| Forests,as one of the largest carbon pools in the ecosystem,play a crucial role in the global carbon cycle.The carbon storage capacity of forests can measure their carbon sequestration ability,and accurately estimating forest carbon storage can provide a reference for the quality evaluation of forest resources.With the development of sensor technology and satellite technology,remote sensing data has gradually become an important technical means for estimating forest carbon storage.Based on the accurate measurement of forest carbon storage,further estimation of the economic value of forest carbon storage and carbon sequestration can improve the quality of forest management and level of management,which is of great significance for China "Dual Carbon" strategy and ecological civilization construction.This paper focuses on the Chinese fir plantation forests in Lutou Forest Farm in Hunan Province.Based on the investigation data of forest management units,Landsat8 and Sentinel-2 optical remote sensing images are used in combination with four common models(K-nearest neighbor,multilayer perceptron,support vector machine,and random forest)to estimate the carbon stock of Chinese fir plantation forests in Lutou Forest Farm and to generate the carbon stock distribution map at the management unit scale.Furthermore,considering the existing management status of Chinese fir forests in the forest farm,the optimal rotation period,carbon sequestration cost,and expected land value of Chinese fir plantation forests under different accounting models are estimated using the Land Expectation Value method,Chinese fir growth models,and the wood trading price data in Hunan Province,respectively.The dynamic effects of carbon prices and discount rates on the rotation period are also analyzed.The research results are as follows:(1)The combined feature set of Landsat-8 and Sentinel-2 data can provide a more accurate estimation of carbon stocks in small groups of fir plantation forests.For carbon storage estimation in Chinese fir forests based on single optical images,the optimal estimation model using Landsat-8 imagery achieved an accuracy of R2=0.54,RMSE=12.40 t/ha,rRMSE=37.26%.The optimal estimation model using Sentinel-2 imagery achieved an accuracy of R2=0.23,RMSE=16.04 t/ha,rRMSE=48.19%.Therefore,Landsat-8 imagery is better suited for estimating carbon storage in Chinese fir forests than Sentinel-2 imagery.The accuracy of carbon storage estimation in Chinese fir forests is significantly improved when combining Landsat 8 and Sentinel-2 data sources.Among the models used,the multilayer perceptron(MLP)model demonstrated the best performance for carbon storage estimation,with an optimal R2 of 0.62,RMSE of 11.21 t/ha,and rRMSE of 33.68%.(2)Incorporating carbon sequestration benefits can significantly enhance the expected value of forest management in forest lands.Compared to the single timber revenue scenario,the combined revenue of timber and carbon sequestration in the two accounting models increased by 2 percentage and 10 percentage,respectively.The optimal rotation period for both accounting models was 21 years.While the ideal accounting model resulted in a decline in combined profits due to carbon taxes,the annual accounting model increased the expected value of forest land by 8%compared to the ideal model.However,carbon sequestration costs were higher in this model.Forest managers can effectively reduce carbon sequestration costs by choosing the ideal accounting model.Sensitivity analysis of the rotation period showed that the optimal rotation period and carbon sequestration costs of Chinese fir plantations were negatively correlated with the discount rate and positively correlated with carbon prices.Therefore,reasonable growth of carbon sequestration prices can help extend the optimal rotation period,promote carbon sequestration growth,and achieve ecological and economic benefits in Chinese fir forestry management. |
PDF Full Text Request