Abstract
Satellite data becomes an important tool for monitoring global change in forest cover. Further development of remote sensing technologies creates opportunities for solving more complex problems requiring multi-time analysis of satellite data. Assessment of success reforestation after a disturbance in forest cover is such an important task. The traditional method of an assessment of successful reforestation is laying out the ground plots, which task requires significant time and resources. Fieldworks and transfer of land to forest cover land is carried out according to the method, which is developed by the Federal Agency for Forestry of Russia. This method has various criteria of success reforestation for every region. Arkhangelsk region, Vologda region and Republic of Karelia became the territories for research. Forest vegetation of this region belongs to the taiga zone and is divided into five groups: the area of pre-tundra forests and sparse taiga, northern taiga, middle taiga and south taiga. International forest classification relates this area to boreal forest. The task of transfer land to forest cover land can be optimized by using remote sensing data. This research shows analysis of recovery of the normalized difference vegetation index, the shortwave vegetation index and the normalized burn ratio in the framework of reforestation objects. Filed data was collected for every object and this data includes a number of young trees, average height and species composition. Processing of a considerable number of satellite imageries requires significant computing power because of the Google Earth Engine platform using for analysis data. The most suitable index was chosen in the analysis of the obtained data for the development of an automatic method for transfer land to forest cover land. The most suitable index for dividing lands on forest cover and nonforest cover lands is the shortwave vegetation index. Optimal threshold for transferring land is achievement of recovery index of 80 % from initial values before disturbance. The automatic method was developed using unsupervised classification and threshold values of recovery index.For citation: Karpov A., Waske B. Method for Transferring Non-Forest Cover to Forest Cover Land Using Landsat Imageries. Lesnoy Zhurnal [Russian Forestry Journal], 2020, no. 3, pp. 83–92. DOI: 10.37482/0536-1036-2020-3-83-92Funding: This research was undertaken as part of the project “Forest Monitoring in the Arkhangelsk Region, Using Multisensory Remote Sensing Data” funded by the Russian Ministry of Education and Science and the German Academic Exchange Service in the framework of the Michail‐Lomonosov‐Programme (project no. АААА-А19-119020590052-2).
Highlights
Land use change is the major driver of global environmental change resulting in loss of biological diversity, changing the global carbon cycle, affecting local climate and the hydrological cycle [15]
Using of multitemporal analysis of the normalized difference vegetation index (NDVI), the shortwave vegetation index (SWVI), the normalized burn ratio (NBR) and components of Tasseled Cap transformations is usual for assessment vegetation cover, but the spectral index values depend on the phenology of vegetation and climatic conditions [6, 7]
We propose to calculate the degree of reforestation as a percentage according to the following formula: R = I obs, I pre where Iobs – index at the moment of field observation; Ipre – pre-disturbance index
Summary
Land use change is the major driver of global environmental change resulting in loss of biological diversity, changing the global carbon cycle, affecting local climate and the hydrological cycle [15]. Achievement of 80 % of the initial value of the spectral index requires a different period for the extremely cold and cold temperature zones for the North American boreal forest region. Analysis Tasseled Cap component shows that spectral indeces after disturbance can change for three decades in the territory of the North American boreal forest region. Using the spectral index requires field data for calibration indices for the assessment of reforestation and finding a relationship between a value of the index and real conditions of forest regeneration such as structure of recovered vegetation, number, and tree species per pixel and tree canopy. The practical issue of using satellite imageries for the assessment reforestation is finding the threshold for transfer cutting and burned areas to forest cover land. The procedure of state monitoring defines required conditions such as number, average height of trees and parameters of ground plots for transferring non-forest to forest land (Table). High-resolution imageries of Google and Yandex services and Hansen’s forest loss dataset were used for drawing the border of cuttings and burned areas
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
