Spatial variability of dry spells duration statistical distributions

Abstract

Dry spells duration and its extent in space, is a key factor in water resources problems. In order to modelling the empirical distribution of dry spells (DS) frequencies observed in Sicily (i.e. in a typical Mediterranean climate), Agnese et al. (2014) successfully applied the two-parameter polylogarithm-series distribution. Because of the strong seasonality characterising Sicily’s rainfall regime, statistical analysis was separately applied to two data sets, referred to as “dry” and “wet” seasons, respectively. In this work, a similar analysis was carried out for a set of 26 DS time-series recorded in a large area (about 30000 km2), including Piedmont and the Aosta Valley. Area altitude ranges between 100 to almost 5000 m a.s.l., and the Mediterranean sea is at about 20 km distance. Main climate character in the whole area is colder and more humid than that of Sicily, and it varies noticeably within the area itself, likely because of the complex role of geomorphology on rain and snow occurrence. Also, partition between rain and snow is evaluated. A subdivision in homogeneous subareas is found which is related to the main weather types occurring. In this work the focus is on the longer dry spells.