Abstract
Fires are frequent in boreal forests affecting forest areas. The detection of forest disturbances and the monitoring of forest restoration are critical for forest management. Vegetation phenology information in remote sensing images may interfere with the monitoring of vegetation restoration, but little research has been done on this issue. Remote sensing and the geographic information system (GIS) have emerged as important tools in providing valuable information about vegetation phenology. Based on the MODIS and Landsat time-series images acquired from 2000 to 2018, this study uses the spatio-temporal data fusion method to construct reflectance images of vegetation with a relatively consistent growth period to study the vegetation restoration after the Greater Hinggan Mountain forest fire in the year 1987. The influence of phenology on vegetation monitoring was analyzed through three aspects: band characteristics, normalized difference vegetation index (NDVI) and disturbance index (DI) values. The comparison of the band characteristics shows that in the blue band and the red band, the average reflectance values of the study area after eliminating phenological influence is lower than that without eliminating the phenological influence in each year. In the infrared band, the average reflectance value after eliminating the influence of phenology is greater than the value with phenological influence in almost every year. In the second shortwave infrared band, the average reflectance value without phenological influence is lower than that with phenological influence in almost every year. The analysis results of NDVI and DI values in the study area of each year show that the NDVI and DI curves vary considerably without eliminating the phenological influence, and there is no obvious trend. After eliminating the phenological influence, the changing trend of the NDVI and DI values in each year is more stable and shows that the forest in the region was impacted by other factors in some years and also the recovery trend. The results show that the spatio-temporal data fusion approach used in this study can eliminate vegetation phenology effectively and the elimination of the phenology impact provides more reliable information about changes in vegetation regions affected by the forest fires. The results will be useful as a reference for future monitoring and management of forest resources.
Highlights
Forests play an irreplaceable role in maintaining the ecological balance of the terrestrial biosphere due to their wide coverage, complex distribution, and species diversity [1,2], multi-function, and multi-value characteristics [3].Fires are one of the serious disturbances globally and are prevalent in boreal forests [4]
The date corresponding to the midpoint of the vegetation growth period in each year was obtained from the vegetation index
Taking the forest restoration in the Greater Hinggan Mountain area after the “5.6 fire” in 1987 as an example, based on the MODIS and Landsat time-series images acquired from 2000 to 2018, this study took the midpoint of the vegetation growth period of each year as the target date and used the STARFM fusion algorithm to construct reflectance images of vegetation with relatively consistent growth periods
Summary
Forests play an irreplaceable role in maintaining the ecological balance of the terrestrial biosphere due to their wide coverage, complex distribution, and species diversity [1,2], multi-function, and multi-value characteristics [3]. Fires are one of the serious disturbances globally and are prevalent in boreal forests [4]. The burning of forests severely causes local economic losses. Forest fires have long term environmental and climate impacts. A certain frequency and intensity of fire can maintain the balance of forest ecology and play an essential role in preserving biodiversity. With climate change and global warming, the frequency of forest fires is increasing and receives increasing attention as an integral part of global environmental change studies [11,12]
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
