Statistical and neural network approach for estimating monthly evapotranspiration at the international institute of tropical agriculture, Ibadan, Nigeria - a comparative study

Abstract

Evapotranspiration (ET) is one of the main components of the hydrological cycle as it accounts for more than two-thirds of the precipitation losses at the global scale. Reliable estimates of actual Evapotranspiration are crucial for effective watershed modelling and water resource management, yet direct measurements of the Evapotranspiration losses are difficult and expensive. The major objective of this study was to investigate the potential of the classical linear regression and neural network (NN) technique to estimate evapotranspiration, and to examine if a trained neural network with limited input variables can estimate ET efficiently. The study utilized daily climatic data of temperature, relative humidity, sunshine hours, wind speed, and rainfall for ten years collected from the International Institute of Tropical Agriculture. (IITA) Ibadan, Nigeria. Linear regression models in terms of the climatic parameters influencing the regions and, optimal neural network architectures considering these climatic parameters as inputs were developed. The linear regression models showed a satisfactory performance in the monthly estimation in the region selected for the present study. The NN models, however, consistently showed a slightly improved performance over linear regression models. The results also indicated that even with limited climatic variables an ANN can estimate ET accurately.