Retrieval of Total Column Water Vapor from Electro-L No. 3 Satellite Data Using Neural Networks

A technique is suggested for retrieving the total column water vapor from measurements of the MSU-GS instrument mounted on the Electro-L No. 3 geostationary satellite with the use of an artificial neural network. The comparison of the total column water vapor retrieved from the MSU-GS data with MODIS data and AERONET measurements showed their good agreement. The root mean square error is 0.311 cm when comparing with the MODIS data and 0.409 cm when comparing with the AERONET data; the correlation is 98.2% and 84.7%, respectively. The results confirm the effectiveness of the technique for solving problems of atmospheric physics.