Since the launch of the Google Earth Engine(GEE)cloud platform in 2010,it has been widely used,leading to a wealth of valuable information.However,the potential of GEE for forest resource management has not been fully...Since the launch of the Google Earth Engine(GEE)cloud platform in 2010,it has been widely used,leading to a wealth of valuable information.However,the potential of GEE for forest resource management has not been fully exploited.To extract dominant woody plant species,GEE combined Sen-tinel-1(S1)and Sentinel-2(S2)data with the addition of the National Forest Resources Inventory(NFRI)and topographic data,resulting in a 10 m resolution multimodal geospatial dataset for subtropical forests in southeast China.Spectral and texture features,red-edge bands,and vegetation indices of S1 and S2 data were computed.A hierarchical model obtained information on forest distribution and area and the dominant woody plant species.The results suggest that combining data sources from the S1 winter and S2 yearly ranges enhances accuracy in forest distribution and area extraction compared to using either data source independently.Similarly,for dominant woody species recognition,using S1 winter and S2 data across all four seasons was accurate.Including terrain factors and removing spatial correlation from NFRI sample points further improved the recognition accuracy.The optimal forest extraction achieved an overall accuracy(OA)of 97.4%and a maplevel image classification efficacy(MICE)of 96.7%.OA and MICE were 83.6%and 80.7%for dominant species extraction,respectively.The high accuracy and efficacy values indicate that the hierarchical recognition model based on multimodal remote sensing data performed extremely well for extracting information about dominant woody plant species.Visualizing the results using the GEE application allows for an intuitive display of forest and species distribution,offering significant convenience for forest resource monitoring.展开更多
Wetland restoration had been implemented for more than two decades in Sanjiang Plain,Northeast China.To assess the restoration effi ciency of wetland vegetation,we investigated plants composition of community,plant sp...Wetland restoration had been implemented for more than two decades in Sanjiang Plain,Northeast China.To assess the restoration effi ciency of wetland vegetation,we investigated plants composition of community,plant species diversity and aboveground biomass of restored sites in a chronosequence of restoration(1,2,3,6,8,11,15 and 25 years)in the Sanjiang Nature Reserve.As comparison,we also investigated the same factors in a cropland and a natural marsh adjacent to the restored sites.The results showed that wetland plant species can invade quickly after croplands were abandoned when there were suitable hydrology conditions.On the early stage of the restoration,weeds were main plant species in the restored sites.Wetland species appeared at the same time but diff ered from the dominant species from the adjacent natural marshes.Common native wetland species could dominance the community after 3-year restoration.Species richness and diversity increased on the early stage,and then decreased to the similar level of the natural marsh with the extension of restoration.Plant biomass could restore easier than the species composition and diversity.Our results indicated that plant species composition and diversity of abandoned reclaimed wetlands can restore gradually by natural succession in Sanjiang Plain.However,25-year restoration site had similarity index of only 56%with the natural marsh,which revealed that two decades are not enough for complete restoration of vegetation.展开更多
Twenty-three secondary forest communities with different structure were selected in Mao＇er Mountain National Park of Heilongjiang Province, China to study the relationship between diversity of forest plant species an...Twenty-three secondary forest communities with different structure were selected in Mao＇er Mountain National Park of Heilongjiang Province, China to study the relationship between diversity of forest plant species and environmental gradient. The forest plant species diversity was analyzed by the diversity index, and the environmental factors was quantified by the method of Whittaker＇s quantification of environmental gradient. Meanwhile, β-diversity indexes of communities were calculated with similar measurements. The results showed that the Shannon-wiener diversity index of forest plant species increased with the increase of the environmental gradient, and the β-diversity indexes of communities showed a liner increase along with the change of environmental gradient.展开更多
基金supported by the National Technology Extension Fund of Forestry,Forest Vegetation Carbon Storage Monitoring Technology Based on Watershed Algorithm ([2019]06)Fundamental Research Funds for the Central Universities (No.PTYX202107).
文摘Since the launch of the Google Earth Engine(GEE)cloud platform in 2010,it has been widely used,leading to a wealth of valuable information.However,the potential of GEE for forest resource management has not been fully exploited.To extract dominant woody plant species,GEE combined Sen-tinel-1(S1)and Sentinel-2(S2)data with the addition of the National Forest Resources Inventory(NFRI)and topographic data,resulting in a 10 m resolution multimodal geospatial dataset for subtropical forests in southeast China.Spectral and texture features,red-edge bands,and vegetation indices of S1 and S2 data were computed.A hierarchical model obtained information on forest distribution and area and the dominant woody plant species.The results suggest that combining data sources from the S1 winter and S2 yearly ranges enhances accuracy in forest distribution and area extraction compared to using either data source independently.Similarly,for dominant woody species recognition,using S1 winter and S2 data across all four seasons was accurate.Including terrain factors and removing spatial correlation from NFRI sample points further improved the recognition accuracy.The optimal forest extraction achieved an overall accuracy(OA)of 97.4%and a maplevel image classification efficacy(MICE)of 96.7%.OA and MICE were 83.6%and 80.7%for dominant species extraction,respectively.The high accuracy and efficacy values indicate that the hierarchical recognition model based on multimodal remote sensing data performed extremely well for extracting information about dominant woody plant species.Visualizing the results using the GEE application allows for an intuitive display of forest and species distribution,offering significant convenience for forest resource monitoring.
基金the National Natural Science Foundation of China(31870443)the Fundamental Research Funds for the Central Universities(2572016CA03)the Heilongjiang Postdoctoral Funds(LBH-Z16014).
文摘Wetland restoration had been implemented for more than two decades in Sanjiang Plain,Northeast China.To assess the restoration effi ciency of wetland vegetation,we investigated plants composition of community,plant species diversity and aboveground biomass of restored sites in a chronosequence of restoration(1,2,3,6,8,11,15 and 25 years)in the Sanjiang Nature Reserve.As comparison,we also investigated the same factors in a cropland and a natural marsh adjacent to the restored sites.The results showed that wetland plant species can invade quickly after croplands were abandoned when there were suitable hydrology conditions.On the early stage of the restoration,weeds were main plant species in the restored sites.Wetland species appeared at the same time but diff ered from the dominant species from the adjacent natural marshes.Common native wetland species could dominance the community after 3-year restoration.Species richness and diversity increased on the early stage,and then decreased to the similar level of the natural marsh with the extension of restoration.Plant biomass could restore easier than the species composition and diversity.Our results indicated that plant species composition and diversity of abandoned reclaimed wetlands can restore gradually by natural succession in Sanjiang Plain.However,25-year restoration site had similarity index of only 56%with the natural marsh,which revealed that two decades are not enough for complete restoration of vegetation.
基金Foundation project： The paper was supported by National Natural Science Foundation of China （39899370）.
文摘Twenty-three secondary forest communities with different structure were selected in Mao＇er Mountain National Park of Heilongjiang Province, China to study the relationship between diversity of forest plant species and environmental gradient. The forest plant species diversity was analyzed by the diversity index, and the environmental factors was quantified by the method of Whittaker＇s quantification of environmental gradient. Meanwhile, β-diversity indexes of communities were calculated with similar measurements. The results showed that the Shannon-wiener diversity index of forest plant species increased with the increase of the environmental gradient, and the β-diversity indexes of communities showed a liner increase along with the change of environmental gradient.
