Due to the recent and severe downtrend in sea ice coverage, Arctic-derived moisture serves as new, increasingly important, water source for the northern hemisphere. Feedback and exchange processes between the different hydrological compartments of the Arctic might be tracked by stable water isotopologues (H216O, H218O, HD16O). This is possible as evaporative sources, phase changes and transport history have a specific imprint on the isotopic compositions. The MOSAiC drift experiment offered the unique possibility to tackle the main hydrological processes occurring in the Central Arctic, covering a complete seasonal cycle, including the understudied Arctic winter. A Cavity Ring Down Spectrometer (CRDS) was installed on board of RV Polarstern and atmospheric humidity, δ18O, δD and d-excess were observed continuously from October 2019 to October 2020. Simultaneously, isotopic changes of water vapour have been measured by international partners at several land-based Arctic stations. A first analysis of the Polarstern isotopic vapour dataset reveals a range of 30‰ (min=-48.4; max=-11.4; mean=-32.4) variations in δ18Ο of atmospheric water vapour. A clear seasonal cycle with the most depleted values occurring in the dry and cold winter months and increasingly enriched values in spring, peaking in August is noticed. Strong, positive correlation is found with both local specific humidity (r2 = 0.87) and air temperature (r2=0.81). Several short-term events on synoptical time scales with abrupt fluctuations in the isotopic composition are detected throughout the entire dataset, especially during the freeze up phase (Oct-Nov) and the transition from frozen conditions to summer melt (Apr-Jun). Preliminary comparison of the Polarstern data with measurements from different Arctic stations indicates a strong influence of sea ice coverage on the isotopic signal. For an in-depth understanding of the observed isotopic changes, we quantitatively compare the measured isotopic signatures with model results from an ECHAM6 atmosphere simulation, which includes explicit water isotope diagnostics. For this simulation, pressure and temperature fields have been nudged to ERA5 data. The model-data comparison assesses the capability of this state-of-the-art AGCM to capture the first-order evaporation/condensation processes and their seasonal evolution. However, both a systematic overestimation of winter values and overall decreased variability of modeled isotope values as compared to the observation is found. Investigation of such discrepancies may help to identify deficits in the representation of the fine-scale exchange processes characterizing the central-Arctic water cycle.