Optimal irrigation water allocation for a center pivot using remotely sensed data

Efficient irrigation can help avoid crop water stress, undesirable leaching of nutrients, and yield reduction due to water shortage, and runoff and soil erosion due to over irrigation. Gains in water use efficiency can be achieved when water application is precisely matched to the spatially distributed crop water demand. This is important to produce high quality crops and otherwise conserve water for greatest efficiency of use. Irrigation efficiency is a term which defines irrigation performance based on indicators such as irrigation uniformity, crop production, economic return and water resources sustainability. The present paper introduces a modeling approach for optimal water allocation to a center pivot irrigation unit in consideration of these types of indicators. Landsat images, weather data and field measurements were used to develop a soil water balance model using Artificial Neural Networks (ANN). The model includes two main modules, one for soil water forecasting and one for optimization of water allocation. The optimization module uses Genetic Algorithms (GA) to identify optimal crop water application rates based on the crop type, growing stage and sensitivity to water stress. The forecasting module allocates water through time across the area covered by the center pivot considering the results from the previous period of irrigation and the operational limitations of the center pivot irrigation system. The model was tested for a farm equipped with a modern sprinkler irrigation system in Scipio, Utah. The solution obtained from the model used up to 30 percent less water without reducing the benefits realized by the irrigator than traditional operating procedures