Prediction of evapotranspiration for optimum water use in agriculture: a case study of Keiskamma irrigation scheme, Eastern Cape, South Africa

Abstract

Evapo-transpiration (ET) is one of the crucial elements of the hydrological cycle which expedites constant precipitation through the process of condensation. The accurate prediction of ET is essential for irrigated agriculture as it informs appropriate planning and contributes positively to the daily supervision of the irrigation scheme. However, because of the limited data in arid and semi-arid regions, which have been used widely in the traditional Penman-Monteith approach, alternative, reliable and more powerful techniques are used to predict ET. South Africa is one of many countries that fall under the semi-arid zones where the degree of evapo-transpiration is more than the rainfall rate. The aim of this study was to predict evapo-transpiration in the Keiskammahoek Irrigation Scheme located in Eastern Cape, South Africa, using three, time series, prediction models, namely, Auto Regressive Integrated Moving Average (ARIMA), Artificial Neural Networks (ANNs) and Hybrid (ARIMA-ANNs). ARIMA and ANNs models have been used mostly over the years to predict the linear and non-linear time series, and the Hybrid model, developed by Zhang was also applied because of its ability to capture both the linear and non-linear time series. The 18 years (2001 to 2018) ET time series data was extracted from Google Earth Engine, using java script, at Keiskammahoek Irrigation Scheme. Prior to the prediction of ET time series, the time series data were analysed to understand the behaviour of the time series. A detailed time analysis of Keiskamma River Streamflow and Sandile Dam monthly volume, which are water supply sources close to the study area, were analysed, using time series analysis methods such as: the Break for Additive Seasonal (BFAS) and Trend, Wavelength Analysis, Wavelet Coherence, Correlation Statistics, Theil-Sen plots, Man-Kendall Test, Sequential Mann-Kendall Test and Multi-Linear Regression Analysis. Furthermore, tele-connection analysis between the satellite-derived et time series for the study area and other parameters, such as the Normalised Difference Vegetation Index (NDVI), Normalised Difference Water Index (NDWI), Normalised Difference Drought Index (NDDI), and Precipitation(P), was performed. Through use of the Mann-Kendall Trend Test, it was noted that et has increased over the years with the z-score reaching +3.898 which is greater than +1.96 which indicates the significance of the trend, in contrast to the z-score for precipitation which is equal to - 2.6134, indicating a significant decrease in P in the study area over the 18-year period. This trend, and its significance, were noted also using the Sequential Mann-Kendall method. Using the Multi Linear Regression, a statistically significant relationship was also noted between ET, with p-value < 2X10-16 and NDVI, p-value equal to 7.89 x 10- 11, for Stream Flow p-value equal to 2.32X10-06, for P p-value equal to < 2Ex10-16 and for NDDI p-value equal to 0.0208. The significant relationship between these variables is indicated by a p-value less than 0.05.. Using ARIMA, ANNs and Hybrid(ARIMAANNs), the ET at Keiskammahoek Irrigation Scheme was predicted successfully for 3 years (2015 to 2018). Furthermore, the three models were combined to assess the quality of prediction further and ET was once more predicted successfully. To select the best performing model for prediction of ET, the predicted results for three applied modelling techniques were evaluated using four, well accepted, model performance statistics, namely, Root Mean Square Error (RMSE), Mean Absolute Error (MAE), Mean Absolute Percentage Error (MAPE) and Pearson’s Correlation Coefficient (R). The results of this study shows that, the hybrid (ARIMA-ANN) model outperformed both the ARIMA and ANN consecutively with less values of the statistical performance evaluation show-ing RMSE = 33.80, MAE = 27.02, MAPE = 17.31, and R = 0.94 compared to higher values of ARIMA and ANN In general, these forecasting results show the superiority of the Hybrid (ARIMA-ANN) model over ARIMA and ANN.