Integration of Weather Radar and Rain Gauge Data: A Statistical and Geostatistical Approach

For hydraulic-hydrological models to yield results consistent with reality, it is crucial to analyze the quality of the input parameters, aiming to identify any unusual behavior in the data series. Furthermore, combining multiple data sources can be an effective strategy for achieving better results. Combining rainfall estimates from weather radar with surface network measurements can help reduce uncertainties. Data from January, February, March, October, November, and December (rainy months in Brazil’s Southeast region) of 2016 from the Camorim Pequeno station in Angra dos Reis – RJ, Brazil, was analyzed for their fit to Log-Normal and Log-Gamma distributions, the presence or not of outliers, the heteroscedasticity or homoscedasticity, and the shape of the cross-covariograms and cross-correlograms. Cross-covariograms and cross-correlograms between radar and automatic tipping bucket rain gauge data were obtained, demonstrating the correlation of these variables, considering a time difference of 5 to 20 minutes between them.