Comparison of radiative signals between precipitating and non-precipitating clouds in frontal and typhoon domains over East Asia

Satellite visible and infrared observations have their own advantages in cloud and precipitation remote sensing, for which the technique for discriminating precipitating clouds (PCs) and non-precipitating clouds (N-PCs) relying on such passive measurements is vital. Combined measurements from precipitation radar (PR) and visible/infrared scanner (VIRS) aboard the Tropical Rainfall Measuring Mission (TRMM) satellite were analyzed to clarify visible/infrared signals from PCs and N-PCs, where PCs and N-PCs were distinguished by PR from the cloud clusters in frontal and typhoon domains. Frequency statistics are employed to examine the characteristic difference of radiances between PCs and N-PCs, which is essential for satellite-based PC identification and in turn rainfall estimation. The frequency distribution of PCs and N-PCs were found to be approximately similar, except acceptable discrepancy of 0.63 µm reflectance (RF1) and 10.8 µm brightness temperature (TB4) in frontal regions, and on RF1 in typhoons, indicating the finite capability of individual visible or infrared signals to identify PCs. The reflectance ratio of 0.63 µm to 1.6 µm (RFT12) is more satisfactory for differentiating PCs and N-PCs than any single signal, while the brightness temperature differences between infrared channels supply no additional benefit. Based on both probability of detection (POD) and false alarm ratio (FAR), the composite use of RFT12 and TB4 proves to be the optimal strategy for discriminating PCs and N-PCs in both frontal and typhoon domains, in which the GOES Precipitation Index (GPI) scheme was revealed to be impracticable.