Research On Surface Energy Exchange And Evapotranspiration Of Typical Wetlands In The Permafrost Region Of The Greater Khingan Mountains

Permafrost wetlands and the atmosphere continue to exchange energy and materials such as sensible heat,latent heat,and evapotranspiration,forming a unique wetland ecosystem and unique climate environment.The changes in energy exchange affect the development of the permafrost wetland ecological environment.Regional climate change has an important impact.Therefore,the energy flux and water vapor exchange on permafrost wetland underlying surface are deeply studied.However,there is no study on the land surface water and heat fluxes of this ecosystem in China.Based on eddy covariance method,the energy flux and evapotranspiration of a typical permafrost wetland in Daxinganling region were observed in the growing season for two years,in order to reveal the change rules and influencing factors of the energy flux and evapotranspiration of permafrost wetland ecosystem,and provide more reliable actual observation data for further study of the energy and water cycle of permafrost wetland,which can be used in Daxinganling region Optimization of land surface model parameterization scheme and verification of remote sensing inversion data.The main results are as follows:(1)The diurnal variation of each component of energy exchange flux(sensible heat flux,latent heat flux,net radiation flux and soil heat flux)in the growing season of wetland in the permafrost region of Daxinganling is of"single peak"type,reaching the peak around noon,with obvious seasonal variation characteristics.The average values in the 2018 growing season were 40.1±30.2 W m^-2,47.8±27.5 W m^-2,143.3±58.7 W m^-2,and 8.8±5.6 W m^-2,respectively.The average values in the 2019growing season were 34.7±18.4 W m^-2,53.5±23.9 W m^-2,116.4±5.2 W m^-2 and7.5±5.0 W m^-2.(2)On the seasonal scale,sensible heat flux is mainly affected by radiation flux,while latent heat flux is mainly affected by radiation flux and temperature.On the daily scale,the sensible heat flux is strongly affected by the air temperature in addition to the radiation flux.Besides the radiation flux and the temperature,the latent heat flux is also affected by the wind speed.Based on the daily average data of air temperature,radiation flux and saturated water vapor pressure,combined with the multiple linear regression equation obtained by the least square method,the seasonal variation of sensible and latent heat fluxes in permafrost wetlands can be well simulated.(3)During the growing season,the latent heat flux was dominant,accounting for33%?45%of the net radiation.In 2018 and 2019,the energy closure rates are 0.83 and 0.78,respectively.The energy closure rates are better in the middle of the growing season,and worse in the early and late growing season.The results of energy closure rate in the whole growing season are close to the average value(0.79)reported by FLUXNET.(4)The main driving factors of evapotranspiration in the permafrost region of Daxinganling are radiation,temperature and air saturated vapor pressure difference.Meanwhile,the influence of groundwater level on evapotranspiration is greatly weakened due to the capillary transmission of water by wetland vegetation roots.At the same time,in spring and summer,if there is high air temperature and less precipitation in the wetland,it will significantly increase the annual evapotranspiration and change the hydrological process of the wetland.(5)The seasonal variation of vegetation coefficient is obvious,and the diurnal fluctuation is large,and the early and late growth period is less than the middle growth period.At the same time,the actual evapotranspiration is higher than the potential evapotranspiration in the period of high precipitation,that is,the vegetation coefficient is greater than 1.Three different potential evapotranspiration models(Penman-Monteith method,Priestley-Taylor method and Makkink method)were used to estimate the actual evapotranspiration in combination with the vegetation coefficient.The estimated results showed that the RMSE and MAE were 0.82?1.07 mm d^-1.This is because the conventional vegetation coefficient only considers the seasonal fluctuation,but does not consider the diurnal fluctuation.Through the correlation analysis,it is found that there is a good correlation between the vegetation coefficient and temperature,so this paper uses temperature to correct the vegetation coefficient,and uses the data in 2018 to correct the vegetation coefficient,and the data in 2019 to verify the simulation results,the simulation error is reduced,RMSE and MAE are between 0.57?0.79 mm d^-1,the priority of potential evapotranspiration model for simulation of permafrost wetland evapotranspiration is:Penman-Monteith method>Priestley-Taylor method>Makkink method.