Simulated Doppler Radar Observations of Inhomogeneous Clouds: Application to the EarthCARE Space Mission

Abstract

A new simulation technique for spaceborne Doppler radar observations that was developed specifically for inhomogeneous targets is presented. Cloud inhomogeneity affects Doppler observations in two ways. First, line-of-sight velocities within the instantaneous field of view are unequally weighted. As the large forward motion of a spaceborne radar contributes to these line-of-sight velocities this causes biases in observed Doppler speeds. Second, receiver voltages now have time-varying stochastical properties, increasing the inaccuracy of Doppler observations. The new technique predicts larger inaccuracies of observed Doppler speeds than the traditional random signal simulations based on the inverse Fourier transform. The accuracy of Doppler speed observations by a spaceborne 95-GHz radar [as part of the proposed European Space Agency (ESA)/Japan Aerospace Exploration Agency (JAXA)/National Institute for Information and Communications Technology (NICT) EarthCARE mission] is assessed through simulations for realistic cloud scenes based on observations made by ground-based cloud-profiling radars. Close to lateral cloud boundary biases as large as several meters per second occur. For half of the cloud scenes investigated, the distribution of the in-cloud bias has an rms of 0.5 m s?1, implying that a bias in excess of 0.5 m s?1 will not be uncommon. An algorithm to correct the bias in observed Doppler observations, based on the observed gradient of reflectivity along track, is suggested and shown to be effective; that is, the aforementioned rms bias reduces to 0.14 m s?1.