Parameterization of Entrainment Rate for Cumulus Clouds with WRF Simulation

Abstract

By using Weather Research and Forecasting Model (WRF) to simulate a southwest vortex precipitation process, this work studies the correlations between entrainment rate (λ) and dynamical parameters in the cloud and further fit λ. We relate the probability density distribution (PDF) to the parameterization of λ and find that the greater the probability, the larger the slope of the logarithmic liner function. The slope of the log-linear fitting function in fitting decreases for developing and enhancing cumulus clouds, which is related to the increase in updraft motion and the decrease in λ. Then, we group clouds according to cloud top heights and calculate average λ and dynamic parameters, and the results indicate that when only one dynamic parameter is used, vertical wind velocity (w) is more suitable than buoyancy (B) to be used to fit λ. The fitting functions combing one single parameter and more parameters by principal components regression are compared with two traditional schemes, and we found that λ obtained by our fitting schemes are between the two traditional schemes. Because the principal component regression method takes into account the interaction between more dynamic factors and entrainment, the fitting function, including w and B, is suitable to be applied to fit λ in the parameterization scheme for cumulus clouds.