The representation of atmospheric processes in northeast Greenland in CMIP6 models

Since the end of the 20th century, Greenland has been the largest contributor to sea level rise.  As temperatures continue to increase, this tendency is projected to continue.  This has resulted in numerous studies which evaluate present and future conditions of the Greenland Ice Sheet, many of which use general circulation models (GCMs). The majority of these focus on sea level rise and/or surface mass balance. While some analyses of atmospheric processes have been undertaken, these have typically been over a larger scale (Arctic or Greenland).  This has led to a lack of regional studies of atmospheric processes and how they are represented in GCMs, particularly over northeast Greenland, an area of increased interest in both its glaciology and atmosphere.

To address this, 67 CMIP6 GCM realizations were subject to the Pickler and Mölg (2021) model selection procedure to determine the most suitable realization over northeast Greenland.  The historical simulation of these realizations were evaluated for: (i) their ability to capture the space-time climatic anomalies over 1979-2014 with respect to ERA5 reanalysis data and (ii) their ability to simulate the mean climatic state of northeast Greenland with respect to four automated weather stations over 2009-2020.  MPI-ESM1-2-HR r6i1p1f1 was found to rank highest and ACCESS-ESM1-5 r10i1p1f1 lowest.

The 67 realizations were then evaluated on their ability to capture two important processes influencing the region: the North Atlantic Oscillation (NAO) and the Greenland blocking (GBI).  All realizations were able to simulate the NAO during boreal winter, while all failed to capture the GBI during boreal summer.  Furthermore, the ability of the top and bottom ranked realizations to simulate precipitation, katabatic winds, sea ice, and warm-air events were examined. This analysis reveals key differences between the representation of regional climates within the GCMs, which highlights the need for a rigorous selection procedure prior to estimating future changes.