Estimating extreme areal precipitation in Norway from a gridded dataset

Abstract

To obtain estimates of extreme areal precipitation in Norway, the Norwegian Meteorological Institute currently applies a statistical method that combines measured point precipitation, empirical growth factors, and areal reduction factors. We here suggest performing statistical analysis directly on areal 24-h precipitation from a gridded dataset covering the period from 1957 to the present. Grid-based methods provide increased objectivity and consistency, and enable estimation in ungauged catchments. The proposed method fits the generalized extreme value (GEV) distribution to areal precipitation series in order to estimate precipitation return levels required for design values for flooding and dam safety. The study includes an investigation of the spatial variation of extreme precipitation in Norway, as reflected by the GEV shape parameter. Our results suggest that this parameter varies spatially according to the dominating precipitation systems and, most probably, to the degree of orographic enhancement.