Estimation of Crop Water Stress Index using Machine Learning Techniques to Enhance Irrigation Scheduling of Wheat

Código

I-EBHE0019

Autores

Aschalew Cherie Workneh

Tema

WG 1.04: From local to large scale human-water dynamics

Resumo

Estimation of crop water stress index (CWSI) to quantify the crop water stress is a cost- effective, non-destructive, and easy to use technique. The CWSI is derived from canopy temperature of crops. This study addresses the applicability of CWSI to detect the onset of the crop water stress of wheat crop. In this study, empirical CWSI (EP-CWSI) method was used for estimation of CWSI. The estimated EP-CWSI was compared with Machine learning (ML) techniques which are getting greater attention in recent times for estimation of CWSI. The study employed two ML techniques, adaptive neuro-fuzzy inference system (ANFIS) and self-organizing maps (SOM) for determination of the CWSI of wheat crop. Field experiments were conducted with varying irrigation water applications during two seasons in 2022 and 2023 at the irrigation field laboratory at the Civil Engineering Department, Indian Institute of Technology Roorkee, India. The ANFIS and SOM-simulated CWSI values were compared with EP-CWSI. Multiple regression analysis was used to determine the upper and lower CWSI reference temperatures. The upper CWSI reference temperature was found to be a function of crop height and wind speed, while the lower CWSI reference temperature was a function of crop height, air vapor pressure deficit, and wind speed. The performance of ANFIS and SOM were compared based on mean absolute error (MAE), mean bias error (MBE), root mean squared error (RMSE), index of agreement (d), Nash-Sutcliffe efficiency (NSE), and coefficient of correlation (R2). Both models successfully estimated the CWSI of the paddy crop, with higher correlation coefficients and lower statistical errors. However, the ANFIS (R2=0.81, NSE=0.73, d=0.94, RMSE=0.04, MAE= 0.00-1.76 and MBE=-2.13-1.32) outperformed the SOM model (R2=0.77, NSE=0.68, d=0.90, RMSE=0.05, MAE= 0.00-2.13 and MBE=-2.29-1.45). Overall, the findings suggest that ANFIS is a promising tool for accurately estimating CWSI in wheat crop compared to SOM.