Spatial variability of near-surface air temperature in the Copernicus Arctic regional reanalysis

This study investigates spatial variability of near-surface air temperature at scales of tens of kilometres in the European Arctic, using the Copernicus Arctic Regional Reanalysis (CARRA) data from 1991 to 2022. It examines factors influencing this variability, compares model-derived variability with observed subgrid variability, and evaluates its impact on model verification. Key findings reveal that spatial variability is strongly influenced by geographical features and seasonal changes. The highest seasonal spatial variability is found during winter, particularly near coasts, mountainous regions, and sea ice edges where persistent temperature gradients drive baseline variability. In contrast, in inland Scandinavia, where climatic temperature gradients are less pronounced, situation-dependent variability predominates. The study highlights these distinct patterns of variability in Ny-& Aring;lesund, Svalbard, and Sodankyl & auml;, Finland. In Sodankyl & auml;, high variability is found under cold and stable conditions, with variability peaking above the surface due to spatial differences in temperature inversion development. The observed spatial variability largely aligns with the modeled CARRA data, particularly in terms of the relationship between baseline and situation-dependent variability. Furthermore, considerable subgrid variability observed in Sodankyl & auml; accounts for a substantial portion of the difference between the model grid box value and observed point value. These findings emphasize the importance of considering spatial variability in weather forecasting, model verification, model intercomparisons, and process studies.