A Comparison of Spatial Interpolation Techniques in Rainfall and Temperature Estimation
Alejandro M. de Asis and Kenneth S. Artes1
Agricultural operations and many engineering works need spatially distributed estimates of temperature and rainfall. A common problem is the conversion of point measurements made in meteorologic stations into appropriate, spatially-distributed variables.
This research aimed to determine the best-suited, GIS-aided spatial interpolator specific for temperature and rainfall for selected climatic and synoptic stations in the Philippines. Three interpolators, namely kriging, inverse distance weighted averaging, and splining were tested using cross-validation techniques and summary statistics. Results showed the superiority of splining in estimating spatial monthly rainfall. Inverse distance weighted averaging on the other hand was found to be best suited for mean monthly temperature estimation.
1Assistant Professor and University Research Associate, respectively, Agrometeorology and Farm Structures Division (AFSD), Institute of Agricultural Engineering (IAE), College of Engineering and Agro-industrial Technology (CEAT), University of the Philippines Los Baños (UPLB)