Systematic large-scale secondary circulations in a regional climate model

Regional climate models (RCMs) are used to add the effects of nonresolved scales to coarser resolved model simulations by using a finer grid within a limited domain. We identify large-scale secondary circulations (SCs) relative to the driving global climate model (GCM) in an RCM simulation over Europe. By applying a clustering technique, we find that the SC depends on the large-scale flow prescribed by the driving GCM data. Evidence is presented that the SC is caused by the different representations of orographic effects in the RCM and the GCM. Flow modifications in the RCM caused by the Alps lead to large-scale vortices in the SC fields. These vortices are limited by the RCM boundaries, causing artificial boundary-parallel flows. The SC is associated with geopotential height and temperature anomalies between RCM and GCM and has the potential to produce systematic large-scale biases in RCMs.