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.

Reference Evapotranspiration Estimation Model Based on Temperature and Humidity

Optimal Methods for Estimating Shortwave and Longwave Radiation to Accurately Calculate Reference Crop Evapotranspiration in the High-Altitude of Central Tibet

A Prediction of Reference Crop Evapotranspiration Based on Generalized Regression Neural Network and Particle Swarm Optimization Algorithm

Accessible Remote Sensing Data Based Reference Evapotranspiration Estimation Modelling

Optimal Alternative for Quantifying Reference Evapotranspiration in Northern Xinjiang

Improved Remote Sensing Reference Evapotranspiration Estimation Using Simple Satellite Data and Machine Learning.

Nation-scale reference evapotranspiration estimation by using deep learning and classical machine learning models in China

Machine learning-based estimation of evapotranspiration under adaptation conditions: a case study in Heilongjiang Province, China

Research on methods for estimating reference crop evapotranspiration under incomplete meteorological indicators

Modelling Seasonal Evapotranspiration of Arid Lands in China

Simple and Applicable Method for Estimating Evapotranspiration and Its Components in Arid Regions

Comparison of CLDAS and Machine Learning Models for Reference Evapotranspiration Estimation under Limited Meteorological Data

[Calculating Method for Crop Water Requirement Based on Air Temperature].

Calculation Methods of Crop Reference Evapotranspiration Under the Extreme Weather

Parameter regionalization based on machine learning optimizes the estimation of reference evapotranspiration in data deficient area

Simulation of Daily Reference Evapotranspiration Based on Least Squares Support Vector Machine Optimized by Coupled Simulated Annealing

Estimating Growing Season Evapotranspiration and Transpiration of Major Crops over a Large Irrigation District from HJ-1A/1B Data Using a Remote Sensing-Based Dual Source Evapotranspiration Model

A Method to Estimate Evapotranspiration in Greenhouse Conditions by Artificial Neural Networks Using Limited Climate Parameters

Research on ET0 Influence Factors and ANFIS Forecast Model

Calculation of Reference Evapotranspiration Based on FAO-Blaney-Criddle Method in Hetao Irrigation District.

Estimating daily reference evapotranspiration based on limited meteorological data using deep learning and classical machine learning methods