Combining regional flood frequency models with geometric pooling: an application to the Po River District

This dataset contains the flood quantile distribution parameters underlying the regional flood frequency analysis of the Po River District (northern Italy), based on a harmonized dataset of annual maximum peak flows from 226 catchments. For each of the 226 sites and for each of nine return periods (T = 2, 5, 10, 20, 50, 100, 200, 500, 1000 years), the file provides the parameters of the predictive probability distribution of the flood quantile q_T estimated by three independent regional models, together with their geometric pool ("Geom Pool"): SSEM (Spatially Smooth Estimation Method, Laio et al. 2011) RoI (Region of Influence, index-flood approach) VAPI (updated North-Western VAPI method, De Michele et al. 2001) Geom Pool: geometric (logarithmic) pooling of the three models' predictive distributions, approximated by a three-parameter log-normal distribution Distributions:- SSEM, RoI, and Geom Pool are represented as three-parameter log-normal (generalized normal, version 2) distributions, with parameters xi (location), alpha (scale), and kappa (shape).- VAPI is represented as a normal distribution, with parameters mu and sigma. File structure:- Sheet 1: station code ("codice"), drainage area (km2), and coordinates of the catchment closing section for each of the 226 sites.- Remaining sheets: one sheet per return period and model, reporting, for every station, the distribution parameters and the corresponding quantiles at probability levels 0.025, 0.05, 0.1, 0.25, 0.5, 0.75, 0.9, 0.95, 0.975. This dataset supports the results presented in: Cafiero, L., Spadoni, A., Castellarin, A., Demichele, C., Laio, F., and Viglione, A., "Combining regional flood frequency models with geometric pooling: an application to the Po River District" (submitted to Hydrological Sciences Journal in 2026).