Description: This data represents an assessment of forest composition used in the ecological assessment of upland hardwood systems by the GCPO LCC. We used a combination of remote sensing products including 2011 National Land Cover Database (NLCD) forest classes and the 2011 MAV forest classification layer produced by the Lower Mississippi Valley Joint Venture to delineate patches of all forest types in the GCPO LCC. We used NLCD as the primary data source when assessing forests outside the GCPO LCC MAV subgeography, and the LMVJV forest classification as the primary data source for forest assessment within the MAV. NLCD was developed using 2011 Landsat TM imagery, with forest classes including only areas with trees exceeding 5 m (16 ft) in height and where trees compose at least 20% of the total vegetation cover. We first clipped the 2011 NLCD to a 10 km buffer around the GCPO LCC geographic boundary, then resampled the data from 30 m resolution to 250 m resolution using a nearest neighbor algorithm. We resampled to 250 m to allow the forest classification to be assessed with other forested wetland condition data developed at a 250 m resolution from MODIS satellite imagery. Once data were at 250 m resolution we then reclassified the data to extract NLCD Deciduous Forest (41), Evergreen Forest (42), Mixed Forest (43), and Woody Wetlands (90)) classes as a single forest value. We next assessed the LMVJV forest classification data for the MAV, using 2011 Landsat-based classification supplemented with known reforestation patches and aggregated across 90 m breaks. To produce this product Mitchell et al. used 11 cloud free Landsat 5 TM scenes from Oct-Nov 2011 in combination with ancillary data, then used object-based image analysis to segment out classify forests and other land cover features. This analysis was supplemented with spatial data on regenerating forest planted under the U.S. Department of Agriculture Wetland Reserve Program (now part of the Agricultural Conservation Enhancement Program), Conservation Reserve Program, and other conservation easement lands, which is often misclassified in national mapping products. We converted vector polygon data to a 30 m resolution then resampled up to 250 m resolution raster using a nearest neighbor algorithm. We clipped this layer to the GCPO LCC MAV subgeography boundary. We then mosaicked the LMVJV forest classification to 2011 NLCD forest classes using LMVJV forest as the primary operator, resulting in a 250 m resolution forest “mask” that combined the two datasets within the GCPO. To address the configuration endpoint for the upland hardwoods system targeting woodlands and forest in heavily (>70%) forested 10 km landscapes we ran a focal statistics procedure in ArcGIS assessing the mean percent cover of the composite forest layer described above with a 40 cell radius neighborhood for analysis (250 m pixel x 40 cells = 10 km) throughout the GCPO. We then reclassified the mean forest cover output to >70% and extracted back through the hardwood woodland and forest masks to calculate acres on the pixels of each hardwood system that have >70% forest in surrounding10km landscape.
Definition Expression: N/A
Copyright Text: Gulf Coastal Plains and Ozarks Landscape Conservation Cooperative with supporting data from the 2011 National Land Cover Dataset and the Lower Mississippi Valley Joint Venture 2011 Forest Characterization Dataset.
Name: Upland hardwood woodland in >70% forest landscape
Display Field:
Type: Raster Layer
Geometry Type: null
Description: This data represents an assessment of forest composition used in the ecological assessment of upland hardwood systems by the GCPO LCC. We used a combination of remote sensing products including 2011 National Land Cover Database (NLCD) forest classes and the 2011 MAV forest classification layer produced by the Lower Mississippi Valley Joint Venture to delineate patches of all forest types in the GCPO LCC. We used NLCD as the primary data source when assessing forests outside the GCPO LCC MAV subgeography, and the LMVJV forest classification as the primary data source for forest assessment within the MAV. NLCD was developed using 2011 Landsat TM imagery, with forest classes including only areas with trees exceeding 5 m (16 ft) in height and where trees compose at least 20% of the total vegetation cover. We first clipped the 2011 NLCD to a 10 km buffer around the GCPO LCC geographic boundary, then resampled the data from 30 m resolution to 250 m resolution using a nearest neighbor algorithm. We resampled to 250 m to allow the forest classification to be assessed with other forested wetland condition data developed at a 250 m resolution from MODIS satellite imagery. Once data were at 250 m resolution we then reclassified the data to extract NLCD Deciduous Forest (41), Evergreen Forest (42), Mixed Forest (43), and Woody Wetlands (90)) classes as a single forest value. We next assessed the LMVJV forest classification data for the MAV, using 2011 Landsat-based classification supplemented with known reforestation patches and aggregated across 90 m breaks. To produce this product Mitchell et al. used 11 cloud free Landsat 5 TM scenes from Oct-Nov 2011 in combination with ancillary data, then used object-based image analysis to segment out classify forests and other land cover features. This analysis was supplemented with spatial data on regenerating forest planted under the U.S. Department of Agriculture Wetland Reserve Program (now part of the Agricultural Conservation Enhancement Program), Conservation Reserve Program, and other conservation easement lands, which is often misclassified in national mapping products. We converted vector polygon data to a 30 m resolution then resampled up to 250 m resolution raster using a nearest neighbor algorithm. We clipped this layer to the GCPO LCC MAV subgeography boundary. We then mosaicked the LMVJV forest classification to 2011 NLCD forest classes using LMVJV forest as the primary operator, resulting in a 250 m resolution forest “mask” that combined the two datasets within the GCPO. To address the configuration endpoint for the upland hardwoods system targeting woodlands and forest in heavily (>70%) forested 10 km landscapes we ran a focal statistics procedure in ArcGIS assessing the mean percent cover of the composite forest layer described above with a 40 cell radius neighborhood for analysis (250 m pixel x 40 cells = 10 km) throughout the GCPO. We then reclassified the mean forest cover output to >70% and extracted back through the hardwood woodland and forest masks to calculate acres on the pixels of each hardwood system that have >70% forest in surrounding10km landscape.
Definition Expression: N/A
Copyright Text: Gulf Coastal Plains and Ozarks Landscape Conservation Cooperative with supporting data from the 2011 National Land Cover Dataset and the Lower Mississippi Valley Joint Venture 2011 Forest Characterization Dataset.
Name: Upland hardwood forest in >70% forest landscape
Display Field:
Type: Raster Layer
Geometry Type: null
Description: This data represents an assessment of forest composition used in the ecological assessment of upland hardwood systems by the GCPO LCC. We used a combination of remote sensing products including 2011 National Land Cover Database (NLCD) forest classes and the 2011 MAV forest classification layer produced by the Lower Mississippi Valley Joint Venture to delineate patches of all forest types in the GCPO LCC. We used NLCD as the primary data source when assessing forests outside the GCPO LCC MAV subgeography, and the LMVJV forest classification as the primary data source for forest assessment within the MAV. NLCD was developed using 2011 Landsat TM imagery, with forest classes including only areas with trees exceeding 5 m (16 ft) in height and where trees compose at least 20% of the total vegetation cover. We first clipped the 2011 NLCD to a 10 km buffer around the GCPO LCC geographic boundary, then resampled the data from 30 m resolution to 250 m resolution using a nearest neighbor algorithm. We resampled to 250 m to allow the forest classification to be assessed with other forested wetland condition data developed at a 250 m resolution from MODIS satellite imagery. Once data were at 250 m resolution we then reclassified the data to extract NLCD Deciduous Forest (41), Evergreen Forest (42), Mixed Forest (43), and Woody Wetlands (90)) classes as a single forest value. We next assessed the LMVJV forest classification data for the MAV, using 2011 Landsat-based classification supplemented with known reforestation patches and aggregated across 90 m breaks. To produce this product Mitchell et al. used 11 cloud free Landsat 5 TM scenes from Oct-Nov 2011 in combination with ancillary data, then used object-based image analysis to segment out classify forests and other land cover features. This analysis was supplemented with spatial data on regenerating forest planted under the U.S. Department of Agriculture Wetland Reserve Program (now part of the Agricultural Conservation Enhancement Program), Conservation Reserve Program, and other conservation easement lands, which is often misclassified in national mapping products. We converted vector polygon data to a 30 m resolution then resampled up to 250 m resolution raster using a nearest neighbor algorithm. We clipped this layer to the GCPO LCC MAV subgeography boundary. We then mosaicked the LMVJV forest classification to 2011 NLCD forest classes using LMVJV forest as the primary operator, resulting in a 250 m resolution forest “mask” that combined the two datasets within the GCPO. To address the configuration endpoint for the upland hardwoods system targeting woodlands and forest in heavily (>70%) forested 10 km landscapes we ran a focal statistics procedure in ArcGIS assessing the mean percent cover of the composite forest layer described above with a 40 cell radius neighborhood for analysis (250 m pixel x 40 cells = 10 km) throughout the GCPO. We then reclassified the mean forest cover output to >70% and extracted back through the hardwood woodland and forest masks to calculate acres on the pixels of each hardwood system that have >70% forest in surrounding10km landscape.
Definition Expression: N/A
Copyright Text: Gulf Coastal Plains and Ozarks Landscape Conservation Cooperative with supporting data from the 2011 National Land Cover Dataset and the Lower Mississippi Valley Joint Venture 2011 Forest Characterization Dataset.