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.

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

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

The Wet Meadows and Fens Geodatabase of Sequoia and Kings Canyon National Parks (SEKI) was produced over a three month period (late February through early May 2013). The resulting products are intended to: (1) Identify and classify all target wetlands (wet meadows and fens) >0.5 hectare in the parks and provide an estimate of the proportion of each feature which is peat accumulating. Over 2300 polygons represent wetlands in three mapping classes: wet meadows, fen-meadow complexes, and fens. (2) Identify locations of fens within park boundaries. More than 600 fen points identify locations where fens are known or presumed to occur. Mapped polygons are based on photo interpretation of Natural Color and Color Infrared NAIP aerial photography from 2005, 2009, 2010, and 2012. Ancillary data informing mapping decisions included: the SEKI Vegetation map, 7.5-minute USGS topographic maps, the National Wetlands Inventory Wetland Layer, the SEKI Rivers and Streams Layer, and 2012 spatial data depicting known fen locations produced by the SEKI Plant Ecology program. Overall spatial accuracy of the resulting map products is equivalent to 1:24000 National Map Accuracy Standards. Overall classification accuracy was estimated to be 83% (90% CI = 78 – 89%) based on field assessments of 138 polygons. Revised mapping class and peat accumulation attributes based on accuracy assessment are included. These geospatial data are provided in Esri File Geodatabase format containing: (1) a polygon feature class depicting wet meadow and fen delineations; and (2) a point feature class indicating the locations of fens within the park boundary. Detailed metadata are provided in XML format and basic metadata are provided in PDF format. Users of these data should refer to: Pyrooz NN and Others. 2015. Wet meadow and fen mapping of Sequoia and Kings Canyon National Parks: A photo interpretation mapping project of wetland resources. Natural Resource Report. NPS/SIEN/NRR—2015/968. National Park Service. Fort Collins, Colorado. https://irma.nps.gov/DataStore/Reference/Profile/2222143

The Wet Meadows and Fens Geodatabase of Sequoia and Kings Canyon National Parks (SEKI) was produced over a three month period (late February through early May 2013). The resulting products are intended to: (1) Identify and classify all target wetlands (wet meadows and fens) >0.5 hectare in the parks and provide an estimate of the proportion of each feature which is peat accumulating. Over 2300 polygons represent wetlands in three mapping classes: wet meadows, fen-meadow complexes, and fens. (2) Identify locations of fens within park boundaries. More than 600 fen points identify locations where fens are known or presumed to occur. Mapped polygons are based on photo interpretation of Natural Color and Color Infrared NAIP aerial photography from 2005, 2009, 2010, and 2012. Ancillary data informing mapping decisions included: the SEKI Vegetation map, 7.5-minute USGS topographic maps, the National Wetlands Inventory Wetland Layer, the SEKI Rivers and Streams Layer, and 2012 spatial data depicting known fen locations produced by the SEKI Plant Ecology program. Overall spatial accuracy of the resulting map products is equivalent to 1:24000 National Map Accuracy Standards. Overall classification accuracy was estimated to be 83% (90% CI = 78 – 89%) based on field assessments of 138 polygons. Revised mapping class and peat accumulation attributes based on accuracy assessment are included. These geospatial data are provided in Esri File Geodatabase format containing: (1) a polygon feature class depicting wet meadow and fen delineations; and (2) a point feature class indicating the locations of fens within the park boundary. Detailed metadata are provided in XML format and basic metadata are provided in PDF format. Users of these data should refer to: Pyrooz NN and Others. 2015. Wet meadow and fen mapping of Sequoia and Kings Canyon National Parks: A photo interpretation mapping project of wetland resources. Natural Resource Report. NPS/SIEN/NRR—2015/968. National Park Service. Fort Collins, Colorado. https://irma.nps.gov/DataStore/Reference/Profile/2222143

These data were converted from the originally delivered Microsoft Access PLOTs database from the Vegetation Mapping Inventory Project of Sequoia and Kings Canyon National Parks. These comma-delimited data tables contain(s) vegetation mapping plot classification and accuracy assessment data, as well as summary information about the data itself. If a table is empty, then it was empty in the original database.

These data were converted from the originally delivered Microsoft Access PLOTs database from the Vegetation Mapping Inventory Project of Sequoia and Kings Canyon National Parks. These comma-delimited data tables contain(s) vegetation mapping plot classification and accuracy assessment data, as well as summary information about the data itself. If a table is empty, then it was empty in the original database.

Severn River Nature Reserve vegetation mapping was undertaken by Dr John T. Hunter in 2000 by contract for the NPWS Northern Tableland Region. The Clayton Chase additions to the Severn River Nature Reserve were mapped in 2005. Severn River Nature Reserve is located 50km north west of Glen Innes. The vegetation of Severn River Nature Reserve is described and mapped (scale 1:50000). Nine communities are defined based on PATN analysis and two specialised communities are broadened in their circumscription based on previously surveyed sites. Nine communities are mapped based on ground truthing, air photo interpretation and altitude. Most communities are of a Woodland structure, however Forests exist along with Shrublands and Herbfields. The distribution of communities is related to Aspect and Physiography mainly but also past disturbances and soil depth. Many of the communities show considerable variation and intergrade along common boundaries. A number of specialised communities are thought to be largely restricted to the reserve and nearby areas. Six communities previously mapped within Severn River Nature Reserve by Hunter (2000) were found within Clayton Chase. Two additional communities were sampled and described for this area in 2005. VIS_ID4754

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.

During summer 2019, botanists with the Maine Natural Areas Program collected data from 94 vegetation plots for plant community characterization. The sampling data were entered into the National Park Service PLOTS version 4.0 (National Park Service 2015) for analyses to characterize vegetation associations in the U.S. National Vegetation Classification. An accuracy assessment was performed on the draft version of the vegetation map layer. During the summer of 2020, field crews collected data from 107 stratified and randomly selected sites for evaluating the accuracy of the vegetation map layer for those map classes representing U.S. National Vegetation Classification associations. The accuracy assessment field data were then compared to the vegetation map data. Results from the accuracy assessment study show an overall accuracy of 87.6% (kappa index of 87.0%) based on an analysis of data from 105 of the 107 accuracy assessment sites.