A Bayesian framework to derive consistent intensity-duration-frequency curves from multiple data sources

As a warming climate leads to more frequent heavy rainfall, the importance of accurate rainfall statistics is increasing. Rainfall statistics are often presented as intensity-duration-frequency (IDF) curves showing the rainfall intensity (return level) that can be expected at a location for a duration, and the frequency of this intensity (return period). IDF curves are commonly constructed by fitting generalized extreme value (GEV) distributions to observed annual maximum rainfall for several target durations, where the available observation data sources may vary for the different durations. As the estimation is performed independently across durations, the resulting IDF curves may be inconsistent across durations and return periods. We discuss how consistent estimates across the different durations may be derived by post-processing independently obtained Bayesian posterior distributions for each duration. The proposed methods are evaluated for simulated data and for Norwegian rainfall data from 83 locations, for 16 durations between 1 minute and 24 hours, where the post-processing yields consistent and accurate estimates.