Remote Sensing Estimation Of Potential Evapotranspiration In Poyang Lake Wetland Based On Penman-Monteith Model

Evapotranspiration is closely related to surface water and energy cycle,it is one of the important ways of surface water dissipation.With the rapid development of remote sensing technology,remote sensing has become an important channel to provide parameters closely related to surface evapotranspiration and replaced ground evaporation measurement partially,to achieve a wide range of land surface energy and water dynamic monitoring,in agriculture,hydrology,climate,ecology and other fields play an important role.Therefore,it is of great significance to study the remote sensing estimation method of surface evapotranspiration for the optimal allocation of regional water resources and the construction of ecological civilization.Based on Landsat satellite images and conventional ground meteorological data in2002,2005,2013 and 2016,a remote sensing estimation model of land surface evapotranspiration based on Penman-Monteith model is established in this paper.The potential evapotranspiration of Poyang Lake wetland on month,season and year scales is estimated.The results of the remote sensing estimation model are validated by the observed meteorological data of Poyang station,which is the ground observation station within the study area.The evapotrion of water surface in Poyang Lake is calculated by an empirical model developed by Hehai University.Subsequently,the factors affecting the evapotranspiration of the wetland in Poyang Lake and the evapotranspiration laws of different terrain types were analyzed.The main contents and achievements of the study are as follows:Firstly,Data preprocessing,in which the remote sensing data is mainly completed with ENVI5.3 software radiometric calibration,atmospheric correction,geometric correction and image clipping;by comparing the advantages and disadvantages of the three interpolation methods,namely inverse distance interpolation,Kriging interpolation,natural domain interpolation,and the use of conditions,select the inverse distance interpolation,single point gas.Image data are interpolated into surface data that cover the whole area.Secondly,Using Landsat remote sensing data of 2002,2005,2013 and 2016,the land use status of Poyang Lake wetland was classified by SVM classification algorithm.The results show that the total precision of the classification maps in 2002,2005,2013 and2016 are 98.20%,93.29%,91.03% and 91.03% respectively.The overall accuracy is96.03%;the Kappa coefficient of the classification map is 0.92 in 2003 and 2005,it is 0.89 in 2013,and 0.95 in 2016,which shows that the extracted land use map has good reliability.Thirdly,The solar radiation Ra,theoretical sunshine hours N,net short-wave radiation Rns and net long-wave radiation Rnl were calculated using meteorological data,and thebroad-wave albedo alpha and normalized vegetation index NDVI were inverted using Landsat-8 image data,then the net radiation Rn,soil heat flux G and potential heat flux were calculated.Table the spatial distribution characteristics of factors.Fourthly,Penman-Monteith model was used to calculate the potential evapotranspiration of Poyang Lake wetland.The results showed that the annual evapotranspiration of the wetland was 1289.60 mm in 2002,1402.23 mm in 2005,1559.29 mm in 2013 and 1881.02 mm in 2016.Finally,the potential evapotranspiration of different terrain types in four years was analyzed,and the results showed that the evapotranspiration of different terrain types was: water > mud Beach > grassland > farmland > bare soil >Town > hills and unused land.Finially,the remote sensing estimation model of evapotranspiration based on Penman-Monteith formula is implemented on the remote sensing image processing software,and the evapotranspiration of Poyang Lake wetland is estimated.The estimated results were verified by the measured data of meteorological stations,and the correlation between different meteorological factors and potential evapotranspiration was analyzed.The correlation coefficients of evapotranspiration and NDVI were 0.31-0.82,0.14-0.67,0.43-0.87,0.43-0.71,0.15-0.49 and 0.76-0.28 respectively.