Résumé
Aim: A study was conducted to develop a new approach of non-stress baseline and evaluate different approach of determining non stress baseline for Crop Water Stress Index (CWSI).Methodology: To observe maximum and minimum canopy temperature, two plots with three replications were maintained at fully stress (no irrigation) and no stress (fully irrigated) conditions, respectively. Portable weather station with data logger was installed at the middle of field to record continuous data at half an hour interval. Gypsum block soil moisture sensor were buried at 20 cm interval up to 100 cm and connected with data logger. Infrared radiometer was used to measure canopy temperature. Results: The evaluation was based on five statistical indicators; correlation coefficient (R2), root mean square error (RMSE), mean absolute error (MAE), mean bias error (MBE) and Wilmot index of agreement (d). It was found that new regression base line performed better than other existing approaches. The calculated value of RMSE, MAE, MBE, d and R2 were 0.65, 0.51, 0.02, 0.79 and 0.83, respectively, for new baseline approach. Interpretation: The newly developed multiple regression equation (incorporating three weather parameters (Rn, VPD and u) and two crop growth parameters (h and CC)) helped to reduce error in the calculation of non-stress baseline for wheat crop in humid subtropical climate