Validation of the Version 05 Level 2 precipitation products from the GPM Core Observatory and constellation satellite sensors. (4th December 2017)

• Record Type: Journal Article
• Authors: Kidd, Christopher; Tan, Jackson; Kirstetter, Pierre‐Emmanuel; Petersen, Walter A.
• Is Part Of: Quarterly journal of the Royal Meteorological Society. Volume 144(2018)Supplement 1
• Page Start: 313
• Page End: 328
• Publication Date: 2017-12-04
• Subjects: Global Precipitation Measurement -- GPROF -- intercomparisons -- precipitation -- satellite -- validation; Meteorology -- Periodicals; 551.5
• DOI: 10.1002/qj.3175
• Languages: English
• ISSNs: 0035-9009

Abstract

Abstract : The accurate measurement of precipitation, an essential source of fresh water, is key for assessing water resources and precipitation‐related hazards across the globe. The measurement of global precipitation is hampered by the inadequate spatial distribution of ground‐based observations such as gauges and weather radars. However, satellite instruments can provide global observations from which precipitation estimates may be generated. The satellites of the Global Precipitation Measurement (GPM) constellation carry passive microwave sensors that provide an essential set of observations from which the Goddard PROFiling (GPROF 2017v1) scheme is used to retrieve precipitation. The regional consistency and differences in the satellite retrievals need to be documented in order to refine and improve global precipitation estimates which form the basis of widely used gridded precipitation products. Here the Level 2 instantaneous swath‐based precipitation products generated by the GPROF scheme are evaluated using standard descriptive and statistical scores against national/international surface radar and dense gauge datasets, over the United States and western Europe. Results show that over Europe the current GPROF retrieval technique tends to overestimate the occurrence of light precipitation, leading to an overestimation of the volumetric contribution by light precipitation intensities, while it underestimates moderate to heavy precipitation. Over the USA the overestimation of light precipitation is reduced, with a more pronounced overestimation of moderate precipitation intensities, and an underestimation of heavier precipitation intensities. Abstract : Extent of radar coverage used in this study for (a) western Europe and (b) eastern United States. The light grey‐shaded region represents the areas excluded from the analysis to ensure good conditions for comparison. The locations of the well‐calibrated dense gauge networks are also shown for Pocomoke (PCMK) and WegenerNet (WEGN).