Evapotranspiration data assimilation with genetic algorithms and SWAP model for on-demand irrigation

Evapotranspiration (ET) is one of the indicators of water use efficiency. Periodic information of ET based on remote sensing is useful for an on-demand irrigation (ODI) management. The main objective of this paper was to develop an ET data assimilation scheme to optimize the parameters of an agro-hydrology model for ODI scheduling. The soil, water, atmosphere, and plant (SWAP) simulation model has been utilized for this purpose. We computed remote sensing-based ET for a wheat field in the Sirsa Irrigation Circle, Haryana, in India using 18 cloud-free moderate resolution imaging spectroradiometer images taken between December 2001 and April 2002. The surface energy balance algorithm for land (SEBAL) was used for this purpose. Because ET estimates from SEBAL provide information on the surface soil moisture state, they were treated as observations to estimate unknown parameters of the SWAP model via a stochastic data assimilation (genetic algorithm) approach. The SWAP parameters were optimized by minimizing the residuals between SEBAL and SWAP model-based ET values. The optimized parameters were used as input to SWAP to estimate soil water balance for ODI scheduling. The results showed that the selected parameters (i.e. sowing, harvesting, and irrigation scheduling dates) were successfully estimated with the data assimilation methodology. The SWAP model produced reasonable states of water balance by assimilating ET observations. The root mean square of error was 0.755 and 2.132 cm³/cm³ for 0–15 and 15–30 cm soil depths the same layers, respectively. With optimized parameters for ODI, SWAP predicted higher yield and water use efficiency than traditional farmer’s irrigation criteria. The data assimilation methodology produced can be considered as an operational tool at the field scale to schedule irrigation or predict irrigation requirements from remote sensing-based ET.