Remote Sensing Estimation In Tumote Plain Farmland Steam Based On The SEBAL Model

In order to verify the applicability of the SEBAL model in Tumote Youqi,Baotou City,Inner Mongolia.Based on 13 Landsat8 remote sensing images from April to September in crop growth period from 2015 to 2016,the surface parameters(NDVI,vegetation coverage,emissivity,surface albedo,etc.)were retrieved and analyzed using remote sensing data and meteorological data The SEBAL model was established,the amount of evapotranspiration was estimated,and the time scale of instantaneous evapotranspiration was expanded to obtain the daily evapotranspiration and monthly evapotranspiration.The estimation results were validated and evaluated by using FAO Penman-Monteith formula and field data.The main influential factors were analyzed by principal component analysis and multiple stepwise regression.get conclusion:1.Overall,the surface parameters,in addition to surface albedo,NDVI,surface emissivity,surface temperature,surface net radiation,sensible heat flux,soil heat flux,evapotranspiration changes are "first Increase and decrease."2.Soil in the study area can be divided into two types: 0 ~ 60 cm is silty loam soi,60 ~ 80 cm is sandy loam soil.Leaf area index and field water storage during the growth period are "first increase and then decrease" trend.The changes of ET24 are obviously seasonal.The relative evapotranspiration of the SEBAL model and the FAO Penman-Monteith formula are 5.66% and 7.76%,respectively.In contrast,the SEBAL model is more accurate and the estimation process is easier.The variation law of monthly evapotranspiration obtained by time scale expansion is in accordance with the change of crop water requirement of Tumote Right Banner in Baotou,Inner Mongolia.3.The correlation between meteorological factors and evapotranspiration are: net radiation> sunshine time> maximum temperature> relative humidity> minimum temperature> wind speed.Relevant parameters of the correlation o f ET: the largest NDVI,followed by vegetation cover,and finally the surface temperature.The principal component analysis obtained comprehensive evaluation value Z,Z value can replace 10 kinds of influencing variables.Multiple regression analysis was carried out to obtain the optimal equation of the regression equation jointly constructed by maximu m temperature,surface temperature and vegetation coverage.