Statistical Post-Processing with Standardized Anomalies Based on a 1 km Gridded Analysis

Abstract

Statistical post-processing is necessary to correct systematic errors of numerical weather prediction models, especially in complex terrains such as the Alps. However, this post-processing is usually applied on every grid point individually, which can be computationally expensive. We want to present a method to forecast all grid points of a certain region simultaneously to expedite operational forecast times. The presented post-processing is part of the project SAPHIR, which provides forecasts from nowcasting up to +72 hours lead time with the same spatial resolution as the analysis. The used analysis is the Integrated Nowcasting through Comprehensive Analysis (INCA) system provided by ZAMG with a spatial resolution of 1 km. The post-processed variables are temperature, precipitation, wind and relative humidity. As a result highly resolved forecasts are presented with a similar performance to station-based forecasts.