Retrieval of water vapor properties from measurements of differential attenuation at microwaves

The assimilation of measures in the Numerical Weather Prediction (NWP) models provide a fundamental contribution in the reconstruction of the correct dynamics of atmospheric events. A better knowledge of the water vapour (WV) distribution of the atmosphere is of paramount importance in NWP capabilities at short time scale. In this paper, a new technique to retrieve the vertical profiles of WV from measurements of differential attenuation signal at microwaves, is discussed. Recent studies show that integrated water vapor (IWV) is strictly related to the normalized differential attenuation undergone by two-tone signals, called spectral sensitivity. This innovative approach, called Normalized Differential Spectral Attenuation (NDSA), allows to evaluate the IWV from attenuation measurements performed in K U /K bands along microwave links crossing the troposphere and it was demonstrated that can be successfully applied from space by using one (or more) couple(s) of co-rotating or counter-rotating LEO (Low Earth Orbiting) satellites. The possibility to convert an angular profile of IWV to the vertical WV distribution was already tested for co-rotating satellites configuration in previous projects funded by ESA, however the measurement has never been performed yet. For retrieving the WV profiles, a forward model was used to simulate the ray-tracing of a microwave signal from a transmitter to a receiver in atmosphere with horizontal inhomogeneities and then a retrieval code was implemented to invert the measurements of IWV calculated along the path length. A retrieval code, based on the optimal estimation method, was implemented to invert the IWV measurements. The retrieval performance was tested by using synthetic measures generated from selected scenarios for two different configurations with the receiver on board an aircraft and the transmitter placed at 10 m above the sea level and on the top of mount Cimone at 2165 m, respectively.