Remote Sensing-based Evaluation Of Irrigation Efficiency And Crop Water Use Efficiency Over Irrigation District In Arid Region

Quantitative assessment of irrigation efficiency and crop water use efficiency is the base of high-efficient use of agricultural water resources.A representative irrigation district in arid region,the main region of Hetao Irrigation District in Inner Mongolia,is taken as the study region in this dissertation.Data from remote sensing,field experiment and on-site survery are used as input for remote sensing-based evapotranspiration(ET)model and maize cover mapping and yield estimation models.Based on modeling results,spatio-temporal variation of evapotranspiration,spatial distribution and the annual variation of maize distribution and yield are analyzed from 2003 to 2012.Moreover,irrigation efficiency and maize water use efficiency are assessed quantitatively.First,two dual source ET models(TSEB and HTEM)fed with MODIS data are applied in Hetao Irrigation District from 2003 to 2012 to examine the spatio-temporal patterns of evapotranspiration in the study region.Detailed comparison between the two ET models is provided with field scale and region scale test in this dissertation.Results indicate that the HTEM model is more capable of estimating evapotranspiration in this region.To evaluate the irrigation efficiency of Hetao Irrigation District,a new evaluation indicator,coefficient of irrigation water effective utilization,is proposed.The difference of evapotranspiration and effective precipitation in irrigated land during the crop growing season is considered as the effective use of irrigation water,and the ratio of effective use of irrigation water to net water diversion or total water diversion to the irrigation district is defined as the coefficient of irrigation water effective utilization based on net water diversion or total water diversion(?N and ?I).A study is carried out to estimate the coefficient of irrigation water effective utilization throughout the period 2003-2012.Results show that ?N tends to increase with the implementation of water-saving irrigation projects.And ?N decreases with precipitation and net water diversion.The reduction of net water diversion had a greater effect than the water-saving irrigation projects on ?N.To further explore the spatio-temporal patterns of the major crop(maize)in Hetao Irrigation District,representative MODIS-NDVI time series of maize is obtained during a field survey,which is fitted with an asymmetric logistic curve to obtain the characteristic phenological and vegetation indices.A maize classifier is proposed based on pheonology-vegetation characteristic values space.Results show that the classifier can be use effectively for maize mapping over multiple planting years and in areas with a complex planting structure.At last,maize evapotransipiration during its growth period is estimated based on previous study.A yield estimation model of maize is established based on crop water production function.Thus,the spatio-temporal patterns of maize water use efficiency in Hetao Irrigation District from 2003 to 2012 can be obtained.Results show that simulated maize growth period agrees well with measured value.Maize yield can be calculated based on Jensen function with high accuracy.The study of this thesis indicates that remote sensing data and models provide an effective and feasible approach for the quantitative evaluation of irrigation efficiency and water use efficiency over irrigation district in arid region.