Microwave radiometry for the detection of snowfall

V. GALLIGANI; CARLOS JIMENEZ; ERIC DEFER; PRIGENT, CATHERINE

Paris

Workshop; CALIPSO, CloudSat, EarthCARE ESA Joint Workshop; 2012

The quantification of falling snow on a global basis is important to capture the full atmospheric water cycle and to completely characterize the Earth´s energy and radiation budget. In principle, microwave frequencies are sensitive to scattering from frozen hydrometeors, with the sensitivity depending to a large degree on the size of the hydrometeor with respect to the observing wavelength. For this reason, sensors such as the Microwave Humidity Sounder (MHS) on board MetOp providing observations at 89 GHz, 157 GHz, 183 ± 1 GHz, 183 ± 3 GHz and 190 GHz, has the potential to be used for snowfall characterization. Active microwave measurements responding to cloud particles, light rain, and snow events are also available through the CPR radar (94 GHz) aboard CloudSat.In reality, the estimation of snowfall form the present suite of observations is still at a very early stage. The main difficulties encountered in characterizing snowfall from space are (a) the weak signal from snowfall with respect to the background emission from the snow covered land, (b) separating the signal related to the emission/scattering from other cloud hydrometeors from the snowfall signal; and (c) the complex variability and lack of parameterizations of the microphysical properties, and thus radiative properties, of snow particles (e.g. shape, size, density, wetness, and the related dielectric properties). This study seeks to improve the understanding of the relationship between the physical properties of snowfall and the radiative properties associated with radar reflectivities and sounder radiances. In contrast to other studies based on analyzing only one type of observation we exploit here the synergy of active and passive observations from a database of (a) collocated real observations from CloudSat and MHS, and (b) simulated observations for the same sensors from atmospheric scenes generated by a cloud resolving model (MESO-NH) capable of calculating complete hydrometeor profiles, including snow.Present work is going in two directions. First the Advanced Radiative Transfer Simulator (ARTS) has been used to simulate both the active and passive observations corresponding to snowfall situations reproduced by Meso-NH. Simulations have also been calculated for frequencies for planned sensors (118 GHz – 874 GHz) to further explore the potential of the submillimeter/millimeter range in the context of snowfall characterization. Special attention is being paid to the sensitivity of the simulations to snow physical properties, especially those variables affecting the scattering properties of the snowfall particles. Secondly, the hydrometeors profiles from selected relevant CPR inversions have been used to simulate the coincident MHS observations from the collocation database. The difference between the MHS simulations and that of the observations is helping us constraining the physical properties of snow in the radiative transfer model. Our major findings will be presented.Finally, we expect to develop a snow retrieval scheme for the passive microwave band on a data base of simulations derived from realistic meteorological situations. Coupling of passive and active observations will possibly be examined.