Abstract: The importance of accurately estimating crop water requirement for irrigation forecast and agricultural water management has been widely recognized. Although it has been broadly adopted to determine crop evapotranspiration (ET_c) via meteorological data and crop coefficient, most of the data in whether forecast are qualitative rather than quantitative except air temperature. Therefore, in this study, how to estimate ET_c precisely only using air temperature data in forecast was explored, the accuracy of estimation based on different time scales was also investigated, which was believed to be beneficial to local irrigation forecast as well as optimal management of water and soil resources. Three parameters of Hargreaves equation and two parameters of McClound equation were corrected by using meteorological data of Xinxiang from 1970 to 2010, and Hargreaves equation was selected to calculate reference evapotranspiration (ET₀) during the growth period of winter wheat. A model of calculating crop water requirement was developed to predict ET_c at time scales of 1, 3, and 7 d intervals through combining Hargreaves equation and crop coefficient model based on air temperature. Results showed that the correlation coefficients between measured and predicted values of ET_c reached 0.883 (1 d), 0.933 (3 d), and 0.959 (7 d), respectively. The consistency indexes were 0.94, 0.95 and 0.97, respectively, which showed that forecast error decreased with the increasing time scales. Forecasted accuracy with an error less than 1 mm·d^-1 was more than 80%, and that less than 2 mm·d^-1 was greater than 90%. This study provided sound basis for irrigation forecast and agricultural management in irrigated areas since the forecasted accuracy at each time scale was relatively high.