The team from WP2 has finished Deliverable 2.9. Maps of Intensity-durantion-frequency (IDF) Curves for Extreme Precipitation for the BINGO Research Sites.
Different types of precipitation events can have different effects, depending on several factors. Two important factors are the intensity of the event and its duration. To plan and manage risk, we need to have good predictions of the frequency of extreme events with a specific intensity and duration. Intensity-duration-frequency (IDF) curves are a tool to estimate and visualize these relations. Fig. 1 shows the IDF curve for annual maximum precipitation at the station Saedalen at the Research Site Bergen. The different colours are related to the expected frequency of events, e.g. 2yr return-level (50% exceedance probability, red), 50yr return-level (2% exceedance probability, green) and 100yr return-level (1% exceedance probability, blue). As smoothness of this relation across various durations is generally a good assumption, we estimate consistent IDF curves using a duration-dependent Generalized Extreme Value Distribution. The resulting IDF curves (coloured lines) are considered more reliable than frequencies obtained from only a single duration.
As the smoothness argument is expected to hold also in space, we extended this approach and set up a statistical model for “spatial IDF curves”. Fig. 2 shows an example for the Research Site Wupper. As different return-levels (exceedance probabilities) and durations cannot be visualized simultaneously on a map, we show exemplarily the 100yr return-level for 1h events (left) and 3day events(right).
You can find all the deliverables here.