The two Arctic wintertime boundary layer states: Disentangling the role of cloud and wind regimes in reanalysis and observations during MOSAiC

The central Arctic atmosphere during winter comprises two distinct synoptic states: a radiatively clear state, which is linked to clear sky, strong surface cooling and temperature inversions; and a radiatively opaque state, which is linked to mixed-phase clouds, weak surface radiative cooling, and more neutrally-buoyant boundary layers. Weather and climate models are often reported to lack the representation of processes associated with these states, but most prior work has treated the problem as an aggregate of synoptic conditions. Here, we disaggregate the Arctic states in an evaluation of ERA5 reanalysis and compare to observations from the MOSAiC drift campaign over the central Arctic sea ice from November 2019 – March 2020. Combining near surface winds and liquid water path (LWP), nine different classes describing synoptic conditions spanning the states are derived. Results show that the clear state is primarily formed by weak and moderate winds and the absence of liquid-bearing clouds, while strong wind cases and enhanced LWP forms the occurrence peak in the radiatively opaque state. ERA5 struggles to reproduce these basic statistics, shows too weak sensitivity of thermal radiation to synoptic forcing, and for similar LWP amounts, it overestimates both upward and downward longwave radiation due to a warm bias near the surface. This warm bias has a pronounced vertical structure and is largest in clear and calm conditions, owing to the lack of surface inversions in ERA5. Under strong synoptic forcing, the warm bias is constant with height and discrepancies in mixed-phase cloud altitude appear. Thus, biases in each state are partially opposing in a manner that makes them overlap unrealistically, masking the distinctions that are known to form the first-order variability of the Arctic winter energy budget. Separating between different synoptic conditions in conjunction with the classical two radiative states classification is therefore regarded a useful step for isolating dominant processes for evaluation of the Arctic troposphere in models.