Christine Achberger

Projecting Future Local Precipitation and Its Extremes for Sweden

Abstract A procedure to obtain future local precipitation characteristics focused on extreme conditions has been developed based on a weather generator. The method involves six major steps: (1) the weather generator was calibrated using observed daily precipitation at 220 wedish stations during 1961–2004; (2) present and future daily precipitation characteristics for the wedish stations from two global climate models, namely and HadCM3, were used to calculate weather generator parameters for the present and future climates at global climate model spatial scales; (3) the ratio of the weather generator parameters for the present climate simulated by the global climate models to those calculated for each station falling into the global climate model grid box were computed for all the stations; (4) these ratios were also assumed to be valid in the future climate, that way the future parameters for each station for the global climate model projected future climate could be calculated; (5) using the estimated future parameters of the weather generator, the future daily precipitation at each station could be simulated by the weather generator; (6) the simulated daily precipitation was used to compute eight indices describing mean and extreme precipitation climates. The future mean and extreme precipitation characteristics at the stations under the Second Report on Emission Scenarios A2 scenario were obtained and presented. An overall increasing trend for frequency and intensity of the indices are identified for the majority of the stations studied. The developed downscaling methodology is relatively simple but useful in deriving local precipitation changes, including changes in the precipitation extremes.

Characterizing and visualizing spatio-temporal patterns in hourly precipitation records

We develop new techniques to summarise and visualise spatial patterns of coincidence in weather events such as more or less heavy precipitation at a network of meteorological stations. The cosine similarity measure, which has a simple probabilistic interpretation for vectors of binary data, is generalised to characterise spatial dependencies of events that may reach different stations with a variable time lag. More specifically, we reduce such patterns into three parameters (dominant time lag, maximum cross-similarity, and window-maximum similarity) that can easily be computed for each pair of stations in a network. Furthermore, we visualise such three-parameter summaries by using colour-coded maps of dependencies to a given reference station and distance-decay plots for the entire network. Applications to hourly precipitation data from a network of 93 stations in Sweden illustrate how this method can be used to explore spatial patterns in the temporal synchrony of precipitation events.