Medium- to long-term forecast of reference crop evapotranspiration based on the correction of the principal control factor

This study selected the temperature Penman-Monteith (PMT) model to solve the difficulty of reference crop evapotranspiration (ET0) prediction caused by incomplete meteorological data. Then, the principal control factor of the PMT model was determined using several indicators. The results show that the annual average correlation and sensitivity coefficients between T-max and ET0 are the highest. Hence, among the factors, T-max is the largest contributor in contribution proportion and amount to ET0 and the principal control factor of the PMT model. The root mean square error (RMSE), relative error (RE) and mean absolute error (MAE) of the forecast ET0 increased by 0.7, 0.2, and 0.6 mm/day, respectively, after the 16th day compared with those on the 15th day. Thus, a remarkable drop in ET0 forecast precision is observed and is hardly accurate for the purpose of water distribution management in an irrigation district. The RMSE, RE, and MAE of the T-max forecast decrease by 2.0-3.0 degrees C, 0.1-0.2, and 2.0-2.8 degrees C, respectively, and those of the corresponding ET0 forecast decrease by 0.7-0.9, 0.2-0.3, and 0.6-0.7 mm/day, respectively, after T-max correction. Compared with the decaying-average method, the improved decaying-average method can correct T-max better and improve the ET0 prediction accuracy.