Research On The Valuation Of Forest Resources Assets Based On Sentinel-2 Satellite Image

At present,global development has entered a new stage,the importance of natural resources in society is increasing,and the management of natural resources in various countries has taken a new step,forming models with their own characteristics in policy making,market building and system construction.As an important puzzle of natural resource assets,forest resource assets provide forest products and various ecosystem services for human beings and have irreplaceable values.This thesis takes forest resource assets in Ji’an City as the research object,starts from the value connotation of forest resource assets,builds a framework and system of forest resource asset value assessment based on both economic and ecological value perspectives,and introduces a forest stock identification method based on remote sensing technology,with a view to further broadening the research path on the basis of previous studies,innovating ideas and methods of forest resource asset economic value accounting,and at the same time providing support for The study aims to further broaden the research path on the basis of previous studies,innovate the ideas and methods of economic value accounting of forest resources assets,and provide support for the compilation of forest resources balance sheet to reflect the practical value of the study.In the context of the new era,the value connotation of natural resources should be reflected in a new way,and even the unexploited natural resources have value as well.This thesis firstly compares the current research progress at home and abroad to clarify the value connotation and categories of natural resources,and on this basis,determines the value scope of the accounting in this study,establishes the accounting framework,and starts the specific accounting from economic value and ecological value respectively.In the accounting of economic value,forest stock is taken as the basis of forest value calculation,and based on the remote sensing data of Sentinel 2,multiple linear regression,random forest regression and artificial neural network are selected based on the idea of machine learning for modeling and estimation,and the economic value of forest resource assets in Ji’an City is derived by the market value method.The ecological value was calculated by combining the functional value method with the equivalent factor method,and the amount of water connotation,soil conservation,carbon sequestration and oxygen release,and atmospheric purification functions of the assessed objects were calculated,and then the ecological value was derived.The main conclusions of this study are as follows:(1)The accuracy of each accumulation estimation model differs significantly,and the annual economic value of forest resource assets is high with obvious spatial clustering characteristics.The correlation between image single-band emissivity,texture feature factor,vegetation index,slope and forest stock is significant,while the effect of slope direction on forest stock is small.In addition,among the three models established based on the sample points,the multiple stepwise regression model performed the worst;while the artificial neural network model had better accuracy;the random forest model with the number of decision trees and the maximum decision tree depth of 200 and 3,respectively,had the highest accuracy and the best modeling effect,and could be used as the choice for forest stock estimation in Ji’an.The final result is that the forest stock in Ji’an in 2019 is about 130 million cubic meters,and the economic value is about 5.802 billion yuan.The areas with high forest stock and value are mainly distributed in the areas where the Luoxiao Mountain Range and the Summer Mountain Range are located on the east and west sides,and these areas have a wide distribution of mountains and a large woodland area as well as good forest growth conditions.(2)Forest resource assets provide huge amount of ecological service functions.12.355 billion cubic meters of water connotation;8.0597 million tons of soil conservation function,of which 64,235.88 tons,10,800.01 tons and 127,343.41 tons of nutrient nitrogen,phosphorus and potassium are maintained respectively;annual carbon sequestration of forest resource assets in2019 is 4,922,900 tons and 13,180,000 tons of oxygen release;the annual SO2 absorption function is 257,900 tons and the annual dust retention is 39,999,200 tons.(3)The difference in the high and low value of each type of ecosystem services is significant.the total ecological value of forest resource assets in Ji’an City in 2019 was 35.667 billion yuan.The value of carbon fixation and oxygen release is the highest,amounting to19.087 billion yuan,accounting for 53.51% of the total ecological value of the city,respectively;followed by the value of water connotation,8.278 billion yuan,accounting for 23.21%;again,the value of purifying the atmosphere,6.308 billion yuan,accounting for 17.69%;the least is the value of soil conservation,1.994 billion yuan,accounting for 5.59% of the total ecological value of the city,respectively.5.59%.(4)In terms of total value.the total economic and ecological value of forest resource assets in Ji’an reached 41.469 billion yuan in 2019,of which the economic value was 5.802 billion yuan,accounting for 13.99% of the total value,and the ecological value was 35.667 billion yuan,accounting for 86.01% of the total value,with ecological services providing a higher value compared to physical forest products.The unit area value of forest resource assets is22,947.43 Yuan/ha.This study can enrich the theoretical research results on the grid scale and provide the theoretical basis for realizing the protection and rational development of forest resources.On the other hand,the introduction of remote sensing technology-based forest stock assessment is a new method in the related accounting field.This study can overcome the deficiencies of the traditional data base of economic value of forest resources,introduce the method of forest stock assessment based on remote sensing technology,ensure the accuracy of research data while taking into account the efficiency of data acquisition,measure the physical volume,which is the basis of economic value measurement,and further innovate the existing method of forest resource asset value accounting.