Challenges in Mesoscale Prediction of a Nocturnal Stratocumulus-Topped Marine Boundary Layer and Implications for Operational Forecasting

Abstract

A numerical study using the fifth-generation Pennsylvania State University–National Center for Atmospheric Research Mesoscale Model (MM5) was performed to assess the impact of initial and boundary conditions, the parameterization of turbulence transfer and its coupling with cloud-driven radiation, and cloud microphysical processes on the accuracy of mesoscale predictions and forecasts of the cloud-capped marine boundary layer. Aircraft, buoy, and satellite data and the large eddy simulation (LES) results during the Dynamics and Chemistry of Marine Stratocumulus field experiment (DYCOMS II) in July 2001 were used in the assessment. Three of the tested input fields (Eta, NCEP, and ECMWF) show deficiencies, mainly in the thermodynamic structure of the lowest 1500 m of the marine atmosphere. On a positive note, the simulated marine-layer depth showed good agreement with aircraft observations using the Eta fields, while using the NCEP and ECMWF datasets underestimated the marine-layer depth by about 20%–30%. The predicted turbulence kinetic energy (inversion strength) was about 50% of that obtained from the LES results (aircraft observed). As a consequence of moisture overprediction, the predicted liquid water path was twice the observed by 1–2 g kg 1. The sensitivity tests have shown that the selections of turbulence and cloud microphysical schemes significantly influence the turbulence estimates and cloud parameters. Two of the tested turbulence schemes (Eta PBL and Burk–Thompson) did not exhibit the coupling with radiation. The significant differences in the simulated turbulence estimates appear to be a consequence of the use of water-conserving potential temperature variables. The microphysical parameterization, which uses the number concentration of cloud drops in the autoconversion process, simulates a realistic evolution of precipitable hydrometeors in the cloudy marine layer on the positive side, but on the other hand enhances the decoupling in the turbulence structure. This study can provide guidance to operational forecasters concerning accuracy issues of the commonly used large-scale analyses for model initialization, and optimal selection of model parameterizations in order to simulate and forecast the cloudy atmospheric boundary layer over the ocean.