Evaluation of INSAT-3D satellite derived precipitation estimates for heavy rainfall events and its validation with gridded GPM (IMERG) rainfall dataset over the Indian region

In the current study, an attempt has been made to investigate the potential utilization of INSAT-3D satellite derived rainfall estimations i.e., Hydro-Estimator Method (HEM), INSAT Multi-spectral rainfall (IMR) and Quantitative Precipitation Estimation (QPE) for the analysis of heavy rainfall episodes over the Indian region from the operational INSAT Meteorological Data Processing System (IMDPS), New Delhi. The dataset used for the comparison and validations are, Integrated Multi-satellitE Retrievals for GPM (Global Precipitation Measurement) (IMERG) (0.1° × 0.1°) data, IMD's surface in-situ raingauge data and daily merged satellite gauge (DMSG) rainfall data (0.5° × 0.5°). The results show that all three INSAT-3D satellite rainfall products captured the rainfall pattern spatially and temporally very well in term of the occurrence of rainfall events. For the heavy rainfall events, HEM shows good skill and correlation (r > 0.7) in detecting heavy rainfall with an accuracy of ± 20 mm with good pattern matching with actual rainfall, whereas IMR and QPE correlation are less than 0.3 and 0.1 respectively. The performance statistics with IMERG and daily merged satellite gauge rainfall shows that HEM correlates well with both the dataset with a small deviation. The underestimation of heavy rain is severe for higher elevation, suggesting HEM, IMR and QPE are not able to detect orographic enhancement appropriately; however, to some extent, HEM is better able to detect heavy rain events despite the fact that it failed to reach the magnitude of rainfall. Further, the INSAT-3D derived HEM and IMR products with GPM data during initial stages of cyclone 'Roanu' has been considered to illustrate the observing pattern of rainfall. It is found that HEM captured the similar pattern of single-rain-band indicating moderate to heavy rainfall around the cyclone in comparison with GPM. This study is one of the earlier assessments of operational INSAT-3D derived rainfall products and its comparison with IMERG/DMSG and actual rainfall data which shows that HEM is better for heavy rainfall and could be used in future for meteorological and hydrological models while QPE and IMR have better skills for light to moderate rainfall which can be used for agriculture and ground water recharge.