The repository contains data and scripts used in the analysis done for the publication 'Moving on up: A two-step scaling approach for grassland vegetation structure'. This includes the following foldersms_orthos: unoccupied aerial vehicle (UAV) multispectral orthomosaicsUAV_point_clouds: UAV structure-from motion point cloudsS2_ROI_AOA_predictions: Sentinel-2 predictions and area of applicability (AOA) layers for vegetation cover (%) and height (m) between 2019 and 2024 across the region of interest (ROI). These include layers from models developed with either spatial leave-site-out (i.e., space) or stratified (i.e., strat) cross-validation.S2_predictions: Sentinel-2 predictions of vegetation cover (%), height (m) and volume (derived from cover and height layers) between 2019-2024.These rasters extend beyond the ROI to the entirety of the three study provinces. We also have included the annual rate of change (i.e., slope of change per pixel from 2019 to 2024) for each vegetation characteristic.Rscripts: R scripts of all analysis. Workflows to answer each research question can be found by the script name. For example, the script for research question one begins with "RQ1_".csv: Plot-level field measurements.shapes: Shape files of the UAV survey boundary, quadrats within UAV surveys, and the ROI.

Asiatic black bear is a vulnerable species under the IUCN Red List and an umbrella species across the ecosystems in Asia. Here we present results from the first distribution-wide map of Asiatic black bear (Ursus thibetanus) habitat suitability based on high resolution remotely sensed data. We collected 1008 bear presence points between 2005–2022 to train the habitat suitability model. The study area was demarcated by creating a bounding box of 150 km² area around these presence points. To assess the effects of spatial scale, we used land cover variables with higher resolution based on Copernicus product (100 m) and coarser resolution based on MODIS data (250m) to drive the model. Our models projected suitable habitat of 1,697,647 km² (under Model 1) and 1,742,921 km² (under Model 2) for Asiatic black bear across all countries with existing records in IUCN database. Among those countries, we found that China has the maximum habitat (690,310-700,236 km²) while Bangladesh has the least (1,014-1,420 km²) amount of suitable habitat areas for black bears. Additionally, our model projected suitable habitat of 81,274 km² (under Model 1) and 97,907 km² (under Model 2) for Asiatic black bear in countries and regions outside of the IUCN documented range i.e., the Philippines, Oman, Sri Lanka, the United Arab Emirates, and southern parts of India. Though black bears are not yet reported in these areas, potentially due to a combination of historic over-hunting and geographic isolation, the highlighted regions show the potential distribution of the species where habitat conditions can support viable populations. These areas could be having as reintroduction sites corridor development in the future. Our study provides new insights into the conservation status of Asiatic black bear and the environmental variables that determine their distribution at a broad scale. We also provide information for further monitoring, establishment of protected areas, transboundary conservation, and enhancement of habitat connectivity for black bears among regions where high value black bear habitat remains.

Asiatic black bear is a vulnerable species under the IUCN Red List and an umbrella species across the ecosystems in Asia. Here we present results from the first distribution-wide map of Asiatic black bear (Ursus thibetanus) habitat suitability based on high resolution remotely sensed data. We collected 1008 bear presence points between 2005–2022 to train the habitat suitability model. The study area was demarcated by creating a bounding box of 150 km² area around these presence points. To assess the effects of spatial scale, we used land cover variables with higher resolution based on Copernicus product (100 m) and coarser resolution based on MODIS data (250m) to drive the model. Our models projected suitable habitat of 1,697,647 km² (under Model 1) and 1,742,921 km² (under Model 2) for Asiatic black bear across all countries with existing records in IUCN database. Among those countries, we found that China has the maximum habitat (690,310-700,236 km²) while Bangladesh has the least (1,014-1,420 km²) amount of suitable habitat areas for black bears. Additionally, our model projected suitable habitat of 81,274 km² (under Model 1) and 97,907 km² (under Model 2) for Asiatic black bear in countries and regions outside of the IUCN documented range i.e., the Philippines, Oman, Sri Lanka, the United Arab Emirates, and southern parts of India. Though black bears are not yet reported in these areas, potentially due to a combination of historic over-hunting and geographic isolation, the highlighted regions show the potential distribution of the species where habitat conditions can support viable populations. These areas could be having as reintroduction sites corridor development in the future. Our study provides new insights into the conservation status of Asiatic black bear and the environmental variables that determine their distribution at a broad scale. We also provide information for further monitoring, establishment of protected areas, transboundary conservation, and enhancement of habitat connectivity for black bears among regions where high value black bear habitat remains.

The repository contains data and scripts used in the analysis done for the publication 'Moving on up: A two-step scaling approach for grassland vegetation structure'. This includes the following foldersms_orthos: unoccupied aerial vehicle (UAV) multispectral orthomosaicsUAV_point_clouds: UAV structure-from motion point cloudsS2_ROI_AOA_predictions: Sentinel-2 predictions and area of applicability (AOA) layers for vegetation cover (%) and height (m) between 2019 and 2024 across the region of interest (ROI). These include layers from models developed with either spatial leave-site-out (i.e., space) or stratified (i.e., strat) cross-validation.S2_predictions: Sentinel-2 predictions of vegetation cover (%), height (m) and volume (derived from cover and height layers) between 2019-2024.These rasters extend beyond the ROI to the entirety of the three study provinces. We also have included the annual rate of change (i.e., slope of change per pixel from 2019 to 2024) for each vegetation characteristic.Rscripts: R scripts of all analysis. Workflows to answer each research question can be found by the script name. For example, the script for research question one begins with "RQ1_".csv: Plot-level field measurements.shapes: Shape files of the UAV survey boundary, quadrats within UAV surveys, and the ROI.

Myanmar Forest Change dataset is produced by TerraPulse and Smithsonian Conservation Biology Institute. The dataset is based on Landsat satellite image archivebetween 1979 and 2019. The dataset provides separate layers to quantify theforest loss year, forest gain year and forest detection year, all at 30 meter×30 meter pixel resolution.

Myanmar Forest Change dataset is produced by TerraPulse and Smithsonian Conservation Biology Institute. The dataset is based on Landsat satellite image archivebetween 1984 and 2019. The dataset provides separate layers to quantify theforest loss year, forest gain year and forest detection year, all at 30 meter×30 meter pixel resolution.

Myanmar Forest Change dataset is produced by TerraPulse and Smithsonian Conservation Biology Institute. The dataset is based on Landsat satellite image archivebetween 1979 and 2019. The dataset provides separate layers to quantify theforest loss year, forest gain year and forest detection year, all at 30 meter×30 meter pixel resolution.

Myanmar Forest Change dataset is produced by TerraPulse and Smithsonian Conservation Biology Institute. The dataset is based on Landsat satellite image archivebetween 1979 and 2019. The dataset provides separate layers to quantify theforest loss year, forest gain year and forest detection year, all at 30 meter×30 meter pixel resolution.