Sources of temperature biases in Regional Climate Models over complex orography: a general approach

Amid the alarming pace and effects of human-induced climate change, mountainous regions are warming at about twice the global average rate. Modeling climate and climate change scenarios over regions with highly complex topography, such as the Alps, remains a significant challenge for regional climate modeling. Better characterizing the sources of model biases is a major issue, particularly in areas with complex terrain.

We analyze the sources of bias affecting near-surface temperature (TAS) in an ensemble of EURO-CORDEX models, focusing on the Friulian Alps. By examining the vertical structure of atmospheric thermal profiles, we identify and quantify four main sources that contribute to surface temperature biases at specific locations or grid points.

• The first source is related to the ensemble's ability to reproduce free-atmosphere temperatures, such as those at the 500 hPa level.

• The second component accounts for the biased representation of the thermal gradient between the free-atmosphere and the boundary layer top.

• The third component is associated with model errors in the height of the boundary layer top. Under the environmental lapse rate approximation, this component corresponds to the orographic bias at a station or grid point. In the mountainous region we examined, the orographic bias represents a significant source of error.

• The final contribution to the TAS bias stems from the inadequate representation of processes within the boundary layer, which exhibit temporal and spatial variability depending on the type of mountain boundary layer.

We provide seasonal and annual estimates for each TAS bias component and suggest that advanced statistical bias correction techniques, including machine learning approaches, may be particularly effective in addressing the specific challenges posed by the boundary-layer-dependent component of the overall TAS bias.