Interpolation methods applied to the spatialisation of monthly solar irradiation in a region of complex terrain in the state of Rio de Janeiro in the southeast of Brazil

Abstract

Due to the scarcity of weather or solarimetric stations, and in the absence of observed solar irradiation data (Hg), one alternative for obtaining continuous and quality data distributed in space is the use of spatial interpolation. The aim of this study was to assess different methods of spatial interpolation applied to the monthly average daily Hg in the state of Rio de Janeiro, in the southeast of Brazil. Inverse distance weighting (IDW), tension spline (SpT) and ordinary Kriging (OKr) were assessed. Hg time series (2000–2013) were used, measured at 20 automatic weather stations located in the state of Rio de Janeiro and the states of Sao Paulo, Minas Gerais and Espirito Santo. The IDW and SpT methods showed satisfactory performance, and result in precision (coefficient of determination—r2 > 0.90 for IDW and > 0.80 for SpT) and accuracy (refined Willmott index—dref > 0.55 for IDW and SpT) for most months, while OKr showed inadequate precision (r2 0. 50) similar to that of IDW and SpT. The RMSE was less than 10% (1.64 MJ m−2 day−1) for all methods. Under conditions of heterogeneous station distribution, low station density and complex terrain, the OKr method does not produce a precise spatial interpolation of monthly Hg.