RAINFALL RETRIEVAL ALGORITHM USING MODIS CLOUD MASK AND CLOUD PRODUCTS

摘要

The use of visible and infrared bands, which have higher spatial resolution, is explored as an alternative to readily available satellite-derived rainfall estimates based on microwave and radar measurements. Data from the Moderate Resolution Imaging Spectroradiometer (MODIS) were used to develop the rainfall retrieval algorithm. Precipitating clouds were defined using the confident cloudy areas derived from MODIS cloud mask (MOD35) While cloud water path (CWP), cloud effective radius (CER), and cloud top temperature (CTT), which are all derived from the MODIS cloud (MOD06) products, were utilized as proxies to formulate rainfall estimate models. Regression analysis was applied using rainfall measurements from selected synoptic weather stations as training dataset. Among the models created, the model that best estimates the rainfall incorporates CWP, CER and CTT. This model has the highest R-square and the least root-mean-square error (RMSE).