VALIDATION OF PRECIPITATION DATA OF GPM SATELLITE
PRODUCTS OVER UKRAINE

This study provides a description and results of the validation methodology of GPM satellite precipitation data for the territory of Ukraine. Validation was carried out by comparing satellite data with ground weather stations data and included the following stages: data collection and processing, bringing the data to the same spatial and temporal resolution, calculating correlation coefficients and estimation of confusion matrices.

Correlation analysis was carried out for two studied periods (the first period - April-September 2020, the second - April-September 2021). Satellite data were provided by the GPM project of NASA and ground data were provided by 155 meteorological stations of the Ukrainian National Hydrometeorological Network.

Satellite and ground data were calculated to have the comparable values, that is 12 hours precipitation sums, from 6:00 a.m. to 6:00 p.m. For this, weather stations data were obtained as the sum of the amount of precipitation for this period of the day, and the satellite data that were downloaded for every half an hour as intensity mm/h, were converted into mm/day, thus obtaining the precipitation amount for the specified period of the day.

For obtaining reliable results of validation, correlation coefficients were calculated and confusion matrices were built. Confusion matrices are based on the division of precipitation into classes of different intensity. Confusion matrices were calculated for four classes of precipitation (mm): “0-2”, “2-5”, “5-10”, “>10”. Calculations were performed using the software environment for statistical calculations R.

The comparison analysis showed that a larger number of stations (117) has a correlation coefficient 0.5 - 0.8. The coefficients of the confusion matrices showed that low-intensity precipitation or “no precipitation”, as well as high-intensity precipitation, are estimated by the satellite with high accuracy in comparison with the ground-based weather station measurements. So, according to the "specificity" indicator, the highest level of correspondence of satellite data (GPM Late, GPM Early) to ground data has precipitation class >”10 mm”. The low value of the "specificity" for the range of 0-2 mm is explained by the fact that satellite methods are able to detect very low values of precipitation intensity, while the station shows their absence - 0 mm. Thus, according to the values of the confusion matrices, we see that a large part of the values of surface observations of precipitation in the range of 0-2 mm are shown by the satellite method as “2-5 mm” class. We can also see that the “2-5 mm” class by satellite has the most confusion with the 0-2 mm class by station (GPM Late, GPM Early). However, based on the high values of the "recall" for the range of 0-2 mm (GPM Late, GPM Early), we can conclude that when the station does not show this amount of precipitation (0-2 mm), then the satellite also shows another class of values. The high values of the coefficient of "precision" show that the satellite measurements are really reliable, that is, the presence or absence of precipitation will be determined precisely, since the highest value of "precision" is typical for low and high amounts of precipitation (on average 0.76 and 0.66, respectively). Heavy precipitation or no precipitation detected by satellites is confirmed by ground stations in most cases.

Keywords: precipitation, satellite data, ground data, comparison