SPATIOTEMPORAL STORM STRUCTURE AND SCALING PROPERTY ANALYSIS FOR MODELING

Taylor's hypothesis and the multiscaling properties of British rainfal l data are analyzed and Poisson-based cluster models are examined for their ability to reproduce these features, Taylor's hypothesis is foun d to hold for timescales of up to 30-40 min, while at this scale a bre ak is also observed in the temporal scaling properties of moments of s everal orders, The first property is observed for the spatiotemporal m odel, while some of the scaling features of tile time series are repro duced by the point rainfall model, thus demonstrating the ability of h ierarchical models to reproduce self-similar properties of the data.