Molecular Distribution And C-H Isotopes Of N-alkanes Of Lacustrine Sediments Since The Last Deglaciation In The Northeast Tibet Plateau

To investigate issues concerning the utilization of molecular distribution and compound-specific C-H isotopes of n-alkanes in past lacustrine environmental research, we have analyzed 110 lake sediment samples since the last deglaciation from Hurleg Lake and Genggahai Lake located in the northeast Tibet Plateau respectivley. n-alkanes were extracted, separated and GC analyzed as well as GC-IRMS to study compound-specific C-H isotopes.For both Hurleg Lake and Genggahai Lake, carbon number of n-alkanes of most samples distribute in nC14～nC35 with nC22～nC33 show relatively high abundance with obvious odd-over-even preference. There are two molecular distribution models of n-alkanes extracted from both Hurleg and Genggahai Lake with one peak or two peaks. Relative to Hurleg Lake, the content of n-alkanes derived from aquatic plants is greater in Genggahai Lake. In addition, the variations along time series of CPI26-34,ACL27-33,Paq indeies of n-alkanes extracted from these two lake is different obviously, that may indicate the n-alkanes in Hurleg Lake is derived from terrestrial higher plants and aquatic plants, but the n-alkanes in Genggahai Lake mainly derived from a single source.TheδD difference of n-alkanes between aquatic and terrestrial plants has been proposed is a proxy of terrestrial evapotranspiration of the lake environment or a proxy of ratio of lake evaporation to input. However, the applications of this proposed proxy to some lakes were not successfully achieved as expected. For arid and semi-arid regions, either terrestrial evapotranspiration or ratio of lake evaporation to input is primarily controlled by regional aridity extent. OurδD data from Hurleg Lake and Genggahai Lake indicate thatΔD (i.e.,δD difference between mid-chain and long-chain n-alkanes) show completely different trends among these two lakes. Comparatively,ΔD from Hurleg Lake seems to be more reliable, indicating relatively dry climatic conditions during last glacial and relatively wet climatic conditions during Holocene. For Hurleg Lake, further analysis show positive correlations ofδD between nC23 and nC25, nC27 and nC29, nC29 and nC31 n-alkanes, however, no apparent 8D correlation between nC25 and nC27. It implies that mid-chain (nC23 & nC25) and long-chain (nC27, nC29 and nC31) n-alkanes in sediments of Hurleg Lake are indeed derived from two different plant groups, i.e. aquatic and terrestrial higher plants. On the other hand, data from Genggahai Lake show positiveδD correlations between nC23 and nC25, nC25 and nC27, nC27 and nC29, nC29 and nC31 n-alkanes, which indicate that n-alkanes from aquatic plant and terrestrial higher plant origins couldn't be discriminated by chain length of n-alkanes. These results might explain differentΔD trends between these two lakes since last deglaciation. Therefore, our results suggest that it is essential to know the real sources of n-alkanes of different chain length for solid and reliable paleohydrology, paleoclimate and paleoenvironment reconstruction with the proposedΔD indicator.Compound-specific carbon isotopes of n-alkanes from Hurleg Lake show positive correlations between nC23 and nC25, nC25 and nC27, nC27 and nC29, nC29 and nC31. SinceδD data in sediments of Hurleg Lake already indicate that mid-chain n-alkanes are from aquatic plants and long-chain ones are from terrestrial higher plants, water stress should not be the primary factor affectingδ13C of aquatic plants. Thus, variations ofδ13C values of nC23 and nC25 n-alkanes from Hurleg Lake sediments probably responded to temperature variations mainly. Andδ13C values of nC27, nC29 and nC31 derived from terrestrial higher plants indicate increasing relative abundance of C4 plants during the Holocene relative to the last glacial, And temperature should be responsible for increasing relative abundance of C4 plants during the last glacial-Holocene transition. Further comparison of both weigh-averagedδ13C of mid-chain nC23 and nC25 n-alkanes and that of long-chain nC27, nC29 and nC31 with the ice raft debris records from the subpolar North Atlantic shows good consistence. So,δ31C of n-alkanes from Hurleg Lake might be indicators of temperature variation. Albeit problems of chronologies and data resolution making detailed comparison between Hurleg Lake and Genggahai Lake is difficult, overall good correlation has been observed betweenδ13C data from these two lakes.Based on data of molecular distribution and C-H isotopes of n-alkanes from Hurleg Lake, paleoclimate and paleoenvironment since last deglacial of the study area has been preliminary reconstructed.13ka B.P.-9ka B.P., all proxies implied cold and dry climatic conditions; During 9kaB.P.-4.5kaB.P., warm and wet climatic conditions and large oscillation are suggested; During 4.5kaB.P.-2kaB.P., all proxies implyed cold and wet and stable climatic conditions; After 2ka B.P., warm-dry climate trend but still favorable compared to that before 9ka B.P..