Implementing conservation actions on-the-ground is not a straightforward process,especially when faced with high scientific uncertainty due to limited available information. This is especially acute in regions of the ...Implementing conservation actions on-the-ground is not a straightforward process,especially when faced with high scientific uncertainty due to limited available information. This is especially acute in regions of the world that harbor many unique species that have not been well studied,such as the alpine zone of the Hengduan Mountains of Northwest Yunnan (NWY),a global biodiversity hotspot and site of The Nature Conservancy’s Yunnan Great Rivers Project. We conducted a quantitative,but rapid regional-level assessment of the alpine flora across NWY to provide a broad-based understanding of local and regional patterns of the alpine flora,the first large-scale analysis of alpine biodiversity patterns in this region. Multivariate analyses were used to classify the major plant community types and link community patterns to habitat variables. Our analysis indicated that most species had small distributions and/or small population sizes. Strong patterns emerged with higher diversity in the more northern mountains,but beta diversity was high,averaging only 10% among sites. The ordinations indicated that elevation and geographic location were the dominant environ-mental gradients underlying the differences in the species composition among communities. The high beta diversity across the alpine of these mountains implies that conservation strategies ultimately will require the protection of large numbers of species over a large geographical area. However,prioritiza-tion should be given to areas where potential payoffs are greatest. Sites with high species richness also have a greater number of endemic species,and,by focusing efforts on these sites,conservation investments would be maximized by protecting the greatest number of unique species.展开更多
Aims Conduct a quantitative,but rapid,regional-level assessment of the alpine flora across northwest Yunnan(NWY)to provide a broadbased understanding of local and regional patterns of the composition,diversity and hea...Aims Conduct a quantitative,but rapid,regional-level assessment of the alpine flora across northwest Yunnan(NWY)to provide a broadbased understanding of local and regional patterns of the composition,diversity and health of alpine ecosystems across NWY.Methods A stratified random sampling design was employed to select sites across the different mountain ranges of NWY.Vegetation was sampled by stratifying each site by the three major alpine vegetation community types:meadow,dwarf shrub and scree.Two 50-m transects were randomly located within each community type at each sampling site with 101-m^(2) subplots systematically placed along each transect.Environmental variables were recorded at each transect.Multivariate analyses were used to classify the major plant community assemblages and link community patterns to environmental and habitat variables.Important Findings Forb species richness varied from 19 to 105 species per site(21 sites total)with an average of 59 species per site(60 m^(2) sampled per site).Most species were patchily distributed with narrow distributions and/or small population sizes;over half the species occurred at only one or two sites.Distinct species assemblages were identified in the meadow vegetation that was strongly aggregated by geographic location suggesting the presence of distinct phytogeographic zones of the meadow alpine flora.Elevation and geographic location were the dominant environmental gradients underlying the variations in species composition.Jaccard’s coefficient of similarity averaged only 10%among sites indicating there was little similarity in the alpine flora across the region.The alpine vegetation is highly heterogeneous across the complex landscape of the Hengduan Mountains of NWY.Conservation strategies need to take into account the large geographic differences in the flora to maximize protection of biodiversity.展开更多
To address the problems of insufficient dimensionality of electroencephalogram(EEG) feature extraction,the tendency to ignore the importance of different sequential data segments,and the poor generalization ability of...To address the problems of insufficient dimensionality of electroencephalogram(EEG) feature extraction,the tendency to ignore the importance of different sequential data segments,and the poor generalization ability of the model in EEG based emotion recognition,the model of convolutional neural network and bi-directional long short-term memory and self-attention(CNN+Bi LSTM+self-attention) is proposed.This model uses convolutional neural network(CNN) to extract more distinctive features from both spatial and temporal dimensions.The bi-directional long short-term memory(Bi LSTM) is used to further preserve the long-term dependencies between the temporal phases of sequential data.The self-attention mechanism can change the weights of different channels to extract and highlight important information and address the often-ignored importance of different channels and samples when extracting EEG features.The subject-dependent experiment and subject-independent experiment are performed on the database for emotion analysis using physiological signals(DEAP) and collected datasets to verify the recognition performance.The experimental results show that the model proposed in this paper has excellent recognition performance and generalization ability.展开更多
文摘Implementing conservation actions on-the-ground is not a straightforward process,especially when faced with high scientific uncertainty due to limited available information. This is especially acute in regions of the world that harbor many unique species that have not been well studied,such as the alpine zone of the Hengduan Mountains of Northwest Yunnan (NWY),a global biodiversity hotspot and site of The Nature Conservancy’s Yunnan Great Rivers Project. We conducted a quantitative,but rapid regional-level assessment of the alpine flora across NWY to provide a broad-based understanding of local and regional patterns of the alpine flora,the first large-scale analysis of alpine biodiversity patterns in this region. Multivariate analyses were used to classify the major plant community types and link community patterns to habitat variables. Our analysis indicated that most species had small distributions and/or small population sizes. Strong patterns emerged with higher diversity in the more northern mountains,but beta diversity was high,averaging only 10% among sites. The ordinations indicated that elevation and geographic location were the dominant environ-mental gradients underlying the differences in the species composition among communities. The high beta diversity across the alpine of these mountains implies that conservation strategies ultimately will require the protection of large numbers of species over a large geographical area. However,prioritiza-tion should be given to areas where potential payoffs are greatest. Sites with high species richness also have a greater number of endemic species,and,by focusing efforts on these sites,conservation investments would be maximized by protecting the greatest number of unique species.
文摘Aims Conduct a quantitative,but rapid,regional-level assessment of the alpine flora across northwest Yunnan(NWY)to provide a broadbased understanding of local and regional patterns of the composition,diversity and health of alpine ecosystems across NWY.Methods A stratified random sampling design was employed to select sites across the different mountain ranges of NWY.Vegetation was sampled by stratifying each site by the three major alpine vegetation community types:meadow,dwarf shrub and scree.Two 50-m transects were randomly located within each community type at each sampling site with 101-m^(2) subplots systematically placed along each transect.Environmental variables were recorded at each transect.Multivariate analyses were used to classify the major plant community assemblages and link community patterns to environmental and habitat variables.Important Findings Forb species richness varied from 19 to 105 species per site(21 sites total)with an average of 59 species per site(60 m^(2) sampled per site).Most species were patchily distributed with narrow distributions and/or small population sizes;over half the species occurred at only one or two sites.Distinct species assemblages were identified in the meadow vegetation that was strongly aggregated by geographic location suggesting the presence of distinct phytogeographic zones of the meadow alpine flora.Elevation and geographic location were the dominant environmental gradients underlying the variations in species composition.Jaccard’s coefficient of similarity averaged only 10%among sites indicating there was little similarity in the alpine flora across the region.The alpine vegetation is highly heterogeneous across the complex landscape of the Hengduan Mountains of NWY.Conservation strategies need to take into account the large geographic differences in the flora to maximize protection of biodiversity.
基金supported by the National Key Research and Development Program of China (No.2021YFF1200600)the National Natural Science Foundation of China (No.61806146)the Natural Science Foundation of Tianjin City (Nos.18JCYBJC95400 and 19JCTPJC56000)。
文摘To address the problems of insufficient dimensionality of electroencephalogram(EEG) feature extraction,the tendency to ignore the importance of different sequential data segments,and the poor generalization ability of the model in EEG based emotion recognition,the model of convolutional neural network and bi-directional long short-term memory and self-attention(CNN+Bi LSTM+self-attention) is proposed.This model uses convolutional neural network(CNN) to extract more distinctive features from both spatial and temporal dimensions.The bi-directional long short-term memory(Bi LSTM) is used to further preserve the long-term dependencies between the temporal phases of sequential data.The self-attention mechanism can change the weights of different channels to extract and highlight important information and address the often-ignored importance of different channels and samples when extracting EEG features.The subject-dependent experiment and subject-independent experiment are performed on the database for emotion analysis using physiological signals(DEAP) and collected datasets to verify the recognition performance.The experimental results show that the model proposed in this paper has excellent recognition performance and generalization ability.
