ABSTRACT

Cloud schemes in regional and global climate models have to be highly parameterized to allow for long integration times. Because of the non-linearity of many cloud microphysical processes and the unresolved subgrid scale variability of cloud properties and other meteorological variables, cloud schemes are often empirically tuned to obtain agreement with observed top of the atmosphere radiative fluxes. This tuning approach significantly limits the use of these schemes in climate change scenarios in two ways. The parameterizations become scale-dependent and it remains unclear whether this empirical and unphysical tuning to current-day conditions is still valid for global warming scenarios.
A novel approach to deal with subgrid scale variability in cloud parameterizations has been developed. This new approach is based on a scale-aware parameterization of the subgrid scale cloud variability, which is implemented as a column-integral constraint. The method is applied to warm rain processes effectively eliminating autoconversion and accretion as scale-dependent empirical tuning variables. Model integrations with the IPRC Regional Atmospheric Model at different spatial resolutions (between 25 and 200 km) show that the new method provides significant improvements in the scale-dependency of the representation of e.g. the East Pacific stratocumulus fields at typical (regional) climate model resolutions.