journal of volcanology and geothermal researchfolk.uio.no/hensven/mazzini_jvgr_11.pdf · 68 a....

Journal of Volcanology and Geothermal Research 205 (2011) 67–83

Contents lists available at ScienceDirect

Journal of Volcanology and Geothermal Research

j ourna l homepage: www.e lsev ie r.com/ locate / jvo lgeores

Fluid origin, gas fluxes and plumbing system in the sediment-hosted Salton SeaGeothermal System (California, USA)

Adriano Mazzini a,⁎, Henrik Svensen a, Giuseppe Etiope b, Nathan Onderdonk c, David Banks d

a Physics of Geological Processes, University of Oslo, Sem Sælandsvei 24, Box 1048, 0316 Oslo, Norwayb Istituto Nazionale di Geofisica e Vulcanologia, Sezione Roma 2, Via V. Murata, 605, 00143, Italyc Department of Geological Sciences, California State University, Long Beach, 1250 Bellflower Blvd., Long Beach, CA 90840, USAd School of Earth and Environment, University of Leeds, Leeds, LS2 9JT, UK

⁎ Corresponding author.E-mail address: [email protected] (A. Mazz

0377-0273/$ – see front matter © 2011 Elsevier B.V. Aldoi:10.1016/j.jvolgeores.2011.05.008

a b s t r a c t
a r t i c l e i n f o
Article history:Received 12 January 2011Accepted 31 May 2011Available online 12 June 2011

Keywords:Salton Sea Geothermal Systemhydrothermal seepsgas and water geochemistryflux measurementsmantle

The Salton Sea Geothermal System (California) is an easily accessible setting for investigating the interactionsof biotic and abiogenic geochemical processes in sediment-hosted hydrothermal systems. We present newtemperature data and the molecular and isotopic composition of fluids seeping at the Davis-Schrimpf seepfield during 2003–2008. Additionally, we show the first flux data for CO2 and CH4 released throughout thefield from focused vents and diffuse soil degassing.The emitted gases are dominated by CO2 (~98%) and CH4 (~1.5%). By combining δ13CCO2 (as low as −5.4‰)and δ13CCH4 (−32‰ to−17.6‰) with 3He/4He (R/RaN6) and δDCH4 values (−216‰ to−150‰), we suggest,in contrast to previous studies, that CO2 may have a significant Sub-Continental Mantle source, with minimalcrustal contamination, and CH4 seems to be a mixture of high temperature pyrolitic (thermogenic) andabiogenic gas.Water seeps show that δD and δ18O increase proportionally with salinity (Total Dissolved Solids in g/L)ranging from 1–3 g/L (gryphons) to 145 g/L (hypersaline pools). In agreement with elemental analyses, theisotopic composition of the waters indicate a meteoric origin, modified by surface evaporation, with little orno evidence of deep fossil or magmatic components. Very high Cl/Br (N3,000) measured at many seepingwaters suggests that increased salinities result from dissolution of halite crusts near the seep sites.Gas flux measurements from 91 vents (pools and gryphons) give a conservative estimate of ~2,100 kg of CO2

and 11.5 kg of CH4 emitted per day. In addition soil degassing measured at 81 stations (20x20 m grid over51,000 m2) revealed that 7,310 kg/d CO2 and 33 kg/d CH4 are pervasively released to the atmosphere. Theseresults emphasise that diffuse gas emission from soil can be dominant (~75%) even in hydrothermal systemswith large and vigorous gas venting. Sediment-hosted hydrothermal systems may represent an intermediateclass of geologic methane sources for the atmosphere, with emission factors lower than those of sedimentaryseepage in petroleum basins but higher than those of traditional geothermal-volcanic systems; on a globalscale they may significantly contribute to the atmospheric methane budget.

ini).

l rights reserved.

© 2011 Elsevier B.V. All rights reserved.

1. Introduction

Sediment-hosted hydrothermal systems occur in a range ofgeological settings, especially in active rift basins both onshore andoffshore (e.g. Helgeson, 1968; Von Damm et al., 1985; Simoneit, 1988;Jamtveit et al., 2004; Zarate-del Valle and Simoneit, 2005). The fluidsinvolved may include mixtures of biotic gases, due to rapid thermalmaturation of kerogens, with possible oil generation (e.g. Welhan andLupton, 1987) and abiogenic gases derived by late- or post-magmaticpolymerization synthesis and Fischer-Tropsch Type (FTT) reactions

(e.g. Simoneit, 1988; Potter et al., 2004) or from magma degassing(e.g. Werner et al., 2008).

A key example of this type of setting is the Guaymas Basin in theGulf of California (e.g. Simoneit and Galimov, 1984; Welhan andLupton, 1987), where shallow sill intrusions associated with the EastPacific Rise have led to hydrothermal fluid circulation, rapidmaturation of organic matter, oil generation, and seeps at the seafloor. Following this trend further north in the Salton Trough (Fig. 1),sedimentary rocks are deposited in a pull-apart basin intruded byQuaternary basalt and rhyolite bodies that provide the heatcharacterizing the Salton Sea Geothermal System (SSGS) (e.g.,Helgeson, 1968; Elders et al., 1972; Younker et al., 1982).

Although the hydrothermal system has been thoroughly investi-gated (e.g. numerous confidential prospects for geothermal energyproduction, drilling site “State 2.14”, Elders and Sass, 1988; Fig. 1),

http://dx.doi.org/10.1016/j.jvolgeores.2011.05.008

mailto:[email protected]

http://dx.doi.org/10.1016/j.jvolgeores.2011.05.008

http://www.sciencedirect.com/science/journal/03770273

Wister field

Mullet Island

State 2-14

DSF

Red Hill

Power plants

B

A

BA 1 km

C

50 m

North pool

East gryphonWest pool

West gryphon

Ditch120m south

Dav

is R

d.

N33 12'6.12''

N33 11'58.92''

W11

5 34

'43.

5''

W11

5 34

'38.

1''

Artificial lake

Sulphate field

Central gryphon

oil pool

Central pool

East pool

South gryphon

Central East gryphon

Fig. 1. (A) Location map of the south eastern Salton Sea. The main known faults (red lines) and seepages (black dots) are marked as well as the intensity of the thermal-gradientanomaly that has the highest peaks at warmer colours. Compiled after Lynch and Hudnut (2008); Hulen et al. (2002). (B) Google Earth image of the SE Salton Sea where theDavis-Schrimpf seep field (DSF), the Wister field, power plants and rhyolite extrusions are present. Line A-B refers to the section sketched in Fig. 7; (C) overview of the Davis-Schrimpf seep field and the sampling stations.

68 A. Mazzini et al. / Journal of Volcanology and Geothermal Research 205 (2011) 67–83

there are only few studies specifically dealing with the SSGS surfacefluid manifestations. Published works on gas and water geochemistryfrom the Davis-Schrimpf field and the State 2.14 hole suggest that thegas seeping in the SSGS originates from 1) active decarbonationreactions involving sedimentary carbonates (generating CO2 with aδ13C value close to the original carbonate composition, ~-5‰ VPDB),2) rapid high temperature maturation of organic matter (generatingCH4 with a δ13C of −32‰ VPDB); the seeping water instead seems tobe compositionally modified by surface evaporation, but a deep origincan not be excluded (Clayton et al., 1968; Muffler and White, 1968;

Williams andMckibben, 1989; Sturz et al., 1992; Svensen et al., 2007).The gas venting is mainly driven by the very steep geothermalgradients at the South-East end of the Salton Sea (e.g., Helgeson,1968) wheremost of the geothermal energy extraction occurs (Towseand Palmer, 1975). Lynch and Hudnut (2008) hypothesized that faultsassociated with the San Andreas fault system control the location ofthe seeps. The regional CO2 degassing from the Salton Trough is poorlyquantified; based on theoretical estimates of convective heat flowdata, Kerrick et al. (1995) suggested that about 109 mol/y of CO2 isreleased to the atmosphere.

69A. Mazzini et al. / Journal of Volcanology and Geothermal Research 205 (2011) 67–83

In this paper, we present the results of a four year investigationand sampling program of seep gas and water from the main seep fieldin the SSGS, the Davis-Schrimpf field (DSF), in order to give moreconstrained answers to the following main questions: 1) What is theultimate source of the seeping gas? 2) What is the origin of theseeping water? 3) How is the structure of the subsurface plumbingsystem affecting the surface seep temperatures? 4) Is the seepageconfined to the vents or is soil degassing a significant component ofthe total gas release? 5) What is the total emission of the twogreenhouse gases, CO2 and CH4, to the atmosphere?

2. The geothermal field

The SSGS is situated in the Salton Trough in Southern California, ahigh heat flow area with abundant surface manifestations ofhydrothermal seeps. The hydrothermal activity in the Salton Troughoccurs in a pull-apart setting where Quaternary magmatic intrusionsare responsible for high heat flow of up to 600 mW/m2 (e.g. Helgeson,1968; Robinson et al., 1976; Lachenbruch et al., 1985;Williams, 1997).Although the shallow stratigraphy of the basin is well known due toextensive hydrothermal energy exploration and drilling in the 1980's,in particular the Salton Sea Scientific Drilling (Elders and Sass, 1988),only thin sill intrusions have been intersected. Meta-sediments arepresent throughout the deepest drilled intervals (~3000 m) wherethe temperature reaches 350 °C, and the heat source is likely to berelated to deeper intrusions.

Inferred hot intrusions at depth produce contact metamorphism ofpredominantly fluvial and lacustrine sediments, and result intemperatures exceeding 350 °C at a depth of 1400 m in the highestheat flow zone (e.g. Helgeson, 1968; Muffler and White, 1969; Eldersand Sass, 1988; Haaland et al., 2000). Both the salinity andtemperature distribution of hydrothermal fluids from boreholessuggest that there is a fluid density interface between deep salinebrines and less saline shallow waters (Williams and Mckibben, 1989;Williams, 1997). The high salinity of the hydrothermal reservoir fluids(up to 150,000 ppm total dissolved solids) is attributed to dissolutionof evaporites, modified by high temperature fluid-rock interactions(Helgeson, 1968). Seeps and fluid expulsion features are likely aconsequence of the active contact metamorphism. Two seep fields arelocated close to the south-eastern shore line of the Salton Sea (theDavis-Schrimpf and the Wister fields), and seep activity is alignedalong an inferred fault zone (Lynch and Hudnut, 2008). This makesthe Salton Sea area a highly dynamic fluid expulsion zone, anattractive area for studying long term seep dynamics, and isfurthermore a relevant onshore analogue of the offshore WagnerConsag and Guaymas Basin hydrothermal seeps (e.g. Simoneit, 1985;Canet et al., 2010).

At our study area, the Davis-Schrimpf field (Fig. 1C), seeps cover anarea of about 22,000 m2. Here, gas, mud, and water are expelled fromgryphons, pools, and gas vents (Muffler and White, 1968; Sturz et al.,1992; Svensen et al., 2007). Although the DSF is commonly referred asa “mud volcano field”, we stress that both the surface expressions(lack of main craters or large scale mud breccias) and the systemdriving the seeps (hydrothermal CO2-dominated gases) are verydifferent from those typical of hydrocarbon-bearing sedimentaryvolcanism (e.g. Etiope et al., 2007b; Mazzini et al., 2009). For somespecific vents releasing gas, water andmudwe use, however, the term“gryphons” in analogy to similar morphologic manifestations occur-ring in actual mud volcanoes.

3. Fieldwork and methods

Four fieldwork campaigns in December 2003, 2006, 2007, and2008 included extensive sampling of the seeping fluids in the DSF. Atotal of eleven gas seeps and seventeen water seeps were sampled forin situ and laboratory analyses.

3.1. Water density and salinity

The density of expelled mud and waters was measured with acommercial electronic scale, with accuracy better than~2% for therelevant mass of the measured samples.

Water salinity (Total Dissolved Solids) was measured in situ usinga VWR EC300 conductivity sensor; pH was measured by L.T. LutronYK-2001 pH meter.

3.2. Temperature measurements

Water temperature was measured with a hand held TFX 392 SK-5thermometer with a precision of 0.1 °C. Temperature monitoringmethodology and results are described in Svensen et al. (2009),additionally here we report new time series monitoring over asulphate-rich area in the south east of the field where a thermometerwas positioned 0.1 m underground through the period 06 December2007 to 05 April 2008. A night survey for thermal imaging and filmingof different vents was carried out using a Flir System P640HS camerain December 2007.

3.3. Gas Geochemistry

Gases expelled from the seeps in 2003, 2006, and 2007 weresampled into brine-filled glass bottles that were subsequentlyhermetically sealed. Gases (N2, O2, CO2, C1-C4 hydrocarbons) wereanalyzed at the Institute for Energy Technology (IFE), Norway.0.1-1 ml of gas were sampled using a Gerstel MPS2 autosampler andloop injected into a splitt/splittless injector on a Hewlett Packard 5890Series II GC equipped with Molsieve and Poraplot Q columns, 1 flameionization detector (FID) and 2 thermal conductivity detector (TCD).Hydrocarbons were measured by FID. H2, CO2, CO, N2, H2S and O2/Arwere measured by TCD. The precision of the gas composition is betterthan 3% (1 σ) for all compounds.

The stable carbon isotopic compositions of methane and carbondioxide were determined by sampling aliquots with a Gerstel MPS2autosampler, then injected into a split/splittless injector on an Agilent7890 gas chromatograph with a Poraplot Q column, connected to aHorizon IRMS from NU-Instruments. After separation, the compo-nents were burnt to CO2 and water in a 1000 °C furnace over Cu/Ni/Pt.The water was removed by Nafion membrane separation. Repeatedanalyses of standards indicate a reproducibility of δ 13 C values betterthan 0.3‰ VPDB (2 sigma). The hydrogen isotope analysis of methanewere completed by sampling aliquots from the exetainers with aGCPal autosampler and injected into a split/splittless injector on aTrace GC2000 from Thermo, equipped with a Poraplot Q column,connected to a Delta+XP IRMS from Thermo. The components weredecomposed to H2 and cooked in a 1400 °C furnace. The internationalstandard NGS-2 and an in-house standard is used for checkingaccuracy and precision. Repeated analyses of standards indicate areproducibility of δ D values better than 5‰ PDB (2 sigma).

The gas samples collected in December 2008 were analyzed at theIsotech Labs Inc. (Illinois, USA) for C1-C6 hydrocarbons, He, H2, Ar, O2,CO2, N2, (Carle AGC 100–400 TCD-FID GC; detection limits of CO2, Ar,O2: 50 ppmv; N2: 100 ppmv; H2S: 300 ppmv; He H2: 10 ppmv; C1-C5alkanes: 5 ppmv; precision 2% (1 σ), and isotopic compositions δ13C1,δD1, δ13CO2, δ15N (Finnigan Delta Plus XL mass spectrometer,precision±0.1‰ (1 σ) for 13 C and±4‰ (1 σ) for 2H). The hydro-carbon components were separated by GC, and CH4 was thermallyconverted to CO2 and H2 in the pyrolysis furnace at 1450 °C. Theresulting H2 was introduced directly into the mass spectrometer.Results are reported in δ notation, as per mil (‰) deviation relative tothe VPDB (for carbon), VSMOW (for hydrogen) and atmosphere (fornitrogen) standards.

Samples for He analyses were collected in annealed copper tubeswhich were sealed in the field using a cold welding clamp, and


analyzed at the Scripps Institution of Oceanography (USA) usingextraction, purification and measurement protocols described in e.g.Shaw et al. (2003). The helium, neon and carbon dioxide aliquotswerecaptured in glass break-seals and transferred to a MAP noble gas massspectrometer and a manometric measurement/purification linerespectively. In the case of the He+Ne fraction, the gas was furtherpurified using a Ti-getter, charcoal trap (at liquid nitrogen tempera-ture) and a He-cooled refrigeration trap designed to separate He fromNe prior to inlet into the mass spectrometer. Aliquots of air were usedto calibrate the sensitivity and the mass fractionation. Followingpurification and manometric measurement, an aliquot of the CO2 wastransferred to a Thermo-Finnigan DeltaPlus XPmass spectrometer for13 C/12 C analysis. The results are given in the R/Ra notation, where R isthe sample 3He/4He ratio and Ra is the atmospheric 3He/4He ratio,1.39×10-6).

3.4. Water geochemistry

Water and mud from the seeps were collected in plastic vials andfiltered. Anions were analyzed using a Dionex DX-500 ion chromato-graph and major cations analysed with a Perkin-Elmer Optima ICP-OESand trace cations analysed with a Perkin-Elmer Elan DRC-e ICP-MS atthe School of Earth Sciences, University of Leeds (UK). The results arereported as parts per million (ppm) (Table 2A). For oxygen isotopicanalyses, H2Owas equilibratedwith CO2 at 25 °C for 24 hours. The δ18Ocomposition of CO2 will then reflect the isotope composition of theoriginal water sample. The impurities were separated from CO2 on aPoraplot GC column before on-line determination of δ18O using aFinnigan DeltaXP isotope mass spectrometer at IFE. The average valuefor GISP from IAEA is δ18O VSMOW=−24.80±0.10‰ (1 σ). “True” valueis −24.78±0.08‰. Before measuring the δD composition H2O wasreduced with Zn to H2 and ZnO in sealed, evacuated quartz vessels at900 °C. The δD composition was then determined by dual inlet, using aMicromass Optima isotope mass spectrometer. Average value for GISPfrom IAEA is δD VSMOW=−189.71±0.89‰ (1 σ). “True” value is−189.73±0.9‰.

3.5. Flux measurements

In December 2008, CO2 and CH4 fluxes from focused venting(macro-seepage) and diffuse soil degassing (mini-seepage, as definedin Spulber et al., 2010) were measured throughout the DSF. Gas fluxfrom 86 vents (pools or gryphons) was measured by volumetric fluxmeter techniques, time monitoring the funnelled expelled gas in a1.5 l calibrated vial. Flux measurements from 5 smaller and weakvents and in 81 soil degassing sites (invisible miniseepage) wereperformed with a closed-chamber system (West Systems srl, Italy)equippedwith portable CH4 and CO2 detectors andwireless data com-munication to a palm computer (Etiope et al., 2010). The CH4 sensorincludes semiconductor (range 0–2,000 ppmv; lower detection limit:1 ppmv; resolution: 1 ppmv), catalytic (range: 2,000 ppmv - 3% v/v),and thermal conductivity (3% to 100% v/v) detectors. The CO2 detectorhas a double beam infrared sensor (Licor) with an accuracy of 2%,precision±5 ppmv and a full scale range of 20,000 ppmv. The gas fluxis calculated through a linear regression of the gas concentrationbuild-up in the chamber. An accumulation time of ~10 minutesallowed us to detect fluxes down to 30 mg m-2 d-1. Shorter accumu-lation times were applied at locations where high fluxes rapidlysaturated the chamber. These soil degassing measurements coveredan area of ~20,000 m2 following a 20×20 m grid. Eight additionalmeasurements where made in order to extend the 20×20 m grid to80 m from each of the corners. The total area covered thus extends toover 51,000 m2. Gas flux data were elaborated with commercialkriging interpolation software in order to provide 3D maps and anestimate of the total gas miniseepage output. The sum of miniseepage

andmacro-seepage fluxes gives the total emission of CH4 and CO2 intothe atmosphere.

4. Results

4.1. Morphologies and seepage types

Gryphons and pools releasing a mixture of water, gas, and sedimentare commonly clustered in10–20 mdiameter calderas. The venting sitescan be divided into fourmain groups dependingon the temperature andthe density of the expelled water or mud: gryphons, oil pools, waterpools, and thermally anomalous dry zones (Fig. 2, Table A1).

a) Gryphons: conical features resulting from superimposed mud flowsthat commonly reach the height of 2–2.5 meters where the hot(up to 70 °C) dominatingmuddy component is forcefully erupted byconstant or intermittent surges of gas from a central crater.Gryphons can be isolated structures or grouped and are commonlysurrounded by subsiding calderas that host pools (Fig. 2A).

Pools are topographically negative structures that can also be foundas isolated features. Two types of pools can be distinguished.

b) Oil pools where gas is vented through greyish clay-dominatedmud. The grey mud (reaching temperatures up to 46.6 °C) is thesame erupted at gryphon sites suggesting that the oil pools are alsofed with sediments from deeper units. This mud is mixed with asignificant amount of oily components that commonly form afoamy layer floating on the surface of the pools (Fig. 2B).

c) Water pools where gas is vented through reddish-brownish waterdominated mud. The reddish colour of the mud suggests oxidationoccurring on the surface and thus that there is little (if any?)expulsion of clay fromdeeper strata. Their average temperatures arebetween 20 to 30 °C (Fig. 2C).

d) A thermally anomalousdryzone in the south-easternpart of thefieldis characterized by large patches of yellowish sulphate mineraliza-tions (blodite) (Fig. 2D). These features consist of circular drydepressions ranging in size between 1–2 meters. In 2003 a 0.5 mdeep trench was made into one of these circular zones and thestratigraphy compared to the background stratigraphy froma trenchmade 10 meters away Figs. A2 and A3. Locally nodular calcite andanhydrite was discovered in the mud. The surface temperature wasN60 °C but precise measurements are lacking. When investigatedthe following year, the thermal anomaly decreased in intensity andthe calcite and gypsum nodules had dissolved.

4.2. Density, temperature and pH

The density of the expelled water and mud mixtures from 18individual seeps range from 1.1 to 1.7 g/cm3 (Table 1). This is withinthe same range as the data from December 2003 (Svensen et al.,2007). The gryphons muds have densities generally higher than 1.4 g/cm3, while the pools release fluids with a density lower than 1.4 g/cm3

(Fig. A1A, Table 1). Fig. A1B shows the correlation between the hotterand denser gryphons and the colder and less viscous pools. The in situpH (Fig. A1C) varies between 5.6 and 6.8, which is a narrower rangecompared to the 2003 measurements (pH between 5.2 and 6.3). Inaddition to the differences in pH between the 2003 and 2006sampling, the seep field was drier in 2006, with lower water tablesand fewer individual seeps. In general the gryphons with highertemperature also have a higher pH (Fig. A1A).

Infrared imaging (Fig. 3) showed that adjacent seepage sites canhave broad variations in temperature even when the seeps are part ofthe same structure (e.g. Fig. 3A-D). Additionally it was noticed that thehot mud erupted from gryphons looses heat very rapidly after flowingalong the flanks of the structures.

BA

C D

Fig. 2. (A) Gryphons and caldera's pools. Gryphons commonly reach the height of 2–2.5 meters and erupt hot (up to ~70 °C) mud, gas and water. Most gryphons are grouped inclusters surrounded by subsiding calderas. These caldera structures contain cold ~20-30 °C water-dominated pools seeping gas; (B) Oily pool seeping mud, water, and gas at ~40 °C(diameter ~2.5 meters). The depth of the pool was measured as up to four meters; (C) Water dominated isolated cold (~20 °C) pool (i.e. north pool) reaching approximately threemeters in depth and a diameter of ~2 meters. Note the brownish colour compared to hotter oil pools and gryphons; (D) Two of the gentle depressions present in the sulphate fieldwhere blodite mineralization are distributed over the warm (up to 34 °C) soil and where a higher gas flux is present.


4.3. Water geochemistry

The field measurements of the water TDS showed a wide spread insolute content (Table 1), with the four measured gryphons havingvalues from 1.5 to 4.5 g/L whereas mud pools have salinities thatrange from 17.6 to 134 g/L. The field measurements were corrobo-rated with laboratory measurements of the same samples, and thediscrepancy between the two methods (cf. Tables 1 and 2A; the R2 ofthe correlation is 0.56 when excluding the high salinity sample) islikely due to difficulties analysing solutes in clay-rich water. Thehighest TDS, 134 g/L measured in the field, was recorded in a poolwith salt crust along the rims, in agreement with the onset of haliteprecipitation at ~145,000 ppm Cl (Fontes and Matray, 1993). Nearbysurface water reservoirs (artificial lake directly east of the seeps, dykewith running water, and the Alamo River) had measured chlorideconcentrations of 1386, 873, and 518 ppm Cl respectively, over-lapping the lower range of the gryphon waters (N768 ppm Cl)(Fig. 4A-B). All cations except Ba and Br show a positive correlationwith the Cl concentration. The ‘oil pool’ was more saline in 2006,71,042 ppm Cl, compared to 44,692 ppm Cl in 2003, and the Cl/Brratio increased from 3575 to 14,208 over the same period (cf. Svensenet al., 2007). The sulphate concentrations were high, with up to7549 ppm in the ‘oil pool’. Compared to the surface water reservoirs,most of the gryphons and pools have very high Cl/Br ratios (Fig. 4A).The highest value is from the pool water in contact with halite (Cl/Br=35,600), and even a relatively low chlorinity gryphon(Cl=15,658 ppm) has a ratio of 32,620. Note that the Cl/Br ratiosare considerably higher compared to the 2003 water data presentedby Svensen et al. (2007), where the Cl/Br in pools was between 1941and 5099.

The δ18O values of both the seep water and the nearby surfacewaters are in the range −15.1 to +0.7‰ VSMOW (Table 2A) and the

range in δD values is from −45.2 to −112.9‰. The pool samples aremore 18O-enriched than the gryphon waters and when plottedrelative to the global meteoric water line (Craig, 1957), all but onesample plots along a trend commonly attributed to evaporation(Fig. 4C). The Alamo River and the dykewater are among themost 18Odepleted, whereas the seeps and the nearby artificial lake waters are18O enriched. The seep water with the highest measured salinity isenriched in 18O (i.e. δ18ON0.7‰) and has a relatively high δD(−52.3‰), while the most 18O depleted values are from the lowsalinity gryphons (Fig. 4B-C). Note that the deep hydrothermalreservoir brines (Elders and Sass, 1988) are more 18O-enriched (δ18Oof 1.5-4.0‰) than any of the seep waters, and that most pool watersplot between the evaporation trend values and the reservoir brinevalues.

4.4. Gas geochemistry

The main gas component analysed in the seeps is CO2, rangingfrom 97.8 to 99.5 vol. %. Additionally, the seep gas contains CH4 (0.2-1.9 vol. %), N2 (b0.76 vol.%) and trace amounts of ethane, propane andbutane (Table 2B). Stable carbon isotopic compositions of CO2 and CH4

range from −5.4‰ to 0.4‰ and from −32.7‰ to −17.6‰respectively (Table 2B). Isotopic composition of N2 was determinedonly in one sample (δ15N=−0.52‰), as for ethane (δ13C2=−20‰).The molecular composition and the 13C of CO2 and CH4 (Fig. 5A) areindependent of the type of venting structure (gryphons or pools). TheHe isotopic ratio, R/RA, ranged from 6.08 to 6.64 (Table 2B).

4.5. Macroseepage flux at vent sites

Volumetric fluxes of CO2 and CH4 from gryphons and pools havebeen normalized to the average gas composition measured during the

Table 1Field measurements and sampling. *Measurements from Svensen et al. (2007).

Sample ID Structure Comment Level(cm)

Diam.(cm)

T(°C)

pH TDS(g/L)

Density(g/cm3)

CA03-3* Pool North pool 15.0CA03-5* Pool Oil pool 33.9CA06-38 Gryphon Splattering flank 68.2 6.2 4.5 1.6CA06-22 Gryphon East gryphon. 15 cm diameter gryphon close to oil pool 30 64.3 6.3 3.1 1.7CA06-32 Gryphon Big gryphon in westernmost caldera 120 70 58.5 6.3 1.6 1.7CA06-35 Gryphon Same structure sampled in CA06-34 135 40 37.0 6.0 1.5 1.7CA06-40 Gryphon Only gas seepage, but water inside -45 36.0CA06-28 Pool Double pool west of oil pool -50 40 35.0 6.2 33.4 1.3CA06-21 Pool Oil pool -65 70 34.6 5.9 36.4 1.7CA06-37 Gryphon Splatter cone 34.4CA06-36 Pool Small pool, 1.5×2 m -35 150 33.0 6.0 49.7 1.3CA06-27 Pool Double pool west of oil pool -50 40 32.2 6.2 39.0 1.2CA06-24 Pool West pool -70 40 30.8 6.1 19.5 1.3CA06-30 Pool Pool in westernmost caldera, grey, in triple pool -65 20 25.1 6.0 54.3 1.3CA06-31 Pool Pool in westernmost caldera, grey, forming three seepage sites -65 30 24.9 6.1 45.6 1.2CA06-23 Pool -28 50 23.3 6.1 23.2 1.1CA06-34 Pool Muddy, west end, possible oil present (?) -80 200 23.1 5.8 50.3 1.6CA06-26 Pool Pool north of oil pool -20 100 19.4 6.1 17.8 1.2CA06-25 Pool North pool with brownish water -67 200 19.1 6.2 17.6 1.2CA06-29 Pool Pool in westernmost caldera, brown, in triple pool -65 40 18.0 6.0 51.2 1.3CA06-39 Pool Structure present at the base of the gryphon sampled at CA06-38 60 130 14.1 5.8 33.2 1.1CA06-46 Water Salton Sea, tip of boat launching place, red hill, west side 13.7 8.3 6.0 1.0CA06-33 Pool Brown, rimmed with salt crusts -85 400 13.6 5.6 134.0 1.3CA06-44 Water Flowing water in ditch (1 m across) just south of seep field 12.6 7.9 2.9 1.0CA06-45 Water Alamo River, at bridge close to red hill marina 12.0 7.7 1.9 1.0CA06-41 Water Artificial lake next to duck blind 11.7 6.8 3.5 1.0CA07-01 Pool Oil pool 38.3CA07-02 Pool North pool 19.4CA07-03 Gryphon East Gryphon 54.3CA07-04 Gryphon West Gryphon 47.8CA07-05 Gryphon South Gryphon 55.2CA07-06 Gryphon Central Gryphon 62.2CA07-07 Pool Central pool 18.0CA07-08 Pool Artesan pool with multiple seeping points (2 km NW from the SD field) 16.3CA07-09 Pool Artesan pool with multiple seeping points (2 km NW from the SD field) 21.6CA08-01 Pool Oil pool 46.6CA08-02 Pool North pool 23.7CA08-03 Gryphon East Gryphon 45.5CA08-04 Gryphon West Gryphon 46.0CA08-05 Gryphon South Gryphon 36.3CA08-06 Gryphon Central Gryphon 56.3CA08-07 Pool Central pool 17.2CA08-08 Gryphon East central gryphon 55.4CA07-09 Pool West pool


different field campaigns (i.e. 98% CO2, 1.5% CH4). Gas outputs fromindividual vents ranged from 1.4 to 254 kg/day for CO2 and 0.01 to1.4 kg/day for CH4. Integrating all the measurements performed fromthe 86 vents, a total of 2,046 kg/day of CO2 and 11 kg/day of CH4 arereleased from the DSF. If we include the output from five weak vents(see Section 3.5) measured with the closed chamber (62 kg/day forCO2 and 0.5 kg/day for CH4), then the total macro-seepage outputwould be: 2,108 kg/day (~769 ton/year) of CO2 and 11.5 kg/day(~4 ton/year) of CH4 (Fig. 6 A-B).

Note that these values represent a conservative estimate as thetransition from macroseepage to miniseepage is gradual and someless obvious vents have been neglected both at gryphons and poolsites.

4.6. Diffuse gas flux from soil (miniseepage)

Fig. 6C-D shows the spatial distribution derived by kriginginterpolation of diffuse CO2 and CH4 soil degassing measured withthe closed chamber system. CO2 fluxes from soil ranged from 1 to3700 g m-2 d-1. While CO2 seepage was essentially detected at allsampling stations, this was not the case for CH4. The CH4 fluxes from

soil above the detection limit (~30 mg m-2 d-1, imposed by amaximum accumulation time of 10 minutes) were measured in 45%of locations (37 sites ranging from 30 to 8190 mg m-2 d-1). Measure-ments of soil degassing from the sulphate field rapidly saturated thechamber indicating a substantial flux that gradually decreases radiallyoutward from the dry thermal anomaly zones.

The closed chamber measurements also reveal a direct correlationbetween the CH4 and the CO2 fluxes (Fig. 6C-D-E). The average weightratio between the CO2 and CH4 flux is 228. This value is close to theaverage compositional CO2/CH4 weight ratio, 186 (ranging from 145to 231; Table 2B) measured in the laboratory. Therefore gas venting atgryphon or pool sites has roughly the same CO2 and CH4 proportionsof that released by soil degassing. Applying kriging interpolation tothe miniseepage fluxes gives a total emission of 7310 kg/day(~2,700 t/y) of CO2 and 33 kg/day (~12 t/y) of CH4.

4.7. Temperature monitoring

The temperature monitoring of the sulphate field revealedtemperatures varying from 34.1 °C in early December to a minimumof 30.2 °C in early February and rising to 33.9 °C at the end of

BA

C D

E F

Fig. 3. Infrared images (A, C, E) highlight the different temperatures of the seeping fluids even at closely spaced locations indicating a complex plumbing system in the nearsubsurface. (B, D) Isolated pools (ca. 1.5 m in diameter) where focused seepage of fluids occur at several locations. The images also highlight that different water levels are present atclosely spaced sites. (F) Isolated gryphon (~2 m high) with hot mud seeping from the top of the crater and flowing along the flanks of the structure. Although the temperature of themud is higher than 60 °C in the crater, once the gaseous phase is released to the atmosphere, the mud cools down very rapidly as it flows.

Table 2AComposition of waters at vent sites. All values in ppm. Isotopic composition is reported in per mil vs SMOV; nd: not determined.

Sample ID Structure Cl Br SO4 Cl/Br Ca Na K Mg Sr B Ba δ18O δD

CA06-21 Oil pool 71042 5.0 7549 14208 2093 44320 1323 1197 66 186 0.3 -5.4 -81.9CA06-39 Pool 11368 20.9 20 544 146 6860 364 136 26 62 0.4 -2.0 -55.3CA06-23 Pool 15658 0.5 441 32620 291 9555 590 110 41 63 0.4 -2.6 -73.1CA06-24 West pool 15086 31.2 905 483 351 8817 507 116 23 78 0.3 -2.9 -65.6CA06-25 North pool 13330 91.3 56 146 48 7970 408 127 18 61 1.7 -5.7 -74.3CA06-26 Pool 10361 2.5 322 4144 127 6875 402 106 29 65 0.4 -4.2 -72.8CA06-30 Pool 26712 2.5 1539 10685 629 16355 470 697 31 88 0.2 -4.0 -45.2CA06-33 Pool 178000 5.0 5965 35600 1169 100700 2843 7058 48 601 0.4 0.7 -52.3CA06-34 Pool 46818 240.0 1272 195 772 29390 766 1088 40 143 0.3 -4.2 -52.5CA06-38 Gryphon 4685 0.4 1134 12393 157 2967 174 50 8 18 0.8 -8.5 -76.1CA06-22 East gryphon 6923 1.5 475 4569 87 4320 228 68 16 28 0.5 -8.2 -86.3CA06-32 Gryphon 7758 2.5 20 3103 215 5160 264 122 23 33 0.2 -6.1 -72.1CA06-35 Gryphon 768 2.5 515 307 9 881 79 23 6 6.5 0.2 -15.1 -112.9CA06-41 Water 1386 21.9 1173 63 176 915 23 168 6 2.7 0.0 -4.2 -60.5CA06-44 Water 873 6.0 1266 146 231 672 15 148 6 2.4 0.1 -8.1 -81.7CA06-45 Water 518 0.3 916 1990 176 443 3 99 2 4.8 0.1 -10.4 -90.0CA06-46 Water 2371 2.6 2375 912 275 1848 36 263 8 5.4 0.1 -8.9 -82.8CA07-1 Oil pool nd nd nd nd nd nd nd nd nd nd nd -13.8 -90.5CA07-2 North pool nd nd nd nd nd nd nd nd nd nd nd -10.7 -80.4CA07-3 East Gryphon nd nd nd nd nd nd nd nd nd nd nd -10.2 -79.8


10

100

1000

10000

100000

0 50000 100000 150000 200000

Cl/B

r

Cl ppm

-16

-14

-12

-10

-8

-6

-4

-2

0

2

100 1000 10000 100000 1000000

18 δO

(V

SM

OW

)

Cl (ppm)

-120

-110

-100

-90

-80

-70

-60

-50

-40

-30

-16 -12 -8 -4 0 4 8

δD (

VS

MO

W)

18δ O (VSMOW)

Seawater evaporation trend

Hal

ite

‘oil pool’ 2006

‘oil pool’ 2003

2003 Gryphons

2006 Pools

2006 Gryphon

2006 Artificial Lake

2006 Dike

2006 Alamos River

2006 Salton Sea

2003 Pools

SSGS deep brine

18

Glo

bal M

eteo

ric W

ater

Lin

e (δ

D=8

*δO

+10)

Evap

orat

ion

2006 Pools

2006 Gryphon

2006 Artificial Lake

2006 Dike

2006 Alamos River

2006 Salton Sea

2007 Pools

2007 Gryphon

SSGS shallow brine

A

B

C

‘Oil pool’ dilution

‘North pool’ dilution

Fig. 4. Seep water and reference water geochemistry. A) Chloride and bromide datacompared to the evaporation trend for seawater (Fontes and Matray, 1993) and halitecomposition (Mccaffrey et al., 1987). Note the large range in Cl/Br ratios and that highCl/Br cannot be obtained by evaporation. B) δ18O and Cl of seep waters. Note that allseep waters shown here are enriched in Cl compared to the reference samples but thatthe δ18O values overlap. C) Hydrogen and oxygen isotope systematic of the studiedwaters together with the Global Meteoric Water line (Craig, 1957), the evaporationtrend for waters in the Salton Sea area and the hydrothermal reservoir brines of theSSGS (Mckibben et al., 1987). Note that our reference samples plot close to theevaporation trend, and that most of the pool seep waters are enriched in 18O. Thearrows show the geochemistry of the oil and north pool waters between sampling in2006 and 2007 (arrows pointing towards the 2007 samples). Meteoric dilution can becorrelated to heavy rainfall prior to the 2007 sampling campaign.


monitoring in early April. Although the total temperature variation isonly 4 °C, the trend reflects variations in air temperature with a10 days delay (Fig. A4). These limited temperature variations areremarkable considering that the air temperature had 40.5 °C range(i.e. -1.5 °C in mid January up to 39 °C in mid March).

5. Discussion

5.1. A revised interpretation of the gas origin

The large gas-geochemical database obtained in the four fieldcampaigns allows to better constrain the interpretation of the originof gas and to re-evaluate the conclusions suggested by previousstudies. The stable carbon isotopic composition of seeping CO2

(δ13CCO2 from −5.4 to +0.4‰, with a mean of −3.4‰) is in therange of typical CO2 derived from magmatic sources (−6 to −3‰;Kyser, 1986; Sheppard, 1986) and thermal decarbonation reactions(−5 to +4‰). The biotic CO2 contribution from organic matter insediments (kerogen decarboxilation typically from −10 to −25‰,e.g. Jenden et al., 1993) seems therefore to be negligible.

Svensen et al. (2007) suggested a CO2 origin from decarbonationreactions on the basis of the 2003 field data (δ13CCO2=−5.4‰) andearlier work based on gas geochemistry and petrography of themetamorphic sediments (Muffler and White, 1968; Cho et al., 1988;Shearer et al., 1988). It is known, however, that the carbon isotopiccomposition of CO2 alone is not an exhaustive diagnostic tool, as bothmetasediments and the mantle result in overlapping δ13C values.Furthermore, the ongoing decarbonation reactions in the SSGSinvolves recrystallized and modified calcite cement originally derivedfrom lacustrine porewaterswhere the calcite δ13C can be significantly13 C depleted compared to marine carbonates. Moreover, we cannotexclude that metamorphic marine sediments may be present atgreater depth in the Salton Trough.

Clayton and Muffler (1968) report the values of the carbon stableisotopes of carbonate cuttings recovered from a borehole in the SSGS.The δ13C of calcite as pore filling and in veins ranges from −1.4‰ to−7.8‰ (mean value −4.4‰) in the top 2.5 km represented bylacustrine sediments. For a CO2-calcite fractionation factor of 2–3‰ at350 °C (Sheppard, 1986; Chacko et al., 1991), the CO2 derived fromeithermarine or deltaic lacustrine carbonates (e.g. δ13C between−4.4to 2.2‰) would range between −2 and 5‰. These values partlyoverlap the range obtained from the seeping gases in the DSF.However the δ13C values of the lacustrine carbonate cuttings alsooverlap the commonly used MORB values (Kyser, 1986), thus nothelping to differentiate the original source of gas. To sum up, anunambiguous discrimination of carbon sources cannot be deducedfrom the sole δ13CCO2 values.

The helium isotope data indicate a strong mantle component, withR/Ra from 6.08 to 6.64 (Fig. 5B top), which is within the range oftypical Sub-Continental Mantle (SCM; Gautheron and Moreira, 2002),including that of western USA (Dodson et al., 1998). The δ13C valueswould also be consistent with a dominant mantle origin. In theCO2/3He vs δ13C plot (Fig. 5b bottom) Salton gas appears to be close totypical SCM values which are CO2-enriched in comparison to MORB(Sano andMarty, 1995; Kampf et al., 2007; Hahm et al., 2008; Tedescoet al., 2010). Also, increased CO2/3He ratios can be due to fractionationduring gas migration (Hilton, 1996) so that the original deep gas maybewell within the SCM field, and the limestone-derived fraction couldbe minimal or nil. To summarize, the origin of CO2 and He of SaltonSea is similar to the magmatic gas in the Long Valley Caldera (Fig. 5b),which is less than 600 km north of Salton (R/Ra: 6–6.7; CO2/3He:1010-1011; Hilton, 1996). Motyka et al. (1989) reported similar resultsshowing that the Kalawasi group mud volcanoes in Alaska have CO2

dominated seep gases with δ13CCO2 values (−3.1 to 4.8‰) over-lapping those of the DSF and high 3He/4He ratios (i.e. between 2.1 and4.1). Thesemud volcano gases likely derive from amagmatic intrusion

Table 2BMolecular and isotopic composition of gas from vents. Gas composition is in vol. %. Isotopic data: δ13C:‰, VPDB; δD:‰, VSMOW; δ15N:‰, atm. R/Ra=(3He/4He)sample/(3He/4He)atmosphere; Ra=1.39×10-6; bdl: below detection limit, nd: not determined. Samples from CA07 have only hydrocarbons molecular composition reported.

Sample ID Structure C1 C2 C3 iC4 nC4 CO2 N2 δ13C1 δD1 δ13CO2 δ15N δ13C23He/4He (R/RA) (He/Ne)/(He/Ne)air

CA03-3 North pool 0.51 0.01 bdl bdl bdl 99.49 0.00 -32.7 nd -5.3 nd nd nd ndCA03-5 Oil pool 0.49 0.02 bdl bdl bdl 99.49 0.00 -32 nd -5.4 nd -20.1 nd ndCA06-21 Oil pool 1.20 0.01 0.00 bdl bdl 98.79 0.00 -24 nd -4.7 nd nd nd ndCA06-22 East gryphon 1.24 0.01 0.00 bdl bdl 98.74 0.00 -24.1 nd -4.1 nd nd nd ndCA06-35 Gryphon 1.12 0.01 0.00 bdl bdl 98.86 0.00 -21.9 nd -4.4 nd nd nd ndCA06-37 Gryphon 1.18 0.01 0.00 bdl bdl 98.81 0.00 -23.7 nd -4.6 nd nd nd ndCA06-40 Gryphon 0.91 0.01 0.00 bdl bdl 99.07 0.00 -23.8 nd -3.6 nd nd nd ndCA06-32 Gryphon 0.86 0.01 0.00 bdl bdl 99.13 0.00 -18.9 nd -3 nd nd nd ndCA06-23 Pool 1.43 0.01 0.00 bdl bdl 98.55 0.00 -24.2 nd -3.6 nd nd nd ndCA06-24 Pool 1.00 0.01 0.00 bdl bdl 98.99 0.00 -24.2 nd -4.5 nd nd nd ndCA06-25 North pool 1.40 0.04 0.00 bdl bdl 98.55 0.00 -27.6 nd -5 nd nd nd ndCA06-29 Pool 1.08 0.01 0.00 bdl bdl 98.91 0.00 -24 nd -3.9 nd nd nd ndCA06-34 Pool 1.26 0.02 0.00 bdl bdl 98.71 0.00 -17.6 nd -3.8 nd nd nd ndCA07-01 Oil pool 1.07 0.01 0.00 0.00 0.00 nd 0.00 -23 nd -4.7 nd nd 6.64 1671CA07-02 North pool 1.36 0.03 0.00 bdl bdl nd 0.00 -26.4 nd -2.9 nd nd nd ndCA07-03 East Gryphon 1.04 0.01 0.00 0.00 0.00 nd 0.00 -23.5 nd -4.3 nd nd 6.55 1563CA07-04 West Gryphon 1.06 0.01 0.00 0.00 0.00 nd 0.00 -23.1 nd -3.9 nd nd 6.6 456CA07-05 South Gryphon 0.84 0.01 0.00 0.00 0.00 nd 0.00 -23.2 nd -4.4 nd nd 6.08 29.7CA07-06 Central Gryphon 0.27 0.00 bdl bdl bdl nd 0.00 nd nd nd nd nd ndCA07-08 South artesan pool 0.25 bdl bdl bdl bdl nd 0.00 nd nd nd nd nd ndCA07-10 West pool nd nd nd nd nd nd nd nd nd nd nd nd 6.52 107CA08-01A Oil pool 1.67 0.01 0.00 0.00 0.00 98.31 0.00 -24.4 -171 -4.0 nd nd nd ndCA08-02A North pool 1.86 0.03 0.00 0.00 bdl 98.11 0.00 -26.5 -216 -1.6 nd nd nd ndCA08-03A East Gryphon 1.21 0.01 0.00 0.00 0.00 98.08 0.70 -24.2 -159 -2.4 -0.52 nd nd ndCA08-04A West Gryphon 1.45 0.01 0.00 0.00 0.00 97.78 0.76 -23.6 -158 -0.2 nd nd nd ndCA08-05A South Gryphon 1.17 0.01 0.00 0.00 7.72 98.82 0.00 -23.2 -160 0.4 nd nd nd ndCA08-06A Central Gryphon 1.81 0.01 0.00 0.00 0.00 98.18 0.00 -24.8 -150 0.2 nd nd nd ndCA08-07A Central pool 1.48 0.01 0.00 0.00 0.00 97.16 0.35 -23.5 -156 -2.4 nd nd nd ndCA08-08A East central gryphon 1.37 0.01 0.00 0.00 0.00 98.96 0.66 -23.6 -174 2.3 nd nd nd ndWelhan, 1988 Deeep high-T well -26 -270


with additional contributions of CO2 from contact metamorphisms ofdeep sites limestone beds, thus comparable to the outlined DSFscenario.

The methane stable carbon isotopic ratio measured in 2003(δ13CCH4~−32‰) suggested a thermogenic origin (Svensen et al.,2007). Our data collected between 2006–2008 gave considerablyhigher values (from −17.6 to −26.5‰) suggesting oxidation ofthermogenic gas, high-temperature pyrolisis of bituminous materialor an abiogenic source (Fig. 5C). Welhan (1988) reported δ13CCH4values of −26‰ for unoxidised gas and values from −16 to −0.6‰for fractionated gases in deep wells from the SSGS (T=320 °C ±20).The CH4 concentration (b1.5 vol.%) is much lower than that of theGuaymas Basin (about 61 vol.%; Welhan and Lupton, 1987) wherelight hydrocarbons were considered to be mainly thermogenic(δ13CCH4 from −43 to −51‰). A simple calculation of Rayleighfractionation with bacterial oxidation factors (Coleman et al., 1981)suggests that starting from a CH4 with δ13CCH4~−45‰ (as that ofGuaymas Basin), the oxidation of 97% of CH4 (necessary to get 2 vol.%from an initial concentration of 60 vol.%, as that of Guaymas Basin)would produce a residual CH4 with δ13CCH4 much higher than−20‰, similar to the oxidised residual gases reported by Welhan(1988). The isotopic composition of DSF seeping CH4 is similar tothat of gas in deep SSGS wells (Fig. 5C); consistent with the carbonisotopic composition of the evolved CO2, this suggests the lack ofsignificant microbial oxidation during migration to the surface; acomplete thermogenic source is however improbable. The presenceof a mantle helium (R/Ra~6) and additional mantle CO2 togetherwith the high seep water temperatures (up to ~70 °C) suggests thatalso CH4 might have a mantle component (Welhan and Craig, 1983),in addition to the thermogenic one. By using the 13CCH4 vs R/Radiagram (Fig. 5D; Jenden et al., 1993) it appears that Salton has a

higher magmatic component than the other large sediment-hostedhydrothermal system of Guaymas. However, the occurrence ofabiogenic non-magmatic (crustal) CH4 components, due for exampleto thermal reduction of CO2 (from mantle and limestones) to CH4,metamorphism of graphite, thermal decomposition of carbonatesand FTT reactions (e.g. Potter et al., 2004; Fiebig et al., 2007) isentirely possible. The presence of oil in some seeps suggests, in anycase, that high-T pyrolysis of petroleum has also taken place.Multiple origins and consequent mixing of CO2 and CH4 in the Saltongas prevent to quantify the actual sources without significantuncertainty.

The two available isotopic data of N2 (δ15N: -0.52‰) and ethane(δ13C2: -20.1‰) are typical of deep crust or mantle (δ15N from −2to +1‰; Zhu et al., 2000) and of highly mature thermogenic orabiogenic gas (typically with δ13C2 values higher than−25‰; Jendenet al., 1993; Potter et al., 2004). However, additional isotope data onethane coupledwith those of propane and butane are needed to betterassess the thermogenic vs abiogenic contributions.

5.2. Water source

The major and trace element geochemistry of the seep waters canbe interpreted to be controlled by a combination of in situ evaporationand halite dissolution in addition to dilution by meteoric watersduring periods of rain. The elevated Cl/Br ratio of most seeps is bestexplained by input from halite dissolution, as halite has Cl/BrN3000(Mccaffrey et al., 1987). Even rain water spilling off into the poolswould result in high Cl/Br waters in the seeps, as halite crusts areabundant throughout the area. During dry periods, evaporation willincrease the pool salinities without changing the Cl/Br ratio untilhalite precipitates (Fig. 4A). These interpretations are supported by

Fig. 5. Gas analyses results and interpretation. (A) Combined δ13CCH4 and δ13CCO2 with results sub-divided by year of sampling and type of venting structure; (B) He isotopic ratios vs.13CCO2 (top) and CO2/3He vs δ13CCO2 (bottom). Sub-Continental Mantle (SCM) and arc volcanoes (ARC) are from Gautheron and Moreira (2002), Kampf et al. (2007), Hahm et al.(2008); Mid-Ocean Ridge Basalt (MORB), marine limestone and organic sediments are from Sano andMarty (1995). Long Valley Caldera from Hilton (1996). Mixing lines are amongthe SCM, limestone and sediment endmembers. (C) Methane δ13C vs δD. Salton Sea data (seeps) are compared with another major sediment-hosted hydrothermal field (GuaymasBasin) and abiogenic gases. Salton wells: Welhan and Lupton (1987); high-T well:Welhan (1988); Cerro Prieto:Welhan and Lupton (1987); EPR: East Pacific Rise:Welhan and Craig(1983); Guaymas Basin: Welhan (1988); Socorro: Taran et al. (2010); Lost City: Proskurowski et al. (2008); Lovozero: Potter et al. (2004); Zambales, the Philippines: Abrajano et al.(1988); Chimaera, Turkey: Hosgormez et al. (2008). (D) Plot of methane δ13C vs 3He/4He (as R/Ra). Dashed lines encompass commercial biotic gas from America, Asia and Europe asreported by Jenden et al. (1993). Continuous lines are themixing paths between hypothetical crustal andmagmatic endmembers with ratios (CH4/3He)crustal / (CH4/3He) magmaticfrom 1 to 106 (Jenden et al., 1993). East Pacific Rise: Welhan and Craig (1983); Guaymas Basin: Lupton (1983) and Welhan (1988).


the deuterium and oxygen isotopic data, showing a distinctevaporation signature for most of the samples, and of meteoricdilution in two cases (the ‘oil pool’ and the ‘north pool’). Whencomparing the ‘oil pool’ and ‘north pool’ geochemistry from 2006 and2007, both seeps were strongly depleted in 18O in 2007, which can be

explained by dilution by rain water in the weeks prior to sampling.The salinity response to this dilution is not available as major elementanalyses were not performed on the 2007 samples. The geochemicalsignature of the deep geothermal reservoir brines, as reported byElders and Sass (1988), are enriched in 18O compared to pool waters

632450 632500 632550 632600

3674350

3674400

3674450

3674500

3674550

632450 632500 632550 632600

3674350

3674400

3674450

3674500

3674550

DC

No

rth

ing

Easting

-2 -1CO flux: g m d2 CH flux: m4-2 -1g m d

100

100

100

100

100

100

100 100

100

500

500

500

500

1000

500

1000 500

500

500

1000

1000

500

500

500

1000

1000

100050

0Northing

Easting

E

20 40 60 80100

1500.1 0.2 0.5 1

CO flux2 CH flux4

kg/d kg/d

BA

No

rth

ing

Easting

Fig. 6. (A-B) CO2 and CH4 fluxes measured from the vent sites (pools or gryphons) throughout the field. A total of 96 seepage points have been measured; (C-D) 2D plots of CO2 andCH4 with flux distribution of miniseepage from soil (vents excluded); (E) 3D plots of CO2 and CH4 flux distribution of miniseepage from soil (vents excluded). UTM coordinates(zone 11).


image of Fig.�6


by at least 4‰. A key question is: do the pool waters contain asignature of this deep brine, or are all waters meteoric or mixtures ofmeteoric waters affected by shallow water-rock interactions? Sincethe temperature of the seeps at the surface can be as high as 70 °C(i.e. in the gryphons, see Tables A1 and 1), even shallow subsurfaceinteractions between water and clay minerals could result in a 18O-enrichment (Savin and Hsieh, 1998). Moreover, oxygen isotopeexchange between CO2 and H2O at high temperature at depth in thesystem may lead to 18O depletion of the water (Chiodini et al., 2000).Although the seepage of deep brines has been suggested previously(Sturz et al., 1992), and the new gas geochemistry suggests theinvolvement of gases with a very deep origin, our water data remainsinconclusive.

5.3. Seep dynamics and their shallow plumbing system

Our data suggest the existence of a complex subsurface plumbingsystem at the DSF. Although the gas composition is consistentthroughout the field, there are significant variations of water contentand temperature at different sites, in some places between vents asclose as 0.2 meters (e.g. Fig. 3).

Carbon dioxide flows to the surface together with Na-Cl brine,mud, and petroleum, with the individual seeps having differing watertables, gas fluxes, and fluid exit temperatures. Petroleum is present inonly one seep (i.e. oil pool, Table 1), and is likely sourced from hightemperature maturation of organic matter (Svensen et al., 2007).Gryphons are mud-rich and normally hot (22-69 °C) whereas thewater-rich seeps are generally colder (b32 °C) and form pools. Weconclude that the water has a mainly meteoric source, and that hotmud and a perturbed geothermal gradient in the vicinity of the seepfield are responsible for the up to 70 °C exit temperatures. Thishypothesis is supported by previous data by Svensen et al. (2009).These authors conducted broad temperature monitoring of the ventsin the field showing that diurnal air variations control the temper-ature in the water pools, whereas the gryphon temperatures do notcorrelate with air temperature and are instead characterized byabrupt spikes interpreted as surges of hot fluids. Although these siteshave large temperature variations, their gas composition remainsconsistent at most locations. This suggests that although the CO2 fluxis driving the seep activity at both the pools and gryphons, therecorded temperature variations are related to different proportionsof water in the erupted mud. Seeps with higher percentages of watermay allow faster cooling, where hot gas is quickly cooled whenbubbling through a colder high viscosity media (i.e. water) (Svensenet al., 2009).

Infrared images (Fig. 3) and field observations show that isolatedneighbouring pools often have different water tables and tempera-tures. Onderdonk et al. (2011) monitored pool levels over a 28 monthperiod and showed that there are consistent variations of the SaltonSea water level and some of the pools. This suggests that some of thepools may be connected to a shallow water table. Additionally,significant variations in water levels (up to 0.5 m) were observed inpools that were less than 0.3 m apart. This relationship, along with thereported temperature differences indicate that single seeps have acompletely independent plumbing system that extends to significantdepth despite their narrow spacing.

As reported earlier, the soil degassing flux is stronger on theperiphery of active vents and in particular around gryphon sites. Thisobservation supports the hypothesis of Mazzini et al. (2009) thatgryphons and neighbouring pools (i.e. caldera collapse pools) are partof the same plumbing and seepage system where a deep rootedconduit bifurcates at shallow depth before reaching the surface. Weelaborate on this model here by suggesting that around the gryphonclusters, where seepage is focused, a system of shallow radial fracturesproduces pools on the surface or, when the flux is less vigorous, anenhanced diffuse release through soil degassing. However, the precise

subsurface structure of these gryphon clusters and associated calderasis still unknown. Svensen et al. (2009) suggested two models thatimply the gradual collapse around gryphons where deep (model A) orshallow (model B) mud is forcefully erupted and mobilized by themigrating hot gas. The second model is favoured by Onderdonk et al.(2011) who interpreted that the positive volume of mud forming thegryphons is constantly balanced through time by negative volume(subsidence?) of the surrounding caldera by documenting coevealgryphon growth and caldera collapse. This observation may imply anentrainment of shallow mud through the conduit that is periodicallyerupted forming positive gryphons and subsequently subsided andrecycled for a new eruption. None of these models however seems tobe applicable to explain why the isolated pools in the field becomegradually wider and deeper. Two mechanisms are proposed: a) therise of fluids along the feeder conduits allows a continuous re-adjustment of the sedimentary textural mesh resulting in continuouscompaction and collapse at the surface; and b) the rise of CO2-richfluids promotes the continuous dissolution of anhydrite-rich unitsthat are abundantly distributed at least throughout the first 3,000 m(Herzig et al., 1988). Several isolated pools are also visible 2 km to thenorth of the DSF where the casing of abandoned exploration wellshave created the ideal pathway for CO2-rich fluids to reach thesurface.

Elders and Sass (1988) reported data from the borehole “State2-14” that was drilled just 800 m north of the DSF. The drillingpenetrated over 3,220 m of lacustrine sediments and two thin diabaseintrusions were intersected at 2880 and 2896 m depth. Theseintrusions are too thin to explain the extreme thermal anomalies inthe area, but the sills are likely part of a bigger intrusive complexlocated a few kilometres deeper. (Fig. 7). Microseismicity (Boyle et al.,2007) showed that local earthquakes cluster in a conduit-like zonethat most likely represents the region of focused fluid flow. Thesefeatures extend as deep as 5 km (and possibly up to 10 km).

5.4. Carbon dioxide and methane degassing

The CO2/CH4 flux ratio measured by the closed chamber is similarto the compositional CO2/CH4 ratio measured in the vents. This is aremarkable result considering that the error of individual measure-ments can be significant. This also implies that methane degassingthrough the soil is not significantly consumed by methanotrophicbacteria, except where gas fluxes are relatively low (orders of 100-101 mg CH4 m-2 day-1; Fig. 8). This result is similar to observations ofJapanese mud volcanoes (Etiope et al., 2011), but is in contrast withwhat was reported by D'Alessandro et al. (2009) from geothermalareas, who claimed that the methanotrophic effect is substantialeven in high flux degassing zones (N102 mg CH4 m-2 day-1). Webelieve that the major controlling factors are the soil characteristics(wetness, organic matter) and the availability of methanotrophs, andthat the effects of methanotrophic activity in soils increase asresidence times of gas in the soil increase due to lower gas migrationrates. Our conservative estimate reveals that at least 3,400 t/y of CO2

and 16 t/y of CH4 are released throughout the monitored field(0.05 km2). The methane emission factor (i.e. the total outputdivided by the area investigated), which is a fundamental parameterfor regional and global emission estimates of gases from naturalsources (Etiope et al., 2007a, 2011), is then 320 t km-2 y-1: this is atleast one order of magnitude lower than typical emission factors ofCH4-rich seeps and mud volcanoes in sedimentary basins (Etiopeet al., 2011) but up to 2 orders of magnitude higher than those ofnon-sedimentary geothermal and volcanic manifestations (Etiopeet al., 2007a). Sediment-hosted hydrothermal systems may thusrepresent an intermediate class of geologic methane sources, and onglobal scale they may significantly contribute to the atmosphericmethane budget.

2

1

3 Dep

th (

km)

Red Hillintrusion

Davis-Schrimpf Field

State 2-14

CO2 He

A B

CO reservoir2

OilLow TDS fluids

High TDS fluids

Decarbonation CO2?

260 °C

305 °C

355 °C

Sill intrusions

70 °C

Intrusions/Mantle Rocks

0 (km) 0.5 1.0 1.5 2.0 2.5 3.0

300 °C

?355 °C

Thermogenic CH ?4

4

5

?

CH4

abiogenic,

post or late magmatic

10-15?T

herm

al a

nom

aly/

Flu

id fl

ow

Fig. 7. Schematic cross section (not to scale) through points A-B (see Fig. 1B) of the geothermal system crossing the DSF. Temperatures from the State 2–14 borehole are from (Sass et al.,1988), and the temperatures at 914 meters depth at various distances from the Red Hill intrusion are taken from Elders and Sass (1988). The perturbation of the gradient below the seepfield is based on the assumption that the surface perturbation (70 °C in gryphons; Svensen et al., 2009) can be extrapolated. Also shown are the sources of the seep gases, and the source ofpetroleum (Svensen et al., 2007) which is assumed to be shallow since it is only locally present in the seep field. Shallowmeteoric fluids represent themain source ofwater for the systemalthough a deep origin cannot be excluded. The Imperial CO2 gas field is present in the upper part of the section.Mafic intrusive (?) rocks are inferred at 10–15 kmdepth based on seismicinterpretations (Fuis et al., 1982).


More than 75% of emission is from the invisible soil degassing.Indirect estimates by Kerrick et al. (1995), based on convective heatflow, temperatures and CO2 concentrations of reservoir fluids,indicated a total emission of 44,000 t/y of CO2 for the whole Salton

R2 = 0.9601

1

10

100

1000

10000

100000

1000000

1 10 100 1000 10000 100000

CH4 flux detection limit

Methanotrophic consumption?

CO2flux (g m-2 d-1)

CH

4 fl

ux

(mg

m-2

d-1

)

Fig. 8. Correlation between CO2 and CH4 fluxes. CH4 values belowdetection limit are set at5 mgm-2 d-1 only for graphical purpose. CH4 consumption by methanotrophic bacteriamay be significant only at low gas flux soil conditions.

Trough and SSGS (about 200 km2). This is equivalent to an averagediffuse CO2 flux of only 0.6 g m-2 d-1, that is virtually indistinguish-able from the background biologic CO2 flux of soil respiration(typically in the order of units to tens of g m-2 d-1), and four ordersof magnitude lower than the mean CO2 flux of the DSF as reportedhere (133 g m-2 d-1). This suggests that the DSF is an anomalouslyhigh degassing zone within the SSGS. However active venting hasbeen witnessed periodically at various locations north west of theseep field, along the shore of the Salton Sea and offshore in the SaltonSea (e.g. permanent mud pots fields and numerous ephemeral seepsand pots, see Lynch and Hudnut, 2008 for a detailed list). It istherefore likely that the regional assessment by Kerrick et al. (1995)might underestimate the actual volume of gas being released to theatmosphere within the SSGS.

Further flux measurements should be carried out to understandpossible temporal variations of the degassing. We observed that thegas flux was higher in the soil around active vents and in the areas ofdry and more fractured ground; this would suggest that a surveyconducted in a dry season with less compacted soil, would most likelyrecord higher fluxes. It is known, however, that gas seepage mayincrease after rainy periods and it also strongly depends onbarometric pressure and groundwater level (Koch and Heinicke,2007; Spulber et al., 2010).

Overall a NW-SE alignment of the vents and the soil degassingpattern can be inferred. This direction coincides with the direction ofthe Calipatria fault suggested by Lynch and Hudnut (2008). However

1

1,1

1,2

1,3

1,4

1,5

1,6

1,7

1,8

10 30 50 70temp (˚C)

2003 pool

2003 gryphon

2006 gryphon

2006 pool

water river, S. Sea,ditch

6,4

6,6

1

1,1

1,2

1,3

1,4

1,5

1,6

1,7

1,8

5 5,2 5,4 5,6 5,8 6 6,2 6,4 6,6 6,8

g/c

m3

pH

A

B

C

g/c

m3


the data presented in this article are not sufficient to confirm thepresence of this fault and a flux survey over a broader region should becompleted to highlight if the orientation of the main soil degassinghas a clear NW-SE trend.

6. Conclusions

This study in the Davis-Shrimp field represents one of the mostcomplete surveys over a sediment-hosted hydrothermal field. Ourapproach was to combine geochemical and temperature monitoringwith detailed measurements of gas released by soil degassing andfocused venting. The results can be summarized as follow:

• The composition of the gas released is consistent throughout thefield (~98% of CO2, 1.5% of CH4 and b0.5% of C2+) indicating thepresence of the same type of gas mixture in the whole plumbingsystem.

• The CO2 stable carbon and helium isotopic composition (R/RaN6)suggest a considerable Sub-Continental Mantle source, whereas theCH4 is likely to be a mixture of high temperature pyrolitic andabiogenic gas. These results define a deep rooted plumbing systemthat allows rapid release of fluids to the surface.

• Water analyses reveal a strong component of meteoric input both ingryphons and pools, with water geochemistry affected by in situevaporation.

• The concentrations of major elements correlate with Cl, except forBr and Ba. Pool water in apparent equilibrium with halite crustsshow a Cl concentration of 178,000 ppm Cl and an extremely highCl/Br of 35,600, in support of halite dissolution buffering the Cl andBr concentrations.

• The sum of gas released by invisible, diffuse soil degassing (7,310 kg/day of CO2 and 33 kg/day of CH4) and by vents (2,108 kg/day of CO2

and 11.5 kg/day of CH4) results in a significant emission of CO2,3,400 t/y, over a restricted area. An even larger amount of gas is likelyto be released all along the faulted area in the Salton Through. Soildegassing represents thedominant (~75%) component of gas releasedover the monitored field.

• Sediment-hosted hydrothermal systems may represent an interme-diate class of geologic methane sources for the atmosphere, withmethane emissions lower than those of sedimentary seepage inpetroleum basins but higher than those in traditional geothermal-volcanic systems; on global scale they may significantly contribute tothe atmospheric methane budget.

5

5,2

5,4

5,6

5,8

6

6,2

10 30 50 70

pH

temp (˚C)

Acknowledgements

The editor G. Chiodini, A. Sturz, and J. Fiebig are thanked for theirconstructive reviews. We gratefully acknowledge support from theNorwegian Research Council (PETROMAKS grant NFR-163469) toH. Svensen. S. Polteau, O. Galland, Ø. Hammer, A. Nermoen, K. Fristadand S. Planke are thanked for their support during field campaigns andfor the fruitful discussions. Thanks are due to D. Hilton and D. Tedescofor useful discussions.

Appendix A

Fig. A1. (A) pH vs density reveals that the gryphons have density higher than 1.4 g/m3;(B) density vs temperature shows that many denser gryphons have temperaturessignificantly higher than the more viscous pools; (C) pH vs temperature shows thatgryphons with higher temperature also have higher pH.

Table A1Summary of main characteristics of Davis-Schrimpf-field seeps (average values from2003-2007).

Structure Size (m) Height (m) pH* Temp. (°C) Density Salinity (TDS)*

Gryphon 0.5-4 0.4-2.5 5.98 Up to 70 N1.4 36.4Oil pool 2-4 -0.3 5.89 Up to 40 b1.4 2.7Water pool 0.5-6 -0.5 6.10 20-30 b1.4 44.0

Trench

A

BC

D

Trench 123

Reference

BA

C D

Fig. A2. (A) Outline of the sulphate-covered field showing the trench that was dug for stratigraphic purposes. (B) The surface mineralization at the centre of the field is characterizedby the presence of white and yellow sulphate minerals (Tamarugite and Blodite) in addition to halite. (C) Close-up of the trench where distinct layers are labelled A-D (Fig. A3).(D) A reference trench was dug at ten meters distance from the sulphate-covered field, showing the shallow stratigraphy dominated by two brown units and deeper dark grey units.Note the differences between the layers in the two trenches.


References

Abrajano, T.A., Sturchio, N.C., Bohlke, J.K., Lyon, G.L., Poreda, R.J., Stevens, C.M., 1988.Methane-hydrogen gas seeps, Zambales Ophiolite, Philippines: Deep or shalloworigin? Chemical Geology 71 (1–3), 211–222.

Boyle, K.L., Hutchings, L.J., Bonner, B.P., Foxall, W., Kasameyer, P.W., 2007. DoubleDifference Earthquake Locations at the Salton Sea Geothermal Reservoir.Geothermal Resources Council, September 2007(LLNL, UCRL-TR-233538).

Canet, C., Prol-Ledesma, R.M., Dando, P.R., Vázquez-Figueroa, V., Shumilin, E., Birosta, E.,Sánchez, A., Robinson, C.J., Camprubí, A., Tauler, E., 2010. Discovery of massiveseafloor gas seepage along the Wagner Fault, northern Gulf of California.Sedimentary Geology 228 (3–4), 292–303.

Chacko, T., Mayeda, T.K., Clayton, R.N., Goldsmith, J.R., 1991. Oxygen and Carbon IsotopeFractionations between Co2 and Calcite. Geochimica et Cosmochimica Acta 55 (10),2867–2882.

Chiodini, G., Allard, P., Caliro, S., Parello, F., 2000. 18O exchange between steam andcarbon dioxide in volcanic and hydrothermal gases: implications for the source ofwater. Geochimica et Cosmochimica Acta 64 (14), 2479–2488.

Cho, M., Liou, J.G., Bird, D.K., 1988. Prograde Phase-Relations in the State 2–14 WellMetasandstones, Salton-Sea Geothermal-Field, California. Journal of GeophysicalResearch-Solid Earth and Planets 93 (B11), 13081–13103.

Clayton, R.N., Muffler, J.L.P., 1968. Oxygen isotope study of calcite and silicates of theRiver Ranch No. 1 well Salton Sea geotehrmal field, California. American Journal ofScience 266, 968–979.

Coleman, D.D., Risatti, J.B., Schoell, M., 1981. Fractionation of carbon and hydrogenisotopes by methane oxidising bacteria. Geochimica et Cosmochimica Acta 45,1033–1037.

Craig, H., 1957. Isotopic standards for carbon and oxygen and correction factors formass spectrometric analyses of carbon dioxide. Geochimica et Cosmochimica Acta12, 133–149.

D'Alessandro, W., Bellomo, S., Brusca, L., Fiebig, J., Longo, M., Martelli, M., Pecoraino, G.,Salerno, F., 2009. Hydrothermal methane fluxes from the soil at Pantelleria island(Italy). Journal of Volcanology and Geothermal Research 187 (3–4), 147–157.

Dodson, A., DePaolo, D.J., Kennedy, B.M., 1998. Helium isotopes in lithospheric mantle:Evidence from Tertiary basalts of the western USA. Geochimica et CosmochimicaActa 62 (23–24), 3775–3787.

Elders, W.A., Sass, J.H., 1988. The Salton-Sea-Scientific-Drilling-Project. Journal ofGeophysical Research-Solid Earth and Planets 93 (B11), 12953–12968.

Elders, W.A., Biehler, S., Rex, R.W., Robinson, P.T., Meidav, T., 1972. Crustal Spreading inSouthern-California. Science 178 (4056), 15–24.

Etiope, G., Fridriksson, T., Italiano, F., Winwarter, W., Theloke, J., 2007a. Naturalemissions of methane from geothermal and volcanic sources in Europe. Journal ofVolcanology and Geothermal Research 165 (1–2), 76–86.

Etiope, G., Martinelli, G., Caracausi, A., Italiano, F., 2007b. Methane seeps and mudvolcanoes in Italy: gas origin, fractionation and emission to the atmosphere.Geophysical Research Letters 34, L14303. doi:10.1029/2007GL030341.

Etiope, G., Zwahlen, C., Anselmetti, F.S., Kupfer, R., Schubert, C.J., 2010. Origin and flux ofa gas seep in the Northern Alps (Giswil, Switzerland). Geofluids 10, 476–485.

Etiope, G., Nakada, R., Tanaka, K., Yoshida, N., 2011. Gas seepage from Tokamachi mudvolcanoes, onshore Niigata Basin (Japan): origin, post-genetic alterations and CH4-CO2 fluxes. Applied Geochemistry 26, 348–359.

Fiebig, J., Woodland, A.B., Spangenberg, J., Oschmann, W., 2007. Natural evidence forrapid abiogenic hydrothermal generation of CH4. Geochimica et CosmochimicaActa 71 (12), 3028–3039.

Fontes, J.C., Matray, J.M., 1993. Geochemistry and Origin of Formation Brines from theParis Basin, France. 1. Brines Associated with Triassic Salts. Chemical Geology 109(1–4), 149–175.

Fuis, G.S., Mooney, W.D., Healey, J.H., McMechan, G.A., W.J., L., 1982. Crustal structure ofthe Imperial Valley. U.S. Geol. Survey. Prof. Paper, 1254, pp. 25–49.

Gautheron, C., Moreira, M., 2002. Helium signature of the subcontinental lithosphericmantle. Earth and Planetary Science Letters 199 (1–2), 39–47.

Haaland, H.J., Furnes, H., Martinsen, O.J., 2000. Paleogene tuffaceous intervals, GraneField (Block 25/11), Norwegian North Sea: their depositional, petrographical,geochemical character and regional implications. Marine and Petroleum Geology17 (1), 101–118.

Hahm, D., Hilton, D.R., Cho, M., Wei, H., Kim, K.R., 2008. Geothermal He and CO2variations at Changbaishan intra-plate volcano (NE China) and the nature of thesub-continental lithospheric mantle. Geophysical Research Letters 35 (22).

Helgeson, H.C., 1968. Geologic and Thermodynamic Characteristics of Salton SeaGeothermal System. American Journal of Science 266 (3), 129–166.

Herzig, C.T.,Mehegan, J.M., Stelting, C.E., 1988. Lithostratigraphyof theState 2–14Borehole- Salton-Sea-Scientific-Drilling-Project. Journal of Geophysical Research-Solid Earthand Planets 93 (11), 12969–12980.

Hilton, D.R., 1996. The helium and carbon isotope systematics of a continentalgeothermal system: Results from monitoring studies at Long Valley caldera(California, USA). Chemical Geology 127 (4), 269–295.

http://dx.doi.org/10.1029/2007GL030341

0

10

20

30

40

NaCl, NaAl(SO ) *6H O (Tamarugite, white), Na Mg(SO ) *4H2O (Blodite, yellow), 2 2 4

KFeO2

4 2 2

(Potassium ferrite, reddish brown)

Gypsum

Gypsum

Gypsum, Calcite

Mg-Calcite

Mg-Calcite, Ankerite

Refer

ence

Anomaly

Dep

th (

cm)

Authigenic minerals

B

A

C

D

E

F

G

1

2

3

4

Fig. A3. Lithological logs of the reference trench and the thermal anomaly trench in thesulphatefield. The log shows that theuppermost7–8 cminboth sequences are characterizedby oxidized sediments, although tamarugite and blodite are absent in the reference section.The characteristic dark grey coloured mud from the reference section is altered in thethermal anomaly, and several distinct layers (D-G) contain authigenic minerals in nodules(gypsum and calcite). These minerals were not present in the reference section. Mineralcomposition was confirmed by XRD analyses.


Hosgormez, H., Etiope, G., Yalcin, M.N., 2008. New evidence for a mixed inorganic andorganic origin of the Olympic Chimaera fire (Turkey): a large onshore seepage ofabiogenic gas. Geofluids 8 (4), 263–273.

Hulen, J.B., Kaspereit, D., Norton, D.L., Osborn, W., Pulka, F.S., 2002. Refined conceptualmodeling and a new resource estimate for the Salton Sea geothermal field, ImperialValley, California. Geotherm. Resources Coun. Trans., 26, pp. 29–36.

0

5

10

15

20

25

30

35

40

45

24.12.2007 13.01.2008 02.02.200

Air

Tem

per

atu

re (

°C)

Sulfur

Air

Date

Fig. A4. Temperature Time series from December 2007 to April 2008 from the und

Jamtveit, B., Svensen, H., Podladchikov, Y., Planke, S., 2004. Hydrothermal vent complexesassociated with sill intrusions in sedimentary basins. Geological Society, London,Special Publications 234, 233–241.

Jenden, P.D., Hilton, D.R., Kaplan, I.R., Craig, H., 1993. Abiogenic hydrocarbons and mantlehelium in oil and gas fields. In: Howell, D.G. (Ed.), The Future of Energy Gases.(US Geological Survey Professional Paper 1570). United States Government PrintingOffice, Washington, pp. 31–56.

Kampf, H., Geissler, W.H., Brauer, K., 2007. Combined gas-geochemical and receiverfunction studies of the Vogtland/NW Bohemia intraplate mantle degassing field,Central Europe. In: Ritter, J.R.R., Christensen, U.R. (Eds.), Mantle plumes.A Multidisciplinary Approach. Springer-Verlag, Berlin Heidelberg, pp. 127–158.

Kerrick, D.M., McKibben, M.A., Seward, T.M., Caldeira, K., 1995. Convective hydro-thermal CO 2 emission from high heat flow regions. Chemical Geology 121,285–293.

Koch, U., Heinicke, J., 2007. Hydrological influences on long-term gas flow trends atlocations in the Vogtland/NW Bohemian seismic region (German-Czech border).Annals of Geophysics 50 (4), 557–568.

Kyser, T.K., 1986. Stable Isotope Variations in the Mantle. In: Valley, J.M., Taylor, H.P.J.,O'Neil, J.R. (Eds.), Stable Isotopes in High Temperature Geological Pocesses:Reviews in Mineralogy, pp. 141–164.

Lachenbruch, A.H., Sass, J.H., Galanis, S.P., 1985. Heat-Flow in Southernmost California andthe Origin of the Salton Trough. Journal of Geophysical Research-Solid Earth andPlanets 90 (Nb8), 6709–6736.

Lupton, J.E., 1983. Terrestrial Inert-Gases - Isotope Tracer Studies and Clues to PrimordialComponents in the Mantle. Annual Review of Earth and Planetary Sciences 11,371–414.

Lynch, D.K., Hudnut, K.W., 2008. The Wister mud pot lineament: South eastwardextension or abandoned strand of the San Andreas fault? Seismic Society AmericanBulletin 98 (4), 1720–1729.

Mazzini, A., Svensen, H., Planke, S., Guliyev, I., Akhmanov, G.G., Fallik, T., Banks, D., 2009.When mud volcanoes sleep: Insight from seep geochemistry at the Dashgil mudvolcano, Azerbaijan. Marine and Petroleum Geology 26 (9), 1704–1715.

Mccaffrey, M.A., Lazar, B., Holland, H.D., 1987. The Evaporation Path of Seawater and theCoprecipitation of Br- and K+with Halite. Journal of Sedimentary Petrology 57 (5),928–937.

Mckibben, M.A., Williams, A.E., Elders, W.A., Eldridge, C.S., 1987. Saline brines andmetallogenesis in a modern sediment-filled rift: the Salton Sea geothermal system,California, U.S.A. Applied Geochemistry 2, 563–578.

Motyka, R.J., Poreda, R.J., Jeffrey, A.W.A., 1989. Geochemistry, Isotopic Composition, andOrigin of Fluids Emanating from Mud Volcanos in the Copper River Basin, Alaska.Geochimica et Cosmochimica Acta 53 (1), 29–41.

Muffler, J.L.P., White, D.E., 1968. Origin of CO2 in the Salton Sea GeothermalSystem, Southeastern California, U.S.A. XXIII International Geological Congress,pp. 185–194.

Muffler, L.J.P., White, D.E., 1969. Acitve Metamorphism of Upper Cenozoic Sediments inSalton Sea Geothermal Field and Salton Trough Southeastern California. GeologicalSociety of America Bulletin 80 (2), 157–182.

Onderdonk, N., Mazzini, A., Shafer, L., Svensen, H., 2011. Controls on the geomorphicexpression and evolution of gryphons, pools, and caldera features at hydrothermalseeps in the Salton Sea Geothermal Field, southern California. Geomorphology130 (3–4), 327–342.

Potter, J., Rankin, A.H., Treloar, P.J., 2004. Abiogenic Fischer-Tropsch synthesis ofhydrocarbons in alkaline igneous rocks;fluid inclusion, textural and isotopic evidencefrom the Lovozero complex, N.W. Russia. Lithos 75 (3–4), 311–330.

Proskurowski, G., Lilley,M.D., Seewald, J.S., Fruh-Green, G.L., Olson, E.J., Lupton, J.E., Sylva, S.P.,Kelley, D.S., 2008. Abiogenic hydrocarbon production at Lost City hydrothermal field.Science 319 (5863), 604–607.

Robinson, P.T., Elders, W.A., Muffler, L.J.P., 1976. Quaternary Volcanism in Salton SeaGeothermal Field, Imperial-Valley, California. Geological Society of America Bulletin87 (3), 347–360.

8 22.02.2008 13.03.2008 02.04.2008

31

32

33

34

35

Su

lfu

r zo

ne

Tem

per

atu

re (

°C)

erground station at the sulphate field thermal anomaly and air temperature.


Sano, Y., Marty, B., 1995. Origin of Carbon in Fumarolic Gas from Island Arcs. ChemicalGeology 119 (1–4), 265–274.

Sass, J.H., Priest, S.S., Duda, L.E., Carson, C.C., Hendricks, J.D., Robison, L.C., 1988. ThermalRegime of the State 2–14 Well, Salton Sea Scientific Drilling Project. Journal ofGeophysical Research 93 (B11), 12995–13004.

Savin, S.M., Hsieh, J.C.C., 1998. The hydrogen and oxygen isotope geochemistry ofpedogenic clay minerals: principles and theoretical background. Geoderma 82(1–3), 227–253.

Shaw, A.M., Hilton, D.R., Fischer, T.P., Walker, J.A., Alvarado, G.E., 2003. Contrasting He-Crelationships in Nicaragua and Costa Rica: insights into C cycling throughsubduction zones. Earth and Planetary Science Letters 214 (3–4), 499–513.

Shearer, C.K., Papike, J.J., Simon, S.B., Davis, B.L., Laul, J.C., 1988. Mineral Reactions inAltered Sediments from the California State 2–14 Well - Variations in the ModalMineralogy, Mineral Chemistry and Bulk Composition of the Salton-Sea-Scientific-Drilling-Project Core. Journal of Geophysical Research-Solid Earth and Planets 93(B11), 13104–13122.

Sheppard, S.M.F., 1986. Characterization and Isotopic Variations in Natural-Waters.Reviews in Mineralogy 16, 165–183.

Simoneit, B.R.T., 1985. Hydrothermal Petroleum - Genesis, Migration, and Deposition inGuaymas Basin, Gulf of California. Canadian Journal of Earth Sciences 22 (12),1919–1929.

Simoneit, B.R.T., 1988. Petroleum Generation in Submarine Hydrothermal Systems - anUpdate. The Canadian Mineralogist 26, 827–840.

Simoneit, B.R.T., Galimov, E.M., 1984. Geochemistry of interstitial gases in Quaternarysediments of the Gulf of California. Chemical Geology 43 (1–2), 151–166.

Spulber, L., Etiope, G., Baciu, C., Malos, C., Vlad, S.N., 2010. Methane emission from naturalgas seeps and mud volcanoes in Transylvania (Romania). Geofluids 10, 463–475.

Sturz, A.A., Kamps, R.L., Earley, P.J., 1992. Temporal changes in mud volcanoes, SaltonSea Geothermal Area. In: Kharaka, Y.K., Maest, A.S. (Eds.), Water-rock interaction.Balkema, Rotterdam, pp. 1363–1366.

Svensen, H., Karlsen, D.A., Sturz, A., Backer-Owe, K., Banks, D., Planke, S., 2007. Processescontrolling water and hydrocarbon composition in seeps from the Salton Seageothermal system, California, USA. Geology 35 (1), 85–88.

Svensen, H., Hammer, Ø., Mazzini, A., Onderdonk, N., Polteau, S., Planke, S.,Podladchikov, Y.Y., 2009. Dynamics of hydrothermal seeps from the Salton Seageothermal system (California, USA) constrained by temperature monitoring andtime series analysis. Journal of Geophysical Research (Solid Earth) 114, B09201.doi:10.1029/2008JB006247.

Taran, Y.A., Varley, N.R., Inguaggiato, S., Cienfuegos, E., 2010. Geochemistry of H2- andCH4-enriched hydrothermal fluids of Socorro Island, Revillagigedo Archipelago,

Mexico. Evidence for serpentinization and abiogenic methane. Geofluids 10 (4),542–555.

Tedesco, D., Tassi, F., Vaselli, O., Poreda, R.J., Darrah, T., Cuoco, E., Yalire, M.M., 2010. Gasisotopic signatures (He, C, and Ar) in the Lake Kivu region (western branch of theEast African rift system): Geodynamic and volcanological implications. Journal ofGeophysical Research-Solid Earth 115.

Towse, D.F., Palmer, T.D., 1975. Summary of geology at the ERDA-Magma-SDG&Egeothermal test site. UCID-17008, Lawrence Livermore Natl. Lab, Livermore,California.

Von Damm, K.L., Edmond, J.M., Measures, C.I., Grant, B., 1985. Chemistry of submarinehydrothermal solutions at Guaymas Basin, Gulf of California. Geochimica etCosmochimica Acta 49 (11), 2221–2237.

Welhan, J.A., 1988. Origins of methane in hydrothermal systems. Chemical Geology 71,183–198.

Welhan, J.A., Craig, H., 1983. Methane, hydrogen and helium in hydrothermal fluids at21Â°N on the East Pacific Rise. In: Rona, P.A., Bostrom, K., Laubier, L., Smith, K.L.(Eds.), Hydrothermal Processes at Seafloor Spreading Centers. Plenum, New York,pp. 391–409.

Welhan, J.A., Lupton, J.E., 1987. Light-Hydrocarbon Gases in Guaymas Basin Hydro-thermal Fluids - Thermogenic Versus Abiogenic Origin. Aapg Bulletin-AmericanAssociation of Petroleum Geologists 71 (2), 215–223.

Werner, C., Hurwitz, S., Evans, W.C., Lowenstern, J.B., Bergfeld, D., Heasler, H.,Jaworowski, C., Hunt, A., 2008. Volatile emissions and gas geochemistry of HotSpring Basin, Yellowstone National Park, USA. Journal of Volcanology andGeothermal Research 178 (4), 751–762.

Williams, A.E., 1997. Fluid density distribution in a high temperature, stratifiedthermohaline system: Implications for saline hydrothermal circulation. Earth andPlanetary Science Letters 146 (1–2), 121–136.

Williams, A.E., Mckibben, M.A., 1989. A Brine Interface in the Salton-Sea GeothermalSystem, California - Fluid Geochemical and Isotopic Characteristics. Geochimica etCosmochimica Acta 53 (8), 1905–1920.

Younker, L.W., Kasameyer, P.W., Tewhey, J.D., 1982. Geological, Geophysical, andThermal-Characteristics of the Salton-Sea Geothermal-Field, California. Journal ofVolcanology and Geothermal Research 12 (3–4), 221–258.

Zarate-del Valle, P.F., Simoneit, B.R.T., 2005. Hydrothermal bitumen generated fromsedimentary organic matter of rift lakes - Lake Chapala, Citala Rift, western Mexico.Applied Geochemistry 20 (12), 2343–2350.

Zhu, Y.N., Shi, B.Q., Fang, C.B., 2000. The isotopic compositions of molecular nitrogen:implications on their origins in natural gas accumulations. Chemical Geology164 (3–4), 321–330.

http://dx.doi.org/10.1029/2008JB006247

journal of volcanology and geothermal researchfolk.uio.no/hensven/mazzini_jvgr_11.pdf · 68 a....

Documents