1Development of Daily Precipitation Scenarios at KNMI

The use of statistical relationships is a noticeable feature of daily precipitation scenarios produced at KNMI. Recently, atmospheric moisture has been included as predictor variable. Separate models are used to describe rainfall occurrence (probability of a wet day) and the wet-day rainfall amounts. Rela-tionships for daily rainfall at Berne (Switzerland) are presented using atmospheric moisture and circu-lation variables from the NCEP re-analysis data. Both perturbation of the observed record and sto-chastic time series simulation is examined as methods of scenario production. An application is given with data from a time-dependent greenhouse gas forcing experiment with the ECHAM4/OPYC3 model for the periods 1968-1997 and 2070-2099. The estimated changes in the annual and seasonal mean rainfall amounts from the fitted statistical relationships are compared with those in the simulated rainfall of ECHAM4/OPYC3. It is shown that these changes are sensitive to the choice of the moisture variables in the relationship for the wet-day precipitation amounts. Besides the changes in the mean precipitation amounts, the 90th percentile of the distributions of the N-day annual maximum precipita-tion amounts is considered to assess daily precipitation scenarios. The representation of the coefficient of variation of the wet-day precipitation amounts strongly influences the quality of the reproduction of this extreme-value characteristic and its changes. 1