Evaluation of a compound distribution based on weather pattern subsampling for extreme rainfall in Norway

Abstract

Abstract. Simulation methods for design flood analyses require estimates of extreme precipitation for simulating maximum discharges. This article evaluates the multi-exponential weather pattern (MEWP) model, a compound model based on weather pattern classification, seasonal splitting and exponential distributions, for its suitability for use in Norway. The MEWP model is the probabilistic rainfall model used in the SCHADEX method for extreme flood estimation. Regional scores of evaluation are used in a split sample framework to compare the MEWP distribution with more general heavytailed distributions, in this case the Multi Generalized Pareto Weather Pattern (MGPWP) distribution. The analysis shows the clear benefit obtained from seasonal and weather patternbased subsampling for extreme value estimation. The MEWP distribution is found to have an overall better performance as compared with the MGPWP, which tends to overfit the data and lacks robustness. Finally, we take advantage of the split sample framework to present evidence for an increase in extreme rainfall in the southwestern part of Norway during the period 1979-2009, relative to 1948-1978.

Regional frequency analysis of extreme rainfall in Belgium based on radar estimates

“…The EXP distribution is preferred for short periods since estimating the shape parameter is very uncertain. Blanchet et al (2015) found that GPD fails to robustly estimate the tail of the distribution because of a lack of data and unrealistic return levels for very long return periods (when the shape parameter is positive). An additional argument for the EXP model is that it is less affected by observational errors, which plays an important role here.…”

An objective cross-validation framework for mapping rainfall hazard based on rain gauge data

“…The scores relating the tail of the distribution have been proposed and used in Garavaglia et al (2011);Renard et al (2013); Blanchet et al (2015). In the split sample evaluation framework, four scores can be derived of a given score Sc: Sc (12) is the regional score when G , Sc (11) and Sc (22) are obtained symmetrically.…”

“…Thus Blanchet et al (2015) proposed to base comparison on the divergence of the density of the ff (12) to the uniform density. A reasonable estimate of the latter is obtained by computing the 12) show mainly the three above alternatives: either a good fit (flat histogram), or a tendency towards a systematic under-or over-estimation (left-or right-skewed histograms).…”

“…Furthermore, we impose the season-at-risk to be the same for all the stations due to the little extent of the region, defining the season-at-risk as the three bigger regions would be to select the months composing the season-at-risk following the procedure described in Blanchet et al (2015).…”

“…Although the mixture of distributions over weather patterns has revealed efficient in other countries (e.g. in Norway and Canada, Brigode et al, 2014;Blanchet et al, 2015), it might be less relevant in, e.g., monsoon climate regions where the consideration of seasons seems Figure 10. Map of the probability of daily rainfall to exceed 1mm and of the mean of nonzero rainfall in the three WPs of the season-at-risk.…”