|dc.description.abstract||The paleo-environments of South-West Germany during Marine Isotope Stage 3 (MIS 3) and Marine Isotope Stage 2 (MIS 2), from approximately 50 – 23 thousand years ago, were reconstructed at both millennial and sub-annual scale using temporally successive oxygen (δ18O ) and carbon (δ13C) isotopes in the tooth enamel of woolly mammoth (Mammuthus primigenius). The mammoth teeth were discovered in the sediments of the Upper Rhine Graben, and their ages were determined to roughly fit into three time windows: 50ka (Early MIS 3), 40ka (Middle MIS 3) and 23ka (Last Glacial Maximum, LGM) respectively. The δ18O values were used to infer the isotopic compositions of mammoth water source, which in turn tracks local precipitation and air temperature conditions. The δ13C values were analyzed to deduce the mammoth dietary behaviour as well as reconstruction of paleo-vegetation. The approach we used for high-resolution reconstruction was developed during this research, with few prior studies at this scale. This new approach successfully reconstructed the paleo-environments of SW Germany and the paleo-ecology of mammoths, and we obtained the following results:
The oxygen isotopic analysis at millennial scale indicated the trend of 18O depletion in the mammoth water source from MIS 3 to the LGM, with the mean δ18O values ranging from -9.33‰ to -7.12‰. We estimated the mean annual temperature during the LGM from the δ18O values, and it was around 4-10℃ lower than present. Our results were compared to other studies which involved paleo-environmental reconstructions with mammoth remains. We found that, during the 40ka time window, mammoth in SW Germany had 1.1-4.7‰ higher δ18O values compared to those in Sweden, Estonia, Lithuania and Latvia. During the LGM, however, the δ18O values in these regions were similar, while Germany was around 2.2‰ higher than Hungary.
Records of oxygen isotopic oscillations at sub-annual scale were also obtained. The isotopic oscillations are in regular cycles with regional peaks and troughs representing summer and winter seasons respectively. The δ18O values during summers were approximately 9.3 – 14.6 ‰ higher than those during winters. Seasonal differences, indicated by the difference between mean δ18O values of summer and winter, were most enhanced during around 50ka, while LGM seasonality was intermediate between those in 50ka and 40ka intervals.
We also analyzed the paleo-ecology of M. primigenius on their diets and tooth growth. The mean δ13C values for all of our mammoths were around -12.0‰ and -12.5‰, indicating that they primarily consumed C3 plants. This matched with the studies from mammoth feces, as well as pollen records, which indicated that Germany was dominated by semi-desert steppe landscapes from 50 – 23ka. There was a trend towards more positive δ13C values from MIS 3 to the LGM, which may be caused by the C4 plant expansion due to reduced aridity or reduced water availability in a C3 dominated landscape. Mammoth enamel growth rates were also estimated using the seasonal variation of stable isotopes. The mean growth rate was estimated to be around 15mm/year, with a general trend of decreasing speed of growth through time.
Finally, our study has shown the utility of using high resolution stable isotope analyses for reconstructing past environmental and palaeo-ecological changes from mammoth teeth. This method has great potential for expanding our knowledge of both seasonal and millennial scale paleo-environmental changes across the geographic range where mammoth remains are found.||en_US