First Three Years of the Microwave Radiometer aboard Envisat: In-Flight Calibration, Processing, and Validation of the Geophysical Products

Abstract The Envisat microwave radiometer is designed to correct the satellite altimeter data for the excess path delay resulting from tropospheric humidity. Neural networks have been used to formulate the inversion algorithm to retrieve this quantity from the measured brightness temperatures. The learning database has been built with European Centre for Medium-Range Weather Forecasts (ECMWF) analyses and simulated brightness temperatures by a radiative transfer model. The in-flight calibration has been performed in a consistent way by adjusting measurements on simulated brightness temperatures. Finally, coincident radiosonde measurements are used to validate the Envisat wet-tropospheric correction, and this comparison shows the good performances of the method.