Quantifying of the Effect of Rain-Induced Sub-Footprint Scale Wind Variability on the Rapidscat Ku-Band Nrcs

The recent advances in Ku-band wind speed retrievals [Fore, 2014] in the presence of rain [Stiles and Dunbar, 2010] indicate that there is an opportunity to learn more about the physical relationship between the spatial variation of surface winds within a scatterometer resolution cell (wind vector cell; WVC) and the observed backscatter. There is a need for improvement in the ability of this type of scatterometer to estimate wind speed and direction, in the presence of rain. The wind vector products that are currently available, derived using these approaches and without a separate bias adjustment such as Stiles et al. [2014], often have relatively large errors in speed and direction [Bourassa, 2017]. In effect, the presence of rain frequently leads to erroneous solutions by these methods. This effort descried herein is engaged with data from the RapidScat scatterometer mission on the International Space Station. We are assessing the sub-footprint variability in wind and how rain contributes to the area averaged backscatter. The sub-footprint winds are from a new high-resolution wind vector product (hourly, 2.5 km spacing), for which data assimilation of NEXRAD observations should enhance the consistency with sub-footprint variability in rain contributions to the radar cross section. This product is the Real Time Mesoscale Analysis, from NOAA-NCEP. We are comparing the PO.DAAC NRCS products (after adjustment for rain) to the values estimated from the higher resolution wind product, transferred to backscatter, through the RapidScat geophysical model function (GMF). The rain data is available through a special RScat “RadRain” product created by Remote Sensing Systems using a variety of collocated, coincident satellite sensors.