Radial Basis Function (RBF) Interpolation and Investigating its Impact on Rainfall Duration Mapping

Abstract

The missing data in database must be reproduced primarily by appropriate interpolation techniques. Radial basis function (RBF) interpolators can play a significant role in data completion of precipitation mapping. Five RBF techniques were engaged to be employed in compensating the missing data in event-wised dataset of Upper Paramatta River Catchment in the western suburbs of Sydney, Australia. The related shape parameter, C, of RBFs was optimized for first event of database during a cross-validation process. The Normalized mean square error (NMSE), percent average estimation error (PAEE) and coefficient of determination (R2) were the statistics used as validation tools. Results showed that the multiquadric RBF technique with the least error, best suits compensation of the related database.

Keywords

• Function
• Rainfall Duration Mapping
• Shape Function
• Cross validation