The adverse impacts of climate change and the disparity between water availability and demand in agriculture necessitate the development of water-efficient irrigation schedules. The empirically derived crop water stress index (CWSI) is a popular tool for the detection of water stress in crops and the scheduling of water-efficient irrigation regimes. But the sensitivity of the empirical CWSI to the input parameters, i.e., air temperature (Ta), canopy temperature (Tc), and relative humidity (RH), is rarely studied. This study is conducted on wheat crops in the Uttar Pradesh province of India where four irrigation scheduling strategies/treatments are experimented with to study CWSI and its sensitivity to Tc and meteorological parameters (Ta and RH). Two irrigation scheduling strategies correspond to drip irrigation and the remaining two correspond to flood irrigation. The mean CWSI values, derived by empirical approach, are 0.11 and 0.03 for drip irrigated treatments and 0.29 and 0.31 for flood irrigated treatments. To observe the sensitivity of the empirical CWSI to the input parameters, initially, the accuracies of the research-grade equipment, used for taking Tc, Ta, and RH observations, are used as error values in the input parameters. This resulted in 3.5 to 7% errors in the empirically derived CWSI values. A maximum error of 2.7% was observed in CWSI values when an error of 10% was introduced in the RH data. The maximum error in CWSI (52%) was observed for the case when a combined error in Ta, Tc, and RH by −1 °C, 1 °C, and 10%, respectively, were introduced. A random error in the range −1 to 1 °C in Tc resulted in a maximum CWSI error of 21%. Out of the three input parameters, CWSI was found to be least sensitive to RH and with sensitivity to both Ta and Tc showing similar trends. A common pattern of decreasing error in CWSI values for increasing vapor pressure deficit (VPD) was observed for all the studied cases. The results show that the water stress, as reflected by the CWSI values, was significantly lower for the drip irrigation treatments as compared to the flood irrigation treatments. Also, small errors in the input parameters may combine to result in significant errors in empirically derived CWSI. Hence, data quality is critical for the studies that utilize CWSI for irrigation scheduling, especially in humid climatic conditions. [ABSTRACT FROM AUTHOR]