Non-stationary extreme value analysis in a changing climate

Cheng, Linyin; AghaKouchak, Amir; Gilleland, Eric; Katz, Richard W.

doi:10.1007/s10584-014-1254-5

Cited by 426 publications

(304 citation statements)

References 51 publications

Supporting

Mentioning

301

Contrasting

Unclassified

Order By: Relevance

“…This information can be derived from long-term records of weather quantities such as precipitation and temperature by means of statistical extreme value modeling. While extreme value theory provides a methodological framework that is commonly used in various scientific disciplines, such as hydrology (Katz et al, 2002), finance (Embrechts et al, 2003), engineering (Castillo et al, 2005) and climate sciences (Katz, 2010;Cheng et al, 2014), the application of these tools for road network exposure analysis is a relatively uncharted area. In particular, formal comparative assessments of the various statistical methods that can be applied for estimating return levels of extreme events are rare.…”

Section: Introductionmentioning

confidence: 99%

Extreme weather exposure identification for road networks – a comparative assessment of statistical methods

Schlögl

Laaha

2017

Nat. Hazards Earth Syst. Sci.

View full text Add to dashboard Cite

Abstract. The assessment of road infrastructure exposure to extreme weather events is of major importance for scientists and practitioners alike. In this study, we compare the different extreme value approaches and fitting methods with respect to their value for assessing the exposure of transport networks to extreme precipitation and temperature impacts. Based on an Austrian data set from 25 meteorological stations representing diverse meteorological conditions, we assess the added value of partial duration series (PDS) over the standardly used annual maxima series (AMS) in order to give recommendations for performing extreme value statistics of meteorological hazards. Results show the merits of the robust L-moment estimation, which yielded better results than maximum likelihood estimation in 62 % of all cases. At the same time, results question the general assumption of the threshold excess approach (employing PDS) being superior to the block maxima approach (employing AMS) due to information gain. For low return periods (non-extreme events) the PDS approach tends to overestimate return levels as compared to the AMS approach, whereas an opposite behavior was found for high return levels (extreme events). In extreme cases, an inappropriate threshold was shown to lead to considerable biases that may outperform the possible gain of information from including additional extreme events by far. This effect was visible from neither the square-root criterion nor standardly used graphical diagnosis (mean residual life plot) but rather from a direct comparison of AMS and PDS in combined quantile plots. We therefore recommend performing AMS and PDS approaches simultaneously in order to select the best-suited approach. This will make the analyses more robust, not only in cases where threshold selection and dependency introduces biases to the PDS approach but also in cases where the AMS contains non-extreme events that may introduce similar biases. For assessing the performance of extreme events we recommend the use of conditional performance measures that focus on rare events only in addition to standardly used unconditional indicators. The findings of the study directly address road and traffic management but can be transferred to a range of other environmental variables including meteorological and hydrological quantities.

show abstract

Section: Introductionmentioning

confidence: 99%

Extreme weather exposure identification for road networks – a comparative assessment of statistical methods

Schlögl

Laaha

2017

Nat. Hazards Earth Syst. Sci.

View full text Add to dashboard Cite

show abstract

“…Conditioning on a single chosen model was criticized by (Draper 1995;Madigan and Raftery 1994), since it ignores model selection uncertainty and therefore leads to an underestimation of the uncertainty of quantities. Basic issues related to combining discriminatory and regression models are given by Burnham and (Anderson 2002;Gatnar 2008, and for FFA models by Bogdanowicz 2010; Markiewicz et al 2015). Moradkhani (2015, 2016) proposed a multimodel …”

Section: Multimodel Approach To Seasonal Maxima Distributionsmentioning

confidence: 99%

“…We refer to them in our second paper in this issue entitled ''Around and about an application of the GAMLSS package in non-stationary flood frequency analysis''. Here we propose a methodology to take account of non-stationary flood flows and we limit ourselves to three methods and their comparison, although others are possible, e.g., quantile regression methods (Koenker 2005) or Bayesian hierarchical estimation (Cheng et al 2014;Moradkhani 2015, 2016).…”

Section: Introductionmentioning

confidence: 99%

A comparison of three approaches to non-stationary flood frequency analysis

2017

View full text Add to dashboard Cite

Non-stationary flood frequency analysis (FFA) is applied to statistical analysis of seasonal flow maxima from Polish and Norwegian catchments. Three non-stationary estimation methods, namely, maximum likelihood (ML), two stage (WLS/TS) and GAMLSS (generalized additive model for location, scale and shape parameters), are compared in the context of capturing the effect of non-stationarity on the estimation of time-dependent moments and design quantiles. The use of a multimodel approach is recommended, to reduce the errors due to the model misspecification in the magnitude of quantiles. The results of calculations based on observed seasonal daily flow maxima and computer simulation experiments showed that GAMLSS gave the best results with respect to the relative bias and root mean square error in the estimates of trend in the standard deviation and the constant shape parameter, while WLS/TS provided better accuracy in the estimates of trend in the mean value. Within three compared methods the WLS/TS method is recommended to deal with nonstationarity in short time series. Some practical aspects of the GAMLSS package application are also presented. The detailed discussion of general issues related to consequences of climate change in the FFA is presented in the second part of the article entitled ''Around and about an application of the GAMLSS package in non-stationary flood frequency analysis''.Keywords Non-stationary flood frequency analysis Á Maximum likelihood Á Two-stage methodology Á GAMLSS Á Multimodel approach Á Monte Carlo simulations

show abstract

“…Extreme precipitation modelled through EVT usually describes amounts that are far out in the tail of the distribution and associated with low probability, and the estimates may change when new extremes are sampled. Most uses of EVT also assume stationarity, although there are ways to account for trends (Cheng et al, 2014).…”

Section: Introductionmentioning

confidence: 99%

Simple and approximate estimations of future precipitation return values

Benestad

Parding

Mezghani

et al. 2017

Nat. Hazards Earth Syst. Sci.

View full text Add to dashboard Cite

Abstract. We present estimates of future 20-year return values for 24 h precipitation based on multi-model ensembles of temperature projections and a crude method to quantify how warmer conditions may influence precipitation intensity. Our results suggest an increase by as much as 40-50 % projected for 2100 for a number of locations in Europe, assuming the high Representative Concentration Pathway (RCP) 8.5 emission scenario. The new strategy was based on combining physical understandings with the limited information available, and it utilised the covariance between the mean seasonal variations in precipitation intensity and the North Atlantic saturation vapour pressure. Rather than estimating the expected values and interannual variability, we tried to estimate an "upper bound" for the response in the precipitation intensity based on the assumption that the seasonal variations in the precipitation intensity are caused by the seasonal variations in temperature. Return values were subsequently derived from the estimated precipitation intensity through a simple and approximate scheme that combined the 1-year 24 h precipitation return values and downscaled annual wet-day mean precipitation for a 20-year event. The latter was based on the 95th percentile of a multi-model ensemble spread of downscaled climate model results. We found geographical variations in the shape of the seasonal cycle of the wet-day mean precipitation which suggest that different rain-producing mechanisms dominate in different regions. These differences indicate that the simple method used here to estimate the response of precipitation intensity to temperature was more appropriate for convective precipitation than for orographic rainfall.

show abstract

Non-stationary extreme value analysis in a changing climate

Cited by 426 publications

References 51 publications

Extreme weather exposure identification for road networks – a comparative assessment of statistical methods

Extreme weather exposure identification for road networks – a comparative assessment of statistical methods

A comparison of three approaches to non-stationary flood frequency analysis

Simple and approximate estimations of future precipitation return values

Contact Info

Product

Resources

About