PNNL

Abstract

A new parameterization for boundary layer cumulus clouds, called the cumulus potential (CuP) scheme, is introduced. Unlike many other parameterizations, the CuP scheme explicitly links the fair-weather clouds to the boundary-layer turbulence and accounts for the non-local nature of the turbulence. This scheme uses joint probability density functions (JPDFs) of virtual potential temperature and water-vapor mixing ratio, as well as the mean vertical profiles of virtual potential temperature, to predict the amount and size distribution of boundary layer cloud cover. This model considers the diversity of air parcels over a heterogeneous surface, and recognizes that some parcels rise above their lifting condensation level to become cumulus, while other parcels might rise as clear updrafts. This model has several unique features: 1) surface heterogeneity and boundary-layer turbulence is represented using the boundary layer JPDF of virtual potential temperature versus water-vapor mixing ratio, 2) clear and cloudy thermals are allowed to coexist at the same altitude, and 3) a range of cloud-base heights, cloud-top heights, and cloud thicknesses are predicted within any one cloud field, as observed. Using data from Boundary Layer Experiment 1996 and a model intercomparsion study using large eddy simulation (LES) based on the Barbados Oceanographic and Meteorological Experiment (BOMEX), the CuP scheme is compared to three other cumulus parameterizations: one based on relative humidity, a statistical scheme based on the saturation deficit, and a slab model. It is shown that the CuP model does a better job predicting the cloud-base height and the cloud-top height than three other parameterizations. The model also shows promise in predicting cloud cover, and is found to give better cloud-cover estimates than the three other cumulus parameterizations. In ongoing work supported by the US Department of Energy¹s Atmospheric Radiation Measurement Program, the CuP scheme is being implemented in the Weather Research and Forecasting (WRF) model, in which it replaces the ad-hoc trigger function in an existing cumulus parameterization.