Application Of 3PG Carbon Production Model In The GPP Estimation Of Broad-leaved Korean Pine Forest In Changbai Mountain.

The gross primary productivity of vegetation is an important parameter as a vital evaluation factor in terrestrial ecosystem circulation for assessing capacity of forest carbon sequestration and carbon budget.The ability of accurate estimation of vegetation productivity can predict dynamic changes in ecosystems,and help to understand the relationship between ecosystems and climate change,and have data and scientific support for the ecosystem management.These are vital signification for the further study of global carbon cycle processes and ecosystem functions.The GPP estimation model is not universal because it is established and verified in a specific area.Therefore,there required optimization for the specific factors and apply the 3PG carbon production model to the broad-leaved Korean pine forest in Changbai Mountain,it will be better reflected the vegetation characteristics of Changbai Mountain.The study used the multi-source and multi-scale data,and integrate data-models.It integrated the remote sensing data and the ground observation data of flux data and meteorological data.It evaluated model simulation accuracy,and optimized model structure to reduce model errors,which is urgent in GPP estimation study.In this paper,the contribution area of the Changbai Mountain Flux Tower was used as the research area.It used the remote sensing data,meteorological data and Changbai Mountain flux observation data.The 3PG carbon production model was compared with the VPM,VI and EC-LUE models,and the flux observation was used to evaluated model simulation accuracy and optimized the model structure.The main research contents and conclusions of this paper are as follows:(1)In order to solve the relation of the each factor and the GPP,the correlation analysis,the results shows that there were the largest determination coefficient of 3 vegetation indices(NDVI,EVI,LSWI)and latent heat flux with GPP in the factors of the model inputted,followed by the LAI,average temperature,precipitation,VPD and PAR.The correlation of the sensible heat flux with GPP from eddy correlation observations was minimal determination coefficient.(2)The GPP estimation based on 3PG carbon production model,which is compared with GPP estimation accuracy simulated by VPM,VI and EC-LUE light energy models in this paper.This study predicted GPP used these models driven by the 2003-2005 data,then the three-year flux data was used to the verification data,and we compared the model simulation results.We considered these evaluation indexes of R2,RMSE and the fitting slope.The results show that the estimation accuracy using 3PG model was the highest(R2=0.921,RMSE=1.89 gC/m2/d),followed by VI and EC-LUE model,and VPM(R2=0.920,RMSE=2.82 gC/m2/d)model had the lowest estimation accuracy.In terms of the annual average GPP,the 3PG carbon production model is overestimated during the growing season,while other models have significant or slight underestimation during the growing season.(3)In view of the overestimation problem in the prediction of 3PG carbon production model,this paper used the parameter structure of other light energy utilization models to optimize the 3PG carbon production mode.The result showed that the model was optimal in estimating gross primary productivity(GPP)when the temperature,the enhanced vegetation index and the land surface water index was taken as the temperature limit factor,the fraction of photosynthetically active radiation absorption,and the moisture restriction factor respectively it obtained the higher estimation accuracy(R2=0.948,RMSE=0.035 mol·m-2·month-1)than the original model(r2=0.854,RMSE=0.177 mol·m-2·month-1),the optimized 3PG carbon production model can effectively improve the growth season overestimation in the original model.The paper used the 3PG carbon production model to simulate the GPP of broad-leaved Korean pine forest,optimized the model structure and improved the simulation accuracy.The research process deepens our understanding of the 3PG carbon production model and the different structures and parameters of each light energy utilization model.It can clarify the impact of each parameter on GPP estimation sensitivity,in order to provide important reference value for the further predict forest growth and biomass change using 3PG model.