Assessing arctic wetting: Performances of CMIP6 models and projections of precipitation changes
Ziyi Cai,Qinglong You,Hans W. Chen,Ruonan Zhang,Zhiyan Zuo,Deliang Chen,Judah Cohen,James A. Screen
DOI: https://doi.org/10.1016/j.atmosres.2023.107124
IF: 5.965
2023-11-25
Atmospheric Research
Abstract:The Arctic region is experiencing a notable increase in precipitation, known as Arctic wetting, amidst the backdrop of Arctic warming. This phenomenon has implications for the Arctic hydrological cycle and numerous socio-ecological systems. However, the ability of climate models to accurately simulate changes in Arctic wetting has not been thoroughly assessed. In this study, we analyze total precipitation in the Arctic using station data, multiple reanalyses, and 35 models participating in the Coupled Model Intercomparison Project Phase 6 (CMIP6). By employing the moisture budget equation and an evaluation method for model performance with ERA5 reanalysis as a reference, we evaluated the models' capability to reproduce past Arctic wetting patterns. Our findings indicate that most reanalyses and models are able to replicate Arctic wetting. However, the CMIP6 models generally exhibit an overestimation of Arctic wetting during the warm season and an underestimation during the cold season from 1979 to 2014 when compared to the ERA5 reanalysis. Further investigation reveals that the overestimation of wetting during the warm season is largest over the Arctic Ocean's northern part, specifically the Canadian Arctic Archipelago, and is associated with an overestimation of atmospheric moisture transport. Conversely, the models significantly underestimate wetting over the Barents-Kara Sea during the cold season, which can be attributed to an underestimation of evaporation resulting from the models' inadequate representation of sea ice reduction in that region. The models with the best performance in simulating historical Arctic wetting indicate a projected intensification of Arctic wetting, and optimal models significantly reduce uncertainties in future projections compared to the original models, particularly in the cold season and oceanic regions. Our study highlights significant biases in the CMIP6 models' simulation of Arctic precipitation, and improving the model's ability to simulate historical Arctic precipitation could reduce uncertainties in future projections.
meteorology & atmospheric sciences