Accounting for surface ice and snow in the goddard profiling algorithm rain rate retrievals

A recurring challenge for microwave precipitation retrieval algorithms is the separation of precipitation from surface snow. Cold surfaces can have similar radiometric characteristics to suspended precipitation particles, such that surface snow reduces the detection accuracy of precipitation algorithms. Snow detection is typically performed using a collection of heritage algorithms, however surface snow can often go undetected and lead to erroneous precipitation retrievals. This study examines flagging procedures in the Goddard Profiling Algorithm 2010 V2 as it applies to the Advanced Scanning Microwave Radiometer 2 over the United States in the winter of 2014-2015. The legacy algorithms are often not reliable, such that a climatological screen is applied in known snowy regions. The diurnal cycle of brightness temperatures also complicates resolving rain when using the static legacy algorithms.