Comparison of Ceilometer, Satellite, and Synoptic Measurements of Boundary-Layer Cloudiness and the ECMWF Diagnostic Cloud Parameterization Scheme during ASTEX

Abstract

Cloud fraction is a widely used parameter for estimating the effects of boundary-layer cloud on radiative transfer. During the Atlantic Stratocumulus Transition Experiment (ASTEX) during June 1992, ceilometer and satellite-based measurements of boundary-layer cloud fraction were made in the subtropical North Atlantic, a region typified by a 1?2 km deep marine boundary layer with cumulus clouds rising into a broken stratocumulus layer underneath an inversion. Both the diurnal cycle and day-to-day variations in low-cloud fraction are examined. It is shown that ECMWF low cloudiness analyses do not correlate with the observed variations in cloudiness and substantially underestimate the mean low cloudiness. In these analyses, the parameterization of low cloud fraction is primarily based on the inversion strength. A comparison of ECMWF analyses and ASTEX soundings (most of which were assimilated into the analyses) shows that the thermodynamic structure of the boundary layer and the inversion strength are well represented (with some small but significant systematic biases) in the analyses and preserved (again with some biases) in 5-day forecasts. However, even when applied to the actual sounding the ECMWF low cloud scheme cannot predict the observed day-to-day variations or the diurnal cycle in low cloud. Other diagnostic schemes based on lower tropospheric stability, cloud-top entrainment instability, boundary-layer depth, and vertical motion do equally poorly. The only successful predictor of low cloud frontier from sounding information is the relative humidity in the upper part of the boundary layer.