Unmanned aerial systems and crop modeling for irrigation scheduling in the Southern High Plains
Lack of precipitation and irrigation resources threatens sustainable crop production in the Southern High Plains (SHP). Effective and efficient irrigation scheduling is required to improve water use efficiency (WUE) and conserve water. Typical irrigation scheduling methods are based on the reference evapotranspiration (ET0) and soil water balance. A key challenge in irrigation scheduling is the shortage of methods to determine accurate crop coefficients (Kc) and water requirements at different crop growth stages. Unmanned aerial systems (UAS) can provide high-resolution remote sensing images, which potentially enable accurate monitoring of crop growth and irrigation needs. The objectives of this research were: 1) to develop a UAS remote sensing based irrigation scheduling algorithm for cotton (Gossypium hirsutum L.); 2) to evaluate the water use and WUE of UAS remote sensing based irrigation scheduling as compared to the Decision Support System for Agrotechnology Transfer (DSSAT) simulations. The algorithm incorporated the FAO-56 ET0 method and Kc derived from high-resolution UAS images into the computation of crop water requirement. ET0 was calculated with meteorological data using the FAO-56 method. Ground cover was derived from the UAS images using the maximum likelihood classification. This ground cover was equivalent to the Kc at the corresponding crop growth stage. Daily crop evapotranspiration (ETc) is then calculated as the product of ET0 and Kc with adjustment of plant height. The daily water requirement is the difference between ETc and effective rainfall. Irrigation scheduling was implemented weekly based on the water requirement of the previous seven days. This algorithm was integrated into a Python script for irrigation scheduling. The algorithm was implemented in an experiment with four irrigation treatments, high (90% ET), medium (70% ET), low (45% ET), and very low (25% ET), in a research field in the SHP in 2018. The performance of this algorithm was evaluated by comparing with results from a long-term study and simulation from the DSSAT CSM-CROPGRO-Cotton model, with respect to seed cotton yield, water use, WUE, and irrigation water use efficiency (IWUE). Compared with the historical cotton yields, observed seed cotton yields for UAS based irrigation scheduling algorithm were higher by 7.4%, 11.3%, 9.0%, and 3.3% for four irrigation treatments, respectively. Statistical results showed the values of WUE and IWUE from UAS based irrigation scheduling were significantly higher than the DSSAT simulation results. The total irrigation amount of the UAS based irrigation scheduling algorithm was 30% less than the DSSAT simulation results for the high irrigation treatments. Higher yield with less irrigation was achieved using the UAS remote sensing irrigation scheduling as compared to the DSSAT simulation. This study demonstrated that UAS remote sensing has potential in effective irrigation scheduling to improve WUE, conserve water and enhance production sustainability.