Abstract:
Abstract: Real-time irrigation forecast is a key step for management modernization in Hetao Irrigation district. Irrigation forecast lies on the prediction of reference evapotranspiration (ET0). A calculating method for ET0 based on air temperature is needed since air temperature is the most accurate and conventional parameter acquired from weather forecast. In this study, FAO-Blaney-Criddle method based on air temperature for ET0 estimation was proposed, and the parameter of the equation was calibrated every ten days in Jiefangzha Irrigation sub-district of Hetao Irrigation district. FAO-Penman-Monteith equation was employed as standard for ET0 estimation. The meteorological data at Linhe climate station, which represented the climate conditions of Jiefangzha Irrigation sub-district, was utilized for calibration and validation of the parameter. The results showed that the estimated ET0 of FAO-Blaney-Criddle method was close to that of FAO-Penman-Monteith method in growing season (April to September) on 10 days scale. The relative error (RE) of the 10 days average ET0 was less than 5%, and the root mean square error (RMSE) was less than 0.65 mm/d in calibration period. And in validation period, RE was less than 9% and RMSE was less than 0.70 mm/d. It was because the correlation between ET0 and temperature was larger in growing season than non-growing season. In April, August and September, ET0 estimation of FAO-Blaney-Criddle method was closer to ET0 estimation of FAO-Penman-Monteith method, compared with that in May, June and July. In the latter 3 months, ET0 was underestimated by FAO-Blaney-Criddle method when ET0 of FAO-Penman-Monteith method was less than 5 mm/d and it was overestimated when ET0 of FAO-Penman-Monteith method was greater than 6 mm/d. For the 10 days moving average ET0, the Nash-Sutcliffe efficiency coefficient (NSE) of estimation of FAO-Blaney-Criddle method was 0.77. Forecasted accuracy with an error less than 0.5 mm/d was greater than 70%, and that with an error less than 1 mm/d was greater than 95%. The same method and parameter was applied at Hangjinhouqi climate station located in Jiefangzha Irrigation sub-district. The NSE of estimation of FAO-Blaney-Criddle method was 0.75. Forecasted accuracy with an error less than 0.5 mm/d was 68%, and that with an error less than 1 mm/d was greater than 95%. Hence, FAO-Blaney-Criddle method is available for the prediction of ET0 as well as irrigation forecast in Jiefangzha irrigation sub-district. When the parameter of FAO-Blaney-Criddle method was employed to estimate ET0 at Baotou climate station located in Hetao irrigation district but a bit far from Jiefangzha irrigation sub-district, theET0 estimation of FAO-Blaney-Criddle method was still reasonable though it was a bit poor than the estimation at Linhe station and Hangjinhouqi station. NSE of estimation was 0.65, forecasted accuracy with an error less than 0.5 mm/d was 52%, and that with an error less than 1 mm/d was 86%. It demonstrates that FAO-Blaney-Criddle method is available in the whole Hetao Irrigation district. It also suggests that the parameter of FAO-Blaney-Criddle method must be different in different places. And when FAO-Blaney-Criddle method is employed to estimate ET0 in the other area, the parameter should be calibrated using the meteorological data in corresponding area.