Relational database with data supporting an ecological land survey and land cover mapping of Selawik National Wildlife Refuge.

developing effective habitat conservation and management strategies. The relationship between available habitat and waterfowl numbers obtained from aerial survey transects is not well studied. To determine these relationships, multispectral SPOT-5 satellite imagery acquired for Sequoyah National Wildlife Refuge close to the time of waterfowl surveys was used to map habitat conditions. Robust Random Forest classification trees were used to model 16 land cover types using 416 reference locations collected in the field or derived from aerial photos close to or during waterfowl survey dates. The normalized difference vegetation index (NDVI), normalized difference water index (NDWI) and a simple ratio (SR) of red and near infrared bands were used to enhance classification accuracy for key habitat areas and abundance of water. Terrain variables such as slope, solar illumination and cosine transformed aspect derived from a digital elevation model (DEM) were also used to enhance habitat classification. Random Forest (RF) models were also compared to support vector machines (SVM) and cforest (CF) conditional inference trees. We used error matrices and the Kappa agreement statistic (K) to compare model results from each classifier. Results indicated that a tuned RF classifier showed better performance (K=0.73) than SVM (K=0.65) and unbiased cforest trees (K=0.63). Overall class agreement between similar RF and cforest models, designed to reduce predictor variable selection bias, was also relatively low (K=0.47). A final tuned RF model was selected resulting in 75% accuracy overall and was used to map habitat types for the refuge and surrounding landscape. We found that elevation and minimum noise fraction (MNF) bands were the most important predictor variables. MNF bands can help to reduce the number of correlated variables entering into a classification model when a larger number of correlated spectral bands are used. Similar forest types such as riverine, bottomland hardwood, and floodplain forest showed the greatest misclassification error. Overall, the RF model and SPOT-5 leaf-off imagery generated accurate land cover data for assessing habitat conditions during waterfowl surveys.

developing effective habitat conservation and management strategies. The relationship between available habitat and waterfowl numbers obtained from aerial survey transects is not well studied. To determine these relationships, multispectral SPOT-5 satellite imagery acquired for Sequoyah National Wildlife Refuge close to the time of waterfowl surveys was used to map habitat conditions. Robust Random Forest classification trees were used to model 16 land cover types using 416 reference locations collected in the field or derived from aerial photos close to or during waterfowl survey dates. The normalized difference vegetation index (NDVI), normalized difference water index (NDWI) and a simple ratio (SR) of red and near infrared bands were used to enhance classification accuracy for key habitat areas and abundance of water. Terrain variables such as slope, solar illumination and cosine transformed aspect derived from a digital elevation model (DEM) were also used to enhance habitat classification. Random Forest (RF) models were also compared to support vector machines (SVM) and cforest (CF) conditional inference trees. We used error matrices and the Kappa agreement statistic (K) to compare model results from each classifier. Results indicated that a tuned RF classifier showed better performance (K=0.73) than SVM (K=0.65) and unbiased cforest trees (K=0.63). Overall class agreement between similar RF and cforest models, designed to reduce predictor variable selection bias, was also relatively low (K=0.47). A final tuned RF model was selected resulting in 75% accuracy overall and was used to map habitat types for the refuge and surrounding landscape. We found that elevation and minimum noise fraction (MNF) bands were the most important predictor variables. MNF bands can help to reduce the number of correlated variables entering into a classification model when a larger number of correlated spectral bands are used. Similar forest types such as riverine, bottomland hardwood, and floodplain forest showed the greatest misclassification error. Overall, the RF model and SPOT-5 leaf-off imagery generated accurate land cover data for assessing habitat conditions during waterfowl surveys.

This data set provides two landcover and vegetation maps for the Seward Peninsula, Alaska. These maps were produced from existing maps, Landsat imagery, and color infrared aerial photography covering the period 1976-06-01 to 1999-09-01.

This data set provides two landcover and vegetation maps for the Seward Peninsula, Alaska. These maps were produced from existing maps, Landsat imagery, and color infrared aerial photography covering the period 1976-06-01 to 1999-09-01.

This geodatabase contains three (3) rasters, two (2) of which represent landcover for Selawik National Wildlife Refuge and surrounding areas. The third raster contains plot-based ground characteristics for pixels classified with high confidence. The two landcover rasters contain attribute information for soils, vegetation, and ecotypes; and differ slightly in their classifications because one encompasses a broader geographic area (lc_arcn), and therefore some classes are more generalized than in the other (lc_nokose). The classification of local-scale ecosystems (ecotypes) combines physiography (e.g., riverine, coastal), topography (DEM), geology and vegetation from the landcover spectral database derived from the satellite image processing. These layers are used to model ecotypes in a way that best partitions geomorphic, hydrologic, pedologic, and vegetative characteristics. Map projection: Albers Alaska, NAD 83, meters. ***NOTE*** The lc_nokose raster was used for the landcover classifications in the final report, as it is more specific to Selawik National Wildlife Refuge than the other landcover raster (lc_arcn), which includes some surrounding National Park Service lands and differs slightly in its classifications at the pixel level. PDF maps are provided here for reference to help visualize what the data look like before downloading. Full resolution maps can be viewed in the final report (ServCat #49603).

This geodatabase contains three (3) rasters, two (2) of which represent landcover for Selawik National Wildlife Refuge and surrounding areas. The third raster contains plot-based ground characteristics for pixels classified with high confidence. The two landcover rasters contain attribute information for soils, vegetation, and ecotypes; and differ slightly in their classifications because one encompasses a broader geographic area (lc_arcn), and therefore some classes are more generalized than in the other (lc_nokose). The classification of local-scale ecosystems (ecotypes) combines physiography (e.g., riverine, coastal), topography (DEM), geology and vegetation from the landcover spectral database derived from the satellite image processing. These layers are used to model ecotypes in a way that best partitions geomorphic, hydrologic, pedologic, and vegetative characteristics. Map projection: Albers Alaska, NAD 83, meters. ***NOTE*** The lc_nokose raster was used for the landcover classifications in the final report, as it is more specific to Selawik National Wildlife Refuge than the other landcover raster (lc_arcn), which includes some surrounding National Park Service lands and differs slightly in its classifications at the pixel level. PDF maps are provided here for reference to help visualize what the data look like before downloading. Full resolution maps can be viewed in the final report (ServCat #49603).

This data set consists of current 2019 and historic ~1940 vegetative cover classification of Chassahowitzka NWR based on the Florida Natural Area Inventory classification. Also vegetation plot locations and associated vegetation variables to track changes in the vegetative community collected in 2019 ae provided to support the larger project. Exotic plant and rare animal ocations are provided. Data are presented as point and polygon shapefiles in WGS 1984 lat/long. A pdf metafile is included.

This data set consists of current 2019 and historic ~1940 vegetative cover classification of Chassahowitzka NWR based on the Florida Natural Area Inventory classification. Also vegetation plot locations and associated vegetation variables to track changes in the vegetative community collected in 2019 ae provided to support the larger project. Exotic plant and rare animal ocations are provided. Data are presented as point and polygon shapefiles in WGS 1984 lat/long. A pdf metafile is included.