Assessment of continuous flow analysis (CFA) for high-precision profiles of water isotopes in snow cores

Abstract. In order to derive climatic information from stable water isotopes of the very recent past, the signal-to-noise ratio in climate reconstructions from ice cores has to be improved. To this end, understanding of the formation and preservation of the climate signal in stable water isotopes at the surface is required, which in turn requires a substantial number of snow surface profiles. However, due to its high porosity and poor stability surface, snow has been rarely measured; i.e., climate records from firn and ice cores often start at several meter depths, and the few discrete samplings of surface snow required large effort. Here we present a new setup to efficiently measure stable water isotopes in snow profiles utilizing a continuous flow analysis (CFA) system enabling measuring multiple snow cores in a reasonable time and with high quality. The CFA setup is described, and a systematic assessment of the mixing of the isotope signal due to the setup is conducted. We systematically determine the mixing length at different parts of the system. We measure and analyze six snow cores from Kohnen station, Antarctica, and find the largest contribution to mixing to originate in the percolation of meltwater on top of the melt head. In comparison to discrete measurements, we show that our CFA system is able to reasonably analyze highly porous snow cores for stable water isotopes. Still, for future developments we recommend improving the melt head with respect to the strong percolation.