Daily Pan Evaporation Modelling With ANFIS and NNARX

Abstract

Evaporation, as a major component of the hydrologic cycle, plays a key role in water resources development and management in arid and semi-arid climatic regions. Although there are empirical formulas available, their performances are not all satisfactory due to the complicated nature of the evaporation process and the data availability. This paper explores evaporation estimation methods based on nonlinear dynamic neural network model (NNARX ) and adaptive neuro-fuzzy inference system (ANFIS) techniques. It has been found that NNARX and ANFIS techniques have much better performances than the empirical formulas (for the test data set, NNARX R2 = 0.95, ANFIS R2 = 0.94, Meyer R2 = 0.81 and Marciano R2 = 0.68). ANFIS and NNARX models are slightly better albeit the small difference. Although NNARX and ANFIS techniques seem to be powerful, their data input selection process is quite complicated. More studies are needed to gain wider experience about this data selection tool and how it could be used in assessing the validation data.