 This study compared the performance of various models used to predict runoff hydrographs in ungaged catchments. It found that humid catchments were more accurately predicted than arid ones, and larger catchments were more accurately predicted than smaller ones. Additionally, the performance of the models varied depending on the region and the method used to regionalize them. Spatial proximity and similarity methods performed best in humid catchments, while geostatistical methods performed better in arid catchments. Furthermore, dense networks of gauges provided better performance than sparse networks. Finally, the number of regionalized model parameters did not have a significant impact on the accuracy of the prediction. This article was authored by Jay Parajka, Aviglioni, M. Rogger, and others.