The climate change context has raised new problems in the computation of temperature return levels (RLs) in using the statistical extreme value theory. This arises since it is not yet possible to accept the hypothesis that the series of maxima or of high level values are stationary, without at least verifying the assumption. Thus, in this paper, different approaches are tested and compared to derive high order RLs in the nonstationary context. These RLs are computed by extrapolating identified trends, and a bootstrap method is used to estimate confidence intervals. The identification of trends can be made either in the parameters of the extreme value distributions or in the mean and variance of the whole series. Then, a methodology is proposed to test if the trends in extremes can be explained by the trends in mean and variance of the whole dataset. If this is the case, the future extremes can be derived from the stationary extremes of the centered and normalized variable and the changes in mean and variance of the whole dataset. The RL can then be estimated as nonstationary or as stationary for fixed future periods. The work is done for both extreme value methods: block maxima and peak over threshold, and will be illustrated with the example of a long observation time series for daily maximum temperature in France. Copyright © 2010 John Wiley & Sons, Ltd.
environmetrics; poisson; john wiley sons; copyright; parey; scale parameter; nonstationary; exceedance; variance; deols; hoang; bootstrap; location parameter; extrapolating; standard deviation; dataset; temperature return levels; different ways; poisson process; block maxima; nonstationary context; whole dataset; stationary; stationary context; poisson process intensity; block maxima method; future period; nonparametric; parameter; future periods; extreme value distributions; extreme value distribution parameters; shape parameter; optimal trend; right panel; linear trends; second part; climate model simulations; methodology; high number; threshold exceedances; extreme value methods; other hand; bootstrap intervals; whole series; simulation; sample effects; bootstrap technique; temperature series; block length; standard deviation changes; such distances; different approaches; climate change; high temperatures; observation series; seasonal cycle; optimal polynomial trend; similar results; high values; stationary case; whole sample; maximum temperature; climate simulations; klein tank; extrapolation; delta method; extreme values; scale parameters; optimal trends; high threshold; stationary extremes; quai watier; poisson distribution; probability theory; parametric trend; block index; stationary distribution; time evolutions; climatic data; extreme value distribution; independent values; european climate assessment; chatou cedex; wind speed; identical distribution; annual maxima; quadratic trend; summer maxima; block maxima evolution; threshold selection; return level; rare events; value theory; declustering procedure; edge effects; independent exceedances; observation period; pareto distribution; linear trend; mathematical expectation; optimization procedure; bootstrap samples; deols series; whole data series; mathematical tools; maximum temperatures; climate model results; climate models; nearest grid point; extreme events; model values; warm bias; current period; variance evolution; variance evolutions; same procedure