Temperature and precipitation climatologies for Italy and downscaling of the italian secular records onto a high-resolution grid

Italy has an exceptional availability of secular temperature and precipitation data. In this presentation, we discuss a methodology, based on the anomaly method, to project these records onto a high-resolution grid. The key issue of this methodology is the construction of thirty-arc-second resolution monthly temperature and precipitation climatologies [1][2]. These climatologies are obtained by means of a procedure which consists in considering each cell of a high-resolution grid and in using the neighboring stations with geographical characteristics as much similar as possible to those of the cell itself in order to perform a local weighted linear regression (LWLR) of temperature or precipitation versus elevation: the weights are given by the distances of the stations from the grid point and by the level of similarity between the stations cells and the considered grid cell in terms of elevation, slope steepness, slope orientation and distance from the sea. The model performances are compared with the results provided by other widely used interpolation methods, highlighting the best agreement of LWLR method with the observed station normals and a lower leave-one-out-estimated mean absolute error. The better performances of LWLR are even clearer when specific clusters of stations are considered, e.g. high elevation sites. The temporal component to be superimposed to climatologies can be obtained, for each grid point, by means of a distance-weighted average of the anomaly temperature and precipitation records of the neighboring stations. Some case studies of the potentiality of this approach will be presented.