Evaluation of Snowfall Rate Observations from Gauge, Ground Radar, Spaceborne Radar, and Radiometer over the Contiguous United States

This study compares the hourly snowfall rate from gauges operated by National Oceanic and Atmospheric Administration’s (NOAA) Climate Prediction Center (CPC), the Global Precipitation Measurement Mission’s (GPM) dual-frequency precipitation radar (DPR), GPM microwave imager (GMI), and Multiple Radar Multiple Sensors (MRMS) ground radar during the 2014-2015 cold season. The overall performance of each instrument was evaluated using the probability of detection, false alarm rate, correlation, root mean square error, and bias with gauge observations as a reference. Snowfall rates were then separated into categories based on location, and environmental conditions such as wind speed and air temperature. The results show that overall the Ka-band of the DPR has the lowest false alarm rate, while the MRMS has the highest rate of detection. Snowfall rates from DPR, GMI, and MRMS agree better with gauge observations over the eastern half (East of 100°W) of the Contiguous United States (CONUS) in comparison to the western half, as indicated by the larger correlation values, smaller root-mean- square errors, and lower biases for the eastern region. The larger discrepancy over the western CONUS is most likely caused by the severe terrain blockage in that region. Observed snowfall rates also showed better accuracy in warmer environments with calmer winds.