Testing NCEP operational surface layer parameterizations for stable conditions using CASES99 data

Impetus to extend operational forecasting capability beyond weather to air quality prediction has prompted an investigation at the National Center for Environmental Prediction (NCEP) of the stable boundary layer (SBL) parameterizations used in their forecast models. The proposed presentation will report on an evaluation of the surface layer (SL) parameterizations in stable conditions used in NCEP'S GFS and ETA forecast models. These parameterizations, modifications to Monin-Obukhov theory designed to represent the effects of higher ("enhanced") turbulence intensities than inherent in theoretically-based Monin-Obukhov (MO) theory, produce warmer near-surface temperatures than achieved with MO theory. Predicted two-meter temperature and surface heat fluxes from the GFS, ETA and MO parameterizations are compared with observations from four nights of high static stability during the CASES99 field experiment. The results of the investigation extend knowledge of theoretical versus turbulence-enhanced stable SL parameterizations by focusing on a) a particular class of evening, namely highly stable nocturnal boundary layers and b) surface heat flux as well as near-surface temperature. Results show an interesting paradox: while enhanced (GFS and ETA) parameterizations predict warmer two-meter temperatures than MO theory, in agreement with observations, they overpredict the magnitude of surface heat flux. MO theory, on the other hand, accurately predicts the low surface heat fluxes observed on these evenings, while underpredicting two-meter temperatures. Physical reasons for as well as practical implications concerning operational forecasting of this paradox will be discussed. This effort is also in coordination with the GEWEX GABLS stable boundary layer project.