tempera-ture history and the kerogen-to-hydrocarbontransformation (Figure 11C) For an identical thermalframework in terms of modeling input parameterstemperature differences of up to 25degC (45degF) resultin TR differences of more than 20 in the early com-pressional phase before TR values meet again for thepresent-day situation

CONCLUSIONS

Structurally complex areas are attracting increasinginterest in hydrocarbon exploration Exploration riskcan be substantially narrowed by analyzing the his-tory of the basin and its petroleum systems based onstructural evolution We presented a further progressin technology that allows the integration of structuralrestoration and BPSM This combined analysis isseen as a way forward to improve modeling in thrustbelts with stacked stratigraphic sequences Despitelimitations of nonoptimized decompaction whichare added to the general uncertainty related to theindividual methods we modeled the charge and sealhistory of the Monagas fold and thrust belt inVenezuela Calibrated porosity temperature andthermal maturity as well as basin-scale stress pro-vided the regional framework through geologic timein which the petroleum systems evolved such thatthe charge of the known Santa Barbara and Onadofields and other trends could be understood We dis-cussed seal integrity of a potential exploration pros-pect by analyzing structural restoration-derived faultdisplacement through geologic time and an advancedstress forward simulation The uncertainty of using

decompacted or nondecompacted structural restora-tion has been assessed It can have important influ-ence on rock matrix mass balance through geologictime paleogeometry and paleoburial depth andtherefore affect paleotemperatures and paleomatur-ities We conclude that decompaction assuminghydrostatic pressure is the best option in that particu-lar case

REFERENCES CITED

Alberdi M and E Lafargue 1993 Vertical variations of organicmatter content in Guayuta Group (Upper Cretaceous)interior mountain belt eastern Venezuela OrganicGeochemistry v 20 p 425ndash436 doi 1010160146-6380(93)90091-O

Athy L F 1930 Density porosity and compaction of sedimen-tary rocks AAPG Bulletin v 14 p 1ndash24

Aymard R L Pimentel P Eitz P Lopez A ChaouchJ Navarro J Mijares and J G Pereira 1990 Geologic inte-gration and evaluation of Northern Monagas EasternVenezuelan Basin in J Brooks ed Classic petroleum prov-inces The Geologic Society (London) Special Publicationv 50 p 37ndash53

Baur F M Di Benedetto T Fuchs C Lampe andS Sciamanna 2009 Integrating structural geology andpetroleum systems modelingmdashA pilot project from Boliviarsquosfold and thrust belt Marine and Petroleum Geology v 26no 4 p 573ndash579 doi 101016jmarpetgeo200901004

Baur F R Littke H Wielens C Lampe and T Fuchs 2010Basin modeling meets rift analysismdashA numerical modelingstudy from the Jeanne drsquoArc basin offshore NewfoundlandCanada Marine and Petroleum Geology v 27 p 585ndash599doi 101016jmarpetgeo200906003

Beha A R O Thomsen and R Littke 2008 Thermal historyhydrocarbon generation and migration in the Horn Grabenin the Danish North Sea A 2D basin modelling studyInternational Journal of Earth Sciences v 97 p 1087ndash1100doi 101007s00531-007-0247-2

Blumenstein I O B M Krooss R di Primio W RottkeE Muumlller C Westerlage and R Littke 2008Biodegradation in numerical basin modelling A case studyfrom the Gifhorn Trough N-Germany International Journalof Earth Sciences v 97 p 1115ndash1129 doi 101007s00531-007-0272-1

Brandes C A Astorga R Littke and J Winsemann 2008Basin modelling of the Limoacuten back-arc basin (Costa Rica)Burial history and temperature evolution of an island arc-related basin-system Basin Research v 20 p 119ndash142doi 101111bre200820issue-1

Carnevali J O 1988 El Furrial oil field northeasternVenezuela First giant in foreland fold and thrust belts ofwestern hemisphere AAPG Bulletin v 72 68 p

Chevalier Y 1987 Contribution a lrsquoetude geacuteologique de lafrontiereacute Sud-Est de la plaque Caraibe Les zones internesde la chaine Sud-Caraibe sur le transect ile de Margarita-

1348 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

Figure 4 Fault displacement from lateMiocene to present day for all faults onrestored cross sections (A) and for the Piritalthrust only (B) Black and white shading andsize of points show magnitude of displacementin meters

NEUMAIER ET AL 1333

Table1

Basin

andPetro

leum

Syste

mModeling(BPSM)Input

ParametersGeologicAges

andLithologies

Compressib

ility

(GPa

-1)

Perm

eability

(logmD)

Thermal

conductivity

(Wm

K)

Heatcapacity

(kcalkgK)

Age

(Ma)

GeologicTime

Facies

Density

(kg∕m

3 )Initial

Porosity(

)max

min

at1

por

at25por

at20degC

at100degC

at20degC

at100degC

0Pleisto

cene

LasP

iedras

sandsto

ne2720

412747

115

minus18

3395

338

02

024

LasPiedrasshale

2700

4040327

403

minus852

minus3

164

169

021

024

26

Pliocene

LasP

iedras

sandsto

ne2720

412747

115

minus18

3395

338

02

024

LasPiedrasshale

2700

4040327

403

minus852

minus3

164

169

021

024

53

upperM

iocene

2La

Pica

sandsto

ne2720

412747

115

minus18

3395

338

02

024

LaPica

shale

2700

4040327

403

minus852

minus3

164

169

021

024

Morich

itoshale

2700

4040327

403

minus852

minus3

164

169

021

024

8upperM

iocene

1La

Pica

sandsto

ne2720

412747

115

minus18

3395

338

02

024

LaPica

shale

2700

4040327

403

minus852

minus3

164

169

021

024

Morich

itobasal

sandsto

ne2715

4812142

187

minus348

15

317

284

02

024

105

middleMiocene

2Carapitasandsto

ne2715

4812142

187

minus348

15

317

284

02

024

Carapitashale

2705

6330932

331

minus684

minus15

204

201

02

024

12middleMiocene

1Carapitashale

2705

6330932

331

minus684

minus15

204

201

02

024

15lower

Miocene

2Carapitashale

2705

6330932

331

minus684

minus15

204

201

02

024

19lower

Miocene

1lower

Carapitashale

2700

7040327

403

minus852

minus3

164

169

021

024

23Oligocene2

Merecureplatform

sandsto

ne2715

4812142

187

minus348

15

317

284

02

024

27Oligocene1

Merecureplatform

sandsto

ne2715

4812142

187

minus348

15

317

284

02

024

1334 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

339

PaleoEocene

Vidono

sandsto

ne2720

412747

115

minus18

3395

338

02

024

Caratasbasin

alshale

2700

7040327

403

minus852

minus3

164

169

021

024

655

upperC

retaceous4

SanJuan

sandsto

ne2720

412747

115

minus18

3395

338

02

024

74upperC

retaceous3

SanAntoniosiltstone

(SOUR

CERO

CK)

2710

5510422

212

minus628

minus1

201

196

023

026

825

upperC

retaceous2

SanAntoniosiltstone

(SOUR

CERO

CK)

2710

5510422

212

minus628

minus1

201

196

023

026

91upperC

retaceous1

Querecualmudsto

ne(SOUR

CERO

CK)

2700

7040327

403

minus852

minus3

164

169

021

024

996

lower

Cretaceous

2undifflower

Cretaceous

(30

Lst50

Sst

20Sh)

2722

459445

141

minus334

174

305

275

02

023

122

lower

Cretaceous

1undifflower

Cretaceous

(30

Lst50

Sst

20Sh)

2722

459445

141

minus334

174

305

275

02

023

1455

sub-Cretaceous

2undiffsub-Cretaceous

(50

Sh50Sst)

2710

5521537

259

minus516

0255

239

02

024

251

NEUMAIER ET AL 1335

by the influence of the paleowater depth using theapproach of Wygrala (1989) Because of the hightopographic elevation of approximately 3500 m(11500 ft) the maximum altitude during the pastcompressional phase we adjusted the model takinginto account a temperature decrease of 6degCkm (33degF1000 ft) related to the paleoaltitude This resultsin a difference of up to 20degC (36degF) compared to sea-level temperature We then assigned the adjustedsedimentwater interface temperatures to the modelas the upper thermal boundary condition At the baseof the model we assigned a heat flow ranging from42 to 80 mW∕m2 through geologic time The highestheat flow is assumed during the rifting event beforethe basin cooled as a passive margin A thermal eventin the Eocene detected by fission track analysis(Locke and Garver 2005 Perez de Armas 2005)has been addressed by a second peak before a secondcooling phase For uncertainty analysis basal paleo-heat flow has been varied within geological meaning-ful ranges and the thermal calibration data has beenhonored as much as possible

Modeling Results

Basin and petroleum system modeling can be dividedinto simulation of the physical parameters within abasin such as pressure and temperature (basin model-ing) and the subsequent evolution of the source rocksand their derived hydrocarbon compounds (petro-leum systems modeling) In the following sectionswe use this subdivision to describe the BPSM resultscompaction modeling thermal modeling and chargehistory It is important to state that these are predic-tions based on forward modeling as just describedThe results are compared and calibrated to actualmeasurements and observations where available

Compaction Modeling

Figure 5 presents an extraction of the modeled resultsat the location of Well Alpha in the Furrial Trend Inthe absence of direct calibration data for pore pres-sure we undertook a sensitivity analysis to test thedifferent scenarios Using the model we tested howpore pressure in the main reservoir of the FurrialTrend reacts to changes in fault and seal-rock per-meability In a first step fault permeability has beenevaluated Faults were assumed open during the mainfault activity phase and closed afterward In additionwe performed several variations from that base casescenario Figure 5A shows the effect of two end-member scenarios on pore pressure within the mainreservoir of the Furrial Trend (Santa Barbara field) inwhich faults have been modeled to be either open orclosed since their formation A third scenario consistsof thrust faults that are open during the compressionalphase (from 15 Ma on) and closed afterward (from105 Ma on) This scenario has been chosen as theldquomaster scenariordquo because it results in overpressureof 23 MPa which is in the range of pressure reportedin similar structures of the Furrial Trend by Schneideret al (2004) In addition this pressure scenariomatches the calibration data for porosity (Figure 5B)

The 2D distribution of modeled overpressure atpresent day is shown in Figure 6A Apart fromimportant overpressure in the Querecual Formationthe allochthon is almost normally pressured Thethrust wedge is overpressured especially in thelow-permeability Carapita Formation Although

Table 2 Basin and Petroleum System Modeling (BPSM) InputParameters Thermal Boundary Conditions

Age (Ma)

SedimentWaterInterface

Temperature (degC)

Age (Ma)

Basal HeatFlow

(mW∕m2)min max

0 28 28 0 4226 28 28 30 6053 76 23 34 708 184 231 50 50105 62 233 107 6012 31 238 1455 8015 5 242 260 6019 5 24723 234 2527 78 235339 94 236655 146 2831455 148 287255 5 239

1336 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

the autochthon is highly overpressured close tonormal hydrostatic pressures are simulated in theforeland basin The discussed model assumption ofclosed faults in the postcompressional phase resultsin a regional pressure compartmentalization

The simulated pressure evolution through geo-logic time can be seen in Figure 7A A time extractionof a source rock cell in the thrust wedge shows a sud-den increase of both lithostatic and hydrostatic pres-sures that is a consequence of the tectonic burial

load associated with the thrusting In addition themodeled pore pressure buildup even though delayedin time can be directly linked to the thrust event Itcan be clearly seen that the timing of fault permeabil-ity directly impacts changes in the present-day porepressure The earlier the faults close the more over-pressure generated by the massive syntectonic sedi-mentation is kept until present day given adequatesealing lithologies are present The present-day over-pressure is modeled as14 MPa when the faults are

Figure 5 Plots of one-dimensional model output and available calibration data for Well Alpha (Table 3) (A) Dotted lines are hydro-static and lithostatic pressures continuous lines are pore pressure the master scenario is in black (a1 faults closing at 105 Ma) andadditional scenarios in gray (a2 open faults a3 closed faults) (B) modeled porosity (C) borehole temperature and (D) Ro The blackline shows the master scenario and the gray line the alternative high-heat-flow scenario Bold lines in the stratigraphic column arethrusts (T) and unconformities (U)

Table 3 Calibration Data for Well Alpha (Porosity Temperature Ro)

TVD (m) Porosity () TVD (m) Corrected Temperature (degC) TVD (m) Vitrinite Reflectance (Ro)

4350 15 600 40 400 0264400 15 1100 53 1200 0484800 10 1800 68 1485 068

4400 131 1700 0394800 133 2900 048

3100 0493400 0494000 054300 0524800 06

NEUMAIER ET AL 1337

open since 8 Ma as 23 MPa for 105 Ma and as37 MPa for 12 Ma Compared to this order of magni-tude changes in the permeability of the overlyingCarapita shale are negligible for the pore pressureTherefore the controlling parameter for pore pressurein the main reservoir of the Furrial trend is assumedfault permeability and resulting lateral connectivityrather than the permeability of the seal rock

Thermal Modeling

Modeled temperature and thermal maturity(Figure 5C D) have been calibrated with availablewell data Some discontinuities in the modeled Ro

trend are noted (Figure 5D) The upper kinks (egbase Pliocene) are related to unconformities withassociated erosion and the lower one (base UpperCretaceous) is due to the well penetrating a reverse

Figure 6 Two-dimensional model at present day with (A) modeled overpressure and (B) modeled thermal maturity and isothermsSource rocks marked by lithology pattern

1338 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

fault displacing mature rocks upon less mature ones(see Figure 6B for 2D view) There is a mismatchbetween the modeled Ro trend and the data in thedeeper part of the well In the Oligocene rocks themodeled temperature is too low (2degC or 36degF)whereas the modeled Ro is too high (up to 02 Ro)Any attempt to reduce the modeled Ro (eg bychanging thermal conductivity of the MioceneCarapita Formation) also results in a lower modeledpresent-day temperature Therefore the thermal sce-nario used for this calibration is a trade-off betweenmatching present-day temperature and Ro (using thekinetics of Sweeney and Burnham 1990) The basalheat flow assumed in the master scenario (Table 2)results in a modeled present-day surface heat flowranging from 37 mW∕m2 in the southern part of thethrusts to 52 mW∕m2 in parts of the uplifted areaAt the same depth the allochthonous northern partis thermally more mature than the southern autochthon(Figure 6B) because of the recent uplift Locallyinverted trends around the thrusts can be observedfor example at the location of Well Alpha (seeFigure 5D) In the autochthon the Upper Cretaceoussource rocks are entirely in the oil or even gas windowat present day because of the Neogene tectonic loadingand synorogenic deposition In the northern alloch-thon most of the Upper Cretaceous source rocks havebeen in the oil window in the past The modeled ther-mal maturity is in good agreement with the regionalmaturity suggested by Summa et al (2003)

Similar to the pressures the thrusting also affectsthe modeled temperature and Ro In the thrust wedgea direct relationship exists between the thrusting andthe increase of temperature and maturity The thrust-ing is the trigger for source rocks entering the oil win-dow within the thrust wedge (Figure 7B) and thepresent-day thermal maturity has been reached sincethe last 10 Ma Although this recent thermal historycan be considered as relatively well constrained bythe measured Ro in the calibration well this doesnot apply to the past In this thrusted area any ther-mal maturity reached prior to the thrusting onset hasbeen ldquooverprintedrdquo and therefore no thermal calibra-tion is possible for this period However theCretaceous source rocks were not buried deeply dur-ing the passive margin phase and even very hot sce-narios (basal heat flow of 100 mW∕m2) do not

significantly change their maturity (Figure 7B) Inthe uplifted allochthon several wells show low(03) to moderate (1) measured Ro (Parra et al2011) Because the allochthon experienced its maxi-mum burial and the highest temperatures during thepassive margin phase these maturity levels can beassumed as representative of the maturity prior tothe compression In contrast to the thrust wedge theconsequence of the thrusting is inverse for the alloch-thon (Figure 7C) as the hanging wall is uplifted andthe accompanying erosion removes the overburdenThe highest temperatures are simulated to have beenreached just before the thrusting event Then temper-ature decreased and Ro stagnated leaving the sourcerocks in the early mature zone until present day Theassumption of a regionally relatively homogeneousthermal field during the cooling phase of the passivemargin in combination with the relatively low matur-ities measured in the uplifted allochthon makes thehot heat-flow scenario unlikely However a laterallyvarying heat flow cannot be excluded and the conse-quently hotter paleoheat flow would have led to muchhigher temperatures prior to compression resultingalso in a higher present-day maturity level (14)despite the uplift (Figure 7C) Any variations of basalheat flow during the rift period (Late Jurassic to EarlyCretaceous) occurred too early to affect the maturityof the source rocks deposited during the lateCretaceous

Charge History

After calibration of the basinrsquos physical environ-ment the evolution of the petroleum systems wasmodeled Hydrocarbon generation was simulatedthrough geologic time using the activation energydistribution of the assigned source rock kineticsFrom the moment of hydrocarbon expulsiononward secondary migration was modeled usingthe invasion percolation method (Wilkinson 1984)It takes into account the geometry overpressure gra-dient and capillary entry pressures of the rockswhich are opposed to the buoyancy of the migratingoil and gas Figure 8 shows the time evolution of thetotal study area and Figure 9 focuses on the thrustwedge As already mentioned the described chargehistory is based on forward modeling It is consistent

NEUMAIER ET AL 1339

Figure 7 Time extractions (see Figure 6 for location) (A) Pressure development through geologic time for a source rock in the ver-tical continuity of Well Alpha Dotted lines are hydrostatic and lithostatic pressures continuous lines are pore pressure the master sce-nario is in black (a1 faults closing at 105 Ma) and additional scenarios in gray (a2 open faults a3 closed faults a4 faults closing at12 Ma a5 faults closing at 8 Ma) (B) Temperature (continuous lines) and Ro (dotted lines) through geologic time for a source rockin the vertical continuity of Well Alpha the black line shows the master scenario and the gray line the alternative high-heat-flow sce-nario (C) Temperature (continuous lines) and Ro (dotted lines) through geologic time for a source rock in the uplifted allochthonthe black line shows the master scenario and the gray line the alternative high heat flow scenario

1340 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

with the modeled compaction and thermal historyand therefore shows one scenario that explains thecharge of the Santa Barbara field and the forelandbasin Simplifications are inherent to any such work-flow and shall be considered when analyzing themodeling results

Passive Margin PhasemdashMesozoic toOligocene

Modeling results indicate that already during the LateCretaceous any Jurassic source rocks potentiallydeposited in the northern part of the section (Summaet al 2003) have reached temperatures of more than200degC (392degF) and corresponding high levels of matu-rity However the temperature at the base of theQuerecual Formation (Upper Cretaceous) was lessthan 100degC (212degF) leaving the kerogen immatureThe source rock became effective only during lateEocene with temperatures of up to 120degC (248degF)and a (reactive kerogen to hydrocarbon) transforma-tion ratio (TR) of more than 20 in the deeper distalpart of the passive margin Oil was quickly expelledmigrated vertically and seeped to the surface becauseof the general lack of seals in the overburden Duringthe Oligocene the time of deposition of the Merecurereservoir sandstones in the southern platform this sit-uation did not change apart from increased transforma-tion in the deeper basinal part of the passive margin

First oil accumulations are modeled in the Vidontildeosandstones in the basinal part in the late Oligocene(Figure 8A) in which the base of the QuerecualFormation shows TRs of up to 60 at around 130degC(266degF) In the proximal part there is no hydrocarbongeneration Although there is a general northwardtilting slope no southward migration from the distalmargin is modeled this has been checked usingdifferent migration methods including Darcy flowwith scenarios of continuous intrasource carrier bedsThis is in contrast to the studies from Gallango andParnaud (1995) and Schneider (2003) who suggestedlateral migration at that stage The difference couldbe explained by the presence of a regional top seal intheir models However our model suggests that theshallow rocks do not have sufficient sealing capacityto allow lateral hydrocarbon migration During thepassive-margin phase the basin was normally

pressured (Figure 7A) and simulated stresses havenot yet resulted in any fracturing at that stage

Main Compressional PhasemdashEarly to MiddleMiocene

With the onset of compression during early Miocene(Figure 8B) both the Pirital and Furrial thrusts wereformed and the northern distal part of the passivemargin was rapidly uplifted Important overpressuresas well as mechanical rock failure both the result ofthe rapid uplift are simulated in the allochthonDespite this uplift the kerogen-to-hydrocarbon trans-formation continued and the rate even increasedlocally close to the footwall cutoffs and where syn-orogenic deposition provided rapid overburden load-ing Oil accumulations are modeled in the freshlycreated anticlines

The tectonicndashsedimentary wedge between thetwo main thrusts experienced very rapid burial Infact the Upper Cretaceous source rocks were broughtinto depths where temperatures were as high as185degC (365degF) at 19 Ma and 225degC (437degF) at15 Ma Especially close to the cutoff of the Piritalthrust they very quickly reached a high maturity at19 Ma and reached full transformation in most placesat 12 Ma The expelled hydrocarbons migrated intothe sandstones of the Merecure Formation which weresealed by the Carapita Formation The charge toward aseries of recent anticlines is modeled by the fill-to-spillmechanism from north to south (Figure 9C D) Afterthe first oil fill the northern prospect quickly receiveda local gas charge the central structure containedmainly oil Note the presence of a lateral seal that isdue to sandndashshale juxtaposition The southernmoststructure which forms the Santa Barbara field todaywas charged by oil only at 15 Ma (Figure 9D)

The generated hydrocarbons in the southernautochthon were leaking to the surface until theCarapita Formation was compacted enough to holdsome of them (Figure 8C) At 12 Ma the wedgesource rocks were mostly spent but expulsion wasstill ongoing Important rock failure is modeled tohave occurred in the thrusted footwall breaking theformerly sealing Carapita Formation (Figure 10)This is in good agreement with cemented hydraulicfractures observed in a few cores in the Carapita

NEUMAIER ET AL 1341

Figure 8 Two-dimensional model through geologic time (selected time steps only) with modeled transformation ratio for the sourcerock layers isotherms and liquid and vapor migration and accumulation

1342 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

shales from the Furrial Trend (Roure et al 2005)This important fracturing event resulted in thedestruction of the modeled accumulation (Figure 9E)Hydrocarbon generation and expulsion were alsohappening in the southern foreland basin at transfor-mation ratios of up to 30

Late Compressional and PostcompressionalPhasemdashLate Miocene to Present Day

The late compressional phase was marked by the par-tial destruction of the allochthonous petroleum systemsby erosion (Figure 8E F) Many traps disappeared and

Figure 8 Continued

NEUMAIER ET AL 1343

the hydrocarbon generation within the source rocksstopped The erosion triggered an important sedimenttransfer toward the south contributing to further burialand increased temperatures in the thrust wedge and theforeland basin During late Miocene the petroleumsystem in the wedge matured An important northwardtilting was triggered by flexural tectonic loading andthe loading of syntectonic sediments (Summa et al2003) In combination with the presence of an effec-tive regional top seal the compacted CarapitaFormation allowed a change from dominantly verticalmigration into lateral migration First accumulationsare modeled in the Neogene synorogenic sequence

reservoirs in the south and important southwardmigration toward the foreland basin started(Figure 8G) These trends continued through Plioceneand Pleistocene times to present day filling theOnado field (Figure 8H) and other known fields inthe south (Figure 1A) of the Eastern Venezuela Basin

Prospectivity

We modeled the charge of the Santa Barbara field inthe Furrial Trend and of the Onado field in the fore-land basin (Figure 8H) Other prospects in the thrustwedge were also investigated The structure called

Figure 9 Detail of two-dimensionalmodel through geologic time (seeFigure 8) showing modeled transforma-tion ratio for the source rock layers iso-therms and liquid and vapor migrationand accumulation The Santa Barbara fieldand the prospect are shown by ellipses

1344 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

ldquoprospectrdquo (Figure 9G) looks similar to that of theSanta Barbara field because both are anticlines andthe reservoir and seal formations are the sameHowever they differ in several aspects First theprospect is in the vicinity of the Pirital thrust whichpresents a high risk of impacting the seal integrityThe prospect location is directly below the area wherewe presume the maximum of late tectonic activityoccurred (Figure 4) As modeled the CarapitaFormation the seal of the prospect was already brit-tle when this deformation occurred Calculatedstresses also predict rock failure from 12 Ma onward(Figure 10) Even though charge continued ourmodel suggests continuous leakage of the structure(Figure 9EndashG) The second major difference is thatthe source rock in the fault block of the prospect ismodeled to have already started generating and expel-ling hydrocarbons during the Oligocene and earlyMiocene (Figure 9A) before the trap formed and the

seal became effective (Figures 9C 10) Even thoughgeneration expulsion and migration continued aftertrap formation the prospect was not charged withthe earlier expelled oil In addition the present-daymaturity of the source rocks within the prospectrsquosfault block is higher than in the source rocks withinthe tectonic block of the Santa Barbara field poten-tially indicating expulsion of more gas A southndashnorth trend can be seen on the modeled accumulationcomposition with more gas toward the allochthon(Figure 9) The modeled composition of the prospectprior to seal failure was much richer in gas than inthe neighboring Santa Barbara field (Figure 9D) Infact simulation scenarios with unbreached sealsresulted in accumulation of predominantly gas Thecombination of both seal integrity and charge compo-sition concerns makes this a risky prospect In addi-tion potential diagenetic fluids related to the faultfluid flow might have altered the reservoir quality

Figure 10 Two-dimensional model (at 12 Ma) with modeled stress to failure Mohr circles (with σ standing for normal stress and τfor shear stress) and petroleum system events charts for the Santa Barbara field (left) and the prospect (right) Model predicts rock fail-ure in the prospect from about 12 Ma For further details see text

NEUMAIER ET AL 1345

Apart from the Onado oil field only minor accu-mulations are modeled in the southern Neogene syn-orogenic sequence The proportion of gas is higherthan in the Furrial trend The accumulations at thesouthern model edge are a modeling artifact repre-senting hydrocarbons (mainly oil) that in reality arethought to have charged the southern foreland basin(from mid to late Miocene times to present day)However parts of the migrated oil and gas could havebeen trapped in the Upper Cretaceous San JuanFormation south of the horst structure

The potential for conventional hydrocarbon accu-mulations in the allochthon is very limited by theimportant fracturing that occurred during the upliftperiod The basal sandstones of the Morochito piggy-back basin north of the Pirital thrust are not chargedin our 2D model because of the absence of structuraltraps Nevertheless there is a potential for strati-graphic traps to have been charged as migrationthrough the basin occurred However seal strengthmight be critical at this low-compaction stage Thesame applies for the Las Piedras Formation We thinkthere might be potential for shale oil because theQuerecual Formation is at shallow depths but showsmaturity within the oil window From the thermalmaturity point of view any potential Jurassic sourcerocks as discussed by Summa et al (2003) mightbe a gas source or shale gas target

UNCERTAINTY ANALYSIS

There is often significant uncertainty in basin geom-etry and depth of stratigraphic horizons based onuncertain seismic interpretation and velocity models(Kukla et al 2000 Brandes et al 2008 Baur et al2010) Restoring of present-day geometries to obtainpaleogeometries adds further uncertainty Additionaluncertainty is related to thermal boundary conditionssuch as paleoheat flow (Beha et al 2008) and togeochemical properties such as source rock character-istics and petroleum generation kinetics and calcula-tion of biodegradation and petroleum density(Blumenstein et al 2008) In addition to the porepressure and thermal sensitivity analysis justdescribed we varied initial TOC within a range of2 to 10 A variation of initial TOC obviously

directly scales the amount of initial kerogen availablefor transformation into hydrocarbons Neither thecharge volume nor composition of the Santa Barbarafield nor the wedge prospects are significantly sensi-tive to variation of initial TOC these traps are filledto spill in each scenario However the amount ofhydrocarbons charging the foreland basin (eg theOnado field) is reduced in the case of low initialTOC A high initial TOC only increased the modeledamount of surface seepage The sensitivity of the ini-tial HI is analogous to the sensitivity of the ini-tial TOC

In addition to the more classical uncertaintyand sensitivity analysis described we focused onquantitative analysis of the decompaction uncer-tainty when performing structural restoration Thisuncertainty is particular to the applied workflowof using structurally restored paleogeometries inBPSM Note that this uncertainty is significantlylower when using the backstripping method withthe optimization algorithms discussed One methodto analyze the adequacy of the decompactionmethod is comparison of the output rock matrixmass balance In the backstripping approach therock matrix mass is kept constant of a model cellwith a simulated decrease in porosity controlledby compaction laws and the corresponding reduc-tion of the layer thickness When using fixed pale-ogeometries from structural restoration the cellthickness is imposed However the decrease inporosity is calculated and the mass is adapted tofit the imposed cell thickness That means that therock matrix mass is constant only if ideal decom-paction has been applied

Figure 11A shows the percentage variation fromthe present-day matrix mass (100) of a given sourcerock cell through geologic time The dotted line rep-resents the ideal case with constant rock matrix massWe used two end-member decompaction scenarios todescribe the decompaction-related uncertainty that isintroduced when fixed paleogeometries replace back-stripping idealized hydrostatic decompaction assum-ing no overpressure and idealized lithostaticdecompaction assuming no effective stress (nodecompaction resulting in constant cell thicknessthrough time) In the latter case with decreasingporosity during compaction the rock matrix mass

1346 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

increases until it reaches the present-day mass Inother words rock mass matrix has been highly under-estimated in the past (in this particular case by morethan 20) This is a source rock so the kerogen massdecreases by the same factor This yields an underes-timation of more than 20 in TOC Note the rockmatrix mass is constant for time steps prior to23 Ma in which backstripping optimization couldbe performed toward the earliest paleogeometry Forthe second end-member scenario of idealized hydro-static decompaction the rock matrix mass is nearlyconstant which is a good decompaction adequacy

indicator for that given cell in the model Howeverthe so-measured adequacy changes throughout themodel In some areas the rock matrix mass has beenunderestimated by about 50 for nondecompactedpaleogeometries however in the case of hydrostaticdecompaction values generally range from 80 to120 of the present-day rock matrix mass Thatmeans in this particular case that hydrostatic decom-paction is more adequate (or less wrong) than nodecompaction at all The decompaction adequacycan be roughly correlated with pressure changesthrough geologic time

Figure 11 Time extractions for comparison of two end-member decompaction scenarios (see Figure 6 for location) (A) Rock matrixmass conservation through geologic time (B) Source rock burial depth through geologic time (C) Temperature and transformation ratiothrough geologic time

NEUMAIER ET AL 1347

Inadequate decompaction results in wrong depthfor stratigraphic horizons in the paleogeometriesFigure 11B shows the burial depth of the same inves-tigated source rock cell We can see that during thepassive margin phase there is a slight difference inpaleoburial depth for the two end-member scenariosDuring the compressional phase the differencechanges dramatically reaching more than 1000 m(3281 ft) in some places Toward present day bothcurves meet again because the input to the two resto-ration scenarios was the same (identical present-dayburial depth) This difference in compaction calcula-tion has severe implications on the modeled tempera-ture history and the kerogen-to-hydrocarbontransformation (Figure 11C) For an identical thermalframework in terms of modeling input parameterstemperature differences of up to 25degC (45degF) resultin TR differences of more than 20 in the early com-pressional phase before TR values meet again for thepresent-day situation

CONCLUSIONS

Structurally complex areas are attracting increasinginterest in hydrocarbon exploration Exploration riskcan be substantially narrowed by analyzing the his-tory of the basin and its petroleum systems based onstructural evolution We presented a further progressin technology that allows the integration of structuralrestoration and BPSM This combined analysis isseen as a way forward to improve modeling in thrustbelts with stacked stratigraphic sequences Despitelimitations of nonoptimized decompaction whichare added to the general uncertainty related to theindividual methods we modeled the charge and sealhistory of the Monagas fold and thrust belt inVenezuela Calibrated porosity temperature andthermal maturity as well as basin-scale stress pro-vided the regional framework through geologic timein which the petroleum systems evolved such thatthe charge of the known Santa Barbara and Onadofields and other trends could be understood We dis-cussed seal integrity of a potential exploration pros-pect by analyzing structural restoration-derived faultdisplacement through geologic time and an advancedstress forward simulation The uncertainty of using

decompacted or nondecompacted structural restora-tion has been assessed It can have important influ-ence on rock matrix mass balance through geologictime paleogeometry and paleoburial depth andtherefore affect paleotemperatures and paleomatur-ities We conclude that decompaction assuminghydrostatic pressure is the best option in that particu-lar case

REFERENCES CITED

Alberdi M and E Lafargue 1993 Vertical variations of organicmatter content in Guayuta Group (Upper Cretaceous)interior mountain belt eastern Venezuela OrganicGeochemistry v 20 p 425ndash436 doi 1010160146-6380(93)90091-O

Athy L F 1930 Density porosity and compaction of sedimen-tary rocks AAPG Bulletin v 14 p 1ndash24

Aymard R L Pimentel P Eitz P Lopez A ChaouchJ Navarro J Mijares and J G Pereira 1990 Geologic inte-gration and evaluation of Northern Monagas EasternVenezuelan Basin in J Brooks ed Classic petroleum prov-inces The Geologic Society (London) Special Publicationv 50 p 37ndash53

Baur F M Di Benedetto T Fuchs C Lampe andS Sciamanna 2009 Integrating structural geology andpetroleum systems modelingmdashA pilot project from Boliviarsquosfold and thrust belt Marine and Petroleum Geology v 26no 4 p 573ndash579 doi 101016jmarpetgeo200901004

Baur F R Littke H Wielens C Lampe and T Fuchs 2010Basin modeling meets rift analysismdashA numerical modelingstudy from the Jeanne drsquoArc basin offshore NewfoundlandCanada Marine and Petroleum Geology v 27 p 585ndash599doi 101016jmarpetgeo200906003

Beha A R O Thomsen and R Littke 2008 Thermal historyhydrocarbon generation and migration in the Horn Grabenin the Danish North Sea A 2D basin modelling studyInternational Journal of Earth Sciences v 97 p 1087ndash1100doi 101007s00531-007-0247-2

Blumenstein I O B M Krooss R di Primio W RottkeE Muumlller C Westerlage and R Littke 2008Biodegradation in numerical basin modelling A case studyfrom the Gifhorn Trough N-Germany International Journalof Earth Sciences v 97 p 1115ndash1129 doi 101007s00531-007-0272-1

Brandes C A Astorga R Littke and J Winsemann 2008Basin modelling of the Limoacuten back-arc basin (Costa Rica)Burial history and temperature evolution of an island arc-related basin-system Basin Research v 20 p 119ndash142doi 101111bre200820issue-1

Carnevali J O 1988 El Furrial oil field northeasternVenezuela First giant in foreland fold and thrust belts ofwestern hemisphere AAPG Bulletin v 72 68 p

Chevalier Y 1987 Contribution a lrsquoetude geacuteologique de lafrontiereacute Sud-Est de la plaque Caraibe Les zones internesde la chaine Sud-Caraibe sur le transect ile de Margarita-

1348 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

Table1

Basin

andPetro

leum

Syste

mModeling(BPSM)Input

ParametersGeologicAges

andLithologies

Compressib

ility

(GPa

-1)

Perm

eability

(logmD)

Thermal

conductivity

(Wm

K)

Heatcapacity

(kcalkgK)

Age

(Ma)

GeologicTime

Facies

Density

(kg∕m

3 )Initial

Porosity(

)max

min

at1

por

at25por

at20degC

at100degC

at20degC

at100degC

0Pleisto

cene

LasP

iedras

sandsto

ne2720

412747

115

minus18

3395

338

02

024

LasPiedrasshale

2700

4040327

403

minus852

minus3

164

169

021

024

26

Pliocene

LasP

iedras

sandsto

ne2720

412747

115

minus18

3395

338

02

024

LasPiedrasshale

2700

4040327

403

minus852

minus3

164

169

021

024

53

upperM

iocene

2La

Pica

sandsto

ne2720

412747

115

minus18

3395

338

02

024

LaPica

shale

2700

4040327

403

minus852

minus3

164

169

021

024

Morich

itoshale

2700

4040327

403

minus852

minus3

164

169

021

024

8upperM

iocene

1La

Pica

sandsto

ne2720

412747

115

minus18

3395

338

02

024

LaPica

shale

2700

4040327

403

minus852

minus3

164

169

021

024

Morich

itobasal

sandsto

ne2715

4812142

187

minus348

15

317

284

02

024

105

middleMiocene

2Carapitasandsto

ne2715

4812142

187

minus348

15

317

284

02

024

Carapitashale

2705

6330932

331

minus684

minus15

204

201

02

024

12middleMiocene

1Carapitashale

2705

6330932

331

minus684

minus15

204

201

02

024

15lower

Miocene

2Carapitashale

2705

6330932

331

minus684

minus15

204

201

02

024

19lower

Miocene

1lower

Carapitashale

2700

7040327

403

minus852

minus3

164

169

021

024

23Oligocene2

Merecureplatform

sandsto

ne2715

4812142

187

minus348

15

317

284

02

024

27Oligocene1

Merecureplatform

sandsto

ne2715

4812142

187

minus348

15

317

284

02

024

1334 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

339

PaleoEocene

Vidono

sandsto

ne2720

412747

115

minus18

3395

338

02

024

Caratasbasin

alshale

2700

7040327

403

minus852

minus3

164

169

021

024

655

upperC

retaceous4

SanJuan

sandsto

ne2720

412747

115

minus18

3395

338

02

024

74upperC

retaceous3

SanAntoniosiltstone

(SOUR

CERO

CK)

2710

5510422

212

minus628

minus1

201

196

023

026

825

upperC

retaceous2

SanAntoniosiltstone

(SOUR

CERO

CK)

2710

5510422

212

minus628

minus1

201

196

023

026

91upperC

retaceous1

Querecualmudsto

ne(SOUR

CERO

CK)

2700

7040327

403

minus852

minus3

164

169

021

024

996

lower

Cretaceous

2undifflower

Cretaceous

(30

Lst50

Sst

20Sh)

2722

459445

141

minus334

174

305

275

02

023

122

lower

Cretaceous

1undifflower

Cretaceous

(30

Lst50

Sst

20Sh)

2722

459445

141

minus334

174

305

275

02

023

1455

sub-Cretaceous

2undiffsub-Cretaceous

(50

Sh50Sst)

2710

5521537

259

minus516

0255

239

02

024

251

NEUMAIER ET AL 1335

by the influence of the paleowater depth using theapproach of Wygrala (1989) Because of the hightopographic elevation of approximately 3500 m(11500 ft) the maximum altitude during the pastcompressional phase we adjusted the model takinginto account a temperature decrease of 6degCkm (33degF1000 ft) related to the paleoaltitude This resultsin a difference of up to 20degC (36degF) compared to sea-level temperature We then assigned the adjustedsedimentwater interface temperatures to the modelas the upper thermal boundary condition At the baseof the model we assigned a heat flow ranging from42 to 80 mW∕m2 through geologic time The highestheat flow is assumed during the rifting event beforethe basin cooled as a passive margin A thermal eventin the Eocene detected by fission track analysis(Locke and Garver 2005 Perez de Armas 2005)has been addressed by a second peak before a secondcooling phase For uncertainty analysis basal paleo-heat flow has been varied within geological meaning-ful ranges and the thermal calibration data has beenhonored as much as possible

Modeling Results

Basin and petroleum system modeling can be dividedinto simulation of the physical parameters within abasin such as pressure and temperature (basin model-ing) and the subsequent evolution of the source rocksand their derived hydrocarbon compounds (petro-leum systems modeling) In the following sectionswe use this subdivision to describe the BPSM resultscompaction modeling thermal modeling and chargehistory It is important to state that these are predic-tions based on forward modeling as just describedThe results are compared and calibrated to actualmeasurements and observations where available

Compaction Modeling

Figure 5 presents an extraction of the modeled resultsat the location of Well Alpha in the Furrial Trend Inthe absence of direct calibration data for pore pres-sure we undertook a sensitivity analysis to test thedifferent scenarios Using the model we tested howpore pressure in the main reservoir of the FurrialTrend reacts to changes in fault and seal-rock per-meability In a first step fault permeability has beenevaluated Faults were assumed open during the mainfault activity phase and closed afterward In additionwe performed several variations from that base casescenario Figure 5A shows the effect of two end-member scenarios on pore pressure within the mainreservoir of the Furrial Trend (Santa Barbara field) inwhich faults have been modeled to be either open orclosed since their formation A third scenario consistsof thrust faults that are open during the compressionalphase (from 15 Ma on) and closed afterward (from105 Ma on) This scenario has been chosen as theldquomaster scenariordquo because it results in overpressureof 23 MPa which is in the range of pressure reportedin similar structures of the Furrial Trend by Schneideret al (2004) In addition this pressure scenariomatches the calibration data for porosity (Figure 5B)

The 2D distribution of modeled overpressure atpresent day is shown in Figure 6A Apart fromimportant overpressure in the Querecual Formationthe allochthon is almost normally pressured Thethrust wedge is overpressured especially in thelow-permeability Carapita Formation Although

Table 2 Basin and Petroleum System Modeling (BPSM) InputParameters Thermal Boundary Conditions

Age (Ma)

SedimentWaterInterface

Temperature (degC)

Age (Ma)

Basal HeatFlow

(mW∕m2)min max

0 28 28 0 4226 28 28 30 6053 76 23 34 708 184 231 50 50105 62 233 107 6012 31 238 1455 8015 5 242 260 6019 5 24723 234 2527 78 235339 94 236655 146 2831455 148 287255 5 239

1336 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

the autochthon is highly overpressured close tonormal hydrostatic pressures are simulated in theforeland basin The discussed model assumption ofclosed faults in the postcompressional phase resultsin a regional pressure compartmentalization

The simulated pressure evolution through geo-logic time can be seen in Figure 7A A time extractionof a source rock cell in the thrust wedge shows a sud-den increase of both lithostatic and hydrostatic pres-sures that is a consequence of the tectonic burial

load associated with the thrusting In addition themodeled pore pressure buildup even though delayedin time can be directly linked to the thrust event Itcan be clearly seen that the timing of fault permeabil-ity directly impacts changes in the present-day porepressure The earlier the faults close the more over-pressure generated by the massive syntectonic sedi-mentation is kept until present day given adequatesealing lithologies are present The present-day over-pressure is modeled as14 MPa when the faults are

Figure 5 Plots of one-dimensional model output and available calibration data for Well Alpha (Table 3) (A) Dotted lines are hydro-static and lithostatic pressures continuous lines are pore pressure the master scenario is in black (a1 faults closing at 105 Ma) andadditional scenarios in gray (a2 open faults a3 closed faults) (B) modeled porosity (C) borehole temperature and (D) Ro The blackline shows the master scenario and the gray line the alternative high-heat-flow scenario Bold lines in the stratigraphic column arethrusts (T) and unconformities (U)

Table 3 Calibration Data for Well Alpha (Porosity Temperature Ro)

TVD (m) Porosity () TVD (m) Corrected Temperature (degC) TVD (m) Vitrinite Reflectance (Ro)

4350 15 600 40 400 0264400 15 1100 53 1200 0484800 10 1800 68 1485 068

4400 131 1700 0394800 133 2900 048

3100 0493400 0494000 054300 0524800 06

NEUMAIER ET AL 1337

open since 8 Ma as 23 MPa for 105 Ma and as37 MPa for 12 Ma Compared to this order of magni-tude changes in the permeability of the overlyingCarapita shale are negligible for the pore pressureTherefore the controlling parameter for pore pressurein the main reservoir of the Furrial trend is assumedfault permeability and resulting lateral connectivityrather than the permeability of the seal rock

Thermal Modeling

Modeled temperature and thermal maturity(Figure 5C D) have been calibrated with availablewell data Some discontinuities in the modeled Ro

trend are noted (Figure 5D) The upper kinks (egbase Pliocene) are related to unconformities withassociated erosion and the lower one (base UpperCretaceous) is due to the well penetrating a reverse

Figure 6 Two-dimensional model at present day with (A) modeled overpressure and (B) modeled thermal maturity and isothermsSource rocks marked by lithology pattern

1338 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

fault displacing mature rocks upon less mature ones(see Figure 6B for 2D view) There is a mismatchbetween the modeled Ro trend and the data in thedeeper part of the well In the Oligocene rocks themodeled temperature is too low (2degC or 36degF)whereas the modeled Ro is too high (up to 02 Ro)Any attempt to reduce the modeled Ro (eg bychanging thermal conductivity of the MioceneCarapita Formation) also results in a lower modeledpresent-day temperature Therefore the thermal sce-nario used for this calibration is a trade-off betweenmatching present-day temperature and Ro (using thekinetics of Sweeney and Burnham 1990) The basalheat flow assumed in the master scenario (Table 2)results in a modeled present-day surface heat flowranging from 37 mW∕m2 in the southern part of thethrusts to 52 mW∕m2 in parts of the uplifted areaAt the same depth the allochthonous northern partis thermally more mature than the southern autochthon(Figure 6B) because of the recent uplift Locallyinverted trends around the thrusts can be observedfor example at the location of Well Alpha (seeFigure 5D) In the autochthon the Upper Cretaceoussource rocks are entirely in the oil or even gas windowat present day because of the Neogene tectonic loadingand synorogenic deposition In the northern alloch-thon most of the Upper Cretaceous source rocks havebeen in the oil window in the past The modeled ther-mal maturity is in good agreement with the regionalmaturity suggested by Summa et al (2003)

Similar to the pressures the thrusting also affectsthe modeled temperature and Ro In the thrust wedgea direct relationship exists between the thrusting andthe increase of temperature and maturity The thrust-ing is the trigger for source rocks entering the oil win-dow within the thrust wedge (Figure 7B) and thepresent-day thermal maturity has been reached sincethe last 10 Ma Although this recent thermal historycan be considered as relatively well constrained bythe measured Ro in the calibration well this doesnot apply to the past In this thrusted area any ther-mal maturity reached prior to the thrusting onset hasbeen ldquooverprintedrdquo and therefore no thermal calibra-tion is possible for this period However theCretaceous source rocks were not buried deeply dur-ing the passive margin phase and even very hot sce-narios (basal heat flow of 100 mW∕m2) do not

significantly change their maturity (Figure 7B) Inthe uplifted allochthon several wells show low(03) to moderate (1) measured Ro (Parra et al2011) Because the allochthon experienced its maxi-mum burial and the highest temperatures during thepassive margin phase these maturity levels can beassumed as representative of the maturity prior tothe compression In contrast to the thrust wedge theconsequence of the thrusting is inverse for the alloch-thon (Figure 7C) as the hanging wall is uplifted andthe accompanying erosion removes the overburdenThe highest temperatures are simulated to have beenreached just before the thrusting event Then temper-ature decreased and Ro stagnated leaving the sourcerocks in the early mature zone until present day Theassumption of a regionally relatively homogeneousthermal field during the cooling phase of the passivemargin in combination with the relatively low matur-ities measured in the uplifted allochthon makes thehot heat-flow scenario unlikely However a laterallyvarying heat flow cannot be excluded and the conse-quently hotter paleoheat flow would have led to muchhigher temperatures prior to compression resultingalso in a higher present-day maturity level (14)despite the uplift (Figure 7C) Any variations of basalheat flow during the rift period (Late Jurassic to EarlyCretaceous) occurred too early to affect the maturityof the source rocks deposited during the lateCretaceous

Charge History

After calibration of the basinrsquos physical environ-ment the evolution of the petroleum systems wasmodeled Hydrocarbon generation was simulatedthrough geologic time using the activation energydistribution of the assigned source rock kineticsFrom the moment of hydrocarbon expulsiononward secondary migration was modeled usingthe invasion percolation method (Wilkinson 1984)It takes into account the geometry overpressure gra-dient and capillary entry pressures of the rockswhich are opposed to the buoyancy of the migratingoil and gas Figure 8 shows the time evolution of thetotal study area and Figure 9 focuses on the thrustwedge As already mentioned the described chargehistory is based on forward modeling It is consistent

NEUMAIER ET AL 1339

Figure 7 Time extractions (see Figure 6 for location) (A) Pressure development through geologic time for a source rock in the ver-tical continuity of Well Alpha Dotted lines are hydrostatic and lithostatic pressures continuous lines are pore pressure the master sce-nario is in black (a1 faults closing at 105 Ma) and additional scenarios in gray (a2 open faults a3 closed faults a4 faults closing at12 Ma a5 faults closing at 8 Ma) (B) Temperature (continuous lines) and Ro (dotted lines) through geologic time for a source rockin the vertical continuity of Well Alpha the black line shows the master scenario and the gray line the alternative high-heat-flow sce-nario (C) Temperature (continuous lines) and Ro (dotted lines) through geologic time for a source rock in the uplifted allochthonthe black line shows the master scenario and the gray line the alternative high heat flow scenario

1340 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

with the modeled compaction and thermal historyand therefore shows one scenario that explains thecharge of the Santa Barbara field and the forelandbasin Simplifications are inherent to any such work-flow and shall be considered when analyzing themodeling results

Passive Margin PhasemdashMesozoic toOligocene

Modeling results indicate that already during the LateCretaceous any Jurassic source rocks potentiallydeposited in the northern part of the section (Summaet al 2003) have reached temperatures of more than200degC (392degF) and corresponding high levels of matu-rity However the temperature at the base of theQuerecual Formation (Upper Cretaceous) was lessthan 100degC (212degF) leaving the kerogen immatureThe source rock became effective only during lateEocene with temperatures of up to 120degC (248degF)and a (reactive kerogen to hydrocarbon) transforma-tion ratio (TR) of more than 20 in the deeper distalpart of the passive margin Oil was quickly expelledmigrated vertically and seeped to the surface becauseof the general lack of seals in the overburden Duringthe Oligocene the time of deposition of the Merecurereservoir sandstones in the southern platform this sit-uation did not change apart from increased transforma-tion in the deeper basinal part of the passive margin

First oil accumulations are modeled in the Vidontildeosandstones in the basinal part in the late Oligocene(Figure 8A) in which the base of the QuerecualFormation shows TRs of up to 60 at around 130degC(266degF) In the proximal part there is no hydrocarbongeneration Although there is a general northwardtilting slope no southward migration from the distalmargin is modeled this has been checked usingdifferent migration methods including Darcy flowwith scenarios of continuous intrasource carrier bedsThis is in contrast to the studies from Gallango andParnaud (1995) and Schneider (2003) who suggestedlateral migration at that stage The difference couldbe explained by the presence of a regional top seal intheir models However our model suggests that theshallow rocks do not have sufficient sealing capacityto allow lateral hydrocarbon migration During thepassive-margin phase the basin was normally

pressured (Figure 7A) and simulated stresses havenot yet resulted in any fracturing at that stage

Main Compressional PhasemdashEarly to MiddleMiocene

With the onset of compression during early Miocene(Figure 8B) both the Pirital and Furrial thrusts wereformed and the northern distal part of the passivemargin was rapidly uplifted Important overpressuresas well as mechanical rock failure both the result ofthe rapid uplift are simulated in the allochthonDespite this uplift the kerogen-to-hydrocarbon trans-formation continued and the rate even increasedlocally close to the footwall cutoffs and where syn-orogenic deposition provided rapid overburden load-ing Oil accumulations are modeled in the freshlycreated anticlines

The tectonicndashsedimentary wedge between thetwo main thrusts experienced very rapid burial Infact the Upper Cretaceous source rocks were broughtinto depths where temperatures were as high as185degC (365degF) at 19 Ma and 225degC (437degF) at15 Ma Especially close to the cutoff of the Piritalthrust they very quickly reached a high maturity at19 Ma and reached full transformation in most placesat 12 Ma The expelled hydrocarbons migrated intothe sandstones of the Merecure Formation which weresealed by the Carapita Formation The charge toward aseries of recent anticlines is modeled by the fill-to-spillmechanism from north to south (Figure 9C D) Afterthe first oil fill the northern prospect quickly receiveda local gas charge the central structure containedmainly oil Note the presence of a lateral seal that isdue to sandndashshale juxtaposition The southernmoststructure which forms the Santa Barbara field todaywas charged by oil only at 15 Ma (Figure 9D)

The generated hydrocarbons in the southernautochthon were leaking to the surface until theCarapita Formation was compacted enough to holdsome of them (Figure 8C) At 12 Ma the wedgesource rocks were mostly spent but expulsion wasstill ongoing Important rock failure is modeled tohave occurred in the thrusted footwall breaking theformerly sealing Carapita Formation (Figure 10)This is in good agreement with cemented hydraulicfractures observed in a few cores in the Carapita

NEUMAIER ET AL 1341

Figure 8 Two-dimensional model through geologic time (selected time steps only) with modeled transformation ratio for the sourcerock layers isotherms and liquid and vapor migration and accumulation

1342 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

shales from the Furrial Trend (Roure et al 2005)This important fracturing event resulted in thedestruction of the modeled accumulation (Figure 9E)Hydrocarbon generation and expulsion were alsohappening in the southern foreland basin at transfor-mation ratios of up to 30

Late Compressional and PostcompressionalPhasemdashLate Miocene to Present Day

The late compressional phase was marked by the par-tial destruction of the allochthonous petroleum systemsby erosion (Figure 8E F) Many traps disappeared and

Figure 8 Continued

NEUMAIER ET AL 1343

the hydrocarbon generation within the source rocksstopped The erosion triggered an important sedimenttransfer toward the south contributing to further burialand increased temperatures in the thrust wedge and theforeland basin During late Miocene the petroleumsystem in the wedge matured An important northwardtilting was triggered by flexural tectonic loading andthe loading of syntectonic sediments (Summa et al2003) In combination with the presence of an effec-tive regional top seal the compacted CarapitaFormation allowed a change from dominantly verticalmigration into lateral migration First accumulationsare modeled in the Neogene synorogenic sequence

reservoirs in the south and important southwardmigration toward the foreland basin started(Figure 8G) These trends continued through Plioceneand Pleistocene times to present day filling theOnado field (Figure 8H) and other known fields inthe south (Figure 1A) of the Eastern Venezuela Basin

Prospectivity

We modeled the charge of the Santa Barbara field inthe Furrial Trend and of the Onado field in the fore-land basin (Figure 8H) Other prospects in the thrustwedge were also investigated The structure called

Figure 9 Detail of two-dimensionalmodel through geologic time (seeFigure 8) showing modeled transforma-tion ratio for the source rock layers iso-therms and liquid and vapor migrationand accumulation The Santa Barbara fieldand the prospect are shown by ellipses

1344 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

ldquoprospectrdquo (Figure 9G) looks similar to that of theSanta Barbara field because both are anticlines andthe reservoir and seal formations are the sameHowever they differ in several aspects First theprospect is in the vicinity of the Pirital thrust whichpresents a high risk of impacting the seal integrityThe prospect location is directly below the area wherewe presume the maximum of late tectonic activityoccurred (Figure 4) As modeled the CarapitaFormation the seal of the prospect was already brit-tle when this deformation occurred Calculatedstresses also predict rock failure from 12 Ma onward(Figure 10) Even though charge continued ourmodel suggests continuous leakage of the structure(Figure 9EndashG) The second major difference is thatthe source rock in the fault block of the prospect ismodeled to have already started generating and expel-ling hydrocarbons during the Oligocene and earlyMiocene (Figure 9A) before the trap formed and the

seal became effective (Figures 9C 10) Even thoughgeneration expulsion and migration continued aftertrap formation the prospect was not charged withthe earlier expelled oil In addition the present-daymaturity of the source rocks within the prospectrsquosfault block is higher than in the source rocks withinthe tectonic block of the Santa Barbara field poten-tially indicating expulsion of more gas A southndashnorth trend can be seen on the modeled accumulationcomposition with more gas toward the allochthon(Figure 9) The modeled composition of the prospectprior to seal failure was much richer in gas than inthe neighboring Santa Barbara field (Figure 9D) Infact simulation scenarios with unbreached sealsresulted in accumulation of predominantly gas Thecombination of both seal integrity and charge compo-sition concerns makes this a risky prospect In addi-tion potential diagenetic fluids related to the faultfluid flow might have altered the reservoir quality

Figure 10 Two-dimensional model (at 12 Ma) with modeled stress to failure Mohr circles (with σ standing for normal stress and τfor shear stress) and petroleum system events charts for the Santa Barbara field (left) and the prospect (right) Model predicts rock fail-ure in the prospect from about 12 Ma For further details see text

NEUMAIER ET AL 1345

Apart from the Onado oil field only minor accu-mulations are modeled in the southern Neogene syn-orogenic sequence The proportion of gas is higherthan in the Furrial trend The accumulations at thesouthern model edge are a modeling artifact repre-senting hydrocarbons (mainly oil) that in reality arethought to have charged the southern foreland basin(from mid to late Miocene times to present day)However parts of the migrated oil and gas could havebeen trapped in the Upper Cretaceous San JuanFormation south of the horst structure

The potential for conventional hydrocarbon accu-mulations in the allochthon is very limited by theimportant fracturing that occurred during the upliftperiod The basal sandstones of the Morochito piggy-back basin north of the Pirital thrust are not chargedin our 2D model because of the absence of structuraltraps Nevertheless there is a potential for strati-graphic traps to have been charged as migrationthrough the basin occurred However seal strengthmight be critical at this low-compaction stage Thesame applies for the Las Piedras Formation We thinkthere might be potential for shale oil because theQuerecual Formation is at shallow depths but showsmaturity within the oil window From the thermalmaturity point of view any potential Jurassic sourcerocks as discussed by Summa et al (2003) mightbe a gas source or shale gas target

UNCERTAINTY ANALYSIS

There is often significant uncertainty in basin geom-etry and depth of stratigraphic horizons based onuncertain seismic interpretation and velocity models(Kukla et al 2000 Brandes et al 2008 Baur et al2010) Restoring of present-day geometries to obtainpaleogeometries adds further uncertainty Additionaluncertainty is related to thermal boundary conditionssuch as paleoheat flow (Beha et al 2008) and togeochemical properties such as source rock character-istics and petroleum generation kinetics and calcula-tion of biodegradation and petroleum density(Blumenstein et al 2008) In addition to the porepressure and thermal sensitivity analysis justdescribed we varied initial TOC within a range of2 to 10 A variation of initial TOC obviously

directly scales the amount of initial kerogen availablefor transformation into hydrocarbons Neither thecharge volume nor composition of the Santa Barbarafield nor the wedge prospects are significantly sensi-tive to variation of initial TOC these traps are filledto spill in each scenario However the amount ofhydrocarbons charging the foreland basin (eg theOnado field) is reduced in the case of low initialTOC A high initial TOC only increased the modeledamount of surface seepage The sensitivity of the ini-tial HI is analogous to the sensitivity of the ini-tial TOC

In addition to the more classical uncertaintyand sensitivity analysis described we focused onquantitative analysis of the decompaction uncer-tainty when performing structural restoration Thisuncertainty is particular to the applied workflowof using structurally restored paleogeometries inBPSM Note that this uncertainty is significantlylower when using the backstripping method withthe optimization algorithms discussed One methodto analyze the adequacy of the decompactionmethod is comparison of the output rock matrixmass balance In the backstripping approach therock matrix mass is kept constant of a model cellwith a simulated decrease in porosity controlledby compaction laws and the corresponding reduc-tion of the layer thickness When using fixed pale-ogeometries from structural restoration the cellthickness is imposed However the decrease inporosity is calculated and the mass is adapted tofit the imposed cell thickness That means that therock matrix mass is constant only if ideal decom-paction has been applied

Figure 11A shows the percentage variation fromthe present-day matrix mass (100) of a given sourcerock cell through geologic time The dotted line rep-resents the ideal case with constant rock matrix massWe used two end-member decompaction scenarios todescribe the decompaction-related uncertainty that isintroduced when fixed paleogeometries replace back-stripping idealized hydrostatic decompaction assum-ing no overpressure and idealized lithostaticdecompaction assuming no effective stress (nodecompaction resulting in constant cell thicknessthrough time) In the latter case with decreasingporosity during compaction the rock matrix mass

1346 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

increases until it reaches the present-day mass Inother words rock mass matrix has been highly under-estimated in the past (in this particular case by morethan 20) This is a source rock so the kerogen massdecreases by the same factor This yields an underes-timation of more than 20 in TOC Note the rockmatrix mass is constant for time steps prior to23 Ma in which backstripping optimization couldbe performed toward the earliest paleogeometry Forthe second end-member scenario of idealized hydro-static decompaction the rock matrix mass is nearlyconstant which is a good decompaction adequacy

indicator for that given cell in the model Howeverthe so-measured adequacy changes throughout themodel In some areas the rock matrix mass has beenunderestimated by about 50 for nondecompactedpaleogeometries however in the case of hydrostaticdecompaction values generally range from 80 to120 of the present-day rock matrix mass Thatmeans in this particular case that hydrostatic decom-paction is more adequate (or less wrong) than nodecompaction at all The decompaction adequacycan be roughly correlated with pressure changesthrough geologic time

Figure 11 Time extractions for comparison of two end-member decompaction scenarios (see Figure 6 for location) (A) Rock matrixmass conservation through geologic time (B) Source rock burial depth through geologic time (C) Temperature and transformation ratiothrough geologic time

NEUMAIER ET AL 1347

Inadequate decompaction results in wrong depthfor stratigraphic horizons in the paleogeometriesFigure 11B shows the burial depth of the same inves-tigated source rock cell We can see that during thepassive margin phase there is a slight difference inpaleoburial depth for the two end-member scenariosDuring the compressional phase the differencechanges dramatically reaching more than 1000 m(3281 ft) in some places Toward present day bothcurves meet again because the input to the two resto-ration scenarios was the same (identical present-dayburial depth) This difference in compaction calcula-tion has severe implications on the modeled tempera-ture history and the kerogen-to-hydrocarbontransformation (Figure 11C) For an identical thermalframework in terms of modeling input parameterstemperature differences of up to 25degC (45degF) resultin TR differences of more than 20 in the early com-pressional phase before TR values meet again for thepresent-day situation

CONCLUSIONS

Structurally complex areas are attracting increasinginterest in hydrocarbon exploration Exploration riskcan be substantially narrowed by analyzing the his-tory of the basin and its petroleum systems based onstructural evolution We presented a further progressin technology that allows the integration of structuralrestoration and BPSM This combined analysis isseen as a way forward to improve modeling in thrustbelts with stacked stratigraphic sequences Despitelimitations of nonoptimized decompaction whichare added to the general uncertainty related to theindividual methods we modeled the charge and sealhistory of the Monagas fold and thrust belt inVenezuela Calibrated porosity temperature andthermal maturity as well as basin-scale stress pro-vided the regional framework through geologic timein which the petroleum systems evolved such thatthe charge of the known Santa Barbara and Onadofields and other trends could be understood We dis-cussed seal integrity of a potential exploration pros-pect by analyzing structural restoration-derived faultdisplacement through geologic time and an advancedstress forward simulation The uncertainty of using

decompacted or nondecompacted structural restora-tion has been assessed It can have important influ-ence on rock matrix mass balance through geologictime paleogeometry and paleoburial depth andtherefore affect paleotemperatures and paleomatur-ities We conclude that decompaction assuminghydrostatic pressure is the best option in that particu-lar case

REFERENCES CITED

Alberdi M and E Lafargue 1993 Vertical variations of organicmatter content in Guayuta Group (Upper Cretaceous)interior mountain belt eastern Venezuela OrganicGeochemistry v 20 p 425ndash436 doi 1010160146-6380(93)90091-O

Athy L F 1930 Density porosity and compaction of sedimen-tary rocks AAPG Bulletin v 14 p 1ndash24

Aymard R L Pimentel P Eitz P Lopez A ChaouchJ Navarro J Mijares and J G Pereira 1990 Geologic inte-gration and evaluation of Northern Monagas EasternVenezuelan Basin in J Brooks ed Classic petroleum prov-inces The Geologic Society (London) Special Publicationv 50 p 37ndash53

Baur F M Di Benedetto T Fuchs C Lampe andS Sciamanna 2009 Integrating structural geology andpetroleum systems modelingmdashA pilot project from Boliviarsquosfold and thrust belt Marine and Petroleum Geology v 26no 4 p 573ndash579 doi 101016jmarpetgeo200901004

Baur F R Littke H Wielens C Lampe and T Fuchs 2010Basin modeling meets rift analysismdashA numerical modelingstudy from the Jeanne drsquoArc basin offshore NewfoundlandCanada Marine and Petroleum Geology v 27 p 585ndash599doi 101016jmarpetgeo200906003

Beha A R O Thomsen and R Littke 2008 Thermal historyhydrocarbon generation and migration in the Horn Grabenin the Danish North Sea A 2D basin modelling studyInternational Journal of Earth Sciences v 97 p 1087ndash1100doi 101007s00531-007-0247-2

Blumenstein I O B M Krooss R di Primio W RottkeE Muumlller C Westerlage and R Littke 2008Biodegradation in numerical basin modelling A case studyfrom the Gifhorn Trough N-Germany International Journalof Earth Sciences v 97 p 1115ndash1129 doi 101007s00531-007-0272-1

Brandes C A Astorga R Littke and J Winsemann 2008Basin modelling of the Limoacuten back-arc basin (Costa Rica)Burial history and temperature evolution of an island arc-related basin-system Basin Research v 20 p 119ndash142doi 101111bre200820issue-1

Carnevali J O 1988 El Furrial oil field northeasternVenezuela First giant in foreland fold and thrust belts ofwestern hemisphere AAPG Bulletin v 72 68 p

Chevalier Y 1987 Contribution a lrsquoetude geacuteologique de lafrontiereacute Sud-Est de la plaque Caraibe Les zones internesde la chaine Sud-Caraibe sur le transect ile de Margarita-

1348 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

339

PaleoEocene

Vidono

sandsto

ne2720

412747

115

minus18

3395

338

02

024

Caratasbasin

alshale

2700

7040327

403

minus852

minus3

164

169

021

024

655

upperC

retaceous4

SanJuan

sandsto

ne2720

412747

115

minus18

3395

338

02

024

74upperC

retaceous3

SanAntoniosiltstone

(SOUR

CERO

CK)

2710

5510422

212

minus628

minus1

201

196

023

026

825

upperC

retaceous2

SanAntoniosiltstone

(SOUR

CERO

CK)

2710

5510422

212

minus628

minus1

201

196

023

026

91upperC

retaceous1

Querecualmudsto

ne(SOUR

CERO

CK)

2700

7040327

403

minus852

minus3

164

169

021

024

996

lower

Cretaceous

2undifflower

Cretaceous

(30

Lst50

Sst

20Sh)

2722

459445

141

minus334

174

305

275

02

023

122

lower

Cretaceous

1undifflower

Cretaceous

(30

Lst50

Sst

20Sh)

2722

459445

141

minus334

174

305

275

02

023

1455

sub-Cretaceous

2undiffsub-Cretaceous

(50

Sh50Sst)

2710

5521537

259

minus516

0255

239

02

024

251

NEUMAIER ET AL 1335

by the influence of the paleowater depth using theapproach of Wygrala (1989) Because of the hightopographic elevation of approximately 3500 m(11500 ft) the maximum altitude during the pastcompressional phase we adjusted the model takinginto account a temperature decrease of 6degCkm (33degF1000 ft) related to the paleoaltitude This resultsin a difference of up to 20degC (36degF) compared to sea-level temperature We then assigned the adjustedsedimentwater interface temperatures to the modelas the upper thermal boundary condition At the baseof the model we assigned a heat flow ranging from42 to 80 mW∕m2 through geologic time The highestheat flow is assumed during the rifting event beforethe basin cooled as a passive margin A thermal eventin the Eocene detected by fission track analysis(Locke and Garver 2005 Perez de Armas 2005)has been addressed by a second peak before a secondcooling phase For uncertainty analysis basal paleo-heat flow has been varied within geological meaning-ful ranges and the thermal calibration data has beenhonored as much as possible

Modeling Results

Basin and petroleum system modeling can be dividedinto simulation of the physical parameters within abasin such as pressure and temperature (basin model-ing) and the subsequent evolution of the source rocksand their derived hydrocarbon compounds (petro-leum systems modeling) In the following sectionswe use this subdivision to describe the BPSM resultscompaction modeling thermal modeling and chargehistory It is important to state that these are predic-tions based on forward modeling as just describedThe results are compared and calibrated to actualmeasurements and observations where available

Compaction Modeling

Figure 5 presents an extraction of the modeled resultsat the location of Well Alpha in the Furrial Trend Inthe absence of direct calibration data for pore pres-sure we undertook a sensitivity analysis to test thedifferent scenarios Using the model we tested howpore pressure in the main reservoir of the FurrialTrend reacts to changes in fault and seal-rock per-meability In a first step fault permeability has beenevaluated Faults were assumed open during the mainfault activity phase and closed afterward In additionwe performed several variations from that base casescenario Figure 5A shows the effect of two end-member scenarios on pore pressure within the mainreservoir of the Furrial Trend (Santa Barbara field) inwhich faults have been modeled to be either open orclosed since their formation A third scenario consistsof thrust faults that are open during the compressionalphase (from 15 Ma on) and closed afterward (from105 Ma on) This scenario has been chosen as theldquomaster scenariordquo because it results in overpressureof 23 MPa which is in the range of pressure reportedin similar structures of the Furrial Trend by Schneideret al (2004) In addition this pressure scenariomatches the calibration data for porosity (Figure 5B)

The 2D distribution of modeled overpressure atpresent day is shown in Figure 6A Apart fromimportant overpressure in the Querecual Formationthe allochthon is almost normally pressured Thethrust wedge is overpressured especially in thelow-permeability Carapita Formation Although

Table 2 Basin and Petroleum System Modeling (BPSM) InputParameters Thermal Boundary Conditions

Age (Ma)

SedimentWaterInterface

Temperature (degC)

Age (Ma)

Basal HeatFlow

(mW∕m2)min max

0 28 28 0 4226 28 28 30 6053 76 23 34 708 184 231 50 50105 62 233 107 6012 31 238 1455 8015 5 242 260 6019 5 24723 234 2527 78 235339 94 236655 146 2831455 148 287255 5 239

1336 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

the autochthon is highly overpressured close tonormal hydrostatic pressures are simulated in theforeland basin The discussed model assumption ofclosed faults in the postcompressional phase resultsin a regional pressure compartmentalization

The simulated pressure evolution through geo-logic time can be seen in Figure 7A A time extractionof a source rock cell in the thrust wedge shows a sud-den increase of both lithostatic and hydrostatic pres-sures that is a consequence of the tectonic burial

load associated with the thrusting In addition themodeled pore pressure buildup even though delayedin time can be directly linked to the thrust event Itcan be clearly seen that the timing of fault permeabil-ity directly impacts changes in the present-day porepressure The earlier the faults close the more over-pressure generated by the massive syntectonic sedi-mentation is kept until present day given adequatesealing lithologies are present The present-day over-pressure is modeled as14 MPa when the faults are

Figure 5 Plots of one-dimensional model output and available calibration data for Well Alpha (Table 3) (A) Dotted lines are hydro-static and lithostatic pressures continuous lines are pore pressure the master scenario is in black (a1 faults closing at 105 Ma) andadditional scenarios in gray (a2 open faults a3 closed faults) (B) modeled porosity (C) borehole temperature and (D) Ro The blackline shows the master scenario and the gray line the alternative high-heat-flow scenario Bold lines in the stratigraphic column arethrusts (T) and unconformities (U)

Table 3 Calibration Data for Well Alpha (Porosity Temperature Ro)

TVD (m) Porosity () TVD (m) Corrected Temperature (degC) TVD (m) Vitrinite Reflectance (Ro)

4350 15 600 40 400 0264400 15 1100 53 1200 0484800 10 1800 68 1485 068

4400 131 1700 0394800 133 2900 048

3100 0493400 0494000 054300 0524800 06

NEUMAIER ET AL 1337

open since 8 Ma as 23 MPa for 105 Ma and as37 MPa for 12 Ma Compared to this order of magni-tude changes in the permeability of the overlyingCarapita shale are negligible for the pore pressureTherefore the controlling parameter for pore pressurein the main reservoir of the Furrial trend is assumedfault permeability and resulting lateral connectivityrather than the permeability of the seal rock

Thermal Modeling

Modeled temperature and thermal maturity(Figure 5C D) have been calibrated with availablewell data Some discontinuities in the modeled Ro

trend are noted (Figure 5D) The upper kinks (egbase Pliocene) are related to unconformities withassociated erosion and the lower one (base UpperCretaceous) is due to the well penetrating a reverse

Figure 6 Two-dimensional model at present day with (A) modeled overpressure and (B) modeled thermal maturity and isothermsSource rocks marked by lithology pattern

1338 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

fault displacing mature rocks upon less mature ones(see Figure 6B for 2D view) There is a mismatchbetween the modeled Ro trend and the data in thedeeper part of the well In the Oligocene rocks themodeled temperature is too low (2degC or 36degF)whereas the modeled Ro is too high (up to 02 Ro)Any attempt to reduce the modeled Ro (eg bychanging thermal conductivity of the MioceneCarapita Formation) also results in a lower modeledpresent-day temperature Therefore the thermal sce-nario used for this calibration is a trade-off betweenmatching present-day temperature and Ro (using thekinetics of Sweeney and Burnham 1990) The basalheat flow assumed in the master scenario (Table 2)results in a modeled present-day surface heat flowranging from 37 mW∕m2 in the southern part of thethrusts to 52 mW∕m2 in parts of the uplifted areaAt the same depth the allochthonous northern partis thermally more mature than the southern autochthon(Figure 6B) because of the recent uplift Locallyinverted trends around the thrusts can be observedfor example at the location of Well Alpha (seeFigure 5D) In the autochthon the Upper Cretaceoussource rocks are entirely in the oil or even gas windowat present day because of the Neogene tectonic loadingand synorogenic deposition In the northern alloch-thon most of the Upper Cretaceous source rocks havebeen in the oil window in the past The modeled ther-mal maturity is in good agreement with the regionalmaturity suggested by Summa et al (2003)

Similar to the pressures the thrusting also affectsthe modeled temperature and Ro In the thrust wedgea direct relationship exists between the thrusting andthe increase of temperature and maturity The thrust-ing is the trigger for source rocks entering the oil win-dow within the thrust wedge (Figure 7B) and thepresent-day thermal maturity has been reached sincethe last 10 Ma Although this recent thermal historycan be considered as relatively well constrained bythe measured Ro in the calibration well this doesnot apply to the past In this thrusted area any ther-mal maturity reached prior to the thrusting onset hasbeen ldquooverprintedrdquo and therefore no thermal calibra-tion is possible for this period However theCretaceous source rocks were not buried deeply dur-ing the passive margin phase and even very hot sce-narios (basal heat flow of 100 mW∕m2) do not

significantly change their maturity (Figure 7B) Inthe uplifted allochthon several wells show low(03) to moderate (1) measured Ro (Parra et al2011) Because the allochthon experienced its maxi-mum burial and the highest temperatures during thepassive margin phase these maturity levels can beassumed as representative of the maturity prior tothe compression In contrast to the thrust wedge theconsequence of the thrusting is inverse for the alloch-thon (Figure 7C) as the hanging wall is uplifted andthe accompanying erosion removes the overburdenThe highest temperatures are simulated to have beenreached just before the thrusting event Then temper-ature decreased and Ro stagnated leaving the sourcerocks in the early mature zone until present day Theassumption of a regionally relatively homogeneousthermal field during the cooling phase of the passivemargin in combination with the relatively low matur-ities measured in the uplifted allochthon makes thehot heat-flow scenario unlikely However a laterallyvarying heat flow cannot be excluded and the conse-quently hotter paleoheat flow would have led to muchhigher temperatures prior to compression resultingalso in a higher present-day maturity level (14)despite the uplift (Figure 7C) Any variations of basalheat flow during the rift period (Late Jurassic to EarlyCretaceous) occurred too early to affect the maturityof the source rocks deposited during the lateCretaceous

Charge History

After calibration of the basinrsquos physical environ-ment the evolution of the petroleum systems wasmodeled Hydrocarbon generation was simulatedthrough geologic time using the activation energydistribution of the assigned source rock kineticsFrom the moment of hydrocarbon expulsiononward secondary migration was modeled usingthe invasion percolation method (Wilkinson 1984)It takes into account the geometry overpressure gra-dient and capillary entry pressures of the rockswhich are opposed to the buoyancy of the migratingoil and gas Figure 8 shows the time evolution of thetotal study area and Figure 9 focuses on the thrustwedge As already mentioned the described chargehistory is based on forward modeling It is consistent

NEUMAIER ET AL 1339

Figure 7 Time extractions (see Figure 6 for location) (A) Pressure development through geologic time for a source rock in the ver-tical continuity of Well Alpha Dotted lines are hydrostatic and lithostatic pressures continuous lines are pore pressure the master sce-nario is in black (a1 faults closing at 105 Ma) and additional scenarios in gray (a2 open faults a3 closed faults a4 faults closing at12 Ma a5 faults closing at 8 Ma) (B) Temperature (continuous lines) and Ro (dotted lines) through geologic time for a source rockin the vertical continuity of Well Alpha the black line shows the master scenario and the gray line the alternative high-heat-flow sce-nario (C) Temperature (continuous lines) and Ro (dotted lines) through geologic time for a source rock in the uplifted allochthonthe black line shows the master scenario and the gray line the alternative high heat flow scenario

1340 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

with the modeled compaction and thermal historyand therefore shows one scenario that explains thecharge of the Santa Barbara field and the forelandbasin Simplifications are inherent to any such work-flow and shall be considered when analyzing themodeling results

Passive Margin PhasemdashMesozoic toOligocene

Modeling results indicate that already during the LateCretaceous any Jurassic source rocks potentiallydeposited in the northern part of the section (Summaet al 2003) have reached temperatures of more than200degC (392degF) and corresponding high levels of matu-rity However the temperature at the base of theQuerecual Formation (Upper Cretaceous) was lessthan 100degC (212degF) leaving the kerogen immatureThe source rock became effective only during lateEocene with temperatures of up to 120degC (248degF)and a (reactive kerogen to hydrocarbon) transforma-tion ratio (TR) of more than 20 in the deeper distalpart of the passive margin Oil was quickly expelledmigrated vertically and seeped to the surface becauseof the general lack of seals in the overburden Duringthe Oligocene the time of deposition of the Merecurereservoir sandstones in the southern platform this sit-uation did not change apart from increased transforma-tion in the deeper basinal part of the passive margin

First oil accumulations are modeled in the Vidontildeosandstones in the basinal part in the late Oligocene(Figure 8A) in which the base of the QuerecualFormation shows TRs of up to 60 at around 130degC(266degF) In the proximal part there is no hydrocarbongeneration Although there is a general northwardtilting slope no southward migration from the distalmargin is modeled this has been checked usingdifferent migration methods including Darcy flowwith scenarios of continuous intrasource carrier bedsThis is in contrast to the studies from Gallango andParnaud (1995) and Schneider (2003) who suggestedlateral migration at that stage The difference couldbe explained by the presence of a regional top seal intheir models However our model suggests that theshallow rocks do not have sufficient sealing capacityto allow lateral hydrocarbon migration During thepassive-margin phase the basin was normally

pressured (Figure 7A) and simulated stresses havenot yet resulted in any fracturing at that stage

Main Compressional PhasemdashEarly to MiddleMiocene

With the onset of compression during early Miocene(Figure 8B) both the Pirital and Furrial thrusts wereformed and the northern distal part of the passivemargin was rapidly uplifted Important overpressuresas well as mechanical rock failure both the result ofthe rapid uplift are simulated in the allochthonDespite this uplift the kerogen-to-hydrocarbon trans-formation continued and the rate even increasedlocally close to the footwall cutoffs and where syn-orogenic deposition provided rapid overburden load-ing Oil accumulations are modeled in the freshlycreated anticlines

The tectonicndashsedimentary wedge between thetwo main thrusts experienced very rapid burial Infact the Upper Cretaceous source rocks were broughtinto depths where temperatures were as high as185degC (365degF) at 19 Ma and 225degC (437degF) at15 Ma Especially close to the cutoff of the Piritalthrust they very quickly reached a high maturity at19 Ma and reached full transformation in most placesat 12 Ma The expelled hydrocarbons migrated intothe sandstones of the Merecure Formation which weresealed by the Carapita Formation The charge toward aseries of recent anticlines is modeled by the fill-to-spillmechanism from north to south (Figure 9C D) Afterthe first oil fill the northern prospect quickly receiveda local gas charge the central structure containedmainly oil Note the presence of a lateral seal that isdue to sandndashshale juxtaposition The southernmoststructure which forms the Santa Barbara field todaywas charged by oil only at 15 Ma (Figure 9D)

The generated hydrocarbons in the southernautochthon were leaking to the surface until theCarapita Formation was compacted enough to holdsome of them (Figure 8C) At 12 Ma the wedgesource rocks were mostly spent but expulsion wasstill ongoing Important rock failure is modeled tohave occurred in the thrusted footwall breaking theformerly sealing Carapita Formation (Figure 10)This is in good agreement with cemented hydraulicfractures observed in a few cores in the Carapita

NEUMAIER ET AL 1341

Figure 8 Two-dimensional model through geologic time (selected time steps only) with modeled transformation ratio for the sourcerock layers isotherms and liquid and vapor migration and accumulation

1342 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

shales from the Furrial Trend (Roure et al 2005)This important fracturing event resulted in thedestruction of the modeled accumulation (Figure 9E)Hydrocarbon generation and expulsion were alsohappening in the southern foreland basin at transfor-mation ratios of up to 30

Late Compressional and PostcompressionalPhasemdashLate Miocene to Present Day

The late compressional phase was marked by the par-tial destruction of the allochthonous petroleum systemsby erosion (Figure 8E F) Many traps disappeared and

Figure 8 Continued

NEUMAIER ET AL 1343

the hydrocarbon generation within the source rocksstopped The erosion triggered an important sedimenttransfer toward the south contributing to further burialand increased temperatures in the thrust wedge and theforeland basin During late Miocene the petroleumsystem in the wedge matured An important northwardtilting was triggered by flexural tectonic loading andthe loading of syntectonic sediments (Summa et al2003) In combination with the presence of an effec-tive regional top seal the compacted CarapitaFormation allowed a change from dominantly verticalmigration into lateral migration First accumulationsare modeled in the Neogene synorogenic sequence

reservoirs in the south and important southwardmigration toward the foreland basin started(Figure 8G) These trends continued through Plioceneand Pleistocene times to present day filling theOnado field (Figure 8H) and other known fields inthe south (Figure 1A) of the Eastern Venezuela Basin

Prospectivity

We modeled the charge of the Santa Barbara field inthe Furrial Trend and of the Onado field in the fore-land basin (Figure 8H) Other prospects in the thrustwedge were also investigated The structure called

Figure 9 Detail of two-dimensionalmodel through geologic time (seeFigure 8) showing modeled transforma-tion ratio for the source rock layers iso-therms and liquid and vapor migrationand accumulation The Santa Barbara fieldand the prospect are shown by ellipses

1344 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

ldquoprospectrdquo (Figure 9G) looks similar to that of theSanta Barbara field because both are anticlines andthe reservoir and seal formations are the sameHowever they differ in several aspects First theprospect is in the vicinity of the Pirital thrust whichpresents a high risk of impacting the seal integrityThe prospect location is directly below the area wherewe presume the maximum of late tectonic activityoccurred (Figure 4) As modeled the CarapitaFormation the seal of the prospect was already brit-tle when this deformation occurred Calculatedstresses also predict rock failure from 12 Ma onward(Figure 10) Even though charge continued ourmodel suggests continuous leakage of the structure(Figure 9EndashG) The second major difference is thatthe source rock in the fault block of the prospect ismodeled to have already started generating and expel-ling hydrocarbons during the Oligocene and earlyMiocene (Figure 9A) before the trap formed and the

seal became effective (Figures 9C 10) Even thoughgeneration expulsion and migration continued aftertrap formation the prospect was not charged withthe earlier expelled oil In addition the present-daymaturity of the source rocks within the prospectrsquosfault block is higher than in the source rocks withinthe tectonic block of the Santa Barbara field poten-tially indicating expulsion of more gas A southndashnorth trend can be seen on the modeled accumulationcomposition with more gas toward the allochthon(Figure 9) The modeled composition of the prospectprior to seal failure was much richer in gas than inthe neighboring Santa Barbara field (Figure 9D) Infact simulation scenarios with unbreached sealsresulted in accumulation of predominantly gas Thecombination of both seal integrity and charge compo-sition concerns makes this a risky prospect In addi-tion potential diagenetic fluids related to the faultfluid flow might have altered the reservoir quality

Figure 10 Two-dimensional model (at 12 Ma) with modeled stress to failure Mohr circles (with σ standing for normal stress and τfor shear stress) and petroleum system events charts for the Santa Barbara field (left) and the prospect (right) Model predicts rock fail-ure in the prospect from about 12 Ma For further details see text

NEUMAIER ET AL 1345

Apart from the Onado oil field only minor accu-mulations are modeled in the southern Neogene syn-orogenic sequence The proportion of gas is higherthan in the Furrial trend The accumulations at thesouthern model edge are a modeling artifact repre-senting hydrocarbons (mainly oil) that in reality arethought to have charged the southern foreland basin(from mid to late Miocene times to present day)However parts of the migrated oil and gas could havebeen trapped in the Upper Cretaceous San JuanFormation south of the horst structure

The potential for conventional hydrocarbon accu-mulations in the allochthon is very limited by theimportant fracturing that occurred during the upliftperiod The basal sandstones of the Morochito piggy-back basin north of the Pirital thrust are not chargedin our 2D model because of the absence of structuraltraps Nevertheless there is a potential for strati-graphic traps to have been charged as migrationthrough the basin occurred However seal strengthmight be critical at this low-compaction stage Thesame applies for the Las Piedras Formation We thinkthere might be potential for shale oil because theQuerecual Formation is at shallow depths but showsmaturity within the oil window From the thermalmaturity point of view any potential Jurassic sourcerocks as discussed by Summa et al (2003) mightbe a gas source or shale gas target

UNCERTAINTY ANALYSIS

There is often significant uncertainty in basin geom-etry and depth of stratigraphic horizons based onuncertain seismic interpretation and velocity models(Kukla et al 2000 Brandes et al 2008 Baur et al2010) Restoring of present-day geometries to obtainpaleogeometries adds further uncertainty Additionaluncertainty is related to thermal boundary conditionssuch as paleoheat flow (Beha et al 2008) and togeochemical properties such as source rock character-istics and petroleum generation kinetics and calcula-tion of biodegradation and petroleum density(Blumenstein et al 2008) In addition to the porepressure and thermal sensitivity analysis justdescribed we varied initial TOC within a range of2 to 10 A variation of initial TOC obviously

directly scales the amount of initial kerogen availablefor transformation into hydrocarbons Neither thecharge volume nor composition of the Santa Barbarafield nor the wedge prospects are significantly sensi-tive to variation of initial TOC these traps are filledto spill in each scenario However the amount ofhydrocarbons charging the foreland basin (eg theOnado field) is reduced in the case of low initialTOC A high initial TOC only increased the modeledamount of surface seepage The sensitivity of the ini-tial HI is analogous to the sensitivity of the ini-tial TOC

In addition to the more classical uncertaintyand sensitivity analysis described we focused onquantitative analysis of the decompaction uncer-tainty when performing structural restoration Thisuncertainty is particular to the applied workflowof using structurally restored paleogeometries inBPSM Note that this uncertainty is significantlylower when using the backstripping method withthe optimization algorithms discussed One methodto analyze the adequacy of the decompactionmethod is comparison of the output rock matrixmass balance In the backstripping approach therock matrix mass is kept constant of a model cellwith a simulated decrease in porosity controlledby compaction laws and the corresponding reduc-tion of the layer thickness When using fixed pale-ogeometries from structural restoration the cellthickness is imposed However the decrease inporosity is calculated and the mass is adapted tofit the imposed cell thickness That means that therock matrix mass is constant only if ideal decom-paction has been applied

Figure 11A shows the percentage variation fromthe present-day matrix mass (100) of a given sourcerock cell through geologic time The dotted line rep-resents the ideal case with constant rock matrix massWe used two end-member decompaction scenarios todescribe the decompaction-related uncertainty that isintroduced when fixed paleogeometries replace back-stripping idealized hydrostatic decompaction assum-ing no overpressure and idealized lithostaticdecompaction assuming no effective stress (nodecompaction resulting in constant cell thicknessthrough time) In the latter case with decreasingporosity during compaction the rock matrix mass

1346 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

increases until it reaches the present-day mass Inother words rock mass matrix has been highly under-estimated in the past (in this particular case by morethan 20) This is a source rock so the kerogen massdecreases by the same factor This yields an underes-timation of more than 20 in TOC Note the rockmatrix mass is constant for time steps prior to23 Ma in which backstripping optimization couldbe performed toward the earliest paleogeometry Forthe second end-member scenario of idealized hydro-static decompaction the rock matrix mass is nearlyconstant which is a good decompaction adequacy

indicator for that given cell in the model Howeverthe so-measured adequacy changes throughout themodel In some areas the rock matrix mass has beenunderestimated by about 50 for nondecompactedpaleogeometries however in the case of hydrostaticdecompaction values generally range from 80 to120 of the present-day rock matrix mass Thatmeans in this particular case that hydrostatic decom-paction is more adequate (or less wrong) than nodecompaction at all The decompaction adequacycan be roughly correlated with pressure changesthrough geologic time

Figure 11 Time extractions for comparison of two end-member decompaction scenarios (see Figure 6 for location) (A) Rock matrixmass conservation through geologic time (B) Source rock burial depth through geologic time (C) Temperature and transformation ratiothrough geologic time

NEUMAIER ET AL 1347

Inadequate decompaction results in wrong depthfor stratigraphic horizons in the paleogeometriesFigure 11B shows the burial depth of the same inves-tigated source rock cell We can see that during thepassive margin phase there is a slight difference inpaleoburial depth for the two end-member scenariosDuring the compressional phase the differencechanges dramatically reaching more than 1000 m(3281 ft) in some places Toward present day bothcurves meet again because the input to the two resto-ration scenarios was the same (identical present-dayburial depth) This difference in compaction calcula-tion has severe implications on the modeled tempera-ture history and the kerogen-to-hydrocarbontransformation (Figure 11C) For an identical thermalframework in terms of modeling input parameterstemperature differences of up to 25degC (45degF) resultin TR differences of more than 20 in the early com-pressional phase before TR values meet again for thepresent-day situation

CONCLUSIONS

Structurally complex areas are attracting increasinginterest in hydrocarbon exploration Exploration riskcan be substantially narrowed by analyzing the his-tory of the basin and its petroleum systems based onstructural evolution We presented a further progressin technology that allows the integration of structuralrestoration and BPSM This combined analysis isseen as a way forward to improve modeling in thrustbelts with stacked stratigraphic sequences Despitelimitations of nonoptimized decompaction whichare added to the general uncertainty related to theindividual methods we modeled the charge and sealhistory of the Monagas fold and thrust belt inVenezuela Calibrated porosity temperature andthermal maturity as well as basin-scale stress pro-vided the regional framework through geologic timein which the petroleum systems evolved such thatthe charge of the known Santa Barbara and Onadofields and other trends could be understood We dis-cussed seal integrity of a potential exploration pros-pect by analyzing structural restoration-derived faultdisplacement through geologic time and an advancedstress forward simulation The uncertainty of using

decompacted or nondecompacted structural restora-tion has been assessed It can have important influ-ence on rock matrix mass balance through geologictime paleogeometry and paleoburial depth andtherefore affect paleotemperatures and paleomatur-ities We conclude that decompaction assuminghydrostatic pressure is the best option in that particu-lar case

REFERENCES CITED

Alberdi M and E Lafargue 1993 Vertical variations of organicmatter content in Guayuta Group (Upper Cretaceous)interior mountain belt eastern Venezuela OrganicGeochemistry v 20 p 425ndash436 doi 1010160146-6380(93)90091-O

Athy L F 1930 Density porosity and compaction of sedimen-tary rocks AAPG Bulletin v 14 p 1ndash24

Aymard R L Pimentel P Eitz P Lopez A ChaouchJ Navarro J Mijares and J G Pereira 1990 Geologic inte-gration and evaluation of Northern Monagas EasternVenezuelan Basin in J Brooks ed Classic petroleum prov-inces The Geologic Society (London) Special Publicationv 50 p 37ndash53

Baur F M Di Benedetto T Fuchs C Lampe andS Sciamanna 2009 Integrating structural geology andpetroleum systems modelingmdashA pilot project from Boliviarsquosfold and thrust belt Marine and Petroleum Geology v 26no 4 p 573ndash579 doi 101016jmarpetgeo200901004

Baur F R Littke H Wielens C Lampe and T Fuchs 2010Basin modeling meets rift analysismdashA numerical modelingstudy from the Jeanne drsquoArc basin offshore NewfoundlandCanada Marine and Petroleum Geology v 27 p 585ndash599doi 101016jmarpetgeo200906003

Beha A R O Thomsen and R Littke 2008 Thermal historyhydrocarbon generation and migration in the Horn Grabenin the Danish North Sea A 2D basin modelling studyInternational Journal of Earth Sciences v 97 p 1087ndash1100doi 101007s00531-007-0247-2

Blumenstein I O B M Krooss R di Primio W RottkeE Muumlller C Westerlage and R Littke 2008Biodegradation in numerical basin modelling A case studyfrom the Gifhorn Trough N-Germany International Journalof Earth Sciences v 97 p 1115ndash1129 doi 101007s00531-007-0272-1

Brandes C A Astorga R Littke and J Winsemann 2008Basin modelling of the Limoacuten back-arc basin (Costa Rica)Burial history and temperature evolution of an island arc-related basin-system Basin Research v 20 p 119ndash142doi 101111bre200820issue-1

Carnevali J O 1988 El Furrial oil field northeasternVenezuela First giant in foreland fold and thrust belts ofwestern hemisphere AAPG Bulletin v 72 68 p

Chevalier Y 1987 Contribution a lrsquoetude geacuteologique de lafrontiereacute Sud-Est de la plaque Caraibe Les zones internesde la chaine Sud-Caraibe sur le transect ile de Margarita-

1348 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

by the influence of the paleowater depth using theapproach of Wygrala (1989) Because of the hightopographic elevation of approximately 3500 m(11500 ft) the maximum altitude during the pastcompressional phase we adjusted the model takinginto account a temperature decrease of 6degCkm (33degF1000 ft) related to the paleoaltitude This resultsin a difference of up to 20degC (36degF) compared to sea-level temperature We then assigned the adjustedsedimentwater interface temperatures to the modelas the upper thermal boundary condition At the baseof the model we assigned a heat flow ranging from42 to 80 mW∕m2 through geologic time The highestheat flow is assumed during the rifting event beforethe basin cooled as a passive margin A thermal eventin the Eocene detected by fission track analysis(Locke and Garver 2005 Perez de Armas 2005)has been addressed by a second peak before a secondcooling phase For uncertainty analysis basal paleo-heat flow has been varied within geological meaning-ful ranges and the thermal calibration data has beenhonored as much as possible

Modeling Results

Basin and petroleum system modeling can be dividedinto simulation of the physical parameters within abasin such as pressure and temperature (basin model-ing) and the subsequent evolution of the source rocksand their derived hydrocarbon compounds (petro-leum systems modeling) In the following sectionswe use this subdivision to describe the BPSM resultscompaction modeling thermal modeling and chargehistory It is important to state that these are predic-tions based on forward modeling as just describedThe results are compared and calibrated to actualmeasurements and observations where available

Compaction Modeling

Figure 5 presents an extraction of the modeled resultsat the location of Well Alpha in the Furrial Trend Inthe absence of direct calibration data for pore pres-sure we undertook a sensitivity analysis to test thedifferent scenarios Using the model we tested howpore pressure in the main reservoir of the FurrialTrend reacts to changes in fault and seal-rock per-meability In a first step fault permeability has beenevaluated Faults were assumed open during the mainfault activity phase and closed afterward In additionwe performed several variations from that base casescenario Figure 5A shows the effect of two end-member scenarios on pore pressure within the mainreservoir of the Furrial Trend (Santa Barbara field) inwhich faults have been modeled to be either open orclosed since their formation A third scenario consistsof thrust faults that are open during the compressionalphase (from 15 Ma on) and closed afterward (from105 Ma on) This scenario has been chosen as theldquomaster scenariordquo because it results in overpressureof 23 MPa which is in the range of pressure reportedin similar structures of the Furrial Trend by Schneideret al (2004) In addition this pressure scenariomatches the calibration data for porosity (Figure 5B)

The 2D distribution of modeled overpressure atpresent day is shown in Figure 6A Apart fromimportant overpressure in the Querecual Formationthe allochthon is almost normally pressured Thethrust wedge is overpressured especially in thelow-permeability Carapita Formation Although

Table 2 Basin and Petroleum System Modeling (BPSM) InputParameters Thermal Boundary Conditions

Age (Ma)

SedimentWaterInterface

Temperature (degC)

Age (Ma)

Basal HeatFlow

(mW∕m2)min max

0 28 28 0 4226 28 28 30 6053 76 23 34 708 184 231 50 50105 62 233 107 6012 31 238 1455 8015 5 242 260 6019 5 24723 234 2527 78 235339 94 236655 146 2831455 148 287255 5 239

1336 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

the autochthon is highly overpressured close tonormal hydrostatic pressures are simulated in theforeland basin The discussed model assumption ofclosed faults in the postcompressional phase resultsin a regional pressure compartmentalization

The simulated pressure evolution through geo-logic time can be seen in Figure 7A A time extractionof a source rock cell in the thrust wedge shows a sud-den increase of both lithostatic and hydrostatic pres-sures that is a consequence of the tectonic burial

load associated with the thrusting In addition themodeled pore pressure buildup even though delayedin time can be directly linked to the thrust event Itcan be clearly seen that the timing of fault permeabil-ity directly impacts changes in the present-day porepressure The earlier the faults close the more over-pressure generated by the massive syntectonic sedi-mentation is kept until present day given adequatesealing lithologies are present The present-day over-pressure is modeled as14 MPa when the faults are

Figure 5 Plots of one-dimensional model output and available calibration data for Well Alpha (Table 3) (A) Dotted lines are hydro-static and lithostatic pressures continuous lines are pore pressure the master scenario is in black (a1 faults closing at 105 Ma) andadditional scenarios in gray (a2 open faults a3 closed faults) (B) modeled porosity (C) borehole temperature and (D) Ro The blackline shows the master scenario and the gray line the alternative high-heat-flow scenario Bold lines in the stratigraphic column arethrusts (T) and unconformities (U)

Table 3 Calibration Data for Well Alpha (Porosity Temperature Ro)

TVD (m) Porosity () TVD (m) Corrected Temperature (degC) TVD (m) Vitrinite Reflectance (Ro)

4350 15 600 40 400 0264400 15 1100 53 1200 0484800 10 1800 68 1485 068

4400 131 1700 0394800 133 2900 048

3100 0493400 0494000 054300 0524800 06

NEUMAIER ET AL 1337

open since 8 Ma as 23 MPa for 105 Ma and as37 MPa for 12 Ma Compared to this order of magni-tude changes in the permeability of the overlyingCarapita shale are negligible for the pore pressureTherefore the controlling parameter for pore pressurein the main reservoir of the Furrial trend is assumedfault permeability and resulting lateral connectivityrather than the permeability of the seal rock

Thermal Modeling

Modeled temperature and thermal maturity(Figure 5C D) have been calibrated with availablewell data Some discontinuities in the modeled Ro

trend are noted (Figure 5D) The upper kinks (egbase Pliocene) are related to unconformities withassociated erosion and the lower one (base UpperCretaceous) is due to the well penetrating a reverse

Figure 6 Two-dimensional model at present day with (A) modeled overpressure and (B) modeled thermal maturity and isothermsSource rocks marked by lithology pattern

1338 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

fault displacing mature rocks upon less mature ones(see Figure 6B for 2D view) There is a mismatchbetween the modeled Ro trend and the data in thedeeper part of the well In the Oligocene rocks themodeled temperature is too low (2degC or 36degF)whereas the modeled Ro is too high (up to 02 Ro)Any attempt to reduce the modeled Ro (eg bychanging thermal conductivity of the MioceneCarapita Formation) also results in a lower modeledpresent-day temperature Therefore the thermal sce-nario used for this calibration is a trade-off betweenmatching present-day temperature and Ro (using thekinetics of Sweeney and Burnham 1990) The basalheat flow assumed in the master scenario (Table 2)results in a modeled present-day surface heat flowranging from 37 mW∕m2 in the southern part of thethrusts to 52 mW∕m2 in parts of the uplifted areaAt the same depth the allochthonous northern partis thermally more mature than the southern autochthon(Figure 6B) because of the recent uplift Locallyinverted trends around the thrusts can be observedfor example at the location of Well Alpha (seeFigure 5D) In the autochthon the Upper Cretaceoussource rocks are entirely in the oil or even gas windowat present day because of the Neogene tectonic loadingand synorogenic deposition In the northern alloch-thon most of the Upper Cretaceous source rocks havebeen in the oil window in the past The modeled ther-mal maturity is in good agreement with the regionalmaturity suggested by Summa et al (2003)

Similar to the pressures the thrusting also affectsthe modeled temperature and Ro In the thrust wedgea direct relationship exists between the thrusting andthe increase of temperature and maturity The thrust-ing is the trigger for source rocks entering the oil win-dow within the thrust wedge (Figure 7B) and thepresent-day thermal maturity has been reached sincethe last 10 Ma Although this recent thermal historycan be considered as relatively well constrained bythe measured Ro in the calibration well this doesnot apply to the past In this thrusted area any ther-mal maturity reached prior to the thrusting onset hasbeen ldquooverprintedrdquo and therefore no thermal calibra-tion is possible for this period However theCretaceous source rocks were not buried deeply dur-ing the passive margin phase and even very hot sce-narios (basal heat flow of 100 mW∕m2) do not

significantly change their maturity (Figure 7B) Inthe uplifted allochthon several wells show low(03) to moderate (1) measured Ro (Parra et al2011) Because the allochthon experienced its maxi-mum burial and the highest temperatures during thepassive margin phase these maturity levels can beassumed as representative of the maturity prior tothe compression In contrast to the thrust wedge theconsequence of the thrusting is inverse for the alloch-thon (Figure 7C) as the hanging wall is uplifted andthe accompanying erosion removes the overburdenThe highest temperatures are simulated to have beenreached just before the thrusting event Then temper-ature decreased and Ro stagnated leaving the sourcerocks in the early mature zone until present day Theassumption of a regionally relatively homogeneousthermal field during the cooling phase of the passivemargin in combination with the relatively low matur-ities measured in the uplifted allochthon makes thehot heat-flow scenario unlikely However a laterallyvarying heat flow cannot be excluded and the conse-quently hotter paleoheat flow would have led to muchhigher temperatures prior to compression resultingalso in a higher present-day maturity level (14)despite the uplift (Figure 7C) Any variations of basalheat flow during the rift period (Late Jurassic to EarlyCretaceous) occurred too early to affect the maturityof the source rocks deposited during the lateCretaceous

Charge History

After calibration of the basinrsquos physical environ-ment the evolution of the petroleum systems wasmodeled Hydrocarbon generation was simulatedthrough geologic time using the activation energydistribution of the assigned source rock kineticsFrom the moment of hydrocarbon expulsiononward secondary migration was modeled usingthe invasion percolation method (Wilkinson 1984)It takes into account the geometry overpressure gra-dient and capillary entry pressures of the rockswhich are opposed to the buoyancy of the migratingoil and gas Figure 8 shows the time evolution of thetotal study area and Figure 9 focuses on the thrustwedge As already mentioned the described chargehistory is based on forward modeling It is consistent

NEUMAIER ET AL 1339

Figure 7 Time extractions (see Figure 6 for location) (A) Pressure development through geologic time for a source rock in the ver-tical continuity of Well Alpha Dotted lines are hydrostatic and lithostatic pressures continuous lines are pore pressure the master sce-nario is in black (a1 faults closing at 105 Ma) and additional scenarios in gray (a2 open faults a3 closed faults a4 faults closing at12 Ma a5 faults closing at 8 Ma) (B) Temperature (continuous lines) and Ro (dotted lines) through geologic time for a source rockin the vertical continuity of Well Alpha the black line shows the master scenario and the gray line the alternative high-heat-flow sce-nario (C) Temperature (continuous lines) and Ro (dotted lines) through geologic time for a source rock in the uplifted allochthonthe black line shows the master scenario and the gray line the alternative high heat flow scenario

1340 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

with the modeled compaction and thermal historyand therefore shows one scenario that explains thecharge of the Santa Barbara field and the forelandbasin Simplifications are inherent to any such work-flow and shall be considered when analyzing themodeling results

Passive Margin PhasemdashMesozoic toOligocene

Modeling results indicate that already during the LateCretaceous any Jurassic source rocks potentiallydeposited in the northern part of the section (Summaet al 2003) have reached temperatures of more than200degC (392degF) and corresponding high levels of matu-rity However the temperature at the base of theQuerecual Formation (Upper Cretaceous) was lessthan 100degC (212degF) leaving the kerogen immatureThe source rock became effective only during lateEocene with temperatures of up to 120degC (248degF)and a (reactive kerogen to hydrocarbon) transforma-tion ratio (TR) of more than 20 in the deeper distalpart of the passive margin Oil was quickly expelledmigrated vertically and seeped to the surface becauseof the general lack of seals in the overburden Duringthe Oligocene the time of deposition of the Merecurereservoir sandstones in the southern platform this sit-uation did not change apart from increased transforma-tion in the deeper basinal part of the passive margin

First oil accumulations are modeled in the Vidontildeosandstones in the basinal part in the late Oligocene(Figure 8A) in which the base of the QuerecualFormation shows TRs of up to 60 at around 130degC(266degF) In the proximal part there is no hydrocarbongeneration Although there is a general northwardtilting slope no southward migration from the distalmargin is modeled this has been checked usingdifferent migration methods including Darcy flowwith scenarios of continuous intrasource carrier bedsThis is in contrast to the studies from Gallango andParnaud (1995) and Schneider (2003) who suggestedlateral migration at that stage The difference couldbe explained by the presence of a regional top seal intheir models However our model suggests that theshallow rocks do not have sufficient sealing capacityto allow lateral hydrocarbon migration During thepassive-margin phase the basin was normally

pressured (Figure 7A) and simulated stresses havenot yet resulted in any fracturing at that stage

Main Compressional PhasemdashEarly to MiddleMiocene

With the onset of compression during early Miocene(Figure 8B) both the Pirital and Furrial thrusts wereformed and the northern distal part of the passivemargin was rapidly uplifted Important overpressuresas well as mechanical rock failure both the result ofthe rapid uplift are simulated in the allochthonDespite this uplift the kerogen-to-hydrocarbon trans-formation continued and the rate even increasedlocally close to the footwall cutoffs and where syn-orogenic deposition provided rapid overburden load-ing Oil accumulations are modeled in the freshlycreated anticlines

The tectonicndashsedimentary wedge between thetwo main thrusts experienced very rapid burial Infact the Upper Cretaceous source rocks were broughtinto depths where temperatures were as high as185degC (365degF) at 19 Ma and 225degC (437degF) at15 Ma Especially close to the cutoff of the Piritalthrust they very quickly reached a high maturity at19 Ma and reached full transformation in most placesat 12 Ma The expelled hydrocarbons migrated intothe sandstones of the Merecure Formation which weresealed by the Carapita Formation The charge toward aseries of recent anticlines is modeled by the fill-to-spillmechanism from north to south (Figure 9C D) Afterthe first oil fill the northern prospect quickly receiveda local gas charge the central structure containedmainly oil Note the presence of a lateral seal that isdue to sandndashshale juxtaposition The southernmoststructure which forms the Santa Barbara field todaywas charged by oil only at 15 Ma (Figure 9D)

The generated hydrocarbons in the southernautochthon were leaking to the surface until theCarapita Formation was compacted enough to holdsome of them (Figure 8C) At 12 Ma the wedgesource rocks were mostly spent but expulsion wasstill ongoing Important rock failure is modeled tohave occurred in the thrusted footwall breaking theformerly sealing Carapita Formation (Figure 10)This is in good agreement with cemented hydraulicfractures observed in a few cores in the Carapita

NEUMAIER ET AL 1341

Figure 8 Two-dimensional model through geologic time (selected time steps only) with modeled transformation ratio for the sourcerock layers isotherms and liquid and vapor migration and accumulation

1342 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

shales from the Furrial Trend (Roure et al 2005)This important fracturing event resulted in thedestruction of the modeled accumulation (Figure 9E)Hydrocarbon generation and expulsion were alsohappening in the southern foreland basin at transfor-mation ratios of up to 30

Late Compressional and PostcompressionalPhasemdashLate Miocene to Present Day

The late compressional phase was marked by the par-tial destruction of the allochthonous petroleum systemsby erosion (Figure 8E F) Many traps disappeared and

Figure 8 Continued

NEUMAIER ET AL 1343

the hydrocarbon generation within the source rocksstopped The erosion triggered an important sedimenttransfer toward the south contributing to further burialand increased temperatures in the thrust wedge and theforeland basin During late Miocene the petroleumsystem in the wedge matured An important northwardtilting was triggered by flexural tectonic loading andthe loading of syntectonic sediments (Summa et al2003) In combination with the presence of an effec-tive regional top seal the compacted CarapitaFormation allowed a change from dominantly verticalmigration into lateral migration First accumulationsare modeled in the Neogene synorogenic sequence

reservoirs in the south and important southwardmigration toward the foreland basin started(Figure 8G) These trends continued through Plioceneand Pleistocene times to present day filling theOnado field (Figure 8H) and other known fields inthe south (Figure 1A) of the Eastern Venezuela Basin

Prospectivity

We modeled the charge of the Santa Barbara field inthe Furrial Trend and of the Onado field in the fore-land basin (Figure 8H) Other prospects in the thrustwedge were also investigated The structure called

Figure 9 Detail of two-dimensionalmodel through geologic time (seeFigure 8) showing modeled transforma-tion ratio for the source rock layers iso-therms and liquid and vapor migrationand accumulation The Santa Barbara fieldand the prospect are shown by ellipses

1344 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

ldquoprospectrdquo (Figure 9G) looks similar to that of theSanta Barbara field because both are anticlines andthe reservoir and seal formations are the sameHowever they differ in several aspects First theprospect is in the vicinity of the Pirital thrust whichpresents a high risk of impacting the seal integrityThe prospect location is directly below the area wherewe presume the maximum of late tectonic activityoccurred (Figure 4) As modeled the CarapitaFormation the seal of the prospect was already brit-tle when this deformation occurred Calculatedstresses also predict rock failure from 12 Ma onward(Figure 10) Even though charge continued ourmodel suggests continuous leakage of the structure(Figure 9EndashG) The second major difference is thatthe source rock in the fault block of the prospect ismodeled to have already started generating and expel-ling hydrocarbons during the Oligocene and earlyMiocene (Figure 9A) before the trap formed and the

seal became effective (Figures 9C 10) Even thoughgeneration expulsion and migration continued aftertrap formation the prospect was not charged withthe earlier expelled oil In addition the present-daymaturity of the source rocks within the prospectrsquosfault block is higher than in the source rocks withinthe tectonic block of the Santa Barbara field poten-tially indicating expulsion of more gas A southndashnorth trend can be seen on the modeled accumulationcomposition with more gas toward the allochthon(Figure 9) The modeled composition of the prospectprior to seal failure was much richer in gas than inthe neighboring Santa Barbara field (Figure 9D) Infact simulation scenarios with unbreached sealsresulted in accumulation of predominantly gas Thecombination of both seal integrity and charge compo-sition concerns makes this a risky prospect In addi-tion potential diagenetic fluids related to the faultfluid flow might have altered the reservoir quality

Figure 10 Two-dimensional model (at 12 Ma) with modeled stress to failure Mohr circles (with σ standing for normal stress and τfor shear stress) and petroleum system events charts for the Santa Barbara field (left) and the prospect (right) Model predicts rock fail-ure in the prospect from about 12 Ma For further details see text

NEUMAIER ET AL 1345

Apart from the Onado oil field only minor accu-mulations are modeled in the southern Neogene syn-orogenic sequence The proportion of gas is higherthan in the Furrial trend The accumulations at thesouthern model edge are a modeling artifact repre-senting hydrocarbons (mainly oil) that in reality arethought to have charged the southern foreland basin(from mid to late Miocene times to present day)However parts of the migrated oil and gas could havebeen trapped in the Upper Cretaceous San JuanFormation south of the horst structure

The potential for conventional hydrocarbon accu-mulations in the allochthon is very limited by theimportant fracturing that occurred during the upliftperiod The basal sandstones of the Morochito piggy-back basin north of the Pirital thrust are not chargedin our 2D model because of the absence of structuraltraps Nevertheless there is a potential for strati-graphic traps to have been charged as migrationthrough the basin occurred However seal strengthmight be critical at this low-compaction stage Thesame applies for the Las Piedras Formation We thinkthere might be potential for shale oil because theQuerecual Formation is at shallow depths but showsmaturity within the oil window From the thermalmaturity point of view any potential Jurassic sourcerocks as discussed by Summa et al (2003) mightbe a gas source or shale gas target

UNCERTAINTY ANALYSIS

There is often significant uncertainty in basin geom-etry and depth of stratigraphic horizons based onuncertain seismic interpretation and velocity models(Kukla et al 2000 Brandes et al 2008 Baur et al2010) Restoring of present-day geometries to obtainpaleogeometries adds further uncertainty Additionaluncertainty is related to thermal boundary conditionssuch as paleoheat flow (Beha et al 2008) and togeochemical properties such as source rock character-istics and petroleum generation kinetics and calcula-tion of biodegradation and petroleum density(Blumenstein et al 2008) In addition to the porepressure and thermal sensitivity analysis justdescribed we varied initial TOC within a range of2 to 10 A variation of initial TOC obviously

directly scales the amount of initial kerogen availablefor transformation into hydrocarbons Neither thecharge volume nor composition of the Santa Barbarafield nor the wedge prospects are significantly sensi-tive to variation of initial TOC these traps are filledto spill in each scenario However the amount ofhydrocarbons charging the foreland basin (eg theOnado field) is reduced in the case of low initialTOC A high initial TOC only increased the modeledamount of surface seepage The sensitivity of the ini-tial HI is analogous to the sensitivity of the ini-tial TOC

In addition to the more classical uncertaintyand sensitivity analysis described we focused onquantitative analysis of the decompaction uncer-tainty when performing structural restoration Thisuncertainty is particular to the applied workflowof using structurally restored paleogeometries inBPSM Note that this uncertainty is significantlylower when using the backstripping method withthe optimization algorithms discussed One methodto analyze the adequacy of the decompactionmethod is comparison of the output rock matrixmass balance In the backstripping approach therock matrix mass is kept constant of a model cellwith a simulated decrease in porosity controlledby compaction laws and the corresponding reduc-tion of the layer thickness When using fixed pale-ogeometries from structural restoration the cellthickness is imposed However the decrease inporosity is calculated and the mass is adapted tofit the imposed cell thickness That means that therock matrix mass is constant only if ideal decom-paction has been applied

Figure 11A shows the percentage variation fromthe present-day matrix mass (100) of a given sourcerock cell through geologic time The dotted line rep-resents the ideal case with constant rock matrix massWe used two end-member decompaction scenarios todescribe the decompaction-related uncertainty that isintroduced when fixed paleogeometries replace back-stripping idealized hydrostatic decompaction assum-ing no overpressure and idealized lithostaticdecompaction assuming no effective stress (nodecompaction resulting in constant cell thicknessthrough time) In the latter case with decreasingporosity during compaction the rock matrix mass

1346 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

increases until it reaches the present-day mass Inother words rock mass matrix has been highly under-estimated in the past (in this particular case by morethan 20) This is a source rock so the kerogen massdecreases by the same factor This yields an underes-timation of more than 20 in TOC Note the rockmatrix mass is constant for time steps prior to23 Ma in which backstripping optimization couldbe performed toward the earliest paleogeometry Forthe second end-member scenario of idealized hydro-static decompaction the rock matrix mass is nearlyconstant which is a good decompaction adequacy

indicator for that given cell in the model Howeverthe so-measured adequacy changes throughout themodel In some areas the rock matrix mass has beenunderestimated by about 50 for nondecompactedpaleogeometries however in the case of hydrostaticdecompaction values generally range from 80 to120 of the present-day rock matrix mass Thatmeans in this particular case that hydrostatic decom-paction is more adequate (or less wrong) than nodecompaction at all The decompaction adequacycan be roughly correlated with pressure changesthrough geologic time

Figure 11 Time extractions for comparison of two end-member decompaction scenarios (see Figure 6 for location) (A) Rock matrixmass conservation through geologic time (B) Source rock burial depth through geologic time (C) Temperature and transformation ratiothrough geologic time

NEUMAIER ET AL 1347

Inadequate decompaction results in wrong depthfor stratigraphic horizons in the paleogeometriesFigure 11B shows the burial depth of the same inves-tigated source rock cell We can see that during thepassive margin phase there is a slight difference inpaleoburial depth for the two end-member scenariosDuring the compressional phase the differencechanges dramatically reaching more than 1000 m(3281 ft) in some places Toward present day bothcurves meet again because the input to the two resto-ration scenarios was the same (identical present-dayburial depth) This difference in compaction calcula-tion has severe implications on the modeled tempera-ture history and the kerogen-to-hydrocarbontransformation (Figure 11C) For an identical thermalframework in terms of modeling input parameterstemperature differences of up to 25degC (45degF) resultin TR differences of more than 20 in the early com-pressional phase before TR values meet again for thepresent-day situation

CONCLUSIONS

Structurally complex areas are attracting increasinginterest in hydrocarbon exploration Exploration riskcan be substantially narrowed by analyzing the his-tory of the basin and its petroleum systems based onstructural evolution We presented a further progressin technology that allows the integration of structuralrestoration and BPSM This combined analysis isseen as a way forward to improve modeling in thrustbelts with stacked stratigraphic sequences Despitelimitations of nonoptimized decompaction whichare added to the general uncertainty related to theindividual methods we modeled the charge and sealhistory of the Monagas fold and thrust belt inVenezuela Calibrated porosity temperature andthermal maturity as well as basin-scale stress pro-vided the regional framework through geologic timein which the petroleum systems evolved such thatthe charge of the known Santa Barbara and Onadofields and other trends could be understood We dis-cussed seal integrity of a potential exploration pros-pect by analyzing structural restoration-derived faultdisplacement through geologic time and an advancedstress forward simulation The uncertainty of using

decompacted or nondecompacted structural restora-tion has been assessed It can have important influ-ence on rock matrix mass balance through geologictime paleogeometry and paleoburial depth andtherefore affect paleotemperatures and paleomatur-ities We conclude that decompaction assuminghydrostatic pressure is the best option in that particu-lar case

REFERENCES CITED

Alberdi M and E Lafargue 1993 Vertical variations of organicmatter content in Guayuta Group (Upper Cretaceous)interior mountain belt eastern Venezuela OrganicGeochemistry v 20 p 425ndash436 doi 1010160146-6380(93)90091-O

Athy L F 1930 Density porosity and compaction of sedimen-tary rocks AAPG Bulletin v 14 p 1ndash24

Aymard R L Pimentel P Eitz P Lopez A ChaouchJ Navarro J Mijares and J G Pereira 1990 Geologic inte-gration and evaluation of Northern Monagas EasternVenezuelan Basin in J Brooks ed Classic petroleum prov-inces The Geologic Society (London) Special Publicationv 50 p 37ndash53

Baur F M Di Benedetto T Fuchs C Lampe andS Sciamanna 2009 Integrating structural geology andpetroleum systems modelingmdashA pilot project from Boliviarsquosfold and thrust belt Marine and Petroleum Geology v 26no 4 p 573ndash579 doi 101016jmarpetgeo200901004

Baur F R Littke H Wielens C Lampe and T Fuchs 2010Basin modeling meets rift analysismdashA numerical modelingstudy from the Jeanne drsquoArc basin offshore NewfoundlandCanada Marine and Petroleum Geology v 27 p 585ndash599doi 101016jmarpetgeo200906003

Beha A R O Thomsen and R Littke 2008 Thermal historyhydrocarbon generation and migration in the Horn Grabenin the Danish North Sea A 2D basin modelling studyInternational Journal of Earth Sciences v 97 p 1087ndash1100doi 101007s00531-007-0247-2

Blumenstein I O B M Krooss R di Primio W RottkeE Muumlller C Westerlage and R Littke 2008Biodegradation in numerical basin modelling A case studyfrom the Gifhorn Trough N-Germany International Journalof Earth Sciences v 97 p 1115ndash1129 doi 101007s00531-007-0272-1

Brandes C A Astorga R Littke and J Winsemann 2008Basin modelling of the Limoacuten back-arc basin (Costa Rica)Burial history and temperature evolution of an island arc-related basin-system Basin Research v 20 p 119ndash142doi 101111bre200820issue-1

Carnevali J O 1988 El Furrial oil field northeasternVenezuela First giant in foreland fold and thrust belts ofwestern hemisphere AAPG Bulletin v 72 68 p

Chevalier Y 1987 Contribution a lrsquoetude geacuteologique de lafrontiereacute Sud-Est de la plaque Caraibe Les zones internesde la chaine Sud-Caraibe sur le transect ile de Margarita-

1348 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

the autochthon is highly overpressured close tonormal hydrostatic pressures are simulated in theforeland basin The discussed model assumption ofclosed faults in the postcompressional phase resultsin a regional pressure compartmentalization

The simulated pressure evolution through geo-logic time can be seen in Figure 7A A time extractionof a source rock cell in the thrust wedge shows a sud-den increase of both lithostatic and hydrostatic pres-sures that is a consequence of the tectonic burial

load associated with the thrusting In addition themodeled pore pressure buildup even though delayedin time can be directly linked to the thrust event Itcan be clearly seen that the timing of fault permeabil-ity directly impacts changes in the present-day porepressure The earlier the faults close the more over-pressure generated by the massive syntectonic sedi-mentation is kept until present day given adequatesealing lithologies are present The present-day over-pressure is modeled as14 MPa when the faults are

Figure 5 Plots of one-dimensional model output and available calibration data for Well Alpha (Table 3) (A) Dotted lines are hydro-static and lithostatic pressures continuous lines are pore pressure the master scenario is in black (a1 faults closing at 105 Ma) andadditional scenarios in gray (a2 open faults a3 closed faults) (B) modeled porosity (C) borehole temperature and (D) Ro The blackline shows the master scenario and the gray line the alternative high-heat-flow scenario Bold lines in the stratigraphic column arethrusts (T) and unconformities (U)

Table 3 Calibration Data for Well Alpha (Porosity Temperature Ro)

TVD (m) Porosity () TVD (m) Corrected Temperature (degC) TVD (m) Vitrinite Reflectance (Ro)

4350 15 600 40 400 0264400 15 1100 53 1200 0484800 10 1800 68 1485 068

4400 131 1700 0394800 133 2900 048

3100 0493400 0494000 054300 0524800 06

NEUMAIER ET AL 1337

open since 8 Ma as 23 MPa for 105 Ma and as37 MPa for 12 Ma Compared to this order of magni-tude changes in the permeability of the overlyingCarapita shale are negligible for the pore pressureTherefore the controlling parameter for pore pressurein the main reservoir of the Furrial trend is assumedfault permeability and resulting lateral connectivityrather than the permeability of the seal rock

Thermal Modeling

Modeled temperature and thermal maturity(Figure 5C D) have been calibrated with availablewell data Some discontinuities in the modeled Ro

trend are noted (Figure 5D) The upper kinks (egbase Pliocene) are related to unconformities withassociated erosion and the lower one (base UpperCretaceous) is due to the well penetrating a reverse

Figure 6 Two-dimensional model at present day with (A) modeled overpressure and (B) modeled thermal maturity and isothermsSource rocks marked by lithology pattern

1338 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

fault displacing mature rocks upon less mature ones(see Figure 6B for 2D view) There is a mismatchbetween the modeled Ro trend and the data in thedeeper part of the well In the Oligocene rocks themodeled temperature is too low (2degC or 36degF)whereas the modeled Ro is too high (up to 02 Ro)Any attempt to reduce the modeled Ro (eg bychanging thermal conductivity of the MioceneCarapita Formation) also results in a lower modeledpresent-day temperature Therefore the thermal sce-nario used for this calibration is a trade-off betweenmatching present-day temperature and Ro (using thekinetics of Sweeney and Burnham 1990) The basalheat flow assumed in the master scenario (Table 2)results in a modeled present-day surface heat flowranging from 37 mW∕m2 in the southern part of thethrusts to 52 mW∕m2 in parts of the uplifted areaAt the same depth the allochthonous northern partis thermally more mature than the southern autochthon(Figure 6B) because of the recent uplift Locallyinverted trends around the thrusts can be observedfor example at the location of Well Alpha (seeFigure 5D) In the autochthon the Upper Cretaceoussource rocks are entirely in the oil or even gas windowat present day because of the Neogene tectonic loadingand synorogenic deposition In the northern alloch-thon most of the Upper Cretaceous source rocks havebeen in the oil window in the past The modeled ther-mal maturity is in good agreement with the regionalmaturity suggested by Summa et al (2003)

Similar to the pressures the thrusting also affectsthe modeled temperature and Ro In the thrust wedgea direct relationship exists between the thrusting andthe increase of temperature and maturity The thrust-ing is the trigger for source rocks entering the oil win-dow within the thrust wedge (Figure 7B) and thepresent-day thermal maturity has been reached sincethe last 10 Ma Although this recent thermal historycan be considered as relatively well constrained bythe measured Ro in the calibration well this doesnot apply to the past In this thrusted area any ther-mal maturity reached prior to the thrusting onset hasbeen ldquooverprintedrdquo and therefore no thermal calibra-tion is possible for this period However theCretaceous source rocks were not buried deeply dur-ing the passive margin phase and even very hot sce-narios (basal heat flow of 100 mW∕m2) do not

significantly change their maturity (Figure 7B) Inthe uplifted allochthon several wells show low(03) to moderate (1) measured Ro (Parra et al2011) Because the allochthon experienced its maxi-mum burial and the highest temperatures during thepassive margin phase these maturity levels can beassumed as representative of the maturity prior tothe compression In contrast to the thrust wedge theconsequence of the thrusting is inverse for the alloch-thon (Figure 7C) as the hanging wall is uplifted andthe accompanying erosion removes the overburdenThe highest temperatures are simulated to have beenreached just before the thrusting event Then temper-ature decreased and Ro stagnated leaving the sourcerocks in the early mature zone until present day Theassumption of a regionally relatively homogeneousthermal field during the cooling phase of the passivemargin in combination with the relatively low matur-ities measured in the uplifted allochthon makes thehot heat-flow scenario unlikely However a laterallyvarying heat flow cannot be excluded and the conse-quently hotter paleoheat flow would have led to muchhigher temperatures prior to compression resultingalso in a higher present-day maturity level (14)despite the uplift (Figure 7C) Any variations of basalheat flow during the rift period (Late Jurassic to EarlyCretaceous) occurred too early to affect the maturityof the source rocks deposited during the lateCretaceous

Charge History

After calibration of the basinrsquos physical environ-ment the evolution of the petroleum systems wasmodeled Hydrocarbon generation was simulatedthrough geologic time using the activation energydistribution of the assigned source rock kineticsFrom the moment of hydrocarbon expulsiononward secondary migration was modeled usingthe invasion percolation method (Wilkinson 1984)It takes into account the geometry overpressure gra-dient and capillary entry pressures of the rockswhich are opposed to the buoyancy of the migratingoil and gas Figure 8 shows the time evolution of thetotal study area and Figure 9 focuses on the thrustwedge As already mentioned the described chargehistory is based on forward modeling It is consistent

NEUMAIER ET AL 1339

Figure 7 Time extractions (see Figure 6 for location) (A) Pressure development through geologic time for a source rock in the ver-tical continuity of Well Alpha Dotted lines are hydrostatic and lithostatic pressures continuous lines are pore pressure the master sce-nario is in black (a1 faults closing at 105 Ma) and additional scenarios in gray (a2 open faults a3 closed faults a4 faults closing at12 Ma a5 faults closing at 8 Ma) (B) Temperature (continuous lines) and Ro (dotted lines) through geologic time for a source rockin the vertical continuity of Well Alpha the black line shows the master scenario and the gray line the alternative high-heat-flow sce-nario (C) Temperature (continuous lines) and Ro (dotted lines) through geologic time for a source rock in the uplifted allochthonthe black line shows the master scenario and the gray line the alternative high heat flow scenario

1340 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

with the modeled compaction and thermal historyand therefore shows one scenario that explains thecharge of the Santa Barbara field and the forelandbasin Simplifications are inherent to any such work-flow and shall be considered when analyzing themodeling results

Passive Margin PhasemdashMesozoic toOligocene

Modeling results indicate that already during the LateCretaceous any Jurassic source rocks potentiallydeposited in the northern part of the section (Summaet al 2003) have reached temperatures of more than200degC (392degF) and corresponding high levels of matu-rity However the temperature at the base of theQuerecual Formation (Upper Cretaceous) was lessthan 100degC (212degF) leaving the kerogen immatureThe source rock became effective only during lateEocene with temperatures of up to 120degC (248degF)and a (reactive kerogen to hydrocarbon) transforma-tion ratio (TR) of more than 20 in the deeper distalpart of the passive margin Oil was quickly expelledmigrated vertically and seeped to the surface becauseof the general lack of seals in the overburden Duringthe Oligocene the time of deposition of the Merecurereservoir sandstones in the southern platform this sit-uation did not change apart from increased transforma-tion in the deeper basinal part of the passive margin

First oil accumulations are modeled in the Vidontildeosandstones in the basinal part in the late Oligocene(Figure 8A) in which the base of the QuerecualFormation shows TRs of up to 60 at around 130degC(266degF) In the proximal part there is no hydrocarbongeneration Although there is a general northwardtilting slope no southward migration from the distalmargin is modeled this has been checked usingdifferent migration methods including Darcy flowwith scenarios of continuous intrasource carrier bedsThis is in contrast to the studies from Gallango andParnaud (1995) and Schneider (2003) who suggestedlateral migration at that stage The difference couldbe explained by the presence of a regional top seal intheir models However our model suggests that theshallow rocks do not have sufficient sealing capacityto allow lateral hydrocarbon migration During thepassive-margin phase the basin was normally

pressured (Figure 7A) and simulated stresses havenot yet resulted in any fracturing at that stage

Main Compressional PhasemdashEarly to MiddleMiocene

With the onset of compression during early Miocene(Figure 8B) both the Pirital and Furrial thrusts wereformed and the northern distal part of the passivemargin was rapidly uplifted Important overpressuresas well as mechanical rock failure both the result ofthe rapid uplift are simulated in the allochthonDespite this uplift the kerogen-to-hydrocarbon trans-formation continued and the rate even increasedlocally close to the footwall cutoffs and where syn-orogenic deposition provided rapid overburden load-ing Oil accumulations are modeled in the freshlycreated anticlines

The tectonicndashsedimentary wedge between thetwo main thrusts experienced very rapid burial Infact the Upper Cretaceous source rocks were broughtinto depths where temperatures were as high as185degC (365degF) at 19 Ma and 225degC (437degF) at15 Ma Especially close to the cutoff of the Piritalthrust they very quickly reached a high maturity at19 Ma and reached full transformation in most placesat 12 Ma The expelled hydrocarbons migrated intothe sandstones of the Merecure Formation which weresealed by the Carapita Formation The charge toward aseries of recent anticlines is modeled by the fill-to-spillmechanism from north to south (Figure 9C D) Afterthe first oil fill the northern prospect quickly receiveda local gas charge the central structure containedmainly oil Note the presence of a lateral seal that isdue to sandndashshale juxtaposition The southernmoststructure which forms the Santa Barbara field todaywas charged by oil only at 15 Ma (Figure 9D)

The generated hydrocarbons in the southernautochthon were leaking to the surface until theCarapita Formation was compacted enough to holdsome of them (Figure 8C) At 12 Ma the wedgesource rocks were mostly spent but expulsion wasstill ongoing Important rock failure is modeled tohave occurred in the thrusted footwall breaking theformerly sealing Carapita Formation (Figure 10)This is in good agreement with cemented hydraulicfractures observed in a few cores in the Carapita

NEUMAIER ET AL 1341

Figure 8 Two-dimensional model through geologic time (selected time steps only) with modeled transformation ratio for the sourcerock layers isotherms and liquid and vapor migration and accumulation

1342 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

shales from the Furrial Trend (Roure et al 2005)This important fracturing event resulted in thedestruction of the modeled accumulation (Figure 9E)Hydrocarbon generation and expulsion were alsohappening in the southern foreland basin at transfor-mation ratios of up to 30

Late Compressional and PostcompressionalPhasemdashLate Miocene to Present Day

The late compressional phase was marked by the par-tial destruction of the allochthonous petroleum systemsby erosion (Figure 8E F) Many traps disappeared and

Figure 8 Continued

NEUMAIER ET AL 1343

the hydrocarbon generation within the source rocksstopped The erosion triggered an important sedimenttransfer toward the south contributing to further burialand increased temperatures in the thrust wedge and theforeland basin During late Miocene the petroleumsystem in the wedge matured An important northwardtilting was triggered by flexural tectonic loading andthe loading of syntectonic sediments (Summa et al2003) In combination with the presence of an effec-tive regional top seal the compacted CarapitaFormation allowed a change from dominantly verticalmigration into lateral migration First accumulationsare modeled in the Neogene synorogenic sequence

reservoirs in the south and important southwardmigration toward the foreland basin started(Figure 8G) These trends continued through Plioceneand Pleistocene times to present day filling theOnado field (Figure 8H) and other known fields inthe south (Figure 1A) of the Eastern Venezuela Basin

Prospectivity

We modeled the charge of the Santa Barbara field inthe Furrial Trend and of the Onado field in the fore-land basin (Figure 8H) Other prospects in the thrustwedge were also investigated The structure called

Figure 9 Detail of two-dimensionalmodel through geologic time (seeFigure 8) showing modeled transforma-tion ratio for the source rock layers iso-therms and liquid and vapor migrationand accumulation The Santa Barbara fieldand the prospect are shown by ellipses

1344 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

ldquoprospectrdquo (Figure 9G) looks similar to that of theSanta Barbara field because both are anticlines andthe reservoir and seal formations are the sameHowever they differ in several aspects First theprospect is in the vicinity of the Pirital thrust whichpresents a high risk of impacting the seal integrityThe prospect location is directly below the area wherewe presume the maximum of late tectonic activityoccurred (Figure 4) As modeled the CarapitaFormation the seal of the prospect was already brit-tle when this deformation occurred Calculatedstresses also predict rock failure from 12 Ma onward(Figure 10) Even though charge continued ourmodel suggests continuous leakage of the structure(Figure 9EndashG) The second major difference is thatthe source rock in the fault block of the prospect ismodeled to have already started generating and expel-ling hydrocarbons during the Oligocene and earlyMiocene (Figure 9A) before the trap formed and the

seal became effective (Figures 9C 10) Even thoughgeneration expulsion and migration continued aftertrap formation the prospect was not charged withthe earlier expelled oil In addition the present-daymaturity of the source rocks within the prospectrsquosfault block is higher than in the source rocks withinthe tectonic block of the Santa Barbara field poten-tially indicating expulsion of more gas A southndashnorth trend can be seen on the modeled accumulationcomposition with more gas toward the allochthon(Figure 9) The modeled composition of the prospectprior to seal failure was much richer in gas than inthe neighboring Santa Barbara field (Figure 9D) Infact simulation scenarios with unbreached sealsresulted in accumulation of predominantly gas Thecombination of both seal integrity and charge compo-sition concerns makes this a risky prospect In addi-tion potential diagenetic fluids related to the faultfluid flow might have altered the reservoir quality

Figure 10 Two-dimensional model (at 12 Ma) with modeled stress to failure Mohr circles (with σ standing for normal stress and τfor shear stress) and petroleum system events charts for the Santa Barbara field (left) and the prospect (right) Model predicts rock fail-ure in the prospect from about 12 Ma For further details see text

NEUMAIER ET AL 1345

Apart from the Onado oil field only minor accu-mulations are modeled in the southern Neogene syn-orogenic sequence The proportion of gas is higherthan in the Furrial trend The accumulations at thesouthern model edge are a modeling artifact repre-senting hydrocarbons (mainly oil) that in reality arethought to have charged the southern foreland basin(from mid to late Miocene times to present day)However parts of the migrated oil and gas could havebeen trapped in the Upper Cretaceous San JuanFormation south of the horst structure

The potential for conventional hydrocarbon accu-mulations in the allochthon is very limited by theimportant fracturing that occurred during the upliftperiod The basal sandstones of the Morochito piggy-back basin north of the Pirital thrust are not chargedin our 2D model because of the absence of structuraltraps Nevertheless there is a potential for strati-graphic traps to have been charged as migrationthrough the basin occurred However seal strengthmight be critical at this low-compaction stage Thesame applies for the Las Piedras Formation We thinkthere might be potential for shale oil because theQuerecual Formation is at shallow depths but showsmaturity within the oil window From the thermalmaturity point of view any potential Jurassic sourcerocks as discussed by Summa et al (2003) mightbe a gas source or shale gas target

UNCERTAINTY ANALYSIS

There is often significant uncertainty in basin geom-etry and depth of stratigraphic horizons based onuncertain seismic interpretation and velocity models(Kukla et al 2000 Brandes et al 2008 Baur et al2010) Restoring of present-day geometries to obtainpaleogeometries adds further uncertainty Additionaluncertainty is related to thermal boundary conditionssuch as paleoheat flow (Beha et al 2008) and togeochemical properties such as source rock character-istics and petroleum generation kinetics and calcula-tion of biodegradation and petroleum density(Blumenstein et al 2008) In addition to the porepressure and thermal sensitivity analysis justdescribed we varied initial TOC within a range of2 to 10 A variation of initial TOC obviously

directly scales the amount of initial kerogen availablefor transformation into hydrocarbons Neither thecharge volume nor composition of the Santa Barbarafield nor the wedge prospects are significantly sensi-tive to variation of initial TOC these traps are filledto spill in each scenario However the amount ofhydrocarbons charging the foreland basin (eg theOnado field) is reduced in the case of low initialTOC A high initial TOC only increased the modeledamount of surface seepage The sensitivity of the ini-tial HI is analogous to the sensitivity of the ini-tial TOC

In addition to the more classical uncertaintyand sensitivity analysis described we focused onquantitative analysis of the decompaction uncer-tainty when performing structural restoration Thisuncertainty is particular to the applied workflowof using structurally restored paleogeometries inBPSM Note that this uncertainty is significantlylower when using the backstripping method withthe optimization algorithms discussed One methodto analyze the adequacy of the decompactionmethod is comparison of the output rock matrixmass balance In the backstripping approach therock matrix mass is kept constant of a model cellwith a simulated decrease in porosity controlledby compaction laws and the corresponding reduc-tion of the layer thickness When using fixed pale-ogeometries from structural restoration the cellthickness is imposed However the decrease inporosity is calculated and the mass is adapted tofit the imposed cell thickness That means that therock matrix mass is constant only if ideal decom-paction has been applied

Figure 11A shows the percentage variation fromthe present-day matrix mass (100) of a given sourcerock cell through geologic time The dotted line rep-resents the ideal case with constant rock matrix massWe used two end-member decompaction scenarios todescribe the decompaction-related uncertainty that isintroduced when fixed paleogeometries replace back-stripping idealized hydrostatic decompaction assum-ing no overpressure and idealized lithostaticdecompaction assuming no effective stress (nodecompaction resulting in constant cell thicknessthrough time) In the latter case with decreasingporosity during compaction the rock matrix mass

1346 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

increases until it reaches the present-day mass Inother words rock mass matrix has been highly under-estimated in the past (in this particular case by morethan 20) This is a source rock so the kerogen massdecreases by the same factor This yields an underes-timation of more than 20 in TOC Note the rockmatrix mass is constant for time steps prior to23 Ma in which backstripping optimization couldbe performed toward the earliest paleogeometry Forthe second end-member scenario of idealized hydro-static decompaction the rock matrix mass is nearlyconstant which is a good decompaction adequacy

indicator for that given cell in the model Howeverthe so-measured adequacy changes throughout themodel In some areas the rock matrix mass has beenunderestimated by about 50 for nondecompactedpaleogeometries however in the case of hydrostaticdecompaction values generally range from 80 to120 of the present-day rock matrix mass Thatmeans in this particular case that hydrostatic decom-paction is more adequate (or less wrong) than nodecompaction at all The decompaction adequacycan be roughly correlated with pressure changesthrough geologic time

Figure 11 Time extractions for comparison of two end-member decompaction scenarios (see Figure 6 for location) (A) Rock matrixmass conservation through geologic time (B) Source rock burial depth through geologic time (C) Temperature and transformation ratiothrough geologic time

NEUMAIER ET AL 1347

Inadequate decompaction results in wrong depthfor stratigraphic horizons in the paleogeometriesFigure 11B shows the burial depth of the same inves-tigated source rock cell We can see that during thepassive margin phase there is a slight difference inpaleoburial depth for the two end-member scenariosDuring the compressional phase the differencechanges dramatically reaching more than 1000 m(3281 ft) in some places Toward present day bothcurves meet again because the input to the two resto-ration scenarios was the same (identical present-dayburial depth) This difference in compaction calcula-tion has severe implications on the modeled tempera-ture history and the kerogen-to-hydrocarbontransformation (Figure 11C) For an identical thermalframework in terms of modeling input parameterstemperature differences of up to 25degC (45degF) resultin TR differences of more than 20 in the early com-pressional phase before TR values meet again for thepresent-day situation

CONCLUSIONS

Structurally complex areas are attracting increasinginterest in hydrocarbon exploration Exploration riskcan be substantially narrowed by analyzing the his-tory of the basin and its petroleum systems based onstructural evolution We presented a further progressin technology that allows the integration of structuralrestoration and BPSM This combined analysis isseen as a way forward to improve modeling in thrustbelts with stacked stratigraphic sequences Despitelimitations of nonoptimized decompaction whichare added to the general uncertainty related to theindividual methods we modeled the charge and sealhistory of the Monagas fold and thrust belt inVenezuela Calibrated porosity temperature andthermal maturity as well as basin-scale stress pro-vided the regional framework through geologic timein which the petroleum systems evolved such thatthe charge of the known Santa Barbara and Onadofields and other trends could be understood We dis-cussed seal integrity of a potential exploration pros-pect by analyzing structural restoration-derived faultdisplacement through geologic time and an advancedstress forward simulation The uncertainty of using

decompacted or nondecompacted structural restora-tion has been assessed It can have important influ-ence on rock matrix mass balance through geologictime paleogeometry and paleoburial depth andtherefore affect paleotemperatures and paleomatur-ities We conclude that decompaction assuminghydrostatic pressure is the best option in that particu-lar case

REFERENCES CITED

Alberdi M and E Lafargue 1993 Vertical variations of organicmatter content in Guayuta Group (Upper Cretaceous)interior mountain belt eastern Venezuela OrganicGeochemistry v 20 p 425ndash436 doi 1010160146-6380(93)90091-O

Athy L F 1930 Density porosity and compaction of sedimen-tary rocks AAPG Bulletin v 14 p 1ndash24

Aymard R L Pimentel P Eitz P Lopez A ChaouchJ Navarro J Mijares and J G Pereira 1990 Geologic inte-gration and evaluation of Northern Monagas EasternVenezuelan Basin in J Brooks ed Classic petroleum prov-inces The Geologic Society (London) Special Publicationv 50 p 37ndash53

Baur F M Di Benedetto T Fuchs C Lampe andS Sciamanna 2009 Integrating structural geology andpetroleum systems modelingmdashA pilot project from Boliviarsquosfold and thrust belt Marine and Petroleum Geology v 26no 4 p 573ndash579 doi 101016jmarpetgeo200901004

Baur F R Littke H Wielens C Lampe and T Fuchs 2010Basin modeling meets rift analysismdashA numerical modelingstudy from the Jeanne drsquoArc basin offshore NewfoundlandCanada Marine and Petroleum Geology v 27 p 585ndash599doi 101016jmarpetgeo200906003

Beha A R O Thomsen and R Littke 2008 Thermal historyhydrocarbon generation and migration in the Horn Grabenin the Danish North Sea A 2D basin modelling studyInternational Journal of Earth Sciences v 97 p 1087ndash1100doi 101007s00531-007-0247-2

Blumenstein I O B M Krooss R di Primio W RottkeE Muumlller C Westerlage and R Littke 2008Biodegradation in numerical basin modelling A case studyfrom the Gifhorn Trough N-Germany International Journalof Earth Sciences v 97 p 1115ndash1129 doi 101007s00531-007-0272-1

Brandes C A Astorga R Littke and J Winsemann 2008Basin modelling of the Limoacuten back-arc basin (Costa Rica)Burial history and temperature evolution of an island arc-related basin-system Basin Research v 20 p 119ndash142doi 101111bre200820issue-1

Carnevali J O 1988 El Furrial oil field northeasternVenezuela First giant in foreland fold and thrust belts ofwestern hemisphere AAPG Bulletin v 72 68 p

Chevalier Y 1987 Contribution a lrsquoetude geacuteologique de lafrontiereacute Sud-Est de la plaque Caraibe Les zones internesde la chaine Sud-Caraibe sur le transect ile de Margarita-

1348 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

open since 8 Ma as 23 MPa for 105 Ma and as37 MPa for 12 Ma Compared to this order of magni-tude changes in the permeability of the overlyingCarapita shale are negligible for the pore pressureTherefore the controlling parameter for pore pressurein the main reservoir of the Furrial trend is assumedfault permeability and resulting lateral connectivityrather than the permeability of the seal rock

Thermal Modeling

Modeled temperature and thermal maturity(Figure 5C D) have been calibrated with availablewell data Some discontinuities in the modeled Ro

trend are noted (Figure 5D) The upper kinks (egbase Pliocene) are related to unconformities withassociated erosion and the lower one (base UpperCretaceous) is due to the well penetrating a reverse

Figure 6 Two-dimensional model at present day with (A) modeled overpressure and (B) modeled thermal maturity and isothermsSource rocks marked by lithology pattern

1338 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

fault displacing mature rocks upon less mature ones(see Figure 6B for 2D view) There is a mismatchbetween the modeled Ro trend and the data in thedeeper part of the well In the Oligocene rocks themodeled temperature is too low (2degC or 36degF)whereas the modeled Ro is too high (up to 02 Ro)Any attempt to reduce the modeled Ro (eg bychanging thermal conductivity of the MioceneCarapita Formation) also results in a lower modeledpresent-day temperature Therefore the thermal sce-nario used for this calibration is a trade-off betweenmatching present-day temperature and Ro (using thekinetics of Sweeney and Burnham 1990) The basalheat flow assumed in the master scenario (Table 2)results in a modeled present-day surface heat flowranging from 37 mW∕m2 in the southern part of thethrusts to 52 mW∕m2 in parts of the uplifted areaAt the same depth the allochthonous northern partis thermally more mature than the southern autochthon(Figure 6B) because of the recent uplift Locallyinverted trends around the thrusts can be observedfor example at the location of Well Alpha (seeFigure 5D) In the autochthon the Upper Cretaceoussource rocks are entirely in the oil or even gas windowat present day because of the Neogene tectonic loadingand synorogenic deposition In the northern alloch-thon most of the Upper Cretaceous source rocks havebeen in the oil window in the past The modeled ther-mal maturity is in good agreement with the regionalmaturity suggested by Summa et al (2003)

Similar to the pressures the thrusting also affectsthe modeled temperature and Ro In the thrust wedgea direct relationship exists between the thrusting andthe increase of temperature and maturity The thrust-ing is the trigger for source rocks entering the oil win-dow within the thrust wedge (Figure 7B) and thepresent-day thermal maturity has been reached sincethe last 10 Ma Although this recent thermal historycan be considered as relatively well constrained bythe measured Ro in the calibration well this doesnot apply to the past In this thrusted area any ther-mal maturity reached prior to the thrusting onset hasbeen ldquooverprintedrdquo and therefore no thermal calibra-tion is possible for this period However theCretaceous source rocks were not buried deeply dur-ing the passive margin phase and even very hot sce-narios (basal heat flow of 100 mW∕m2) do not

significantly change their maturity (Figure 7B) Inthe uplifted allochthon several wells show low(03) to moderate (1) measured Ro (Parra et al2011) Because the allochthon experienced its maxi-mum burial and the highest temperatures during thepassive margin phase these maturity levels can beassumed as representative of the maturity prior tothe compression In contrast to the thrust wedge theconsequence of the thrusting is inverse for the alloch-thon (Figure 7C) as the hanging wall is uplifted andthe accompanying erosion removes the overburdenThe highest temperatures are simulated to have beenreached just before the thrusting event Then temper-ature decreased and Ro stagnated leaving the sourcerocks in the early mature zone until present day Theassumption of a regionally relatively homogeneousthermal field during the cooling phase of the passivemargin in combination with the relatively low matur-ities measured in the uplifted allochthon makes thehot heat-flow scenario unlikely However a laterallyvarying heat flow cannot be excluded and the conse-quently hotter paleoheat flow would have led to muchhigher temperatures prior to compression resultingalso in a higher present-day maturity level (14)despite the uplift (Figure 7C) Any variations of basalheat flow during the rift period (Late Jurassic to EarlyCretaceous) occurred too early to affect the maturityof the source rocks deposited during the lateCretaceous

Charge History

After calibration of the basinrsquos physical environ-ment the evolution of the petroleum systems wasmodeled Hydrocarbon generation was simulatedthrough geologic time using the activation energydistribution of the assigned source rock kineticsFrom the moment of hydrocarbon expulsiononward secondary migration was modeled usingthe invasion percolation method (Wilkinson 1984)It takes into account the geometry overpressure gra-dient and capillary entry pressures of the rockswhich are opposed to the buoyancy of the migratingoil and gas Figure 8 shows the time evolution of thetotal study area and Figure 9 focuses on the thrustwedge As already mentioned the described chargehistory is based on forward modeling It is consistent

NEUMAIER ET AL 1339

Figure 7 Time extractions (see Figure 6 for location) (A) Pressure development through geologic time for a source rock in the ver-tical continuity of Well Alpha Dotted lines are hydrostatic and lithostatic pressures continuous lines are pore pressure the master sce-nario is in black (a1 faults closing at 105 Ma) and additional scenarios in gray (a2 open faults a3 closed faults a4 faults closing at12 Ma a5 faults closing at 8 Ma) (B) Temperature (continuous lines) and Ro (dotted lines) through geologic time for a source rockin the vertical continuity of Well Alpha the black line shows the master scenario and the gray line the alternative high-heat-flow sce-nario (C) Temperature (continuous lines) and Ro (dotted lines) through geologic time for a source rock in the uplifted allochthonthe black line shows the master scenario and the gray line the alternative high heat flow scenario

1340 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

with the modeled compaction and thermal historyand therefore shows one scenario that explains thecharge of the Santa Barbara field and the forelandbasin Simplifications are inherent to any such work-flow and shall be considered when analyzing themodeling results

Passive Margin PhasemdashMesozoic toOligocene

Modeling results indicate that already during the LateCretaceous any Jurassic source rocks potentiallydeposited in the northern part of the section (Summaet al 2003) have reached temperatures of more than200degC (392degF) and corresponding high levels of matu-rity However the temperature at the base of theQuerecual Formation (Upper Cretaceous) was lessthan 100degC (212degF) leaving the kerogen immatureThe source rock became effective only during lateEocene with temperatures of up to 120degC (248degF)and a (reactive kerogen to hydrocarbon) transforma-tion ratio (TR) of more than 20 in the deeper distalpart of the passive margin Oil was quickly expelledmigrated vertically and seeped to the surface becauseof the general lack of seals in the overburden Duringthe Oligocene the time of deposition of the Merecurereservoir sandstones in the southern platform this sit-uation did not change apart from increased transforma-tion in the deeper basinal part of the passive margin

First oil accumulations are modeled in the Vidontildeosandstones in the basinal part in the late Oligocene(Figure 8A) in which the base of the QuerecualFormation shows TRs of up to 60 at around 130degC(266degF) In the proximal part there is no hydrocarbongeneration Although there is a general northwardtilting slope no southward migration from the distalmargin is modeled this has been checked usingdifferent migration methods including Darcy flowwith scenarios of continuous intrasource carrier bedsThis is in contrast to the studies from Gallango andParnaud (1995) and Schneider (2003) who suggestedlateral migration at that stage The difference couldbe explained by the presence of a regional top seal intheir models However our model suggests that theshallow rocks do not have sufficient sealing capacityto allow lateral hydrocarbon migration During thepassive-margin phase the basin was normally

pressured (Figure 7A) and simulated stresses havenot yet resulted in any fracturing at that stage

Main Compressional PhasemdashEarly to MiddleMiocene

With the onset of compression during early Miocene(Figure 8B) both the Pirital and Furrial thrusts wereformed and the northern distal part of the passivemargin was rapidly uplifted Important overpressuresas well as mechanical rock failure both the result ofthe rapid uplift are simulated in the allochthonDespite this uplift the kerogen-to-hydrocarbon trans-formation continued and the rate even increasedlocally close to the footwall cutoffs and where syn-orogenic deposition provided rapid overburden load-ing Oil accumulations are modeled in the freshlycreated anticlines

The tectonicndashsedimentary wedge between thetwo main thrusts experienced very rapid burial Infact the Upper Cretaceous source rocks were broughtinto depths where temperatures were as high as185degC (365degF) at 19 Ma and 225degC (437degF) at15 Ma Especially close to the cutoff of the Piritalthrust they very quickly reached a high maturity at19 Ma and reached full transformation in most placesat 12 Ma The expelled hydrocarbons migrated intothe sandstones of the Merecure Formation which weresealed by the Carapita Formation The charge toward aseries of recent anticlines is modeled by the fill-to-spillmechanism from north to south (Figure 9C D) Afterthe first oil fill the northern prospect quickly receiveda local gas charge the central structure containedmainly oil Note the presence of a lateral seal that isdue to sandndashshale juxtaposition The southernmoststructure which forms the Santa Barbara field todaywas charged by oil only at 15 Ma (Figure 9D)

The generated hydrocarbons in the southernautochthon were leaking to the surface until theCarapita Formation was compacted enough to holdsome of them (Figure 8C) At 12 Ma the wedgesource rocks were mostly spent but expulsion wasstill ongoing Important rock failure is modeled tohave occurred in the thrusted footwall breaking theformerly sealing Carapita Formation (Figure 10)This is in good agreement with cemented hydraulicfractures observed in a few cores in the Carapita

NEUMAIER ET AL 1341

Figure 8 Two-dimensional model through geologic time (selected time steps only) with modeled transformation ratio for the sourcerock layers isotherms and liquid and vapor migration and accumulation

1342 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

shales from the Furrial Trend (Roure et al 2005)This important fracturing event resulted in thedestruction of the modeled accumulation (Figure 9E)Hydrocarbon generation and expulsion were alsohappening in the southern foreland basin at transfor-mation ratios of up to 30

Late Compressional and PostcompressionalPhasemdashLate Miocene to Present Day

The late compressional phase was marked by the par-tial destruction of the allochthonous petroleum systemsby erosion (Figure 8E F) Many traps disappeared and

Figure 8 Continued

NEUMAIER ET AL 1343

the hydrocarbon generation within the source rocksstopped The erosion triggered an important sedimenttransfer toward the south contributing to further burialand increased temperatures in the thrust wedge and theforeland basin During late Miocene the petroleumsystem in the wedge matured An important northwardtilting was triggered by flexural tectonic loading andthe loading of syntectonic sediments (Summa et al2003) In combination with the presence of an effec-tive regional top seal the compacted CarapitaFormation allowed a change from dominantly verticalmigration into lateral migration First accumulationsare modeled in the Neogene synorogenic sequence

reservoirs in the south and important southwardmigration toward the foreland basin started(Figure 8G) These trends continued through Plioceneand Pleistocene times to present day filling theOnado field (Figure 8H) and other known fields inthe south (Figure 1A) of the Eastern Venezuela Basin

Prospectivity

We modeled the charge of the Santa Barbara field inthe Furrial Trend and of the Onado field in the fore-land basin (Figure 8H) Other prospects in the thrustwedge were also investigated The structure called

Figure 9 Detail of two-dimensionalmodel through geologic time (seeFigure 8) showing modeled transforma-tion ratio for the source rock layers iso-therms and liquid and vapor migrationand accumulation The Santa Barbara fieldand the prospect are shown by ellipses

1344 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

ldquoprospectrdquo (Figure 9G) looks similar to that of theSanta Barbara field because both are anticlines andthe reservoir and seal formations are the sameHowever they differ in several aspects First theprospect is in the vicinity of the Pirital thrust whichpresents a high risk of impacting the seal integrityThe prospect location is directly below the area wherewe presume the maximum of late tectonic activityoccurred (Figure 4) As modeled the CarapitaFormation the seal of the prospect was already brit-tle when this deformation occurred Calculatedstresses also predict rock failure from 12 Ma onward(Figure 10) Even though charge continued ourmodel suggests continuous leakage of the structure(Figure 9EndashG) The second major difference is thatthe source rock in the fault block of the prospect ismodeled to have already started generating and expel-ling hydrocarbons during the Oligocene and earlyMiocene (Figure 9A) before the trap formed and the

seal became effective (Figures 9C 10) Even thoughgeneration expulsion and migration continued aftertrap formation the prospect was not charged withthe earlier expelled oil In addition the present-daymaturity of the source rocks within the prospectrsquosfault block is higher than in the source rocks withinthe tectonic block of the Santa Barbara field poten-tially indicating expulsion of more gas A southndashnorth trend can be seen on the modeled accumulationcomposition with more gas toward the allochthon(Figure 9) The modeled composition of the prospectprior to seal failure was much richer in gas than inthe neighboring Santa Barbara field (Figure 9D) Infact simulation scenarios with unbreached sealsresulted in accumulation of predominantly gas Thecombination of both seal integrity and charge compo-sition concerns makes this a risky prospect In addi-tion potential diagenetic fluids related to the faultfluid flow might have altered the reservoir quality

Figure 10 Two-dimensional model (at 12 Ma) with modeled stress to failure Mohr circles (with σ standing for normal stress and τfor shear stress) and petroleum system events charts for the Santa Barbara field (left) and the prospect (right) Model predicts rock fail-ure in the prospect from about 12 Ma For further details see text

NEUMAIER ET AL 1345

Apart from the Onado oil field only minor accu-mulations are modeled in the southern Neogene syn-orogenic sequence The proportion of gas is higherthan in the Furrial trend The accumulations at thesouthern model edge are a modeling artifact repre-senting hydrocarbons (mainly oil) that in reality arethought to have charged the southern foreland basin(from mid to late Miocene times to present day)However parts of the migrated oil and gas could havebeen trapped in the Upper Cretaceous San JuanFormation south of the horst structure

The potential for conventional hydrocarbon accu-mulations in the allochthon is very limited by theimportant fracturing that occurred during the upliftperiod The basal sandstones of the Morochito piggy-back basin north of the Pirital thrust are not chargedin our 2D model because of the absence of structuraltraps Nevertheless there is a potential for strati-graphic traps to have been charged as migrationthrough the basin occurred However seal strengthmight be critical at this low-compaction stage Thesame applies for the Las Piedras Formation We thinkthere might be potential for shale oil because theQuerecual Formation is at shallow depths but showsmaturity within the oil window From the thermalmaturity point of view any potential Jurassic sourcerocks as discussed by Summa et al (2003) mightbe a gas source or shale gas target

UNCERTAINTY ANALYSIS

There is often significant uncertainty in basin geom-etry and depth of stratigraphic horizons based onuncertain seismic interpretation and velocity models(Kukla et al 2000 Brandes et al 2008 Baur et al2010) Restoring of present-day geometries to obtainpaleogeometries adds further uncertainty Additionaluncertainty is related to thermal boundary conditionssuch as paleoheat flow (Beha et al 2008) and togeochemical properties such as source rock character-istics and petroleum generation kinetics and calcula-tion of biodegradation and petroleum density(Blumenstein et al 2008) In addition to the porepressure and thermal sensitivity analysis justdescribed we varied initial TOC within a range of2 to 10 A variation of initial TOC obviously

directly scales the amount of initial kerogen availablefor transformation into hydrocarbons Neither thecharge volume nor composition of the Santa Barbarafield nor the wedge prospects are significantly sensi-tive to variation of initial TOC these traps are filledto spill in each scenario However the amount ofhydrocarbons charging the foreland basin (eg theOnado field) is reduced in the case of low initialTOC A high initial TOC only increased the modeledamount of surface seepage The sensitivity of the ini-tial HI is analogous to the sensitivity of the ini-tial TOC

In addition to the more classical uncertaintyand sensitivity analysis described we focused onquantitative analysis of the decompaction uncer-tainty when performing structural restoration Thisuncertainty is particular to the applied workflowof using structurally restored paleogeometries inBPSM Note that this uncertainty is significantlylower when using the backstripping method withthe optimization algorithms discussed One methodto analyze the adequacy of the decompactionmethod is comparison of the output rock matrixmass balance In the backstripping approach therock matrix mass is kept constant of a model cellwith a simulated decrease in porosity controlledby compaction laws and the corresponding reduc-tion of the layer thickness When using fixed pale-ogeometries from structural restoration the cellthickness is imposed However the decrease inporosity is calculated and the mass is adapted tofit the imposed cell thickness That means that therock matrix mass is constant only if ideal decom-paction has been applied

Figure 11A shows the percentage variation fromthe present-day matrix mass (100) of a given sourcerock cell through geologic time The dotted line rep-resents the ideal case with constant rock matrix massWe used two end-member decompaction scenarios todescribe the decompaction-related uncertainty that isintroduced when fixed paleogeometries replace back-stripping idealized hydrostatic decompaction assum-ing no overpressure and idealized lithostaticdecompaction assuming no effective stress (nodecompaction resulting in constant cell thicknessthrough time) In the latter case with decreasingporosity during compaction the rock matrix mass

1346 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

increases until it reaches the present-day mass Inother words rock mass matrix has been highly under-estimated in the past (in this particular case by morethan 20) This is a source rock so the kerogen massdecreases by the same factor This yields an underes-timation of more than 20 in TOC Note the rockmatrix mass is constant for time steps prior to23 Ma in which backstripping optimization couldbe performed toward the earliest paleogeometry Forthe second end-member scenario of idealized hydro-static decompaction the rock matrix mass is nearlyconstant which is a good decompaction adequacy

indicator for that given cell in the model Howeverthe so-measured adequacy changes throughout themodel In some areas the rock matrix mass has beenunderestimated by about 50 for nondecompactedpaleogeometries however in the case of hydrostaticdecompaction values generally range from 80 to120 of the present-day rock matrix mass Thatmeans in this particular case that hydrostatic decom-paction is more adequate (or less wrong) than nodecompaction at all The decompaction adequacycan be roughly correlated with pressure changesthrough geologic time

Figure 11 Time extractions for comparison of two end-member decompaction scenarios (see Figure 6 for location) (A) Rock matrixmass conservation through geologic time (B) Source rock burial depth through geologic time (C) Temperature and transformation ratiothrough geologic time

NEUMAIER ET AL 1347

Inadequate decompaction results in wrong depthfor stratigraphic horizons in the paleogeometriesFigure 11B shows the burial depth of the same inves-tigated source rock cell We can see that during thepassive margin phase there is a slight difference inpaleoburial depth for the two end-member scenariosDuring the compressional phase the differencechanges dramatically reaching more than 1000 m(3281 ft) in some places Toward present day bothcurves meet again because the input to the two resto-ration scenarios was the same (identical present-dayburial depth) This difference in compaction calcula-tion has severe implications on the modeled tempera-ture history and the kerogen-to-hydrocarbontransformation (Figure 11C) For an identical thermalframework in terms of modeling input parameterstemperature differences of up to 25degC (45degF) resultin TR differences of more than 20 in the early com-pressional phase before TR values meet again for thepresent-day situation

CONCLUSIONS

Structurally complex areas are attracting increasinginterest in hydrocarbon exploration Exploration riskcan be substantially narrowed by analyzing the his-tory of the basin and its petroleum systems based onstructural evolution We presented a further progressin technology that allows the integration of structuralrestoration and BPSM This combined analysis isseen as a way forward to improve modeling in thrustbelts with stacked stratigraphic sequences Despitelimitations of nonoptimized decompaction whichare added to the general uncertainty related to theindividual methods we modeled the charge and sealhistory of the Monagas fold and thrust belt inVenezuela Calibrated porosity temperature andthermal maturity as well as basin-scale stress pro-vided the regional framework through geologic timein which the petroleum systems evolved such thatthe charge of the known Santa Barbara and Onadofields and other trends could be understood We dis-cussed seal integrity of a potential exploration pros-pect by analyzing structural restoration-derived faultdisplacement through geologic time and an advancedstress forward simulation The uncertainty of using

decompacted or nondecompacted structural restora-tion has been assessed It can have important influ-ence on rock matrix mass balance through geologictime paleogeometry and paleoburial depth andtherefore affect paleotemperatures and paleomatur-ities We conclude that decompaction assuminghydrostatic pressure is the best option in that particu-lar case

REFERENCES CITED

Alberdi M and E Lafargue 1993 Vertical variations of organicmatter content in Guayuta Group (Upper Cretaceous)interior mountain belt eastern Venezuela OrganicGeochemistry v 20 p 425ndash436 doi 1010160146-6380(93)90091-O

Athy L F 1930 Density porosity and compaction of sedimen-tary rocks AAPG Bulletin v 14 p 1ndash24

Aymard R L Pimentel P Eitz P Lopez A ChaouchJ Navarro J Mijares and J G Pereira 1990 Geologic inte-gration and evaluation of Northern Monagas EasternVenezuelan Basin in J Brooks ed Classic petroleum prov-inces The Geologic Society (London) Special Publicationv 50 p 37ndash53

Baur F M Di Benedetto T Fuchs C Lampe andS Sciamanna 2009 Integrating structural geology andpetroleum systems modelingmdashA pilot project from Boliviarsquosfold and thrust belt Marine and Petroleum Geology v 26no 4 p 573ndash579 doi 101016jmarpetgeo200901004

Baur F R Littke H Wielens C Lampe and T Fuchs 2010Basin modeling meets rift analysismdashA numerical modelingstudy from the Jeanne drsquoArc basin offshore NewfoundlandCanada Marine and Petroleum Geology v 27 p 585ndash599doi 101016jmarpetgeo200906003

Beha A R O Thomsen and R Littke 2008 Thermal historyhydrocarbon generation and migration in the Horn Grabenin the Danish North Sea A 2D basin modelling studyInternational Journal of Earth Sciences v 97 p 1087ndash1100doi 101007s00531-007-0247-2

Blumenstein I O B M Krooss R di Primio W RottkeE Muumlller C Westerlage and R Littke 2008Biodegradation in numerical basin modelling A case studyfrom the Gifhorn Trough N-Germany International Journalof Earth Sciences v 97 p 1115ndash1129 doi 101007s00531-007-0272-1

Brandes C A Astorga R Littke and J Winsemann 2008Basin modelling of the Limoacuten back-arc basin (Costa Rica)Burial history and temperature evolution of an island arc-related basin-system Basin Research v 20 p 119ndash142doi 101111bre200820issue-1

Carnevali J O 1988 El Furrial oil field northeasternVenezuela First giant in foreland fold and thrust belts ofwestern hemisphere AAPG Bulletin v 72 68 p

Chevalier Y 1987 Contribution a lrsquoetude geacuteologique de lafrontiereacute Sud-Est de la plaque Caraibe Les zones internesde la chaine Sud-Caraibe sur le transect ile de Margarita-

1348 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

fault displacing mature rocks upon less mature ones(see Figure 6B for 2D view) There is a mismatchbetween the modeled Ro trend and the data in thedeeper part of the well In the Oligocene rocks themodeled temperature is too low (2degC or 36degF)whereas the modeled Ro is too high (up to 02 Ro)Any attempt to reduce the modeled Ro (eg bychanging thermal conductivity of the MioceneCarapita Formation) also results in a lower modeledpresent-day temperature Therefore the thermal sce-nario used for this calibration is a trade-off betweenmatching present-day temperature and Ro (using thekinetics of Sweeney and Burnham 1990) The basalheat flow assumed in the master scenario (Table 2)results in a modeled present-day surface heat flowranging from 37 mW∕m2 in the southern part of thethrusts to 52 mW∕m2 in parts of the uplifted areaAt the same depth the allochthonous northern partis thermally more mature than the southern autochthon(Figure 6B) because of the recent uplift Locallyinverted trends around the thrusts can be observedfor example at the location of Well Alpha (seeFigure 5D) In the autochthon the Upper Cretaceoussource rocks are entirely in the oil or even gas windowat present day because of the Neogene tectonic loadingand synorogenic deposition In the northern alloch-thon most of the Upper Cretaceous source rocks havebeen in the oil window in the past The modeled ther-mal maturity is in good agreement with the regionalmaturity suggested by Summa et al (2003)

Similar to the pressures the thrusting also affectsthe modeled temperature and Ro In the thrust wedgea direct relationship exists between the thrusting andthe increase of temperature and maturity The thrust-ing is the trigger for source rocks entering the oil win-dow within the thrust wedge (Figure 7B) and thepresent-day thermal maturity has been reached sincethe last 10 Ma Although this recent thermal historycan be considered as relatively well constrained bythe measured Ro in the calibration well this doesnot apply to the past In this thrusted area any ther-mal maturity reached prior to the thrusting onset hasbeen ldquooverprintedrdquo and therefore no thermal calibra-tion is possible for this period However theCretaceous source rocks were not buried deeply dur-ing the passive margin phase and even very hot sce-narios (basal heat flow of 100 mW∕m2) do not

significantly change their maturity (Figure 7B) Inthe uplifted allochthon several wells show low(03) to moderate (1) measured Ro (Parra et al2011) Because the allochthon experienced its maxi-mum burial and the highest temperatures during thepassive margin phase these maturity levels can beassumed as representative of the maturity prior tothe compression In contrast to the thrust wedge theconsequence of the thrusting is inverse for the alloch-thon (Figure 7C) as the hanging wall is uplifted andthe accompanying erosion removes the overburdenThe highest temperatures are simulated to have beenreached just before the thrusting event Then temper-ature decreased and Ro stagnated leaving the sourcerocks in the early mature zone until present day Theassumption of a regionally relatively homogeneousthermal field during the cooling phase of the passivemargin in combination with the relatively low matur-ities measured in the uplifted allochthon makes thehot heat-flow scenario unlikely However a laterallyvarying heat flow cannot be excluded and the conse-quently hotter paleoheat flow would have led to muchhigher temperatures prior to compression resultingalso in a higher present-day maturity level (14)despite the uplift (Figure 7C) Any variations of basalheat flow during the rift period (Late Jurassic to EarlyCretaceous) occurred too early to affect the maturityof the source rocks deposited during the lateCretaceous

Charge History

After calibration of the basinrsquos physical environ-ment the evolution of the petroleum systems wasmodeled Hydrocarbon generation was simulatedthrough geologic time using the activation energydistribution of the assigned source rock kineticsFrom the moment of hydrocarbon expulsiononward secondary migration was modeled usingthe invasion percolation method (Wilkinson 1984)It takes into account the geometry overpressure gra-dient and capillary entry pressures of the rockswhich are opposed to the buoyancy of the migratingoil and gas Figure 8 shows the time evolution of thetotal study area and Figure 9 focuses on the thrustwedge As already mentioned the described chargehistory is based on forward modeling It is consistent

NEUMAIER ET AL 1339

Figure 7 Time extractions (see Figure 6 for location) (A) Pressure development through geologic time for a source rock in the ver-tical continuity of Well Alpha Dotted lines are hydrostatic and lithostatic pressures continuous lines are pore pressure the master sce-nario is in black (a1 faults closing at 105 Ma) and additional scenarios in gray (a2 open faults a3 closed faults a4 faults closing at12 Ma a5 faults closing at 8 Ma) (B) Temperature (continuous lines) and Ro (dotted lines) through geologic time for a source rockin the vertical continuity of Well Alpha the black line shows the master scenario and the gray line the alternative high-heat-flow sce-nario (C) Temperature (continuous lines) and Ro (dotted lines) through geologic time for a source rock in the uplifted allochthonthe black line shows the master scenario and the gray line the alternative high heat flow scenario

1340 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

with the modeled compaction and thermal historyand therefore shows one scenario that explains thecharge of the Santa Barbara field and the forelandbasin Simplifications are inherent to any such work-flow and shall be considered when analyzing themodeling results

Passive Margin PhasemdashMesozoic toOligocene

Modeling results indicate that already during the LateCretaceous any Jurassic source rocks potentiallydeposited in the northern part of the section (Summaet al 2003) have reached temperatures of more than200degC (392degF) and corresponding high levels of matu-rity However the temperature at the base of theQuerecual Formation (Upper Cretaceous) was lessthan 100degC (212degF) leaving the kerogen immatureThe source rock became effective only during lateEocene with temperatures of up to 120degC (248degF)and a (reactive kerogen to hydrocarbon) transforma-tion ratio (TR) of more than 20 in the deeper distalpart of the passive margin Oil was quickly expelledmigrated vertically and seeped to the surface becauseof the general lack of seals in the overburden Duringthe Oligocene the time of deposition of the Merecurereservoir sandstones in the southern platform this sit-uation did not change apart from increased transforma-tion in the deeper basinal part of the passive margin

First oil accumulations are modeled in the Vidontildeosandstones in the basinal part in the late Oligocene(Figure 8A) in which the base of the QuerecualFormation shows TRs of up to 60 at around 130degC(266degF) In the proximal part there is no hydrocarbongeneration Although there is a general northwardtilting slope no southward migration from the distalmargin is modeled this has been checked usingdifferent migration methods including Darcy flowwith scenarios of continuous intrasource carrier bedsThis is in contrast to the studies from Gallango andParnaud (1995) and Schneider (2003) who suggestedlateral migration at that stage The difference couldbe explained by the presence of a regional top seal intheir models However our model suggests that theshallow rocks do not have sufficient sealing capacityto allow lateral hydrocarbon migration During thepassive-margin phase the basin was normally

pressured (Figure 7A) and simulated stresses havenot yet resulted in any fracturing at that stage

Main Compressional PhasemdashEarly to MiddleMiocene

With the onset of compression during early Miocene(Figure 8B) both the Pirital and Furrial thrusts wereformed and the northern distal part of the passivemargin was rapidly uplifted Important overpressuresas well as mechanical rock failure both the result ofthe rapid uplift are simulated in the allochthonDespite this uplift the kerogen-to-hydrocarbon trans-formation continued and the rate even increasedlocally close to the footwall cutoffs and where syn-orogenic deposition provided rapid overburden load-ing Oil accumulations are modeled in the freshlycreated anticlines

The tectonicndashsedimentary wedge between thetwo main thrusts experienced very rapid burial Infact the Upper Cretaceous source rocks were broughtinto depths where temperatures were as high as185degC (365degF) at 19 Ma and 225degC (437degF) at15 Ma Especially close to the cutoff of the Piritalthrust they very quickly reached a high maturity at19 Ma and reached full transformation in most placesat 12 Ma The expelled hydrocarbons migrated intothe sandstones of the Merecure Formation which weresealed by the Carapita Formation The charge toward aseries of recent anticlines is modeled by the fill-to-spillmechanism from north to south (Figure 9C D) Afterthe first oil fill the northern prospect quickly receiveda local gas charge the central structure containedmainly oil Note the presence of a lateral seal that isdue to sandndashshale juxtaposition The southernmoststructure which forms the Santa Barbara field todaywas charged by oil only at 15 Ma (Figure 9D)

The generated hydrocarbons in the southernautochthon were leaking to the surface until theCarapita Formation was compacted enough to holdsome of them (Figure 8C) At 12 Ma the wedgesource rocks were mostly spent but expulsion wasstill ongoing Important rock failure is modeled tohave occurred in the thrusted footwall breaking theformerly sealing Carapita Formation (Figure 10)This is in good agreement with cemented hydraulicfractures observed in a few cores in the Carapita

NEUMAIER ET AL 1341

Figure 8 Two-dimensional model through geologic time (selected time steps only) with modeled transformation ratio for the sourcerock layers isotherms and liquid and vapor migration and accumulation

1342 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

shales from the Furrial Trend (Roure et al 2005)This important fracturing event resulted in thedestruction of the modeled accumulation (Figure 9E)Hydrocarbon generation and expulsion were alsohappening in the southern foreland basin at transfor-mation ratios of up to 30

Late Compressional and PostcompressionalPhasemdashLate Miocene to Present Day

The late compressional phase was marked by the par-tial destruction of the allochthonous petroleum systemsby erosion (Figure 8E F) Many traps disappeared and

Figure 8 Continued

NEUMAIER ET AL 1343

the hydrocarbon generation within the source rocksstopped The erosion triggered an important sedimenttransfer toward the south contributing to further burialand increased temperatures in the thrust wedge and theforeland basin During late Miocene the petroleumsystem in the wedge matured An important northwardtilting was triggered by flexural tectonic loading andthe loading of syntectonic sediments (Summa et al2003) In combination with the presence of an effec-tive regional top seal the compacted CarapitaFormation allowed a change from dominantly verticalmigration into lateral migration First accumulationsare modeled in the Neogene synorogenic sequence

reservoirs in the south and important southwardmigration toward the foreland basin started(Figure 8G) These trends continued through Plioceneand Pleistocene times to present day filling theOnado field (Figure 8H) and other known fields inthe south (Figure 1A) of the Eastern Venezuela Basin

Prospectivity

We modeled the charge of the Santa Barbara field inthe Furrial Trend and of the Onado field in the fore-land basin (Figure 8H) Other prospects in the thrustwedge were also investigated The structure called

Figure 9 Detail of two-dimensionalmodel through geologic time (seeFigure 8) showing modeled transforma-tion ratio for the source rock layers iso-therms and liquid and vapor migrationand accumulation The Santa Barbara fieldand the prospect are shown by ellipses

1344 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

ldquoprospectrdquo (Figure 9G) looks similar to that of theSanta Barbara field because both are anticlines andthe reservoir and seal formations are the sameHowever they differ in several aspects First theprospect is in the vicinity of the Pirital thrust whichpresents a high risk of impacting the seal integrityThe prospect location is directly below the area wherewe presume the maximum of late tectonic activityoccurred (Figure 4) As modeled the CarapitaFormation the seal of the prospect was already brit-tle when this deformation occurred Calculatedstresses also predict rock failure from 12 Ma onward(Figure 10) Even though charge continued ourmodel suggests continuous leakage of the structure(Figure 9EndashG) The second major difference is thatthe source rock in the fault block of the prospect ismodeled to have already started generating and expel-ling hydrocarbons during the Oligocene and earlyMiocene (Figure 9A) before the trap formed and the

seal became effective (Figures 9C 10) Even thoughgeneration expulsion and migration continued aftertrap formation the prospect was not charged withthe earlier expelled oil In addition the present-daymaturity of the source rocks within the prospectrsquosfault block is higher than in the source rocks withinthe tectonic block of the Santa Barbara field poten-tially indicating expulsion of more gas A southndashnorth trend can be seen on the modeled accumulationcomposition with more gas toward the allochthon(Figure 9) The modeled composition of the prospectprior to seal failure was much richer in gas than inthe neighboring Santa Barbara field (Figure 9D) Infact simulation scenarios with unbreached sealsresulted in accumulation of predominantly gas Thecombination of both seal integrity and charge compo-sition concerns makes this a risky prospect In addi-tion potential diagenetic fluids related to the faultfluid flow might have altered the reservoir quality

Figure 10 Two-dimensional model (at 12 Ma) with modeled stress to failure Mohr circles (with σ standing for normal stress and τfor shear stress) and petroleum system events charts for the Santa Barbara field (left) and the prospect (right) Model predicts rock fail-ure in the prospect from about 12 Ma For further details see text

NEUMAIER ET AL 1345

Apart from the Onado oil field only minor accu-mulations are modeled in the southern Neogene syn-orogenic sequence The proportion of gas is higherthan in the Furrial trend The accumulations at thesouthern model edge are a modeling artifact repre-senting hydrocarbons (mainly oil) that in reality arethought to have charged the southern foreland basin(from mid to late Miocene times to present day)However parts of the migrated oil and gas could havebeen trapped in the Upper Cretaceous San JuanFormation south of the horst structure

The potential for conventional hydrocarbon accu-mulations in the allochthon is very limited by theimportant fracturing that occurred during the upliftperiod The basal sandstones of the Morochito piggy-back basin north of the Pirital thrust are not chargedin our 2D model because of the absence of structuraltraps Nevertheless there is a potential for strati-graphic traps to have been charged as migrationthrough the basin occurred However seal strengthmight be critical at this low-compaction stage Thesame applies for the Las Piedras Formation We thinkthere might be potential for shale oil because theQuerecual Formation is at shallow depths but showsmaturity within the oil window From the thermalmaturity point of view any potential Jurassic sourcerocks as discussed by Summa et al (2003) mightbe a gas source or shale gas target

UNCERTAINTY ANALYSIS

There is often significant uncertainty in basin geom-etry and depth of stratigraphic horizons based onuncertain seismic interpretation and velocity models(Kukla et al 2000 Brandes et al 2008 Baur et al2010) Restoring of present-day geometries to obtainpaleogeometries adds further uncertainty Additionaluncertainty is related to thermal boundary conditionssuch as paleoheat flow (Beha et al 2008) and togeochemical properties such as source rock character-istics and petroleum generation kinetics and calcula-tion of biodegradation and petroleum density(Blumenstein et al 2008) In addition to the porepressure and thermal sensitivity analysis justdescribed we varied initial TOC within a range of2 to 10 A variation of initial TOC obviously

directly scales the amount of initial kerogen availablefor transformation into hydrocarbons Neither thecharge volume nor composition of the Santa Barbarafield nor the wedge prospects are significantly sensi-tive to variation of initial TOC these traps are filledto spill in each scenario However the amount ofhydrocarbons charging the foreland basin (eg theOnado field) is reduced in the case of low initialTOC A high initial TOC only increased the modeledamount of surface seepage The sensitivity of the ini-tial HI is analogous to the sensitivity of the ini-tial TOC

In addition to the more classical uncertaintyand sensitivity analysis described we focused onquantitative analysis of the decompaction uncer-tainty when performing structural restoration Thisuncertainty is particular to the applied workflowof using structurally restored paleogeometries inBPSM Note that this uncertainty is significantlylower when using the backstripping method withthe optimization algorithms discussed One methodto analyze the adequacy of the decompactionmethod is comparison of the output rock matrixmass balance In the backstripping approach therock matrix mass is kept constant of a model cellwith a simulated decrease in porosity controlledby compaction laws and the corresponding reduc-tion of the layer thickness When using fixed pale-ogeometries from structural restoration the cellthickness is imposed However the decrease inporosity is calculated and the mass is adapted tofit the imposed cell thickness That means that therock matrix mass is constant only if ideal decom-paction has been applied

Figure 11A shows the percentage variation fromthe present-day matrix mass (100) of a given sourcerock cell through geologic time The dotted line rep-resents the ideal case with constant rock matrix massWe used two end-member decompaction scenarios todescribe the decompaction-related uncertainty that isintroduced when fixed paleogeometries replace back-stripping idealized hydrostatic decompaction assum-ing no overpressure and idealized lithostaticdecompaction assuming no effective stress (nodecompaction resulting in constant cell thicknessthrough time) In the latter case with decreasingporosity during compaction the rock matrix mass

1346 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

increases until it reaches the present-day mass Inother words rock mass matrix has been highly under-estimated in the past (in this particular case by morethan 20) This is a source rock so the kerogen massdecreases by the same factor This yields an underes-timation of more than 20 in TOC Note the rockmatrix mass is constant for time steps prior to23 Ma in which backstripping optimization couldbe performed toward the earliest paleogeometry Forthe second end-member scenario of idealized hydro-static decompaction the rock matrix mass is nearlyconstant which is a good decompaction adequacy

indicator for that given cell in the model Howeverthe so-measured adequacy changes throughout themodel In some areas the rock matrix mass has beenunderestimated by about 50 for nondecompactedpaleogeometries however in the case of hydrostaticdecompaction values generally range from 80 to120 of the present-day rock matrix mass Thatmeans in this particular case that hydrostatic decom-paction is more adequate (or less wrong) than nodecompaction at all The decompaction adequacycan be roughly correlated with pressure changesthrough geologic time

Figure 11 Time extractions for comparison of two end-member decompaction scenarios (see Figure 6 for location) (A) Rock matrixmass conservation through geologic time (B) Source rock burial depth through geologic time (C) Temperature and transformation ratiothrough geologic time

NEUMAIER ET AL 1347

Inadequate decompaction results in wrong depthfor stratigraphic horizons in the paleogeometriesFigure 11B shows the burial depth of the same inves-tigated source rock cell We can see that during thepassive margin phase there is a slight difference inpaleoburial depth for the two end-member scenariosDuring the compressional phase the differencechanges dramatically reaching more than 1000 m(3281 ft) in some places Toward present day bothcurves meet again because the input to the two resto-ration scenarios was the same (identical present-dayburial depth) This difference in compaction calcula-tion has severe implications on the modeled tempera-ture history and the kerogen-to-hydrocarbontransformation (Figure 11C) For an identical thermalframework in terms of modeling input parameterstemperature differences of up to 25degC (45degF) resultin TR differences of more than 20 in the early com-pressional phase before TR values meet again for thepresent-day situation

CONCLUSIONS

Structurally complex areas are attracting increasinginterest in hydrocarbon exploration Exploration riskcan be substantially narrowed by analyzing the his-tory of the basin and its petroleum systems based onstructural evolution We presented a further progressin technology that allows the integration of structuralrestoration and BPSM This combined analysis isseen as a way forward to improve modeling in thrustbelts with stacked stratigraphic sequences Despitelimitations of nonoptimized decompaction whichare added to the general uncertainty related to theindividual methods we modeled the charge and sealhistory of the Monagas fold and thrust belt inVenezuela Calibrated porosity temperature andthermal maturity as well as basin-scale stress pro-vided the regional framework through geologic timein which the petroleum systems evolved such thatthe charge of the known Santa Barbara and Onadofields and other trends could be understood We dis-cussed seal integrity of a potential exploration pros-pect by analyzing structural restoration-derived faultdisplacement through geologic time and an advancedstress forward simulation The uncertainty of using

decompacted or nondecompacted structural restora-tion has been assessed It can have important influ-ence on rock matrix mass balance through geologictime paleogeometry and paleoburial depth andtherefore affect paleotemperatures and paleomatur-ities We conclude that decompaction assuminghydrostatic pressure is the best option in that particu-lar case

REFERENCES CITED

Alberdi M and E Lafargue 1993 Vertical variations of organicmatter content in Guayuta Group (Upper Cretaceous)interior mountain belt eastern Venezuela OrganicGeochemistry v 20 p 425ndash436 doi 1010160146-6380(93)90091-O

Athy L F 1930 Density porosity and compaction of sedimen-tary rocks AAPG Bulletin v 14 p 1ndash24

Aymard R L Pimentel P Eitz P Lopez A ChaouchJ Navarro J Mijares and J G Pereira 1990 Geologic inte-gration and evaluation of Northern Monagas EasternVenezuelan Basin in J Brooks ed Classic petroleum prov-inces The Geologic Society (London) Special Publicationv 50 p 37ndash53

Baur F M Di Benedetto T Fuchs C Lampe andS Sciamanna 2009 Integrating structural geology andpetroleum systems modelingmdashA pilot project from Boliviarsquosfold and thrust belt Marine and Petroleum Geology v 26no 4 p 573ndash579 doi 101016jmarpetgeo200901004

Baur F R Littke H Wielens C Lampe and T Fuchs 2010Basin modeling meets rift analysismdashA numerical modelingstudy from the Jeanne drsquoArc basin offshore NewfoundlandCanada Marine and Petroleum Geology v 27 p 585ndash599doi 101016jmarpetgeo200906003

Beha A R O Thomsen and R Littke 2008 Thermal historyhydrocarbon generation and migration in the Horn Grabenin the Danish North Sea A 2D basin modelling studyInternational Journal of Earth Sciences v 97 p 1087ndash1100doi 101007s00531-007-0247-2

Blumenstein I O B M Krooss R di Primio W RottkeE Muumlller C Westerlage and R Littke 2008Biodegradation in numerical basin modelling A case studyfrom the Gifhorn Trough N-Germany International Journalof Earth Sciences v 97 p 1115ndash1129 doi 101007s00531-007-0272-1

Brandes C A Astorga R Littke and J Winsemann 2008Basin modelling of the Limoacuten back-arc basin (Costa Rica)Burial history and temperature evolution of an island arc-related basin-system Basin Research v 20 p 119ndash142doi 101111bre200820issue-1

Carnevali J O 1988 El Furrial oil field northeasternVenezuela First giant in foreland fold and thrust belts ofwestern hemisphere AAPG Bulletin v 72 68 p

Chevalier Y 1987 Contribution a lrsquoetude geacuteologique de lafrontiereacute Sud-Est de la plaque Caraibe Les zones internesde la chaine Sud-Caraibe sur le transect ile de Margarita-

1348 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

Figure 7 Time extractions (see Figure 6 for location) (A) Pressure development through geologic time for a source rock in the ver-tical continuity of Well Alpha Dotted lines are hydrostatic and lithostatic pressures continuous lines are pore pressure the master sce-nario is in black (a1 faults closing at 105 Ma) and additional scenarios in gray (a2 open faults a3 closed faults a4 faults closing at12 Ma a5 faults closing at 8 Ma) (B) Temperature (continuous lines) and Ro (dotted lines) through geologic time for a source rockin the vertical continuity of Well Alpha the black line shows the master scenario and the gray line the alternative high-heat-flow sce-nario (C) Temperature (continuous lines) and Ro (dotted lines) through geologic time for a source rock in the uplifted allochthonthe black line shows the master scenario and the gray line the alternative high heat flow scenario

1340 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

with the modeled compaction and thermal historyand therefore shows one scenario that explains thecharge of the Santa Barbara field and the forelandbasin Simplifications are inherent to any such work-flow and shall be considered when analyzing themodeling results

Passive Margin PhasemdashMesozoic toOligocene

Modeling results indicate that already during the LateCretaceous any Jurassic source rocks potentiallydeposited in the northern part of the section (Summaet al 2003) have reached temperatures of more than200degC (392degF) and corresponding high levels of matu-rity However the temperature at the base of theQuerecual Formation (Upper Cretaceous) was lessthan 100degC (212degF) leaving the kerogen immatureThe source rock became effective only during lateEocene with temperatures of up to 120degC (248degF)and a (reactive kerogen to hydrocarbon) transforma-tion ratio (TR) of more than 20 in the deeper distalpart of the passive margin Oil was quickly expelledmigrated vertically and seeped to the surface becauseof the general lack of seals in the overburden Duringthe Oligocene the time of deposition of the Merecurereservoir sandstones in the southern platform this sit-uation did not change apart from increased transforma-tion in the deeper basinal part of the passive margin

First oil accumulations are modeled in the Vidontildeosandstones in the basinal part in the late Oligocene(Figure 8A) in which the base of the QuerecualFormation shows TRs of up to 60 at around 130degC(266degF) In the proximal part there is no hydrocarbongeneration Although there is a general northwardtilting slope no southward migration from the distalmargin is modeled this has been checked usingdifferent migration methods including Darcy flowwith scenarios of continuous intrasource carrier bedsThis is in contrast to the studies from Gallango andParnaud (1995) and Schneider (2003) who suggestedlateral migration at that stage The difference couldbe explained by the presence of a regional top seal intheir models However our model suggests that theshallow rocks do not have sufficient sealing capacityto allow lateral hydrocarbon migration During thepassive-margin phase the basin was normally

pressured (Figure 7A) and simulated stresses havenot yet resulted in any fracturing at that stage

Main Compressional PhasemdashEarly to MiddleMiocene

With the onset of compression during early Miocene(Figure 8B) both the Pirital and Furrial thrusts wereformed and the northern distal part of the passivemargin was rapidly uplifted Important overpressuresas well as mechanical rock failure both the result ofthe rapid uplift are simulated in the allochthonDespite this uplift the kerogen-to-hydrocarbon trans-formation continued and the rate even increasedlocally close to the footwall cutoffs and where syn-orogenic deposition provided rapid overburden load-ing Oil accumulations are modeled in the freshlycreated anticlines

The tectonicndashsedimentary wedge between thetwo main thrusts experienced very rapid burial Infact the Upper Cretaceous source rocks were broughtinto depths where temperatures were as high as185degC (365degF) at 19 Ma and 225degC (437degF) at15 Ma Especially close to the cutoff of the Piritalthrust they very quickly reached a high maturity at19 Ma and reached full transformation in most placesat 12 Ma The expelled hydrocarbons migrated intothe sandstones of the Merecure Formation which weresealed by the Carapita Formation The charge toward aseries of recent anticlines is modeled by the fill-to-spillmechanism from north to south (Figure 9C D) Afterthe first oil fill the northern prospect quickly receiveda local gas charge the central structure containedmainly oil Note the presence of a lateral seal that isdue to sandndashshale juxtaposition The southernmoststructure which forms the Santa Barbara field todaywas charged by oil only at 15 Ma (Figure 9D)

The generated hydrocarbons in the southernautochthon were leaking to the surface until theCarapita Formation was compacted enough to holdsome of them (Figure 8C) At 12 Ma the wedgesource rocks were mostly spent but expulsion wasstill ongoing Important rock failure is modeled tohave occurred in the thrusted footwall breaking theformerly sealing Carapita Formation (Figure 10)This is in good agreement with cemented hydraulicfractures observed in a few cores in the Carapita

NEUMAIER ET AL 1341

Figure 8 Two-dimensional model through geologic time (selected time steps only) with modeled transformation ratio for the sourcerock layers isotherms and liquid and vapor migration and accumulation

1342 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

shales from the Furrial Trend (Roure et al 2005)This important fracturing event resulted in thedestruction of the modeled accumulation (Figure 9E)Hydrocarbon generation and expulsion were alsohappening in the southern foreland basin at transfor-mation ratios of up to 30

Late Compressional and PostcompressionalPhasemdashLate Miocene to Present Day

The late compressional phase was marked by the par-tial destruction of the allochthonous petroleum systemsby erosion (Figure 8E F) Many traps disappeared and

Figure 8 Continued

NEUMAIER ET AL 1343

the hydrocarbon generation within the source rocksstopped The erosion triggered an important sedimenttransfer toward the south contributing to further burialand increased temperatures in the thrust wedge and theforeland basin During late Miocene the petroleumsystem in the wedge matured An important northwardtilting was triggered by flexural tectonic loading andthe loading of syntectonic sediments (Summa et al2003) In combination with the presence of an effec-tive regional top seal the compacted CarapitaFormation allowed a change from dominantly verticalmigration into lateral migration First accumulationsare modeled in the Neogene synorogenic sequence

reservoirs in the south and important southwardmigration toward the foreland basin started(Figure 8G) These trends continued through Plioceneand Pleistocene times to present day filling theOnado field (Figure 8H) and other known fields inthe south (Figure 1A) of the Eastern Venezuela Basin

Prospectivity

We modeled the charge of the Santa Barbara field inthe Furrial Trend and of the Onado field in the fore-land basin (Figure 8H) Other prospects in the thrustwedge were also investigated The structure called

Figure 9 Detail of two-dimensionalmodel through geologic time (seeFigure 8) showing modeled transforma-tion ratio for the source rock layers iso-therms and liquid and vapor migrationand accumulation The Santa Barbara fieldand the prospect are shown by ellipses

1344 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

ldquoprospectrdquo (Figure 9G) looks similar to that of theSanta Barbara field because both are anticlines andthe reservoir and seal formations are the sameHowever they differ in several aspects First theprospect is in the vicinity of the Pirital thrust whichpresents a high risk of impacting the seal integrityThe prospect location is directly below the area wherewe presume the maximum of late tectonic activityoccurred (Figure 4) As modeled the CarapitaFormation the seal of the prospect was already brit-tle when this deformation occurred Calculatedstresses also predict rock failure from 12 Ma onward(Figure 10) Even though charge continued ourmodel suggests continuous leakage of the structure(Figure 9EndashG) The second major difference is thatthe source rock in the fault block of the prospect ismodeled to have already started generating and expel-ling hydrocarbons during the Oligocene and earlyMiocene (Figure 9A) before the trap formed and the

seal became effective (Figures 9C 10) Even thoughgeneration expulsion and migration continued aftertrap formation the prospect was not charged withthe earlier expelled oil In addition the present-daymaturity of the source rocks within the prospectrsquosfault block is higher than in the source rocks withinthe tectonic block of the Santa Barbara field poten-tially indicating expulsion of more gas A southndashnorth trend can be seen on the modeled accumulationcomposition with more gas toward the allochthon(Figure 9) The modeled composition of the prospectprior to seal failure was much richer in gas than inthe neighboring Santa Barbara field (Figure 9D) Infact simulation scenarios with unbreached sealsresulted in accumulation of predominantly gas Thecombination of both seal integrity and charge compo-sition concerns makes this a risky prospect In addi-tion potential diagenetic fluids related to the faultfluid flow might have altered the reservoir quality

Figure 10 Two-dimensional model (at 12 Ma) with modeled stress to failure Mohr circles (with σ standing for normal stress and τfor shear stress) and petroleum system events charts for the Santa Barbara field (left) and the prospect (right) Model predicts rock fail-ure in the prospect from about 12 Ma For further details see text

NEUMAIER ET AL 1345

Apart from the Onado oil field only minor accu-mulations are modeled in the southern Neogene syn-orogenic sequence The proportion of gas is higherthan in the Furrial trend The accumulations at thesouthern model edge are a modeling artifact repre-senting hydrocarbons (mainly oil) that in reality arethought to have charged the southern foreland basin(from mid to late Miocene times to present day)However parts of the migrated oil and gas could havebeen trapped in the Upper Cretaceous San JuanFormation south of the horst structure

The potential for conventional hydrocarbon accu-mulations in the allochthon is very limited by theimportant fracturing that occurred during the upliftperiod The basal sandstones of the Morochito piggy-back basin north of the Pirital thrust are not chargedin our 2D model because of the absence of structuraltraps Nevertheless there is a potential for strati-graphic traps to have been charged as migrationthrough the basin occurred However seal strengthmight be critical at this low-compaction stage Thesame applies for the Las Piedras Formation We thinkthere might be potential for shale oil because theQuerecual Formation is at shallow depths but showsmaturity within the oil window From the thermalmaturity point of view any potential Jurassic sourcerocks as discussed by Summa et al (2003) mightbe a gas source or shale gas target

UNCERTAINTY ANALYSIS

There is often significant uncertainty in basin geom-etry and depth of stratigraphic horizons based onuncertain seismic interpretation and velocity models(Kukla et al 2000 Brandes et al 2008 Baur et al2010) Restoring of present-day geometries to obtainpaleogeometries adds further uncertainty Additionaluncertainty is related to thermal boundary conditionssuch as paleoheat flow (Beha et al 2008) and togeochemical properties such as source rock character-istics and petroleum generation kinetics and calcula-tion of biodegradation and petroleum density(Blumenstein et al 2008) In addition to the porepressure and thermal sensitivity analysis justdescribed we varied initial TOC within a range of2 to 10 A variation of initial TOC obviously

directly scales the amount of initial kerogen availablefor transformation into hydrocarbons Neither thecharge volume nor composition of the Santa Barbarafield nor the wedge prospects are significantly sensi-tive to variation of initial TOC these traps are filledto spill in each scenario However the amount ofhydrocarbons charging the foreland basin (eg theOnado field) is reduced in the case of low initialTOC A high initial TOC only increased the modeledamount of surface seepage The sensitivity of the ini-tial HI is analogous to the sensitivity of the ini-tial TOC

In addition to the more classical uncertaintyand sensitivity analysis described we focused onquantitative analysis of the decompaction uncer-tainty when performing structural restoration Thisuncertainty is particular to the applied workflowof using structurally restored paleogeometries inBPSM Note that this uncertainty is significantlylower when using the backstripping method withthe optimization algorithms discussed One methodto analyze the adequacy of the decompactionmethod is comparison of the output rock matrixmass balance In the backstripping approach therock matrix mass is kept constant of a model cellwith a simulated decrease in porosity controlledby compaction laws and the corresponding reduc-tion of the layer thickness When using fixed pale-ogeometries from structural restoration the cellthickness is imposed However the decrease inporosity is calculated and the mass is adapted tofit the imposed cell thickness That means that therock matrix mass is constant only if ideal decom-paction has been applied

Figure 11A shows the percentage variation fromthe present-day matrix mass (100) of a given sourcerock cell through geologic time The dotted line rep-resents the ideal case with constant rock matrix massWe used two end-member decompaction scenarios todescribe the decompaction-related uncertainty that isintroduced when fixed paleogeometries replace back-stripping idealized hydrostatic decompaction assum-ing no overpressure and idealized lithostaticdecompaction assuming no effective stress (nodecompaction resulting in constant cell thicknessthrough time) In the latter case with decreasingporosity during compaction the rock matrix mass

1346 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

increases until it reaches the present-day mass Inother words rock mass matrix has been highly under-estimated in the past (in this particular case by morethan 20) This is a source rock so the kerogen massdecreases by the same factor This yields an underes-timation of more than 20 in TOC Note the rockmatrix mass is constant for time steps prior to23 Ma in which backstripping optimization couldbe performed toward the earliest paleogeometry Forthe second end-member scenario of idealized hydro-static decompaction the rock matrix mass is nearlyconstant which is a good decompaction adequacy

indicator for that given cell in the model Howeverthe so-measured adequacy changes throughout themodel In some areas the rock matrix mass has beenunderestimated by about 50 for nondecompactedpaleogeometries however in the case of hydrostaticdecompaction values generally range from 80 to120 of the present-day rock matrix mass Thatmeans in this particular case that hydrostatic decom-paction is more adequate (or less wrong) than nodecompaction at all The decompaction adequacycan be roughly correlated with pressure changesthrough geologic time

Figure 11 Time extractions for comparison of two end-member decompaction scenarios (see Figure 6 for location) (A) Rock matrixmass conservation through geologic time (B) Source rock burial depth through geologic time (C) Temperature and transformation ratiothrough geologic time

NEUMAIER ET AL 1347

Inadequate decompaction results in wrong depthfor stratigraphic horizons in the paleogeometriesFigure 11B shows the burial depth of the same inves-tigated source rock cell We can see that during thepassive margin phase there is a slight difference inpaleoburial depth for the two end-member scenariosDuring the compressional phase the differencechanges dramatically reaching more than 1000 m(3281 ft) in some places Toward present day bothcurves meet again because the input to the two resto-ration scenarios was the same (identical present-dayburial depth) This difference in compaction calcula-tion has severe implications on the modeled tempera-ture history and the kerogen-to-hydrocarbontransformation (Figure 11C) For an identical thermalframework in terms of modeling input parameterstemperature differences of up to 25degC (45degF) resultin TR differences of more than 20 in the early com-pressional phase before TR values meet again for thepresent-day situation

CONCLUSIONS

Structurally complex areas are attracting increasinginterest in hydrocarbon exploration Exploration riskcan be substantially narrowed by analyzing the his-tory of the basin and its petroleum systems based onstructural evolution We presented a further progressin technology that allows the integration of structuralrestoration and BPSM This combined analysis isseen as a way forward to improve modeling in thrustbelts with stacked stratigraphic sequences Despitelimitations of nonoptimized decompaction whichare added to the general uncertainty related to theindividual methods we modeled the charge and sealhistory of the Monagas fold and thrust belt inVenezuela Calibrated porosity temperature andthermal maturity as well as basin-scale stress pro-vided the regional framework through geologic timein which the petroleum systems evolved such thatthe charge of the known Santa Barbara and Onadofields and other trends could be understood We dis-cussed seal integrity of a potential exploration pros-pect by analyzing structural restoration-derived faultdisplacement through geologic time and an advancedstress forward simulation The uncertainty of using

decompacted or nondecompacted structural restora-tion has been assessed It can have important influ-ence on rock matrix mass balance through geologictime paleogeometry and paleoburial depth andtherefore affect paleotemperatures and paleomatur-ities We conclude that decompaction assuminghydrostatic pressure is the best option in that particu-lar case

REFERENCES CITED

Alberdi M and E Lafargue 1993 Vertical variations of organicmatter content in Guayuta Group (Upper Cretaceous)interior mountain belt eastern Venezuela OrganicGeochemistry v 20 p 425ndash436 doi 1010160146-6380(93)90091-O

Athy L F 1930 Density porosity and compaction of sedimen-tary rocks AAPG Bulletin v 14 p 1ndash24

Aymard R L Pimentel P Eitz P Lopez A ChaouchJ Navarro J Mijares and J G Pereira 1990 Geologic inte-gration and evaluation of Northern Monagas EasternVenezuelan Basin in J Brooks ed Classic petroleum prov-inces The Geologic Society (London) Special Publicationv 50 p 37ndash53

Baur F M Di Benedetto T Fuchs C Lampe andS Sciamanna 2009 Integrating structural geology andpetroleum systems modelingmdashA pilot project from Boliviarsquosfold and thrust belt Marine and Petroleum Geology v 26no 4 p 573ndash579 doi 101016jmarpetgeo200901004

Baur F R Littke H Wielens C Lampe and T Fuchs 2010Basin modeling meets rift analysismdashA numerical modelingstudy from the Jeanne drsquoArc basin offshore NewfoundlandCanada Marine and Petroleum Geology v 27 p 585ndash599doi 101016jmarpetgeo200906003

Beha A R O Thomsen and R Littke 2008 Thermal historyhydrocarbon generation and migration in the Horn Grabenin the Danish North Sea A 2D basin modelling studyInternational Journal of Earth Sciences v 97 p 1087ndash1100doi 101007s00531-007-0247-2

Blumenstein I O B M Krooss R di Primio W RottkeE Muumlller C Westerlage and R Littke 2008Biodegradation in numerical basin modelling A case studyfrom the Gifhorn Trough N-Germany International Journalof Earth Sciences v 97 p 1115ndash1129 doi 101007s00531-007-0272-1

Brandes C A Astorga R Littke and J Winsemann 2008Basin modelling of the Limoacuten back-arc basin (Costa Rica)Burial history and temperature evolution of an island arc-related basin-system Basin Research v 20 p 119ndash142doi 101111bre200820issue-1

Carnevali J O 1988 El Furrial oil field northeasternVenezuela First giant in foreland fold and thrust belts ofwestern hemisphere AAPG Bulletin v 72 68 p

Chevalier Y 1987 Contribution a lrsquoetude geacuteologique de lafrontiereacute Sud-Est de la plaque Caraibe Les zones internesde la chaine Sud-Caraibe sur le transect ile de Margarita-

1348 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

with the modeled compaction and thermal historyand therefore shows one scenario that explains thecharge of the Santa Barbara field and the forelandbasin Simplifications are inherent to any such work-flow and shall be considered when analyzing themodeling results

Passive Margin PhasemdashMesozoic toOligocene

Modeling results indicate that already during the LateCretaceous any Jurassic source rocks potentiallydeposited in the northern part of the section (Summaet al 2003) have reached temperatures of more than200degC (392degF) and corresponding high levels of matu-rity However the temperature at the base of theQuerecual Formation (Upper Cretaceous) was lessthan 100degC (212degF) leaving the kerogen immatureThe source rock became effective only during lateEocene with temperatures of up to 120degC (248degF)and a (reactive kerogen to hydrocarbon) transforma-tion ratio (TR) of more than 20 in the deeper distalpart of the passive margin Oil was quickly expelledmigrated vertically and seeped to the surface becauseof the general lack of seals in the overburden Duringthe Oligocene the time of deposition of the Merecurereservoir sandstones in the southern platform this sit-uation did not change apart from increased transforma-tion in the deeper basinal part of the passive margin

First oil accumulations are modeled in the Vidontildeosandstones in the basinal part in the late Oligocene(Figure 8A) in which the base of the QuerecualFormation shows TRs of up to 60 at around 130degC(266degF) In the proximal part there is no hydrocarbongeneration Although there is a general northwardtilting slope no southward migration from the distalmargin is modeled this has been checked usingdifferent migration methods including Darcy flowwith scenarios of continuous intrasource carrier bedsThis is in contrast to the studies from Gallango andParnaud (1995) and Schneider (2003) who suggestedlateral migration at that stage The difference couldbe explained by the presence of a regional top seal intheir models However our model suggests that theshallow rocks do not have sufficient sealing capacityto allow lateral hydrocarbon migration During thepassive-margin phase the basin was normally

pressured (Figure 7A) and simulated stresses havenot yet resulted in any fracturing at that stage

Main Compressional PhasemdashEarly to MiddleMiocene

With the onset of compression during early Miocene(Figure 8B) both the Pirital and Furrial thrusts wereformed and the northern distal part of the passivemargin was rapidly uplifted Important overpressuresas well as mechanical rock failure both the result ofthe rapid uplift are simulated in the allochthonDespite this uplift the kerogen-to-hydrocarbon trans-formation continued and the rate even increasedlocally close to the footwall cutoffs and where syn-orogenic deposition provided rapid overburden load-ing Oil accumulations are modeled in the freshlycreated anticlines

The tectonicndashsedimentary wedge between thetwo main thrusts experienced very rapid burial Infact the Upper Cretaceous source rocks were broughtinto depths where temperatures were as high as185degC (365degF) at 19 Ma and 225degC (437degF) at15 Ma Especially close to the cutoff of the Piritalthrust they very quickly reached a high maturity at19 Ma and reached full transformation in most placesat 12 Ma The expelled hydrocarbons migrated intothe sandstones of the Merecure Formation which weresealed by the Carapita Formation The charge toward aseries of recent anticlines is modeled by the fill-to-spillmechanism from north to south (Figure 9C D) Afterthe first oil fill the northern prospect quickly receiveda local gas charge the central structure containedmainly oil Note the presence of a lateral seal that isdue to sandndashshale juxtaposition The southernmoststructure which forms the Santa Barbara field todaywas charged by oil only at 15 Ma (Figure 9D)

The generated hydrocarbons in the southernautochthon were leaking to the surface until theCarapita Formation was compacted enough to holdsome of them (Figure 8C) At 12 Ma the wedgesource rocks were mostly spent but expulsion wasstill ongoing Important rock failure is modeled tohave occurred in the thrusted footwall breaking theformerly sealing Carapita Formation (Figure 10)This is in good agreement with cemented hydraulicfractures observed in a few cores in the Carapita

NEUMAIER ET AL 1341

Figure 8 Two-dimensional model through geologic time (selected time steps only) with modeled transformation ratio for the sourcerock layers isotherms and liquid and vapor migration and accumulation

1342 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

shales from the Furrial Trend (Roure et al 2005)This important fracturing event resulted in thedestruction of the modeled accumulation (Figure 9E)Hydrocarbon generation and expulsion were alsohappening in the southern foreland basin at transfor-mation ratios of up to 30

Late Compressional and PostcompressionalPhasemdashLate Miocene to Present Day

The late compressional phase was marked by the par-tial destruction of the allochthonous petroleum systemsby erosion (Figure 8E F) Many traps disappeared and

Figure 8 Continued

NEUMAIER ET AL 1343

the hydrocarbon generation within the source rocksstopped The erosion triggered an important sedimenttransfer toward the south contributing to further burialand increased temperatures in the thrust wedge and theforeland basin During late Miocene the petroleumsystem in the wedge matured An important northwardtilting was triggered by flexural tectonic loading andthe loading of syntectonic sediments (Summa et al2003) In combination with the presence of an effec-tive regional top seal the compacted CarapitaFormation allowed a change from dominantly verticalmigration into lateral migration First accumulationsare modeled in the Neogene synorogenic sequence

reservoirs in the south and important southwardmigration toward the foreland basin started(Figure 8G) These trends continued through Plioceneand Pleistocene times to present day filling theOnado field (Figure 8H) and other known fields inthe south (Figure 1A) of the Eastern Venezuela Basin

Prospectivity

We modeled the charge of the Santa Barbara field inthe Furrial Trend and of the Onado field in the fore-land basin (Figure 8H) Other prospects in the thrustwedge were also investigated The structure called

Figure 9 Detail of two-dimensionalmodel through geologic time (seeFigure 8) showing modeled transforma-tion ratio for the source rock layers iso-therms and liquid and vapor migrationand accumulation The Santa Barbara fieldand the prospect are shown by ellipses

1344 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

ldquoprospectrdquo (Figure 9G) looks similar to that of theSanta Barbara field because both are anticlines andthe reservoir and seal formations are the sameHowever they differ in several aspects First theprospect is in the vicinity of the Pirital thrust whichpresents a high risk of impacting the seal integrityThe prospect location is directly below the area wherewe presume the maximum of late tectonic activityoccurred (Figure 4) As modeled the CarapitaFormation the seal of the prospect was already brit-tle when this deformation occurred Calculatedstresses also predict rock failure from 12 Ma onward(Figure 10) Even though charge continued ourmodel suggests continuous leakage of the structure(Figure 9EndashG) The second major difference is thatthe source rock in the fault block of the prospect ismodeled to have already started generating and expel-ling hydrocarbons during the Oligocene and earlyMiocene (Figure 9A) before the trap formed and the

seal became effective (Figures 9C 10) Even thoughgeneration expulsion and migration continued aftertrap formation the prospect was not charged withthe earlier expelled oil In addition the present-daymaturity of the source rocks within the prospectrsquosfault block is higher than in the source rocks withinthe tectonic block of the Santa Barbara field poten-tially indicating expulsion of more gas A southndashnorth trend can be seen on the modeled accumulationcomposition with more gas toward the allochthon(Figure 9) The modeled composition of the prospectprior to seal failure was much richer in gas than inthe neighboring Santa Barbara field (Figure 9D) Infact simulation scenarios with unbreached sealsresulted in accumulation of predominantly gas Thecombination of both seal integrity and charge compo-sition concerns makes this a risky prospect In addi-tion potential diagenetic fluids related to the faultfluid flow might have altered the reservoir quality

Figure 10 Two-dimensional model (at 12 Ma) with modeled stress to failure Mohr circles (with σ standing for normal stress and τfor shear stress) and petroleum system events charts for the Santa Barbara field (left) and the prospect (right) Model predicts rock fail-ure in the prospect from about 12 Ma For further details see text

NEUMAIER ET AL 1345

Apart from the Onado oil field only minor accu-mulations are modeled in the southern Neogene syn-orogenic sequence The proportion of gas is higherthan in the Furrial trend The accumulations at thesouthern model edge are a modeling artifact repre-senting hydrocarbons (mainly oil) that in reality arethought to have charged the southern foreland basin(from mid to late Miocene times to present day)However parts of the migrated oil and gas could havebeen trapped in the Upper Cretaceous San JuanFormation south of the horst structure

The potential for conventional hydrocarbon accu-mulations in the allochthon is very limited by theimportant fracturing that occurred during the upliftperiod The basal sandstones of the Morochito piggy-back basin north of the Pirital thrust are not chargedin our 2D model because of the absence of structuraltraps Nevertheless there is a potential for strati-graphic traps to have been charged as migrationthrough the basin occurred However seal strengthmight be critical at this low-compaction stage Thesame applies for the Las Piedras Formation We thinkthere might be potential for shale oil because theQuerecual Formation is at shallow depths but showsmaturity within the oil window From the thermalmaturity point of view any potential Jurassic sourcerocks as discussed by Summa et al (2003) mightbe a gas source or shale gas target

UNCERTAINTY ANALYSIS

There is often significant uncertainty in basin geom-etry and depth of stratigraphic horizons based onuncertain seismic interpretation and velocity models(Kukla et al 2000 Brandes et al 2008 Baur et al2010) Restoring of present-day geometries to obtainpaleogeometries adds further uncertainty Additionaluncertainty is related to thermal boundary conditionssuch as paleoheat flow (Beha et al 2008) and togeochemical properties such as source rock character-istics and petroleum generation kinetics and calcula-tion of biodegradation and petroleum density(Blumenstein et al 2008) In addition to the porepressure and thermal sensitivity analysis justdescribed we varied initial TOC within a range of2 to 10 A variation of initial TOC obviously

directly scales the amount of initial kerogen availablefor transformation into hydrocarbons Neither thecharge volume nor composition of the Santa Barbarafield nor the wedge prospects are significantly sensi-tive to variation of initial TOC these traps are filledto spill in each scenario However the amount ofhydrocarbons charging the foreland basin (eg theOnado field) is reduced in the case of low initialTOC A high initial TOC only increased the modeledamount of surface seepage The sensitivity of the ini-tial HI is analogous to the sensitivity of the ini-tial TOC

In addition to the more classical uncertaintyand sensitivity analysis described we focused onquantitative analysis of the decompaction uncer-tainty when performing structural restoration Thisuncertainty is particular to the applied workflowof using structurally restored paleogeometries inBPSM Note that this uncertainty is significantlylower when using the backstripping method withthe optimization algorithms discussed One methodto analyze the adequacy of the decompactionmethod is comparison of the output rock matrixmass balance In the backstripping approach therock matrix mass is kept constant of a model cellwith a simulated decrease in porosity controlledby compaction laws and the corresponding reduc-tion of the layer thickness When using fixed pale-ogeometries from structural restoration the cellthickness is imposed However the decrease inporosity is calculated and the mass is adapted tofit the imposed cell thickness That means that therock matrix mass is constant only if ideal decom-paction has been applied

Figure 11A shows the percentage variation fromthe present-day matrix mass (100) of a given sourcerock cell through geologic time The dotted line rep-resents the ideal case with constant rock matrix massWe used two end-member decompaction scenarios todescribe the decompaction-related uncertainty that isintroduced when fixed paleogeometries replace back-stripping idealized hydrostatic decompaction assum-ing no overpressure and idealized lithostaticdecompaction assuming no effective stress (nodecompaction resulting in constant cell thicknessthrough time) In the latter case with decreasingporosity during compaction the rock matrix mass

1346 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

increases until it reaches the present-day mass Inother words rock mass matrix has been highly under-estimated in the past (in this particular case by morethan 20) This is a source rock so the kerogen massdecreases by the same factor This yields an underes-timation of more than 20 in TOC Note the rockmatrix mass is constant for time steps prior to23 Ma in which backstripping optimization couldbe performed toward the earliest paleogeometry Forthe second end-member scenario of idealized hydro-static decompaction the rock matrix mass is nearlyconstant which is a good decompaction adequacy

indicator for that given cell in the model Howeverthe so-measured adequacy changes throughout themodel In some areas the rock matrix mass has beenunderestimated by about 50 for nondecompactedpaleogeometries however in the case of hydrostaticdecompaction values generally range from 80 to120 of the present-day rock matrix mass Thatmeans in this particular case that hydrostatic decom-paction is more adequate (or less wrong) than nodecompaction at all The decompaction adequacycan be roughly correlated with pressure changesthrough geologic time

Figure 11 Time extractions for comparison of two end-member decompaction scenarios (see Figure 6 for location) (A) Rock matrixmass conservation through geologic time (B) Source rock burial depth through geologic time (C) Temperature and transformation ratiothrough geologic time

NEUMAIER ET AL 1347

Inadequate decompaction results in wrong depthfor stratigraphic horizons in the paleogeometriesFigure 11B shows the burial depth of the same inves-tigated source rock cell We can see that during thepassive margin phase there is a slight difference inpaleoburial depth for the two end-member scenariosDuring the compressional phase the differencechanges dramatically reaching more than 1000 m(3281 ft) in some places Toward present day bothcurves meet again because the input to the two resto-ration scenarios was the same (identical present-dayburial depth) This difference in compaction calcula-tion has severe implications on the modeled tempera-ture history and the kerogen-to-hydrocarbontransformation (Figure 11C) For an identical thermalframework in terms of modeling input parameterstemperature differences of up to 25degC (45degF) resultin TR differences of more than 20 in the early com-pressional phase before TR values meet again for thepresent-day situation

CONCLUSIONS

Structurally complex areas are attracting increasinginterest in hydrocarbon exploration Exploration riskcan be substantially narrowed by analyzing the his-tory of the basin and its petroleum systems based onstructural evolution We presented a further progressin technology that allows the integration of structuralrestoration and BPSM This combined analysis isseen as a way forward to improve modeling in thrustbelts with stacked stratigraphic sequences Despitelimitations of nonoptimized decompaction whichare added to the general uncertainty related to theindividual methods we modeled the charge and sealhistory of the Monagas fold and thrust belt inVenezuela Calibrated porosity temperature andthermal maturity as well as basin-scale stress pro-vided the regional framework through geologic timein which the petroleum systems evolved such thatthe charge of the known Santa Barbara and Onadofields and other trends could be understood We dis-cussed seal integrity of a potential exploration pros-pect by analyzing structural restoration-derived faultdisplacement through geologic time and an advancedstress forward simulation The uncertainty of using

decompacted or nondecompacted structural restora-tion has been assessed It can have important influ-ence on rock matrix mass balance through geologictime paleogeometry and paleoburial depth andtherefore affect paleotemperatures and paleomatur-ities We conclude that decompaction assuminghydrostatic pressure is the best option in that particu-lar case

REFERENCES CITED

Alberdi M and E Lafargue 1993 Vertical variations of organicmatter content in Guayuta Group (Upper Cretaceous)interior mountain belt eastern Venezuela OrganicGeochemistry v 20 p 425ndash436 doi 1010160146-6380(93)90091-O

Athy L F 1930 Density porosity and compaction of sedimen-tary rocks AAPG Bulletin v 14 p 1ndash24

Aymard R L Pimentel P Eitz P Lopez A ChaouchJ Navarro J Mijares and J G Pereira 1990 Geologic inte-gration and evaluation of Northern Monagas EasternVenezuelan Basin in J Brooks ed Classic petroleum prov-inces The Geologic Society (London) Special Publicationv 50 p 37ndash53

Baur F M Di Benedetto T Fuchs C Lampe andS Sciamanna 2009 Integrating structural geology andpetroleum systems modelingmdashA pilot project from Boliviarsquosfold and thrust belt Marine and Petroleum Geology v 26no 4 p 573ndash579 doi 101016jmarpetgeo200901004

Baur F R Littke H Wielens C Lampe and T Fuchs 2010Basin modeling meets rift analysismdashA numerical modelingstudy from the Jeanne drsquoArc basin offshore NewfoundlandCanada Marine and Petroleum Geology v 27 p 585ndash599doi 101016jmarpetgeo200906003

Beha A R O Thomsen and R Littke 2008 Thermal historyhydrocarbon generation and migration in the Horn Grabenin the Danish North Sea A 2D basin modelling studyInternational Journal of Earth Sciences v 97 p 1087ndash1100doi 101007s00531-007-0247-2

Blumenstein I O B M Krooss R di Primio W RottkeE Muumlller C Westerlage and R Littke 2008Biodegradation in numerical basin modelling A case studyfrom the Gifhorn Trough N-Germany International Journalof Earth Sciences v 97 p 1115ndash1129 doi 101007s00531-007-0272-1

Brandes C A Astorga R Littke and J Winsemann 2008Basin modelling of the Limoacuten back-arc basin (Costa Rica)Burial history and temperature evolution of an island arc-related basin-system Basin Research v 20 p 119ndash142doi 101111bre200820issue-1

Carnevali J O 1988 El Furrial oil field northeasternVenezuela First giant in foreland fold and thrust belts ofwestern hemisphere AAPG Bulletin v 72 68 p

Chevalier Y 1987 Contribution a lrsquoetude geacuteologique de lafrontiereacute Sud-Est de la plaque Caraibe Les zones internesde la chaine Sud-Caraibe sur le transect ile de Margarita-

1348 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

Figure 8 Two-dimensional model through geologic time (selected time steps only) with modeled transformation ratio for the sourcerock layers isotherms and liquid and vapor migration and accumulation

1342 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

shales from the Furrial Trend (Roure et al 2005)This important fracturing event resulted in thedestruction of the modeled accumulation (Figure 9E)Hydrocarbon generation and expulsion were alsohappening in the southern foreland basin at transfor-mation ratios of up to 30

Late Compressional and PostcompressionalPhasemdashLate Miocene to Present Day

The late compressional phase was marked by the par-tial destruction of the allochthonous petroleum systemsby erosion (Figure 8E F) Many traps disappeared and

Figure 8 Continued

NEUMAIER ET AL 1343

the hydrocarbon generation within the source rocksstopped The erosion triggered an important sedimenttransfer toward the south contributing to further burialand increased temperatures in the thrust wedge and theforeland basin During late Miocene the petroleumsystem in the wedge matured An important northwardtilting was triggered by flexural tectonic loading andthe loading of syntectonic sediments (Summa et al2003) In combination with the presence of an effec-tive regional top seal the compacted CarapitaFormation allowed a change from dominantly verticalmigration into lateral migration First accumulationsare modeled in the Neogene synorogenic sequence

reservoirs in the south and important southwardmigration toward the foreland basin started(Figure 8G) These trends continued through Plioceneand Pleistocene times to present day filling theOnado field (Figure 8H) and other known fields inthe south (Figure 1A) of the Eastern Venezuela Basin

Prospectivity

We modeled the charge of the Santa Barbara field inthe Furrial Trend and of the Onado field in the fore-land basin (Figure 8H) Other prospects in the thrustwedge were also investigated The structure called

Figure 9 Detail of two-dimensionalmodel through geologic time (seeFigure 8) showing modeled transforma-tion ratio for the source rock layers iso-therms and liquid and vapor migrationand accumulation The Santa Barbara fieldand the prospect are shown by ellipses

1344 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

ldquoprospectrdquo (Figure 9G) looks similar to that of theSanta Barbara field because both are anticlines andthe reservoir and seal formations are the sameHowever they differ in several aspects First theprospect is in the vicinity of the Pirital thrust whichpresents a high risk of impacting the seal integrityThe prospect location is directly below the area wherewe presume the maximum of late tectonic activityoccurred (Figure 4) As modeled the CarapitaFormation the seal of the prospect was already brit-tle when this deformation occurred Calculatedstresses also predict rock failure from 12 Ma onward(Figure 10) Even though charge continued ourmodel suggests continuous leakage of the structure(Figure 9EndashG) The second major difference is thatthe source rock in the fault block of the prospect ismodeled to have already started generating and expel-ling hydrocarbons during the Oligocene and earlyMiocene (Figure 9A) before the trap formed and the

seal became effective (Figures 9C 10) Even thoughgeneration expulsion and migration continued aftertrap formation the prospect was not charged withthe earlier expelled oil In addition the present-daymaturity of the source rocks within the prospectrsquosfault block is higher than in the source rocks withinthe tectonic block of the Santa Barbara field poten-tially indicating expulsion of more gas A southndashnorth trend can be seen on the modeled accumulationcomposition with more gas toward the allochthon(Figure 9) The modeled composition of the prospectprior to seal failure was much richer in gas than inthe neighboring Santa Barbara field (Figure 9D) Infact simulation scenarios with unbreached sealsresulted in accumulation of predominantly gas Thecombination of both seal integrity and charge compo-sition concerns makes this a risky prospect In addi-tion potential diagenetic fluids related to the faultfluid flow might have altered the reservoir quality

Figure 10 Two-dimensional model (at 12 Ma) with modeled stress to failure Mohr circles (with σ standing for normal stress and τfor shear stress) and petroleum system events charts for the Santa Barbara field (left) and the prospect (right) Model predicts rock fail-ure in the prospect from about 12 Ma For further details see text

NEUMAIER ET AL 1345

Apart from the Onado oil field only minor accu-mulations are modeled in the southern Neogene syn-orogenic sequence The proportion of gas is higherthan in the Furrial trend The accumulations at thesouthern model edge are a modeling artifact repre-senting hydrocarbons (mainly oil) that in reality arethought to have charged the southern foreland basin(from mid to late Miocene times to present day)However parts of the migrated oil and gas could havebeen trapped in the Upper Cretaceous San JuanFormation south of the horst structure

The potential for conventional hydrocarbon accu-mulations in the allochthon is very limited by theimportant fracturing that occurred during the upliftperiod The basal sandstones of the Morochito piggy-back basin north of the Pirital thrust are not chargedin our 2D model because of the absence of structuraltraps Nevertheless there is a potential for strati-graphic traps to have been charged as migrationthrough the basin occurred However seal strengthmight be critical at this low-compaction stage Thesame applies for the Las Piedras Formation We thinkthere might be potential for shale oil because theQuerecual Formation is at shallow depths but showsmaturity within the oil window From the thermalmaturity point of view any potential Jurassic sourcerocks as discussed by Summa et al (2003) mightbe a gas source or shale gas target

UNCERTAINTY ANALYSIS

There is often significant uncertainty in basin geom-etry and depth of stratigraphic horizons based onuncertain seismic interpretation and velocity models(Kukla et al 2000 Brandes et al 2008 Baur et al2010) Restoring of present-day geometries to obtainpaleogeometries adds further uncertainty Additionaluncertainty is related to thermal boundary conditionssuch as paleoheat flow (Beha et al 2008) and togeochemical properties such as source rock character-istics and petroleum generation kinetics and calcula-tion of biodegradation and petroleum density(Blumenstein et al 2008) In addition to the porepressure and thermal sensitivity analysis justdescribed we varied initial TOC within a range of2 to 10 A variation of initial TOC obviously

directly scales the amount of initial kerogen availablefor transformation into hydrocarbons Neither thecharge volume nor composition of the Santa Barbarafield nor the wedge prospects are significantly sensi-tive to variation of initial TOC these traps are filledto spill in each scenario However the amount ofhydrocarbons charging the foreland basin (eg theOnado field) is reduced in the case of low initialTOC A high initial TOC only increased the modeledamount of surface seepage The sensitivity of the ini-tial HI is analogous to the sensitivity of the ini-tial TOC

In addition to the more classical uncertaintyand sensitivity analysis described we focused onquantitative analysis of the decompaction uncer-tainty when performing structural restoration Thisuncertainty is particular to the applied workflowof using structurally restored paleogeometries inBPSM Note that this uncertainty is significantlylower when using the backstripping method withthe optimization algorithms discussed One methodto analyze the adequacy of the decompactionmethod is comparison of the output rock matrixmass balance In the backstripping approach therock matrix mass is kept constant of a model cellwith a simulated decrease in porosity controlledby compaction laws and the corresponding reduc-tion of the layer thickness When using fixed pale-ogeometries from structural restoration the cellthickness is imposed However the decrease inporosity is calculated and the mass is adapted tofit the imposed cell thickness That means that therock matrix mass is constant only if ideal decom-paction has been applied

Figure 11A shows the percentage variation fromthe present-day matrix mass (100) of a given sourcerock cell through geologic time The dotted line rep-resents the ideal case with constant rock matrix massWe used two end-member decompaction scenarios todescribe the decompaction-related uncertainty that isintroduced when fixed paleogeometries replace back-stripping idealized hydrostatic decompaction assum-ing no overpressure and idealized lithostaticdecompaction assuming no effective stress (nodecompaction resulting in constant cell thicknessthrough time) In the latter case with decreasingporosity during compaction the rock matrix mass

1346 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

increases until it reaches the present-day mass Inother words rock mass matrix has been highly under-estimated in the past (in this particular case by morethan 20) This is a source rock so the kerogen massdecreases by the same factor This yields an underes-timation of more than 20 in TOC Note the rockmatrix mass is constant for time steps prior to23 Ma in which backstripping optimization couldbe performed toward the earliest paleogeometry Forthe second end-member scenario of idealized hydro-static decompaction the rock matrix mass is nearlyconstant which is a good decompaction adequacy

indicator for that given cell in the model Howeverthe so-measured adequacy changes throughout themodel In some areas the rock matrix mass has beenunderestimated by about 50 for nondecompactedpaleogeometries however in the case of hydrostaticdecompaction values generally range from 80 to120 of the present-day rock matrix mass Thatmeans in this particular case that hydrostatic decom-paction is more adequate (or less wrong) than nodecompaction at all The decompaction adequacycan be roughly correlated with pressure changesthrough geologic time

Figure 11 Time extractions for comparison of two end-member decompaction scenarios (see Figure 6 for location) (A) Rock matrixmass conservation through geologic time (B) Source rock burial depth through geologic time (C) Temperature and transformation ratiothrough geologic time

NEUMAIER ET AL 1347

Inadequate decompaction results in wrong depthfor stratigraphic horizons in the paleogeometriesFigure 11B shows the burial depth of the same inves-tigated source rock cell We can see that during thepassive margin phase there is a slight difference inpaleoburial depth for the two end-member scenariosDuring the compressional phase the differencechanges dramatically reaching more than 1000 m(3281 ft) in some places Toward present day bothcurves meet again because the input to the two resto-ration scenarios was the same (identical present-dayburial depth) This difference in compaction calcula-tion has severe implications on the modeled tempera-ture history and the kerogen-to-hydrocarbontransformation (Figure 11C) For an identical thermalframework in terms of modeling input parameterstemperature differences of up to 25degC (45degF) resultin TR differences of more than 20 in the early com-pressional phase before TR values meet again for thepresent-day situation

CONCLUSIONS

Structurally complex areas are attracting increasinginterest in hydrocarbon exploration Exploration riskcan be substantially narrowed by analyzing the his-tory of the basin and its petroleum systems based onstructural evolution We presented a further progressin technology that allows the integration of structuralrestoration and BPSM This combined analysis isseen as a way forward to improve modeling in thrustbelts with stacked stratigraphic sequences Despitelimitations of nonoptimized decompaction whichare added to the general uncertainty related to theindividual methods we modeled the charge and sealhistory of the Monagas fold and thrust belt inVenezuela Calibrated porosity temperature andthermal maturity as well as basin-scale stress pro-vided the regional framework through geologic timein which the petroleum systems evolved such thatthe charge of the known Santa Barbara and Onadofields and other trends could be understood We dis-cussed seal integrity of a potential exploration pros-pect by analyzing structural restoration-derived faultdisplacement through geologic time and an advancedstress forward simulation The uncertainty of using

decompacted or nondecompacted structural restora-tion has been assessed It can have important influ-ence on rock matrix mass balance through geologictime paleogeometry and paleoburial depth andtherefore affect paleotemperatures and paleomatur-ities We conclude that decompaction assuminghydrostatic pressure is the best option in that particu-lar case

REFERENCES CITED

Alberdi M and E Lafargue 1993 Vertical variations of organicmatter content in Guayuta Group (Upper Cretaceous)interior mountain belt eastern Venezuela OrganicGeochemistry v 20 p 425ndash436 doi 1010160146-6380(93)90091-O

Athy L F 1930 Density porosity and compaction of sedimen-tary rocks AAPG Bulletin v 14 p 1ndash24

Aymard R L Pimentel P Eitz P Lopez A ChaouchJ Navarro J Mijares and J G Pereira 1990 Geologic inte-gration and evaluation of Northern Monagas EasternVenezuelan Basin in J Brooks ed Classic petroleum prov-inces The Geologic Society (London) Special Publicationv 50 p 37ndash53

Baur F M Di Benedetto T Fuchs C Lampe andS Sciamanna 2009 Integrating structural geology andpetroleum systems modelingmdashA pilot project from Boliviarsquosfold and thrust belt Marine and Petroleum Geology v 26no 4 p 573ndash579 doi 101016jmarpetgeo200901004

Baur F R Littke H Wielens C Lampe and T Fuchs 2010Basin modeling meets rift analysismdashA numerical modelingstudy from the Jeanne drsquoArc basin offshore NewfoundlandCanada Marine and Petroleum Geology v 27 p 585ndash599doi 101016jmarpetgeo200906003

Beha A R O Thomsen and R Littke 2008 Thermal historyhydrocarbon generation and migration in the Horn Grabenin the Danish North Sea A 2D basin modelling studyInternational Journal of Earth Sciences v 97 p 1087ndash1100doi 101007s00531-007-0247-2

Blumenstein I O B M Krooss R di Primio W RottkeE Muumlller C Westerlage and R Littke 2008Biodegradation in numerical basin modelling A case studyfrom the Gifhorn Trough N-Germany International Journalof Earth Sciences v 97 p 1115ndash1129 doi 101007s00531-007-0272-1

Brandes C A Astorga R Littke and J Winsemann 2008Basin modelling of the Limoacuten back-arc basin (Costa Rica)Burial history and temperature evolution of an island arc-related basin-system Basin Research v 20 p 119ndash142doi 101111bre200820issue-1

Carnevali J O 1988 El Furrial oil field northeasternVenezuela First giant in foreland fold and thrust belts ofwestern hemisphere AAPG Bulletin v 72 68 p

Chevalier Y 1987 Contribution a lrsquoetude geacuteologique de lafrontiereacute Sud-Est de la plaque Caraibe Les zones internesde la chaine Sud-Caraibe sur le transect ile de Margarita-

1348 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

shales from the Furrial Trend (Roure et al 2005)This important fracturing event resulted in thedestruction of the modeled accumulation (Figure 9E)Hydrocarbon generation and expulsion were alsohappening in the southern foreland basin at transfor-mation ratios of up to 30

Late Compressional and PostcompressionalPhasemdashLate Miocene to Present Day

The late compressional phase was marked by the par-tial destruction of the allochthonous petroleum systemsby erosion (Figure 8E F) Many traps disappeared and

Figure 8 Continued

NEUMAIER ET AL 1343

the hydrocarbon generation within the source rocksstopped The erosion triggered an important sedimenttransfer toward the south contributing to further burialand increased temperatures in the thrust wedge and theforeland basin During late Miocene the petroleumsystem in the wedge matured An important northwardtilting was triggered by flexural tectonic loading andthe loading of syntectonic sediments (Summa et al2003) In combination with the presence of an effec-tive regional top seal the compacted CarapitaFormation allowed a change from dominantly verticalmigration into lateral migration First accumulationsare modeled in the Neogene synorogenic sequence

reservoirs in the south and important southwardmigration toward the foreland basin started(Figure 8G) These trends continued through Plioceneand Pleistocene times to present day filling theOnado field (Figure 8H) and other known fields inthe south (Figure 1A) of the Eastern Venezuela Basin

Prospectivity

We modeled the charge of the Santa Barbara field inthe Furrial Trend and of the Onado field in the fore-land basin (Figure 8H) Other prospects in the thrustwedge were also investigated The structure called

Figure 9 Detail of two-dimensionalmodel through geologic time (seeFigure 8) showing modeled transforma-tion ratio for the source rock layers iso-therms and liquid and vapor migrationand accumulation The Santa Barbara fieldand the prospect are shown by ellipses

1344 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

ldquoprospectrdquo (Figure 9G) looks similar to that of theSanta Barbara field because both are anticlines andthe reservoir and seal formations are the sameHowever they differ in several aspects First theprospect is in the vicinity of the Pirital thrust whichpresents a high risk of impacting the seal integrityThe prospect location is directly below the area wherewe presume the maximum of late tectonic activityoccurred (Figure 4) As modeled the CarapitaFormation the seal of the prospect was already brit-tle when this deformation occurred Calculatedstresses also predict rock failure from 12 Ma onward(Figure 10) Even though charge continued ourmodel suggests continuous leakage of the structure(Figure 9EndashG) The second major difference is thatthe source rock in the fault block of the prospect ismodeled to have already started generating and expel-ling hydrocarbons during the Oligocene and earlyMiocene (Figure 9A) before the trap formed and the

seal became effective (Figures 9C 10) Even thoughgeneration expulsion and migration continued aftertrap formation the prospect was not charged withthe earlier expelled oil In addition the present-daymaturity of the source rocks within the prospectrsquosfault block is higher than in the source rocks withinthe tectonic block of the Santa Barbara field poten-tially indicating expulsion of more gas A southndashnorth trend can be seen on the modeled accumulationcomposition with more gas toward the allochthon(Figure 9) The modeled composition of the prospectprior to seal failure was much richer in gas than inthe neighboring Santa Barbara field (Figure 9D) Infact simulation scenarios with unbreached sealsresulted in accumulation of predominantly gas Thecombination of both seal integrity and charge compo-sition concerns makes this a risky prospect In addi-tion potential diagenetic fluids related to the faultfluid flow might have altered the reservoir quality

Figure 10 Two-dimensional model (at 12 Ma) with modeled stress to failure Mohr circles (with σ standing for normal stress and τfor shear stress) and petroleum system events charts for the Santa Barbara field (left) and the prospect (right) Model predicts rock fail-ure in the prospect from about 12 Ma For further details see text

NEUMAIER ET AL 1345

Apart from the Onado oil field only minor accu-mulations are modeled in the southern Neogene syn-orogenic sequence The proportion of gas is higherthan in the Furrial trend The accumulations at thesouthern model edge are a modeling artifact repre-senting hydrocarbons (mainly oil) that in reality arethought to have charged the southern foreland basin(from mid to late Miocene times to present day)However parts of the migrated oil and gas could havebeen trapped in the Upper Cretaceous San JuanFormation south of the horst structure

The potential for conventional hydrocarbon accu-mulations in the allochthon is very limited by theimportant fracturing that occurred during the upliftperiod The basal sandstones of the Morochito piggy-back basin north of the Pirital thrust are not chargedin our 2D model because of the absence of structuraltraps Nevertheless there is a potential for strati-graphic traps to have been charged as migrationthrough the basin occurred However seal strengthmight be critical at this low-compaction stage Thesame applies for the Las Piedras Formation We thinkthere might be potential for shale oil because theQuerecual Formation is at shallow depths but showsmaturity within the oil window From the thermalmaturity point of view any potential Jurassic sourcerocks as discussed by Summa et al (2003) mightbe a gas source or shale gas target

UNCERTAINTY ANALYSIS

There is often significant uncertainty in basin geom-etry and depth of stratigraphic horizons based onuncertain seismic interpretation and velocity models(Kukla et al 2000 Brandes et al 2008 Baur et al2010) Restoring of present-day geometries to obtainpaleogeometries adds further uncertainty Additionaluncertainty is related to thermal boundary conditionssuch as paleoheat flow (Beha et al 2008) and togeochemical properties such as source rock character-istics and petroleum generation kinetics and calcula-tion of biodegradation and petroleum density(Blumenstein et al 2008) In addition to the porepressure and thermal sensitivity analysis justdescribed we varied initial TOC within a range of2 to 10 A variation of initial TOC obviously

directly scales the amount of initial kerogen availablefor transformation into hydrocarbons Neither thecharge volume nor composition of the Santa Barbarafield nor the wedge prospects are significantly sensi-tive to variation of initial TOC these traps are filledto spill in each scenario However the amount ofhydrocarbons charging the foreland basin (eg theOnado field) is reduced in the case of low initialTOC A high initial TOC only increased the modeledamount of surface seepage The sensitivity of the ini-tial HI is analogous to the sensitivity of the ini-tial TOC

In addition to the more classical uncertaintyand sensitivity analysis described we focused onquantitative analysis of the decompaction uncer-tainty when performing structural restoration Thisuncertainty is particular to the applied workflowof using structurally restored paleogeometries inBPSM Note that this uncertainty is significantlylower when using the backstripping method withthe optimization algorithms discussed One methodto analyze the adequacy of the decompactionmethod is comparison of the output rock matrixmass balance In the backstripping approach therock matrix mass is kept constant of a model cellwith a simulated decrease in porosity controlledby compaction laws and the corresponding reduc-tion of the layer thickness When using fixed pale-ogeometries from structural restoration the cellthickness is imposed However the decrease inporosity is calculated and the mass is adapted tofit the imposed cell thickness That means that therock matrix mass is constant only if ideal decom-paction has been applied

Figure 11A shows the percentage variation fromthe present-day matrix mass (100) of a given sourcerock cell through geologic time The dotted line rep-resents the ideal case with constant rock matrix massWe used two end-member decompaction scenarios todescribe the decompaction-related uncertainty that isintroduced when fixed paleogeometries replace back-stripping idealized hydrostatic decompaction assum-ing no overpressure and idealized lithostaticdecompaction assuming no effective stress (nodecompaction resulting in constant cell thicknessthrough time) In the latter case with decreasingporosity during compaction the rock matrix mass

1346 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

increases until it reaches the present-day mass Inother words rock mass matrix has been highly under-estimated in the past (in this particular case by morethan 20) This is a source rock so the kerogen massdecreases by the same factor This yields an underes-timation of more than 20 in TOC Note the rockmatrix mass is constant for time steps prior to23 Ma in which backstripping optimization couldbe performed toward the earliest paleogeometry Forthe second end-member scenario of idealized hydro-static decompaction the rock matrix mass is nearlyconstant which is a good decompaction adequacy

indicator for that given cell in the model Howeverthe so-measured adequacy changes throughout themodel In some areas the rock matrix mass has beenunderestimated by about 50 for nondecompactedpaleogeometries however in the case of hydrostaticdecompaction values generally range from 80 to120 of the present-day rock matrix mass Thatmeans in this particular case that hydrostatic decom-paction is more adequate (or less wrong) than nodecompaction at all The decompaction adequacycan be roughly correlated with pressure changesthrough geologic time

Figure 11 Time extractions for comparison of two end-member decompaction scenarios (see Figure 6 for location) (A) Rock matrixmass conservation through geologic time (B) Source rock burial depth through geologic time (C) Temperature and transformation ratiothrough geologic time

NEUMAIER ET AL 1347

Inadequate decompaction results in wrong depthfor stratigraphic horizons in the paleogeometriesFigure 11B shows the burial depth of the same inves-tigated source rock cell We can see that during thepassive margin phase there is a slight difference inpaleoburial depth for the two end-member scenariosDuring the compressional phase the differencechanges dramatically reaching more than 1000 m(3281 ft) in some places Toward present day bothcurves meet again because the input to the two resto-ration scenarios was the same (identical present-dayburial depth) This difference in compaction calcula-tion has severe implications on the modeled tempera-ture history and the kerogen-to-hydrocarbontransformation (Figure 11C) For an identical thermalframework in terms of modeling input parameterstemperature differences of up to 25degC (45degF) resultin TR differences of more than 20 in the early com-pressional phase before TR values meet again for thepresent-day situation

CONCLUSIONS

Structurally complex areas are attracting increasinginterest in hydrocarbon exploration Exploration riskcan be substantially narrowed by analyzing the his-tory of the basin and its petroleum systems based onstructural evolution We presented a further progressin technology that allows the integration of structuralrestoration and BPSM This combined analysis isseen as a way forward to improve modeling in thrustbelts with stacked stratigraphic sequences Despitelimitations of nonoptimized decompaction whichare added to the general uncertainty related to theindividual methods we modeled the charge and sealhistory of the Monagas fold and thrust belt inVenezuela Calibrated porosity temperature andthermal maturity as well as basin-scale stress pro-vided the regional framework through geologic timein which the petroleum systems evolved such thatthe charge of the known Santa Barbara and Onadofields and other trends could be understood We dis-cussed seal integrity of a potential exploration pros-pect by analyzing structural restoration-derived faultdisplacement through geologic time and an advancedstress forward simulation The uncertainty of using

decompacted or nondecompacted structural restora-tion has been assessed It can have important influ-ence on rock matrix mass balance through geologictime paleogeometry and paleoburial depth andtherefore affect paleotemperatures and paleomatur-ities We conclude that decompaction assuminghydrostatic pressure is the best option in that particu-lar case

REFERENCES CITED

Alberdi M and E Lafargue 1993 Vertical variations of organicmatter content in Guayuta Group (Upper Cretaceous)interior mountain belt eastern Venezuela OrganicGeochemistry v 20 p 425ndash436 doi 1010160146-6380(93)90091-O

Athy L F 1930 Density porosity and compaction of sedimen-tary rocks AAPG Bulletin v 14 p 1ndash24

Aymard R L Pimentel P Eitz P Lopez A ChaouchJ Navarro J Mijares and J G Pereira 1990 Geologic inte-gration and evaluation of Northern Monagas EasternVenezuelan Basin in J Brooks ed Classic petroleum prov-inces The Geologic Society (London) Special Publicationv 50 p 37ndash53

Baur F M Di Benedetto T Fuchs C Lampe andS Sciamanna 2009 Integrating structural geology andpetroleum systems modelingmdashA pilot project from Boliviarsquosfold and thrust belt Marine and Petroleum Geology v 26no 4 p 573ndash579 doi 101016jmarpetgeo200901004

Baur F R Littke H Wielens C Lampe and T Fuchs 2010Basin modeling meets rift analysismdashA numerical modelingstudy from the Jeanne drsquoArc basin offshore NewfoundlandCanada Marine and Petroleum Geology v 27 p 585ndash599doi 101016jmarpetgeo200906003

Beha A R O Thomsen and R Littke 2008 Thermal historyhydrocarbon generation and migration in the Horn Grabenin the Danish North Sea A 2D basin modelling studyInternational Journal of Earth Sciences v 97 p 1087ndash1100doi 101007s00531-007-0247-2

Blumenstein I O B M Krooss R di Primio W RottkeE Muumlller C Westerlage and R Littke 2008Biodegradation in numerical basin modelling A case studyfrom the Gifhorn Trough N-Germany International Journalof Earth Sciences v 97 p 1115ndash1129 doi 101007s00531-007-0272-1

Brandes C A Astorga R Littke and J Winsemann 2008Basin modelling of the Limoacuten back-arc basin (Costa Rica)Burial history and temperature evolution of an island arc-related basin-system Basin Research v 20 p 119ndash142doi 101111bre200820issue-1

Carnevali J O 1988 El Furrial oil field northeasternVenezuela First giant in foreland fold and thrust belts ofwestern hemisphere AAPG Bulletin v 72 68 p

Chevalier Y 1987 Contribution a lrsquoetude geacuteologique de lafrontiereacute Sud-Est de la plaque Caraibe Les zones internesde la chaine Sud-Caraibe sur le transect ile de Margarita-

1348 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

the hydrocarbon generation within the source rocksstopped The erosion triggered an important sedimenttransfer toward the south contributing to further burialand increased temperatures in the thrust wedge and theforeland basin During late Miocene the petroleumsystem in the wedge matured An important northwardtilting was triggered by flexural tectonic loading andthe loading of syntectonic sediments (Summa et al2003) In combination with the presence of an effec-tive regional top seal the compacted CarapitaFormation allowed a change from dominantly verticalmigration into lateral migration First accumulationsare modeled in the Neogene synorogenic sequence

reservoirs in the south and important southwardmigration toward the foreland basin started(Figure 8G) These trends continued through Plioceneand Pleistocene times to present day filling theOnado field (Figure 8H) and other known fields inthe south (Figure 1A) of the Eastern Venezuela Basin

Prospectivity

We modeled the charge of the Santa Barbara field inthe Furrial Trend and of the Onado field in the fore-land basin (Figure 8H) Other prospects in the thrustwedge were also investigated The structure called

Figure 9 Detail of two-dimensionalmodel through geologic time (seeFigure 8) showing modeled transforma-tion ratio for the source rock layers iso-therms and liquid and vapor migrationand accumulation The Santa Barbara fieldand the prospect are shown by ellipses

1344 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

ldquoprospectrdquo (Figure 9G) looks similar to that of theSanta Barbara field because both are anticlines andthe reservoir and seal formations are the sameHowever they differ in several aspects First theprospect is in the vicinity of the Pirital thrust whichpresents a high risk of impacting the seal integrityThe prospect location is directly below the area wherewe presume the maximum of late tectonic activityoccurred (Figure 4) As modeled the CarapitaFormation the seal of the prospect was already brit-tle when this deformation occurred Calculatedstresses also predict rock failure from 12 Ma onward(Figure 10) Even though charge continued ourmodel suggests continuous leakage of the structure(Figure 9EndashG) The second major difference is thatthe source rock in the fault block of the prospect ismodeled to have already started generating and expel-ling hydrocarbons during the Oligocene and earlyMiocene (Figure 9A) before the trap formed and the

seal became effective (Figures 9C 10) Even thoughgeneration expulsion and migration continued aftertrap formation the prospect was not charged withthe earlier expelled oil In addition the present-daymaturity of the source rocks within the prospectrsquosfault block is higher than in the source rocks withinthe tectonic block of the Santa Barbara field poten-tially indicating expulsion of more gas A southndashnorth trend can be seen on the modeled accumulationcomposition with more gas toward the allochthon(Figure 9) The modeled composition of the prospectprior to seal failure was much richer in gas than inthe neighboring Santa Barbara field (Figure 9D) Infact simulation scenarios with unbreached sealsresulted in accumulation of predominantly gas Thecombination of both seal integrity and charge compo-sition concerns makes this a risky prospect In addi-tion potential diagenetic fluids related to the faultfluid flow might have altered the reservoir quality

Figure 10 Two-dimensional model (at 12 Ma) with modeled stress to failure Mohr circles (with σ standing for normal stress and τfor shear stress) and petroleum system events charts for the Santa Barbara field (left) and the prospect (right) Model predicts rock fail-ure in the prospect from about 12 Ma For further details see text

NEUMAIER ET AL 1345

Apart from the Onado oil field only minor accu-mulations are modeled in the southern Neogene syn-orogenic sequence The proportion of gas is higherthan in the Furrial trend The accumulations at thesouthern model edge are a modeling artifact repre-senting hydrocarbons (mainly oil) that in reality arethought to have charged the southern foreland basin(from mid to late Miocene times to present day)However parts of the migrated oil and gas could havebeen trapped in the Upper Cretaceous San JuanFormation south of the horst structure

The potential for conventional hydrocarbon accu-mulations in the allochthon is very limited by theimportant fracturing that occurred during the upliftperiod The basal sandstones of the Morochito piggy-back basin north of the Pirital thrust are not chargedin our 2D model because of the absence of structuraltraps Nevertheless there is a potential for strati-graphic traps to have been charged as migrationthrough the basin occurred However seal strengthmight be critical at this low-compaction stage Thesame applies for the Las Piedras Formation We thinkthere might be potential for shale oil because theQuerecual Formation is at shallow depths but showsmaturity within the oil window From the thermalmaturity point of view any potential Jurassic sourcerocks as discussed by Summa et al (2003) mightbe a gas source or shale gas target

UNCERTAINTY ANALYSIS

There is often significant uncertainty in basin geom-etry and depth of stratigraphic horizons based onuncertain seismic interpretation and velocity models(Kukla et al 2000 Brandes et al 2008 Baur et al2010) Restoring of present-day geometries to obtainpaleogeometries adds further uncertainty Additionaluncertainty is related to thermal boundary conditionssuch as paleoheat flow (Beha et al 2008) and togeochemical properties such as source rock character-istics and petroleum generation kinetics and calcula-tion of biodegradation and petroleum density(Blumenstein et al 2008) In addition to the porepressure and thermal sensitivity analysis justdescribed we varied initial TOC within a range of2 to 10 A variation of initial TOC obviously

directly scales the amount of initial kerogen availablefor transformation into hydrocarbons Neither thecharge volume nor composition of the Santa Barbarafield nor the wedge prospects are significantly sensi-tive to variation of initial TOC these traps are filledto spill in each scenario However the amount ofhydrocarbons charging the foreland basin (eg theOnado field) is reduced in the case of low initialTOC A high initial TOC only increased the modeledamount of surface seepage The sensitivity of the ini-tial HI is analogous to the sensitivity of the ini-tial TOC

In addition to the more classical uncertaintyand sensitivity analysis described we focused onquantitative analysis of the decompaction uncer-tainty when performing structural restoration Thisuncertainty is particular to the applied workflowof using structurally restored paleogeometries inBPSM Note that this uncertainty is significantlylower when using the backstripping method withthe optimization algorithms discussed One methodto analyze the adequacy of the decompactionmethod is comparison of the output rock matrixmass balance In the backstripping approach therock matrix mass is kept constant of a model cellwith a simulated decrease in porosity controlledby compaction laws and the corresponding reduc-tion of the layer thickness When using fixed pale-ogeometries from structural restoration the cellthickness is imposed However the decrease inporosity is calculated and the mass is adapted tofit the imposed cell thickness That means that therock matrix mass is constant only if ideal decom-paction has been applied

Figure 11A shows the percentage variation fromthe present-day matrix mass (100) of a given sourcerock cell through geologic time The dotted line rep-resents the ideal case with constant rock matrix massWe used two end-member decompaction scenarios todescribe the decompaction-related uncertainty that isintroduced when fixed paleogeometries replace back-stripping idealized hydrostatic decompaction assum-ing no overpressure and idealized lithostaticdecompaction assuming no effective stress (nodecompaction resulting in constant cell thicknessthrough time) In the latter case with decreasingporosity during compaction the rock matrix mass

1346 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

increases until it reaches the present-day mass Inother words rock mass matrix has been highly under-estimated in the past (in this particular case by morethan 20) This is a source rock so the kerogen massdecreases by the same factor This yields an underes-timation of more than 20 in TOC Note the rockmatrix mass is constant for time steps prior to23 Ma in which backstripping optimization couldbe performed toward the earliest paleogeometry Forthe second end-member scenario of idealized hydro-static decompaction the rock matrix mass is nearlyconstant which is a good decompaction adequacy

indicator for that given cell in the model Howeverthe so-measured adequacy changes throughout themodel In some areas the rock matrix mass has beenunderestimated by about 50 for nondecompactedpaleogeometries however in the case of hydrostaticdecompaction values generally range from 80 to120 of the present-day rock matrix mass Thatmeans in this particular case that hydrostatic decom-paction is more adequate (or less wrong) than nodecompaction at all The decompaction adequacycan be roughly correlated with pressure changesthrough geologic time

Figure 11 Time extractions for comparison of two end-member decompaction scenarios (see Figure 6 for location) (A) Rock matrixmass conservation through geologic time (B) Source rock burial depth through geologic time (C) Temperature and transformation ratiothrough geologic time

NEUMAIER ET AL 1347

Inadequate decompaction results in wrong depthfor stratigraphic horizons in the paleogeometriesFigure 11B shows the burial depth of the same inves-tigated source rock cell We can see that during thepassive margin phase there is a slight difference inpaleoburial depth for the two end-member scenariosDuring the compressional phase the differencechanges dramatically reaching more than 1000 m(3281 ft) in some places Toward present day bothcurves meet again because the input to the two resto-ration scenarios was the same (identical present-dayburial depth) This difference in compaction calcula-tion has severe implications on the modeled tempera-ture history and the kerogen-to-hydrocarbontransformation (Figure 11C) For an identical thermalframework in terms of modeling input parameterstemperature differences of up to 25degC (45degF) resultin TR differences of more than 20 in the early com-pressional phase before TR values meet again for thepresent-day situation

CONCLUSIONS

Structurally complex areas are attracting increasinginterest in hydrocarbon exploration Exploration riskcan be substantially narrowed by analyzing the his-tory of the basin and its petroleum systems based onstructural evolution We presented a further progressin technology that allows the integration of structuralrestoration and BPSM This combined analysis isseen as a way forward to improve modeling in thrustbelts with stacked stratigraphic sequences Despitelimitations of nonoptimized decompaction whichare added to the general uncertainty related to theindividual methods we modeled the charge and sealhistory of the Monagas fold and thrust belt inVenezuela Calibrated porosity temperature andthermal maturity as well as basin-scale stress pro-vided the regional framework through geologic timein which the petroleum systems evolved such thatthe charge of the known Santa Barbara and Onadofields and other trends could be understood We dis-cussed seal integrity of a potential exploration pros-pect by analyzing structural restoration-derived faultdisplacement through geologic time and an advancedstress forward simulation The uncertainty of using

decompacted or nondecompacted structural restora-tion has been assessed It can have important influ-ence on rock matrix mass balance through geologictime paleogeometry and paleoburial depth andtherefore affect paleotemperatures and paleomatur-ities We conclude that decompaction assuminghydrostatic pressure is the best option in that particu-lar case

REFERENCES CITED

Alberdi M and E Lafargue 1993 Vertical variations of organicmatter content in Guayuta Group (Upper Cretaceous)interior mountain belt eastern Venezuela OrganicGeochemistry v 20 p 425ndash436 doi 1010160146-6380(93)90091-O

Athy L F 1930 Density porosity and compaction of sedimen-tary rocks AAPG Bulletin v 14 p 1ndash24

Aymard R L Pimentel P Eitz P Lopez A ChaouchJ Navarro J Mijares and J G Pereira 1990 Geologic inte-gration and evaluation of Northern Monagas EasternVenezuelan Basin in J Brooks ed Classic petroleum prov-inces The Geologic Society (London) Special Publicationv 50 p 37ndash53

Baur F M Di Benedetto T Fuchs C Lampe andS Sciamanna 2009 Integrating structural geology andpetroleum systems modelingmdashA pilot project from Boliviarsquosfold and thrust belt Marine and Petroleum Geology v 26no 4 p 573ndash579 doi 101016jmarpetgeo200901004

Baur F R Littke H Wielens C Lampe and T Fuchs 2010Basin modeling meets rift analysismdashA numerical modelingstudy from the Jeanne drsquoArc basin offshore NewfoundlandCanada Marine and Petroleum Geology v 27 p 585ndash599doi 101016jmarpetgeo200906003

Beha A R O Thomsen and R Littke 2008 Thermal historyhydrocarbon generation and migration in the Horn Grabenin the Danish North Sea A 2D basin modelling studyInternational Journal of Earth Sciences v 97 p 1087ndash1100doi 101007s00531-007-0247-2

Blumenstein I O B M Krooss R di Primio W RottkeE Muumlller C Westerlage and R Littke 2008Biodegradation in numerical basin modelling A case studyfrom the Gifhorn Trough N-Germany International Journalof Earth Sciences v 97 p 1115ndash1129 doi 101007s00531-007-0272-1

Brandes C A Astorga R Littke and J Winsemann 2008Basin modelling of the Limoacuten back-arc basin (Costa Rica)Burial history and temperature evolution of an island arc-related basin-system Basin Research v 20 p 119ndash142doi 101111bre200820issue-1

Carnevali J O 1988 El Furrial oil field northeasternVenezuela First giant in foreland fold and thrust belts ofwestern hemisphere AAPG Bulletin v 72 68 p

Chevalier Y 1987 Contribution a lrsquoetude geacuteologique de lafrontiereacute Sud-Est de la plaque Caraibe Les zones internesde la chaine Sud-Caraibe sur le transect ile de Margarita-

1348 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

ldquoprospectrdquo (Figure 9G) looks similar to that of theSanta Barbara field because both are anticlines andthe reservoir and seal formations are the sameHowever they differ in several aspects First theprospect is in the vicinity of the Pirital thrust whichpresents a high risk of impacting the seal integrityThe prospect location is directly below the area wherewe presume the maximum of late tectonic activityoccurred (Figure 4) As modeled the CarapitaFormation the seal of the prospect was already brit-tle when this deformation occurred Calculatedstresses also predict rock failure from 12 Ma onward(Figure 10) Even though charge continued ourmodel suggests continuous leakage of the structure(Figure 9EndashG) The second major difference is thatthe source rock in the fault block of the prospect ismodeled to have already started generating and expel-ling hydrocarbons during the Oligocene and earlyMiocene (Figure 9A) before the trap formed and the

seal became effective (Figures 9C 10) Even thoughgeneration expulsion and migration continued aftertrap formation the prospect was not charged withthe earlier expelled oil In addition the present-daymaturity of the source rocks within the prospectrsquosfault block is higher than in the source rocks withinthe tectonic block of the Santa Barbara field poten-tially indicating expulsion of more gas A southndashnorth trend can be seen on the modeled accumulationcomposition with more gas toward the allochthon(Figure 9) The modeled composition of the prospectprior to seal failure was much richer in gas than inthe neighboring Santa Barbara field (Figure 9D) Infact simulation scenarios with unbreached sealsresulted in accumulation of predominantly gas Thecombination of both seal integrity and charge compo-sition concerns makes this a risky prospect In addi-tion potential diagenetic fluids related to the faultfluid flow might have altered the reservoir quality

Figure 10 Two-dimensional model (at 12 Ma) with modeled stress to failure Mohr circles (with σ standing for normal stress and τfor shear stress) and petroleum system events charts for the Santa Barbara field (left) and the prospect (right) Model predicts rock fail-ure in the prospect from about 12 Ma For further details see text

NEUMAIER ET AL 1345

Apart from the Onado oil field only minor accu-mulations are modeled in the southern Neogene syn-orogenic sequence The proportion of gas is higherthan in the Furrial trend The accumulations at thesouthern model edge are a modeling artifact repre-senting hydrocarbons (mainly oil) that in reality arethought to have charged the southern foreland basin(from mid to late Miocene times to present day)However parts of the migrated oil and gas could havebeen trapped in the Upper Cretaceous San JuanFormation south of the horst structure

The potential for conventional hydrocarbon accu-mulations in the allochthon is very limited by theimportant fracturing that occurred during the upliftperiod The basal sandstones of the Morochito piggy-back basin north of the Pirital thrust are not chargedin our 2D model because of the absence of structuraltraps Nevertheless there is a potential for strati-graphic traps to have been charged as migrationthrough the basin occurred However seal strengthmight be critical at this low-compaction stage Thesame applies for the Las Piedras Formation We thinkthere might be potential for shale oil because theQuerecual Formation is at shallow depths but showsmaturity within the oil window From the thermalmaturity point of view any potential Jurassic sourcerocks as discussed by Summa et al (2003) mightbe a gas source or shale gas target

UNCERTAINTY ANALYSIS

There is often significant uncertainty in basin geom-etry and depth of stratigraphic horizons based onuncertain seismic interpretation and velocity models(Kukla et al 2000 Brandes et al 2008 Baur et al2010) Restoring of present-day geometries to obtainpaleogeometries adds further uncertainty Additionaluncertainty is related to thermal boundary conditionssuch as paleoheat flow (Beha et al 2008) and togeochemical properties such as source rock character-istics and petroleum generation kinetics and calcula-tion of biodegradation and petroleum density(Blumenstein et al 2008) In addition to the porepressure and thermal sensitivity analysis justdescribed we varied initial TOC within a range of2 to 10 A variation of initial TOC obviously

directly scales the amount of initial kerogen availablefor transformation into hydrocarbons Neither thecharge volume nor composition of the Santa Barbarafield nor the wedge prospects are significantly sensi-tive to variation of initial TOC these traps are filledto spill in each scenario However the amount ofhydrocarbons charging the foreland basin (eg theOnado field) is reduced in the case of low initialTOC A high initial TOC only increased the modeledamount of surface seepage The sensitivity of the ini-tial HI is analogous to the sensitivity of the ini-tial TOC

In addition to the more classical uncertaintyand sensitivity analysis described we focused onquantitative analysis of the decompaction uncer-tainty when performing structural restoration Thisuncertainty is particular to the applied workflowof using structurally restored paleogeometries inBPSM Note that this uncertainty is significantlylower when using the backstripping method withthe optimization algorithms discussed One methodto analyze the adequacy of the decompactionmethod is comparison of the output rock matrixmass balance In the backstripping approach therock matrix mass is kept constant of a model cellwith a simulated decrease in porosity controlledby compaction laws and the corresponding reduc-tion of the layer thickness When using fixed pale-ogeometries from structural restoration the cellthickness is imposed However the decrease inporosity is calculated and the mass is adapted tofit the imposed cell thickness That means that therock matrix mass is constant only if ideal decom-paction has been applied

Figure 11A shows the percentage variation fromthe present-day matrix mass (100) of a given sourcerock cell through geologic time The dotted line rep-resents the ideal case with constant rock matrix massWe used two end-member decompaction scenarios todescribe the decompaction-related uncertainty that isintroduced when fixed paleogeometries replace back-stripping idealized hydrostatic decompaction assum-ing no overpressure and idealized lithostaticdecompaction assuming no effective stress (nodecompaction resulting in constant cell thicknessthrough time) In the latter case with decreasingporosity during compaction the rock matrix mass

1346 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

increases until it reaches the present-day mass Inother words rock mass matrix has been highly under-estimated in the past (in this particular case by morethan 20) This is a source rock so the kerogen massdecreases by the same factor This yields an underes-timation of more than 20 in TOC Note the rockmatrix mass is constant for time steps prior to23 Ma in which backstripping optimization couldbe performed toward the earliest paleogeometry Forthe second end-member scenario of idealized hydro-static decompaction the rock matrix mass is nearlyconstant which is a good decompaction adequacy

indicator for that given cell in the model Howeverthe so-measured adequacy changes throughout themodel In some areas the rock matrix mass has beenunderestimated by about 50 for nondecompactedpaleogeometries however in the case of hydrostaticdecompaction values generally range from 80 to120 of the present-day rock matrix mass Thatmeans in this particular case that hydrostatic decom-paction is more adequate (or less wrong) than nodecompaction at all The decompaction adequacycan be roughly correlated with pressure changesthrough geologic time

Figure 11 Time extractions for comparison of two end-member decompaction scenarios (see Figure 6 for location) (A) Rock matrixmass conservation through geologic time (B) Source rock burial depth through geologic time (C) Temperature and transformation ratiothrough geologic time

NEUMAIER ET AL 1347

Inadequate decompaction results in wrong depthfor stratigraphic horizons in the paleogeometriesFigure 11B shows the burial depth of the same inves-tigated source rock cell We can see that during thepassive margin phase there is a slight difference inpaleoburial depth for the two end-member scenariosDuring the compressional phase the differencechanges dramatically reaching more than 1000 m(3281 ft) in some places Toward present day bothcurves meet again because the input to the two resto-ration scenarios was the same (identical present-dayburial depth) This difference in compaction calcula-tion has severe implications on the modeled tempera-ture history and the kerogen-to-hydrocarbontransformation (Figure 11C) For an identical thermalframework in terms of modeling input parameterstemperature differences of up to 25degC (45degF) resultin TR differences of more than 20 in the early com-pressional phase before TR values meet again for thepresent-day situation

CONCLUSIONS

Structurally complex areas are attracting increasinginterest in hydrocarbon exploration Exploration riskcan be substantially narrowed by analyzing the his-tory of the basin and its petroleum systems based onstructural evolution We presented a further progressin technology that allows the integration of structuralrestoration and BPSM This combined analysis isseen as a way forward to improve modeling in thrustbelts with stacked stratigraphic sequences Despitelimitations of nonoptimized decompaction whichare added to the general uncertainty related to theindividual methods we modeled the charge and sealhistory of the Monagas fold and thrust belt inVenezuela Calibrated porosity temperature andthermal maturity as well as basin-scale stress pro-vided the regional framework through geologic timein which the petroleum systems evolved such thatthe charge of the known Santa Barbara and Onadofields and other trends could be understood We dis-cussed seal integrity of a potential exploration pros-pect by analyzing structural restoration-derived faultdisplacement through geologic time and an advancedstress forward simulation The uncertainty of using

decompacted or nondecompacted structural restora-tion has been assessed It can have important influ-ence on rock matrix mass balance through geologictime paleogeometry and paleoburial depth andtherefore affect paleotemperatures and paleomatur-ities We conclude that decompaction assuminghydrostatic pressure is the best option in that particu-lar case

REFERENCES CITED

Alberdi M and E Lafargue 1993 Vertical variations of organicmatter content in Guayuta Group (Upper Cretaceous)interior mountain belt eastern Venezuela OrganicGeochemistry v 20 p 425ndash436 doi 1010160146-6380(93)90091-O

Athy L F 1930 Density porosity and compaction of sedimen-tary rocks AAPG Bulletin v 14 p 1ndash24

Aymard R L Pimentel P Eitz P Lopez A ChaouchJ Navarro J Mijares and J G Pereira 1990 Geologic inte-gration and evaluation of Northern Monagas EasternVenezuelan Basin in J Brooks ed Classic petroleum prov-inces The Geologic Society (London) Special Publicationv 50 p 37ndash53

Baur F M Di Benedetto T Fuchs C Lampe andS Sciamanna 2009 Integrating structural geology andpetroleum systems modelingmdashA pilot project from Boliviarsquosfold and thrust belt Marine and Petroleum Geology v 26no 4 p 573ndash579 doi 101016jmarpetgeo200901004

Baur F R Littke H Wielens C Lampe and T Fuchs 2010Basin modeling meets rift analysismdashA numerical modelingstudy from the Jeanne drsquoArc basin offshore NewfoundlandCanada Marine and Petroleum Geology v 27 p 585ndash599doi 101016jmarpetgeo200906003

Beha A R O Thomsen and R Littke 2008 Thermal historyhydrocarbon generation and migration in the Horn Grabenin the Danish North Sea A 2D basin modelling studyInternational Journal of Earth Sciences v 97 p 1087ndash1100doi 101007s00531-007-0247-2

Blumenstein I O B M Krooss R di Primio W RottkeE Muumlller C Westerlage and R Littke 2008Biodegradation in numerical basin modelling A case studyfrom the Gifhorn Trough N-Germany International Journalof Earth Sciences v 97 p 1115ndash1129 doi 101007s00531-007-0272-1

Brandes C A Astorga R Littke and J Winsemann 2008Basin modelling of the Limoacuten back-arc basin (Costa Rica)Burial history and temperature evolution of an island arc-related basin-system Basin Research v 20 p 119ndash142doi 101111bre200820issue-1

Carnevali J O 1988 El Furrial oil field northeasternVenezuela First giant in foreland fold and thrust belts ofwestern hemisphere AAPG Bulletin v 72 68 p

Chevalier Y 1987 Contribution a lrsquoetude geacuteologique de lafrontiereacute Sud-Est de la plaque Caraibe Les zones internesde la chaine Sud-Caraibe sur le transect ile de Margarita-

1348 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

Apart from the Onado oil field only minor accu-mulations are modeled in the southern Neogene syn-orogenic sequence The proportion of gas is higherthan in the Furrial trend The accumulations at thesouthern model edge are a modeling artifact repre-senting hydrocarbons (mainly oil) that in reality arethought to have charged the southern foreland basin(from mid to late Miocene times to present day)However parts of the migrated oil and gas could havebeen trapped in the Upper Cretaceous San JuanFormation south of the horst structure

The potential for conventional hydrocarbon accu-mulations in the allochthon is very limited by theimportant fracturing that occurred during the upliftperiod The basal sandstones of the Morochito piggy-back basin north of the Pirital thrust are not chargedin our 2D model because of the absence of structuraltraps Nevertheless there is a potential for strati-graphic traps to have been charged as migrationthrough the basin occurred However seal strengthmight be critical at this low-compaction stage Thesame applies for the Las Piedras Formation We thinkthere might be potential for shale oil because theQuerecual Formation is at shallow depths but showsmaturity within the oil window From the thermalmaturity point of view any potential Jurassic sourcerocks as discussed by Summa et al (2003) mightbe a gas source or shale gas target

UNCERTAINTY ANALYSIS

There is often significant uncertainty in basin geom-etry and depth of stratigraphic horizons based onuncertain seismic interpretation and velocity models(Kukla et al 2000 Brandes et al 2008 Baur et al2010) Restoring of present-day geometries to obtainpaleogeometries adds further uncertainty Additionaluncertainty is related to thermal boundary conditionssuch as paleoheat flow (Beha et al 2008) and togeochemical properties such as source rock character-istics and petroleum generation kinetics and calcula-tion of biodegradation and petroleum density(Blumenstein et al 2008) In addition to the porepressure and thermal sensitivity analysis justdescribed we varied initial TOC within a range of2 to 10 A variation of initial TOC obviously

directly scales the amount of initial kerogen availablefor transformation into hydrocarbons Neither thecharge volume nor composition of the Santa Barbarafield nor the wedge prospects are significantly sensi-tive to variation of initial TOC these traps are filledto spill in each scenario However the amount ofhydrocarbons charging the foreland basin (eg theOnado field) is reduced in the case of low initialTOC A high initial TOC only increased the modeledamount of surface seepage The sensitivity of the ini-tial HI is analogous to the sensitivity of the ini-tial TOC

In addition to the more classical uncertaintyand sensitivity analysis described we focused onquantitative analysis of the decompaction uncer-tainty when performing structural restoration Thisuncertainty is particular to the applied workflowof using structurally restored paleogeometries inBPSM Note that this uncertainty is significantlylower when using the backstripping method withthe optimization algorithms discussed One methodto analyze the adequacy of the decompactionmethod is comparison of the output rock matrixmass balance In the backstripping approach therock matrix mass is kept constant of a model cellwith a simulated decrease in porosity controlledby compaction laws and the corresponding reduc-tion of the layer thickness When using fixed pale-ogeometries from structural restoration the cellthickness is imposed However the decrease inporosity is calculated and the mass is adapted tofit the imposed cell thickness That means that therock matrix mass is constant only if ideal decom-paction has been applied

Figure 11A shows the percentage variation fromthe present-day matrix mass (100) of a given sourcerock cell through geologic time The dotted line rep-resents the ideal case with constant rock matrix massWe used two end-member decompaction scenarios todescribe the decompaction-related uncertainty that isintroduced when fixed paleogeometries replace back-stripping idealized hydrostatic decompaction assum-ing no overpressure and idealized lithostaticdecompaction assuming no effective stress (nodecompaction resulting in constant cell thicknessthrough time) In the latter case with decreasingporosity during compaction the rock matrix mass

1346 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

increases until it reaches the present-day mass Inother words rock mass matrix has been highly under-estimated in the past (in this particular case by morethan 20) This is a source rock so the kerogen massdecreases by the same factor This yields an underes-timation of more than 20 in TOC Note the rockmatrix mass is constant for time steps prior to23 Ma in which backstripping optimization couldbe performed toward the earliest paleogeometry Forthe second end-member scenario of idealized hydro-static decompaction the rock matrix mass is nearlyconstant which is a good decompaction adequacy

indicator for that given cell in the model Howeverthe so-measured adequacy changes throughout themodel In some areas the rock matrix mass has beenunderestimated by about 50 for nondecompactedpaleogeometries however in the case of hydrostaticdecompaction values generally range from 80 to120 of the present-day rock matrix mass Thatmeans in this particular case that hydrostatic decom-paction is more adequate (or less wrong) than nodecompaction at all The decompaction adequacycan be roughly correlated with pressure changesthrough geologic time

Figure 11 Time extractions for comparison of two end-member decompaction scenarios (see Figure 6 for location) (A) Rock matrixmass conservation through geologic time (B) Source rock burial depth through geologic time (C) Temperature and transformation ratiothrough geologic time

NEUMAIER ET AL 1347

Inadequate decompaction results in wrong depthfor stratigraphic horizons in the paleogeometriesFigure 11B shows the burial depth of the same inves-tigated source rock cell We can see that during thepassive margin phase there is a slight difference inpaleoburial depth for the two end-member scenariosDuring the compressional phase the differencechanges dramatically reaching more than 1000 m(3281 ft) in some places Toward present day bothcurves meet again because the input to the two resto-ration scenarios was the same (identical present-dayburial depth) This difference in compaction calcula-tion has severe implications on the modeled tempera-ture history and the kerogen-to-hydrocarbontransformation (Figure 11C) For an identical thermalframework in terms of modeling input parameterstemperature differences of up to 25degC (45degF) resultin TR differences of more than 20 in the early com-pressional phase before TR values meet again for thepresent-day situation

CONCLUSIONS

Structurally complex areas are attracting increasinginterest in hydrocarbon exploration Exploration riskcan be substantially narrowed by analyzing the his-tory of the basin and its petroleum systems based onstructural evolution We presented a further progressin technology that allows the integration of structuralrestoration and BPSM This combined analysis isseen as a way forward to improve modeling in thrustbelts with stacked stratigraphic sequences Despitelimitations of nonoptimized decompaction whichare added to the general uncertainty related to theindividual methods we modeled the charge and sealhistory of the Monagas fold and thrust belt inVenezuela Calibrated porosity temperature andthermal maturity as well as basin-scale stress pro-vided the regional framework through geologic timein which the petroleum systems evolved such thatthe charge of the known Santa Barbara and Onadofields and other trends could be understood We dis-cussed seal integrity of a potential exploration pros-pect by analyzing structural restoration-derived faultdisplacement through geologic time and an advancedstress forward simulation The uncertainty of using

decompacted or nondecompacted structural restora-tion has been assessed It can have important influ-ence on rock matrix mass balance through geologictime paleogeometry and paleoburial depth andtherefore affect paleotemperatures and paleomatur-ities We conclude that decompaction assuminghydrostatic pressure is the best option in that particu-lar case

REFERENCES CITED

Alberdi M and E Lafargue 1993 Vertical variations of organicmatter content in Guayuta Group (Upper Cretaceous)interior mountain belt eastern Venezuela OrganicGeochemistry v 20 p 425ndash436 doi 1010160146-6380(93)90091-O

Athy L F 1930 Density porosity and compaction of sedimen-tary rocks AAPG Bulletin v 14 p 1ndash24

Aymard R L Pimentel P Eitz P Lopez A ChaouchJ Navarro J Mijares and J G Pereira 1990 Geologic inte-gration and evaluation of Northern Monagas EasternVenezuelan Basin in J Brooks ed Classic petroleum prov-inces The Geologic Society (London) Special Publicationv 50 p 37ndash53

Baur F M Di Benedetto T Fuchs C Lampe andS Sciamanna 2009 Integrating structural geology andpetroleum systems modelingmdashA pilot project from Boliviarsquosfold and thrust belt Marine and Petroleum Geology v 26no 4 p 573ndash579 doi 101016jmarpetgeo200901004

Baur F R Littke H Wielens C Lampe and T Fuchs 2010Basin modeling meets rift analysismdashA numerical modelingstudy from the Jeanne drsquoArc basin offshore NewfoundlandCanada Marine and Petroleum Geology v 27 p 585ndash599doi 101016jmarpetgeo200906003

Beha A R O Thomsen and R Littke 2008 Thermal historyhydrocarbon generation and migration in the Horn Grabenin the Danish North Sea A 2D basin modelling studyInternational Journal of Earth Sciences v 97 p 1087ndash1100doi 101007s00531-007-0247-2

Blumenstein I O B M Krooss R di Primio W RottkeE Muumlller C Westerlage and R Littke 2008Biodegradation in numerical basin modelling A case studyfrom the Gifhorn Trough N-Germany International Journalof Earth Sciences v 97 p 1115ndash1129 doi 101007s00531-007-0272-1

Brandes C A Astorga R Littke and J Winsemann 2008Basin modelling of the Limoacuten back-arc basin (Costa Rica)Burial history and temperature evolution of an island arc-related basin-system Basin Research v 20 p 119ndash142doi 101111bre200820issue-1

Carnevali J O 1988 El Furrial oil field northeasternVenezuela First giant in foreland fold and thrust belts ofwestern hemisphere AAPG Bulletin v 72 68 p

Chevalier Y 1987 Contribution a lrsquoetude geacuteologique de lafrontiereacute Sud-Est de la plaque Caraibe Les zones internesde la chaine Sud-Caraibe sur le transect ile de Margarita-

1348 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

increases until it reaches the present-day mass Inother words rock mass matrix has been highly under-estimated in the past (in this particular case by morethan 20) This is a source rock so the kerogen massdecreases by the same factor This yields an underes-timation of more than 20 in TOC Note the rockmatrix mass is constant for time steps prior to23 Ma in which backstripping optimization couldbe performed toward the earliest paleogeometry Forthe second end-member scenario of idealized hydro-static decompaction the rock matrix mass is nearlyconstant which is a good decompaction adequacy

indicator for that given cell in the model Howeverthe so-measured adequacy changes throughout themodel In some areas the rock matrix mass has beenunderestimated by about 50 for nondecompactedpaleogeometries however in the case of hydrostaticdecompaction values generally range from 80 to120 of the present-day rock matrix mass Thatmeans in this particular case that hydrostatic decom-paction is more adequate (or less wrong) than nodecompaction at all The decompaction adequacycan be roughly correlated with pressure changesthrough geologic time

Figure 11 Time extractions for comparison of two end-member decompaction scenarios (see Figure 6 for location) (A) Rock matrixmass conservation through geologic time (B) Source rock burial depth through geologic time (C) Temperature and transformation ratiothrough geologic time

NEUMAIER ET AL 1347

Inadequate decompaction results in wrong depthfor stratigraphic horizons in the paleogeometriesFigure 11B shows the burial depth of the same inves-tigated source rock cell We can see that during thepassive margin phase there is a slight difference inpaleoburial depth for the two end-member scenariosDuring the compressional phase the differencechanges dramatically reaching more than 1000 m(3281 ft) in some places Toward present day bothcurves meet again because the input to the two resto-ration scenarios was the same (identical present-dayburial depth) This difference in compaction calcula-tion has severe implications on the modeled tempera-ture history and the kerogen-to-hydrocarbontransformation (Figure 11C) For an identical thermalframework in terms of modeling input parameterstemperature differences of up to 25degC (45degF) resultin TR differences of more than 20 in the early com-pressional phase before TR values meet again for thepresent-day situation

CONCLUSIONS

Structurally complex areas are attracting increasinginterest in hydrocarbon exploration Exploration riskcan be substantially narrowed by analyzing the his-tory of the basin and its petroleum systems based onstructural evolution We presented a further progressin technology that allows the integration of structuralrestoration and BPSM This combined analysis isseen as a way forward to improve modeling in thrustbelts with stacked stratigraphic sequences Despitelimitations of nonoptimized decompaction whichare added to the general uncertainty related to theindividual methods we modeled the charge and sealhistory of the Monagas fold and thrust belt inVenezuela Calibrated porosity temperature andthermal maturity as well as basin-scale stress pro-vided the regional framework through geologic timein which the petroleum systems evolved such thatthe charge of the known Santa Barbara and Onadofields and other trends could be understood We dis-cussed seal integrity of a potential exploration pros-pect by analyzing structural restoration-derived faultdisplacement through geologic time and an advancedstress forward simulation The uncertainty of using

decompacted or nondecompacted structural restora-tion has been assessed It can have important influ-ence on rock matrix mass balance through geologictime paleogeometry and paleoburial depth andtherefore affect paleotemperatures and paleomatur-ities We conclude that decompaction assuminghydrostatic pressure is the best option in that particu-lar case

REFERENCES CITED

Alberdi M and E Lafargue 1993 Vertical variations of organicmatter content in Guayuta Group (Upper Cretaceous)interior mountain belt eastern Venezuela OrganicGeochemistry v 20 p 425ndash436 doi 1010160146-6380(93)90091-O

Athy L F 1930 Density porosity and compaction of sedimen-tary rocks AAPG Bulletin v 14 p 1ndash24

Aymard R L Pimentel P Eitz P Lopez A ChaouchJ Navarro J Mijares and J G Pereira 1990 Geologic inte-gration and evaluation of Northern Monagas EasternVenezuelan Basin in J Brooks ed Classic petroleum prov-inces The Geologic Society (London) Special Publicationv 50 p 37ndash53

Baur F M Di Benedetto T Fuchs C Lampe andS Sciamanna 2009 Integrating structural geology andpetroleum systems modelingmdashA pilot project from Boliviarsquosfold and thrust belt Marine and Petroleum Geology v 26no 4 p 573ndash579 doi 101016jmarpetgeo200901004

Baur F R Littke H Wielens C Lampe and T Fuchs 2010Basin modeling meets rift analysismdashA numerical modelingstudy from the Jeanne drsquoArc basin offshore NewfoundlandCanada Marine and Petroleum Geology v 27 p 585ndash599doi 101016jmarpetgeo200906003

Beha A R O Thomsen and R Littke 2008 Thermal historyhydrocarbon generation and migration in the Horn Grabenin the Danish North Sea A 2D basin modelling studyInternational Journal of Earth Sciences v 97 p 1087ndash1100doi 101007s00531-007-0247-2

Blumenstein I O B M Krooss R di Primio W RottkeE Muumlller C Westerlage and R Littke 2008Biodegradation in numerical basin modelling A case studyfrom the Gifhorn Trough N-Germany International Journalof Earth Sciences v 97 p 1115ndash1129 doi 101007s00531-007-0272-1

Brandes C A Astorga R Littke and J Winsemann 2008Basin modelling of the Limoacuten back-arc basin (Costa Rica)Burial history and temperature evolution of an island arc-related basin-system Basin Research v 20 p 119ndash142doi 101111bre200820issue-1

Carnevali J O 1988 El Furrial oil field northeasternVenezuela First giant in foreland fold and thrust belts ofwestern hemisphere AAPG Bulletin v 72 68 p

Chevalier Y 1987 Contribution a lrsquoetude geacuteologique de lafrontiereacute Sud-Est de la plaque Caraibe Les zones internesde la chaine Sud-Caraibe sur le transect ile de Margarita-

1348 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

Inadequate decompaction results in wrong depthfor stratigraphic horizons in the paleogeometriesFigure 11B shows the burial depth of the same inves-tigated source rock cell We can see that during thepassive margin phase there is a slight difference inpaleoburial depth for the two end-member scenariosDuring the compressional phase the differencechanges dramatically reaching more than 1000 m(3281 ft) in some places Toward present day bothcurves meet again because the input to the two resto-ration scenarios was the same (identical present-dayburial depth) This difference in compaction calcula-tion has severe implications on the modeled tempera-ture history and the kerogen-to-hydrocarbontransformation (Figure 11C) For an identical thermalframework in terms of modeling input parameterstemperature differences of up to 25degC (45degF) resultin TR differences of more than 20 in the early com-pressional phase before TR values meet again for thepresent-day situation

CONCLUSIONS

Structurally complex areas are attracting increasinginterest in hydrocarbon exploration Exploration riskcan be substantially narrowed by analyzing the his-tory of the basin and its petroleum systems based onstructural evolution We presented a further progressin technology that allows the integration of structuralrestoration and BPSM This combined analysis isseen as a way forward to improve modeling in thrustbelts with stacked stratigraphic sequences Despitelimitations of nonoptimized decompaction whichare added to the general uncertainty related to theindividual methods we modeled the charge and sealhistory of the Monagas fold and thrust belt inVenezuela Calibrated porosity temperature andthermal maturity as well as basin-scale stress pro-vided the regional framework through geologic timein which the petroleum systems evolved such thatthe charge of the known Santa Barbara and Onadofields and other trends could be understood We dis-cussed seal integrity of a potential exploration pros-pect by analyzing structural restoration-derived faultdisplacement through geologic time and an advancedstress forward simulation The uncertainty of using

decompacted or nondecompacted structural restora-tion has been assessed It can have important influ-ence on rock matrix mass balance through geologictime paleogeometry and paleoburial depth andtherefore affect paleotemperatures and paleomatur-ities We conclude that decompaction assuminghydrostatic pressure is the best option in that particu-lar case

REFERENCES CITED

Alberdi M and E Lafargue 1993 Vertical variations of organicmatter content in Guayuta Group (Upper Cretaceous)interior mountain belt eastern Venezuela OrganicGeochemistry v 20 p 425ndash436 doi 1010160146-6380(93)90091-O

Athy L F 1930 Density porosity and compaction of sedimen-tary rocks AAPG Bulletin v 14 p 1ndash24

Aymard R L Pimentel P Eitz P Lopez A ChaouchJ Navarro J Mijares and J G Pereira 1990 Geologic inte-gration and evaluation of Northern Monagas EasternVenezuelan Basin in J Brooks ed Classic petroleum prov-inces The Geologic Society (London) Special Publicationv 50 p 37ndash53

Baur F M Di Benedetto T Fuchs C Lampe andS Sciamanna 2009 Integrating structural geology andpetroleum systems modelingmdashA pilot project from Boliviarsquosfold and thrust belt Marine and Petroleum Geology v 26no 4 p 573ndash579 doi 101016jmarpetgeo200901004

Baur F R Littke H Wielens C Lampe and T Fuchs 2010Basin modeling meets rift analysismdashA numerical modelingstudy from the Jeanne drsquoArc basin offshore NewfoundlandCanada Marine and Petroleum Geology v 27 p 585ndash599doi 101016jmarpetgeo200906003

Beha A R O Thomsen and R Littke 2008 Thermal historyhydrocarbon generation and migration in the Horn Grabenin the Danish North Sea A 2D basin modelling studyInternational Journal of Earth Sciences v 97 p 1087ndash1100doi 101007s00531-007-0247-2

Blumenstein I O B M Krooss R di Primio W RottkeE Muumlller C Westerlage and R Littke 2008Biodegradation in numerical basin modelling A case studyfrom the Gifhorn Trough N-Germany International Journalof Earth Sciences v 97 p 1115ndash1129 doi 101007s00531-007-0272-1

Brandes C A Astorga R Littke and J Winsemann 2008Basin modelling of the Limoacuten back-arc basin (Costa Rica)Burial history and temperature evolution of an island arc-related basin-system Basin Research v 20 p 119ndash142doi 101111bre200820issue-1

Carnevali J O 1988 El Furrial oil field northeasternVenezuela First giant in foreland fold and thrust belts ofwestern hemisphere AAPG Bulletin v 72 68 p

Chevalier Y 1987 Contribution a lrsquoetude geacuteologique de lafrontiereacute Sud-Est de la plaque Caraibe Les zones internesde la chaine Sud-Caraibe sur le transect ile de Margarita-

1348 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela

Peninsule drsquo Araya (Venezuela) PhD Universiteacute deBretagne Occidentale Brest 494 p (in French)

Erlich R N and S F Barrett 1992 Petroleum geology of theEastern Venezuela foreland basin in R W Macqueen andD A Leckie eds Foreland basins and fold belts AAPGMemoir 55 p 341ndash362

Gallango O M Escandon M Alberdi F Parnaud and J-CPascual 1992 Hydrodynamism crude oil distribution andgeochemistry of the stratigraphic column in a transect ofthe eastern Venezuela basin (abs) Annual meeting of theGeological Society of America Cincinnati October 26ndash29Abstracts with Programs A214 p

Gallango O and F Parnaud 1995 Two-dimensional computermodeling of oil generation and migration in a transect ofthe Eastern Venezuela basin in A J Tankard R Suaacuterezand H J Welsink eds Petroleum Basins of SouthAmerica AAPG Memoir 62 p 727ndash740

Hantschel T and A Kauerauf 2009 Fundamentals of Basin andPetroleum Systems Modeling Berlin Springer-Verlag 476 p

Hantschel T B Wygrala M Fuecker and A Neber 2012Modeling basin-scale geomechanics through geologic time(abs) International Petroleum Technology ConferenceBangkok

Hedberg J D 1950 Geology of the eastern Venezuela basin(Anzoategui-Monagas-Sucre-eastern Guarico portion)Geological Society of America Bulletin v 61 p 1173ndash1216doi 1011300016-7606(1950)61[1173GOTEVB]20CO2

James K H 1990 The Venezuelan hydrocarbon habitat inJ Brooks ed Classic petroleum provinces The GeologicSociety (London) Special Publication v 50 p 9ndash35

James K H 2000a The Venezuelan hydrocarbon habitat part1 Tectonics structure paleogeography and source rocksJournal of Petroleum Geology v 23 no 1 p 5ndash53 doi101111jpg200023issue-1

James K H 2000b The Venezuelan hydrocarbon habitatpart 2 Hydrocarbon occurrences and generated accumulatedvolumes Journal of Petroleum Geology v 23 no 2p 133ndash164

Krause H and K H James 1989 Hydrocarbon resources ofVenezuela Their source rock and structural habitat inG E Ericksen M T Canas-Pinochet and J A Reinemundeds Geology of the Andes and its relation to hydrocarbonand mineral resources Circum-Pacific Council for Energyand Mineral Resources Earth Sciences Series v 11p 405ndash414

Kukla P A S C Edwards and K K Reimann 2000 3Dpetroleum systems modelling Bridging the gap betweenbasin and reservoir scale modelling-examples from offshoreWestern Australia Extended Abstract 4258 AAPGInternational Conference October 15ndash18 2000 Baliaccessed May 14 2014 wwwsearchanddiscoverycomabstractshtml2000intlabstracts219

Locke B D and J I Garver 2005 Thermal evolution of theeastern Serrania del Interior foreland fold and thrust beltnortheastern Venezuela based on apatite fission track analy-sis in H G AveacuteLallemant and V Baker Sisson edsCaribbeanndashSouth American plate interactions VenezuelaThe Geologic Society of America Special Paper v 394p 315ndash328

Maerten L and F Maerten 2006 Chronologic modeling offaulted and fractured reservoirs using geomechanicallybased restoration Technique and industry applicationsAAPG Bulletin v 90 no 8 p 1201ndash1226 doi 10130602240605116

Magoon L B and W G Dow 1994 The petroleumsystem in L B Magoon andW G Dow eds The petroleumsystemmdashfrom source to trap AAPG Memoir 60 p 3ndash24

Parnaud F Y Gou J -C Pascual I Truskowski O GallangoH Passalacqua and F Roure 1995 Petroleum geologyof the central part of the Eastern Venezuela basin in A JTankard R Suaacuterez-Soruco and H J Welsink edsPetroleum basins of South America AAPG Memoir 62p 741ndash756

Parra M G J Sanchez L Montilla O J Guzman J Namsonand M I Jacome 2011 The Monagas fold-thrust belt ofeastern Venezuela Marine and Petroleum Geology v 28no 1 p 70ndash80 doi 101016jmarpetgeo201001001

Passalacqua H F Fernandez Y Gou and F Roure 1995 Deeparchitecture and strain partitioning in the eastern VenezuelanRanges in A J Tankard R Suaacuterez-Soruco and H JWelsink eds South American basins AAPG Memoir 62p 667ndash679

Pepper A S and P J Corvi 1995 Simple kinetic models ofpetroleum formation Part I oil and gas from KerogenMarine and Petroleum Geology Geological Society v 12no 3 p 291ndash319 doi 1010160264-8172(95)98381-E

Perez de Armas J 2005 Tectonic and thermal history of thewestern Serrania del Interior foreland fold and thrust beltand Guarico basin north-central Venezuela Implicationsof new apatite fission-track analysis and seismic interpreta-tion in H G Aveacute Lallemant and V Baker Sisson edsCaribbeanndashSouth American Plate interactions VenezuelaThe Geological Society of America Special Paper v 394p 271ndash314

Pindell J L and K D Tabbutt 1995 Mesozoic-CenozoicAndean paleogeography and regional controls on hydrocar-bon systems in A J Tankard R Suaacuterez-Soruco and H JWelsink eds Petroleum basins of South America AAPGMemoir 62 p 101ndash128

Prieto R and G Valdes 1992 El Furrial oil field A new giantin an old basin in M T Halbouty ed Giant oil and gasfields of the decade 1978ndash1988 AAPG Special VolumesM 54 p 155ndash161

Roure F N Bordas-Lefloch J Toro C Aubourg NGuilhaumou E Hernandez S Lecornec-Lance C RiveroP Robion and W Sassi 2003 Petroleum systems and reser-voir appraisal in the sub-Andean basins (eastern Venezuelaand eastern Colombian foothills) in C Bartolini R TBuffler and J Blickwede eds The Circum-Gulf ofMexico and the Caribbean Hydrocarbon habitats basin for-mation and plate tectonics AAPG Memoir 79 p 750ndash775

Roure F R Swennen F Schneider J L Faure H Ferket NGuilhaumou K Osadetz P Robion and V Vandeginste2005 Incidence and importance of tectonics andnatural fluid migration on reservoir evolution in forelandfold-and-thrust belts Oil and Gas Science and Technology-Revue de lrsquoIFP v 60 no 1 p 67ndash106 doi 102516ogst2005006

NEUMAIER ET AL 1349

Schneider F 2003 Basin modeling in complex area Examplesfrom eastern Venezuelan and Canadian foothills Oil andGas Science and Technology-Revue de lrsquoIFP v 58 no 2p 313ndash324 doi 102516ogst2003019

Schneider F M Pagel and E Hernandez 2004 Basin modelingin a complex area Example from the Eastern Venezuelanfoothills in R Swennen F Roure and J W Granath edsDeformation fluid flow and reservoir appraisal in forelandfold and thrust belts AAPG Hedberg Series no 1p 357ndash369

Summa L L E D Goodman M Richardson I O Norton andA R Green 2003 Hydrocarbon systems of NortheasternVenezuela Plate through molecular scale-analysis of thegenesis and evolution of the Eastern VenezuelaBasin Marine and Petroleum Geology v 20 nos 3ndash4p 323ndash349 doi 101016S0264-8172(03)00040-0

Sweeney J J and A K Burnham 1990 Evaluation of a simplemodel of vitrinite reflectance based on chemical kineticsAAPG Bulletin v 74 no 10 p 1559ndash1570

Talukdar S O Gallango and A Ruggiero 1988 Generationand migration of oil in the Maturin subbasin easternVenezuelan basin in L Matavelli and L Novelli edsAdvances in organic geochemistry Organic Geochemistryv 13 p 537ndash547

Terzaghi K 1923 Die Berechnung der Durchlaumlssi-gkeitsziffer des Tones im Verlauf der hydro-dynamischen Spannungserscheinungen SitzungsberAkademieWissenschaft Wien Math-naturwissenschaftKlasseIIa Nr 132 p 125ndash138 (in German)

Wilkinson D 1984 Percolation model of immiscible displace-ment in the presence of buoyancy forces Physical Review Av 30 no 1 p 520ndash531 doi 101103PhysRevA30520

Wygrala B P 1989 Integrated study of an oil field in thesouthern Po basin northern Italy ForschungszentrumJuumllich Reports p 2313

Young G A 1978 Potential resources in the eastern VenezuelaBasin Boletin Informativo AsociacionVenezolano deGeologia Mineria y Petroleo v 20 p 1ndash38

1350 Integrated Charge and Seal Assessment Monagas Fold and Thrust Belt Venezuela