Physical characteristics of frozen hydrometeors inferred with parameter estimation

Abstract. Frozen hydrometeors are found in a huge range of shapes and sizes, with variability on much smaller scales than those typical model grid boxes or satellite fields view. Neither models nor situ measurements can fully describe this variability, so assumptions have to be made applications including atmospheric modelling radiative transfer. In work, parameter estimation has been used optimise six different relevant frozen passive microwave This covers cloud overlap, convective water content particle size distribution (PSD), the large-scale snow snow, an initial exploration ice representation (particle shape PSD combined). These parameters were simultaneously adjusted find best fit between simulations from European Centre for Medium-range Weather Forecasts (ECMWF) assimilation system near-global observations covering frequency 19 190 GHz. The choices overlap particularly well constrained (or identifiable), there was even constraint PSD. practical output is set improved version 13.0 Radiative Transfer TOVS scattering package (RTTOV-SCATT), taking into account newly available such as aggregates hail, additional options. explored full space using efficient assumption linearly additive perturbations. helped illustrate issues multiple minima cost function, non-Gaussian errors, that would make it hard implement same approach standard data weather forecasting. Nevertheless, systems grow more complex, likely necessary part development process.