Active and passive microwave observations of snowfall from space

One of the main goals of the Global Precipitation Measurement (GPM) mission is to improve snowfall retrieval accuracy, as snowfall is, the predominant component of the global precipitation amount at mid and high latitudes. The GPM Core Observatory (GPM-CO) is equipped with two instruments: the GMI, the most advanced conical precipitation radiometer with respect to both channel assortment and spatial resolution; and the Dual-frequency Precipitation Radar (DPR) [composed of two radars, a Ku-band Precipitation Radar (KuPR) (13.6-GHz) and a Ka-band (35.5-GHz) Precipitation Radar (KaPR)]. Advancements in snowfall detection and accuracy in quantitative estimates of snowfall rates at mid-high latitudes is expected from both the GMI and DPR. Moreover, thanks to the exploitation of the high-frequency channels (> 100 GHz) available on most of radiometers in the GPM constellation, providing very good coverage at mid-high latitudes (hourly or less), snowfall monitoring is now possible. Among these radiometers, the Advanced Technology Microwave Sounder (ATMS) onboard Suomi-NPP is the most advanced cross track radiometer with 22 channels, 5 of which in the 183 GHz oxygen absorption band. On the other hand, CloudSat carries the W-band (94GHz) Cloud Profiling Radar (CPR) that has collected data since its 2006 launch. While CPR was designed as primarily a cloud remote sensing mission, its high-latitude coverage (up to ;82° latitude) and hi gh radar sensitivity (~-28dBZ) make it very suitable for snowfall-related research. We will show the results of a study where CPR is used to: 1) assess snowfall detection and estimate capabilities of DPR; 2) analyze snowfall signatures in the high frequency channels of the passive microwave radiometers in relation to fundamental environmental conditions. A number of global datasets made of coincident observations of snowfall producing clouds from the spaceborne radars DPR and CPR and from the most advanced radiometers available (GMI and ATMS) are analyzed. We have assessed the snowfall detection and estimation capabilities of DPR, comparing its observations and precipitation products with those available from CPR. We have estimated that DPR radars miss a very large fraction of snowfall precipitation (more than 90% of the events and around 70% of the precipitating snowfall mass). This is due mostly to the sensitivity limits of the DPR radar and secondly to the effect of the DPR radar side lobe clutter. An algorithm that combines the measured reflectivities from the two Ku-band and Ka-band radars exploiting the weak signals related to snowfall has been developed. Results from this study will be presented, showing improved DPR detection capabilities up to more than 50% of the snowfall mass obtained with the newly developed algorithm. Moreover the coincident observations of ATMS - CPR and of the GMI - DPR have been analyzed in order to study the multichannel brightness temperature signal related to snowfall. The main results of this study show that the high frequency channels (and the 183 GHz band channels in particular) can be successfully used in order to identify and quantify snowfall. The degree of success strongly depends on the type of surface background which requires proper detection and identification. Moreover, some ancillary data must be used (i.e. the columnar water vapor content is of paramount importance) for the correct use of the measurements towards snowfall detection. In this context an algorithm for surface classification of snow over land and ice over ocean using primarily the PMW low frequency channels is proposed and will be presented.