Interannual variability of Greenland winter precipitation sources: 2. Effects of North Atlantic Oscillation variability on stable isotopes in precipitation

Abstract

A new Lagrangian moisture source diagnostic is applied to identify the atmospheric conditions relevant for the fractionation of stable water isotopes during evaporation over the ocean and subsequent transport to Greenland. Northern Hemisphere winter months with positive and negative North Atlantic Oscillation (NAO) index are studied on the basis of ERA‐40 reanalysis data. Diagnosed moisture transport conditions are supplied to a Rayleigh‐type isotope fractionation model to derive estimates for the isotopic composition of stable isotopes in winter precipitation on the Greenland plateau for the two NAO phases. Because of changes in atmospheric circulation, moisture source locations for precipitation in Greenland vary strongly for different phases of the NAO. The mean source SST is ∼5.0 K warmer during negative NAO months compared to the positive phase. This signal is considerably stronger than what would result from interannual SST variability at a spatially fixed moisture source. Furthermore, moisture transport takes place at warmer temperatures during NAO negative conditions. Simulated average isotopic depletion of Greenland precipitation is less negative by 3.8 ± 6.8‰ for δ18O during the negative compared to the positive NAO phase. Comparison with ice core data from central Greenland for three winters shows good agreement between observed and simulated variability. The synoptic interplay of the initial conditions at the moisture sources and of the atmospheric transport conditions leads to enhanced NAO‐related interannual variability of stable isotopes. This could be important for understanding rapid shifts in stable isotopes during past climates. The isotope modeling applied here, however, considerably underestimates the absolute level of isotopic depletion. The offset is attributed to approximations in the model and uncertainties in the comparison with observational data. The high spatial resolution of the Lagrangian method reveals the nonhomogeneous structure of isotope NAO variability over the Greenland ice sheet. The results are therefore potentially useful for selecting new ice core drilling sites with maximum NAO variability.