Flexible and consistent quantile estimation for intensity–duration–frequency curves

Abstract. Assessing the relationship between intensity, duration, and frequency (IDF) of extreme precipitation is required for design water management systems. However, when modeling sub-daily extremes, there are commonly only short observation time series available. This problem can be overcome by applying duration-dependent formulation generalized value (GEV) distribution which fits an IDF model with a range durations simultaneously. The originally proposed GEV exhibits power-law-like behavior quantiles takes care deviation from this scaling relation (curvature) sub-hourly (Koutsoyiannis et al., 1998). We suggest that more flexible might to wide (1 min 5 d). Therefore, we extend following two features: (i) different slopes (multiscaling) (ii) power law large (flattening), newly introduced in study. Based on quantile skill score, investigate performance resulting respect benefit individual features (curvature, multiscaling, flattening) simulated empirical data. provide detailed information duration probability ranges specific or systematic combination leads improvements stations case study area Wupper catchment (Germany). Our results show allowing curvature multiscaling improves very long durations, respectively, but disadvantages other ranges. In contrast, flattening average improvement medium 1 h d, without affecting regimes. Overall, new parametric form offers enhanced consistently describing relations over has not been done before as most existing studies focus longer than day do address (2–5