Analysis On Land Resource Asset Accounting In Nanjing Based On Google Earth Engine

With the rapid development of China’s economy,a series of problems such as waste of resources,environmental pollution and ecological damage have arisen,which seriously restrict the sustainable development of China’s economy.Against this background,the Third Plenary Session of the 18 th Central Committee of the CPC proposed " compiling a balance sheet of natural resources",aiming to strengthen the accounting and management of natural resources,protect the ecological environment,and achieve sustainable development of resources,environment and economy and society.Natural resource asset accounting is the primary task of compiling natural resource balance sheet.In order to promote the construction of ecological civilization and establish China’s natural resources balance sheet system,it is also necessary to conduct scientific accounting of China’s natural resources assets in China.Land resources assets accounting is an important part of natural resources accounting.Land resources assets accounting includes physical quantity accounting and value quantity accounting.The current land resource asset accounting has problems such as difficulty in data acquisition,low processing efficiency,and unclear accounting method system.The GEE cloud platform has rich geospatial data sets and powerful cloud computing capability,which can timely and efficiently grasp the dynamic changes of land resources,and provide data support and technical support for the accounting of physical and value quantities of land resources assets.Therefore,based on the GEE cloud platform,this paper adopts the combination of remote sensing data and auxiliary feature data for the extraction of multi-temporal natural resource asset information and proposes an operable land resource asset value accounting system to account for the value of land resource assets.In this paper,taking Nanjing as an example,based on the GEE cloud platform,the Landsat 8 surface reflectance dataset with 30 m resolution is used as the research data.By combining remote sensing image data and other auxiliary feature data,the random forest algorithm is used to classify land resources in Nanjing,and the influence of different auxiliary features on improving the accuracy of the random forest classification model is comprehensively analyzed.Secondly,the current situation and dynamic changes of the physical quantity of land resource assets in Nanjing are analyzed from the perspective of time and space with the help of Arc GIS software.Finally,based on physical quantity accounting,an operable land resource asset value accounting system is proposed from the economic,ecological and social functions of land resources to realize a multidimensional land resource asset value accounting study and finally obtain the land resource asset value in Nanjing.The findings of this study are as follows:(1)Based on the massive database and strong data capability of GEE,five features including spectral features,spectral indics,texture features,topographic features and soil feature are obtained from the open source datasets.Then the RFE method is used for feature selection and auxiliary features are optimized.The optimized auxiliary features were used as classification input features(M1-M6),and the random forest algorithm was used to classify the land resources in Nanjing.The results show that the overall accuracy of classification can be improved regardless of the type of auxiliary features,with the classification model containing all features having the highest overall accuracy at 94%.(2)From the perspective of land use dynamics,land comprehensiveness,and land use transfer characteristics,the current situation and changes of land resources and assets in Nanjing are comprehensively analyzed.The results show that the overall structure of land use types in Nanjing is relatively stable during 2015-2018,among which the proportion of arable land is maintained at about 50%.The area of construction land,grassland and unused land increased,while the area of cultivated land,forest land and water area decreased.While cultivated land flows into construction land,a large amount of construction land flows into arable land,and the main reason for these changes is that in the process of urban expansion,there may be some industrial and mining land planning and finishing making some construction land reclaimed as arable land.The overall trend of land use degree in Nanjing during 2015-2018 is on the rise,which indicates that land use in Nanjing is generally in the development period and land resources have development potential.At the same time,the increase of land use degree will inevitably increase the pressure on resources and ecological environment.Therefore,the intensity of land development should be controlled,land space should be reasonably planned,and the land resource management system should be strengthened.(3)Based on land use information,the value of land resource assets in Nanjing was accounted for from three aspects: economic value,ecological value and social value using the revenue reduction method,equivalent factor method and alternative market method,respectively.From the study of the time-series dynamics of land resource values: during 2015-2018,the overall land resource values in Nanjing showed an increasing trend,and the land resource asset values increased by a total of 4,702.92 billion yuan,or 60.05%.In terms of the composition of land resource asset value,the asset value of arable land resources accounts for 81.26% of the total value of land resource assets in Nanjing,which dominates the land resources in Nanjing.In terms of the contribution of each district to the asset value of land resources in Nanjing,the highest contribution is made by Liuhe District and the lowest contribution is made by Gulou District.The results of the study are important scientific guidance for changing the concept of economic development,rational planning of land resources and strengthening the sustainable development of ecosystems.