The Holocene paleolimnology of Lake Issyk-Kul, Kyrgyzstan: trace element and stable isotope composition of ostracodes

https://doi.org/10.1016/S0031-0182(01)00339-XGet rights and content

Abstract

The stable isotope and trace element geochemistry of ostracode shells from Lake Issyk-Kul, Kyrgyzstan, provide an important new paleoclimate record for central Asia. The δ18O, δ13C, Sr/Ca, Mg/Ca and U/Ca composition of shells from the species Candona neglecta found in two piston cores constrains the paleohydrology of the lake from 8700 to 1000 calendar yr before present. These data, combined with robust chronologies based on 16 AMS 14C dates, suggest that the region went from a relatively moist climate in the early Holocene to a more arid climate in the late Holocene. Intense freshwater input, presumably glacial runoff, flowed into the lake from 8700 to 8300 cal yr BP as evident from the δ18O and δ13C values. From 8300 to 6900 cal yr BP δ18O and Sr/Ca values indicate that the lake was relatively fresh and an open basin, and U/Ca values suggest that the lake was well-mixed with well-oxygenated bottom waters. There is a transition zone between 6900 and 4900 cal yr BP when δ18O and Sr/Ca values rapidly increase and Mg/Ca and U/Ca values rapidly decrease. During this period the lake evolved from an open, freshwater, well-mixed lake to a closed, more saline, less well-mixed lake. For the remainder of the record δ18O, Sr/Ca, Mg/Ca and U/Ca are relatively constant, implying conditions similar to modern conditions, although the slight increasing trend in Sr/Ca may suggest an increase in salinity over this period. The δ13C values were approximately in equilibrium with atmospheric CO2 in the early part of the record but became relatively negative in the later part of the record. This indicates that the dominant mechanism controlling the isotopic composition of dissolved inorganic carbon shifted from exchange with the atmosphere to variations in biological productivity.

Introduction

Paleoclimate records from central Asia have not been fully exploited even though the potential for sensitive climate records in this region is high. Two major pressure cells, the Siberian High to the north and the Southwest Asian Low to the southwest, interact over the region and changes in their intensity or location would strongly affect central Asian climate. These changes should be preserved in paleoclimate proxy records. Previous work (e.g. Lake Baikal: Colman et al., 1995; Lake Qinghai: Lister et al., 1991; Lake Manas: Rhodes et al., 1996; central China Loess deposits: Kukla and An, 1989; Lakes Sumxi and Bangong: Gasse et al., 1996) indicate that while there are some similarities in the paleoclimate records from the region, there are also differences in the perceived patterns of climate change throughout central Asia.

The current study is an analysis of sediments from Lake Issyk-Kul, a deep lake with a potentially long and continuous record of paleoclimate change. Its location in the heart of the Tien Shan mountains offers a unique opportunity to study the climate history of a region in central Asia that is dominated by a continental climate regime and is relatively far from the monsoons to the south and southeast.

Lake Issyk-Kul is located in Kyrgyzstan, a former Soviet Republic (Fig. 1), and is the 11th largest lake in the world by volume and the 5th deepest (Herdendorf, 1990). It has a volume of 1730 km3, a surface area of 6247 km2, and a maximum depth of 668 m (Fig. 2). Issyk-Kul is set in a compressional basin (Klerkx et al., 1999) between two mountain ranges, the Kungey Alatoo to the north and the Terskey Alatoo to the south, both of which are part of the Tien Shan. The depression that the lake fills has existed since the Miocene, although the initial formation of the Tien Shan dates from the Early Paleozoic (Grosswald et al., 1994).

Atmospheric circulation over Lake Issyk-Kul is controlled by the Siberian High Pressure Cell and the Southwest Indian Low. During winter months, the southwest branch of the Siberian High extends south into the Tien Shan, bringing cold, dry air from Mongolia into the Issyk-Kul basin. During the summer months, frontal cyclonic circulation develops when cold air masses from the north occlude southern cyclones from the southwest, generating precipitation. This circulation pattern brings maximum, albeit minor, precipitation to the basin during the summer months and driest conditions during the winter months (Dando, 1987, Aizen et al., 1995).

The lake is in a semi-arid setting and local meteorological data suggest that moisture is mostly derived from west of the basin (Romanovsky, 1990, Aizen et al., 1995). A strong gradient of rainfall exists from west to east across the lake: the western shore of the lake near Balykchy receives 120 mm/yr of rainfall, while the eastern shore near Karakol receives 720 mm/yr (Romanovsky, 1990). The surrounding mountain ranges contain alpine glaciers that extended nearly to the present lakeshore during the last glaciation (Grosswald et al., 1994). The sources of freshwater for the major rivers that flow into the lake are rainfall, snowmelt, groundwater, and glacial melt (Aizen et al., 1995).

The seasonal air temperature range in the region (−23 to 41°C) is reduced near the lake (−2 to 18°C; Romanovsky, 1990). Issyk-Kul never freezes due to its brackish salinity (6‰) and high-altitude insolation (Romanovsky, 1990). Surface water temperatures range from 4.4°C in winter to 21.4°C in summer, and there is a permanent thermocline located above 100 m (Romanovsky, 1990; E. Ralph, personal communication). The lake is highly oligotrophic; transparency measured by Secchi disk depths in mid-lake typically fall in the range of 13–22 m (Romanovsky, 1990). Convective winter turnover of the lake water occurs every 2 out of 3 yr, and oxygen concentrations range from 6.8 to 9.6 mg O2/l (Romanovsky, 1990). Oxygen concentration decreases with depth, with the deep bottom waters (600 m) at about 70% saturation (E. Ralph and M. Vollmer, personal communication).

Geologic maps of the basin indicate that granitic, metamorphic, folded sedimentary rocks, and glacial debris dominate the watershed and contribute to the composition of the lake. Limestone and marble are uncommon. Lake Issyk-Kul waters have a pH of about 8.6–8.9 and are oversaturated with respect to calcite (Alekin, 1946). They are dominated, in order of abundance, by the cations Na+, Mg2+, and Ca2+ and the anions SO42−, Cl, HCO3, and CO32− (Romanovsky, 1990).

The lake is currently a closed basin lake and would overflow at an elevation of 1620 m, 13 m above the current lake level. During overflow lake waters would extend westward into the Boom Gorge and flow down the Chui River (Fig. 2). The Chui River flows to the west and is ultimately a part of the internal drainage of the Aral–Caspian hydrographic region (Aizen et al., 1997).

Fluctuations in Issyk-Kul’s level, resulting from variable evaporation and precipitation in the basin, are common in the recent and ancient past. Instrumental records indicate that lake level has declined about 3 m since 1926. While Soviet era hydrological projects caused some of this decline, natural processes such as increased evaporation caused by increased temperatures generated part of the decline (Romanovsky, 1990). Preceding the 20th century fall the lake had been documented at just below the outflow level in 1856 AD by the direct observations of Semenov (Semenov, 1858). According to maps of Jesuit missionaries Issyk-Kul may have surpassed the outflow level and become open as early as 1755 AD (Romanovsky, 1990). Remnants of orchard trees, soils and settlements 3–6 m below the current lake level have been dated to the 13th through 15th centuries (Schnitnikov, 1980); lacustrine macrophytes elevated 7.7 m above present lake level have been recently dated at 580±40 14C yr BP, suggesting a rapid mid-14th century rise (Rasmussen et al., 2000). Radiocarbon dating of lacustrine macrophytes from 1623 m elevation indicates that there was another highstand, with overflow, between 1200 and 1400 yr BP (Aleshinskaya et al., 1971, Trofimov, 1978, Berdovskaya and Egorov, 1986).

Pleistocene terraces indicate ancient lake levels much higher than those of the Holocene. A late Pleistocene erosive terrace, with an estimated date of occurrence of 26 ka to 10 ka (based on radiocarbon dates of associated lacustrine sediment) is found 33 m above present lake level (Aleshinskaya et al., 1971). An early or middle Pleistocene wavecut terrace is observed 68 m above the present lake level (Trofimov and Grigina, 1979). Depth profiles of the lake reveal a network of submerged relic river valleys (western and eastern bays, Fig. 2) suggesting a Late Quaternary lake level about 110 m below the present surface (Trofimov, 1978).

Surface sediments in the lake consist of terrigenous muds with abundant carbonates (Alekin, 1946, Korotaev, 1967). Beach-zone lithification of nearshore granitic materials via inorganic calcite and organic monohydrocalcite precipitation forms semi-continuous lake-margin pavements and microbialites (Rasmussen and Romanovsky, 1995, Rasmussen et al., 1996). Deep-basin sediments consist of fine-grained terrigenous muds and silts along with endogenic calcite and ostracodes, as well as material exported from the shallow shelves (Rasmussen et al., 1998, Rasmussen et al., 1999).

The existence of large fluctuations in lake level in the basin’s past, and of abundant organic and inorganic carbonate makes Lake Issyk-Kul an excellent target for paleoclimate reconstruction. Inorganic and organic carbonates (ostracodes) have been used in numerous lake basins to study past climate (e.g. Stuiver, 1970, Johnson et al., 1991, Curtis and Hodell, 1993). This work focuses on sediment from two cores, IK97-10P and IK97-11P, collected from the 240 meter-deep western portion of the basin in 1997 (Fig. 2). These two cores were the longest of 11 recovered from the lake and were relatively undisturbed by turbidites, indicating that they may contain a long and continuous paleoclimate record. This paper will describe the trace element and stable isotope geochemistry of ostracodes from these two cores, while another paper will concentrate on the sedimentology of the core materials (Rasmussen et al.,in preparation).

Section snippets

Methods

Piston cores were collected from the lake in 1997 using a Kullenberg corer (Kelts et al., 1986). Site locations were determined by the satellite global positioning system (GPS). Cores were kept in plastic liners, capped, and shipped back to the US for analysis. Once in the US, the cores were scanned for magnetic susceptibility, split, visually described (Fig. 3), photographed and subsampled for sedimentological and geochemical analyses. Wedges of sediment 3 cm thick were taken at about 10-cm

Geochronology

Sixteen AMS radiocarbon dates were obtained for cores 10P and 11P from the Woods Hole National Ocean Sciences AMS Facility (Table 1; Fig. 4). Six to ten milligrams of clean ostracode shells larger than 150 μm were dated to avoid errors associated with using bulk sediment samples or smaller ostracode shells (e.g. Lake Turkana: Halfman et al., 1994). Ages for the sediment range from 1480±45 to 8940±65 radiocarbon yr BP. All dates were corrected for the δ13C composition of the shells. The

Water

The δ18O and δD compositions of the river waters are probably a good approximation of local rainfall composition because of the relatively short distance the waters travel to flow into the lake. Their δ18O and δD compositions define a local meteoric water line (LMWL, Fig. 6) distinct from both the global meteoric water line (GMWL, Fig. 5) and the weighted mean δ18O and δD compositions of rainfall at the closest International Atomic Energy Agency/World Meteorological Organization (IAEA/WMO)

High freshwater input 8700 to 8300 cal yr BP

At the bottom of both cores δ18O values increase from the lowest values found in either core, −0.8‰, to values of +0.5‰. The δ13C values also increase over this period, although the change is not as strong. There are not any obvious trends in the trace element data, although high-frequency fluctuations appear to be more prevalent in this section of the cores than elsewhere. The low δ18O and δ13C composition of the ostracodes indicates that freshwater input to the basin was particularly strong

Conclusion

The trace element and stable isotope composition of ostracode shells from two cores from Lake Issyk-Kul give insights into the paleolimnology of the basin.

δ18O values and Sr/Ca molar ratios indicate that the lake went from a relatively freshwater open-basin system between 8700 and 6900 cal yr BP to a closed-basin, more saline system for most of the time from 6900 cal yr BP to the present.

δ13C values suggest that when the basin was open and well mixed, the bottom water DIC was closer to

Acknowledgements

We would like to thank Emi Ito and Reed McEwan (Department of Geology and Geophysics, UMN-TC) for help with the stable isotope analyses and Jason Agnich (LLO) for help with the trace element analyses. Yvonne Chan and Lindsey Schoenbohm (LLO) processed the samples for the AMS dates. The crew of the R/V Multor was indispensable. Elise Ralph, Sylvia Barry, Steve Dougherty, T.C.J., K.A.R., and V.V.R. collected the piston cores. Elise Ralph, Sylvia Barry and Steve Dougherty collected the lake water

References (85)

  • T.-L Ku et al.

    Uranium in the open ocean: concentration and isotopic composition

    Deep-Sea Res.

    (1977)
  • G Kukla et al.

    Loess stratigraphy in Central China

    Palaeogeogr. Palaeoclimatol. Palaeoecol.

    (1989)
  • D Langmuir

    Uranium solution-mineral equilibria at low temperatures with applications to sedimentary ore deposits

    Geochim. Cosmochim. Acta

    (1978)
  • H.-C Li et al.

    δ13C-δ18O covariance as a paleohydrological indicator for closed-basin lakes

    Palaeogeogr. Palaeoclimatol. Palaeoecol.

    (1997)
  • G.S Lister et al.

    Lake Qinghai, China: closed-basin lake levels and the oxygen isotope record for ostracoda since the latest Pleistocene

    Palaeogeogr. Palaeoclimatol. Palaeoecol.

    (1991)
  • A Mucci et al.

    The solubility of calcite and aragonite in sulfate free seawater and the seeded growth kinetics and composition of the precipitates at 25°C

    Chem. Geol.

    (1989)
  • T.E Rhodes et al.

    A Late Pleistocene–Holocene lacustrine record from Lake Manas, Zunggar (northern Xinjiang, western China)

    Palaeogeogr. Palaeoclimatol. Palaeoecol.

    (1996)
  • R.D Ricketts et al.

    Climate change in the Turkana basin as deduced from a 4000 year long δ18O record

    Earth Planet. Sci. Lett.

    (1996)
  • J.D Rimstidt et al.

    Distribution of trace elements between carbonate minerals and aqueous solutions

    Geochim. Cosmochim. Acta

    (1998)
  • A Schwalb et al.

    Holocene environments from stable isotope stratigraphy of ostracodes and authigenic carbonate in Chilean Altiplano Lakes

    Palaeogeogr. Palaeoclimatol. Palaeoecol.

    (1999)
  • T.J Shaw et al.

    Redox dynamics in the Chesapeake Bay: The effect on sediment/water uranium exchange

    Geochim. Cosmochim. Acta

    (1994)
  • M.R Talbot

    A review of the palaeohydrological interpretation of carbon and oxygen isotopic ratios in primary lacustrine carbonates

    Chem. Geol.

    (1990)
  • U Von Grafenstein et al.

    Oxygen and carbon isotopes in modern fresh-water ostracod valves: assessing vital offsets and autecological effects of interest for paleoclimate studies

    Palaeogeogr. Palaeoclimatol. Palaeoecol.

    (1999)
  • G Wansard et al.

    Variability in ostracod partition coefficients D(Sr) and D(Mg): Implications for lacustrine palaeoenvironmental reconstructions

    Chem. Geol.

    (1998)
  • J Xia et al.

    Geochemistry of ostracode calcite: Part 2. The effects of water chemistry and seasonal temperature variation on Candona rawsoni

    Geochim. Cosmochim. Acta

    (1997)
  • J Xia et al.

    Geochemistry of ostracode calcite: Part 1. An experimental determination of oxygen isotope fractionation

    Geochim. Cosmochim. Acta

    (1997)
  • V.B Aizen et al.

    Climate, snow cover, glaciers, and runoff in the Tien Shan, Central Asia

    Water Resour. Bull.

    (1995)
  • V.B Aizen et al.

    Climate and hydrologic changes in the Tien Shan, Central Asia

    J. Clim.

    (1997)
  • Alekin, O.A., 1946. Physico-chemical peculiarities of the water of Issyk-Kul Lake (in Russian). Hydrometeorological...
  • Aleshinskaya, Z.V., Bondarev, L.G., Voskoresenskaya, T.N., Leflat, O.N., 1971. Cross-section of the newest deposits in...
  • L Araguás-Araguás et al.

    Stable isotope composition of precipitation over southeast Asia

    J. Geophy. Res.

    (1998)
  • L.V Benson et al.

    Stable isotopes of oxygen and hydrogen in the Truckee River-Pyramid Lake surface-water system. 3. Source of water vapor overlying Pyramid Lake

    Limnol. Oceanogr.

    (1994)
  • Berdovskaya, G.N., Egorov, A.N., 1986. Reconstruction of natural conditions between the Pleistocene and Holocene (in...
  • V.S Brezgunov et al.

    Investigation of the origin and dynamics of waters of large lake basins by means of natural isotopes

    Water Resour. Consult. Bur. N. Y.

    (1980)
  • Bronshtein, Z.S., 1947. Fresh-water Ostracoda, Academy of Sciences of the USSR Publishers, Moscow, published in English...
  • Chivas, A.R., De Deckker, P., Shelley, J.M.G., 1983. Magnesium, strontium, and barium partitioning in nonmarine...
  • Chivas, A.R., De Deckker, P., Cali, J.A., Chapman, A., Kiss, E., Shelley, J.M.G., 1993. Coupled stable-isotope and...
  • A.R Chivas et al.

    Strontium content of ostracodes indicates lacustrine palaeosalinity

    Nature

    (1985)
  • A.R Chivas et al.

    Magnesium and strontium in non-marine ostracode shells as indicators of palaeosalinity and palaeotemperature

    Hydrobiologia

    (1986)
  • S.M Colman et al.

    Continental climate response to orbital forcing from biogenic silica records in Lake Baikal

    Nature

    (1995)
  • H Craig

    Isotopic variations in meteoric waters

    Science

    (1961)
  • Craig, H., Gordon, L.I., 1965. Isotopic Oceanography: Deuterium and Oxygen 18 variations in the Ocean and the Marine...
  • Cited by (0)

    View full text