Classification of daily precipitation patterns on the basis of radar-derived precipitation rates for Saxony, Germany
Abstract:We present a radar-based climatology of precipitation fields summarised into characteristic daily precipitation patterns. These patterns were derived by temporal classification, applying a neural network and data from Saxony during the period from 2004 to 2010. The properties of the dataset (RADOLAN rw-product) are discussed in detail and reviewed with respect to their adequacy for the intended application. The analysis showed a systematic dependence of the precipitation error on the altitude and aggregation period. Accordingly, for future applications of the considered radar product, we recommend the use of a maximal aggregation time step of 24 hours. The classification reveals significant precipitation patterns. Comparison of the qualitative features exhibited by the precipitation patterns, such as the synoptic scale flow direction, pressure distribution and atmospheric humidity, showed general trends as well as distinct spatial and atmospheric properties in dependence of the incidence rate. The lowest statistical qualities were shown by the patterns with the most distinct spatial characteristics due to a low incidence rate and high standard deviations. Nevertheless, the applied method led to a robust classification and the derived patterns appropriately summarized the mean daily precipitation behaviour in Saxony.
Document Type: Research Article
Publication date: October 1, 2012
More about this publication?
- Meteorologische Zeitschrift (originally founded in 1866) is the joint periodical of the meteorological societies of Austria, Germany and Switzerland. It accepts high-quality peer-reviewed manuscripts on all aspects of observational, theoretical and computational research out of the entire field of meteorology, including climatology. Meteorologische Zeitschrift represents a natural forum for the meteorological community of Central Europe and worldwide.
- Editorial Board
- Information for Authors
- Submit a Paper
- Subscribe to this Title
- Ingenta Connect is not responsible for the content or availability of external websites