Consequences to flood management of using different probability distributions to estimate extreme rainfall

Abstract

The design of flood defences, such as pumping stations, takes into consideration the predicted return periods of extreme precipitation depths. Most commonly these are estimated by fitting the Generalised Extreme Value (GEV) or the Generalised Pareto (GP) probability distributions to the annual maxima series or to the partial duration series. In this paper, annual maxima series of precipitation depths obtained from daily rainfall data measured at three selected stations in southeast UK are analysed using a range of probability distributions. These analyses demonstrate that GEV or GP distributions do not always provide the best fit to the data, and that extreme rainfall estimates for long return periods (e.g. 1 in 100 years) can differ by more than 40% depending on the distribution model used. Since a large number of properties in the UK and elsewhere currently benefit from flood defences designed using the GEV or GP probability distributions, the results from this study question whether the level of protection they offer are appropriate in locations where data demonstrate clearly that alternative probability distributions may have a better fit to the local rainfall data. This work: (a) raises awareness of the limitations of common practices in extreme rainfall analysis; (b) suggests a simple way forward to incorporate uncertainties that is easily applicable to local rainfall data worldwide; and thus (c) contributes to improve flood risk management.

Assessing the Impact of Climate Change and Extreme Value Uncertainty to Extreme Flows across Great Britain

“…Methods based on AM or POT series have well-known advantages and disadvantages. For example, the GEV/GL methods generally use a smaller sample size than GP which can affect the uncertainty associated with the estimate and the efficiency of the method [26][27][28], whereas methods that use POT data have a greater sample on which to fit the model. As with the climate models, with any method of extreme value estimation comes a degree of uncertainty, which is associated with both the EV model parameter (i.e., the fitting or sampling error) and the model structure (i.e., the differences between AM approach with the GEV model and POT approach with the GP model).…”

A pan-South-America assessment of avoided exposure to dangerous extreme precipitation by limiting to 1.5 °C warming

“…The present study focuses on the annual maximum 5 d running mean precipitation (Rx5d, mm d −1 ), a frequently used index in flood risk assessments (Seneviratne et al 2012). In this study, dangerous extreme precipitation events are defined as those exceeding the 100 year return values of Rx5d from the historical baseline in each model, respectively, as many flood defenses are designed to withstand floods less severe than the 1 in 100 year event (Esteves 2013). Note that in our analysis, this index is computed as annual values, rather than distinguishing seasonal behaviors.…”

L-Moment-Based Frequency Analysis of High-Flow at the Sungai Langat, Kajang, Selangor, Malaysia

“…In order to perform flood frequency analysis, the L-moment method was used by estimating the annual maximum streamflow data. It was obtained by taking the largest value in each year of interest (Adamowski 2000;Esteves 2013). Beside the L-moment, there are several other methods of performing parameter estimation such as method of moment and maximum likelihood estimation.…”