Validating INSAT-3D atmospheric temperature retrievals over india using radiosonde measurements and other satellite observations

Abstract

Atmospheric temperature profiles retrieved from a geostationary INSAT-3D (Indian National Satellite System) sounder level-2 physical retrieval data are evaluated with respect to radiosonde observations, Atmospheric Infrared Sounder (AIRS), and National Centers for Environmental Prediction and Department of Energy (NCEP-DOE) reanalysis-2 data sets over India and adjoining area. This evolution was carried out in the terms of correlation coefficients, bias, and root-mean-square error (RMSE) at each pressure level from surface 1000 to 100 hPa during a one year period in 2016.The initial validation of INSAT-3D temperature profiles with AIRS and NCEP-DOE reanalysis-2 data were taken as reference and computed for the spatial distribution of correlation coefficient, bias, and RMSE at each pressure level over the Indian region. In addition to that, we also compared the INSAT-3D temperature retrievals with 12 selected IMD (India Meteorological Department) radiosonde observations at different geographical features over the Indian subcontinent during the one year period in 2016. In this analysis, for each station at all pressure levels, the correlation coefficients, biases and their corresponding root-mean-square errors (RMSEs) were carried out between INSAT-3D and radiosonde observations. The results of the inter-comparison reveal that temperature retrievals from the INSAT-3D are a good agreement with AIRS and NCEP-DOE reanalysis-2 over the land region and a slight degradation performance over the ocean- and coastal regions of Arabian Sea and Bay of Bengal below 850 hPa. The degradation performance over the ocean- and coastal areas associated with maybe possibility of undetected clouds and uncertainty in surface emissivity and also it might be attributed to improper bias correction coefficients used for brightness temperature of clear-sky pixels before physical retrievals in INSAT-3D algorithm or inherent deficiency in the retrieval algorithm which mainly depends on initial guess field from NWP models. Furthermore, an evaluation of INSAT-3D with 12 IMD radiosonde observation results reveals that temperature retrievals for each station shows very good agreement at all pressure levels, except slight degradation in performance over the coastal regions (GOA, MDS) and island (AMN, PBL) stations possibly the same reason as that of AIRS/NCEP. Overall, the results show that INSAT-3D temperature retrievals over land are a good agreement with AIRS/NCEP and RSRW then over the ocean- and coastal regions which need to be analyzed.