aus dem Fachbereich Geowissenschaftender Universität Bremen

No. 153

Moustafa, Yaser Ahmed

PALEOCLIMATIC RECONSTRUCTIONS OF THENORTHERN RED SEADURING THE HOLOCENE

INFERRED FROM STAHLE ISOTOPE RECORDS OFMODERN AND FOSSIL CORALS AND MOLLUSCS

Berichte, Fachbereich Geowissenschaften, Universität Bremen, No. 153,102 pages, Bremen 2000

ISSN 0931-0800

The "Berichte aus dem Fachbereich Geowissenschaften" are produced at irregtIlar intervals by the Department

of Geosciences, Bremen University.

They serve for the publication of experimental works, Ph.D.-theses and scientific contributions made by

members of the department.

Reports can be ordered from:

Gise1a Boelen

Sonderforschungsbereich 261

Universität Bremen

Postfach 330440

D 28334 BREMEN

Phone: (49) 421218-4124

Fax: (49) 421218-3116

e-mail: boelen@uni-bremen.de

Citation:

Moustafa, Y.

Paleoclimatic reconstructions of the Northern Red Sea during the Ho1ocene inferred from stable isotope records

of modern and fossil corals and molluscs.

Berichte, Fachbereich Geowissenschaften, Universität Bremen, No. 153, 102 pages, Bremen, 2000.

ISSN 0931-0800

PALEOCLIMATIC RECONSTRUCTIONS OF THE

NORTHERN RED SEA DURING THE HOLOCENE

INFERRED FROM STABLE ISOTOPE RECORDS OF

MODERN AND FOSSIL CORALS AND MOLLUSCS

Dissertation

Zur Erlangung des Doktorgrades der Naturwissenschaften(Dr. rer. nat.)

am Fachbereich Geowissenschaften (FBS)der Universität Bremen

Vorgelegt von

Yaser Ahmed MoustafaBremen

2000

Tag des Kolloquiums:

24 March 2000

Gutacter:

Professor Dr. Gerold Wefer

Professor Dr. Rüdiger Henrich

Prüfer:

Professor Dr. Gotthilf Hempel

Dr. Jürgen Pätzold

In about the year 2260 BC, "HERCHUF" (the commander-in chief of an expedition) reported

the Pharaoh "NEFER-KA-RE, PEPI II", that:

"

""I went fOlih (... ) upon the Oasis Road (... ) and I descended with 300 donkeys laden with (...) all sorts ofbeautiful treasures."

"Ich zog (... ) den Oasenweg hinauf( ... ) und zog mit 300 Eseln wieder herab, die mit (... ) allen schönen Kostbarkeiten beladen waren."

(E. Henfling, Meteorologischer Kalender 1997, Berlin)

Acknowledgements

I would like to thank Professor Gerold Wefer for giving me the opportunity for this thesis and

for his multifarious support of my work. Also, I thank Dr. Jürgen Pätzold for his enthusiastic

support and fruitful discussions during this time. Professor Rüdiger Henrich is gratefully

acknowledged for his commitment as a referee to this dissertation.

Many thanks are due to the Egyptian team of the May 1993 cruise in nOlihern Red Sea,

Professor Yossi Loya and Maoz Fine for logistic support and the drilling the RUS-93 cores

off Sinai Peninsula.

Monika Segl, Wolfgang Bevern and Birgit Meyer-Schack, who were constant sources of

support in the mass spectrometer laboratory, deserve my thanks because without them, the

completion ofthis dissertation would have been difficult.

Many discussions with Drs. Thomas Felis, Hennig Kuhnert and Sylke Draschba provided me

with insights which greatly improved the quality of the dissertation. I also very much

appreciate the constructive scientific discussions with Helge Arz. I thank Wolfgang Metzler,

Gerrit Meinecke, Volker Ratmeyer and Volker Diekamp for technical support in computer

questions and in the photo lab. Also, I thank the rest of the working group "Marine Geology"

for the comfortable atmosphere which accounted for a productive working climate.

I would like to thank P. Grootes, Leibniz Laboratory for Radionuc1ide and Isotope Research,

Kiel University, Germany and J. van der Plicht, Center for Isotope Research, Groningen

University, Netherland for AMS 14C dating, and M. Zuther, Geoscience Department, Bremen

University, Germany for X-ray diffractometry. Also, I thank K.-I-I Baumann and K.

Kaszemeik for the scanning electron microseope.

I sincerely thank my parents and my wife who gave me a lot of help and suppOli during my

studies. I owe special thanks to my brothers and sister who provided consistent moral support

throughout my stay here in Germany.

This research was pmily supported by Red Sea Program (RSP), funded by

"Bundesministerium für Bildung, Wissenschaft, Forschung und Technologie (BMBF)", a

grant from DAAD "Deutsche Akademischer Austausch Dienst), a grant from the University

of Bremen and a grant of the Graduierten Kolleg "StotT-Flüsse in marinen Geosystemen"

funded by the Deutsche Forschungsgemeinschaft (DFG).

TABLE OF CONTENTS

1. SUMMARy 1

2. INTRODUCTION

2.1 Background 3

2.1.1 Stable isotopes in corals and molluscs and their use as climate proxies

2.1.2 Modern and fossil reef terraces along the nOlthern Red Sea

2.1.3 Present oceanography and climate in the Red Sea region

2.1.4 Climate during mid- and late Holocene in the Near East region.

2.2 Corals previous work in the Red Sea (paleoclimatic studies) 8

2.3 Objective ofthis study 9

2.4. Material and Methods 9

2.4.1 Sampling site and collectedmaterials

2.4.2 X-radiography and X-diffractometry

2.4.3 14C dating

2.4.4 Stab1e isotopes

2.5 Thesis structure 11

2.6 References 13

3. PUBLICATIONS

3.1 Twentieth-century coral oxygen isotope record from southern Sinai (northern

Red Sea) 19

3.2 Modern and fossil Holocene marine climate record in molluscs (Tridacna spp.)

from the northern Red Sea .42

3.3 Mid-Holocene stable isotope record of corals from the northern Red Sea....... 61

4. CONCLUSIONS 83

5. APPENDIX 87

1

1. SUMMARY

Fossil corals and molluscs from elevated and submerged reef terraces of the western Gulf of

Aqaba (northern Red Sea) document Holocene c1imate changes in the northern Red Sea. The

stable isotopic cOlnposition of oxygen and carbon in massive modern and fossil corals

(Porites spp.) and also in modern and fossil molluscs (Tridacna spp.) is used to reconstruct

the past climatic and oceanographic changes during the Holocene in the northern Red Sea and

the Gulf of Aqaba.

High-resolution 8180 records of the two coral colonies Porites spp. (RUS-93a and RUS-93b)

reveal ages of the last 96 (1897-1993) and 50 (1943-1993) years, respectively. The colonies

grew about 250 m apart at Ras Umm Sidd off the southern Sinai coast, in the northern Red

Sea proper. The two 8180 records show the same average seasol1al amplitude (0.70%0) and the

same annual growth rate (1 cm/yr). Therefore, they respond similarly to the same

environmental signals. They cOITelate well with each other on a seasonal time scale (r=0.69; at

the 99.5% level) but on the mlliual time scale correlation is only moderate (r=0.40; at the

99.5% level).

To assess the colony oxygen isotope variability in Porites spp. during the twentieth century,

two previous 8180 records; RUS-95 from Felis (1999), RUS-l from Pätzold and Klein

(unpublished data) from the same region were correlated with our 8180 records (RUS-93a and

b). The correlation coefficients between the four coral 8180 records vary on the annual time

scale. These deviations are most likely due to different microenvironments causing variations

in the "vital effect" of the respective coral colonies. Annual means of the different coral 8180

records were compared with instrumentally measured sea-surface temperatures (in four time

intervals, 1907-1916, 1919-1939, 1946-1966 and 1976-1992 as a result of data gaps in the

instrumental record). Apart from the gaps in the data, the correlation coefficients varied from

-0.45 (RUS-95) to -0.60 (RUS-93a). The highest correlation (r=-0.68) is obtained by

averaging the three longest 8180. Therefore, it is required to analyse more than one coral from

a locality to reduce the local variability in coral 8180, and to use averaged data sets which are

more representative for paleoclimate studies as is the case in tree-ring paleoclimate studies.

The 8180/SST gradient varies between 0.15 to 0.18%0, with a mean value ofO.165%0, as is

shown in a previous study (Felis 1999).

The longest record of coral 8180 (RUS-93a) indicates a warmmg of about 1.I°e in the

twentieth-century. Pronounced coherent cyclicities of 10 and 3.6 years are evident in the coral

8180 time series by cross-spectral analysis between RUS-93a and RUS-93b. Also, a highly

coherent 5.8-year cyc1icity is also evident. This period is evident in the coral 8180 of RUS-95

2

core, which is also the most prominent coherent period in the co-spectrum of the North

Atlantic Oscillation Index (NAO) and the EI-Nifio Oscillation Index (ENSO) (Felis 1999).

The 3.6 year period is coherent with a similar cyclicity in the Southern Oscillation Index

(SOl).

The 8180 variations of recent shell carbonate in Tridacna sp. (between 1993-1996) from the

northern Gulf of Aqaba (Red Sea) show a cyclic pattern which agrees with the seasonal cycle

of the measured sea-surface temperature (5.3°C). During the last 6000 years the mean 8180

values of ten recent and fossil Thdacna spp. from the northern Gulf of Aqaba (Red Sea)

varied between 1.20%0 and 2.29%0. This mean 8180 variation (1.1 %0) in Tridacna spp. could

not be explained only with the change in temperature during the last 6000 years BP. We

suggest that a change in 8180 of sea water (i.e. salinity) has played a significant role (about

40%). Thus, the change in the salinity could have amounted to up to 1.5%0 in the last 6000

years. These results are corroborated by marine sediment cores from the Arabian Sea,

hieroglyphical and historical documents of high/low Nile floods during the last about 6000

years BP, and indicate changes in the intensity of the AfricanJAsian monsoon. The nOlihward

migration of the SW monsoon rain zone is the most probable explanation of this 8180 change

in sea water.

During the mid-Holocene (5750 - 4450 14e years BP), near-monthly 8180 records offive fossil

corals (Porites spp.) from the northern Gulf of Aqaba (Red Sea) show changes in climate

seasonality. The average seasonal amplitude of 8180 in these five coral records are higher (1.5

- 2-fold) than in modern corals. This is most probably due to a larger seasonal temperature

contrast and areduction of salinity during the summer seasons in the mid-Holocene times.

Moreover, mid-Holocene 8180 records reveal a reduced average growth rate (up to 45%) as

compared to modern corals. The reduction of coral growth during the mid-Holocene is

probably triggered by an increased input and resuspension of terrestrial sediments. All these

results support the hypothesis of summer monsoon rains reaching the northern Red Sea in

mid-Holocene times when seasonal solar radiation was enhanced. Enhanced seasonalities can

be reconstructed at leastbetween 5750 and 4450 14e years BP. A decrease in the absolute

isotope values of fossil corals occurs at about 5000 yr BP. This date coincides with aperiod of

rapid decline of north African lake and Nile levels indicating a reduced moisture transport

from the Atlantic and Indian oceans. Hence, the 8180 signal of the different corals during the

mid-Holocene could be controlled by intensity variations of the SW monsoon. The phase

relation between 8180 and 813e values of fossil corals are explained with the seasonal time lag

existent between maximum light and maximum temperature, as is the case in recent corals.

The difference between the 8 13e fractionation of mid-Holocene and recent corals could be

attributed to kinetic fractionation effects.

3

2. INTRODUCTION

2.1. Background

2.1.1 Stable isotopes in corals and molluscs and their use as climate proxies

The use of chemical information (e.g. stable isotopes) in corals and molluscs has gained

importance in the last two decades for interpreting past c1imatic changes (Swart and Dodge

1997). The pioneering works of Epstein et al. (1951, 1953) provided the starting point for

stable isotope studies in marine organisms. The very first study on the isotopic composition of

fossil molluscs (Urey et al. 1951) considered the term "vital effect" as a cause for a deviation

from the equilibrium. The first approaches in the field of investigating stable isotopes in coral

skeletons were done in the early 1960s (e.g., Keith and Weber 1965). In the 1970s first

explanations emerged as to which factors influence the fractionation of carbon and oxygen

isotopes in coral skeletons (Weber and Woodhead 1970). The usefulness of chemical

information in coral skeletons became apparent after the discover of the annual density bands

(e.g., Knutson et al. 1972; Macintyre and Smith 1974). These annual density bands in coral

skeletons were similar to tree rings. Hence, the stable isotopes could be analysed within this

annual chronology (e.g. Goreau 1977). Fairbanks and Dodge (1979) suggested that the

correlation of carbon and oxygen in the isotopic composition was determined by the amount

of sunshine and temperature. McConnaughey (1989) systematically studied the significant

kinetic effects that influenced the isotopic fractionation of both carbon and oxygen isotopes.

In the present decade (1990s) the isotopic records have been used to reconstruct the c1imate,

especially the EI Nifio Southern Oscillation (ENSO) (e.g., Cole and Fairbanks 1990 and

Dunbar et al. 1996).

In contrast to corals, molluscs exert only a lllinimal vital effect on the isotopic composition

(Milliman 1974; Jones 1985; Aharon 1991; Wefer and Berger 1991), which allows us not

only to reconstruct the c1imate variability in the past, but also to determine absolute

temperatures when the salinity of the sea water is known. The amplitude of 8180 cyc1es in

shell carbonate from juvenile tridacnid specimens from the Rose Atoll (southern Pacific

Ocean) agrees with the annual temperature variation (Romanek and Grossman 1989).

Although many other factors influence the isotopic composition of mollusc shells, water

temperature, salinity (for oxygen) and the total dissolved carbon content (for carbon) have

been recognised as major factors controlling the oxygen and carbon isotopes in the carbonate

shells of marine molluscs (Anderson and Arthur 1983). Therefore, high-resolution records of

8180 in Tridacna sp. can be used to reconstruct paleotemperatures. Wefer (1985) and Pätzold

et al. (1991) used this method for determining paleotemperatures from Tridacna species and

another mollusc species.

4

The 81S 0 in marine orgamsms vanes as a function of the ISO/160 ratio m seawater and

temperature (Epstein et al. 1953; Grossman and Ku 1986; Wefer and Berger 1991). The

following equations describe this relation between temperature, the 81S0 of calcitic (8 IS

0 C)

and aragonitic (8 1S0 AR) carbonate in the marine organisms and the 81S

0 W ratio in the sea

water.

T(OC) = 16.5- 4.3(8 IS0 C - 81S

0 W) + 0.14(8 Is0 c - 8 IS

O W)2

T(OC) = 21.8- 4.69 (8 ISO AR - 81S

0 W)

(Epstein et al. 1953)

(Grossman and Ku 1986)

The widely recognised relationship between 81S0 of carbonate and temperature for molluscs is

0.22%0/oC for calcite (Epstein et al. 1953) and 0.23%0/oC for aragonite (Grossman and Ku

1986). The gradient of 0.18%0/oC (Gagan et al. 1994) is widely accepted for the temperature

interpretation of Pari/es 81S0 records (Charles et al. 1997; Felis et al. 1998). The isotopic

oxygen compositions in carbonate-shelled organisms have also been used as qualitative

indicators of salinity (e.g., Rostek et al. 1993; Hemleben et al. 1996). This application of

stable isotopes is possible because meteoric waters generally have lower 81S0 values than

marine waters. The signature of isotopic oxygen in water bodies is often related to changes in

the water's salinity as a response to changes in evaporation, precipitation (humidity), and

mixing of waters from different sources (e.g., Rostek et al. 1993). The relationship between

the sea-surface 8 1S0 W versus salinity is different for different oceanic regions (Craig and

Gordon 1965; Fairbanks et al. 1992). For the Red Sea, the sea surface 81S0 W versus salinity is

0.29%0 8ISOw/1 %0 salinity (Craig 1966).

2.1.2 Modern and fossil reef terraces along the northern Red Sea coast

Recent and elevated fossil reef terraces are weIl developed along the coast of southern Sinai

and the northern Gulf of Aqaba. The reefs in the Gulf of Aqaba and the northern Red Sea

proper are mostly of the fringing type. In principle, late Quaternary climatic fluctuations and

eustatic sea-level changes, together with tectonic uplift rates of about 0.1 mm/year are

responsible for the fonnation of the reef sequences (Gvirtzman 1994). Four morphological

terraces can be identified at about 30 m to 1 m above the present sea level in the area between

Ras Mohamed National Park and Ras Nasrani (Gvirtzman et al. 1992; Gvirtzman 1994)

(Figure 2-1). The higher terrace reveals an average age of 310 ka and corresponds to the

interglacial period of the marine isotope stage 9. The second terrace was dated at an average

age of 206 ka, which fonned during the isotope stage 7.1, the third one has been dated at an

average age 122 ka, which was identified to be related to different levels of isotopic stage 5.

The last terrace is about 1 m above the present sea level and has been dated at an average age

5.2 ka.

5

The sedimentary tenaces of the western side of the Gulf of Aqaba at Eilat, reprerenting late

Pleistocene sea levels, occur in different depths. The 50-90 m deep tenace was probably

formed during isotope stage 3 (50000-70000 yr BP) and the 120-140 m depth tenace is

correlated with the last glacial period (18000 yr BP) (Reches et al. 1987). Nir (1996)

suggested that many of the Gulf s reefs represent several periods of growth with no activity

during the Pleistocene low water periods due to glaciation and a high activity during the inter­

glacial periods.

32 33 34

(b)

Sinai

Ras Nasran~ e_ '~V\> •

Ras Umm Sidd ,.Ras Mohamed

Red Sea

29

28

27

Fig. 2-1: a) Schel11atic l11ap ofthe Red Sea showing the Strait ofBab el Mandeb and the Gulf of

Aqaba and Gulf of Suez. b) Study area in the northern Red Sea. The sal11pling locations

are indicated by "cross" symbols. Coral cores RUS-93 a and b were collected from Ras

Umm Sidd (northern Red Sea). The fossil corals (l11id-Holocene), and recent and fossil

Tridacna spp. were collected southwest from Eilat (northern Gulf of Aqaba).

6

The different reef terraces exhibit well-preserved depositional sequences. Earlier studies

indicated that former interglacial sea-Ievel high stands correspond to wetter (rainy) periods in

the Red Sea area (Deuser et al. 1976; Klein et al. 1990). Changes in the atmospheric·

cirfulation and shifting of the monsoonal rain belts are considered to be the reason. In

contrast, the modern reef develops under extremely dry terrestrial conditions.

2.1.3 Present oceanography and climate in the Red Sea region

The Red Sea is a semi-enclosed silled basin (Figure 2-1) and the most saline water body in the

World Oceans (Morcos 1970). The salinity reaches up to 42%0. The climate is extremely arid

due to the very low ratio between precipitation (about 25 mm) and the potential evaporation

ranging from 2000 mm/year in the Red Sea proper to 3650 mm/year in the Gulf of Aqaba

(Reiss and Hottinger 1984). Sea-surface salinities (SSS) of the Red Sea increase from south

(38%0) to north (42%0) due to high average evaporation rates and minimal precipitation and

runoff (Morcos 1970; Ahmed and Sultan 1989). Sea-surface temperatures (SST) are higher in

the south (30°C) and lower in the north (l9°C). The circulation pattern in the Red Sea is

determined by the monsoon (Siedler 1969; MOlTOS 1970) as weIl as by the water exchange

from the Indian Ocean in the south (thermohaline circulation), through the narrow and

shallow (about 140 m depth) Strait of Bab el Mandeb (Eshel and Naik 1997). Thermohaline

forcing plays a dominant role in the Red Sea circulation compared with wind forcing (Eshel et

al. 1994). Thus, the salinity ofthe Red Sea is influenced by the exchange ofwater masses with

the Indian Ocean.

Throughout the year, warm waters of relatively low salinity and highly oxygenated waters

coming from the Red Sea enter the Gulf of Aqaba above the sill of Tiran (about 240 m depth)

and flow northward. The circulation is driven by the evaporative loss in this arid area and is

due to buoyancy flux. The colder and denser, deeper water returns southward to the Red Sea

beneath the inflow to the Gulf (Klinker et al. 1976; Reiss and Hottinger 1984). Therefore,

changes in circulation have a great impact on the material budget and the productivity of the

Gulf of Aqaba (Red Sea).

2.1.4 Climate during mid- and late Holocene in the Near East region.

A moist climate spreads all over East Africa, Arabia, Pakistan, India and Tibet and

characterised the early to late Holocene humid interval. This indicates that the convergence

between the southwest monsoon and the northwesterlies was located over inner Arabia (van

Campo et al. 1982; Ji et al. 1993) (Figure 2-2). Climate change in the Near East region and in

northern East Africa during the Holocene had cultural and demographie dimensions. As a

result of changes in climatic conditions [rom warm/wet to cold/dry in the Near East region,

several cultural and demographie breaks occurred since about 8000 years ago. These cultural

7

and demographie breaks were represented by three types of massive-people migrations. The

nomadie tribes movement in times of drought, the farming populations migration to seek

stable freshwater sourees and the migration of Asiatie-European people to the south due to

less solar radiation were the result of these climatie ehanges during the mid- and late­

Holoeene (Neev and Emery 1995).

40

-.I

Ul 20

p\

0 0(al \

0 40 80 1200 E

Fig.2-2: Schematic climatic maps of Africa and Asia during extreme c1imatic phases (modified

after Ji et al. 1993) show the study area (star). a) Cold (dry) period: in the shaded area,

the climate was drier than at present. ITCZ in Africa and Asia shifted southwards and the

westerlies influenced regions around the Mediterranean. The African/Asian monsoon

weakened. b) Warm (wet) period (i.e. mid-Holocene): the climate ofthe shaded area was

wetter than at present. ITCZ in Africa and Asia shifted to the north and the westerlies also

moved northward. The African/Asian monsoon strengthened.

8

During the dry interval (between 7500-4500 year BP), Nile floods must have been increased

in the eastern Mediterranean region as the monsoonal belt moved northward, hence increasing

the precipitation over the Ethiopian highland and central Africa plateaus (Figure 2-2) (Neev

and Emery 1995). In contrast, a reduced supply of water through the Nile River occurred at

the time of humid phases farther north in the Eastern Mediterranean region, far which an

inverse relationship of humidity and drought is cOIToborated far the Sahara and the mid­

latitude region of the Middle East and Europe (Neev and Emery 1995). Furthermore, the

heavy monsoons in Africa have been recarded in the black organic-rich sediment layers

(sapropels) in the east Mediterranean Sea, a consequence of heavy discharges from the Nile

River (e.g., Rossignol-Strick 1985).

2.2 Previous work on corals in the Red Sea (paleoclimatic studies)

Caral paleoclimate records are now available from the southern as weIl as from the northern

Red Sea. More recently, nearly the last two and a half centuries, a coral oxygen isotope record

from Ras Umm Sidd, northern Red Sea proper, has been obtained (Felis et al. 1998, Felis

1999). By applying the instrumentally measured records and proxies, this coral record

revealed new aspects on Near East climate variability. Felis (1999) deduced that the northern

Red Sea region is more arid during colder periods (i.e. higher evaporation) and more humid

during warmer periods. In contrast, the eastern Mediterranean region in the north is wetter

during colder periods. Schrag (1997) presented a new method for a mare rapid determination

of high-resolution elemental ratios (Sr/Ca, Mg/Ca) to distinguish between temperature and

salinity in the carals with a high-precision. The results (coupled 8180 and Sr/Ca) are used to

study the long-term evolution oftemperature and aridity in the narthern Red Sea

Mareover, Heiss (1994); Heiss et al. (1996); Heiss and Dullo (1997) systematically studied

climate influences on coral growth, carbonate production and stable isotopes from recent and

fossil carals in the northern Red Sea. Klein et al. (1992) discussed the isotope fractionation

behaviour during caral skeleton precipitation in the light of the environmental variables in the

northern Gulf of Aqaba, Red Sea. In addition, Klein et al. (1993) studied the depth-related

timing of density band formation in Porites spp. corals from the northern Gulf of Aqaba

inferred from x-ray chronology and stable isotope compositions. In the southern Red Sea

suggested Klein et al. (1997) that the decadal time-scale variations in the coral skeletal 8180

are closely correlated with both the Indian Ocean SST and variations in the Pacific-based

Southern Oscillation Index.

Klein et al. (1990) found yeIlow-green fluorescent bands in fossil Porites sp. from late

Quaternary reef terraces in southern Sinai. These bands were interpreted as evidence far a

9

wetter climate, as compared with the extreme desert conditions now, with a possible summer

rainfall regime during the late Quaternary.

2.3 Objective of this study

The objective of this work is to investigate the isotopic compositions of oxygen and carbon in

recent coral cores, recent and fossil Holocene molluscs, and mid-Holocene corals from the

northern Red Sea. This study focuses on the following questions:

Are recent data sets of coral 8180 representative for the region, 01' are they primarily

influenced by local environrnental variability?

How did the sea-surface temperature develop in the nOlihern Red Sea during the last 100

years, as inferred from coral 8180?

Did the summer monsoon rains reach the nOlihern Red Sea in mid-Holocene times (i.e. up

to 300 N)?

1f so, which effect did the summer monsoon rains have on the northern Red Sea surface

water and the skeleta18180 ofmarine organisms?

1s the intensity of the summer monsoon evident in the skeletal 8180 of marine organisms

in the northern Red Sea?

Are these results corroborated by other paleo-data sets?

According to the 1GBP-PAGES project (PEP Ir1) which is concerned with climate changes in

Europe and Africa, the northern Red Sea is an impOliant region of global paleoenvironrnental

research. However, marine paleoclimatic records for the Holocene time are rare. Therefore,

the main objective of this dissertation is to infer the climatic changes in the northern Red Sea

during the last 6000 years on the basis of the oxygen isotopic composition in corals and

molluscs. It also aims to assess the colony oxygen isotope variability in Porites spp. anel to

determine how the temperature during the twentieth century developed as inferred from coral

8180 records in the northern Red Sea.

2.4. Material and Methods

2.4.1 Sampling sites and collected materials

The Holocene corals (Porites spp.) and molluscs (Tridacna spp.) were collected 6 km

southwest ofEilat (site 1) at 29°31 'N and 34°56'E (northern end ofthe Gulf of Aqaba) from a

Holocene reef terrace in approximately 250 m distance from the recent shoreline. Moreover, a

recent Tridacna sp. was collected near the same location (in front of the marine laboratory in

Eilat) at a water depth of about 10m in June 1996.

10

The second site "Ras Umm Sidd" is located near the southern tip of the Sinai Peninsula ­

Egypt (27°50.9'N, 34°18.6'E) in the nOlihernmost part of the Red Sea proper. Two coral

cores (RUS-93a and RUS-93b) were taken from two hemispherical coral colonies (Porites

spp.) growing at depths between 3-5 m in the fringing reef at Ras Umm Sidd (M. Shukry,

personal communication, 1998) at the northern end of the Red Sea proper (Figure 2-1). The

colonies grew 250 m apmi from each other. An underwater pneumatic drilling machine with a

5 cm diameter bit was used to drill the cores vertically, parallel to the major axis of coral

growth. The distance between both sampling sites is about 180 km (Figure 2-1).

2.4.2 X-radiography and X-ray diffractometry

The Holocene coral colonies and modern coral cores were sectioned along their longitudinal

axis of growth to obtain slabs of about 5 mm thickness. Also, the Tridacna sampies were

radially sectioned into slaps with a water-cooled rock saw. The X-radiographs of corals were

prepared to reveal annual density bands (Knutson et al. 1972) for the determination of

sampling profiles. X-radiographs were made using a cabinet X-ray system (Faxitron 43855A,

Hewlett-Packard, USA). The coral slabs were exposed to 45 kV, 3 mA, for about 5-10

minutes. The distance between X-ray-source and film was about 50 cm. Agfa Structurix RP-2

type film was used. The X-radiographs revealed quite regular and weIl developed annual

density patterns of alternating bands ofhigh and low density.

To exc1ude sampling of diagenetically altered material, the mineralogy of the Holocene coral

(Porites spp.) and mollusc (Tridacna spp.) sampIes (20 sampies) were determined by X-ray

diffraction analysis on a Philips PW 1800 (Philips, Eindhoven, The Netherlands) X-ray

diffractometer (Cu, 45 kV, 35 mA) at an angle between 20° and 50° (28) with 1/4° 28 per min

(2h) at the Mineralogical Section of the Geoscience Department of the University of Bremen.

Sampies for X-ray diffraction analysis were taken from the same region where isotope

sampies were drilled later. Scanning electron microscopy investigation on some sampies were

done. Generally, the diagenetic history of the fossil reef terraces in the northern Red Sea is

quite weIl known (Strasser et al. 1992; Strasser and Strohmenger 1997).

2.4.3 14C dating

The fossil coral colonies and mollusc shells were dated by AMS 14C. The measurements were

performed at the Leibniz Laboratory for Radionuc1ide Dating and Isotope Research,

Christian-Albrechts-University Kiel, Germany, for all mollusc sampies and coral sampies H2

and F3 (Nadeau et al. 1997). All other sampies were measured at the Center for Isotope

Research, University of Groningen, Netherlands. The radiocarbon method is based on the rate

of decay of the radioactive or unstable carbon isotope 14 C4C), which is formed in the upper

11

atmosphere through the effect of cosmic ray neutrons upon nitrogen 14. 14C enters the earth's

oceans by means of atmospheric exchange and as dissolved carbonate and bicarbonate.

The reaction is:

14N + n => 14C + p; where "n" is a neutron and "p" is a proton.

The 14C dating based on decay equation of 14C

N=No* e-At

Where N is the number of 14C atoms in the sampie, and No the number of 14C atoms in the

time of formation, t is the age of the sampie and A is the decay constant which equals

ln2/Libby half-life (5568 years). The 14C ages were corrected for isotopic fractionation with

l3C values as measured by AMS. The 14C ages can be transformed into calendar years using

the calibration program Calib4 (Stuiver and Reimer 1993; Stuiver et al. 1998).

2.4.4 Stable isotopes

The stable isotopic composition was measured along a profile parallel to the growth direction

of the skeletal material. Sampies for isotopic analyses were drilled along these profiles. For

stable oxygen and carbon isotopic analyses powdered carbonate sampies were reacted with

100% orthophosphoric acid at 75°C to produce carbon dioxide. The isotope measurements

were performed using an automated carbonate preparation device attached to a Finnigan MAT

251 (Finnigan, Bremen, Germany) mass spectrometer. Results are given in the conventional 8

notation relative to the PDB (Belemnite from the Pee Dee Formation of South Carolina)

isotopic standard, calibrated by means of the NBS 19 standard:

8180 (%0) = {[C 80/160)sample - C80/160)standard] / C80/160\tandard} X 1000

8l3 C (%0) = {[(13C/12C)samPle - (13C/12C)stalldard] / (l3C/12C)srandard} X 1000

The precision based on replicate measurements of an internal laboratory standard (Solnhofen

limestone of 63 to 80 /-Lm) was ±0.07 %0 for 8180 and ±0.05 %0 for 813e. All stable isotope

analyses were carried out at the Isotope LaboratS)fy of the Geoscience Department of the

University ofBremen, Germany.

For the time-series analysis of the recent coral cores 8180 data (RUS-93a and b), the

AnalySeries (version 1.1) software package (Paillard et al. 1996) was used. This software was

used to identify periodicities in coral 8180 time series and relating them to known periodicities

of the other coral 8180 data from the northern Red Sea and the climate system. Therefore,

spectral and cross-spectral analyses were applied by using the Blackrnan-Tukey method

(Blackman and Tukey 1958; Jenkins and Watts 1968).

12

2.5 Thesis structure

The following chapters 3.1-3.3 are written in the form of manuscripts to be submitted for

publication or havealready been accepted in reviewed international scientific journals.

Therefore some descriptions ofthe study area and methods occur repeatedly in different parts.

Chapter 3.1.

Twentieth-century coral oxygen isotope records from southern Sinai (northern Red Sea)

To be submitted as:

Moustafa, Y. A, J. Pätzold and G. Wefer, Twentieth-century coral oxygen isotope records

from recent coral reefs off southern Sinai, northern Red Sea

This manuscript tests the correlation between different 8180 records extracted from different

corals growing near to each other in the same environment of the northern Red Sea proper.

Also, to infer the representativeness for regional SST changes, the individual records and an

averaged series are compared with instrumentally measured SST.

Chapter 3.2.

Modern and Holocene marine climate records In molluscs (Tridacna spp.) from the

northern Red Sea

Submitted as:

Moustafa, Y. A, J. Pätzold, Y. Loya, and G. Wefer, Modern and Holocene marine climate

records in molluscs (Tridacna spp.) from the northern Red Sea. To Terra Nova.

This manuscript of the dissertation is concerned with temporal variations m mollusc

(Tridacna spp.) mean 8180 in the Gulf of Aqaba (northern Red Sea) during the last 6000

years, and how these data corroborate other regional paleo-data that relate to the intensity

changes of African and Asian summer monsoon rains.

Chapter 3.3.

Mid-Holocene stable isotope record of corals from the northern Red Sea

Published as:

Moustafa, Y. A, J. Pätzold, Y. Loya, and G. Wefer, Mid-Holocene stable isotope record of

corals from the northern Red Sea. International Journal 01 Earth Sciences, 88: 742-751

(2000)

This paper deals with the seasonal 8 180 amplitude of mid-Holocene corals from the northern

Gulf of Aqaba (Red Sea) as weIl as their growth rates. The seasonal 8 180 amplitudes and the

growth rates of mid-Holocene corals were compared with the seasonal 8180 amplitudes and

13

the growth rates of modern corals. This comparison can help to give a picture about chmate

changes in the northern Red Sea, where the SW monsoon changed and influenced the climate,

during mid-Holocene times.

2.6 References

Aharon P (1991) Recorders of reef environment histories: stable isotopes in cora1s, giant

clams, and calcareous algae. Cora1 Reefs 10: 71-90

Ahmed F, Sultan SAR (1989) Surface heat fluxes and their comparison with the oceanic heat

flow in the Red Sea. Oceanol Acta 12: 33-36

Anderson TF, Arthur MA (1983) Stable isotopes of oxygen and carbon and their application

to sedimentologic and paleoenvironmental problems. In: Arthur MA, Anderson TF,

Kaplan IR, Veizer J, Land LS (Eds), Stable isotopes in sedimentary geology. Soc Econ

Paleon Mineral Short Course 10: 151pp

Blackman RB, Tukey JW, (1958) The measurement of power spectra from the point of view

of communication engineering. 199 pp, Dover Publication, New York

Charles CD, Hunter DE, Fairbanks RG (1997) Interaction between the ENSO and the Asian

monsoon in a coral record of tropical climate. Science 260: 1790-1793

Cole JE, Fairbanks RG (1990) The Southern Oscillation recorded in the 8180 of cora1s from

Tarawa Atoll. Paleoceanogr 5: 669-683

Craig H (1966) Isotopic composition and origin of the Red Sea and Salton Sea geothermal

brines. Science 154: 1544-1548

Craig H, Gordon LI (1965) Deuterium and oxygen-18 variations in the ocean and the marine

atmosphere. In: Tongiorrgi E (ed) Stable isotopes in Oceanographic Studies and

Paleotemperatures. CNR Lab Geol Nucl Pisa, pp 9-130

Deuser WG, Ross EH, Waterman LS (1976) Glacial and pluvial periods: Their relationship

reveald by pleistocene sediments of the Red Sea and Gulf of Aden. Science 191: 1168­

1170

14

Dunbar RB, Linsley BK, Wellington GM (1996) Eastern Pacific corals monitor EI

Nifio/Southern Oscillation, Precipitation, and sea surface temperature variability over

the past 3 centuries. In Jones PD, Bradley RS, Jouzel J (ed.): Climatic variations and

forcing mechanisms of the last 2000 years. Springer, Berlin Heidelberg, NATO ASI

Series 1, 41: 373-405

Epstein S, Buchsbaum R, Lowenstam HA, Urey HC (1951) Carbonate-water isotopic

temperature scale. BuH Geo1 Soc Am 62: 417-426

Epstein S, Buchsbaum R, Lowenstam HA, Urey HC (1953) Revised carbonate-water isotopic

temperature scale. Geol Soc America Bull 64: 1315-13225

Eshel G, Naik NH (1997) Climatological coastal jet collision, intermediate water formation,

and the general circulation ofthe Red Sea. J Phys Oceanogr 27: 1233-1257

Eshel G, Cane MA, Blumenthai MB (1994) Modes of subsurface, intermediate, and deep

water renewal in the Red Sea. J Geophys Res 99: 15941-15952

Fairbanks RC, Dodge RE (1979) Annual periodicity of 180/160 and 13C/12C ratios in the coral

lvJontastrea annularis. Geochim Cosmochim Acta 43: 1009-1020

Fairbanks RC, Charles CD, Wright JD (1992) Origin of global meltwater pulses. In: Taylor

RE, Long A, Kra RS (eds) Radiocarbon After Four Decades. Springer, New York, pp

473-500

Felis T (1999) Climate and ocean variability reconstructed from stable isotope records of

modern subtropical corals (northern Red Sea). Berichte, Fachbereich Geowissen­

schaften, Universität Bremen, Nr 133, 111 Seiten, Bremen

Felis T, Pätzold J, Wefer G, Fine M, Loya Y, Nawar AB (1998) First results of a coral-based

history of recent climate in the northern Red Sea. Zbl Geol Paläont Teil I, B 1/2: 197­

209

Gagan MK, Chivas AR, Isdale PJ (1994) High-resolution isotopic records from corals using

ocean temperature and mass spawning chronometers. Earth Planet Sci Lett 121: 549­

558

15

Goreau TJ (1977) Coral skeletal chemistrY:17hysiological and environmental regulation of

stable isotopes and trace metals in Montasstrea annularis. Proc Roy Soc London 196:

291-315

Grossman EL, Ku T-L (1986) Oxygen and carbon isotope fractionation in biogenic aragonite:

Temperature effects. Chem Geol (Isotope Geoscience Section) 59: 59-74

Gvirtzman, G, 1994: Fluctuations of sea level during the past 400 000 yeal's: the recol'd of

Sinai, Egypt (northern Red Sea). Coral Reefs 1: 203-214

Gvirtzman G, Kronfeld J, Buchbinder B (1992) Dated coral reefs of southern Sinai (Red Sea)

and their implication to late Quaternal'Y sea level. Mal' Geoll08: 29-37

Heiss GA (1994) Coral reefs in the Red Sea: Growth, Production and stable isotopes. Geomar

l'eport 32, unpublished Ph. D. Thesis, Kiel

Heiss GA (1996) Annual band width variation in Porites sp. from Aqaba, Gulf of Aqaba, Red

Sea. Bull Mar Sci 59: 393-403

Heiss GA, Dullo W-C (1997) Stable isotope l'ecord from recent and fossil Porites sp. in the

northern Red Sea. Coral Res Bull 5: 161-169

Hemleben C, Meichner D, Zahn R, Almogi-Labin A, Erlenkeuser H, Hiller B, (1996) Three

hundred eighty thousand year long stable isotope and faunal records from the Red Sea:

Influence of global sea level change on hydrography. Paleoceanogr 11: 147-156

Jenkins GM, Watts DG (1968) Spectral analysis and its applications, 525 pp, Holdenday,

Oakland

Ji J, Petit-Maire N, Yan Z (1993) The last 1000 years: climatic change in arid Asia and Africa.

Glob Planet Change 7: 203-21 0

Jones DS (1985) Growth increment, and geological variations in the molluscan shell. In: T.

W. Broodhed (Ed), Mollusks. University of Tennessee, Knoxville, Tenn Stud Geology

13: 72-87

Keith ML, Weber JN (1965) Systematic relationships between carbon and oxygen isotopes in

carbonates deposited by modern corals ad algae. Science 150: 498-501

16

Klein R, Loya Y, Gvirtzman G, Isdale PJ, Susic M (1990) Seasonal rainfall in the Sinai Desert

during the late Quaternary infened from fluorescent bands in fossil corals. Nature 345:

145-147

Klein R, Pätzold J, Wefer G, Loya Y (1992) Seasonal variations in the stable isotopic

composition and the skeletal density pattern of the coral Porites lobata (Gulf of Eilat,

Red Sea). Mal' Biol 112: 259-263

Klein R, Pätzold J, Wefer G, Loya Y (1993) Depth-related timing of density band formation

in Porites spp. corals from the Red Sea infened from X-ray chronology and stable

isotope composition. Mal' Ecol Prog Ser 97: 99-104

Klein R, Tudhope AW, Chilcot CP, Pätzold J, Abdulkarim Z, Fine M, Fallick AE, Loya Y

(1997) Evaluating southern Red Sea corals as a proxy record for the Asian monsoon.

Earth Planet Sci Lett 148: 381-394

Klink:er J, Reiss Z, Kropach C, Levanon I, Harpaz H, Halicz E, Assaf G (1976) Observations

on the circulation pattern in the Gulf of Elat (Aqaba), Red Sea. Israel J Earth-Sciences

25: 85-103

Knutson DW, Buddemeier RW, Smith SV (1972) Coral chronometers: Seasonal growth

bandsin reefs corals. Science 177: 270-272

Macintyre IG, Smith SV (1974) X-radiographic studies of skeletal development in coral

colonies. Proc 2nd Int Coral Reef Sym: 177-287

McConnaughey T (1989) 13C and 180 isotopic disequilibrium in biological carbonates: 1.

Patterns. Geochim Cosmochim Acta 53: 151-162

Milliman JD (1974) Marine carbonates. Springer, New York, pp 1-375

Morcos SA (1970) Physical and chemical oceanography of the Red Sea. Oceanogr Mal' Biol

Ann Rev 8: 73-202

Nadeau M-J, Schleicher M, Grootes PM, Erlenkeuser H, Gottdang A, Mous DJW, Sarnthein

JM, Willkomm H (1997) The Leibniz-Labor AMS facility at the Christian-Albrechts

University, Kiel, Germany. Nuclear Instruments and Methods in Physics Research B

123: 22-30

17

Neev D, Emery KO (1995) Thedestruction of Sodom, Gomorrah, and Jericho: Geologieal,

climatological, and archaeological background. Oxford University Press, New York

Oxford, pp 1-175

Nil' Y (1996) Sediment transport patterns of southern Sinai coasts and their role in the

Holocene development ofcoral reefs and lagoons. J Coastal Res 12: 70-78

Paillard D, Labeyrie L, Yiou P (1996) Macintosh program performs time-series analysis. Eos

Trans AGU 77: 379

Pätzo1d J, Heinrichs JP, Wolschendorf K, Wefer G (1991) Corre1ation of stable oxygen

isotope temperature record with light attenuation profiles in reef-dwelling Tridacna

shells. Cora1 Reefs 10: 65-69

Reches Z, Erez J, Garfunke1 Z (1937) Sedimentary and tectonic features in the northwestern

Gulf of E1at, Israel. Tectonophysics 141: 169-180

Reiss Z, Hottinger L (1984) The Gulf of Aqaba: Ecological micropaleontology. Springer,

Berlin Heidelberg New York, pp 1-354

Romanek CS, Grossman EL (1989) Stable isotope profiles of Tridacna maxima as

environmental indicators. Palaios 4: 402-413

Rossignol-Strick M (1985) Mediterranean Quaternary sapropels, an immediate response of the

African monsoon to variation of insolation. Palaeogeogr Palaeoclim Palaeecol 49:

237-263

Rostek F, Ruhland G, Bassinot FC, Müller PJ, Labeyrie LD, Lancelot Y, Bard E (1993)

Reconstructing sea surface temperature and salinity using d180 and alkenone records.

Nature 364: 318-322

Schrag DP (1997) Temperature and salinity history of the northern Red Sea from high­

resolution elemental and stable isotope records (abstract). Eos Trans AGU: 78 (46),

F387

Siedler G (1969) General circulation of water masses in the Red Sea. In: Degens ET Ross DA

(Eds) Hot brines and recent heavy metal deposits in the Red Sea, a geochemical and

geophysical account. Springer, Berlin Heidelberg New York, pp 131-137

18

Strasser A, Strohmenger C (1997) Early diagenesis in Pleistocene coral reefs, southern Sinai,

Egypt: responce to tectonics, sea-level and climate. Sedimentology 44: 537-558

Strasser A, Strohmenger C, Davaud E, Bach A (1992) Sequential evolution and diagenesis of

Pleistocene cora1 reefs (South Sinai, Egypt). Sed Geo 78: 59-79

Stuiver M, Reimer PJ (1993) Extended 14C database and revised CALIB radiocarbon

calibration program. Radiocarbon 35: 215-230

Stuiver M, Reimer PJ, Bard E, Beck JW, BUlT, GS, Hughen KA, Kromer B, McCormac FG,

v.d. Plicht J, Spurk, M. (1998) INTCAL98 radiocarbon age calibration, 24,000-0 cal

BP. Radiocarbon 40: 1041-1083

Swart PK, Dodge RE (1997) Climate records in coral skeletons. Proc 8th Int Coral Reef Symp

2: 1695-1696

Urey HC, Lowenstam HA, Epstein S, McKinney CR (1951) Measurements of paleo­

temperatures of the Upper Cretaceous of England, Denmark and the Southeastern

United States. Bull Geol Soc Am 62: 399-416

Van Campo, E, Duplessy JC, Rossignol-Strick M (1982) Climatic conditions deduced from a

150-kyr oxygen isotope-pollen record from the Arabian Sea. Nature 296: 56-59

Weber JN, Woodhead PMJ (1970) C and 0 isotope fractionation in the skeleta1 carbonate of

reefbuilding corals. Chem Geol 6: 93-117

Wefer G (1985) Die -Verteilung stabiler Isotope in Kalkschalen marine Organismen. Geol Jb

A 82: 1-114

Wefer G, Berger WH (1991) Isotope pa1eontology: grow1h and composition of extant

calcareous species. Mar Geol 100: 207-248

3.1

TWENTIETH-CENTURY CORAL OXYGEN ISOTOPE RECORD FROM

SOUTHERN SINAI (NORTHERN RED SEA)

(To be submitted to Journal of Geophysical Research)

ABSTRACT 20

INTRODUCTION 21

STUDY AREA 22

MATERIAL AND METHODS 22

Sampling procedure and isotopic measurements

Chronology development, Data treatment and environmental data

RESULTS , 24

Oxygen isotopes

Comparison with pre-existing local coral Ö180 record

Comparison and calibration with instrumental sea surface temperatures (SSTs)

Long-term trends in Ö180

Spectral analysis and cross-spectral analysis

DISCUSSION 31

CONCLUSIONS 36

ACKNOWLEDGEMENTS 36

REFERENCES 37

19

20

Twentieth-century coral oxygen isotQpe record from the southern

Sinai, northern Red Sea

Moustafa, Y. A., 1. Pätzold, and G. Wefer

Fachbereich Geowissenschaften, University of Bremen, Postfach 330440 28334 Bremen, Germany

Fax: 0049 (421) 2183116

email: yaser@uni-bremen.de

ABSTRACT

In this study we report high-resolution 96 and 50 year records of coral 8180, extracted from

two different Porites colonies (RUS-93a and RUS-93b, respectively). They are about 250 m

apart, from Ras Umm Sidd, off the southern Sinai coast, in the northern Red Sea proper. The

correlation coefficients between fom coral 8180 records from Ras Umm Sidd vary on an

annual time scale. The two additional 8 180 records were obtained previously from two other

colonies (RUS-95, Felis 1999, and RUS-l, Pätzold and Klein, unpublished data), which grew

in the same environment about 500 m away. Therefore, we suggest that local biological

(individual) effects play a maj or role in the coral 8180 records. As a test, we correlate the

different 8180 records with the eOADS sea-surface temperature (at al1l1Ual time scales and

four time intervals, 1907-1916, 1919-1939, 1946-1966 and 1976-1992). The correlation

coefficients are also different among the different 8180 records and eOADS SST on annual

time scales. Excluding the data gaps, the correlation coefficients varied from -0,45 (RUS-95)

to -0.60 (RUS-93a). A 3-coral average composite Ö180 record was constructed from 1907­

1992. Regression analysis between annual SST and this 3-coral average 8 180 record yields a

much better and significant correlation (r=-0.68) also excluding the data gaps. Therefore,

based on this study, it is advisable to analyse more than one coral from a given locality to

obtain a reliable coral 8 180 record that can be used in coral-based paleoclimate studies. The

ÖI80/SST gradient varied between 0.15 to 0.18%%e, with a mean value of 0.165%ofOe as in a

previous study by Felis (1999). The 8 180 record of RUS-93a indicates a warming in the

twentieth century of about 1.1 oe. Pronounced cyclicities of 10 and 3.6 years are evident when

cross-spectral analysis is applied to the bimonthly 8 180 time series of RUS-93a and RUS-93b.

The 3.6-year period is coherent with a similar cyclicity in the Southern Oscillation Index

(SOl). Also, a highly coherent 5.8 cyclicity is evident. This period is also present in the 8180

of RUS-95 (Felis 1999), and happens to be the most prominent period in the co-spectrum of

the Northern Atlantic Oscillation (NAO) and EI-Nino Southern Oscillation (ENSO).

21

INTRODUCTION

Stable-isotope time senes 111 corals are increasingly and widely used for climate

reconstructions of the tropics and subtropics (e.g., Pätzold 1984, 1986; Wefer and Berger

1991; Cole et al. 1993; Dunbar et al. 1994; Quinn et al. 1996, 1998; Charles et al. 1997;

Crowley et al. 1997; Kuhnert et al. 1999a and b). In each of these studies only a single 8180

record from one site was analysed to reconstruct either the local 01' the regional sea-surface

temperature (SST).

Coral paleoclimate records are now available from the southern as weIl as from the northern

Red Sea (Klein et al. 1990, 1992, 1993, 1997; Heiss 1994, 1996; Heiss and Dullo 1997;

Schrag 1997; Schrag et al. 1997). More recent1y, a nearly two and a half centuries of coral

oxygen isotope record from Ras umm Sidd, northern Red Sea proper, has been obtained (Felis

et al. 1998; Felis 1999). With the help of instrumental and proxy records of climate, this coral

record reveals new aspects of Near East climate variability. Felis (1999) deduced that during

co1der periods the northern Red Sea region is more arid (i.e. higher evaporation) and during

warmer periods more wet. In contrast, the eastern Mediterranean region in the north is wetter

during colder periods. Due to the configuration of the coastline of the southeastern edge of the

Mediterranean Sea, the northern Red Sea region lies outside the main path of the winter

rainstorms coming from the west (Westerlies) (Issar 1990; Goodfriend 1991). This westerly

zone moves northward during summer and southward during winter. The above climatic

pattern (cold-arid/warm-less arid) can thus be explained by these latitudinal position

variations of the subtropical westerly jet stream and moisture-bringing North Atlantic

Westerlies to the Eastern Mediterranean region (Felis 1999). The climate of the Red Sea

region is also influenced by the monsoon, a seasonally reversing wind system. The monsoon

is driven by the Hadley circulation of the tropical trade-wind system. In the Intertropical

Convergence Zone (ITCZ), the NE wind of the northern hemisphere meets the SE wind of tbe

southern hemisphere. This is a zone of high insolation with seasonally varying position (i.e. it

shifts northward in summer and southward during the winter). Within the ITCZ humid air

rises, flows poleward and descends in the subtropical high-pressure zones as warm and dry

air, where the subtropical westerly jet stream is located at a height of 12 km (Barry and

Chorley 1998). The Red Sea is the most saline water body in the World Oceans (Morcos

1970). Sea surface salinities (SSS) increase from south to north, in contrast to Sea Surface

Temperature (SST), due to high average evaporation rates of about 200 cm/yr. and minimal

precipitation and runoff (MOl"COS 1970; Ahmed and Sultan 1989). The circulation pattern in

the Red Sea is determined by the Monsoon (Siedler 1969; Morcos 1970) as weIl as by the

water exchange with the Indian Ocean to the south through the narrow and shallow (about 140

m deep) Strait of Bab el Mandeb (Eshel and Naik 1997). However, the thermohaline forcing

(at the Strait of Bab el Mandeb) plays a more dominant role in the Red Sea circulation than

the wind forcing (Eshel et al. 1994).

22

In--this study, we have generated relatively long coral 818

0 records from Ras Umm Sidd,

nOlihern Red Sea proper. They were obtained from cores RUS-93a and RUS-93b,

representing 96 and 50 years of coral growth, respectively. These records will help to verify18

the reproducibility of a coral 8 0 record measured from a 245 year old colony (Felis 1999)

and an 86 year old coral colony (Pätzold and Klein, unpublished data) from the same

environment in the northern Red Sea proper. The objectives of this study are also to: 1)

compare the coral oxygen-isotope records in the northern Red Sea with each other, 2)

compare the coral oxygen-isotope records with the nearest gridded sea-surface temperature in

the northern Red Sea and to deduce an empirical relationship, 3) to correlate the mean annual

8180 record calculated for three coral cores with the nearest grid box from aglobai SST data

set, and possibly to produce a master chronology curve for the corals from the northern Red

Sea, similar to the practise in tree-ring studies (e.g., Cook et al. 1998), and 4) determination of

the long-term trends in 8 180 during the twentieth century.

STUDY AREA

The Red Sea is a long (1932 km), narrow, desert-enclosed basin extending from about 35°E to

about 42°E between 300 N and l2°N with an average width of 280 km and a depth of up to

2800 m. In the north, the V-shaped Sinai Peninsula divides the Red Sea into the shallow Gulf

of Suez and the deep Gulf of Aqaba.

The study area (Ras Umm Sidd) is located at 27°50.9'N, 34°18.6'E in the northernmost part of

the Red Sea proper, on the southern tip of the Sinai Peninsula (Figure 3.1-1). Seasonal ssrs

in the northern Red Sea range between a minimum of about 21°C in winter (February ­

March) and a maximum of about 29°C in summer (August - September), with an average

annual SST of about 25°C. The average annual salinity is about 40.5%0 and the seasonal

salinity variation is 0.5%0. Ras Umm Sidd is a suitable site for coral paleoclimate studies,

because the narrow fringing reefs of the nOlihern Red Sea are exposed directly to open sea

conditions due to a steep submarine slope (Felis 1999) and because a deep water formation

occurs in the northern Red Sea (Cember 1998). Therefore, seasona1ity in cora1 8 180 records is

well developed.

MATERIAL AND METHODS

Sampling procedure and isotopic measurements

Two coral cores (RUS-93a and RUS-93b) were taken from two hemispherica1 cora1 colonies

(Porites spp.) growing at depths between 3-5 m in the fringing reef at Ras Umm Sidd (M.

Shukry, personal communication 1998) (Figure 3.1-1). An underwater pneumatic drilling

machine with a 5 ern-diameter bit was used to drill the cores vertically, parallel to the major

axis of coral growth. The cores were recovered from two different colonies, about 250 m

apart. For more details about the sampling site see Felis et al. (1998) and Felis (1999). The

23

cores were sectioned along their longitudinal axes-to obtain slabs of about 5 mm thickness. X­

radiographs were prepared (Knutson et al. 1972) to reveal the annual density bands far

determining sampling profiles. After determining the sampling profile along the central axis

of the highest bumps (Felis 1999), the aragonitic subsampies were collected by low-speed

drilling using a dentist drill with a I-mm diamond bit. The interval between subsampies is

about 2 mm. This corresponds to a spatial resolution of 5-6 samples/cm and an average

temporal resolution of 5.5 samples/year.

30'E 32 34 36 3834

Mediterranean Sea

32

Red Sea

QDead Sea

28

26

30

Fig. 3.1-1: Map of northern Red Sea showing the grid box 2°*2° centred at 27.00 N, 35.00 E,

from which the COADS SST dataset was obtained (Slutz et aJ. 1985; Woodruff et al.

1993). This grid box includes Ras Umm Sidd, the coral sampling locality (cross).

18To measure the stable oxygen isotope composltlon (8 0), a Filmigan MAT 251 mass

spectrometer connected with an automated carbonate preparation device was used. The

isotopic composition of the carbonate sampies was measured on the C02 gas evolved by

treatment with phosphoric acid (100%) at a constant temperature of 75°C. The ratio of

24

180 /60 is given in %0 relative to the PDB standard. A working standard (Burgbrohl C02 gas)

was used for the measurement of the isotopic ratios of the samples. This standard gas was

calibrated against the PDB standard using standard NBS 19 of the National Bureau of

Standards. Analytical standard deviation is ± 0.07%0 PDB (M. Segl, personal communication

1998).

Chronology development, data treatment and environmental data

A rough age determination was based on the density banding pattern in x-radiographs (each

couplet of a low- and a high density band represents one year). The Ö180 cyclicity presents a

high-resolution age model by assuming that the coral ÖI80 signal depends mainly on

temperature (Kuhnert et al. 1999b). Bence, we constructed a chronology by designating the

maximum Ö180 value within each year as mid-February (the coldest month in the year) (Felis

1999). Linear interpolation of six equidistant values (per year) between these maxima was

applied, using the AnalySeries 1.1 software package (Paillard et al. 1996). This procedure

provided a bimonthly resolution from January to December. This model provided the basis for

comparison with other local ÖI80 coral records (Felis 1999) and regional SST data. The

alIDual cora1 growth rate was calculated as the distance from a maximum Ö180 va1ue in a

certain year to the maximum value of the next year. The seasonal amplitude was obtained

from the difference between maximum and minimum Ö180 within one year.

To examine the relatiol13hips between coralöI8

0 and regional SST, we used instrument-based

records of Comprehensive Ocean Atmosphere Data Set (COADS) SST from the northern Red

Sea (Slutz et al. 1985; Woodruff et al. 1993). The COADS SST record of the 2° grid box,

centred on 27°N, 35°E was used. This box represents the closest marine box to the coral site.

These measurements were taken by ships that pass from Suez in the north, to the Strait of Bab

el Mandeb in the south. The instrumental observation of SST began when the Suez Canal was

opened in 1869~ Bowever, many data gaps occur as a result of wars (world-wide as wen as

regional). Therefore, the longest continuous instrumental observation of this SST data set

on1y comprises aperiod of 21 years.

RESULTS

Oxygen isotopes18

The high-resolution Ö 0 profiles of cores RUS-93a and RUS-93 bare presented in Figure 3.1-

2, and date back to the years 1897 and 1943, respectively. Both records show a marked

seasonality throughout their entire intervals of 96 and 50 years, respectively. The average

seasonal amplitudes are more 01' less the same in both cores (0.67 and 0.70%0, respectively).

By employing a Ö180/o C gradient ofO.165%0 (Felis 1999), the mean Ö180 seasonal amplitude

equals 4.2°C. Recent instrumental SST amplitudes for the northern Red Sea range between

25

6°8aoo 7°8 (Slutz et al. 1985; Woodruff et al. 1993; Rynolds and Smith 1994; Levitus and

Boyer 1994). The growth rate ranges from 0.5 cm/yr to 1.5 cm/yr for RUS-93a and from 0.6

crnlyr to 1.6 cm/yr for RUS-93b, with an average value of about 1cm/yr for both cores. The

seasonal (bimonthly) 8 180 records are in good agreement with each other between 1943 and

1993 (1'=0.66, at the 99.5% level). On a yearly basis, they correlate moderately and still

significantly (r=0.40, at the 99.5% level) (Figure 3.1-3). Applying a 5-year running average

does not improve the data correlation (r=0.40 at the 99.5% level).

2.00(b)..:::-.c

E~<l.>

~ 1.00.!::

~ 1.02 co(9 0

0.00 0...ro 0

ccf2.0.5 o~

Cf) <l.>((1-0<l.> ::J

-4.0 (f).-=:0.

-3.8 Eco ((I

0 -3.6 0.00... -3.4

0

C -3.20

-3.0<Xl

t.O-2.8-2.6

1900 1920 1940 1960 1980 2000Year

2.00(a)..:::-.c

E~<l.>

~ 1.00.!::

~ 1.02 co(9 0

0,00 0...ro 0

c#0,5 o~

Cf) Q)

-4.0((1-0Q) :l

(/).-t=-3.8 0.co-3.6 E

0 ((I

0... -3.40.0

0

C -3.20 -3.0~t.O -2.8

-2.6

1900 1920 1940 1960 1980 2000Year

Fig. 3.1-2: The Ras Umm Sidd coral 8 180 records. a) Bimonthly time series, seasonal amplitude

and annual coral growth rate in the core RUS-93a. b) Bimonthly time series, seasonal

amplitude and annual growth rate in the core RUS-93b. The annual cycles are weil

developed in both time series.

26

(a) -4.0-3.8 -.

RUS-93a -3.6 OJ0

-3.4 0....0

;;-g-3.2 0

-4.0-3.0 0

<Xl

-3.8 ~

ro -2.8 <-0

0 -3.6-2.60..

0 -3.4;;-g0 -3.2'---'

0 -3.0 RUS-93b~<-0 -2.8

-2.6

1940 1950 1960 1970 1980 1990 2000Year

-4.0(b)

-4.0

ro -3.8 RUS-93a -3.8 -.OJ

0 RUS-93b 00.. -3.6 -3.6 0....

0 0;;-g ;;-g0 0

-3.4 -3.4 '---'

0 0~

<Xl

-3.2 -3.2 ~

<-0 c.o

-3.0 -3.0

1940 1950 1960 1970 1980 1990 2000Year

Fig. 3.1-3: Comparison of the Ras Umm Sidd coral 8 180 records (50 years) of this study. a)

bimonthly time series r=0.66 (at 99.5% level). b) Annual time series 1'=0.40 (at 99.5%

level)

18Comparison with a pre-existing localcoral 8 0 record

The 8180 record of thc longer core (RUS-93a) was correlated with another 8180 record

obtained from a coral (RUS-95) growing about 500 m away from our coring site (Felis et al.

1998; Felis 1999). This coral belong to the same species and was also sampled with a

bimonthly resolution. Both records show the same annual amplitude (about 0.7 %0) and

average growth rate (about 1 cm/yr). They also exhibit the same general interannual pattern.

The two 8180 records correlate with 1'=0.69 (at the 99.5% level) on the seasonal time scale

(bimonthly) and with 1'=0.41 (at the 99.5% level) on the annual time scale (Figure 3.1-4). Data

smoothing increases the correlation between the two annual 8180 reco1'ds (1'=0.58, at 99.5%

level) for a 5-year and 11-yea1' running average. The shorter 8 180 coral record (RUS-93b) also

27

conelates with the RUS-95 record. On a seasonal basis the correlation coefficient is 0.62 (at

the 99.5% level), but on the annual time scale, the correlation is not significant (r= -0.16) and

decreases with data smoothing (not shown here). In addition, the annual coral 8180 record

from a fourth colony (RUS-1) in the same environment (Pätzold and Klein, unpublished data)

was correlated with our 1'ecords on the basis of two samples/year. Table 3.1-1 summaries the

correlation coefficients of the four co1'al annual 8 180 1'ecords.

(a)

-4.2CD-4.0 0

-3.8 0..

RUS-95 -3.60

~0"--'

-3.4 0-3.2 co

T-

-4.0 C/O

-3.0 co-3.8 -2.8

'-

---- 0co -3.6 -2.6 000.. -3.40

~ -3.2 RUS-93a

0 -3.0coT-

C/O -2.8-2.6

1900 1920 1940 1960 1980 2000Year

(b) -3.8CDRUS-9500..

-3.6 0

~0"--'

-3.8 0co

-3.4~

C/O

---- coco -3.6 '-0 00.. -3.2 0

0 -3.4~0--0co -3.2~

C/O

-3.0

1900 1920 1940 1960 1980 2000Year

Fig. 3.1-4: Comparison of the coral 8180 records of this study (RUS-93a) with RUS-95 from

Felis (1999). a) Bimonthly time series r= 0.69 (at 99.5% level). b) Annual time series

1'=0.41 (at 99.5% level).

28

Table 3.1-1: Summary of the correlation coefficients (I' values) for the mean annual 0180

records of all the corals collected from Ras Umm Sidd (nolihern Red Sea proper). The

annual records were calculated using bimonthly data from January/February to

NovemberiDecember. The annual record of RUS-l is based on two annual isotope values,

one in summer and one in winter.

RUS-93a RUS-93b RUS-1 RUS-95

RUS-93a +0.40 (n=49) +0.43 (n=86) +0.41 (n=95)

(96 yrs) ---

RUS-93b +0.40 (n=49) 0.00 (n=49) -0.16 (n=49)

(50 yrs) ----

RUS-1 +0.43 (n=86) 0.00 (n=49) +0.19 (n=86)

(87 yrs) ---

RUS-95 +0.41 (n=95) -0.16 (n=49) +0.19 (n=86)

(245 yrs) ---

Comparison and calibration with instrumental sea surface temperatures (SSTs)

As the COADS sea-surface temperature record is non-continuous, the 8180 record of RUS­

93a was carrelated with four different time intervals ofcontinuous COADS SST (1897-1916,

1919-1939, 1946-1966, and 1976-1992). On the bimonthly time scale the correlations are

highly significant at the 99.5% level. The correlation coefficients range from -0.65 to -0.81

for RUS-93a (Figure 3.1-5a). For RUS-93b, the correlation coefficient is -0.73 in the time

interval1946-1966 and -0.71 in the time interval 1976-1992 (not shown). On the annual time

scale, the correlations of RUS-93a vary with regard to the different time intervals but are

nevertheless significant (confidence levels varied from 95% to 99.5%). The correlation

coefficients (r) are -0.62,-0.38, and -0.43 and significant at the 99.5%, 97.5% and 97.5%

levels, for the time intervals 1919-1939, 1946-1966, and 1976-1992, respectively (see also

Table 3.1-3). For the time interval 1898-1916 the correlation is not significant (Figure 3.1-5b).

Correlation of the annual RUS-93b 8 180 record and the annual sea-surface temperature is not

significant (not shown).

As summarised in Table 3.1-2, the bimonthly 8 180 records ofour two coralswere calibrated

with the COADS SSTs (Slutz et al. 1985; Woodruff et al. 1993) far the most recent time

intervals of continuous observations (1946-1966 and 1976-1992). These calibrations show a

SST-coral 8180 relationship that varies from 0.18%o;oC to 0.15%o;oC. When calibrated with

the IGOSS dataset from the time interval of 1982-1992 (Reynolds and Smith 1994) the

relationships varied from 0.12%%C for RUS-93a, to 0.158%%C for RUS-93b.

29

(a)30

Ü~

f-Cf)

25 Cf)

Cf)0«

-4.0 0()

-3.8 20ro -3.600- -3.4e

~ -3.20 -3.0~00 -2.8

-2.6

1900 1920 1940 1960 1980 2000Year

(b)-3.8

ro -3.600-

e -3.4~~

0~ -3.200

-3.0

r=O

1900

r= -0.62

1920

r= -0.38

1940 1960Year

r= -0.43

1980

27

26 Ü~

25f-Cf)Cf)

24 Cf)0«

23 0()

22

2000

Fig. 3.1-5: Time series of RUS-93a 8 180 and the available ship-based sea-surface temperature

(SST) ofthe COADS dataset (Slutz et al. 1985; Woodruffet al. 1993), centred at 27.0oN,

35.0oE. a) Camparisan of the bimonthly RUS-93a coral 8 180 record and the

corresponding SST in the different time intervals. b) as in (a), but considering annua1

mean values.

Long-term trends in 8180

The long-term 8 180 trend for RUS-93a is linear and toward lighter values from 1898 to 1992

(Figure 3.1-4). Regression analysis ofthe annual time scale yields a trend of -0.22%0 in 8180

(1'=0.54 at the 99.5% level), equivalent to an increase of about 1.1 oe in temperature when

0.165%0/oe is used (Felis 1999). In the case of RUS-93b, the long-term trend consists in an

increase of about 0.1 oe during the last 50 years (1943-1992).

30

Table 3.1-2: Linear regression equations and correlation coefficients l' between coral8 180 and

sea-surface temperature (SST) data based on bimonthly (bi) and mean al1lmal (ma)

sampling clensities for RUS-93a anel RUS-93b (time interval: 1946-1966 and 1976-1992),

respectively. The indepenelent variable (x) is coral8 180 anel the elepenclent variable (y) is

SST. The IGOOS sea-surface temperature clata set is satellite-basecl, Cl ° square box

centrecl at 27.5°N, 34.5°E) and obtained from the internet (http://ingriel.lclgo.columbia.edu

/SOURCES/). The designation n. s. means not significant.

Interval (Resolution) Coral core Regression equation r (r2)

COADS SST

1946-1966 (bi) RUS-93a Y(SST°C)=-5A9 (Ö I8O)+6.36 -0.63 (0040)

1976-1992 (bi) RUS-93a . Y(SSTOC)=-6.60(Ö I8O)+2.21 -0.74 (0.55)

1946-1966 (ma) RUS-93a Y(SSTOC)=-1.36(Ö 18O)+20.58 -0.37 (0.14)

1976-1992 (ma) RUS-93a Y(SSTOC)=-1.22(Ö 18O)+20.96 -0044 (0.19)

1946-1966 (bi) RUS-93b Y(SST°C)=-5.87(Ö I8O)+5.32 -0.72 (0.52)

1976-1992 (bi) RUS-93b Y(SSTOC)=-6.6(Ö I8 O)+2.82 -0.71 (0.50)

1946-1966 (ma) RUS-93b Y(SSTOC)=-0.02(Ö 18O)+25.17 n.s.

1976-1992 (ma) RUS-93b Y(SSTOC)=-0.29(Ö 18O)+24.26 n.s.

IGOOS SST

1982-1992 (bi) RUS-93a Y(SSTOC)=-8.19(Ö 18O)-4.02 -0.75 (0.56)

1982-1992 (ma) RUS-93a Y(SSTOC)=-0.26(Ö 18O)+23.73 n.s.

1982-1992 (bi) RUS-93b Y(SSTOC)=-6.34(Ö I8O)+2.95 -0.64 (0041)

1982-1992 (ma) RUS-93b Y(SSTOC)=0.66(Ö I8O)+26.90 n.s.

Spectral analysis and cross-spectral analysis

The Blackman-Tukey spectral analyses of the detrended bimonthly coral Ö180 time series of

RUS-93a and RUS-93b are shown in Figure 3.1-6a. There are several peaks, but the most

significant peaks (minimum of 80% confidence level) are at periods of 10 (RUS-93a) and 3.6

(RUS-93 b) years. The cross-spectral analysis between the two time series indicates that they

are also coherent (at the 80% level) for periods of 10 and 3.6 years (Figure 3.1-6b). Even

more interesting, however, is the highly coherent peak at aperiod of 5.8 years (Figure 3.1-6b).

31

(a) (b)

1.0 RUS-93a 1.0 10 yrs 5.8 yrs 310 yrs yrs

» » ' ~."!:::: 0.8 ."!:::: 0.8

Cf) Cf)

C CQ) 0.6 Q) 0.6'TI 'TI

cuOA

cuOA..... ......- .-

ü üQ) Q)0- 0.2 0- 0.2 RUS-93a

Cf) (J)

0.0 0.0US-93b

0.0 0.1 0.2 0.3 OA 0.5 0.6 0.0 .2 ;3 OA 0.5 0.6

1.2 RUS-93b 1.0» 1.0 0.8."!::::Cf) »c 0.8 üQ) c 0.6'TI Q)

m 0.6 .....Q)..... ..c OAÜ DA 0

Q) 00-

0.2 0.2Cf)

0.0 0.0

0.0 0.1 0.2 0.3 OA 0.5 0.6 0.0 0.1 0.2 0.3 OA 0.5 0.6Frequency Frequency

Fig. 3.1-6: a) The results of detrended bimonthly Blackman-Tukey spectral analysis of the

period 1944-1992 (Jenkins and Watts 1968; Paillard et al. 1996) for RUS-93a (top) and

RUS-93 b (bottom). The periods of statistically sign ificant peaks (80% level) are shown in

units of years. b) Results of B lackman-Tukey cross-spectral analysis of the same period

between the detrended bimonthly coral 8 180 records of RUS-93a and RUS-93b (top) and

coherency - the correlation coefficient as a function of frequency (bottom).

DISCUSSION

Comparison of foul' coral 8 180 records from Ras Umm Sidd revealed similarities in average

seasonal amplitudes and average annual growth rates but also differences in the correlation

coefficients (Table 3.1-1). The average seasonal 8180 amplitude and annual growth rate are

very similar in all bimonthly sampled corals (RUS-93a, RUS-93b; this study and RUS-95;

Felis 1999). Smoothing the 8 180 data either increased the correlation coefficient (1') or did not

affect it at all. Apparently, all corals respond similarly to the same environmental signals.

Kuhneli et al. (1 999b) suggest that differences in the 8 180 records can result from differences

in age determinations, a local variability in the temperature distribution, or due to individual

coral metabolism. Because the corals in Ras Umm Sidd grow less than about 500 m apart

32

from each other, local temperature differences are not presumed to play an important role. We

suggest that differences in biology are primarily essential for the observed differences in the

coral 0180 records ofRas Umm Sidd.

The temperature/0 180 calibration determined to be applied in bimonthly sampling ranged

from 0.18%%C to 0.15%o/OC. Although these values differ from the widely recognised values

of 0.22%%C for calcite (Epstein et al. 1953) and 0.23%%C for aragonite (Grossman and Ku

1986), they are similar to values determined on a weekly scale by Gagan et al. (1994) for

Porites from Great Barrier Reef (0. 18%o;oC) and by Quinn et al. (1998) for a New Caledonian

Porites on a monthly scale (0.172%%C). Interestingly, Felis (1999) determined a value of

0.165%o;oC for RUS-95 originating from the same environment. This value is similar to our

average value. Therefore, we conclude that 0.165%%C is the most reasonable value for

Porites spp. from Ras Umm Sidd, Red Sea. This value is not obtained when the coral 0180 is

calibrated with the IGaS S data set for the time interval of 1982-1992. It should be kept in

mind, however, that the IGOSS data set is satellite-based. However, the correlation between

IGOSS and COADS data sets is very good (r=0.99) on a seasonal time scale, but it is nearly

zero at an annual time scale.

Aglobai SST data set represents a much larger region than the site of coral growth in Ras

Umm Sidd. Therefore, it is necessary to employ the nearest grid box from aglobai SST data

set and compare as many local coral 0180 time series as possible. In this context, we have

correlated each annual coral 0180 record individually with the COADS SST data set (four

time intervals and the time from 1907-1992, except for the data gaps) and the averaged annual

0180 records from the three longest Ras Umm Sidd cores (Table 3.1-3). As shown in Table

3.1-3, at least one time interval has no significant correlation upon comparing the COADS

data set with the individual RUS coral 0180 records. But when the COADS SST data set is

correlated with the averaged annual 0180 record from the three corals, the correlation is quite

good and significant for all time intervals. In the period from 1907 to 1992 (except for the

data gaps in the COADS data set) the correlation coefficient between SST and the mean coral

0180 record improves from -0.45 to -0.68 (Figure 3.1-7). Record RUS-93a shows the highest

correlation coefficient (r=-0.60), because the slope of the 0180 values is similar to the SST

slope at the beginning ofthis century. The long-term trend in 0180 is also comparable with the

SST long-term trend (about 1°C increase). The other records show less similarity with the

temperature trend (Table 3.1-3).

33

Table 3.1-l overview of the correlation coefficients obtained by three different mean annual

coral 8 180 records at Ras UI11I11 Sidd and related to the mean al1l1Ual sea-surface

temperature (COADS SST). Due to gaps in the SST data set, four continuous time

intervals between 1907 and 1992 were chosen. By omitting the data gaps, a single time

interval (1907-1992) is used for easier cOli1parison. The l11ean annual cora18 180 of all the

three cores was calculated. It is evident that by increasing the number of coral records the

correlation is improves. The correlation coefficient increased from -0.45 to -0.68 in the

time interval from 1907 to 1992.

1907-1916 1919-1939 1946-1966 1976-1992 1907-1992 Long-time Trend

r r r r r 8180 (Temp. °C)

RUS-l -0.60 -0.10 -0.09 -0.40 -0.50 0.15%0 (0.88 °C)

(A) (95%) (n.s.) (n.s.) (90%) (99.5%)

RUS-95 -0.55 -0.27 -0.72 -0.49 -0.45 0.07%0 (0.45°C)

(B) (95%) (n.s.) (99.5%) (97.5%) (99.5%)

RUS-93a -0.07 -0.62 -0.38 -0.43 -0.60 0.19%0 (1.1 °C)

(C) (n.s.) (99.5%) (97.5%) (97.5%) (99.5%)

(A+B+C) -0.54 -0.56 -0.60 -0.53 -0.68 0.14%0 (0.84°C)

(95%) (99.5%) (99.5%) (97.5%) (99.5%)

N 10 21 21 17 69

It has been discovered in tree-ring studies that each chronology must be developed from more

than one tree (ideally at least 10 trees). The intention underlying this process is to develop an

average master chronology for a certain site (requirement of the International Tree-Rings Data

Bank "ITRDB"). Similarly, it is good practice to analyse more than one coral from a certain

locality to improve the correlation between the coral 8 180 and regional SST, and hence

develop a master coral chronology for this locality. A 6-coral average composite 8 180 record

provides a better correlation than individual coral 8180 records with monthly SST records

obtained from the Clipperton Atoll, eastern Pacific (Linsley et al. 1999).

The 10-year period found in the 8 180 records of RUS-93a and RUS-93b, and in the cross­

spectral analysis of both 8 180 records may reflect the influence of the North Atlantic

Oscillation (NAO) on the northern Red Sea. A NAO index based on ice-core-data from

Greenland (1648-1991) shows maximum spectral power at periods between 9-11 and between

5-7 years (Appenzeller et al. 1998). A highly coherent 5.8-year cyc1icity is also observed in

our coral records. This period is evident in the coral 8 180 of RUS-95, and it is also the most

prominent period in the co-spectrum of Northern Atlantic Oscillation (NAO) and EI Nifio

Southern Oscillation (ENSO) (Felis 1999).

34

1900 1920 1940 1960 1980 2000

EOo0..

-3.8

-3.6

-3.4

-3.2

26

25

24

23

Go--....I­(f)(f)

(f)o«oü

EOo0..

oco

60CU

C'0(j)

I(f)::J0::::

-3.8

-3.6

-3.4

-3.2

-3.0

(e)

r= -0.60

26

25

24

23

Go--....I­(f)(f)

(f)o«oü

~ -3.4

EO -3.8o0..~ -3.6o--....

I­(f)(f)

(f)o«oü

25

23

24

26

r= -0.50-3.2

-3.0

"'""I(f)::J0::::

EOo0..

oco

60

1.0(j)

I(f)::J0::::

-3.8

-3.6

-3.4

-3.2 -!---r--..,.--..-----.--..:,..-....,..--r--....--.----.

26

25

24

23

Go

I='(f)(f)

(f)o«oü

1900 1920 1940 1960Year

1980 2000

Fig. 3.1-7: Camparisan between annual COADS SST and the various coral 8180 of cores from

Ras Umm Sidd (northern Red Sea). a) SST and coral 8 180 from RUS-95 (Felis, 1999). b)

SST and coral 8 180 from RUS-l (Pätzold and Klein, unpublished data). c) SST and coral

8 180 from RUS-93a (this study). d) SST and mean annual coral 8 180 of all three cores.

35

The cross-spectral analysis of the mean annual coral 8180 of the cores RUS-93 (a and b) and

the annual Southern Oscillation Index (SOl) after Cole (1998) as obtained from the internet

(http://ingrid.ldgo.Columbia.edu/SOURCES/.indices), indicates a highly coherent peak at a

period of3.6 years (Figure 3.1-8). Such a 3.6-year peak has been found in other coral records,

such as from New Guinea (Tudhope et al. 1995), Tarawa (Cole et al. 1993) and New

Caledonia (Quinn et al. 1998). In addition, it has been reported to exist in temperature records

of the northern hemisphere (Mann and Park 1994) and was also demonstrated in the analysis

of Southern Oscillation Index (Allen and Smith 1996). The Southern Oscillation Index (SOl)

is based on the standardised sea-level pressure (SLP) between Tahiti (central Pacific) and

Darwin (Australia). A negative value of the SOl indicates the warm EI Nifio phase of the

Southern Oscillation (SO), a positive value the "cold" La Nifia phase. During high (positive)

SOl, the precipitation during the SW monsoon season is above average. Evidence for

atmospheric teleconnections associated with extreme phases of the SOl is shown in the record

of the Nile floods. Quinn (1992) showed that years of low Nile flood were associated with a

low (negative) SOL We propose that the SW monsoon has also affected the climate in the

northern Red Sea, at least far about the last 50 years (1944-1992). During intensive SW

monsoons the narthern Red Sea receives more fresh water from the south (through the

thermohaline circulation), which influences the 8180 in sea water (salinity signal) and

secondarily, the coral 8180.

8>-

:t:=Cf) 6c(J)

""0

cu 4............()

2 SOl(J)a.

Cf) RUS-93a+b0

0.0 0.1 0.2 3 004 0.5 0.6

1.0

>- 0.8()c 0.6(J)....(J)

004..c0U 0.2

0.0

0.0 0.1 0.2 0.3 004 0.5 0.6Frequency (cycles per year)

Fig. 3.1-8: Blackman-Tukey cross-spectral analysis for the period 1944-1992 between

detrended mean annual coral 8 180 ofRUS-93a and RUS-93b, and the southern oscillation

index (SOl) (upper panel). Coherency (80% level) between the two records occurs at 3.6

years period (lmver panel). The horizontal line in the lower panel shows the 80%

confidence level.

36

CONCLUSIONS

In this study the Ö180 records of two coral Porites spp. cores from the subtropical northern

Red Sea provide a bimonthly resolution for the past 96 and 50 years. These records correlate

differently with other coral Öl80 records from the same site on an annual time scale. We

suggest that individual biological effects influence the ÖI80 records. All bimonthly sampled

corals reflect an averaged environment signal as a result of similar annual growth rates and

seasonal coral Ö180 amplitudes. The longest three annual coral ÖI8 0 records (RUS-l, RUS­

93a and RUS-95) also correlate differently with the regional COADS SST. The correlation

coefficient varied from -0.45 (RUS-95) to -0.60 (RUS-93a). When the COADS SST data set

is correlated with the averaged annual composite 3-coral ÖI80 record between 1907 and 1992,

a highly significant and very good correlation (1'=-0.68) is obtained. Therefore, we advise the

analysis of more than one coral from a certain locality to receive a better coral chronology for

coral-based paleoclimate studies. A long-term temperature rise of 1.1°e between 1907 and

1992 is infelTed from the isotope record (RUS-93a). This long-term trend is comparable with

the long-term trenel in instrumentally measureel COADS SST (about 1°C) in the northern Reel

Sea, excepteel elata gaps. The ÖI80/SST graelient varieel between 0.15 to 0.18%0, with a mean

value of 0.165%0 as in previous stuelies (Felis 1999). Cross-spectral analyses of coral Ö180

recorels of RUS-93a anel RUS-93b show coherences at perioels of 3.6, 5.8 anel 10 years. This

result inelicates the impact of the North Atlantic Oscillation (NAO) anel the Southern

Oscillation (SO) i.e. SW monsoon, on the northern Reel Sea coralö 180.

ACKNOWLEDGEMENTS

We gratefully acknowleelge M. Segl anel B. Meyer-Schack, who performeel the stable-isotope

measurements. We also thank the Government of Egypt anel Mr. A. El-Ibiary (NIOF) for

support anel permission to retrieve core sampies, anel the Gennan Embassy in Cairo for

sending the sampies to Germany. We are most inelebted to T. Felis for provieling his data

(RUS-95) and to 1. Pätzold anel R. Klein far provieling unpubli.shed data (RUS-l). The quality

ofthe manuscript was greatly improveel by the critical reading by J. Bijma. W. HaIe improved

the English of the manuscript. Our work has greatly benefited from discussions with H. Arz,

H. Kuhnert anel T. Felis. This work was partly supporteel by the Red Sea Program on Marine

Sciences (RSP), funeleel by the German Feeleral Ministry for Eelucation, Science, Research and

Technology (BMBF) through grant 03F0151A, anel a grant ofthe Graduierten Kolleg "Stoff­

Flüsse in marinen Geosystemen" funeleel by the Deutsche Farschungsgemeinschaft (DFG).

37

REFERENCES

Ahmed F, Sultan SAR (1989) Surface heat fluxes and their comparison with the oceanic heat

flow in the Red Sea. Oceanol Acta 12: 33-36

Allen MR, Smith LA (1996) Monte Carlo SSA: Detecting irregular oscillations 111 the

presence of colored noise. J Clim 9: 3373-3404

Appenzeller C, Stocker TF, Anklin M (1998) North Atlantic Oscillation dynamics recorded in

Greenland ice cores. Science 282: 446-449

Barry RG Chorley RJ (1998) Atmosphere, Weather and Climate, Routledge, London, i h ed

pp. 1-409

Charles CD, Hunter DE, Fairbanks RG (1997) Interaction between the ENSO and the Asian

monsoon in a cora1 record of tropical climate. Science 277: 925-928

Cember RP (1998) On the sources, formation, and circulation of Red Sea deep water. J

Geophys Res 93: 8175-8191

Cole JE, Fairbanks RG, Shen, GT (1993) Recent variability in the southern oscillation.

Science 260: 1790-1793

Cook ER, D' Arrigo RD, Briffa K R (1998) A reconstruction of the North Atlantic Oscillation

using tree-ring chronologies from North America and Europe. The Holocene 8: 9-17

Crowley TJ, Quinn TM, Taylor FW, Henin C, Joannot P (1997) Evidence for a volcanic

cooling signal in a 335-year coral record from New Caledonia. Paleoceanogr 12: 633­

639

Dunbar RB, Wellington GM, Colgan MW, Glynn PW (1994) Eastern Pacific sea surface

temperature since 1600 AD: The 8180 record of climate variability in Galapagos corals.

Paleoceanogr9: 291-315

Epstein S, Buchsbaum R, Lowenstam HA, Urey HC (1953) Revised carbonate-water isotopic

temperature scale. Geol Soc Am Bu1l64: 1315-13225

Eshel G, Naik NH (1997) Climatological coastal jet collision, intermediate water formation,

and the general circulation ofthe Red Sea. J Phy Oceanogr 27: 1233-1257

38

Eshel G, Cane MA Blumenthal MB (1994) Modes of subsurface, intermediate, and deep

water renewal in the Red Sea. J Geophys Res 99: 15941-15952

Felis T (1999) Climate and ocean variability reconstructed from stable isotope records of

modern subtropical cora1s (northern Red Sea). Berichte, Fachbereich Geowissen­

schaften, Universität Bremen, Nr 133, 111 Seiten, Bremen

Felis T, Pätzold J, Wefer G, Fine M, Loya Y, Nawar AH (1998) First results of a coral-based

history ofrecent climate in the northern Red Sea. Zbl Geol Paläont Teil I, H 1/2: 197­

209

Gagan MK, Chivas AR, Isdale PJ (1994) High-resolution isotopic records from eorals using

ocean temperature and mass-spawning chronometers. Earth and Planet Sei Lett 121:

549-558

GoodfriendGA (1991) Holocene trends in 180 in land snail shells from the Negev Desert and

their implications for changes in rainfall source areas. Quat Res 35: 417-426

Grossman EL., Ku T-L (1986) Oxygen and earbon isotope fractionation in biogenic aragonite:

Temperature effects. Chern Geol 59: 59-74

Heiss GA (1994) Coral reefs in the Red Sea: Growth, Produetion and stable isotopes. Geomar

report 32, Ph. D. Thesis, Kiel

Heiss GA (1996) Annual band width variation in Porites sp. from Aqaba, Gulf of Aqaba, Red

Sea. Bull Mar Sci 59: 393-403

Heiss GA, Dullo W-C (1997) Stable isotope reeord from recent and fossil Forites sp. in the

northern Red Sea. Coral Res Bull5: 161-169

Issar AS (1990) Water shall flow from the rock; hydrology and climate in the lands of the

Bible. Springer, Berlin Heidelberg New York, pp 1-213

Jenkins GM, Watts DG (1968) Spectral analysis and its applications. Holdenday, Oakland, pp

1-525

Klein R, Loya Y, Gvirtzman G, Isdale PJ, Susic M (1990) Seasonal rainfall in the Sinai

Desert during the late Quaternary inferred from fluorescent bands in fossil corals.

Nature 345: 145-147

39

Klein R, Pätzold J, Wefer G, Loya Y (1992) Seasonal variations in the stable isotopic

composition and the skeletal density pattern of the coral Porites lobata (Gulf of Eilat,

Red Sea). Mar Biol 112: 259-263

Klein R, Pätzold J, Wefer G, Loya Y (1993) Depth-related timing of density band formation

in Porites spp. corals from the Red Sea inferred from X-ray chronology and stable

isotope composition. Mar Ecol Prog Ser 97: 99-104

Klein R, Tudhope AW, Chilcot CP, Pätzold J, Abdulkarim Z, Fine M, Falliek AE, Loya Y

(1997) Evaluating southern Red Sea corals as a proxy record for the Asian monsoon.

Earth Planet Sci Lett 148: 381-394

Knutson DW, Buddemeier RW, Smith SV (1972) Coral chronometers: Seasonal growth bands

in reefs corals. Science 177: 270-272

Kulmert H, Pätzold J, Hatcher BG, Wyrwoll K-H., Eisenhauer A, Collins LB, Zhu ZR, Wefer

G (1999a) A 200-year coral stable isotope record from a high-latitude reef off western

Australia. Coral Reefs 18: 1-12

Kuhnet"t H, Pätzold J, Wyrwoll K.-H, Wefer G (1999b) Monitoring climate variability in coral

stable isotopes from Ningaloo Reef, Western Australia. Int J Earth Sci (in press)

Levitus S, Boyer T (1994) World Ocean Atlas 1994, volume 4: Temperature, NOAA Atlas

NESDIS 4, U. S. Department ofCommerce, Washington, D. C.

Linsley BK, Messier RG, Dunbar RB (1999) Assessing between-colony oxygen isotope

variability in the coral Porites lobata at Clipperton Atoll. Coral Reefs 18: 13-27

Mann MA, Park J (1994) Global-scale modes of surface temperature variability on

interannual to century timescales. J Geophys Res 99: 25819-25833

Morcos SA (1970) Physical and chemical oceanography of the Red Sea. Oceanogr Mal' Biol

Ann Rev 8: 73-202

Paillard D, Labeyrie L, Yiou P (1996) Macintosh program performs time-series analysis. Eos

Trans AGU 77: 379

Pätzold J, (1984) Growth rhythms recorded in stable isotopes and density bands in the reef

coral Porites lobata (Cebu, Philippines). Coral Reefs 3: 87-90

40

Pätzold J (1986) Temperature and C02 changes in tropical surface waters of the Philippines

during the past 120 years: record in the stable isotopes of hermatypic corals. Berichte­

Reports, Geol -Paläont. Inst Uni Kiel (in German), Nr 12, pp 1-92, Kiel

Quinn WH (1992) A study of Southern Oscillation-related climatic activity for A. D. 622­

1900 incorporating Nile River flood data. In: Diaz HF, Markgraf V (eds) EI Nifio­

Historical and paleoclimatic aspects of the Southern Oscillation. Cambridge Uni Press,

Cambridge, pp 119-149

Quim1 TM, Crowley TJ, Taylor FW (1996) New stable isotope results from a 173-year coral

from Espiritu Santo, Vanuatu. Geophys Res Lett 23: 3413-3416

Quinn TM, Crowely TJ, Taylor FW, Henin C, Joannot P, Join Y (1998) A multicentury stable

isotope record from a New Caledonia coral: Interannual and decadal sea surface

temperature variability in the southwest Pacific since 1657 A. D. Paleoceanogr 13:

412-426

Reynolds RW, Smith TM (1994) Improved global sea surface temperature analyses. J Clim 7:

929-948

Schrag DP (1997) Temperature and salinity history of the northern Red Sea from high­

resolution elemental and stable isotope records (abstract). Eos Trans AGU: 78 (46),

F387

Schrag DP, Eshel G, Moore MD (1997) Tropics-outer tropics interactions: Lessons from Red

Sea corals I (abstract), Eos Trans AGU: 78 (17), S190

Siedler G (1969) General circulation of water masses in the Red Sea. In: Degens ET, Ross

DA (eds) Hot brines and recent heavy metal deposits in the Red Sea, a geochemical and

geophysical account. Springer, Berlin Heidelberg New York, pp 131-137

Slutz RJ, Lubker SJ, Hiscox JD, Woodruff SD, Jenne RL, Joseph DH, Steuer PM, Elms JD

(1985) Comprehensive Ocean-Atmosphere Data Set. Release 1, 268 pp, NOAA

Environmental Research Laboratories, Climate Research Program, Boulder, CO

Tudhope AW, Shimmield GP, Chi1cott CP, Jebb M, Fallick AE, Dagleish AN (1995) Recent

changes in climate in the far western equatorial Paeifie and their relationship to the

southern Oseillation: Oxygen isotope records from massive eorals, Papua new Guinea.

Earth Planet Sei Lett 136: 575-590

41

Wefe..r G, Berger WH (1991) Isotope palaeontology: growth and composition of extant

calcareous species. Mar Geol100: 207-248

Woodruff SD, Lubker SJ, Wolter K, Worley SJ, Elms JD (1993) Comprehensive Ocean­

Atmosphere Data Set. Release 1a: 1980-1992, Earth System Monitor, Vol 4, No 1,

NOAA

42

3.2

MODERN AND FOSSIL HOLOCENE MARINE CLIMATE RECORD IN

MOLLUSCS (TRIDACNA SPP.) FROM THE NORTHERN RED SEA

(submitted to Terra Nova)

ABSTRACT 43

INTRODUCTION 44

Gastropods and bivalves in the Red Sea

Stable isotopes in molluscs as paleoclimatic records

Afrasian climate during the Holocene

General charaeteristies of the study area and geologie setting

MATERIAL AND METHODS 47

RESULTS 49

DrSCUSSION 51

CONCLUSIONS 54

ACKNOWLEDGEMENTS 54

REFERENCES 55

43

Modern and fossil Holocene marine climate record in molluscs

(Tridacna spp.) from the northern Red Sea

Moustafa, Y. A., J. Pätzold, Y. Loya and G. Wefer

Fachbereich Geowissenschaften, University ofBremen, Postfach 330440 28334 Bremen, Germany

Fax: 0049 (421) 2183116

email: yaser@uni-bremen.de

ABSTRACT

This study focuses on the isotopic composition of oxygen in fossil Tridacna spp. from an

exposed Holocene reef terrace at about 2 m above the present sea-level from the northern

Gulf of Aqaba (Red Sea), covering the late- and mid-Holocene periods from 5460 to 1400 14C

years BP. Arecent Tridacna sp. was collected alive in June 1996 and analysed for comparison

with the fossil sampies. The high-resolution 8180 profile of this recent Tridacna sp. shows a

range of 1.17%0 in 8 180 , which is consistent with today sea-surface temperature (SST) range

of about 5.3°C. The fossil Tridacna spp. 8180 records encompass a range between 1.27%0 and

1.00%0, which is equivalent to seasonal temperature variations between 6°C and 4.5°C. Mean

8180 values of various fossil Tridacna spp. range between 1.20%0 and 2.29%0. These

variations in 8180 (about 1.1 %0) are mainly due to changes in the isotopic composition of sea

water. A likely climatic cause of these mean 8 180 variations in the Tridacna spp. might

consist in an enhancement or a reduction of the Indian and African monsoons, leading to

either a greater or SCal'ce rainfall in the highland of east Africa and the Red Sea region, and

then to the respective salinity changes in the surface layer ofthe northern Red Sea. The results

presented corroborate hieroglyphical and historical documents of high/low Nile levels during

the last about 6000 years BP and also the thickness of varved sediment cores from

northeastern Arabian sea, which were affected by the intensity change of the AfricanlAsian

monsoon.

44

INTRODUCTION

Gastropods and bivalves in the Red Sea

The giant Tridacna squamosa clams are found embedded in reef flats all over the Red Sea.

They reach 30 cm in diameter 01' more and their growth is restricted to clear, shallow water.

Strombus sp. is a conch living in themodern fringing reefs ofthe Red Sea and is also found in

fossil reef terraces along the coast of southern Sinai together with other fossil molluscs like

Tridacna sp. They belong to the Mesogastropoda, a large group occupying a great variety of

niches in the coral reefs of the Red Sea. Strombus is a typical sand-dweller (Mastaller 1987).

Generally, the molluscan fauna of the Red Sea is entirely of Indo-Pacific origin (Mastaller

1987).

Stable isotopes in molJuscs as paleoclimatic records

Molluscs, in general, are believed to exert only a minimal vital effect on isotopic composition

(Milliman 1974; Jones 1985; Wefer and Berger 1991), thus their isotopic values are

representative of paleoenvironmental as well as paleoclimatical conditions. Although many

other factors influence the isotopic composition of mollusc shells, water temperature, salinity

(for oxygen) and the total dissolved carbon content (for carbon) have been recognised as

major factors controlling the isotopic compositions of oxygen and carbon in the carbonate

shells of marine molluscs (Anderson and Arthur 1983). Therefore, high resolution records of

0180 in Strombus sp. and Tridacna sp. can be used to reconstruct paleotemperatures by means

of the calibration equation established by Epstein et al. (1953) 01' Grossman and Ku (1986).

Wefer (1985); Pätzold et al. (1991) used this method to determine paleotemperatures from

Strombus and Tridacna species at different localities around the world.

The isotopic oxygen compositions of carbonate shelled organisms have also been used as

qualitative indicators of salinity (e.g., Rostek et al. 1993; HernIeben et al. 1996). This

application of stable isotopes is possible due to the fact that meteoric waters generally have

lower 0180 values than marine water. The oxygen isotopic signature of closed 01' nearly closed

water bodies is often related to changes in the water's salinity as a response to changes in

evaporation, precipitation (humidity), and mixing of waters from different sources (Rostek et

al. 1993). The dependence of the sea-surface water 0180 on salinity varies in the different

oceanic regions (Craig and Gordon 1965; Fairbanks et al. 1992). For the Red Sea, relationship

between sea-surface water 0180W and salinity amount to a value ofO.29%0 0180w/1%0 salinity

(Craig 1966).

Afrasian climate during the Holocene

Despite their various latitudes, the Afrasian arid areas, are mainly located north of the summer

monsoon zone and are sensitive to the variation of the monsoon system. Ji et al. (1993)

suggested that shOli-tenn (less than 100 years) temperature and rainfall variations occur along

45

the Afrasian arid belt over the last 500 years. Riehl and Meitin (1979) also concluded that

Nile discharges occured at Cairo in a periodicity of approximately 100 years during the last

about 1400 years. The record of Nile discharges provides a measure of the amount and effects

of the summer monsoon rain fall over the highlands of Ethiopia (e.g. Bell 1970 and Quinn

1992). During a strong summer monsoon over India and the east African highland the Nile

discharge would be high and vice versa. Hassan (1981) indicated an apparent cOlTelation

between low Nile discharges and cold climate periods in Europe. Thus, the link between

climatic changes in the NOlihern Hemisphere and Nile levels may also be considered in terms

of the covariation of the Nile flood levels (i.e. summer monsoon) and paleoclimatic changes

in Europe. The correspondence between cold conditions in Europe and low Nile levels, and

vice versa, may thus have been characteristic of the entire Holocene. The variability of the

Holocene climate increased about 5000-6000 years ago, when the winter and mean annual

temperatures were generally low (Frenzel and Pachur 1998).

The past weather and climatic conditions were deciphered from the written reports in ancient

Egypt and Middle East from about 5000 years BP (e.g., Henfling and Pflaumbaum 1991).

Written reports from the Mediterranean region exist only since 1300 years BP. Repolis about

water levels of the river Nile document the humid/arid time intervals in eastern North Africa

(Riehl and Meitin 1979). Generally, all these data are useful to study the climate history of

mid- to late-Holocene tünes in East Africa and Middle East.

General eharaeteristies of the study area and geologie setting

The Red Sea is a semi-enclosed basin separated from the Indian Ocean by the narrow Strait of

Bab al Mandab. The adjacent Indian Ocean influences the hydrography and oceanography of

the Red Sea are strongly by surface and subsurface circulation regimes (Morcos 1970;

Ganssen and Kroon 1991; Eshel et al. 1994). The Gulf of Aqaba is also a semi-enclosed water

body ranging from about 28°N to about 300 N and is connected with the northern Red Sea

tlu'ough the Strait of Tiran (Figure 3.2-1). In the northern Red Sea, seasonal sea-surface

temperatures range between 21°C in winter (February - March) and 29°C in summer (August

-September) as inferred from climatological data (COADS monthly SST measured at 2°

square, centered at 35°E/27°N nOlihern Red Sea) for the last 50 years (Slutz et al. 1985;

Woodruff et al. 1993). Seasonal SSTs off Eilat range between 21°C in winter (February ­

March) and 26°C in summer (August - September). The mean annual rainfall in the northern

Gulf of Aqaba at Eilat is 22 mm/year (Friedman 1968) with extremes ranging between 0-70

mm/year (Mergner and Schuhmacher 1974). Rain only falls in the winter months between

November and March. The scarcity and randomness of the rains at Eilat is due to the local

topography. The Dead Sea rift valley (Figure 3.2-1) receives a minor amount of rain coming

from the Mediterranean Sea and is therefore relatively arid, whereas the adjacent mountains

receive more rain and snow in winter (Issar 1990). The impact of winter rains is high as

evaporation is relatively low during this season. In contrast, under the arid and hot conditions

46

28

29

30

31

3732

36

QDead Sea

35

Red Sea

343332

of the Gulf of Aqaba, evaporation is extremely high (3650 mm/year) and greatly exceeds

precipitation (Reiss and Hottinger 1984). The seasonal salinity amplitude is 0.5%0 and the

average salinity is about 40.5%0 (Paldor and Anati 1979). The cloud cover over Eilat is low

throughout the year. The water circulation in the Gulf of Aqaba is mainly thermohaline,

determined by evaporative loss and buoyancy flux. Throughout the year a considerable

volume of warm, relatively low saline and highly oxygenated waters enter the Gulf of Aqaba

from the Red Sea through the Strait of Tiran (Figure 3.2-1) and flow northward against the

prevailing winds. Cooling and evaporation in the northern sector ofthe Gulfleads to a deeper,

southward return flow of denser waters into the Red Sea (Klinker et al. 1976; Reiss and

Hottinger 1984).

30 oE 31

\'-- ---'~ ___l ___J 2rN

Fig. 3.2-1: Location map ofthe Gulf of Aqaba showing the collection sites (star) of modern and

fossil Tridacna spp. from Eilat (northern Gulf of Aqaba, Red Sea).

The Holocene fossil reef terraces are weIl developed along the coast of southern Sinai

between Ras Mohamed National Park and Ras Nasrani (northern Red Sea). These fossil reefs

have an average age of 5200 year BP (Gvirtzman et al. 1992; Gvirtzman 1994). Friedman

(1965) described samples from a fossil reef in the southwest of Eilat (northern Gulf of Aqaba,

Red Sea) having an age of about 4800 years BP. More recently, Moustafa et al. (1999) dated

fossil coral sampIes (Porites spp.) from an elevated fossil reefterrace about 2 m above present

sea level and 200 m distant from the present shoreline, south of Friedman' s (1965) fossil reef.

47

The AMS 14C ages ofthese fossil corals range<:Lbetween 4450 and 5750 years BP, revealing

an average age of about 5000 years BP. In general, the elevated fossil Holocene reef terraces

are weIl described anddated along the Red Sea and Gulf of Aqaba (especially the Arabian

side) (e.g., Dabbagh et al. 1984; Al-Rifaiy and Cherif 1988; Dullo 1990). The diagentic

history of these terraces is also quite weIl studied (e.g., Strasser et al. 1992; Strasser 1995;

Strasser and Strolunenger 1997). The fossil Holocene reef terrace contains numerous well­

preserved fossils, which are suitable for paleoenvironmental studies. This study is focused on

Tridacna clams from a raised fossil Holocene terrace opposite to the InterUniversity Institute

(IUI) Southwest from Eilat, Israel (Figure 3.2-1). The mean 8180 variability in these molluscs

was used to reconstruct past climate variations in the northern Gulf of Aqaba (Red Sea) from

the mid-Holocene till present.

MATERIAL AND METHODS

Fossil mOllusc sampies (Tridacna spp.) were collected from the top of an exposed Holocene

reef terrace at about 2 m above the present sea level and about 200 m distance from the

present shoreline near the InterUniversity Institute (IUI), which is located about 6 km south of

the city of Eilat (Israel) at the north-western end of the Gulf of Aqaba (Red Sea) at 29°31 'N

and 34°56'E (Figure 3.2-1). For comparison with the fossil sampies, arecent Tridacna sp.

(TO) has been collected alive, adjacent to the Holocene terrace from 10m water depth in July

1996.

The Tridacna spp. sampies used in this study are aragonitic molluscs. The shell sampies were

first rinsed with fresh water and then cleaned in an ultrasonic bath for 30 minutes to remove

detrital clay and sand. The sampie was then washed with distilled water and dried in an oven

at 50°C for 24 hours. The remaining adherent clay and sand on the shells were mechanically

removed. The Tridacna sampies were radially sectioned into slabs with a water-cooled rock

saw.

Nine fossil Tridacna spp. sampies (Tl, T3, T4, T7, T8, T10, T13, T18 and T19) were dated

by AMS 14c. The measurements were performed at the Leibniz Laboratory for Radionuclide

dating and Isotope Research, Christian-Albrechts-University Kiel, Germany (Nadeau et al.

1997). The 14C ages were corrected for isotopic fractionation with l3C values as measured by

AMS.

Since direct visual inspection is often insufficient to determine alteration or diagensis, shell

minerology was determined. All sampies were analysed by X-ray diffraction (XRD). For this

process, pieces (about 4 cm2) of shell material from the inner shelllayer of the Tridacna spp.

sampies were ground to a fine powder using an electric agate mortar and pestle. The powder

sampies were analysed using a Philips PW 1800 X-ray diffractometer (Cu,45 kV, 35 mA) at

48

an angle between 20° and 50° (28) with 1;40 (28) per min (2h). X-ray diffractrometry was

carried out at the mineralogical seetion of the Geoscience Department of the University of

Bremen. Trace amounts of calcite were detected, but the estimated proportion of calcite was

less than 2%. Samples containing more than 2% calcite were rejected. Scanning electron

microscopy investigation on fractured shell pieces confirmed their aragonitic composition.

For stable isotope analyses, powdered subsamples from the inner shell layer of four (TO, Tl,

T3 and T13) ofthe Tridacna spp. shells were produced by grinding a channel into the slabs at

regular intervals perpendicular to the growth bands, using a dental drill with an 0.6 mm

diameter bit (Pätzold et al. 1991) (Figure 3.2-2). The sampling intervals were between 0.3 and

0.5 mm. Six of the fossil Tridacna spp. samples (T4, T7, T8, T10, T18 and T19) were first

measured collectively by excising a segment out of the inner shell layer which was

subsequently powdered and thoroughly homogenised. The measurement was repeated five

times for each sample, except for T8 and T7, which were measured four times and forty times,

respectively.

Fig. 3.2-2: Transmitted light photograph of the transversal section ofTridacna spp. is shown.

The profiles (light band) of TO (Ieft) and Tl (right) against the growth direction which

were used for stable oxygen isotope analyses are indicated. The black bar on the upper

side is 5 cm.

Stable isotopic analyses were performed in the Stable Isotope Laboratory of the Geoscience

Department at the University of Bremen. Samples were reacted with 100% orthophosphoric

acid at 75°C in a Finnigan MAT automatie carbonate preparation device to produce carbon

dioxide. The 180 /160 and 13 C/12C ratios ofthe evolved CO2 were determined using aFinnigan

MAT 251 triple-collector isotope ratio mass spectrometer. Oxygen and carbon isotopic

compositions are commonly expressed as parts per mil (%0) relative to the PDB standard

reported in the (8) notation calibrated by means of the NBS 19 standard. Precision (±crs) was

determined using an internallaboratary standard (Solnhofen limestone) and was ±0.07%0 far

the 8180 values (M. Segl personal communication)

49

RESDLTS

Table (3.2-1) summarises the 14C-datings of the various shell sampies and the eorresponding

ealibrated ages (Stuiver and Reimer 1993, Stuiver et al. 1998). As the sampies are still

aragonitie in eomposition, there is strong evidenee that the oxygen isotope values are original

and have not been altcred by diagensis. The 0180 values are plotted against the sampie

distanee (in em) from the inner shell layer to the outer shell layer (Figure 3.2-3). The 0180

data from the modern Tridacna sp. sampie (TO) show a regular cyclic pattern over a three­

year growth period, from about July/August 1993 till Mai/June 1996. The 0180 values range

between 1.74 and 0.57%0, with a mean value of 1.20%0. The fossil Tridacna sp. sampie (Tl;

1500 14C years BP) also shows a regular eyclie pattern over a growth period of 6 years. The

0180 values range between 2.07 and 0.47%0, the average seasonal 0180 amplitude is 1.27%0,

and the mean 0180 is 1.23%0 (Figure 3.2-3). For both sampies T3 (3460 14C years BP) and

T13 (4840 14C years BP), the 0180 profiles exhibit a two-year growth period. The 0180 values

range between 2.09 and 0.70%0 in T3, and between 2.48 and 1.48%0 in T13. The average 0180

value of both sampies is 1.48 and 1.95%0, respeetively (Figure 3.2-3). Table (3.2-2)

summarises the average 0180 values of the various Tridacna spp. sampies.

Table 3.2-1: AMS 14C ages of fossil molluscs from Eilat (northern Gulf of Aqaba) and Sinai

(Southern Gulf of Aqaba). The AMS datings were carried out at the Leibniz-laboratory

AMS facility in Kiel, Germany (Nadeau et al. 1997). The 14C ages were corrected for a

reservoir effect of 400 years and transformed into calendar years using the calibration

program Calib4 (Stuiver and Reimer 1993; StLliver et al. 1998).

Nr. Sampie Lab. Nr. Sampie Type AMS I'+C Age Calibrated Age

Identification (years BP) (years BP)

1 TO ---------- Tridacna sp. Modern 1993-96 AD

2 T8 KIA7239 Tridacna sp. 1400 ± 30 935-

3 Tl KIA3252 Tridacna sp. 1500 ± 40 1047

4 T10 KIA7240 Tridacna sp. 2820 ± 30 2603

5 T19 KIA7245 Tridacna sp. 3000 ± 40 2753

6 T18 KIA7244 Tridacna sp. 3360 ± 40 3221

7 T3 KIA7235 Tridacna sp. 3460 ± 30 3340

8 T4 KIA7236 Tridacna sp. 3980 ± 30 3969

9 T13 KIA7241 Tridacna sp. 4840 ± 40 5131

10 T7 KIA7237 Tridacna sp. 5460 ± 40 5860

0.0

5460 4840 3980 3460 3360 3000 2820 1500 1400 Recentyr SP yr SP yr SP yr sr yr SP yr sr yr sr yr SP yr SP 1996 AD

0.5(T7) (TI3) (T4) (T3) (TI8) (TI9) (TIO) (TI) (T8) (TO)

co 10 •Cl0., . ...

0 ,,~ •~ . ." •9 ••

1.5 ~J1<-0 • •~

~• •

ü •'"~ 2.0 • • ",Ir= • ••• ••••

2.5 •• •••

Growth direetion-"3.0 li i i ij i 11 i rrrrrJ liiiilJiii JTTTTl JTTTTl JTTTTl I I i i I I i i JTTTTl0 30 0.0 0.5 0 5 0.0 04 0 5 0 5 0 5 0 1 0 5 0 1

no. em no. em no. no. no. em no. em

Fig. 3.2-3: Stable isotopic composition of oxygen in shells of Tridacna spp. from Eilat. The

profiles are along the inner shell layer of sampies TO, Tl, T3 and Tl3 (growth direction is

from right to left). T4, T7, T8, Tl 0, T18, and T19 were also sampled from the inner shell

layer. The powder was thoroughly homogenised and bulk sampies were measured

between four and forty times each.

Table 3.2-2: Summary of statistics for the oxygen isotope analyses of fossil Holocene and

recent Tridacna spp. from Eilat (nolihern Gulf of Aqaba, Red Sea).

Sampie Identification Min. Max. Mean (%0) N

(Tridacna spp.) 0 180 (%0) 0 180 (%0) (± St. Dev.)

*) High resolution samples

TO 0.57 1.74 1.20 (±0.38) 36

Tl 0.47 2.08 1.23 (±0.48) 50

T3 0.70 2.09 1.48 (±0.41) 13

TB 1.48 2.48 1.95 (±0.34) 11

**) Bulk samples

T8 1.62 1.84 1.72 (±0.09) 4

T10 2.57 2.74 2.67 (±0.06) 5

T19 2.18 2.41 2.29 (±0.1 0) 5

T18 1.99 2.08 2.04 (±0.04) 5

T4 1.25 1.57 1.39 (±0.13) 5

T7 0.99 1.57 1.29 (±0.14) 40

50

51

DISCUSSION

Analyses of modern and fossil Tridacna spp. sampies from the northern Gulf of Aqaba

indicate that there are significant differences in the mean cOlnposition of oxygen isotopes and

the seasonal amplitude in 8180 among samples of different ages. The 8180 values in all four

high-resolution samp1ed Tridacna spp. shells show a markedly cyclic pattern, with the highest

amplitude in Tl (1500 14e years BP) of 1.59%0, averaging at 1.27%0. By applying the

aragonite temperature equation of Grossman and Ku (1986), the Tridacna sp. (TO) shows an

offset ofO.05%0, taking into account that the mean 8180 ofthe modern Tridacna sp. is 1.20%0

and that the mean 8180 in water sampies from the Gulf of Aqaba is 1.68%0 relative to PDB

standard (Klein et al. 1992). Therefore, the mean 8180 value ofthe modern Tridacna sp. (TO)

indicates that the biological fractionation effects in Tridacna sp. are nearly zero (considering a

measurement error of ±0.07%0). The mollusc is therefore a reliable indicator of SST and

8180W. Mean seasonal sea-surface temperature (SST) amplitudes between 1988-1995 are

about 5.3°e at Ei1at (Genin A, personal communication, 1996) and match the seasona1

amplitude derived from this recent Tridacna sp. (1.17%0), assuming that a temperature change

of l°e is equiva1ent to 0.22%0 in 8180 (Epstein et al. 1953). The prominent variability in the

8180 profile for the recent Tridacna sp. (TO) primarily reflects the seasonal range ofsurface­

water temperature (5.3°e) of the nortl1ern Gulf of Aqaba. The seasonal variation (0.5%0) in

the salinity ofthe surface water ofthe Gulf of Aqaba has a minor effect on the 8180 profile of

the recent Tridacna sp. The seasonal temperature variation calculated from the 8180

amplitude of fossil Tridacna shells are about 4.5°e in the mid-Holocene and about 6°e 1500

14e years BP.

The mean 8 180 values ofthe various Tridacna spp. shells during the last 6000 years BP show

a variation of about 1.1 %0 (Figure 3.2-4a), ranging from 1.20 to 2.29%0. This wou1d represent

a temperature variation during the Holocene of 5°e, which obviously seems to be too high.

Oxygen and carbon isotopes in planktic frominifera (G. ruber) and alkenone-derived paleo­

sea-surface temperatures (north-eastern Arabian Sea) reveal that maximum SST amplitudes of

about 3°e have been observed during the past 5000 years BP, suggesting that intensive

monsoonal influences have also occured during the late Holocene (Doose-Rolinski and

Lückge 1999). If this temperature change occurs in the northern Red Sea to this extent on1y

60% of the Tridacna 8 180 signal during the last 6000 years BP can be explained. Therefore,

the variation of 8180 in Tridacna spp. shells throughout the Holocene may rather be an

additional result of fluctuations in the isotopic composition of oxygen in the surface sea water

of the northern Red Sea.

52

1.2(a)

-- 300(1) 1.400- --e 1.6 E~ -S.00 1.8 200

(j)<Xl >

<,()(J)

ro ~c 2.0ü Zro27/2 High Nile""0

100'C 2.2 .. '/f-

2.4Low Nile

Nile level (ern)

6.0--5 4.0(b)c

0 2.0-.;::;ro.s>

0.0(J)""0 -300(J)

-2.0zE-4.0 "-'

(J)

>(J)

-350 ro(J)

Cf)

""0ro(J)

0

CO0 -2.2 -4000-

e~e.......... -2.00<Xl~

<,() -1.8....(J).0:J -1.6....

<..9 6000 5000 4000 3000 2000 1000 0cal year BP

Fig. 3.2-4: (a) Mean 0 180 val lies of Tridacna spp. shell carbonate from Eilat (northern Gulf of

Aqaba) plotted versus ages. The Nile level data between about 4500 yr BP and 5000 yr

BP is shown after Henfling andPflallmballm (1991), between abollt 2700 yr BP and 4050

yr BP is after Westendorf and Henfling (1989) based on Hieroglyphical docllments. (b)

The Nile deviation of about the last 1400 years is from Riehl and Meitfn (1979) based on

Nilometer measurements in sOllthern Cairo (Egypt) records. (C) Dead Sea levels inferred

from radiol11etric-dated levels relative to l11ean sea level of the ocean (Neev and Emery

1995). (d) 0180 values in G. ruber during the last 6000 years BP, diamter: 315-400 flm,

20 specil11ens (Sirocko et al 1993).

53

Accarding to the hieroglyphical documents, the Nile floods decreasecLin narthern Egypt from

about 4970 to 4450 years BP. A reduction of the SW summer Monsoon influencing the Nile

discharge has been suggested as an explanation (Henfeling and Pflaumbaum 1991).

Additionally, the reports offamines point to a recession ofthe Nile level at around 4050 years

BP. Between 3950 years BP andabout 3400 years BP, the Nile level rose again, reflecting a

more humid climate (Henfeling and Pflaumbaum 1991). After 3400 years BP the Nile level

decreased till about 2700 years BP (Figure 3.2-4a). As consequence, increased costs of living

were reported in Egypt (Westendorf and Henfling 1989). Altogether, these data (i.e. between

4970 to 4450 cal. years BP) reveal variations similar to the results we have obtained with

Tridacna spp. 01' resemble the Tridacna spp. 8180 data between 4000 cal. years BP and 2700

cal. years BP (Figure 3.2-4a). Similarly, the historical data of the Nilometer measurements in

southern Cairo (Egypt) for approximately the last 1400 cal. years BP coincide with the 8180

variation of the Tridacna spp. Between 1400 and 1000 years BP, the Nilometer shows apart

from the 100 year cycles in the data an upward trend. Between 1000 years BP and about 500

years BP a downward trend is recorded (Riehl and Meitin 1979) (Figure 3.2-4b). Since 500

years BP the Nile level has risen again. The deviations in the Nile level is obtained by

dividing the sum of the flood discharges by the mean flood discharge (Riehl and Meitin

1979).Moreover, the mean Tridacna 8180 data correlate with the level changes in the Dead

Sea (Figure 3.2-4c) (Neev and Emery 1995) and with the changes in 8180 of G. rubel' from

sediment core recovered from the western Arabian Sea (Figure 3.2-4d) (Sirocko et al. 1993)

in the last 6000 years.

Additionally, von Rad et al. (1999) concluded that precipitation decreased off the coast of

southern Pakistan, particularly in the north-eastern Arabian Sea (indicated by thinning varved

sediments) from 5000-3900 years BP, from 2200-1900 years BP, at about 1000 years BP and

from 700-400 years BP. The intervening periods represent precipitation maxima (von Rad

1999) The time interval (5000-3900 years BP) coincides with the reduction of the Nile River

discharge (Henfling and Pflaumbaum 1991) and the beginning of aridification in the Near

East and Middle East that is documented in the archaeological and soil-stratigraphic data from

northern and southern Mesopotamia (e.g. Weiss et al. 1993). The African lakes (Ethiopian and

Chad) show a decline in water levels between 5000 and 4000 years BP (e.g., Gasse 1977;

Gasse and Street, 1978; Gillespie et al., 1983; Gasse and Van Campo, 1994). In summary, the

mean Tridacna 8180 data are corroborated many other regional paleo-data related to the last

6000 years.

However, the mean 8180 of arecent (between 1993-1996) Tridacna sp. (1.20%0) is

comparable to the mean 8180 of a mid-Holocene Tridacna sp. (5460 14C years BP - 5860 cal

years BP) (Figure 3.2-4a). A reason far this similarity may lie in the present intensity of the

summer monsoon rains. Between 1993 and 1996, a time span which is also covered by the

54

recent Tridacna sp. record, the Nasser lake level in southern Egypt increased about 6 m

(Charon et al. 1999). During the 1990s, the inflow into lake Nasser increased steadily. Lake

Nasser lies across the EgyptianJSudanese boundary (500 km in the west of the Red Sea) and

its level fluctuated substantially as a direct response to the summer monsoon rainfall over

Ethiopia. This rain contributes to about 85% of the water that reaches the lake.

CONCLUSIONS

The present study provides a relatively accurate paleoclimate reconstruction during the mid­

and late-Holocene times based on 8 180 values measured in molluscs (Tridacna spp.) from the

northern Gulf of Aqaba (Red Sea). The amplitude of seasonal 8180 (1.17%0) cycles in recent

Tridacna sp. shell carbonates agree with the seasonal temperature variation (5.3°C) in the

nOlihern Gulf of Aqaba (Red Sea). The seasonal 8180 amplitude (temperature) of fossil

Tridacna spp. varied from 1%0 (4.5°C) in the mid-Holocene to 1.27%0 (6°C) in the late­

Holocene. Mean isotope compositions of oxygen in recent and fossil molluscs indicate that

8180 of Tridacna varied from 1.20%0 to 2.29%0 (about 1.1%0 variation in ( 180) during the

late- and mid- Holocene. These mean 8180 variations in Tridacna spp. during the last 6000

years BP also represent a variation of 8180 in sea surface water (i.e. salinity). The variable

intensity of the AfricanJAsian SW monsoon is responsible for precipitation fluctuations over

East Africa and Arabia and hence cause the 8180 variation of sea water. This change probably

causes a SSS variation of maximally 1.5%0 in the northern Red Sea during the last 6000 years

BP.

ACKNOWLEDGEMENTS

All isotopic measurements were carried out at the stable isotope laboratory of the Geoscience

Department at the University of Bremen, Germany. Special thanks go to M. Segl for

performing the isotopic measurements. We also thank H. Arz, T. Felis and H. Kuhnert for

critical discussions. Critical reading of the manuscript by H. Arz greatly improved its quality.

We are grateful to M. Zuther for X-ray diffraction analyses. We are most indebted to E.

Henfling for providing us with his Nile data. This work was supported partly by "Deutscher

Akademischer Austauschdienst" (DAAD) in a scholarship, a grant from the University of

Bremen and a grant from the Graduierten Kolleg "Stoff-Flüsse in marinen Geosystemen"

funded by the Deutsche Forschungsgemeinschaft (DFG) for the first author. This work is also

part of the Red Sea Program on Marine Sciences, funded by the German Ministry for

Education, Science, Research and Technology (BMBF), grant 03F0151A.

55

REFERENCES

AI-Rifaiy IA, CherifOH (1988) The fossil coral reefs of AI-Aqaba, Jordan. Facies 18: 219­

230

Anderson TF, Arthur MA (1983) Stable isotopes of oxygen and carbon and their application

to sedimentologic and paleoenvironmental problems. In: Arthur MA, Anderson TF,

Kaplan IR, Veizer J, Land LS (Eds) Stable isotopes in sedimentm'y geology. Soc Econ

Paleon Mineral Short Course 10: pp 1-151

Bell B (1970) The oldest records ofthe Nile floods. The Geogr J 136: 569-573

Charon B, Mutugudde R, Allan T (1999) Indian Ocean climate event brings floods to East

Africa's lakes and the Sudd Marsh. Geophys Res Lett 26: 1031-1034

Craig H (1966) Isotopic composition and origin of the Red Sea and Salton Sea geothermal

brines. Science 154: 1544-1548

Craig H, Gordon LI (1965) Deuterium and oxygen-18 variations in the ocean and the marine

atmosphere. In: Tongiorrgi E (ed) Stable isotopes in Oceanographic Studies and

Paleotemperatures. CNR Lab Geol Nucl Pisa, pp 9-130

Dabbagh A, Emmermann R, Hötzl H, Jado AR, Lippolt W, Kollmann W, Mosel' H, Rauert W,

Zötl JG (1984) The development of Tihamat Asir during the Quaternary. In: Jado AR,

Zötl JG (eds) Quaternary period in Saudi Arabia. Springer, Berlin Heidelberg New

York, pp 150-174

Doose-Rolinski H, Lückge A (1999) High-resolution hydrographie studies of the laminated

late Holocene and last glacial maximum of the northeastern Arabian Sea. Eos Trans

AGU, 80(17), Fall Meet Suppl, F575

Dullo W-C (1990) Facies fossil record, and age of Pleistocene reefs from the Red Sea (Saudi

Arabia). Facies 22: 1-46

Epstein S, Buchsbaum HA, Lowenstam HA, Urey HC (1953) Revised carbonate-water

isotopic temperature scale. Geol Soc Am Bull 64: 1315-1325

Eshel G, Cane MA, Blumenthai MB (1994) Modes of subsurface, intermediate, and deep

water renewal in the Red Sea. J Geophys Res 99: 15941-15952

56

Fairbanks RC, Charles CD, Wright JD (1992) Origin of global meltwater pulses. In: Taylor

RE, Long A, Kra RS (eds) Radiocarbon After Four Decades. Springer, New York, pp

473-500

Frenzel B, Pachur H -J (1998) Terrestrial palaeoecological research in the framework of

PAGES. In: Ehlers E, Krafft T (eds) German Global Change Research, National

Committee on Global Change Research. pp 60-65

Friedman GM (1965) A fossil shoreline reef in the Gulf of Elat (Aqaba). Isr J Emih Sei 14:

86-90

Friedman GM (1968) Occurrence and stability relationships of aragonite, high-magnesium

calcite under deep-sea conditions. Geol Soc Am Bull 76: 1191-1196

Ganssen G, Kroon D (1991) Evidence for Red Sea surface circulation from oxygen isotopes

of modern surface waters and planktonic foraminiferal tests. Paleoceanogr 6: 73-82

Gasse F (1977) Evolution of lake Abhe (Ethiopia and T.F.A.I.) from 70.000 B.P. Nature 270:

42-45

Gasse F, Street-Perrot FA (1978) Late Quaternary lake level fluctuations and environments of

the northern rift Valley and Afar region (Ethiopia and Djibouti). Pa1eogeogr

Paleoclimatol Paleoecol 24: 279-325

Gasse F, Van Campo E (1994) Abrupt post-glacial climate events in west Asia and North

Africa monsoon domains. Earth Planet Sci Lett 126: 435-456

Gillespie R, Street-Perrot FA, Switzur R (1983) Post-glacial arid episodes in Ethiopia have

implications for climate prediction. Nature 306: 680-683

Grossman EL, Ku TL (1986) Oxygen and carbon isotope fractionation in biogenic aragonite:

Temperature effects. Chel11 Geol 59: 59-74

Gvirtzman G (1994) Fluctuations of sea level during the past 400 ka years: The record of

Sinai, Egypt. Coral Reefs 13: 203-214

Gvirtzman G, Kronfeld J, Buchbinder B (1992) Dated coral reefs of southern Sinai (Red Sea)

and their implication to late Quaternary sea levels. Mar Geol 108: 29-37

57

Hassan FA (1981) Historical Nile floods and their implications for climatic change. Science

212: 1142-1144

Hemleben C, Meichner D, Zahn R, Almogi-Labin A, Erlenkeuser H, Hiller B (1996) Three

hundred eighty thousand year long stable isotope and faunal records from the Red Sea:

Influence of global sea level change on hydrography. Paleoceanogr 11: 147-156

Henfling E, Pflaumbaum H (1991) Neue Aspekte zur klimatischen Interpretation der hohen

pharaonischen Nilflutmarken am 2. Katarakt aus ägyptologischer und geomor­

phologischer Sicht. Würzburg Geogr Arb 80: 87-109

Issar AS (1990) Water shall flow from the rock; hydrology and climate in the lands of the

Bible. Springer, Berlin Heidelberg New York, pp 1-213

Ji J, Petit-Maire N, Yan Z (1993) The last 1000 years: Climatic change in arid Asia and

Africa. Glob and Planet Change 7: 203-210

Jones DS (1985) Growth increment, and geological variations in the molluscan shell. In: T.

W. Broodhed (Ed) Molluscs, University of Tennessee, Knoxville, Tenn Stud Geology

13: 72-87

Klein R, Pätzold J, Wefer G, Loya Y (1992) Seasonal variations 111 the stable isotopic

composition and the skeletal density pattern of the coral Porites lobata (Gulf of Eilat,

Red Sea). Mar. Biol 112: 259-263

Klinker J, Reiss Z, Kropach C, Levanon I, Harpaz H, Halicz E, Assaf G (1976) Observations

on the circulation pattern in the Gulf of Elat (Aqaba), Red Sea. Isr J Emih Sci 25: 85­

103

Mastaller J (1987) Molluscs of the Red Sea. In: Edwards AJ, Head SM (eds) Key

environments Red Sea. Pergamon Press, Oxford, New York, pp 194-214

Mergner H, Schuhmacher H (1974) Morphologie, Ökologie und Zonierung von Korallen­

riffen bei Aqaba, (Golf von Aqaba, Rotes Meer). Helgoländer Wiss Meeresunters 26:

238-358

Milliman JD (1974) Marine carbonates. Springer, Berlin Heidelberg New York, pp 1-375

Morcos SA (1970) Physical and chemical oceanography of the Red Sea. Oceanogr Mal' Biol

Ann Rev 8: 73-202

58

Moustafa YA, Pätzold J, Loya Y, Wefer G (1999) Mid-Holocene stable isotope_ record of

corals from the northern Red Sea. Int J Earth Sci (in press)

Nadeau M-J, Schleicher M, Grootes PM, Erlenkeuser H, Gottdang A, Mous DJW, Samthein

JM, Willkomm H (1997) The Leibniz-Labor AMS facility at the Christian-Albrechts

University, Kiel, Germany. Nuclear Instruments and Methods in Physics Research B

123: 22-30

Neev D, Emery KO (1995) The destruction of Sodom, Gomorrah, and Jericho: Geological,

climatological, and archaeological background. Oxford University Press, New York

Oxford, pp 1-175

Paldor N, Anati DA (1979) Seasonal variations of temperature and salinity in the Gulf of Elat

(Aqaba). Deep-Sea Res 26: 661-672

Pätzold J, Heinrichs JP, Wolschendorf K, Wefer G (1991) Correlation of stable oxygen

isotope temperature record with light attenuation profiles in reef-dwelling Tridacna

shells. Coral Reefs 10: 65-69

Quinn WH (1992) A study of Southern Oscillation-related climatic activity for A. D. 622­

1900 incorporating Nile River flood data. In: EI Nino-Historical and paleoclimatic

aspects of the Southern Oscillation, edited by H. F. Diaz and V. Markgraf, pp. 119-149,

Cambridge Univ. Press, Call1bridge

Reiss Z, Hottinger L (1984) The Gulf of Aqaba: Ecological micropaleontology. Springer,

Berlin Heidelberg New York, pp 1-354

-Riehl H, Meitin J (1979) Discharge of the Nile River: A barometer of short-period climate

variation. Science 206: 1178-1179

Rostek F, Ruhland G, Bassinot FC, Müller PJ, Labeyrie LD, Lancelot Y, Bard E (1993)

Reconstructing sea surface temperature and salinity using 8 180 and alkenone records.

Nature 364: 318-322

Sirocko F, Samthein M, Erlenkeuser H, Lange H, Arnold M, Duplessy JC (1993) Century­

scale events in monsoonal climate over the past 24,000 years. Nature 364: 322-324

59

Slutz RJ, Lubker SJ, Hiscox JD, Woodruff SD, Jenne RL, Joseph DH, Steuer PM, Elms JD

(1985) Comprehensive Ocean-Atmosphere Data Set. Release 1, 268 pp., NOAA

Environmental Research Laboratories, Climate Research Program, Boulder, CO.

Strasser A (1995) Pleistocene reefs in Sinai: eustatic and climatic controls, Proc 2nd Europ

Regional Meeting, ISRS; Publ Serv Geol Lux Vol XXIX, 15-16, Luxembourg.

Strasser A, Strohmenger C (1997) Early diagenesis in Pleistocene coral reefs, southern Sinai,

Egypt: response to tectonics, sea-level and climate. Sedimentology 44: 537-558

Strasser A, Strohmenger C, Davaud E, Bach A (1992) Sequential evolution and diagenesis of

Pleistocene coral reefs (south Sinai, Egypt). Sed Geol 78: 59- 79

Stuiver M, Pol~ch HA (1977) Discussion: Reporting of 14C data. Radiocarbon 19: 355-363

Stuiver M, Reimer PJ (1993) Extended 14C database and revised CALIB radiocarbon

calibration program. Radiocarbon 35: 215-230

Stuiver M, Reimer PJ, Bard E, Beck JW, Burr, GS, Hughen KA, Kromer B, McCormac FG,

v.d. Plicht J, Spurk M (1998) INTCAL98 radiocarbon age calibration, 24,000-0 cal BP.

Radiocarbon 40: 1041-1083

von Rad U, Schaaf M, Michels KH, Schulz H, Berger WH, Sirocko F (1999) A 5000-yr

record of climate change in varved sediments from the oxygen minimum zone off

Pakistan, Northeastern Arabian Sea. Quat Res 51: 39-53

Wefer G (1985) Die Verteilung stabiler Isotope in Kalkschalen mariner Organismen. Geol Jb

A 82: 1-114

Wefer G, Berger WH (1991) Isotope palaeontology: growth and composition of extant

calcareous species. Mal' Geol 100: 207-248

Weiss H, Courty M-A, Wetterstrom W, Guichard F, Senior L, Meadow R, Currnow A (1993)

The genesis and collapse of third millennium north Mesopotamian civilisation. Science

261: 995-1004

Westendorf W, Henfling E (1989) Das Klima im Alten Ägypten nach Hieroglyphen­

aufzeichungen in: Klimaforschungprogramm des Bundesminsterium für Forschung und

Technologie "Statusseminar": 493-494

Woodruff SD, Lubker SJ, Wolt~r K, Worley SJ, Elms JD (1993) Comprehensive Ocean­

Atmosphere Data Set. Release 1a: 1980-1992, Earth System Monitor: 4, NOAA.

60

61

3.3

MID-HOLOCENE STABLE ISOTOPE RECORD OF CORALS FROM THE

NORTHERN RED SEA

(International Journal ofEarth Sciences, 88: 742-751, 2000)

ABSTRACT 62

INTRODUCTION 63

Stable isotopes in corals as paleoc1imatic records

Present climate and circulation pattern

The African monsoon and climate during the mid-Holocene

MATERIAL AND METHODS 64

RESULTS 6714C d .atmg

X-ray diffractometry

Growth rates

Stable Isotope Analyses

Stable oxygen isotopes

Stable carbon isotopes

DISCUSSION 72

CONCLUSIONS 75

ACKNOWLEDGEMENTS 75

REFERENCES 76

62

Mid-Holocene stable isotope record of corals from the northern

Red Sea

Moustafa, Y. A., J. Pätzold, Y. Loya and G. Wefer

Fachbereich Geowissenschaften, University ofBremen, Postfach 330440 28334 Bremen,

Germany

Fax: 0049(421)2183116

email: yaser@uni-bremen.de

ABSTRACT

We present a study based on X-radiography chronologies and the stable isotopic composition

of fossil Porites spp. corals from the northern Gulf of Aqaba (Red Sea) covering the mid­

Holocene period from 5750 to 4450 14C years BP (before present). The stable oxygen and

carbon isotopic compositions of five specimens reveal regular annual periodicities. Compared

to modern Porites spp. from the same environment, the average seasonal 8 180 amplitude of

the fossil corals is higher (by about 0.35 to 0.60 %0) whereas annual growth rates are lower

(by about 3.5 to 2 mm/year). This suggests stronger seasonality of sea surface temperatures

and an increased variability of the oxygen isotopic composition of the sea water due to

changes in the precipitation and evaporation regime during the mid-Holocene. Most likely,

summer monsoon rains reached the nOlihern end of the Red Sea at that time. Average annual

coral growth rates are diminished probably due to an increased input and resuspension of

terrestrial debris to the shallow marine environment during more humid conditions. Our

results corroborate published reports of paleodata and model simulations suggesting a

northward migration of the African monsoon giving rise to increased seasonalities during the

mid-Holocene over northeastern Africa and Arabia.

63

INTRODUCTION

Stable isotopes in corals as paleoclimatic records

Coral stable isotope time series are increasingly used for climate reconstructions. Depending

on the oceanographic and climatic settings, their Ö180 reflect variations in sea surface

temperature (SST) (e.g., Wefer and Berger 1991; Dunbar et al. 1994; Druffel and Griffin

1993; Wellington et al. 1996), or sea surface salinity (SSS) (e.g., Cole and Fairbanks 1990;

Linsley et al. 1994), 01' a combination of both temperature and salinity (e.g., Gagan et al.

1994; Quinn et al. 1996; Klein et al. 1997). Salinity-related effects such as fluctuations in the

patterns of rainfall, evaporation and water mass transport have a large impact on the Ö180 of

coral skeletons (Beck 1998). A multi-proxy approach combining ÖI80 and Sr/Ca ratios has

already been used to deconvolute the effects of SST and SSS, especially for fossil corals

(Beck et al. 1997; Gagan et al. 1998). In contrast, corals ö13C is more complex and difficult to

interpret. A variety of factors such as Ö13 C of seawater, physiological effects, growth rate, and

light intensity influence the Ö13C of coral skeletons (e.g., Swart 1983; Swart et al. 1996,

McConnaughey 1989).

Present climate and circulation pattern

Mediterranean countries are influenced presently by the Westerlies Zone which brings

rainstorms from the northern Atlantic and North Sea through Europe and the Mediterranean

Sea during winter (Goodfriend 1991). This zone moves northward during summer and

southward during winter. However, due to the coastline configuration of the southeastern

edge of the Mediterranean Sea, the deserts of northern Egypt, Sinai, Negev, and southern

Jordan lie outside the main path of the rainstorms coming from the west (Issar 1990). The

mean ammal rainfall in the northern Gulf of Aqaba at Eilat is 22 mm/year (Friedman 1968)

with extremes ranging between 0-70 mm/year (Mergner and Schuhmacher 1974). Rain only

falls in winter months between November and March. The scarcity and randomness of the

rains at Eilat is due to the local topography. The Dead Sea rift valley (Figure 3.3-1a) receives

a minor amount of rain coming from the Mediterranean Sea and is therefore relatively arid,

whereas the adj acent mountains receive more rain and snow in winter (Issar 1990). The

impact of winter rains is high as evaporation is relatively low during this season. On the

contrary, under the arid and hot conditions of the Gulf of Aqaba, evaporation is extremely

high (3650 mm/year) and greatly exceeds precipitation (Reiss and Hottinger 1984). Seasonal

SST off Eilat range between a minimum of about 21°C in winter (February - March) and a

maximum of about 26°C in summer (August-September). The average salinity is about

40.5%0. The cloud cover in Eilat is low throughout the year. The water circulation in the Gulf

of Aqaba is mainly thermohaline, determined by evaporative loss and buoyancy flux.

Throughout the year a considerable volume of warm, relatively low saline and highly

oxygenated waters enter the Gulf of Aqaba from the Red Sea through the Strait of Tiran

64

(Figure 3.3-1 a) and flow northward against the prevailing winds. Cooling and evaporation in

the northern sector of the Gulf leads to a deeper, southward return flow of denseI' waters into

the Red Sea (Klinker et al. 1976; Reiss and Hottinger 1984).

The African monsoon and climate during the mid-Holocene

The Holoeene c1imatie optimum in north Afriea and Saudia Arabia has been identified as a

humid period between 8-5 ka BP (Gasse and Dodo 1997; Glennie et al. 1994). Paleo lake

studies as well as geomorphologie and biostratigraphie data indieate that the Sahara was

considerably more humid than at present. Extensive vegetation has been reconstrueted for the

early to mid-Holoeene. This has eommonly been attributed to a northward shift of the

monsoonal eireulation during the period of maximum summer solar radiation in the Northern

Hemisphere, resulting in enhanced summer precipitation over north Africa (Harrison et al.

1997). According to the results of the Paleoclimate Modelling Intereomparison Projeet

(PMIP) which studied the mid-Holoeene African monsoon ehanges, the amplification of the

temperature seasonal cycle for the northern eontinents is a direct response to the insolation

foreing (Braconnot et al. 1997). Ritchie et al. (1985) suggested a humid tropical climate with

annual monsoonal rainfall of at least 400 mm during mid-Holoeene based on sediment and

pollen evidenee from the eastern Sahara. A progressive increase in aridity with annual

preeipitation declining hom 300 mm at 6000 years BP to less than 100 mm at 4500 years BP

were dedueed. Presently, the alUOlmt of rainfall in the eastern Sahara ranges between 0-5

mm/year.

The COHMAP Members (1988) concluded that the north Afriean-Eurasian landmass was 2 to

4°C warmer during mid-Holocene than at present, whieh enhanced the land-oeean thermal

eontrast and strengthened the monsoonal rainfall over the Sahara, Arabia, and southern and

eastern Asia. Lorenz et al. (1996) modelled summer and winter temperatures for the Holoeene

climate optimum at 6000 years BP, and suggested that summer temperature inereased by 2°C

and winter temperatures were generally lower by abouf2°C partieularly in northern Afriea

and Arabia. Flohn (1991) stated that the end of the mid-Holocene moist period in the Near

East should have been aeeompanied by the end of oeeasional rainfall during summer half­

year, reaehing the Negev from the south.

MATERIAL AND METHODS

Fossil eoral sampies were taken from the top ofan exposed Holoeene reefterrace at about 2 m

above present sea level and about 200 m distance from the present shoreline near the Inter

University Institute (IUI), whieh is loeated about 6 km south of the City of Eilat (Israel) at the

northwestern end ofthe Gulfof Aqaba at 29°31'N and 34°56'E (Figure 3.3-1b,c). Ten fossil

coral colonies were dated by AMS 14C. The measurements were performed at the Leibniz

Laboratory for RadiOl1Uclide Dating and Isotope Research, Christian-Albrechts-University

65

Kiel, Germany, for sampies H2 and E3. All other sampies were measured at the Center for

Isotope Research, University of Groningen, Netherlands. The 14C ages were corrected for

isotopic fractionation with 13C values as measured by AMS. The data were not corrected for

possible reservoir effects. Taking into account changes in total dissolved inorganic carbon

(TC) and estimates of primary production, aging time of water below the thermocline in the

central part ofthe Gulfis calculated to be in the range ofO.5-2 years (Shemesh et al. 1994).

34.,....--------.,--,,-.,....----.--...,----::-:-:--:-:----~---_rr_7""""'r1

32

30

28

26·

22'N+---r-..,..-~-J.;,-__r-.....-Jo._1I-------------------------...J

24 26 28 30 32 34 J(, 38'E

Fig. 3.3-1: a) Location map ofthe northern end ofthe Gulfof Aqaba, b) study area southwest of

Eilat, c) geomorphological features adjacent to the modern and fossil reefs next to the

InterU niversity Institute Cl UI).

The coral colonies were cut into slabs about 5 mm thick parallel to the dominant axis of

growth. The slabs were X-rayed using a cabinet X-ray system (Faxitron 43855A) in order to

visualize the density growth patterns. The coral slabs were exposured at 45 kV, 3 mA, for

about 10min. The X-radiographs revealed regular and weIl developed annual density patterns

of alternating bands of high and low density (for method see e.g., Knutson et al. 1972;

Hudson et al. 1976).

A set of five coral colonies were selected for further analysis. The mineralogy of these coral

sampies were determined by X-ray diffraction analysis on a Philips PW 1800 (Philips,

Eindhoven, The Netherlands) X-ray diffractometer (Cu, 45 kV, 35 mA) at an angle between

20° and 50° (28) with 1/4° 28 per min (2h) at the Mineralogical Section of the Geoscience

Department of the University of Bremen. Sampies for X-ray diffraction analysis were taken

from the same region where isotope sampies were drilled later.

66

For stable oxygen and carbon isotopic analyses the coral colonies (F8, F3, H2, H5 and F24)

were sampled at high resolution along the growth direction. The sampies were taken by

grinding a channel into the slabs at regular intervals between 0.3-0.6 mm depending on band

width. A dental drill with a 0.6 mm diameter rounded (flower-shaped) bit was used. The

drilling depth was about 2 mm. An example of such an isotopic profile is indicated in the X­

radiograph of sampie no. H2 (4600 ±50 years BP) (Figure 3.3-2).

Fig. 3.3-2: Exal11ple ofaX-radiography of a fossil coral (H2, 4600 ±50 14C years BP) colony.

The alternating growth bands of high (dark bands) and low density (I ight bands) can be

seen. The sal11pling profile far stable isotope l11easurel11ents is indicated by a white line.

For stable oxygen and carbon isotopic analyses powdered carbonate sampies were reacted

with 100% orthophosphoric acid at 75°C to produce carbon dioxide. The isotope

measu~ements were performed using an automated carbonate preparation device attached to a

Finnigan MAT 251 (Finnigan, Bremen, Germany) mass spectrometer. Results are given in the

conventional Ö notation relative to the PDB (Belemnite from the Pee Dee Formation of South

Carolina) isotopic standard, calibrated by means ofthe NBS 19 standard:

Ö180 (%0) = {[(180 /160 )sal11ple - C80/160)standard] I (180 /160 )standard} X 1000

Öl3 C (%0) = {[( l3C/12C)sal11ple - (l3CI 12C)standard] I (l3C/ 12C)standard} X 1000

The precision based on replicate measurements of an internal laboratory standard (Solnhofen

limestone of 63 to 80 flm) was ±0.07 %0 for Ö180 and ±O.OS %0 far Öl3 C. All stable isotope

analyses were carried out at the Isotope Laboratory of the Geoscience Department of the

University of Bremen, Germany.

67

RESULTS

14C dating

The AMS 14C ages often fossil corals (Porites spp.) ranged between 4450 and 5750 years BP

(Table 3.3-1). This reefterrace shows the same age as many other elevated terraces along the

northern Red Sea (e.g., Dabbagh et al. 1984; Al-Rifaiy and Cherif 1988; Dullo 1990;

Gvirtzman et al. 1992; Gvirtzman 1994). Friedman (1965) described sampies from a fossil

reef southwest of Eilat having about the same age (4770 ± 140 years BP).

Table 3.3-1: AMS 14C ages of fossil coral colonies frol11 Eilat, nOlihern end of the Gulf of

Aqaba. The datings were performed at the Leibniz-Laboratory (KIA) in Kiel, Germany,

and the Center for Isotope Research (GrA) Groningen, Netherlands.

Sampie Lab No. I'+C_ Age

Identification (years BP)

1 H2 KIA 1881 4600 ± 50

2 F3 KIA 1882 4890 ± 40

3 H5 GrA 7840 4600 ± 60

4 F6 GrA 7824 4960 ± 60

5 F7 GrA 7825 5140 ± 60

6 F8 GrA 7827 5750 ± 60

7 F9 GrA 7829 5100 ± 60

8 Fll GrA 7830 5370 ± 60

9 F23 GrA 7832 4920 ± 90

10 F24 GrA 7833 4450 ± 60

X-ray diffractometry

X-ray diffractometry shows that most samples are still aragonitic in composition, except for

sampie F24 (4450 ±60 years BP) which contains traces of calcite. We conclude, that the

sampies were not subject to diagenetic alterations, which might have increased 14C content. In

addition, the aragonitic minerology is taken as evidence that the stable isotopic composition

has not been altered and can thus be used for paleoclimatic reconstructions.

68

5750 yr SP (F8)-4

0-3 - WNWVVvv~

GO -2 -

-1 -

U o -~~~GO 1-

21 2 3 4 5 6 7 8 9

4600 yr SP (H2)-4

0 -3

~V)

-2

-1

U 0

~V)

2

4890 yr SP (F3)

123 4 5 6 7 8 9 10 11 12 13 14 15 16 1718

4600 yr SP (H5)

2 3 456 7 8 9 10 11 12 13 14 15 16 1 2 3 4 5 6 7 8 9 10 11

4450 yr SP (F24)-4

0 -3

~GO -2

-1

U0

~GO

21 2 3 4 5 8 9 10 11 12 13 14 15 16

Modern coral (Eilat-1) Modern coral (84)-4

0 -3

~CA)

V) -2

-1

0U~V)

21 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 1 2 3 4 5

Number of annual cycles

Fig. 3.3-3: The stable oxygen (L1pper line) and carbon (lower line) isotopic compositions and

time series of five fossil corals from the mid-Holocene years BP and two modern corals

from the northern Red Sea (S4 from Klein et al. 1992; Eilat-1 from Felis et al. 1998b).

The growth direction is from left to right.

69

Growth rates

The X-radiographs of our coral colonies reveal continuous growth records of up to 18 years.

The al1lmal growth rates were determined from the seasonal cycles of the stable oxygen

isotopes. Despite low al1lmal growth rates, our drilling technique allowed a nearly monthly

sampling resolution (Figure 3.3-3). The mid-Holocene corals show lower mean annual growth

rates (between 3.4 ± 0.7 and 5.7 ± 1.4 mm/year) than corals from the modern reef environment

at Eilat (between 7.1 ± 1.9 and 7.9 ± 1.6 mm/year), except for colony F8 (7.2 ± 1.5 mm/year

at 5750 ±60 years BP) which has annual growth rates comparable to the modern colonies

(Figure 3.3-4a). At the northeastern coast ofthe Gulf of Aqaba, adjacent to the City of Aqaba,

modern corals reveal even higher growth rates. Corals from water depths between 2-7 m grow

at rates between 8.7 (±1.4) and 14.4 (±1.2) mm/yr (Heiss 1996). Local effects in the

circulation pattern at the northern end of the Gulf seem to be responsible for these differences

in growth rate. Whereas the reefs off Aqaba are locally affected by upwelling of water

masses, downwelling occurs in front of the reefs at Eilat (Mergener and Schuhmacher 1974;

Genin et al. 1995). Increased plankton abundance in the upwelling area off Aqaba seems to

favor the coral reef growth.

Stable Isotope Anaiyses

Stable oxygen isotopes

Five Parites sp. (F8, F3, H2, H5 and F24) colonies were studied for stable carbon and oxygen

isotope composition. Thc lengths of the isotope records records vary between 9 and 18 years.

The isotopic records of two modern corals are also for reference (Klein et al. 1992; Felis et al.

1998b) (Figure 3.3-3). All mid-Holocene colonies show clear seasonal variations. The stable

oxygen isotope data are summarized in Table 3.3-2. The values of the fossil corals are heavier

(between -2.13 and -2.76 %0 on average) compared to the modern corals (-2.77 and -3.04%0

on average) and the mean seasonal 8 180 amplitudes (difference between minima and maxima)

of the mid-Holocene corals (between 1.1%0 and 1.35%0) is greater than in modern corals

(between 0.71 and 0.75%0) (Figure 3.3-4b and 5). Sampies drilled in high density bands were

found to be emiched in 180 compared to low-density bands. We conclude that the high­

density bands were deposited during winter, wbereas tbe low-density bands were deposited

during summer. This density pattern is similar to tbat of modern Parites lobata (3 m water

depth) from Eilat (Klein et al. 1992; Klein et al. 1993) and modern Porites sp. (4.5 m water

depth) from Aqaba (Heiss 1994).

12

11

10

9

8'2.c 7E.S-'" 6i§

"" 5;Ee

<.9 4

3

-~

~ .........

'H", -- T -~

---'-

--...... -

:....... ~.

T-- r---

'-,- ...........

$ s~I- c.....,-....

-r

----'---:

70

a)o

5750 yr SP 4890 yr SP 4600 yr SP 4600 yr SP 4450 yr SP Modern Modern

T ,-L--,

I,.--L

T -'- .....

~ f---

f·····I-- ...... •......

'= -~ ~

"'T I'-~

~T =

~-'- ~

H g-~

.-L

2.0

1.8

1.6

1A

l'" 1.2-0

;~Ci. ~O 1.0E(1J -00

roc 0.80<f>(1J

'"(j) 0.6

OA

0.2

0.0

b)5750 yr SP 4890 yr SP 4600 yr SP 4600 yr SP _ 4450 yr SP Modern Modern

Fig. 3.3-4: a) Box plot of growth rate of fossil and modern coral sampies showing the mean,

mean ± standard deviation (lcr), minimum and maximum growth rate of each sampie.

The mean growth rate of the fossil sam pies are lower than that of the modern corals.

Note, that the growth rates of the corals from 4600 yr SP have the same mean and ranges.

b) Box plot of 8 180 seasonal amplitude in fossil and modern coral sampies showing the

mean, mean ± standard deviation (lcr), minimum and maximum seasonal amplitude for

each sampie. The mean seasonal amplitude of fossil coralsare clearly higher than for

modern corals.

71

Table 3.3-2: Summary of statistics for stable oxygen isotope analyses of mid-Holocene and

modern corals from Eilat, nOlihern Gulf of Aqaba

* Data from Klein et al. (1992)

** Data from Felis et al. (1 998b)

oj~o (%oPDB) Maximum Minimum Range Mean St. Dev. Sampies

F8 (5750 year BP) -1.39 -3.27 1.89 -2.41 0.44 127

F3 (4890 year BP) -1.75 -3.57 1.82 -2.67 0.42 192

H2 (4600 year BP) -1.27 -3.09 1.82 -2.24 0.45 150

H5 (4600 year BP) -1.16 -2.90 1.74 -2.13 0.48 75

F24 (4450 year BP) -1.45 -2.94 1.49 -2.21 0.39 190

S4* (Modern) -2.19 -3.22 1.03 -2.77 0.29 35

Eilat-1 ** (Modern) -2.37 -3.59 1.22 -3.04 0.28 121

Stable carbon isotopes

The earbon isotope eomposition shows distinet annual periodieity (Figure 3.3-3) although it is

less well developed in the eolonies F8 and H2. The seasonal amplitudes of oJ3e (Table 3) are

mueh more variable than in oxygen isotope eycles. Generally, the mid-Holoeene eolonies

show heavier values in oJ3e (between -0.68 and 0.46 %0 on average) than modern eorals

(between -1.78 and -1.42 %0 on average). The phase relationship between earbon and oxygen

isotopes is very similar in fossil and modern reeords. Figure 3 shows that the most positive

öJ3e values generally lag the most positive 0 180 values by about 2-3 months and that

minimum öJ3e oeeur during autumn or winter.

Table 3.3-3: Summary of statistics for stable carbon isotope analysis of mid-Holocene and

modern corals from Eilat, nOlihern Gulf of Aqaba

* Data from Klein ci al. (1992)

** Data from Felis et al. (1 998b)

OLle (%oPDB) Maximum Minimum Range Mean St. Dev. Sampies

F8 (5750 year BP) 1.20 -0.75 1.95 0.22 0.29 127

F3 (4890 year BP) 0.86 -1.99 2.85 -0.48 0.56 192

H2 (4600 year BP) 0.18 -1.39 1.57 -0.68 0.31 150

H5 (4600 year BP) 1.73 -2.29 4.03 0.08 1.01 75

F24 (4450 year BP) 1.64 -0.75 2.39 0.46 0.62 190

S4* (Modern) -0.44 -2.40 1.96 -1.42 0.37 35

Eilat-l ** (Modern) 0.44 -2.90 2.46 -1.78 0.52 121

72

DISCUSSION

The 8180 of marine organisms vanes as a function of the 180 /160 ratio of seawater and

temperature (Epstein et al. 1953; Wefer and Berger 1991). The oxygen isotopic ratio of ocean

surface water reveals spatial and temporal variability which is linked to changes in

evaporation, precipitation and to atmospheric and oceanic water mass transport (e.g., Rohling

and Bigg 1998). Thus it is closely connected to changes in SSS. The oxygen isotope ratios of

coral aragonite skeletons are not secreted in equilibrium with the surrounding seawater, but

exhibit biological depletion of 180 during calcification. This "vital effect" seems to be species

dependent and is more or less constant. Thus, whereas absolute temperature reconstructions

are uncertain, relative seasonal temperature variations can be resolved with high resolution

providing there is no change in sea-surface salinity (SSS) (e.g., Weber and Woodhead 1972;

Fairbanks and Dodge 1979; Pätzold 1984).

-45750 yr SP 4890 yr SP 4600 yr SP 4600 yr SP 4450 yr SP Modern Modern

-3 .~m00...

0-2

~0~(,Q

-1

o9 1 18 16 1 11 1 16 1 16 1

Number of annual cycies

Fig. 3.3-5: Stable oxygen isotope time series from fossil (4450 - 5750 14C yr BP)and modern

corals. The fossil corals show a considerably higher seasonal 8180 amplitude (about 1.7

times) as weil as heaiver mean 8180 (about 0.5%0) compared to modern corals from the

same location (Klein et al. 1992; Felis et al. 1998b).

Modern oxygen isotope cycles of corals from Eilat do not resolve the temperature variations

between about 21 and 26°C (i.e. 0.9%0 in ( 180) due to a salinity increase during summer

which dampens the isotope signal (Felis et al. 1998a). The modern seasonal salinity amplitude

is 0.5%0 (Paldor and Anati 1979; Wolf-Vecht et a1. 1992). This seasonal change in salinity

conesponds to a change in the isotope composition of seawater, and finally results in a

reduction ofthe coral 8180 amplitude by about 12% in modern corals in the northern Red Sea

(Felis et al. 1998a).

73

The fossil coral records clearly document that the seasonal cycles of oxygen isotopes in

annual growth bands in the northern Red Sea were amplified during the mid-Holocene.

Despite a reduction in growth rates, seasonal amplitudes of (5180 cycles are enhanced in mid­

Holocene corals. Due to the detailed sampling procedure an almost monthly resolution was

achieved for both fossil and modern corals. It is anticipated that this sampling procedure fully

resolved the recorded seasonal isotope cycles.

Model simulations of the earth' s orbital parameters demonstrate that the seasonal cycle of

solar radiation was enhanced during the early and mid-Holocene in the NOlihern Hemisphere

(Kutzbach and Street-Perrott 1985). Solar radiation was increased during summer and

decreased during winter by about 5 and 2 % at 6000 and 3000 years BP, respectively, giving

rise to cooler winters and warmer summers. As a consequence the thermal contrast between

Northern Hemisphere continents and the ocean increased and amplified the monsoonal

circulation. A major intensification of the summer monsoon combined with increased

southwesterly winds increased the transport of moisture from the oceans onto the northern

land masses. More recent simulations with a climate model that asynchronously couples the

atmosphere and the ocean show that summer monsoon precipitation increased as far north as

23°N and up to 300 N in northern Africa (Kutzbach and Liu 1997). The suggested lowering of

sea surface temperatures in winter and increased heating during summer is consistent with our

findings at 29°N. In addition, increased precipitation during the summer monsoon season will

have lowered the coral oxygen isotope signal in the warm season. The difference between the

average seasonal amplitudes of fossil and modern corals ranges from 0.35 to 0.6%0 which

could imply an increase of seasonal SST amplitude of about 2-3.5°C, if entirely related to

temperature. The gradient of 0.18%0/OC (Gagan et al. 1994) is widely accepted for

temperature interpretation of Porites (5180 records (Charles et al. 1997; Felis et al. 1998a). If

the change is interpreted as a pure temperature signal it would imply lowest SST down to 15

or 16°C, which is umealistic since coral growth ceases at temperatures below about 18°C. On

the other hand, if this difference were entirely related to salinity, it would be equivalent to a 1­

2%0 change in salinity (0.29%0 0 180/%0 salinity; Craig 1966). The enhanced seasonal (5180

signals probably reflect a combined effect of temperature and salinity. Their relative

contribution, however, remains to be resolved.

A potential tool to separate temperature and salinity effects and to determine absolute sea

sm'face temperatures is the use of an additional coral proxy thermometer. Sr/Ca ratios in

corals also vary as a function of temperature. The Sr/Ca ratio of ocean water is considered

constant over larger time scales although some work suggests that ocean water also reveals

variability in Sr/Ca ratio. Sr/Ca ratios of corals mayaiso exhibit some growth rate dependence

(deVilliers et al. 1995). Calibrations of the Sr/Ca thermometer reveal consistent resuIts in

many areas despite some discrepancies (deVilliers et al. 1995; Shen et al. 1996). Recent

74

applications of a multi-proxy approach using 8180 and Sr/Ca in coral skeletons provided~

convincing results considering paleotemperatures and variations of evaporation, precipitation,

and ocean surface salinity (Gagan et al. 1998; Beck 1998). Further trace elements analysis of

fossil coral records from the early and late Holocene from the northern Red Sea will help to

elucidate climatic changes in the Near East region.

In addition, the 8 180 signal of the different corals during mid-Holocene (Figure 3.3-5) could

be controlled through variations in the intensity of the SW monsoon. Many climate proxy data

indicate that major changes occur at about 5000 years BP. After the hieroglyphical

documents, the Nile floods fall as a result of the reduction of the SW summer monsoon from

3018 to 2500 years BC (about 4970 to 4450 years BP) in southern Egypt (Westendorf and

Henf1ing 1989; Henfling and Pflaumbaum 1991). As a result of this reduction in the intensity

of the SW monsoon, the mean 8 180 signal in the fossil corals have been reduced by about of

0.5%0 between 4890 and 4450 14C years BP (Figure 3.3-5 and Table 3.3-2). The African lakes

(Ethiopian and Chad) show a decreasing in water levels between 5000 and 4000 years BP

(e.g., Gasse 1977; Gasse and Street-Perrot 1978; Gillespie et al. 1983; Gasse and Van Campo

1994). Also, the Dead Sea level according to Mount Sedom caves exhibit a retread level in the

same time window (Neev and Emery 1995; Frumkin 1997). During the third millennium BC,

the archaeological and soil-stratigraphic data indicate a collapse of rain-fed agriculture

civilisation ofnorthern and southern Mesopotamia (i.e. Subir and Akkadian empire) (Weiss et

al. 1993). More recently, a study based on simulation of sahm"an vegetation in the mid­

Holocene shows an abrupt decrease in the fraction of saharan vegetation cover between 6000

and 4000 years BP which was simulated by using an atmosphere-vegetation model (Claussen

et al. 1999). All these studies may explain the change of corals 8 180 between 4890 and 4600

years BP. Despite the average change, the seasonal 8 180 of fossil corals stays amplified

afterwards. This indicates that the climate between 4600 and 4450 years BP was still wetter

than at present.

Changes in the absolute isotope values could also reflect changes in coral growth rate.

Variations in growth and calcification rate have an impact on the fractionation of stable

isotopes (Land et al. 1975; Pätzold 1986; McConnaughey 1989). Areduction in the growth

rate entails heavier isotope values of oxygen and carbon. Since ammal growth rates of the

mid-Holocene corals were reduced by up to 45% compared to the average modern values

(Figure 3.3-4a) heavier isotope values are expected. Indeed, both oxygen and carbon isotope

signals reveal heavier values in the fossil records. The reduction of coral growth during the

mid-Holocene is probably triggered by increased input and resuspension of terrestrial

sediments. The constant energy expenditure for removal of sediment particles and reduction

of light decreases growth rate (Dodge et al. 1974). Outcrops of the Holocene reef formation at

75

Eilat are characterized by increased interlayering of gravel deposits and intensification of

beach rock formation. Both are taken as signs for more humid conditions.

The difference in the 8 l3C fractionation between mid-Holocene and modern corals (Table 3.3­

3) could be attributed to kinetic fractionation effects. McConnaughey (1989) suggested that a

kinetic effect due to slow skeletal growth rate and hence reduced metabolism results in

heavier 8 13C. In the other hand, Klein et al. (1992, 1993) found the same phase relation

between 8 l3 C and 8180 values and explained it with the seasonal time lag between maximum

light and maximum temperature. Solar irradiance in northern Red Sea reaches its minimum

between December and January, whereas minimum seawater temperatures are recorded

between February and March. Pätzold (1984, 1986) described a similar shift between the

seasonal variation of carbon and oxygen signals of modern and fossil mid-Holocene Porites

lobata from Cebu in the Philippines.

CONCLUSIONS

The stable isotope composition of five mid-Holocene coral colonies from the northern Gulf of

Aqaba were compared to modern corals from the same environment. The results indicate that

the seasonal 8 180 amplitude was greater than in modern corals. This is most probably due to a

larger seasonal temperature contrast and a reduction of salinity during the summer season for

the mid-Holocene. However, 8180 alone cannot resolve the relative contribution of SST and

SSS, and additional tracers are needed. Our results support the hypotheses of summer

monsoon rains reaching the northern Red Sea during mid-Holocene times while seasonal solar

radiation was enhanced. A climatic change seems to have occurred between about 4900 and

4600 years BP. This date coincides with aperiod of rapid fall in north African lake levels

indicating reduction of moisture transport from the ocean. Enhanced seasonalities can be

reconstructed at least until4450 14C years BP.

ACKNOWLEDGEMENTS

We thank T. Felis and M. Fine for the collection of the coral sampies. Special thanks are due

to M. Segl for perfonl1ing the stable isotopic measurements and preparation of the sampie for

AMS 14C dating. AMS 14C measurements were carried out at the Leibniz Laboratory for

Radi011Uclide Dating and Isotope Research, Christian-Albrechts-University Kiel, Germany,

and the Center for Isotope Research, University of Groningen, Netherlands. All stable

isotopic measurements were carried out at the isotope laboratory at Bremen University,

Germany. We thank also S. Draschba, T. Felis and H. Kuhnert for critical discussions, their

help during preparation of the X-radiographs and guidance to take the isotope sampies. 1.

Bijma improved the English of the manuscript. The manuscript was substantially improved by

76

the review comments of T. Correge and C. Dullo. This work was partly supported by the

"Red Sea Program" (RSP) of the Bundesministerium für Bildung, Wissenschaft, Forschung

und Technologie (BMBF, 03FO 151 A6), a grant from the University of Bremen, and a grant of

the Graduierten Kolleg "Stoff-Flüsse in marinen Geosystemen" funded by the Deutsche

Forschungsgemeinschaft (DFG).

REFERENCES

Al-Rifaiy IA, Cherif OH (1988) The fossil coral reefs of Al-Aqaba, Jordan. Facies 18: 219­

230

Beck JW, Recy J, Tay10r F, Edwards RL, Cabioch G (1997) Abrupt changes in early

Holocene tropica1 sea surface temperature derived from coral records. Nature 385: 705­

707

Beck W (1998) Warmer and wetter 6000 years ago? Science 279: 1003-1004

Braconnot P, Joussaume S, DeNoblet N, Ramstein G (1997) Mid-Holocene and last glacia1

maximum African monsoon changes as simulated within PMIP. Paleomonsoons Project

Workshop at Siwa Oasis, Egypt, 11-22 Jan (Abstracts)

COHMAP Members (1988) Climatic changes of the last 18,000 years: Observations and

model simulations. Science 241: 1043-1052

Charles CD, Hunter DE, Fairbanks RG (1997) Interaction between the ENSO and the Asian

monsoon in a coral record oftropical climate. Science 260: 1790-1793

Claussen M, Kubatzki C, Brovkin V, Hoelzmann P, Pachur H-J (1999) Simulation of an

abrupt change in Sahm"an vegetation in the mid-Holocene. Geophys Res Lett 26: 2037­

2040

Cole JE, Fairbanks RG (1990) The Southern Oscillation recorded in the 8 180 of cora1s from

Tarawa Atoll. Paleoceanogr 5: 669-683

Craig H (1966) Isotopic composition and origin of the Red Sea and Salton Sea geothermal

brines. Science 154: 1544-1548

77

Dabbagh A, Emmermann R, Hötzl H, Jado AR, Lippolt W, Kollmann W, Moser H, Raue±:t

W, Zötl JG (1984) The development of Tihamat Asir during the Quaternary. In: Jado

AR, Zötl JG (eds): Quaternary period in Saudi Arabia. Springer, Berlin Heidelberg New

York, pp 150-174

deVilliers S, Nelson BK, Chivas AR (1995) Biological controls on coral Sr/Ca and 8180

reconstructions ofsea surface temperatures. Science 269: 1247-1249

Dodge RE, Aller RA, Thompson J (1974) Coral growth related to resuspension of bottom

sediments. Nature 247: 574-577

Druffel ERM, Griffin S (1993) Large variations of surface ocean radiocarbon: Evidence of

circulation changes in the southern Pacific. J Geophys Res 98: 20,249-20,259

Dullo W-C (1990) Facies fossil record, and age of Pleistocene reefs from the Red Sea (Saudi

Arabia). Facies 22: 1-46

Dunbar RB, Wellington GM, Colgan MW, Glynn PW (1994) Eastern Pacific sea surface

temperature since 1600 AD: The 8180 record of climate variability in Galapagos corals.

Paleoceanogr 9: 291-315

Epstein S, Buchsbaum R, Lowenstam HA, Urey HC (1953) Revised carbonate-water isotopic

temperature scale. BuH Geol Soc Am 64: 1315-1325

Fairbanks RG, Dodge RE (1979) Annual periodicity of the 180 /160 and 13C/12C ratios in the

eoral Montastrea annularis. Geoehim Cosmochim Acta 43: 1009-1020

Felis T, Pätzold J, Wefer G, Fine M, Loya Y, Nawar, AH (1998a) First results of a eoral­

based history of recent climate in the northern Red Sea. Zbl Geol Paläont Teil 1 H1/2:

197-207

Felis T, Pätzold J, Wefer G, Loya Y (1998b) Vertical mixing and plankton blooms recorded

in coral skeletal carbon isotopes. J Geophys Res 103, C13: 30,731-30,740

Flohn H (1991) Towards a physical interpretation of the end of the Holocene moist period in

the Near East. Erdkunde 45: 163-167

Friedman GM (1965) A fossil shoreline reef in the Gulf of Elat (Aqaba). Isr J Earth Sei 14:

86-90

78

Friedman GM (1968) Occurrence and stability relationships of aragonite, high-magnesium

calcite under deep-sea conditions. Geol Soc Am Bull 76: 1191-1196

Frumkin A (1997) The Holocene histroy of Dead Sea levels. In: Niemi TM, Ben-Avraham Z,

Gat JR (eds): The Dead Sea: The lake and its setting. Oxford University Press, New

York Oxford, pp 237-248

Gagan MK, Chivas AR, Isdale PJ (1994) High-resolution isotopic records from corals using

ocean temperature and mass spawning chronometers. Earth Planet Sci Lett 121: 549­

558

Gagan MK, Ayliffe LK, Hop1ey D, Cali JA, Mortimer GE, Chappell J, McCulloch MT, Head

MJ (1998) Temperature and surface-ocean water balance of the mid-Holocene tropical

western Pacific. Science 279: 1014-1018

Gasse F (1977) Evolution of lake Abhe (Ethiopia and T.F.A.I.) from 70.000 B.P. Nature 270:

42-45

Gasse F, Street-Perrot FA (1978) Late Quaternary lake level fluctuations and environments of

the northern rift Valley and Afar region (Ethiopia and Djibouti). Paleogeogr

Paleoclimatol Paleoecol 24: 279-325

Gasse F, Van Campo E (1994: Abrupt post-glacial climate events in west Asia and North

Africa monsoon domains. Earth Planet Sci Lett 126: 435-456

Gasse F, Dodo A (1997) Combined surface and groundwater palaeohydrology in the Sahara

and the Sahel for understanding past climate change. Paleomonsoons Project Workshop

at Siwa Oasis Egypt 11-22 Jan (Abstracts)

Genin A, Lazar B, Brenner S (1995) Vertical mixing and coral death in the Red Sea following

the eruption of Mount Pinatubo. Nature 377: 507-510.

Gillespie R, Street-Perrot FA, Switzur R (1983) Post-glacial arid episodes in Ethiopia have

implications for climate prediction. Nature 306: 680-683

Goodfriend GA (1991) Holocene trends in 180 in land snail shells from the Negev Desert and

their implications for changes in rainfall source areas. Quat Res 35: 417-426

79

Glennie KW, Pugh JM, Goodall TM (1994~ Late Quaternary Arabian Desert models of

Permian Rotliegend reservoirs. Expl Bull 274/3: 1-19

Gvirtzman G (1994) Fluctuations of the sea level during the past 400 000 years: the record of

Sinai, Egypt (northern Red Sea). Coral Reefs 1: 203-214

Gvirtzman G, Kronfeld J, Buchbinder B (1992) Dated coral reefs of southern Sinai (Red Sea)

and their implication to late Quaternary sea levels. Mal' Geoll08: 29-37

HalTison SP, Kutzbach JE, Coe M, Foley J, Hoelzmann P, Jolly D, Liu Z, Prentice IC (1997)

Land-ocean-atmosphere interactions and the amplification of the African monsoon in

the mid-Holocene. Paleomonsoons Project Workshop at Siwa Oasis Egypt 11-22 Jan

(Abstracts)

Heiss GA (1994) Coral reefs in the Red Sea: Growth, production and stable isotopes.

GEOMAR Report 32, Ph.D. thesis, Kiel University

Heiss GA (1996) Annual band width variation in Porites sp. from Aqaba, Gulf of Aqaba, Red

Sea. Bull Mal' Sci 59: 393-403

Henfling E, Pflaumbaum H (1991) Neue Aspekte zur klimatischen Interpretation der hohen

pharaonischen Nilflutmarken am 2. Katarakt aus ägyptologischer und geomor­

phologischer Sicht. Würzburg Geogr Arb 80: 87-109

Hudson JH, Shinn EA, Halley RB, Lidz B (1976) Sclerochronology: a tool for interpreting

past environments. Geology 4: 361-364

Issar AS (1990) Water shall flow from the rock; hydralogy and climate in the lands of the

Bible. Springer, Berlin Heidelberg New York, pp 1-213

Klein R, Pätzold J, Wefer G, Loya Y (1992) Seasonal variations m the stable isotopic

composition and the skeletal density pattern of the coral Porites lobata (Gulf of Eilat,

Red Sea). Mal' Biol 112: 259-263

Klein R, Pätzold J, Wefer G, Loya Y (1993) Depth-related timing of density band formation

in Porites spp. corals from the Red Sea inferred from X-ray chronology and stable

isotope composition. Mal' Ecol Prag SeI' 97: 99-104

80

KleinR, Tudhope AW, Chilcott CP, Pätzold J, Abdulkarim Z, Fine M, Fallick AE, L<2Ya Y

(1997) Evaluating southern Red Sea corals as a proxy record from the Asian monsoon.

Earth Planet Sci Lett 148: 381-394

Klinker J, Reiss Z, Kropach C, Levanon I, Harpaz H, Halicz E, Assaf G (1976) Observations

on the circulation pattern in the Gulf of Elat (Aqaba), Red Sea. Isr J Earth Sci 25: 85­

103

Knutson DW, Buddemeier RW, Smith SV (1972) Coral chronometers: seasonal growth bands

in reef corals. Science 177: 270-272

Kutzbach JE, Street-Perrott FA (1985) Milankovitch forcing of fluctuations in the level of

tropicallakes from 18 to 0 kyr BP. Nature 317: 130-134

Kutzbach JE, Liu Z (1997) Response of the African monsoon to orbital forcing and ocean

feedbacks in the middle Holocene. Science 278: 440-443

Land LS, Lang JC, Barnes DJ (1975) Extension rate: a primary control on the isotopic

composition of West Indian (Jamaican) scleractinian reef coral skeletons. Mal' Biol 33:

221-233

Linsley BK, Dunbar RB, Wellington GM, Mucciarone DA (1994) A coral-based

reconstruction of Intertropical Convergence Zone variability over Central America since

1707. J Geophys Res 99: 9977-9994

Lorenz S, Grieger B, Helbig PH, Herterich K (1996) Investigating the sensitivity of the

atmospheric general circulation model ECHAM 3 to paleoclimatic boundary conditions.

- Geol Rundsch 85: 513-524

McConnaughey T (1989) I3 C and 180 isotopic disequilibrium in biological carbonates: 1.

Patterns. Geochini Cosmochim Acta 53: 151-162

Mergner H, Schuhmacher H (1974) Morphologie, Ökologie und Zonierung von

Korallenriffen bei Aqaba, (Golf von Aqaba, Rotes Meer). Helgoländer Wiss

Meeresunters 26: 238-358

Neev D, Emery KO (1995) The destruction of Sodom, Gomorrah, and Jericho: Geological,

climatological, and archaeological background. Oxford University Press, New York

Oxford, pp 1-175

81

Paldor N, Anati DA (1979) Seasonal~ariations of temperature and salinity in the Gulf of Elat

(Aqaba). Deep-Sea Res 26/6A: 661-672

Pätzold J (1984) Grawth rhythms recorded in stable isotopes and density bands in the reef

coral Porites lobata (Cebu, Philippines). Coral Reefs 3: 87-90

Pätzold J (1986) Temperature and CO2 changes in tropical surface waters of the Philippines

during the past 120 years: record in the stable isotopes of hermatypic corals. Berichte­

Reports, Geol-Paläont Inst Universität Kiel (in German) 12: pp 1-92

Quinn TM, Crowley TJ, Taylor FW (1996) New stable isotope results fram a 173-year coral

record from Espititu, Vanuatu. Geophys Res Lett 23: 3413-3416

Reiss Z, Hottinger L (1984) The Gulf of Aqaba: Ecological micropaleontology. Springer,

Berlin, pp 354

Ritchie JC, Eyles CH, Haynes CV (1985) Sediment and pollen evidence for an early to mid­

Holocene humid period in the eastern Sahara. Nature 314: 352-355

Rohling EJ, Bigg GR (1998) Paleosalinity and 8 180: A critical assessment. J Geophys Res

103: 1307-1318

Shemesh A, Luz B, Erez, J (1994): Carbon isotopes, dissolved oxygen, and the carbonate

system in the northern Gulfof Aqaba (Elat). Isr J Earth Sci 43: 145-155

Shen C-C, Lee T, Chen C-Y, Wang C-H, Dai C-F, Li L-A (1996) The calibration of D[Sr/Ca]

versus sea surface temperature relationship for Porites corals. Geochim Cosmochim

Acta 60: 3849-3858

Swart PK (1983) Carbon and oxygen isotope fractionation in scleractinian corals: A review.

Earth Sci Rev 19: 51-80

Swart PK, Leder JJ, Szmant AM, Dodge RE (1996) The origin of variations in the isotopic

record of scleractinian corals: II Carbon. Geochim Cosmochim Acta 60: 2871-2885

Weber JN, Woodhead PMJ (1972) Temperature dependence of oxygen-18 concentration in

reef coral carbonates. J Geophys Res 77: 463-705

82

Wefer G, Berger WH (1991) Isotope paleontology: growth and composition of extant

calcareous species. Mal' Geol 100: 207-248

Weiss H, Courty M-A, Wetterstrom W, Guichard F, Senior L, Meadow R, Currnow A (1993)

The genesis and collapse of third millennium north Mesopotamian civilisation; Science.

261: 995-1004

Wellington GM, Dunbar RB, Merlen G (1996) Calibration of stable oxygen isotope signatures

in Galfipagos corals. Pa1eoceanogr 11: 467-480

Westendorf W, Henfling E, (1989): Das Klima im Alten Ägypten nach Hieroglyphen­

aufzeichungen in: Klimaforschungprogramm des Bundesminsterium für Forschung und

Technologie "Statusseminar", 493-494

Wolf-Vecht A, Pa1dor N, Brenner S (1992) Hydrographie indications of advectionlconvection

effects in the Gu1fofElat. Deep-Sea Res 39: 1393-1401

83

4. CONCLUSIONS

4.1. General conclusions

From stable isotope analyses of oxygen in fossil eorals (Porites spp.), and fossil and reeent

molluses (Tridacna spp.) from the Gulf of Aqaba (northern Red Sea), and reeent eoral eares

(Porites spp.) from Ras Umm Sidd (northern Red Sea proper) the following eonclusions on

climate and oeean variability in the nOl1hern Red Sea during the mid- to late Holoeene have

been drawn:

a) Recent coral cores (Porites sp.)

The 8180 reeords oftwo coral Porites spp. eares from the subtropieal northern Red Sea proper

provide a bimonthly resolution for the past 96 and 50 years (manuseript 3.1). These reeords

eonelate differently with other eoral 8180 reeords from the same site (Felis 1999; Pätzold and

Klein unpublished data) on an annual time seale. Possibly, local biological and miero­

environmental effects have an influence on the 8 180 recards. All bimonthly sampled eorals

refleet an average environment signal as a result of similar annual growth rates and seasonal

coral 8 180 amplitudes. The longest three annual coral 8180 records also correlate differently

with the regional COADS SSTs. The correlation coefficients varied from -0.45 (RUS-95) to

-0.60 (RUS-93a). When the COADS SST data set is eorrelated with the average of the three

caral 8180 records between 1907 and 1992, a highly signifieant and very good correlation is

reeeived (r=-0.68). It is advisable to analyse more than one earal from a certain locality to

assurne a bettel' caral chronology to be applied in coral-based paleoclimate studies. A long­

term temperature rise of 1.1 °C between 1907 and 1992 is inferred from the isotope reeard

(RUS-93a). This long-term trend is comparable with the long-term trend in instrumentally

measured COADS SST (about 1°C) in the northern Red Sea (except for the data gaps). The

8180/SST gradient varied between 0.15 to 0.18%0, with a mean value of 0.165%0 as has been

shown in a previous study (Felis 1999). Cross-speetral analysis of two coral 8 180 recards

(RUS-93a and RUS-93b) shows a coherency at periods of 3.6, 5.8 and 10 years. These

speetral analysis results indicate the impact of the North Atlantie Oscillation (NAO) and the

Southern Oscillation (SO) i.e. SW monsoon, on the climate variability in the n011hern Red

Sea, which affeeted signifieantly the eoral 8 180 at least during the last 50 years. Furthermore,

the results support the previous finding of Felis (1999) that both the NAO and ENSO

eontributed signifieantly to the climate variability of the Middle East, approximately during

the past 250 years.

84

b) Mid-Holocene to recent molluscs (Tridacna sp.)

The manuscript 3.2 shows that the average amplitude of seasonal 8180 (1.17%0) cycles in

recent Tridacna sp. shell carbonates from the northern Gulf of Aqaba (Red Sea) agree with

mean seasonal temperature (5.3°C) variation measured in the nOlihern Red Sea region. The

seasonal 8 180 amplitude (temperature) of fossil Tridacna spp. varied from 1%0 (4.5°C) in the

mid-Holocene to 1.27%0 (6°C) in the late-Holocene. The mean oxygen isotope compositions

in recent and fossil Tridacna spp. indicate that during the last 6000 years BP the 8180

variation was 1.1 %0. The mean 8180 variations in Tridacna shells during the last 6000 years

BP also represent the variation of 8180 in sea surface water (i.e. salinity). The variable

intensity ofthe AfricanlAsian SW monsoon is responsible for this 8180 variation in sea water

(manuscript 3.1). This change could be caused a salinity variation of up to 1.5%0 in the

northern Red Sea during the last 6000 years BP.

c) Mid-Holocene corals (Porites sp.)

The manuscript 3.3 provides the first marine paleoclimatic record in the northern Red Sea

(Middle East - NE Africa region) during the mid-Holocene. However, the terrestrial records

are dominant. In this study, the seasonal 8 180 amplitudes of five mid-Holocene coral colonies

are higher than in recent corals. This is most probably due to a larger seasonal temperature

contrast and a reduction of salinity during the summer season for the mid-Holocene. Our

results support the hypotheses of summer monsoon rains reaching the nOlihern Red Sea

during mid-Holocene times, whereas seasonal solar radiation was enhanced. A climatic

change seemed to have occurred between about 4900 and 4600 years BP. This date coincides

with aperiod of rapid decline in north African lakes and Nile levels, indicating a reduction of

moisture transport from the ocean. Enhanced seasonalities can be reconstructed at least until

4450 14C years BP. Both oxygen and carbon isotope signals display heavier values in the

fossil records. The reduction of coral growth during the mid-Holocene is probably triggered

by an increased input and the resuspension ofterrestrial sediments.

85

4.2 Future application of fossil corals in paleoclimatic studies

The northern Red Sea is located on the Pole-Equator-Pole Transect that runs through Europe,

the Middle East and Africa (PEP UI). PEP III is an IGBP-PAGES project concerned with

climate change in Europe and Africa. One of the most important question of this project is:

Can we identify wet speIls during the late Holocene in the region of the drylands across North

Africa and the Arabic Peninsula? Therefore, the northern Red Sea is an important region of

global paleoenvironmental research. Molluscs and corals from modern and fossil reefs of the

northern Red Sea provide important high resolution paleoclimatic information during the

Holocene times and support other marine and terrestrial records of past climate variability

from the regional to the global scale.

Paleoclimatic work on corals and molluscs 1S 111 progress in the northern Red Sea. More

recently, mid-Holocene and marine isotope stage 5.1 corals and molluscs were collected from

the raised reef terraces of Aqaba (northern Gulf of Aqaba, Jordan) (AI-Rousan and Felis,

personal communication, 1999). These reefterraces were intensively studied by AI-Rifaiy and

Cherif (1988) and Dullo (1990). It is planned to generate long 8 180 time series (hopefully 100

years), which will provide more information on the climate variability in the Middle East

during the mid-Ho1ocene and the interglacial stage 5 by using spectral analysis. Also,

sediment cores were collected at the northern end of the Gulf of Aqaba during the Meteor

cruise (M 44/3) in April 1999 (Pätzold et al. 2000). The combination of the coral- and

mollusc-based paleoclimatic information with high-resolution records of these marine

sediment cores is expected to shed more light on the climatic and oceanic variability in the

northern Red Sea during the Holocene. Work on these sediment cores is already in progress

(AI-Rousan and Arz, personal communication, 1999).

However, 8 180 alone cannot resolve the relative contribution of SST and SSS. Therefore

additional ~tracers are needed. Sr/Ca ratios in corals vary as a function of temperature. The

Sr/Ca ratio of ocean water is considered constant over larger time scales, although some work

suggests that ocean water also reveals a variability in the Sr/Ca ratio. Sr/Ca ratios in coral

skeletons may also exhibit some growth rate dependence (deVilliers et al. 1995). Calibrations

of the Sr/Ca thermometer reveal consistent results in many areas despite some discrepancies

(deViIliers et al. 1995; Shen et al. 1996). Recent applications ofa multi-proxy approach using

8180 and Sr/Ca in mid-Holocene coral skeletons from the tropical western Pacific provided

convincing results considering paleotemperatures and variations of evaporation, precipitation,

and ocean surface salinity (Gagan et al. 1998). Moreover, coral Sr/Ca records representing

different periods during the early- and mid-Holocene were used to reconstruct abrupt SST

changes in this region (Beck et al. 1997). Schrag 1997 presented a new procedure for more

rapid determination of high-resolution elemental ratios (Sr/Ca, Mg/Ca) with high precision to

distinguish between temperature and salinity effects in cora1s. The results (coupled 8180 and

86

Sr/Catare used to study the long-term evolution of temperature and aridity in the northern

Red Sea. Further trace element analyses of fossil coral records from the northern Red Sea

reaching back to the early and late Holocene will contribute to the elucidation of climatic

changes in the Near East region.

4.3 References

Al-Rifaiy IA, Cherif, OH (1988) The fossil coral reefs of Al-Aqaba, Jordan. Facies 18: 219­

230

Beck JW, Recy J, Taylor F, Edwards RL, Cabioch G (1997) Abrupt changes in early

Holocene tropical sea surface temperature derived from coral records. Nature 385: 705­

707

deVilliers S, Nelson BK, Chivas AR (1995) Biological controls on coral Sr/Ca and 0180

reconstructions ofsea surface temperatures. Science 269: 1247-1249

Dullo W-C (1990) Facies fossil record, and age of Pleistocene reefs from the Red Sea (Saudi

Arabia). Facies 22: 1-46

Felis T (1999) Climate and ocean variability reconstructed from stable isotope records of

modern subtropical corals (northern Red Sea). Berichte, Fachbereich

Geowissenschaften, Universität Bremen, NI' 133, 111 Seiten, Bremen

Gagan MK, Ayliffe LK, Hopley D, Cali JA, Mortimer GE, Chappell J, McCulloch MT, Head

MJ (1998) Temperature and surface-ocean water balance of the mid-Holocene tropical

western Pacific. Science 279: 1014-1018

Pätzold J, Abd El-Wahab Farha 0, Abu-Ouf M, Al Hazmi YMM, Al-Rousan SA, Arz HW,

Bagabas KAA, Bassek D, Blaschk H, Böke W, Donner B, Edel' W, Felis T, Gayed

HYK, Gutowski M, Hemleben ChI', Hübner H, Hübscher Chr, Kadi KA, Kästner R,

Klauke S, Körner SO, Kuhlmann H, Lützeler T, Meir S, Melegy MM, Moammer MO,

Mohamuda AZ, Mokhtar TA, Moos ChI', Omar M, Rasheed M, Rosaik U, Salem M,

Schmidt M, Scmitt M, Shata AM, Themann S (2000) Report and preliminary results of

Meteor-cruise M 44/3, AQABA (JORDAN) - SAFAGA (EGYPT) - HAlFA

(ISRAEL), 12.03.-07.04.1999. Berichte, Fachbereich Geowissenschaften, Universität

Bremen, No 149, RV METEOR Cruise 44 Leg 3,1999

87

Schrag DP (1997) Temperature and salinity history of the northern Red Sea from high­

resolution e1emental and stable isotope records (abstract). Eos Trans AGU: 78 (46),

F387

Shen C-C, Lee T, Chen C-Y, Wang C-H, Dai C-F, Li L-A. (1996) The calibration ofD[Sr/Ca]

versus sea surface temperature relationship for Porites corals. Geochim Cosmochim

Acta 60: 3849-3858

5. ApPENDIX: DATA TABLES

5.1. Fossil corals (POI'ües spp.)

5.1.1 Stable isotope data

o'c 0'0(%0 PDß) (%0 PDß)

I 0.00 -0.07 -3.042 0.06 -0.13 -2.423 0.12 0.14 -1.934 0.18 0.01 -2.295 0.24 0.16 -1.966 0.30 0.21 -2.047 0.36 0.21 -2.018 0.42 0.42 -1.819 0.48 0.53 -2.1210 0.54 0.49 -2.27II 0.60 0.52 -2.5812 0.66 0.48 -2.6113 0.72 0.46 -2.9214 0.78 0.44 -2.8215 0.84 0.21 -2.9216 0.90 0.10 -2.7117 0.96 0.03 -2.6518 1.02 0.02 -24419 1.08 0.02 -2.1820 1.14 0.06 -2.0521 1.20 0.32 -1.7822 1.25 0.54 -1.6923 1.30 0.51 -2.0224 1.35 0.15 -2.3825 1.40 -0.02 -2.6826 1.45 0.01 -2.9027 1.50 0.15 -3.0328 1.55 0.15 -3.0229 1.60 0.24 -2.8230 1.65 0.07 -2.8331 1.70 0.09 -2.6232 1.75 0.24 -2.2433 1.80 0.08 -2.3534 1.85 027 -1.9135 1.90 0.30 -2.0636 1.95 0.26 -2.0137 2.00 0.04 -2.2438 2.05 -0.05 -2.5939 2.10 0.13 -2.6540 2.15- 0.14 -2.7741 2.20 0.07 -2.9642 2.25 0.23 -2.9643 2.30 0.34 -2.8744 2.35 0.21 -2.8345 2.40 0.14 -3.0446 2.45 0.09 -2.7147 2.50 0.29 -2.6248 2.55 0.21 -2.4449 2.60 0.16 -2.2450 2.65 0.13 -2.1151 2.70 0.12 -1.9452 277 0.15 -1.90'i" 2.84 0.27 -2.05_Cl

54 2.91 0.41 -2.2655 2.98 0.26 -2.6156 3.05 0.03 -2.7957 3.12 0.05 -2.9158 3.19 -0.12 -3.2059 3.26 -0.03 -3.1960 3.33 -0.12 -3.06

88

StißGwle Profile depth OC 0"0

(CI11) (%0 PDß) (%0 PDß)

64 0.12 0.97 -2.1465 0.16 0.75 -2.2066 0.20 0.97 -1.9567 0.24 1.20 -1.5168 0.28 0.71 -1.8669 0.32 0.81 -1.6270 0.36 0.39 -2.0271 0.40 0.24 -2.4172 0.45 -0.09 -2.5973 0.50 0.06 -2.8474 0.55 0.00 -2.7575 0.60 0.01 -2.5576 0.65 0.15 -2.4077 0.70 0.14 -2.3078 0.75 0.46 -1.8379 0.80 0.76 -1.5780 0.85 0.14 -2.2081 0.90 0.03 -2.3182 0.96 0.22 -2.6083 1.02 0.06 -2.9584 1.08 0.10 -2.8485 1.14 -0.23 -2.7886 1.20 -0.27 -2.7187 1.26 -0.24 -2.5488 1.32 -0.31 -2.4689 1.38 -0.33 -2.2890 1.44 -0.24 -2.0891 1.50 0.06 -1.7792 1.56 0.01 -1.8793 1.62 0.04 -2.1294 1.68 0.06 -2.4195 1.74 0.07 -2.5796 1.80 -0.04 -2.7697 1.86 0.15 -2.8198 1.92 0.10 -2.8999 1.98 0.10 -2.85100 2.04 -0.01 -2.91101 2.10 -0.18 -2.73102 2.15 -0.75 -1.88103 2.20 002 -189104 2.25 0.18 -1.86105 2.30 0.14 -1.80106 2.35 0.08 -2.12107 2.40 -0.13 -2.40108 2.45 -0.02 -2.75109 2.50 0.12 -3.05110 2.55 0.01 -3.27111 2.60 0.22 -3.11112 2.66 0.61 -2.74113 2.72 0.41 -2.62114 2.78 0.33 -2.14115 2.84 0.33 -1.74116 2.90 0.52 -1.62117 2.96 0.62 -1.71118 3.02 0.61 -2.16119 3.08 0.53 -2.56120 3.14 0.68 -2.60121 3.20 0.64 -2.69122 3.26 0.74 -2.49123 3.32 0.40 -2.65124 3.38 0.72 -1.81125 3.44 0.89 -1.39126 3.50 0.72 -1.61127 3.56 0.39 -1.94

/)'C /) 0(%0 PDß) (%0 PDß)

I 0.00 -1.0 I -3.512 0.03 -1.37 -3.493 0.06 -1.04 -3.414 0.09 -1.09 -3.325 0.12 -1.14 -3.346 0.15 -1.00 -3.117 0.18 -1.43 -2.908 0.21 -1.54 -2.609 0.24 -1.44 -2.2910 0.27 -0.95 -2.1611 0.31 -0.15 -2.0412 0.35 0.09 -2.3613 0.39 -1.35 -2.6414 0.43 0.21 -2.9115 0.47 -0.14 -3.0716 0.51 -0.66 -2.9217 0.55 -1.00 -2.4718 059 -0.26 -2.0719 0.63 0.02 -2.2920 0.67 0.06 -2.6021 0.72 0.08 -3.0622 0.77 -0.70 -2.6823 0.82 -0.80 -2.3924 0.87 -0.40 -2.1325 0.92 -0.08 -2.0826 0.97 -0.35 -2.7427 1.02 -0.56 -3.3128 1.07 -0.45 -3.0929 1.12 -0.61 -2.6230 1.17 -0.63 -2.0831 1.23 -0.09 -1.9232 1.29 0.05 -2.0833 1.35 -0.96 -3.0134 1.41 -0.72 -3.1435 1.47 -0.81 -3.1236 1.53 -1.13 -2.5937 1.59 -1.22 -2.1138 1.65 -0.97 -1.8339 1.71 -0.07 -1.8740 1.77 -0.38 -2.6641 1.82 -0.32 -3.0342 1.87 -0.48 -2.8343 1.92 -0.77 -2.5044 1.97 -0.77 -2.1645 2.02 0.07 -1.7546 2.07 -0.01 -2.7947 2.12 -0.13 -3.0148 2.17 -0.75 -2.5049 2.22 -0.87 -2.4550 2.27 -1.28 -2.3651 2.32 -1.06 -2.1952 2.37 -0.68 -2.2453 2.42 -0.18 -2.2154 2.47 0.19 -2.6855 2.52 0.02 -3.2256 2.57 -0.06 -3.1457 2.62 -0.29 -2.9158 2.67 -0.49 -2.4959 2.72 -0.63 -2.2460 2.77 -0.47 -2.1161 2.82 -0.34 -2.4162 2.87 0.17 -2.4663 2.92 0.03 -2.9764 2.97 -0.18 -3.2465 3.02 -0.44 -3.1766 3.07 -0.94 -2.7067 3.12 -0.91 -2.5168 3.17 -0.80 -2.1069 3.22 -0.77 -2.2870 3.27 -0.23 -2.0071 3.32 -0.44 -2.6772 3.37 -0.46 -2.9873 3.41 -0.29 -3.10

89

s\~ii~eProti le depth /)C /) 0

(cm) (%0 PDß) (%0 PDß)

74 3.47 -0.18 -3.0875 3.52 -0.45 -29076 3.57 -0.69 -2.7377 3.62 -0.80 -2.3578 3.67 -0.67 -2.3179 3.72 -0.20 -2.0780 3.77 -0.15 -2.3181 3.81 -0.05 -3.2082 3.85 -0.17 -3.1783 3.89 -0.41 -3.1484 3.93 -0.58 -3.0685 3.97 -0.87 -2.6686 4.10 -0.72 -2.7487 4.05 -0.67 -2.2488 4.09 -0.65 -2.0989 4.13 -0.33 -2.0090 4.17 -0.12 -2.2291 4.22 0.06 -2.6092 4.27 -0.15 -3.0893 4.32 -0.09 -3.2794 4.37 -0.15 -3.2195 4.42 -0.54 -3.1596 4.47 -0.98 -2.9997 4.52 -1.61 -25598 4.57 -1.99 -2.5599 4.62 -1.71 -2.23100 4.67 -1.65 -2.23101 4.71 -1.12 -2.17102 4.75 -0.13 -2.53103 4.79 -0.14 -3.04104 4.83 -0.12 -3.19105 4.87 0.05 -3.37106 4.91 -0.03 -3.35107 4.95 0.10 -3.28108 4.99 0.07 -3.23109 5.03 -0.15 -3.18110 5.07 -0.32 -3.03111 5.11 -0.33 -3.09112 5.15 -0.42 -3.00113 519 -0.54 -3.00114 5.23 -1.00 -2.74115 5.27 -1.35 -2.70116 5.31 -1.26 -2.63117 5.35 -1.39 -2.41118 5.39 -1.35 -2.17119 5.43 -1.45 -2.21120 5.47 -1.21 -2.17121 5.50 -0.36 -2.59122 5.53

-0.11 -2.37

123 5.56 0.12 -2.65124 5.59 0.23 -2.93125 562 0.08 -286126 5.65 -0.29 -3.14127 5.68 -1.15 -2.84128 5.71 -0.60 -298129 5.74 -1.07 -2.90130 5.77 -0.82 -2.82131 5.82 -1.29 -2.46132 5.87 -1.36 -2.45133 5.92 -1.19 -2.52134 5.97 -0.69 -2.55135 6.02 -0.36 -2.72136 6.07 -0.12 -2.94137 6.12 -0.06 -3.01138 6.17 0.02 -3.34139 6.22 0.24 -3.57140 6.27 0.15 -3.50141 630 -0.29 -3.38142 6.33 -0.67 -3.09143 6.36 -1.06 -3.04144 6.39 -1.40 -2.82145 6.42 -1.65 -2.57146 6.45 -1.60 -2.66

90

Profile depth 8"C 81'0 Profile depth OC 8~0

(C111) (%0 POß) (%0 POß) (C111) (%0 POß) (%0 POß)

147 6.48 -1.35 -2.30 170 7.27 0.49 -2.80

148 6.51 -1.41 -2.40 171 7.31 0.43 -2.70

149 6.54 -1.13 -2.54 172 7.35 0.32 -3.14

150 6.57 -0.77 -2.44 173 7.39 0.26 -3.05

151 6.61 -0.74 -2.55 174 7.43 0.05 -2.94

152 6.65 -0.19 -2.91 175 7.47 -0.18 -2.63

153 6.69 -0.24 -2.96 176 7.51 -0.40 -2.48

154 6.73 -0.20 -3.27 177 755 -0.21 -2.18155 6.77 -0.08 -3.34 178 7.59 -0.01 -2.05156 6.81 -0.11 -3.37 179 7.63 -0.70 -2.37157 6.85 -0.17 -3.20 180 7.67 0.24 -2.26158 6.89 -0.36 -3.24 181 7.71 0.41 -2.14159 6.93 -0.34 -3.10 182 7.75 0.15 -2.71160 6.97 -0.77 -2.95 183 7.79 0.51 -2.74161 7.00 -0.91 -2.78 184 7.83 0.47 -2.82162 7.03 -0.69 -2.19 185 7.87 0.33 -2.78163 7.06 -0.76 -2.39 186 7.91 -0.06 -2.54164 7.09 -1.02 -2.43 187 7.95 0.06 -2.64165 7.12 -0.84 -2.40 188 7.99 -0.27 -2.43166 715 -0.64 -2.34 189 8.03 0.47 -1.97167 7.18 -0.58 -2.29 190 8.07 0.49 -2.34168 7.21 -0.05 -2.36 191 8.17 0.86 -283169 7.24 0.23 -2.65 192 8.27 0.77 -2.66

OIC 01'0(%0 POß) (%0 POß)

1 0.00 -0.19 -2.152 0.05 -0.33 -2.023 0.10 -0.36 -1.674 0.15 -0.35 -1.935 0.20 -0.65 -1.736 0.25 -0.58 -1.587 030 -0.69 -1.988 0.35 -0.61 -2.139 0.40 -0.44 -2.2810 0.45 -0.38 -2.6811 0.50 -0.48 -2.7012 053 -0.48 -2.9913 0.56 -0.57 -30914 0.61 -0.62 -2.8515 0.64 -1.02 -2.9416 0.67 -1.04 -2.8317 0.70 -0.74 -2.7718 0.73 -0.65 -2.1119 0.76 -0.29 -1.6020 0.79 -0.63 -1.6121 0.82 -0.78 -1.6222 0.85 -0.84 -I. 7723 0.88 -0.89 -2.0624 0.91 -0.60 -1.9525 0.94 -0.83 -2.2726 0.97 -0.68 -2.6027 1.00 -0.68 -2.7828 1.03 -0.71 -2.6529 1.06 -0.73 -2.6130 1.09 -1.08 -2.1731 1.12 -0.60 -1.9832 1.15 -0.61 -1.4833 1.18 -0.66 -1.2734 1.21 -0.85 -1.4835 1.24 -0.99 -1.6936 1.27 -1.02 -1.9137 1.30 -1.0 I -2.4338 1.33 -1.04 -2.82

~~~leProlile depth olTC 81~0

(C111) (%0 POß) (%0 POß)

39 1.36 -0.89 -2.6940 1.39 -0.86 -3.0441 1.41 -1.39 -3.0642 1.43 -0.97 -2.7143 1.45 -0.79 -2.6344 1.47 -0.76 -2.1145 1.49 -0.58 -1.9146 1.51 -0.52 -1.5647 1.53 -0.67 -1.4848 1.55 -0.94 -1.7949 1.57 -0.97 -1.9850 1.59 -0.85 -1.9951 1.62 -0.88 -2.5452 1.65 -0.85 -2.6753 1.68 -0.52 -2.6154 1.71 -0.08 -2.2655 1.74 -0.04 -2.1256 1.77 0.18 -1.9957 1.80 -0.40 -1.5258 1.83 -0.53 -1.7959 1.86 -0.64 -1.8660 1.89 -0.76 -2.0961 1.92 -0.62 -2.4562 1.95 -0.35 -2.3563 1.98 -0.21 -2.7164 2.01 -0.43 -2.5265 2.04 -0.16 -2.1366 2.07 -0.14 -1.7067 2.10 -0.31 -1.8068 2.13 -0.33 -2.0969 2.16 -0.46 -2.7070 2.19 -0.22 -2.4971 2.23 -0.08 -21872 2.27 -0.01 -1.7773 2.31 -0.49 -1.8074 2.35 -0.75 -1.9075 2.39 -0.98 -2.2576 2.43 -0.79 -2.77

- s~i;~eProfile depth 8C 8'"0

(cm) (%0 POß) (%0 POß)

77 2.47 -0.50 -2.7778 2.51 -0.46 -2.8079 2.55 -0.97 -2.6580 2.59 -0.97 -2.2581 2.63 -0.70 -1.9282 267 -0.65 -1.6883 2.71 -0.40 -1.6784 2.75 -1.00 -2.4185 2.79 -1.09 -2.7286 2.83 -1.08 -2.9287 2.87 -1.18 -2.9988 2.91 -1.14 -2.9689 2.95 -1.11 -2.8390 299 -1.11 -2.6791 3.02 -0.88 -2.3692 3.05 -0.89 -2.3493 3.08 -0.84 -1.9494 3.11 -1.04 -1.9595 3.14 -1.16 -2.1196 3.17 -1.00 -2.3297 3.20 -1.03 -2.5798 3.23 -0.58 -2.5299 3.26 -0.54 -2.71100 3.29 -0.80 -2.71101 3.34 -0.95 -2.43102 3.39 -0.82 -2.08103 3.44 -0.68 -1.82104 3.49 -0.81 -1.85105 3.54 -0.68 -2.21106 3.59 -0.24 -2.66107 3.64 -0.20 -2.41108 3.69 -0.23 -1.98109 374 -0.46 -1.57110 3.79 -0.72 -1.83111 3.83 -0.64 -2.04112 3.87 -0.79 -2.40113 3.91 -0.93 -2.86114 3.95 -1.02 -2.89

Sampie 8'C 8"0t:w~~~ (%0 POß) (%0 POß)

1 0.00 0.81 -1.892 0.06 0.56 -2.113 0.12 0.30 -2.534 0.18 0.22 -2.555 0.24 -0.38 -2.756 0.30 -0.96 -2.427 0.36 -0.40 -1.878 0.42 0.67 -1.549 0.48 1.l8 -1.8110 0.54 1.34 -2.20II 0.60 0.80 -2.4812 0.65 0.24 -27313 0.70 -0.57 -2.5914 0.75 -0.73 -1.9415 0.80 0.39 -1.3716 0.85 0.42 -1.8317 0.90 1.55 -1.6418 0.95 1.58 -2.1019 1.00 1.37 -1.7920 1.05 0.74 -2.5621 1.12 -0.37 -2.8522 1.19 -0.74 -2.4323 1.26 0.28 -1.7424 1.33 1.59 -1.3725 1.40 1.64 -2.3726 1.47 0.03 -2.7027 1.54 -0.07 -1.79

91

~~~,~eProfile depth 8C 81°0

(cm) (%0 POß) (%0 POß)

115 3.99 -0.98 -2.53116 4.03 -0.79 -1.91117 4.07 -0.85 -1.68118 4.11 -1.14 -1.72119 4.15 -1.20 -2.09120 4.19 -1.15 -2.43121 4.25 -1.05 -2.95122 4.31 -1.09 -3.02123 4.37 -0.85 -2.68124 4.43 -0.69 -194125 4.49 -0.57 -1.57126 4.55 -0.76 -1.80127 461 -0.81 -2.53128 4.67 -0.69 -2.56129 4.73 -0.72 -2.11130 4.78 -0.58 -166131 4.83 -0.90 -158132 4.87 -1.26 -166133 4.91 -103 -191134 4.95 -0.95 -2.46135 4.99 -1.19 -2.89136 5.03 -0.60 -2.50137 5.07 -0.24 -2.22138 5.11 -0.07 -163139 5.15 -0.48 -160140 5.19 -0.88 -196141 5.24 -0.94 -2.42142 5.29 -0.84 -2.71143 5.34 -0.54 -2.40144 5.39 -0.17 -2.05145 5.44 -0.19 -185146 5.49 -0.38 -2.05147 5.54 -0.64 -2.54148 5.59 -0.22 -2.56149 5.64 -0.30 -2.41150 5.69 -0.21 -187

Sampie Profile depth 8 C 8'·0~~~tl (cm) (%0 POß) (%0 POß)

28 161 1.44 -1.7129 1.68 0.50 -2.4830 1.75 - -0.93 -2.2731 1.79 -0.32 -1.5932 1.83 0.72 -12533 1.87 1.55 -1.1634 1.91 1. 73 -1.9335 1.95 1.31 -2.0736 1.99 0.80 -2.5037 2.03 -0.06 -2.9038 2.06 -0.90 -2.7839 2.09 -1.02 -2.2640 2.12 0.14 -1.3441 2.16 1.06 -1.3542 2.20 1.21 -1.6943 2.24 1.51 -2.0644 2.28 1.17 -2.2245 2.32 0.50 -2.3546 2.36 0.34 -2.2847 2.40 -0.36 -2.4248 2.44 -1.37 -2.3849 2.48 -0.95 -1.6650 2.52 0.36 -1.3751 2.58 1.12 -1.7852 2.64 0.59 -2.4753 2.70 -0.77 -2.7854 2.76 -0.77 -1.91

92

Profile depth 8' e 8"0 Protlle depth 8 e 81'0(cm) (%0 PDß) (%0 POß) (cm) i%o POß) (%0 POß)

55 2.82 0.34 -1.37 66 3.54 -1.76 -2.06

56 2.94 0.88 -1.76 67 3.60 -0.18 -1.49

57 3.00 0.54 -2.48 68 3.66 0.26 -1.74

58 3.06 -0.43 -2.87 69 3.72 0.19 -2.17

59 3.12 -0.94 -2.88 70 3.78 -0.30 -2.32

60 318 -1.23 -2.33 71 3.83 -1.15 -2.69

61 3.24 -028 -1.28 72 3.88 -2.01 -2.88

62 3.30 0.28 -1.85 73 3.93 -2.13 -2.48

63 3.36 0.33 -2.24 74 3.98 -2.29 -2.25

64 3.42 -0.59 -2.82 75 4.03 -1.62 -1.67

65 3.48 -1.52 -2.83

8'e 8180(%0 POß) (%0 PDß)

I 0.00 0.96 -2.012 0.06 0.73 -1.813 0.12 -0.09 -2.214 0.18 0.34 -2.475 0.24 0.04 -2.696 0.30 0.27 -2.627 0.36 0.86 -2.508 0.42 1.23 -2.419 0.48 1.29 -2.1510 0.54 1.28 -1.8711 0.60 1.07 -1.8212 0.64 0.56 -1.6313 0.68 0.57 -1.7714 0.72 0.01 -1.9815 0.76 0.28 -1.9616 0.80 0.00 -2.2917 0.84 0.12 -2.5318 0.88 0.57 -2.4519 0.92 0.78 -2.5120 0.96 1.11 -2.3021 1.00 1.06 -2.4522 1.04 1.14 -2.1223 1.08 1.23 -21224 1.12 0.79 -1.7025 1.16 0.56 -1.7426 1.20 0.35 -1.5127 1.24 -0.11 -1.8728 1.28 0.07 -1.7929 1.32 0.18 -2.2130 1.36 -0.04 -2.6631 1.40 -0.13 -2.3932 1.45 -0.39 -2.4933 1.50 0.33 -2.9334 1.55 0.56 -2.5435 1.60 0.77 -2.5036 1.65 0.92 -2.2737 1.70 0.94 -2.0938 1.75 0.37 -1.6639 1.80 -0.33 -1.7140 1.85 -0.49 -1.8141 1.90 -0.38 -1.8042 1.94 -0.41 -2.2443 1.98 -0.16 -2.3544 2.02 0.07 -2.4445 2.06 0.23 -2.6346 2.10 0.13 -2.5147 2.14 0.27 -2.3048 2.18 -0.02 -2.3349 2.22 0.06 -2.1650 2.26 -0.12 -2.3251 2.30 -0.28 -2.7752 2.36 0.38 -2.6453 2.42 0.36 -2.9454 2.48 1.19 -2.53

~~iiliProfile depth 8e 8180

(cm) (%0 POß) (%0 POß)

55 2.54 1.64 -2.0056 2.60 1.49 -1.5857 2.6658 2.72 0.31 -1.5559 2.78 -0.93 -2.5360 2.84 -0.58 -2.4061 2.90 -0.06 -2.7062 2.94 0.04 -2.9063 2.98 UO -2.6964 3.02 1.32 -2.3965 3.06 1.41 -2.0566 3.10 1.30 -2.0867 3.14 1.20 -1.6468 3.18 1.04 -1.4569 3.22 0.12 -1.5370 3.26 -0.29 -1.9271 3.30 -0.37 -2.0872 3.35 -0.75 -1.6973 3.40 -0.62 -2.7674 3.45 -0.37 -3.0475 3.50 -0.44 -3.0176 3.55 -0.14 -3.2477 3.60 OA3 -3.0178 3.65 091 -2.8779 3.70 0.88 -2.7680 3.75 lAI -2.3581 3.80 U8 -1.9982 3.87 -0.07 -1.5683 3.94 -0.62 -1.7584 4.01 -1.34 -2.1985 4.08 -1.13 -2.4286 4.15 -1.22 -2.7587 4.22 -0.98 -2.9088 4.29 -0.64 -3.0189 4.36 0.17 -3.0190 4.43 0.45 -2.8691 4.50 0.80 -2.6992 4.56 1.55 -2.0893 4.62 1.62 -1.8394 4.68 1.52 -1.6495 4.74 0.75 -1.2596 4.80 -0.51 -1.7197 4.86 -U3 -2.2898 4.92 -0.81 -2.6199 4.98 -0.72 -2.94100 5.04 0.13 -2.92101 5.10 0.97 -2.40102 5.14 1.79 -1.80103 5.18 1.34 -1.44104 5.22 0.66 -1.45105 5.26 0.14 -1.48106 5.30 -0.81 -2.26107 5.34 -0.75 -2.50108 5.38 -0.39 -2.62

ti!iil\~Profile depth 8 C 8'0

(cm) (%0 POS) (%0 POS)

109 5.42 -0.16 -2.85110 5.46 0.39 -2.92111 5.50 0.24 -2.85112 5.54 0,62 -2.72113 5.58 0.66 -2.32114 5.62 0.67 -2.16115 5.66 0.67 -2.06116 5.70 0.79 -1.67117 5.74 0.42 -1.46118 5.78 0.24 -1.43119 5.82 -0.58 -1.68120 5,86 -0.59 -2.04121 5,90 -0.21 -2.31122 5,95 0.10 -2.60123 6,00 0.55 -2.68124 6.05 0.60 -2.69125 6.10 094 -2.35126 6.15 1.07 -2.24127 6.20 1.06 -1.66128 6.25 1.00 -1.25129 6.30 0.21 -134130 6.35 -0.08 -1.64131 6.40 0.18 -1.43132 6.45 0.40 -2.08133 6.50 0.88 -1.83134 6.55 1.86 -0.85135 6.60 1.67 -1.55136 6.65 1.92 -1.10137 6.70 1.51 -1.41138 6.75 0.89 -1.41139 6.80 0.67 -1.30140 6.85 0.08 -1.66141 6.90 -0.42 -1.68142 6.96 -1.18 -2.23143 7.02 -0.06 -2.60144 7.08 -0.03 -2.50145 7.14 0.34 -2.45146 7.20 0,17 -2.38147 7.26 0.19 -2.48148 7.32 1.04 -1.75149 7.38 1.24 -137

93

~iil1Profi le depth 8C 8"0

(cm) (%0 POS) (%0 POS)

150 7.44 1.30 -1.16151 7.50 1.04 -0.85152 7.55 0.55 -1.11153 7.60 -0.33 -1.63154 7.65 -130 -2.28155 7.70 -0.39 -2,75156 7.75 -0.01 -2.64157 7.80 -0.02 -2.66158 7,85 0.80 -2.39159 7.90 1.05 -1.86160 7.95 0.99 -1.35161 8,00 -0.11 -1.53162 8.06 -0.64 -2.31163 8.12 -0.91 -2.71164 8.18 -0.82 -2.80165 8.24 -0.57 -307166 8.30 0.35 -2.79167 8.36 1.18 -2.33168 8.42 134 -2.15169 8.48 1.22 -1.66170 8.54 0.67 -1.09171 8.60 -0.22 -130172 8.67 -1.16 -1.89173 8.74 -1.03 -2.51174 8.81 -0.94 -2.59175 8.88 -0,09 -2.56176 8.95 0.44 -2.33177 9.02 0.93 -2.06178 9.09 1.27 -1,77179 9.16 0.85 -1.32180 9.23 -0.61 -1.90181 9.30 -0.92 -2.47182 9.36 -0.47 -2.55183 9.42 -0.01 -2.90184 9.48 -0.04 -2.88185 9.54 0.23 -2.88186 960 0:68 -2.75187 9.66 1.14 -2.44188 9.72 1.18 -2.18189 9,78 -0,01 -1.80190 9.84 0.14 -1.91

94

5.1.2 Annual Growth rate

Year Growth rate cm/yr Growth rate cm/yr Growth rate cl11/yr Growth rate cm/yr Growth rate cm/yr

1 0.84 0.47 0.51 0.36 0.682 0.66 0.20 0.42 0.26 0.443 0.75 0.35 0.35 0.25 0.704 0.94 0.39 0.27 0.40 0.365 0.50 0.41 0.27 0.30 0.566 0.52 0.30 0.20 0.43 0.577 1.02 0.40 0.44 0.47 0.818 0.69 0.45 0.37 021 0.629 0.65 0.44 0.36 0.29 0.4810 0.44 0.30 0.42 0.5911 0.47 0.33 0.40 0.4012 0.55 0.42 0.5713 0.78 0.34 0.6814 0.57 0.28 0.5415 0.59 0.33 0.5716 0.54 0.25 0.6117 0.4818 0.44

5.1.3 Seasonal amplitude

Seasonal ampl itude Seasonal ampl itude%0 %0

Seasonal amplitude%0

Year

123456789101112131415161718

1.111.251.051.301.331.381.131.471.07

1.030.590.681.211.201.261.031.101.111.281.200.961.121.080.940.761.240.86

1.511.181.541.191.191.011.031.330.760.831.321.340.981.300.930.76

1.181.210.961.321.560.981.411.601.341.21

0.741.140.921.141.791.451.691.481.260.831.751.911.721.511.58 ­0.47

5.2. Modern corals (Porites spp.)

5.2.1 Stable isotope data

95

0"0(%oPDB)

0.00 -3.960.20 -3.710.40 1993.125 -3.050.60 1992.858 -3630.75 1992.658 -3.530.95 1992.392 -3.471.15 1992.125 -3.101.35 1991.914 -3.451.55 1991.704 -3.691.75 1991.493 -3.811.95 1991.283 -3.352.10 1991.125 -3.142.30 1990.914 -3.522.45 1990.757 -3.762.65 1990.546 -3.552.85 1990.336 -3.113.05 1990.125 -3.093.20 1989.911 -3.303.40 1989.625 -3.563.55 1989.41 I -3.713.75 1989.125 -3.213.90 1988.982 -3.364.05 1988.839 -3.444.25 1988.649 -3.804.40 1988.506 -3.904.60 1988.315 -3.724.80 1988.125 -3.235.00 1987.914 -3.3 I5.20 1987.704 -3.635.40 1987.493 -3.765.55 1987.336 -3.285.75 1987.125 -3.135.95 1986.858 -3.626.10 1986.658 -3.906.30 1986.392 -3.706.50 1986.125 -3.216.70 1985.935 -3.476.90 1985.744 -3.577.10 1985.554 -3.547.25 1985.411 -3.837.40 1985.268 -3.487.55 1985.125 -3.317.75 1984.951 -3.357.95 1984.777 -3.808.10 1984.647 -3.708.30 1984.473 -3.798.50 1984.299 -3.378.70 1984.125 -3.128.85 1983.911 -3.409.00 1983.696 -3.579.20 1983.411 -3.849.40 1983.125 -2.979.60 1982.725 -3.729.75 1982.425 -3.979.90 1982.125 -3.3210.10 1981.817 -3.4 I10.25 1981.587 -3.8110.40 1981.356 -3.8810.55 1981.125 -3.2210.75 1980.875 -3.5110.90 1980.688 -3.5111.05 1980.500 -3.86

Core depth Age (years) 0"0(cm) (%oPDB)

11.20 1980.313 -3.9111.35 1980.125 -3.1411.50 1980.000 -3.1511.65 1979.875 -3.5311.85 1979.708 -3.5912.00 1979.583 -3.7712.15 1979.458 -3.8212.35 1979.292 -3.6212.55 1979.125 -2.9312.70 1978.975 -3.1512.90 1978.775 -3.4913.05 1978.625 -3.8413.25 1978.425 -3.0713.40 1978.275 -3.0513.55 1978.125 -2.6913.75 1977.890 -3.4613.95 1977.654 -3.9114.10 1977.478 -3.7414.25 1977 .30 I -3.4214.40 1977.125 -3.1014.60 1976.935 -3.1714.75 1976.792 -3.5814.95 1976.601 -3.7415.10 1976.458 -3.6815.25 1976.315 -3.6915.45 1976.125 -3.0815.60 1975.989 -31915.75 1975.852 -3.3315.90 1975.716 -3.7416.00 1975.625 -3.8516.20 1975.443 -3.6016.40 1975.261 -3.1116.55 1975.125 -3.1016.75 1974.858 -3.5616.95 1974.592 -3.7717.10 1974.392 -3.7717.30 1974.125 -3.1117.50 1973.890 -3.4617.65 1973.713 -3.6017.85 1973.478 -3.7818.00 1973.30 I -3.6118.15 1973.125 -3.1618.30 1972.975 -3.3918.50 1972.775 -3.7818.65 1972.625 -36618.80 1972.475 -3.6619.00 1972.275 -3.5519.15 1972.125 -3.0819.30 1971.958 -3.5619.45 1971.792 ~3.63

19.60 1971.625 -3.6119.75 1971.458 -3.6419.90 1971.292 -3.5920.05 1971.125 -3.1620.20 1970.911 -3.3320.40 1970.625 -3.5920.55 1970.411 -3.8320.75 1970.125 -3.2020.95 1969.935 -3.4721. 10 1969.792 -4.0921.25 1969.649 -3.7221.45 1969.458 -3.73

Core depth Age (years) 0"0(cm) (%oPDB)

21.60 1969.315 -3.4521.80 1969.125 -3.2222.00 1968.971 -3.2822.15 1968.856 -3.4222.30 1968.740 -3.4722.50 1968.587 -3.7222.65 1968.471 -3.5922.80 1968.356 -35323.00 1968.202 -3.2623.10 1968.125 -3.2223.35 1967.908 -3.4223.50 1967.777 -3.5923.70 1967.603 -3.6623.90 1967.429 -3.3324.10 1967.255 -3.1324.25 1967.125 -2.9424.40 1967.005 -3.4624.60 1966.845 -3.4624.80 1966.685 -3.5825.00 1966.525 -3.4425.10 1966.445 -3.4925.30 1966.285 -3.2325.50 1966.125 -3.1425.70 1965.971 -3.3225.90 1965.817 -3.6526.10 1965.663 -3.5626.30 1965.510 -3.8426.50 1965.356 -3.2926.60 1965.279 -2.9926.80 1965. I25 -2.9726.90 1965.025 -3.3027.10 1964.825 -3.4627.20 1964.725 -3.4427.40 1964.525 -3.7127.50 1964.425 -3.6627.70 1964.225 -3.3527.80 1964.125 -3.0028.00 1963.982 -3.0228.10 1963.911 -3.3528.30 1963.768 -3.7528.40 1963.696 -3.7328.60 1963.554 -3.8928.80 1963.411 -3.7328.90 1963.339 -3.5729.10 1963.196 -3.3729.20 1963.125 -3.1729.40 1962.875 -3.3429.50 1962.750 -3.4229.60 1962.625 -3.6229.80 1962.375 -3.7330.00 1962.125 -3.5230.15 1961.989 -3.6230.30 1961.852 -3.7230.50 1961.670 -3.6330.60 1961.580 -3.4330.80 1961.398 -3.1930.90 1961.307 -3.0231.10 1961.125 -2.7231.30 1960925 -3.2631.40 1960.825 -3.5631.60 1960.625 -3.7931.80 1960.425 -3.93

Core depth Age (years) 8'"0(cm) (%oPDB)

32.00 1960.225 -3.52 47.00 1944.569 -3453210 1960.125 -2.94 47.20 1944.347 -38532.30 1959.925 -3.10 4740 1944.125 -3.0132.50 1959.725 -3.36 47.60 1943.914 -3.1932.70 1959.525 -3.62 47.80 1943.704 -3.8232.90 1959.325 -348 48.10 1943.388 -3.3133.10 1959.125 -346 48.35 1943.125 -3.1233.30 1958.935 -3.54 48.55 1942.925 -34533.50 1958.744 -3.71 48.75 1942.725 -3.6933.65 1958.601 -3.55 48.95 1942.525 -3.6333.85 1958411 -3.31 49.15 1942.325 -3.5534.00 1958.268 -3.05 49.35 1942.125 -3.2534.15 1958.125 -3.06 49.60 1941.925 -3.3134.35 1957.890 -3.19 49.80 1941.765 -3.3034.55 1957.654 -3.35 50.00 1941.605 -3.8034.80 1957.360 -3.54 50.20 1941445 -3.923500 1957.125 -3.0 I 5040 1941.285 -36035.20 1956.965 -3.33 50.60 1941.125 -3173540 1956.805 -346 50.80 1940.903 -3.553560 1956.645 -349 51.05 1940.625 -3.7635.80 1956485 -3.68 51.30 1940.347 -3.6036.05 1956.285 -3.59 51.50 1940.125 -3.0236.25 1956.125 -3.16 51.70 1939.890 -3.7136.50 1955.933 -3.33 51.95 1939.596 -3.9736.70 1955.779 -3.61 52.10 1939419 -3.7336.90 1955.625 -3.79 52.35 1939.125 -3.333715 1955433 -3.83 52.55 1938.890 -34637.35 1955.279 -3.59 52.80 1938.596 -37437.55 1955.125 -3.09 53.00 1938360 -34837.75 1954.951 -3.31 53.20 1938.125 -3.0037.90 1954.821 -3.38 5340 1937.875 -34638.10 1954.647 -3.70 53.60 1937.625 -3.9138.30 1954473 -4.04 53.75 1937438 -3.5438.50 1954.299 -3.54 54.00 1937.125 -3.2638.70 1954.125 -3.05 54.20 1936.903 -3.5538.90 1953.935 -3.16 5445 1936.625 -3.9039.10 1953.744 -3.64 54.65 1936403 -3.7939.35 1953.506 -369 54.90 1936.125 -3.0939.55 1953.3 15 -347 55.15 1935.875 -3.3739.75 1953.125 -3.03 55.35 1935.675 -3.8139.95 1952.935 -3.52 55.55 1935475 -3.7140.15 1952.744 -3.66 55.75 1935.275 -3.5540.40 1952.506 -3.90 55.90 1935.125 -3.0940.60 1952.315 -347 56.15 1934.917 -3.5740.80 1952.125 -2.91 56.35 1934.750 -3.6441.00 1951.875 -3.44 5655 1934.583 -3.8141.20 1951.625 -3.74 56.75 1934.417 -3.6641.40 1951.375 -3.95 56.95 1934.250 -3.4641.60 1951.125 -3.28 57.10 1934.125 -3.1741.80 1950.971 -3.29 57.35 1933.847 -3.224200 1950.817 -3.64 57.55 1933.625 -3.4642.30 1950.587 -3.84 57.80 1933.347 -3.7542.50 1950.433 -3.67 58.00 1933.125 -3.3142.70 1950.279 -3.40 58.20 1932.914 -3.8142.90 1950.125 -3.06 58.40 1932.704 -3.6343.10 1949.725 -3.39 58.55 1932.546 -3.8943.30 1949.325 -3.78 58.75 1932.336 -3.4643.40 1949.125 -3.14 58.95 1932.125 -3.1143.60 1948.625 -3.59 59.10 1932.010 -3.2743.80 1948.125 -2.95 59.30 1931.856 -3.7244.00 1947.839 -3.74 59.45 1931.740 -3.7144.30 1947.411 -3.04 59.70 1931.548 -3.6844.50 1947.125 -3.02 5995 1931.356 -3.4944.70 1946.839 -3.81 60.10 1931.240 -3.3144.90 1946.554 -3.58 60.25 1931.125 -3.2045.20 1946.125 -3.22 60.45 1930.951 -3504540 1945.971 -3.34 60.60 1930.821 -3.6245.60 1945.817 -360 60.80 1930.647 -3.8545.80 1945663 -3.75 61.00 1930.473 -37246.10 1945.433 -3.93 61.20 1930.299 -34946.30 1945.279 -3.46 61.40 1930.125 -3.0946.50 1945.125 -2.86 61.60 1929.935 -3.1746.70 1944.903 -3.56 61.80 1929.744 -345

96

Core depth Age (years) 8'0(cm) (%oPDB)

62.05 1929.506 -3.5762.25 1929.315 -3.4462.45 1929.125 -31762.65 1928.982 -31962.80 1928875 -3.2863.05 1928.696 -3.4163.25 1928.554 -3.546345 1928.411 -3.5163.65 1928.268 -3.2763.85 1928.125 -3.0864.05 1927.977 -3.2864.25 1927.829 -3.7864.45 1927.681 -3.6964.65 1927.532 -3.8364.80 1927.421 -3436505 1927236 -3.0965.20 1927.125 -3.0465.45 1926.953 -3.2665.60 1926.849 -3.5465.85 1926.677 -3.8466.00 1926.573 -3.9466.20 1926.435 -3.6166.45 1926.263 -3.1566.65 1926.125 -2.9166.85 1925.987 -3.2167.05 1925.849 -3.4967.30 1925.677 -3.5867.50 1925539 -3.8867.70 1925.401 -3.5267.90 1925.263 -3.1468.10 1925.125 -3.0168.30 1924.935 -3.4568.50 1924.744 -3.8068.75 1924.506 -3.5668.95 1924.3 15 -34769.15 1924.125 -3.1469.40 1923.940 -3.4069.60 1923.792 -3.6569.85 1923.606 -38670.10 1923.421 -3.5670.30 1923.273 -3.2270.50 1923.125 -3.1670.70 1922.951 -3.3770.85 1922.821 -3.6571.05 1922.647 -3.7671.25 1922.473 -3.3971.45 1922.299 -3.1571.65 1922.125 -3.1471.85 1921.987 -3.2572.10 1921.815 -3.4572.30 1921.677 -3.5772.50 1921.539 -3.5872.70 1921.401 -3.5572.90 1921.263 -3.0973.10 1921.125 -3.0473.30 1920.982 -3.3673.50 1920.839 -3.2973.60 1920.768 -3.4673.80 1920.625 -3.4074.10 1920.411 -3.1774.30 1920.268 -3.0774.50 1920.125 -2.9074.70 1919.943 -3.5674.90 1919.761 -3.6475.00 1919.670 -3.3975.20 1919.489 -3.1575.40 1919.307 -3.1275.60 1919.125 -3.0875.80 1918.971 -3.3176.10 1918.740 -3.4876.30 1918.587 -3.7376.50 1918.433 -36076.70 1918.279 -3.30

Core depth Age (years) 8'0(%oPDB) (cm) (%oPDB)

76.90 1918.125 -3.02 85.00 1911.971 -3.2977.10 1917.935 -3.58 85.20 1911.817 -3337730 1917.744 -3.51 85.40 1911.663 -3.6077.55 1917.506 -3.66 85.60 1911.510 -3.5677.75 1917315 -3.45 85.90 1911.279 -3.3277.95 1917.125 -3.00 86.10 1911.125 -3.0178.15 1916.965 -3.18 86.40 1910.911 -3.4678.35 1916.805 -3.26 86.60 1910.768 -3.6078.60 1916.605 -3.64 8690 1910.554 -3.4978.80 1916.445 -3.54 87.10 1910.411 -32879.00 1916.285 -3.44 87.30 1910.268 -2.897920 1916.125 -2.74 87.50 1910.125 -2.5379.40 1915.982 -3.08 87.80 1909.875 -30679.60 1915.839 -3.33 88.00 1909.708 -3.1579.80 1915.696 -3.39 88.20 J909.542 -3.2580.10 1915.482 -3.49 88.50 1909.292 -3.1280.35 1915.304 -331 88.70 1909.125 -2.8080.60 1915.125 -2.72 88.90 1908.943 -3.1680.85 1914.958 -336 89.10 1908.761 -3.5281.05 1914.825 -3.45 89.40 1908.489 -3.4381.25 1914.692 -3.52 89.60 . 1908.307 -3.2681.45 1914.558 -3.41 89.80 1908.125 -2.8781.70 1914.392 -3.15 90.00 1907.958 -3.1781.90 1914.258 -3.01 90.20 1907.792 -3.5082.10 1914.125 -2.99 90.40 1907.625 -3.498235 1913.933 -3.26 90.60 1907.458 -3.2382.55 1913.779 -3.49 90.80 1907.292 -3.1482.75 1913.625 -3.49 91.00 1907.125 -2.8082.95 1913.471 -3.54 91.20 1906.958 -2.9083.15 1913.317 -3.21 91.40 1906.792 -3.4083.40 1913.125 -2.79 91.60 1906.625 -3.3983.65 1912.946 -3.32 91.80 1906.458 -3.4783.90 1912.768 -3.33 92.00 1906.292 -3.4284.10 1912.625 -3.49 92.20 1906.125 -3.0584.35 1912.446 -329 92.40 1905.977 -3.0784.55 1912.304 -3.01 92.60 1905.829 -3.4384.80 1912.125 -2.86 92.80 1905.681 -3.52

97

Core depth Age (years) 8"0(cm) (%oPDB)

93.00 1905532 -3.5693.20 1905.384 -3.3693.40 1905.236 -3.2493.55 1905.125 -2.8593.75 1904.958 -28993.95 1904.792 -3.2694.05 1904.708 -3.6094.20 1904.583 -3.7294.40 1904.417 -3.5894.60 1904.250 -3.5294.75 1904.125 -30894.90 1903.852 -3.4495.10 1903.489 -3.4495.30 1903.125 -3.0895.55 1902.625 -3.759580 1902.125 -3.3996.00 1901.875 -3.4796.20 1901.625 -3.6696.40 1901.375 -3.1396.60 1901.l25 -2.8796.80 1900.817 -3.6696.95 1900.587 -3.6997.10 1900.356 -2.9497.25 1900.125 -2.8097.40 1899.958 -3.6697.65 1899.681 -3.5397.90 1899.403 -3.2498.15 1899.125 -2.7398.35 1898.914 -3.0898.50 1898.757 -33398.70 1898.546 -3.5998.90 1898.336 -3.1699.10 1898.125 -2.8399.30 1897.875 -3.4399.50 1897.625 -3.5299.70 1897375 -3.4299.90 1897.125 -2.82

8"0(%oPDB)

0.00 -3.650.15 -3.510.35 1993.125 -2.770.50 1992.938 -3.120.65 1992.750 -3.600.80 1992.563 -3.781.00 1992.313 -3.35l.l5 1992.125 -2.981.35 1991.935 -3.131.50 1991.792 -3.591.70 1991.601 -3.541.85 1991.458 -3.812.00 1991.315 -3.562.20 1991.125 -3.112.35 1990.958 -3.182.50 1990.792 -3.532.70 1990.569 -3,632.95 1990.292 -3.513.10 1990.125 -3.083.30 1989.971 -3.253.50 1989.817 -3.533.65 1989.702 -3.673.85 1989.548 -3.414.00 1989.433 -3.364.20 1989.279 -3.28

Core depth Age (years) 8 '0(cm) (%oPDB)

4.40 1989.125 -2.644.60 1988.903 -3.054.80 1988.681 -3354.95 1988.514 -3.585.20 1988.236 -3.785.30 1988.125 -3.005.55 1987.813 -3.245.70 1987.625 -3.885.90 1987.375 -3.676.10 1987.125 -3386.30 1986.925 -3.446.50 1986.725 -3.706.70 1986.525 -3.586.95 1986.275 -3.337.10 1986.125 -3.277.30 1985943 -3.597.45 1985.807 -3.797.60 1985.670 -3.367.80 1985.489 -3.278.00 1985.307 -3.178.20 1985.125 -2.948.35 1984.911 -3.398.50 1984.696 -3.558.70 1984.411 -3.66

Core depth Age (years) 81'0(cm) (%oPDB)

890 1984.125 -2.979.10 198'3.903 -3.429.25 1983.736 -3.679.45 1983.514 -3.799.65 1983.292 -3.559.80 1983.125 -2.989.95 1982.949 -3.4510.15 1982.713 -3.6610.30 1982.537 -3.6710.50 1982.301 -3.6010.65 1982.125 -3.1310.80 1981.982 -3.1811.00 1981.792 -3.631l.l0 1981.696 -3.2511.40 1981.411 -3.4811.50 1981.315 -3.0711.70 1981.125 -3.0411.90 1980.943 -3.601210 1980.761 -3.7912.20 1980.670 -3.6712.50 1980398 -3.5412.60 1980307 -3.15

12.80 1980.125 -3.0913.00 1979.903 -3.5313.20 1979.681 -3.6813.35 1979.514 -3.5013.50 1979.347 -3.2713.70 1979125 -2.9613.90 1978.958 -3.0714.10 1978.792 -3.1414.20 1978.708 -3.6514.40 1978.542 -3.5714.60 1978.375 -3.2714.90 1978.125 -2.9915.10 1977.958 -3.4015.30 1977.792 -3.5815.50 1977.625 -3.5715.70 1977.458 -3.6315.90 1977.292 -3.1916.10 1977.125 -3.0016.20 1977.042 -3.1416.50 1976.792 -3.4716.70 1976.625 -3.631690 1976.458 -3.4717.10 1976.292 ·3.2617.30 1976.125 -2.9617.50 1975.925 -3.5717.70 1975.725 -3.6917.90 1975.525 -3.5018.10 1975.325 -3.2418.30 1975.125 -3.0818.50 1974.943 -3.5318.70 1974.761 -3.5218.90 1974.580 -3.5719.10 1974.398 -3.2519.40 1974.125 -3.0019.60 1973.925 -3.2519.80 1973.725 -3.6820.00 1973.525 -3.5420.20 1973.325 -3.4020.40 1973.125 -2.8320.60 1972.971 -3.2820.80 1972.817 -3.5721.00 1972.663 -3.6821.30 1972.433 -3.5221.50 1972.279 -3.2021.70 1972.125 -3.0421.90 1971.925 -3.1822.10 1971.725 -3.4922.30 1971.525 -3.6222.50 1971.325 -3.2422.70 1971.125 -3.0322.80 1971.048 -3.1323.00 1970.894 -3.6023.20 1970.740 -3.7023.40 1970.587 -3.6423.60 1970.433 -3.5023.80 1970.279 -3.0824.00 1970.125 -3.0124.20 1969.925 -3.3624.40 1969.725 -3.5724.60 1969.525 -3.6024.80 1969.325 -3.4425.00 1969.125 -2.8225.20 1968925 -3.0025.40 1968.725 -3.4125.60 1968.525 -3.7725.80 1968.325 -3.1126.00 1968.125 -2.9926.20 1967.925 ~3.55

26.50 1967.625 -3.57

Core depth Age (years) 81'0(em) (%oPDB)

26.60 1967.525 -3.4426.80 1967.325 -3.392700 1967.125 -2.9227.20 1966.958 -3.1927.40 1966.792 -3.5727.60 1966.625 -3.6527.80 1966.458 -3.4928.00 1966.292 -3.1428.20 1966.125 -2.9328.40 1965.725 -3.2328.50 1965.525 -3.0928.70 1965.125 -2.9028.90 1964.943 -3.2029.05 1964.807 -3.5829.25 1964.625 -3.8329.40 1964.489 -3.5629.60 1964.307 -3.2129.80 1964.125 -2.7730.00 1963.958 -3.2730.20 1963.792 -3.7130.40 1963.625 -3.7530.60 1963.458 -3.7830.80 1963.292 -3.4431.00 1963.125 -3.1531.20 1962.996 -3.2831.40 1962.867 -3.7231.60 1962.738 -3.8531.80 1962.609 -3.6232.00 1962.480 -3.8132.15 1962.383 -3.5832.30 1962.286 -3.2432.55 1962.125 -3.0532.70 1961.975 -3.6232.90 1961.775 -3.7233.05 1961.625 -3.8033.25 1961.425 -3.6433.40 1961.275 -3.1333.55 1961.125 -2.6333.70 1960.995 -3.2333.90 1960821 -3.4334.10 1960.647 -3.9934.30 1960.473 -36034.50 1960.299 -3.3334.70 1960.125 -3.2034.90 1959.903 -3.3035.05 1959.736 -3.3935.20 1959.569 -3.5835.40 1959.347 -3.6235.60 1959.125 -3.4735.70 1958.958 -3.493590 1958.625 -3.7036.00 1958.458 -3.1536.20 1958.125 -3.0036.40 1957.839 -3.2936.50 1957.696 -3.6636.70 1957.411 -3.3036.90 1957.125 -2.8937.10 1956.839 -3.6137.20 1956.696 -3.6037.40 1956.411 -3.3537.60 1956.125 -3.3437.80 1955.792 -3.8538.00 1955.458 -3.6538.20 1955.125 -3.0838.40 1954.839 -3.5238.50 1954.696 -3.7838.70 1954.411 -3.6538.90 1954.125 -2.8639.10 1953.839 -3.63

98

Core depth Age (years) 8"0(em) (%oPDB)

39.30 1953.554 -3.6139.40 1953.411 -3.4139.60 1953.125 -3.1139.80 1952.903 -3.5240.00 1952.681 -37140.20 1952.458 -3.7040.40 1952.236 -3.3640.50 1952.125 -2.9340.70 1951.903 -3.4940.80 1951.792 -3.7241.00 1951.569 -3.9041.20 1951.347 -3.4741.40 1951.125 -3.0441.60 1950.943 -3.2041.80 1950.761 -3.7542.00 1950.580 -3.7342.15 1950.443 -3.5042.30 1950.307 -3.1742.50 1950.125 -2.9142.60 1950.014 -3.6542.80 1949.792 -3.7043.00 1949.569 -3.6943.20 1949.347 -3.4843.40 1949.125 -3.1843.50 1949.058 -3.2443.70 1948.925 -3.4043.90 1948.792 -3.5744.05 1948.692 -3.7644.30 1948.525 -3.7144.50 1948.392 -3.2944.70 1948.258 -3.1144.90 1948.125 -2.9545.10 1948.000 -2.9845.40 1947.813 -3.2345.55 1947.719 -3.5645.75 1947.594 -3.2745.90 1947.500 -3.2346.10 1947.375 -2.8446.30 1947.250 -3.1346.50 1947.125 -2.8546.70 1946.943 -3.1146.90 1946.761 -3.4047.05 1946.625 -3.5347.20 1946.489 -3.1847.40 1946.307 -3.2347.60 1946.125 -2.9847.80 1945.958 -3.4748.00 1945.792 -3.4948.10 1945.708 -3.5948.30 1945.542 -3.5548.50 1945.375 -3.3748.60 1945.292 -3.0848.80 1945.125 -2.964900 1944.971 -3.4049.10 1944.894 -3.6749.30 1944.740 -3.6449.50 1944.587 -3.9749.70 1944.433 -35349.90 1944.279 -3.3050.10 1944.125 -2.9650.40 1943.825 -3.3250.50 1943.725 -3.3450.70 1943.525 -3.6950.90 1943.325 -3.3251.10 1943.125 -2.5451.30 -3.0551.50 -3.49

5.2.3 Growth rate

Year Growth rate Growth rate Year Growth rate Growth rate(cl1l/yr) (cl1l/yr) (cm/yr) (cl11/yr)ßl[~\!~ 1~1:~~~~ ßw~~~ R~l~~~fi

1898 0.50 1930 0951899 0.50 1931 0,951900 0.50 1932 0.951901 0.55 1933 0,951902 0.65 1934 0.951903 0.65 1935 1.001904 0.70 1936 1.001905 0.70 1937 1.001906 0.70 1938 1.001907 0.70 1939 1.001908 0.70 1940 1.001909 0.75 1941 1.001910 0.75 1942 1.051911 0.75 1943 1.051912 0.80 1944 1.05 0.601913 0.80 1945 1.05 0.601914 0.80 1946 1.05 0.701915 0.80 1947 1.05 0.701916 0.80 1948 1.05 0.701917 0.80 1949 1.05 0.701918 0.85 1950 1.05 0.701919 0.85 1951 1.05 0.801920 0.85 1952 1.10 0.801921 0.85 1953 1.10 0.851922 0.85 1954 1.10 0.901923 0.90 1955 1.10 0.901924 0.90 1956 1.15 0.901925 0.90 1957 1.15 0.901926 0.90 1958 1.15 0.901927 090 1959 1.15 0.901928 0.90 1960 1.15 0901929 0.95 1961 1.20 0.90

99

Year Growth rate Growth rate(cm/yr) (cl1l/yr)ii~E{{ii ßI~~j

1962 1.20 1.001963 1.20 1.001964 1.20 1.001965 1.20 1.001966 1.20 1.001967 1.25 1.001968 1.25 1.001969 1.25 1.001970 1.25 1.001971 1.30 1.051972 1.30 1.051973 1.30 1.101974 1.30 1.101975 1.30 1.101976 1.30 1.101977 1.30 1.101978 1.30 1.101979 1.30 1.151980 1.35 1.201981 1.35 1.201982 1.35 1.201983 1.40 1.201984 1.40 1.201985 1.40 1.201986 1.40 1.301987 1.40 1.301988 1.40 1.301989 1.45 1,301990 1.45 1.501991 1.45 1.551992 1.50 1.60

5.2.4 Seasonal amplitude

100

1898 0.611899 0.631900 0.671901 0.591902 0.321903 0.301904 0.591905 0.691906 0.351907 0.611908 0.521909 0.351910 0.891911 0.481912 0.461913 0.591914 0.381915 0.561916 0.791917 0.561918 0.591919 0.241920 0.311921 0.451922 0.451923 0.611924 0.441925 0.671926 0.931927 0.691928 0.371929 0.35

Year Seas. al11p.(%oPDB)

B:liJ:~g~3

Seas. Al11p.(%oPDB)

t\M~2;{g

Year Seas. a111p. Seas. a111p.(%oPDB) (%oPDB)

ßtl~fl~) t\~][~§

1930 0.691931 0.421932 0.661933 0.271934 0.631935 0.571936 0.741937 0.521938 0651939 0.601940 0.601941 0.581942 0.391943 0.441944 0.40 0.941945 0.82 0.481946 0.35 0.321947 0.45 0.351948 0.47 0.761949 0.35 0.481950 0.75 0.541951 0.50 0.811952 0.82 0.671953 0.53 0.421954 0.75 0.761955 0.66 0.581956 0.36 0.201957 0.31 0.521958 0.45 0.521959 0.46 0.291960 0.85 0.631961 0.61 0.95

Year Seas. amp. Seas. amp.(%oPDB) (%oPDB)

!RJiI$lg~ !R~~~~p

1962 0.60 0.441963 0.66 0.561964 0.62 0.851965 0.60 0.241966 0.30 0.641967 0.50 0.531968 0.46 0.561969 0.48 0.731970 0.38 0.561971 0.43 0.501972 0.46 0.551973 0.48 0.631974 0.60 0.511975 0.62 0.371976 0.62 0.511977 0.72 0.511978 0.85 0.501979 0.78 0.581980 0.56 0.451981 0.54 0.291982 0.51 0.521983 0.63 0.661984 0.56 0.541985 0.50 0.291986 0.57 0.321987 0.49 0.461988 0.60 0.461989 0.34 0.751990 0.46 0.501991 0.54 0.451992 0.34 0.75

101

5.3. Fossil and recent molluscs (Tridacna spp., Northern Gulf of Aqaba)

5.3.1 High-resolution sampies

TO-1TO-2TO-3TO-4TO-5TO-6TO-7TO-8TO-9

TO-10TO-11TO-12

0.000.030.060.090.120.150.180.210.240.270.300.33

1.171.251.331421.621.641.691.711.741.391.150.87

Sampie Oepth 8 '0Illunber (cm) (%0 POß)

TO-13 0.36 0.88TO-14 0.39 0.63TO-15 042 0.65TO-16 045 0.91TO-17 048 1.14TO-18 0.51 1.51TO-19 0,53 1,72TO-20a 0.57 1.29TO-20 0.60 1.70TO-21 0.63 1.51TO-22 0,66 1.14TO-23 0.69 0.58

SampIe Oepth 8"0Illunber (cm) (%0 POß)

TO-24 0.72 0.57TO-25 0.75 0.72TO-26 0.78 0.92TO-27 0.81 1.17TO-28 0.84 147TO-29 0,87 1.71TO-30 0.90 1.64TO-31 0.93 1.21TO-32 0.96 0.92TO-33 0.99 0.89TO-34 1.02 0.80TO-35 1.05 0.65

81 0Illunber (%0 POß)

T1-1 0.00 1.27T1-2 0.03 1.79T1-3 0.06 1.59T1-4 0.09 0.86T1-5 0.12 1.38T1-6 0.15 1.73T1-7 0.18 1.56T1-8 0.21 1.87T1-9 0.24 1.14

T1-10 0.27 0.63T1-11 0.30 0.70T1-12 0.33 1.23T1-13 0.36 1.98T1-14 0.39 1.13T1-15 042 0.67T1-16 045 0.55T1-17 0.48 1.01

Sampie Oepth 8 '0munber (cm) (%0 POß)

T1-18 0.51 1.88T1-19 0.54 1.31T1-20 0.57 0.98T1-21 0.60 0.91T1-22 0.63 1.15T1-23 0.66 2.08T1-24 0.69 1,77T1-25 0.72 1.17T1-26 0.75 0.65T1-27 0.78 0,62T1-28 0.81 0.50T1-29 0.84 0.67T1-30 0.87 0.97T1-31 0.90 1.35T1-32 093 1.69T1-33 0.96 1.66T1-34 0.99 1.84

Sampie Oepth 8'0munber (cm) (%0 POß)

T1-35 102 1.64T1-36 1.05 1.29T1,37 1.08 1.08T1-38 1.11 0.91T1-39 1.14 0.79T1-40 1.17 0.62T1-41 1.20 1.01T1-42 1.24 145T1-43 1.27 1.82T1-44 1.30 2.06T1-45 1.33 1.97T1-46 1.36 1,57T1-47 1.39 1.35T1-48 142 0.79T1-49 145 0.63T1-50 147 047

Oepth 81'0(cm) (%0 POß)

T3-1T3-2T3-3T3-4T3-5T3-5

T13-1T13-2T13-3T13-4T13-5T13-6

0.050.090.140.180.230.23

0.140.210.280.350.420.49

1.861471441.021.981.98

81"0(%0 POß)

2.031.691.661.912482.45

Sampie Oepth 81'01111mber (cm) (%0 POß)

T3-6 0.27 149T3-7 0.32 0,70T3-8 0.36 1.81T3-9 041 2.09T3-10 045 149T3-11 0.50 1.52

Sampie Oepth 81KO1111mber (cm) (%0 POß)

T13-7 0.56 148T13-8 0.63 1.57T13-9 0.70 245

T13-10 0.77 1.70T13-11 0.84 1.97

Sampie1111mber

T3-12T3-13

0.540.59

1.520.76

5.3.1.2 Bulk sampies

T8b-1 1.72T8b-2 1.62T8b-3 1.84T8b-4 1.70

T10b-1 2.74T10b-2 2.57T10b-3 2.65T10b-4 2.70T10b-5 2.69

Sampie 0 (%0)

T19b-1 2.20T19b-2 2.37

Sampie ßTI'0 (%0)

T19b-3 2.41T19b-4 2.18T19b-5 2.28

Sampie 8"0 (%0)

T18b-1 1.99T18b-2 2.07T18b-3 2.01T18b-4 2.08T18b-5 2.08

Sampie öTifQ(%o)

T4b-1 1.30T4b-2 1.42T4b-3 1.41T4b-4 1.25T4b-5 1.57

'Sall1pTe 0'0 (%0)

T7b-1 1.02T7b-2 1.17T7b-3 1.18T7b-4 1.38T7b-5 1.44T7b-6 1.56T7b-7 1.48T7b-8 1.27T7b-9 1.28T7b-10 1.11T7b-11 1.27T7b-12 1.23T7b-13 1.57T7b-14 1.00T7b-15 1.51T7b-16 1.52T7b-17 1.16T7b-18 1.23T7b-19 1.31T7b-20 1.19T7b-21 1.32

102

Sampie 0 0(%0)

T7b-22 1.41T7b-23 1.28T7b-24 1.32T7b-25 1.25T7b-26 1.35T7b-27 1.37T7b-28 1.39T7b-29 1.11T7b-30 1.24T7b-31 1.14T7b-32 1.17T7b-33 1.13T7b-34 1.37T7b-35 1.35T7b-36 1.33T7b-37 1.29T7b-38 1.31T7b-39 1.32T7b-40 1.28