近20年来肥城市林地时空变化及其驱动因子OA北大核心CSTPCD

[Objective]Monitoring forest resources and its spatio-temporal changes using time-series remote sensing datasets and extracting the driving factors has a great significance in achieving scientific management and efficient utilization of regional forest land resources.[Method]This study identified Feicheng City,Shandong Province as a case study,extracted the distribution information of land type from 2000 to 2020 by decision tree classification based on the 21-period Landsat remote sensing datasets,which was preprocessed by radiation calibration and atmospheric correction,and also analyzed spatio-temporal change and its driving factors of forest based on the renovated IPAT model.[Result]The extracted forest land information based on decision tree classification achieved a high accuracy,which yield more than 94.7%of user's accuracy.In the past 20 years,the area of forest land fluctuated with a net increase of 17 463.54 hm2,and the increased forest land was concentrated in the central parts and northern mining area of Feicheng City,while the decreased forest land was centralized in the edge part of the central forest land,thus indicating the spatial characteristics of concentrated distribution in the central and northern parts and sporadic distribution in other areas.The renovated IPAT model showed that the endowment value of forest land was the main driving factor for the area change in Feicheng City(the contribution larger than 80.3%),and the degree of affluence was the secondary driving factor.The contribution of both factors exceeded 92%of the total contribution of the four factors in the renovated IPAT model.In contrast,forest industry and the population posed little influence.[Conclusion]Forest resources could be accurately extracted and monitored based on decision tree classification,and the endowment value of forest land has significant influences on the spatio-temporal evolution of forest land.