Improving satellite precipitation detection using Global Precipitation Measurement mission (GPM) and Ground Based Radar (NMQ)

Global Precipitation Measurement (GPM) Mission is a NASA/JAXA core observatory satellite which launched on February 28th, 2014 and has since been collecting precipitation data worldwide. As a harbinger of wider global precipitation measurement accuracy, the GPM follows in the footsteps of NASA's Tropical Rainfall Measurement Mission (TRMM), but with greater latitudinal range and higher precipitation detection precision. During GPM's infancy, we must ask: how accurate is the retrieval product that we are receiving? This question is answered by correlating rainfall data from the first four months (March 4th-July 7th) in GPM's operation with the National Mosaic & Multi-sensor Quantitative Precipitation Estimate (NMQ) on ground radar detection network. This project aims to identify inconsistency between the two sources: satellite, and ground-based radar. Both datasets, GMI (GPM Microwave Imager) and NMQ, are constrained to sift out missing and poor quality data—the datasets are then averaged over an identical footprint at the pixel level and weighted with a Gaussian distribution. After being matched up, the data points are plotted against one another and found to have a correlation coefficient of .64 (with ideal correlation of 1.0) over 120 days (March 4th to July 7th). Time series analysis of the rain rate correlation between data sets shows an increasing rate of consistency over time (towards summer months). Specific rain rates of 0.0-2.5, 2.5-7.6, and 7.6-10.0 (mm/hr) are plotted, elucidating areas of inconsistency going from mixed phase and light rain to summertime heavy convective precipitation. A link between this result and the seasonal variability is observed. Ice particle formation in the upper structure of clouds, associated with heavier rainfall, is well detected by GMI and thus —as precipitation becomes heavier, the GMI Microwave Imager functions in better accord with NMQ. This conclusion highlights problematic light rainfall data and will thrust further the task of providing accurate precipitation documentation to areas of the globe with inadequate radar and rain gauge capability.