Near-Real-Time Estimation of Water Vapor Column From MSG-SEVIRI Thermal Infrared Bands: Implications for Land Surface Temperature Retrieval

Abstract

The Meteosat Second Generation-Spinning Enhanced Visible and Infrared Imager (MSG-SEVIRI) instrument provides observations of half the globe every 15 min, at low spatial resolution. These data are an invaluable tool to observe daily to yearly cycle of land surface temperature (LST), as well as for various early warning systems. However, advanced algorithms for LST estimation requires a previous estimation of the water vapor (WV) column above the observed pixel, for which no instantaneous retrieval methods are yet available, and therefore hinders their implementation in a near-real-time processing chain for MSG-SEVIRI data. This work analyzes three different formulations for such WV retrieval, which are compared to independent WV estimates obtained from radiosoundings. The best suited algorithm is then selected for WV estimation and compared with the results obtained with a previous noninstantaneous algorithm [23] . This comparison shows that, in spite of retrieval errors higher than the ones reported in the literature, the estimated WV compares relatively well with in situ data, while allowing for an instantaneous estimation of WV column (every 15 min). Error propagation analysis and direct comparison show that the observed increase in WV estimation error has a negligible influence on LST retrieval. Therefore, this algorithm is well suited to be implemented in a near-real-time processing chain for MSG-SEVIRI data.