Deuterium excess of Antarctic snow and precipitation origin

Investigation and a thorough understanding of the climate of the past are necessary to assess the present or future possible climate changes. Valuable information about the former climate is stored in ice cores from the large ice sheets of Greenland and Antarctica. In particular, the ratio of stable isotopes (different types of water molecules) of snow, 18-O and deuterium, are related to air temperature and are thus used for the climatic interpretation of the ice cores. However, the isotope content does not depend on the temperature alone; other factors, such as seasonality and origin of precipitation are also important. To resolve this problem, a variable combining 18-O and deuterium information and known as deuterium excess, d, is used to study the source areas of precipitation. The value of d is mainly determined by sea surface temperature, relative humidity, and wind speed in the source area. By testing under which assumptions for the prevailing conditions in the source area the d values found in snow can be reproduced by simple isotope models, information about the source area can be obtained. The range of possible values for these variables is surprisingly small. Most of the studies devoted to deuterium excess are on large time scales (glacial to interglacial changes). In this study, data from the German Antarctic base "Neumayer" will be used for a study on a small time scale. Samples of freshly fallen snow have been collected over a period of 20 years. High wind speeds at Neumayer often tend to redistribute the fallen snow to a certain extent, leading to some "depositional noise". Therefore, a trajectory model is first used to calculate the pathways of the airmasses that bring precipitation to Neumayer. Different trajectory classes will be defined and the mean deuterium excess of the snow for these classes will be calculated. An isotope model will then be used to model the observed deuterium excess values of the snow samples. Since the deuterium excess depends strongly on water vapor saturation conditions at the source area of precipitation, which are mostly unknown, further the phase difference between deuterium and deuterium excess will be investigated using data from a shallow firn core that covers the time period 1892-1981. This phase difference is less dependent on the relative humidity in the source area and is thus a more independent constraint from which to infer information on the vapor source contained in the excess. This new approach of combining a trajectory model with an isotope model and studying the deuterium excess on a precipitation-event base will improve the classical interpretation of water stable isotopes in ice cores.

Investigation and a thorough understanding of the climate of the past are necessary to assess the present or future possible climate changes. Valuable information about the former climate is stored in ice cores from the large ice sheets of Greenland and Antarctica. In particular, the ratio of stable isotopes (different types of water molecules) of snow, 18-O to 16-O (oxygen) and deuterium (heavy hydrogen) to H, respectively, are related to air temperature and are thus used for the climatic interpretation of ice cores. However, the stable isotope ratio changes during evaporation and condensation processes, thus it does not depend on temperature alone, but on the complete precipitation history, and other factors, such as seasonality and origin of precipitation are also important. To resolve this problem, a variable combining 18O and deuterium information, the so-called deuterium excess, d, is used to study the source areas of precipitation. In this project, data from the German Antarctic wintering base "Neumayer" were used to investigate the atmospheric processes that are important for ice core interpretation on a precipitation-event based time scale. Samples of freshly fallen snow, which can be directly related to the meteorological conditions prevailing during and before the snowfall, have been collected over a period of more than 20 years. As a first step, a so-called trajectory model was used to calculate the pathways that bring precipitation to Neumayer. Different trajectory ("pathway") classes were defined, for which precipitation showed clearly distinguishable mean stable isotope values. The amount of open water underneath the trajectory played a major role for the quality of the 18O- temperature relationship. Next the trajectory model was combined with an isotope model. The isotope model calculated the change in isotope ratios during precipitation formation and transport to the deposition site. Since it is a simple model that does not contain atmospheric dynamics (air motion), it was combined with the trajectory model. Still it does not fully account for the complex 3-dimensional dynamics. Thus good agreement between measurements and model was found only for the cases, where the atmospheric conditions were similar to the model assumptions. The model was also used to calculate the mean annual course of isotope values. This calculation was compared to the fresh snow data and also to data from a shallow firn core from the vicinity of Neumayer. Of special interested was here the difference in the month of the maximum value of 18-O and deuterium excess, respectively. This "phase lag" could be correctly calculated by the model (good agreement between model results and fresh snow data) and was dependent on the trajectory class. It was also different for the fresh snow samples and for the core data. This means: 1. There is a temporal change in this phase lag due to "post-depositional" processes (condensation and evaporation processes between the snow grains) in the snow cover. 2. A change in this phase lag found in an ice core is not only due to such post-depositional processes, but can also mean a change in the oceanic origin of precipitation. This is important, because it changes the stable isotope ratio without a change in temperature, which leads to errors in the climatic interpretation of the ice cores.