Radar-Derived Climatology of Precipitation Intensities in Czechia: improvement due to daily totals from rain gauges

0518766 - ÚFA 2020 US eng A - Abstrakt
Bližňák, Vojtěch - Kašpar, Marek - Müller, Miloslav - Zacharov, Petr, jr.
Radar-Derived Climatology of Precipitation Intensities in Czechia: improvement due to daily totals from rain gauges.
39th International conference on radar meteorology. Boston: American Meteorological Society, 2019.
[International conference on radar meteorology /39./. 16.09.2019-20.09.2019, Nara]
Institucionální podpora: RVO:68378289
Klíčová slova: weather radar * rain gauges * adjustment * precipitation climatology * Czech Republic
Obor OECD: Meteorology and atmospheric sciences
https://cscenter.co.jp/icrm2019/program/data/ICRMprogram.html

Radar-derived precipitation estimates play a crucial rule not only in meteorology but also in climatology
and hydrology. In combination with rain gauge measurements they can be considered as the most
accurate precipitation product with a high horizontal and temporal resolution. However, the main
limitation is that radar data are available for the past few decades only and are completely missing for
historical precipitation events, which restricts some climatological applications.
The contribution will evaluate a 17-year period of adjusted radar-derived precipitation estimates in the
warm parts (April-October) of the years 2002-2018 on the area of the Czech Republic (CR). The rain rates
will be calculated using radar reflectivity data at 2 km above sea level (CAPPI 2 km) estimated from
weather radar measurements performed by two Czech C-band Doppler radars (Brdy, Skalky) every 5 or 10
minutes in 1 km by 1 km square boxes over the whole area of the CR. Radar-derived daily precipitation
will be merged with daily rain gauge measurements and resultant adjusted daily sums of precipitation will
be divided into 10-min precipitation in the ratio of 10-min radar-derived precipitation. Subsequently,
10-min precipitation estimates will be accumulated in various lengths ranging from 30 min to 24 hours.
For climatological purposes seasonal as well as monthly precipitation sums will be corrected by a simple
correction method (Fairman et al., 2015). The method is based on multiplication of accumulated sums by
a ratio between the number of all theoretically possible 10-min measurements and all available 10-min
precipitation measurements in order to take into account a number of missing values that can reach
approximately 15-20 % in average. With respect to the spatial distribution a probability of precipitation
greater than a given threshold will be calculated for every pixel as a seasonal and monthly average during
the whole period. An example of monthly precipitation sums averaged over 10 years (2002-2011) for
summer months (June-August) is depicted in Figure 1.
The accumulation and average of adjusted radar-derived precipitation and interpolated values of
precipitation measured by rain gauges will be compared and their difference will be expressed by various
verification techniques (e.g., RMSE, BIAS, etc.). Besides precipitation totals accumulated over longer
period, attention will be also paid to the analysis of extreme sub-daily precipitation events due to their
high variability in space and time. Gridded information based on adjusted radar-derived precipitation will
be validated with independent 1-h rain rate records at selected stations and for selected precipitation
extremes.
Trvalý link: http://hdl.handle.net/11104/0303830