Abstract
Abstract. The use of radar for precipitation measurement in mountainous regions is complicated by many factors, especially beam shielding by terrain features, which, for example, reduces the visibility of the shallow precipitation systems during the cold season. When extrapolating the radar measurements aloft for quantitative precipitation estimation (QPE) at the ground, these must be corrected for the vertical change of the radar echo caused by the growth and transformation of precipitation. Building on the availability of polarimetric data and a hydrometeor classification algorithm, this work explores the potential of machine learning methods to study the vertical structure of precipitation in Switzerland and to propose a more localised vertical profile correction. It first establishes the ground work for the use of machine learning methods in this context: from volumetric data of 30 precipitation events, vertical cones with 500 m vertical resolution are extracted. It is shown that these cones can well represent the vertical structure of different types of precipitation events (stratiform, convective, snowfall). The reflectivity data and the hydrometeor proportions from the extracted cones constitute the input for the training of artificial neural networks (ANNs), which are used to predict the vertical change in reflectivity. Lower height levels are gradually removed in order to test the ANN's ability to extrapolate the radar measurements to the ground level. It is found that ANN models using the information on hydrometeor proportions can predict from altitudes between 500 and 1000 m higher than the ANN based on only reflectivity data. In comparison to more traditional vertical profile correction techniques, the ANNs show less prediction errors made from all height levels up to 4000 m a.s.l., above which the ANNs lose predictive skill and the performance levels off to a constant value.
Highlights
Precipitation constitutes a key meteorological variable for ecosystems and societies; both as a primary input for freshwater resources and as a potential threat to infrastructure and human lives
The aim of this study is to propose a more localised vertical profile correction technique using machine learning (ML) and information on hydrometeor proportions to predict the vertical change in reflectivity, referred to as growth and decay (GD)
An important part of the work consisted of establishing the foundations for the use of ML for the investigation of the vertical structure of precipitation
Summary
Precipitation constitutes a key meteorological variable for ecosystems and societies; both as a primary input for freshwater resources and (in deficit or excess) as a potential threat to infrastructure and human lives. The existing VPR correction methods can be subdivided into three broad types (Germann, 2000; Zhang and Qi, 2010) which are based on climatology, spatiotemporal averages and modelled VPRs. Climatological VPRs are based on radar data averaged over long time periods (days, seasons, years) and over a certain spatial area (radar volume or well-visible regions) (Joss and Pittini, 1991; Joss and Lee, 1995). Compared to the climatological VPRs, spatiotemporally averaged VPRs can better capture the temporal variations in reflectivity since these are based on a few volume scans only and regularly updated They remain computationally inexpensive, and, among the few countries who correct for VPR in the operational processing, several are using some version of spatiotemporally averaged profiles (Koistinen, 1991; Joss and Lee, 1995; Germann and Joss, 2002).
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
