Simultaneous Retrieval of Sea Surface Wind Speed and Sea Surface Temperature from a Multi-frequency Scanning Microwave Radiometer

Derivation of geophysical parameters from satellite measured brightness-temperature (TB) is an important aspect of satellite remote sensing. Primarily, this involves development of complex inversion algorithms and empirical relations comprising TB and in situ data for parameter retrieval and algorithm validation. In the present work, an Artificial Neural Network model has been attempted to simultaneously obtain sea surface wind speed (WS) and sea surface temperature (SST) utilizing TB from 8 channels (including vertical and horizontal polarizations) of Multi-frequency Scanning Microwave Radiometer on board Indian Remote Sensing Satellite (IRS-P4) and deep sea ocean buoys in the North Indian Ocean region. The ANN obtained values are then compared with actual in situ observations as a test for the performance of the model. It is concluded that the ANN model is able to provide good estimates of WS and SST within acceptable error limits. The goal of the present work is to pre-establish the suitability of ANN approach for geophysical parameter retrieval from satellite measured TB in the Indian context particularly keeping in view the forth coming satellite launches like Megha Tropiques and Oceansat-3.