How Does a Dry General Circulation Model Represent Atmospheric Blocking?

Abstract While dry dynamics has been a key framework for understanding blocking events, recent studies have highlighted the significant role of moist processes in blocking life cycles. However, it remains unclear whether moist processes introduce unique characteristics in the blocking life cycle that cannot be captured by dry dynamics or whether they are a source of extra energy for synoptic eddies that are needed to form realistic blocking circulations. Our approach to address the above squestion is to perform a long-term simulation of a dry general circulation model (GCM) with synoptic eddy energy and mean flow characteristics close to the winter climatology of ERA5 reanalysis. Blocking frequencies are found to be reduced by about one-third in the model compared to ERA5 using a blocking detection based on potential vorticity (PV). This moderate underestimation of blocking frequency in the dry model is due to rarer favorable conditions to initiate a blocking anticyclone. The model needs episodes of weaker eastward flows than in ERA5 to compensate for the weaker westward expansion of the blocking anticyclone associated with the smaller divergent winds. The main characteristics of the blocks (intensity, size, duration) are similar in the two datasets despite some slight structural differences: ERA5 blocks are more Ω shaped, and blocks in the dry model are more dipole shaped. Even though the origins of the air parcels forming a blocking anticyclone are very different in the two datasets, the dry model compensates for its less efficient transport of low-PV air from low levels by transporting more efficiently low-PV air from lower latitudes at upper levels.