Understanding the Uncertainty of Satellite Passive Microwave Precipitation Products (Invited)

Over the past four decades, satellite passive microwave (PMW) observations have played a critical role in understanding the global rainfall – a trend set to continue. The long-lasting effort in retrieval development, with continuously maintained sensor constellation, ensured the availability of highly-accurate PMW estimates of global mean precipitation rate. This accuracy, however, does not apply when the same PMW products are combined and used at regional or seasonal scales. The inconsistency in sensors’ systematic errors, product postprocessing, and the resolution at which the combined sensor data are considered result in rather difficult to untangle uncertainty propagation. Nevertheless, high demand for quantified errors of precipitation products remains. The present work reviews some known uncertainty sources in satellite PMW precipitation products and offers insights into their quantitative evaluation. Uncertainty origins relevant to a wide range of satellite datasets –including machine learning-based Level-2 and multi-sensor Level-3 products– will be considered, with a discussion focusing on deriving their estimates.