Chesapeake Conservancy 20260108 Chesapeake Bay Watershed Forest Fragmentation raster digital data This analysis groups land use classes from the 2021-2022 Chesapeake Bay Land Use/Land Cover (LULC) Database 2024 Edition (Claggett et al. 2025) into fragmentation classes based on the framework used by the Morphological Spatial Pattern Analysis (MSPA) tool (Soille and Vogt 2022). Additional post-processing steps include addition and aggregation of fragmentation classes to make the fragmentation data more accessible to their intended audience. The methods of this analysis were developed in consultation with members of US Forest Service (USFS) and the Chesapeake Bay Program’s Forestry Workgroup (CBP FWG). Note that our use of the term “forest” here is only for the purposes of this particular analysis to characterize fragmentation across the landscape. For definitions and estimates of total forest area, refer to the State of Chesapeake Forests Storymap and the LULC data documentation (McDonald et al. 2025). For additional information about the forest fragmentation analysis, please refer to the forest fragmentation technical documentation. The State of Chesapeake Forests Forest Fragmentation analysis characterizes fragmentation of tree cover across the entire Chesapeake Bay watershed at a 1-meter resolution. This dataset includes 8 classes: Islets: Small, isolated groups of tree cover too small to contain transitional forest (i.e., <30 m across). Connected Canopy: Small groups of tree cover <30m across that are connected to the Edge of a patch containing transitional forest. Edge: External perimeter of patch, 0-15m. Transitional: Tree cover between 15-45m from perimeter of patch. Interior: Tree cover between 45m-100m from perimeter of patch. Core: Tree cover >100m from perimeter of patch. Canopy Gap: Non-treed area within transitional, interior, or core forest Perforation: 15m Edge associated with a canopy gap. 2021 2022 ground condition Complete None planned -80.5410 -74.5810 42.9873 36.6694 ISO 19115 Topic Category environment None Forest None Cheseapeake Bay Watershed No access Constraints. Please see Distribution Information for details The organizations responsible for generating and funding this dataset make no representations of any kind including, but not limited to the warranties of merchantability or fitness for a particular use, nor are any such warranties to be implied with respect to the data. The data is provided on the condition that neither the U.S. Forest Service nor the U.S. Government, or any of the collaborating organizations shall be held liable for any damages resulting from the authorized or unauthorized use of the data. Although every effort has been made to ensure the accuracy of information, errors may be reflected in data supplied. Possible errors or limitations of the data have been documented but the list is not exhaustive. The user must be aware of data conditions and bear responsibility for the appropriate use of the information with respect to possible errors, original map scale, collection methodology, currency of data, and other conditions. Credit should always be given to the data source when this data is transferred, altered, or used for analysis. Use of the term “forest” is only for the purposes of this particular analysis to characterize fragmentation across the landscape. Total region areas from our attributed vector polygons should not be interpreted as total forested area. For definitions and estimates of total forest area, refer to the State of Chesapeake Forests Storymap and the LULC data documentation (McDonald et al. 2025). These data were developed as a reference resource to help visualize and understand forest fragmentation patterns across the Chesapeake Bay watershed. The 2021/2022 land use/land cover data from which this fragmentation analysis was derived may contain errors that affect reported areas. Chesapeake Conservancy and US Forest Service do not guarantee the accuracy of the fragmentation data. For more information about accuracy of the underlying land use data, refer to the land use accuracy section of the LULC data documentation (McDonald et al. 2025). Katherine Brownson physical
1750 Forest Drive
Suite 130
Annapolis MD 21401 USA Katherine.Brownson@usda.gov Chesapeake Conservancy, U.S. Forest Service, Chesapeake Bay Program Forestry Workgroup The accuracy for this analysis is based on the underlying LULC data.Various steps have been taken to ensure accuracy, including consideration of the underlying rationale that inform mapping classifications and extensive visual inspections of the data by multiple team members and members of the Forestry Workgroup to address issues such as large area or systematic misclassifications. Additionally, automated workflows were developed to track and capture certain erroneous classifications in the workflows. These are brief descriptions of the processing steps: (1) Primary forest fragmentation raster (2) Attributed region polygon layer (3) Secondary, ancillary fragmentation raster 2025 Primary forest fragmentation raster: The original input of this analysis is 2021-2022 land use/land cover (LULC) data from the Chesapeake Bay Land Use/Land Cover (LULC) Database 2024 Edition (Claggett et al. 2025). First, LULC data were reclassified into what Morphological Spatial Pattern Analysis (MSPA, Soille and Vogt 2022) calls “foreground” (the region of interest that is being assessed) and “background” (inverse of foreground) pixels. Listed below are the LULC classes and corresponding pixel values, chosen in collaboration with USFS and FWG, that were used as the foreground for this analysis. All other LULC classes were considered background. Tree Canopy over Turf Grass: 26 Forest (all classes): 40, 54, 64, 74 Forested Other (all classes): 41, 53, 63, 73 Harvested Forest (all classes): 45, 46, 55, 65, 75 The data were then tiled, with gaps smaller than 9 square meters filled in, and then run through MSPA using the following parameters: Foreground connectivity: 4 Edge width: 15 pixels (with 1 m pixel size) Transition: 0 Intext: 1 Disk: 0 Statistics: 0 Post-processing including the addition of 45m and 100m edge widths, aggregation of some MSPA classes, some corrections to account for edge effects, and then mosaicking of the tiles. Additional detail can be found in the technical documentation. 2025 Attributed region polygon layer: For this part of the analysis, a “region” is defined as a contiguous group of tree cover containing edge, transitional/interior/core forest, any tree cover attached to it, and any gaps within. This definition includes connected canopy and canopy gaps and excludes islets from the attributed vector layer. To generate region boundaries, each masked exterior tile was vectorized after removal of islets. Polygon boundaries were then used to tabulate the area of each fragmentation class per region, and polygon tiles were merged together into a single vector layer with some corrections to account for edge effects. The information contained in the attributed region polygons can be used to calculate other fragmentation metrics per region as desired. Total region area and an intactness ratio (core:region area) have already been pre-calculated. Note that total region area should not be interpreted as total forested area. For definitions and estimates of total forest area, refer to the State of Chesapeake Forests Storymap and the LULC data documentation (McDonald et al. 2025). Additional detail can be found in the technical documentation. 2025 Secondary, ancillary fragmentation raster: The secondary ancillary fragmentation raster was created in much the same way as the primary fragmentation raster, except with all canopy gaps filled in instead of only those smaller than 9 square meters. This layer is intended to facilitate users’ ability to fill in additional canopy gaps in the primary output according to their specific needs and preferences. Some users might be interested in doing so based upon gap size and/or fragmenting LULC class. Additional detail is available in the technical documentation. 2025 Vector G-polygon 1 Albers Conical Equal Area coordinate pair meters North American Datum of 1983 (NAD 83) Geodetic Reference System 1980 6378137.000000 298.257222 Producer Defined OID Internal object identifier. Producer Defined Sequential unique whole numbers that are automatically generated. Value Unique numeric values contained in each raster cell. Producer Defined 3 Islet Small, isolated groups of tree cover too small to contain transitional forest (i.e., <30 m across) Producer defined 4 Connected Canopy Small groups of tree cover <30m across that are connected to the Edge of a patch containing transitional forest Producer defined 5 Edge External perimeter of patch, 0-15m Producer defined 6 Transitional Tree cover between 15-45m from perimeter of patch Producer defined 7 Interior Tree cover between 45m-100m from perimeter of patch Producer defined 8 Core Tree cover >100m from perimeter of patch Producer defined 9 Canopy Gap Non-treed area within transitional, interior, or core forest Producer defined 10 Perforation 15m Edge associated with a canopy gap Producer defined 13 Islet inside canopy gap Small, isolated groups of tree cover too small to contain transitional forest (i.e., <30 m across) that are located inside a canopy gap Producer defined 14 Connected canopy inside canopy gap Small groups of tree cover <30m across that are connected to the Edge of a patch containing transitional forest and located inside a canopy gap Producer defined 16 Transitional inside canopy gap Tree cover between 15-45m from the perimeter of a patch that is located inside a canopy gap Producer defined 17 Interior inside gap Tree cover between 45m-100m from the perimeter of a patch that is located inside a canopy gap Producer defined 18 Core inside canopy gap Tree cover over 100m from the perimeter of a patch that is located inside a canopy gap Producer defined Count Number of raster cells with this value. Producer Defined 908044.0 70478168468.0 ClassName Fragmentation class definitions are based upon definitions provided in Soille & Vogt’s (2022) description of MSPA’s outputs, modified according to the methods of this analysis. For the purposes of defining these classes, the term “patch” refers to the combination of transitional/interior/core forest pixels and its surrounding edge. Producer Defined Islet Small, isolated groups of tree cover too small to contain transitional forest (i.e., <30 m across) Producer defined Connected Canopy Small groups of tree cover <30m across that are connected to the Edge of a patch containing transitional forest Producer defined Edge External perimeter of patch, 0-15m Producer defined Transitional Tree cover between 15-45m from perimeter of patch Producer defined Interior Tree cover between 45m-100m from perimeter of patch Producer defined Core Tree cover >100m from perimeter of patch Producer defined Canopy gap Non-treed area within transitional, interior, or core forest Producer defined Perforation 15m Edge associated with a canopy gap Producer defined Islet inside gap Small, isolated groups of tree cover too small to contain transitional forest (i.e., <30 m across) that are located inside a canopy gap Producer defined Connected Canopy inside gap Small groups of tree cover <15m across that are connected to the Edge of a patch containing transitional forest and located inside a canopy gap Producer defined Transitional inside gap Tree cover between 15-45m from the perimeter of a patch that is located inside a canopy gap Producer defined Interior inside gap Tree cover between 45m-100m from the perimeter of a patch that is located inside a canopy gap Producer defined Core inside gap Tree cover over 100m from the perimeter of a patch that is located inside a canopy gap Producer defined Red Red value for symbology renderer Producer Defined 0 255 Green Green value for symbology renderer Producer Defined 0 200 Blue Blue value for symbology renderer Producer Defined 0 255 Alpha Alpha value for symbology renderer Producer Defined 255 255 Katherine Brownson physical
1750 Forest Drive
Annapolis MD 21401 USA Katherine.Brownson@usda.gov Digital Data 20260108 Katherine Brownson physical
Suite 130
Annapolis MD 21401 Katherine.Brownson@usda.gov FGDC Content Standard for Digital Geospatial Metadata FGDC-STD-001-1998 20260828 14202400 TRUE