Precipitation Estimation from Radar and Radiometric Observations from Trmm Data Using Artificial Neural Networks

Artificial Neural Network (ANN) technique has been used for the estimation of precipitation, mainly from passive and active microwave measurements from space. ANN has been used to estimate precipitation using TRMM Microwave Imager (TMI) onboard Tropical Rainfall Measuring Mission (TRMM) satellite. A precipitation algorithm designed to generate rainfall estimates using a combination of TMI and TRMM Precipitation Radar (PR) data has been developed. The inputs for the ANN are brightness temperatures (BT) from TMI and the output is PR-rainfall. The networks are trained and cross validated. Once the training is complete, the independent data sets (which were not included in the training) were used to test the performance of the network. Instantaneous precipitation estimates demonstrates correlations of around 0.82 to 0.97 with independent test data sets for the Indian regions. Multiple regression (MR) technique is also used to estimate precipitation using the same database and the results are compared with those obtained from ANN. The model developed can be used for the estimation of precipitation at high spatial and temporal resolutions on instantaneous basis .