ESTIMATION OF EVAPOTRANSPIRATION RATE IN THE SAHELIAN REGION OF NIGERIA USING GENERALIZED REGRESSION NEURAL NETWORK AND FEED FORWARD NEURAL NETWORK
Artificial Neural Network (ANN) has been employed by researchers in obtaining accurate estimates of evapotranspiration rate. Generalized Regression Neural Network (GRNN) and Feed Forward Back Propagation Neural Network (FFBP NN) were used to estimate evapotranspiration rate in Kano State, Northern Nigeria to ascertain its modelling accuracy under less input parameters. A 25-year monthly - time step of climatological data was collected from IITA (International Institute of Tropical Agriculture) station. The data was grouped into 12 different input combination with training and validation sets. GRNN results indicate the lowest performance ranking as the lone solar radiation input combination (GRNNSr) having R = 0.3603, R2 = 0.1298, MSE = 1.4356, RMSE = 1.982 and the highest as the input combination of Temperature and Wind Speed (GRNNTW) with R= 0.7925, R2 = 0.6281, MSE = 0.6048, RMSE = 0.7777. The two–layered Feed Forward Neural Network (FFBN) with 10 hidden neurons, also agreed with the GRNN in the ranking of the inputs combinations. The input combination of Temperature, Wind speed and Solar radiation had the best performance under the FFBP NN with R = 0.9111, R2 = 0.8301, MSE = 0.4011, RMSE = 0.6333. While the input combination of Solar radiation and Humidity, had the lowest performance under the FFBP NN with R = 0.4662, R2 = 0.2177, MSE = 1.8259, RMSE = 1.3512. Based on the comparisons, FFBN showed the highest potential in estimating evapotranspiration in the Sahelian region of Nigeria under limited climatological input parameters.