Precipitations signatures on Synthetic Aperture Radar imagery at X and Ka bands: detection and quantification

The use of spaceborne microwave synthetic aperture radars (SARs) is becoming a well-established tool in several Earth remote sensing disciplines, such as flood monitoring, earthquakes analysis, ground target classification and many others. Satellite SARs ensure high spatial resolution imaging (on the order of meters) in almost all-weather conditions and suitable coverage (especially new generation of SAR based missions). Indeed, for frequencies above C-band, the impact of precipitating clouds may significantly impair the signal backscattered from the ground (e.g. Ferrazzoli and Schiavon 1997) even if the probability is quite low (Danklmayer et al. 2009). Nonetheless, the impact of precipitating clouds on both amplitude and phase-compressed SAR signals cannot be neglected, as recently reassessed by using X-band SARs (X-SAR) currently in orbit, such as COSMO-SkyMed (CSK) and TerraSAR-X (TSX) (e.g. Marzano et al. 2010, Baldini et al. 2014). On the other side, this sensitivity paves the way to the use of SARs as an instrument for observing and quantifying atmospheric precipitations. This work aims at assessing the impacts of the atmospheric precipitations on X-SAR images and proposing innovative and enhanced X-SARs products, such as precipitation maps and cloud masks, exploiting the high spatial resolution, revisit time and coverage of last generation X-SARs. X-SARs give an unprecedented opportunity to final users (e.g. hydrologists) to inject observations of rain fields at catchment scale and at a spatial resolution as never before, into their models for flood forecasting. Nevertheless, research is still at an early stage and several issues have to be addressed. The developed algorithm allows precipitating areas to be distinguished by flooded and permanent water surfaces or wet snow absorption, all of them looking “dark” (low backscattering) in SAR images. The identification of such areas is critical and necessary both for floods analysis and precipitation retrieval, because misinterpretations could cause severe estimation errors in both fields. This pre-processing algorithm is mainly based on image segmentation techniques and fuzzy logic (e.g. Pulvirenti et al. 2014 and Mori et al. 2012). Ancillary data, such as a local incidence angle map and a land cover map are also used. The second step of the procedure consists of a precipitation retrieval algorithm, developed in previous works by Marzano et al. (2010, 2011), and up to now applied only to pixels where rain is known to be present. This methodology has been applied to 14 CSK and TSX study cases, acquired within the European FP7 project EartH2Observe over Italy and United States. These areas have been selected for the possibility of observing both hurricane-like intense events and continental mid-latitude ones. Moreover, they offer the opportunity to verify and validate the proposed methodology using weather radars or rain gauge networks. Results obtained until now show fairly good performances of the precipitation maps, both in terms of cell localization and quantification. A couple of examples are reported in the included file. The above investigations, both theoretical and experimental, suggest exploring the possibility offered by instruments operating at higher frequencies. In order to investigate the potential of Ka SARs, we have further developed the 2D numerical simulator of SAR images in presence of rain, described in (Marzano et al., 2012), to extend it to Ka-band and polarimetric observations (Mori et al. 2015). This has been performed in the frame of an ESA funded project [Contract ESTEC N. 4000109477/13/nl/lvh] supporting the feasibility analysis of a spaceborne Ka-SAR system. Indeed, Ka-band shows a very high sensitivity to atmospheric hydrometeors and this offers interesting possibilities towards multifrequency atmospheric precipitations retrieval. In this work simulated Ka-SAR numerical scenarios will be also discussed.