Short-term forecasting of daily reference evapotranspiration using the Hargreaves–Samani model and temperature forecasts
MetadataShow full item record
Luo, Y.; Chang, X.; Peng, S.; Khan, S.; Wang, W.; Zheng, Q.; Cai, Xueliang. 2014. Short-term forecasting of daily reference evapotranspiration using the Hargreaves–Samani model and temperature forecasts. Agricultural Water Management, 136:42-51. doi: http://dx.doi.org/10.1016/j.agwat.2014.01.006
Permanent link to this item: http://hdl.handle.net/10568/58367
Accurate daily reference evapotranspiration (ET0) forecasting is necessary for real-time irrigation forecasting. We proposed a method for short-term forecasting of ET0 using the locally calibrated Hargreaves–Samani model and temperature forecasts. Daily meteorological data from four stations in China for the period 2001–2013 were collected to calibrate and validate the Hargreaves–Samani (HS) model against the Penman–Monteith (PM) model, and the temperature forecasts for a 7-day horizonin 2012–2013 were collected and entered into the calibrated HS model to forecast the ET0. The pro-posed method was tested through comparisons between ET0 forecasts and ET0calculated from observed meteorological data and the PM model. The correlation coefficients between observed and forecasted temperatures for all stations were all greater than 0.94, and the accuracy of the minimum temperature forecast (error within ±2 C) ranged from 60.48% to 76.29% and the accuracy of the maximum tempera-ture forecast ranged from 50.18% to 62.94%. The accuracy of the ET0 forecast (error within ±1.5 mm day-1) ranged from 77.43% to 90.81%, the average values of the mean absolute error ranged from 0.64 to1.02 mm day-1, the average values of the root mean square error ranged from 0.87 to 1.36 mm day-1,and the average values of the correlation coefficient ranged from 0.64 to 0.86. The sources of errors were the error in the temperature forecasts and the fact that the effects of wind speed and relative humidity were not considered in the HS model. The applications illustrated that the proposed method could provide daily ET0forecasts with a certain degree of accuracy for real-time irrigation forecasts.