PNNL

Abstract

Large spatial inhomogeneity in shallow convection raises uncertainties in estimation of the domain-averaged cloud fraction profile (CFP). Here, numerical model simulations of shallow convection over land, coupled with a sophisticated radar simulator, are used to address this issue, and an objective method for estimating domain-averaged CFP is proposed. The analysis clearly indicates that profiling observations are inadequate to provide reliable CFP estimates. The use of Scanning Cloud Radar (SCR) performing a sequence of cross-wind horizon-to-horizon scans is not straightforward due to the strong dependency of its sensitivity with range. The retrieved CFP by a novel methodology that uses the observed statistics of SCR hydrometeor detections with height to estimate optimum ranges shows very good agreement with the model CFP. On average of CFP estimates more than 40 min of SCR observations are required to converge on the model domain average. The proposed technique is expected to improve our ability to compare model and observations in broken cloud conditions.