Zbyněk Sokol, Daniela Řezáčová (ÚFA AV ČR)
Meteorologické zprávy, 54, 2001, 2, 46-56.
English Summary:
The paper summarises the results of testing various multiple regression models applied to the adjustment of radar-based daily precipitation to ground measured precipitation amount. The statistical treatment of concern follows the concept of the classified multiple regression where different regression models are developed for different precipitation classes. The decision about the application of the concrete regression model depends on the ground-based precipitation estimate. To estimate the ground-based precipitation the one step correction (weighted average) is applied. Predictors used in the regression models depend on pixel coordinates relative to the radar position, on the rough radar precipitation and in some cases also on the interpolated ground precipitation.
The radar-based precipitation was determined from maximum reflectivity measured by Skalky radar (C band Doppler radar Gematronik METEOR360AC). Ground precipitation was reported by more than 600 ground stations giving daily sums. All data come from the period 1996-1998 and the first two years of data are used as developmental data set. The set includes the data from the extreme precipitation period in 07/1997 when a large part of the eastern Czech Republic was flooded. The 1998 data are applied to the independent verification of the statistical models. The flash flood period (22-23/07/1998) when an organised convective system produced extreme local precipitation (maximum 209mm in about 10h) is involved in the verification set. The inclusion of large precipitation in the both data sets, the developmental as well as verification one, makes possible the improvement of the radar-based precipitation even in the range of large precipitation values.
The results of classified regression were compared with results obtained by (a) rough radar based estimation, (b) ground based interpolation, (c) non classified multiple regression, and (d) bulk regression model. The improvement expressed by bias and RMSE is large especially in the warm period and for the classes with large precipitation (30-50mm, >50mm). In winter the interpolated ground precipitation gives results of the nearly the same quality as the adjusted radar. Tests with various number of ground stations used to provide the first guess precipitation show that the adjusted radar gives better results in the range of 81 stations. Such number of automatic ground stations is available in the Czech Republic. The best technigue based on the use of 81 stations is found suitable for the adjustment of daily radar-based precipitation and it is proposed to be used operationally in the Czech Weather Service.