Upstream Large-Scale Control of Low-Cloud Climatology

The adjustment of the atmospheric boundary layer (ABL) to upstream large-scale conditions is an essential physical process in the control of marine low-cloud coverage (LCC). It is well known that upstream conditions can significantly control LCC, with the adjustment times of the ABL up to a couple of days. However, it is unknown if the effects of upstream conditions on LCC manifest themselves on climatological timescales; are they significant for LCC climatology, or negligible compared to the internal variability of the ABL?

Parameterizations of LCC in general circulation models (GCMs) incorporate advected dynamical variables, but do not consider the evolving state of the ABL. Studies of the climate response of LCC to warming focus on the in-situ relationship between these variables and LCC, not taking upstream conditions into account. If the control of upstream conditions is significant on a climatological scale, their inclusion may significantly improve LCC representation in GCMs, as well as improve calculations of climate sensitivity.

Using monthly data, we determine back-trajectories and use supervised machine learning statistical models with feature selection capabilities to determine the influence of in-situ and upstream large-scale conditions on the LCC simultaneously. We find significant upstream effects controlling LCC in some physical regimes, coinciding with significant improvements in statistical prediction skill.