资源环境科技发展态势分析平台

Accurate representation of cloud vertical overlap in climate models is particularly significant for predicting the total cloud fraction (TCF) and calculating radiative budget. It refers to the parameterization of overlap parameter-decorrelation length scale L-but the potential of dynamical factors in developing parameterization of L has still received far less attention. Using ground-based radar observation over Atmospheric Radiation Measurement Southern Great Plains site, here long-term seasonal-averaged L is retrieved and shows a very high anticorrelation with TCF from different data sets, indicating that TCF is sensitive to the way of cloud overlap. Therefore, combined with meteorological reanalysis data set, a robust multiple regression model between L and dynamical factors is built and exhibits smaller TCF bias compared with previous parameterization of L. Contribution calculation further verifies that atmospheric instability contributes 70% of L variation, indicating that it dominates the long-term variation of L over Southern Great Plains site. This finding implies that dynamical factors should be taken into account in the parameterization of L.