Which rainfall metric is more informative about the flood simulation performance? A comprehensive assessment on 1318 basins over Europe

The global availability of satellite rainfall products (SRPs) at an increasingly high temporal/spatial resolution has made possible their exploitation in hydrological applications, especially over in-situ data scarce regions. In this context, understand how uncertainties transfer from SRPs into flood simulation, through the hydrological model, is a main research question. SRPs accuracy is normally characterized by comparing them with ground observations via the calculation of categorical (e.g., threat score, false alarm ratio, probability of detection) and/or continuous (e.g., bias, root mean square error, Nash-Sutcliffe index, Kling-Gupta efficiency index, correlation coefficient) metrics. However, whether these metrics are informative about the associated performance in flood simulations (when the SRP is used as input to an hydrological model) is an underdiscussed research topic. This study aims to relate the accuracy of different SRPs both in terms of rainfall and in terms of flood simulation. That is, the following research question are addressed: is (are) there appropriate performance metric (s) to drive the choice of the best performing rainfall product for flood simulation? To answer this question three SRPs, namely the Tropical Rainfall Measurement Mission Multi-satellite Precipitation Analysis, TMPA; the Climate Prediction Center Morphing algorithm, CMORPH, and the SM2RAIN algorithm applied to the ASCAT (Advanced SCATterometer) soil moisture product, SM2RAIN-ASCAT, have been used as input into a lumped hydrologic model (MISDc, "Modello Idrologico Semi-Distribuito in continuo") on 1318 basins over Europe with different physiographic characteristics. Results have suggested that, among the continuous metrics, correlation coefficient and Kling-Gupta efficiency index are not reliable scores to select rainfall product performing best for hydrological modelling whereas bias and root mean square error seem more appropriate. In particular, by constraining the relative bias to values lower than 0.2 and the relative root mean square error to values lower than 2, good hydrological performances (Kling-Gupta efficiency index on discharge greater than 0.5) are ensured for almost 75 % of the basins fulfilling these criteria. Conversely, the categorical scores have not provided suitable information to address the SRPs selection for hydrological modelling.