Estimates of Point Rainfall Extremes from Satellite Precipitation Products: Application and Testing in Northeastern Italy

Reliable estimates of extreme precipitation are fundamental for understanding, predicting, and mitigating natural disasters. However, the inference of extreme precipitation magnitudes at the global scale is severely constrained by the low and uneven density of direct rainfall observations. Satellite-based rainfall estimates offer a promising source of information to support extreme value analysis but are hindered by high estimation uncertainty and coarse spatial resolutions. The coarse scale of global datasets, with grid sizes typically ranging from 100 to 600 km², prevents direct comparisons with point-scale extreme value estimates because point and area-averaged statistics inherently differ by construction.

This study addresses this limitation by systematically applying a downscaling method for extreme-value statistics based on the theory of random fields and the Metastatistical Extreme Value Distribution (MEVD). We utilize a large dataset from approximately 200 rain gauges in Northeastern Italy and multiple satellite precipitation products, including IMERG, CMORPH, CHIRPS, SM2RAIN, MSWEP, and PERSIANN. Downscaling, based on the autocorrelation structure of the precipitation fields, is performed for each individual product on the grid cells corresponding to the available rain gauges. 

We compare downscaled estimates of daily 50-year return period event magnitudes with those derived from rain gauge time series, for individual products as well as for central tendency statistics of the ensemble. Additionally, we quantify the frequency distribution of estimation errors associated with different products and with their ensemble.

This research was supported by the "raINfall exTremEs and their impacts: from the local to the National ScalE" (INTENSE) project, funded by the European Union - Next Generation EU in the framework of PRIN (Progetti di ricerca di Rilevante Interesse Nazionale) programme (grant 2022ZC2522).

MM was also supported by the RETURN Extended Partnership and received funding from the European Union Next-GenerationEU (National Recovery and Resilience Plan – NRRP, Mission 4, Component 2, Investment 1.3 – D.D. 1243 2/8/2022, PE0000005)