The representativeness problem of a station net applied to the verification of a precipitation forecast based on areas

The results and conclusions obtained in a verification process can be strongly influenced by its methodology, especially when some assumptions are made in order to avoid the limitations of the observational data (i.e., the lack of representativeness of the data or poor area coverage). Several aspects of the verification process must be studied if reliable conclusions are to be obtained. This paper discusses the problems related to the development of a verification method for a precipitation forecast defined over continuous areas using observational data from discrete points in space (meteorological stations). To tackle this asymmetry, the concept of the station net quality is first introduced. The verification approach is based on deterministic categorical scores. The effects of the number of stations, the spatial distribution of the station net and the precipitation thresholds used in the verification process are shown. Using different sets of precipitation stations, the scores show a strong sensitivity to the station net quality. The suitability of the scores used in this study is also analysed and the verification process is finally adapted to compensate for undesired effects. Copyright (C) 2007 Royal Meteorological Society.