The files linked to this reference are the geospatial data created as part of the completion of the baseline vegetation inventory project for the NPS park unit. Current format is ArcGIS file geodatabase but older formats may exist as shapefiles. In 2009, Kass Green & Associates (KGA) was chosen by the NPS to map the vegetation of Grand Canyon National Park and the Lake Mead National Recreation Area – administered portions of the Grand Canyon Parashant National Monument using a National Vegetation Classification Standard (NVCS) compliant classification. Mapping tools and techniques used included remotely sensed digital airborne NAIP imagery, image segmentation, ancillary data, GIS biophysical modeling, photo interpretation, and field visits. This section of the report summarizes the methods used by KGA to map the vegetation of the project area. The vegetation was mapped in three phases based on floristic similarity and difficulties dealing with the logistical barriers presented by extreme terrain. Phase 1 included the high elevation forests and woodlands on rims of the eastern portion of the mapping area; Phase 2 was the inner canyon areas of the eastern and central mapping area; Phase 3 was most of the rim- and canyon areas west of Parashant Canyon. Each phase was mapped and accuracy assessed as a unit. The final mapping activities involved reconciling map classes, boundaries and accuracy assessment among all phases. The final map contained 87 map classes with a minimum mapping unit of 0.5 hectares across more than 560,000 hectares. The classes included 41 NVC Associations, 36 NVC Alliances, seven NVC Group-level classes and three classes of unvegetated surfaces (built-up, water, and bare soil/rock). Accuracy assessment (AA) was done for 1847 map segments, distributed based on map class abundances. Accuracy by map class varied between 13% (Pinyon – Juniper / Talus or Canyon Slope Scrub) and 100% (Douglas Fir / Snowberry Forest and 5 others); project-wide accuracy was 77%. Roughly one-third of the mis-identified samples were among closely– related vegetation types. Others were among classes which were found in similar habitats (e.g., constrained tributary beds) and had very similar spectral signatures.

The files linked to this reference are the geospatial data created as part of the completion of the baseline vegetation inventory project for the NPS park unit. Current format is ArcGIS file geodatabase but older formats may exist as shapefiles. Vegetation and land use were interpreted to as detailed a level possible from high-resolution, 9” x 9” stereo pairs of 1:12,000-scale true color aerial photography. Polygons representing vegetation or land use map classes were delineated directly on-screen through heads-up digitizing using ArcGIS editing tools and transferred to a spatial database. The project used the program standard minimum mapping unit of 0.5 ha with few exceptions. Fifty-four map classes represented by 31,497 polygons were developed for CANY. A total of 30,329 polygons represent 41 natural or semi-natural vegetation map classes covering 94.8% of the mapping project area. One map class was documented as point locations only. Twelve additional land use/land cover and geologic map classes describe 1,168 polygons (3.7% of polygons and 5.2% of the area). Average polygon size across all map classes is 9.9 ha (24.4 acres). Lands within CANY make up 135,204 ha (334,096 acres) or 79.2% of the total project area. Blackbrush Shrubland had the highest number of polygons (3600) polygons covering 9.4% of the mapping area. The Potholed and Jointed Sandstone Woodland Complex is the most common map class, delineated on 30,495 ha (75,354 acres) or 17.9% of the project area.

The files linked to this reference are the geospatial data created as part of the completion of the baseline vegetation inventory project for the NPS park unit. Current format is ArcGIS file geodatabase but older formats may exist as shapefiles. Vegetation and land use were interpreted to as detailed a level possible from high-resolution, 9” x 9” stereo pairs of 1:12,000-scale true color aerial photography. Polygons representing vegetation or land use map classes were delineated directly on-screen through heads-up digitizing using ArcGIS editing tools and transferred to a spatial database. The project used the program standard minimum mapping unit of 0.5 ha with few exceptions. Fifty-four map classes represented by 31,497 polygons were developed for CANY. A total of 30,329 polygons represent 41 natural or semi-natural vegetation map classes covering 94.8% of the mapping project area. One map class was documented as point locations only. Twelve additional land use/land cover and geologic map classes describe 1,168 polygons (3.7% of polygons and 5.2% of the area). Average polygon size across all map classes is 9.9 ha (24.4 acres). Lands within CANY make up 135,204 ha (334,096 acres) or 79.2% of the total project area. Blackbrush Shrubland had the highest number of polygons (3600) polygons covering 9.4% of the mapping area. The Potholed and Jointed Sandstone Woodland Complex is the most common map class, delineated on 30,495 ha (75,354 acres) or 17.9% of the project area.

The files linked to this reference are the geospatial data created as part of the completion of the baseline vegetation inventory project for the NPS park unit. Current format is ArcGIS file geodatabase but older formats may exist as shapefiles. The mapping component of the BRCA project used a combination of methods to interpret and delineate vegetation and land use polygons. The USGS applied an electronic segmentation method (e-Cognition software) to create preliminary linework on features with high-contrast photo-signatures. Using the preliminary linework as a baseline starting point, the primary photointerpreter drew polygons directly on screen through heads-up digitizing using ArcGIS editing tools. Additionally, trained photointerpreters assisting the primary photointerpreter drew polygons on Mylar overlays covering 1m resolution, 1:12,000-scale, 9 x 9-inch true-color aerial photographs. This process enabled the photointerpreters to view the landscape in stereo in order to identify finer details. The linework drawn on Mylar overlays was then transferred into digital media by heads-up digitizing using ArcGIS software. The park and environs were interpreted and mapped to the same level of detail.

The files linked to this reference are the geospatial data created as part of the completion of the baseline vegetation inventory project for the NPS park unit. Current format is ArcGIS file geodatabase but older formats may exist as shapefiles. The mapping component of the BRCA project used a combination of methods to interpret and delineate vegetation and land use polygons. The USGS applied an electronic segmentation method (e-Cognition software) to create preliminary linework on features with high-contrast photo-signatures. Using the preliminary linework as a baseline starting point, the primary photointerpreter drew polygons directly on screen through heads-up digitizing using ArcGIS editing tools. Additionally, trained photointerpreters assisting the primary photointerpreter drew polygons on Mylar overlays covering 1m resolution, 1:12,000-scale, 9 x 9-inch true-color aerial photographs. This process enabled the photointerpreters to view the landscape in stereo in order to identify finer details. The linework drawn on Mylar overlays was then transferred into digital media by heads-up digitizing using ArcGIS software. The park and environs were interpreted and mapped to the same level of detail.

This reference contains the imagery data information used in the completion of the baseline vegetation inventory project for Grand Canyon National Park (GRCA) and Parashant National Monument (PARA). This project relied on both the 2007 and 2010 NAIP imagery for image segmentation and classification. Two core imagery datasets used in this project were: National Agriculture Image Program (NAIP) 2010 DMC 4 band, 8 bit DigitalOrtho Quarter Quads (DOQQs) for the entire project area. In areas were there was significant shadowing and/or some mis-registration, the project augmented the aerial information with NAIP 2007 ADS40 4 band, 8 bit DOQQs for the entire project area.

This reference contains the imagery data information used in the completion of the baseline vegetation inventory project for Grand Canyon National Park (GRCA) and Parashant National Monument (PARA). This project relied on both the 2007 and 2010 NAIP imagery for image segmentation and classification. Two core imagery datasets used in this project were: National Agriculture Image Program (NAIP) 2010 DMC 4 band, 8 bit DigitalOrtho Quarter Quads (DOQQs) for the entire project area. In areas were there was significant shadowing and/or some mis-registration, the project augmented the aerial information with NAIP 2007 ADS40 4 band, 8 bit DOQQs for the entire project area.

The files linked to this reference are the geospatial data created as part of the completion of the baseline vegetation inventory project for the NPS park unit. Current format is ArcGIS file geodatabase but older formats may exist as shapefiles. The project team also developed a spatial vegetation map database representing CACH, with three different map-class schemas: base, group, and management map classes. The base map classes represented the finest level of spatial detail. Photointerpreters delineated initial polygons through manual interpretation of 2003/2004 1:12,000-scale true color aerial photography supplemented by occasional computer screen digitizing on a mosaic of digitized aerial photos. These polygons were labeled with base map classes during photointerpretation. Field visits verified interpretation concepts. The vegetation map database includes, 53 base map classes, which consist of associations and park specials classified with the quantitative analysis, additional associations noted during photointerpretation, non-vegetated land cover, such as infrastructure, land use, and geological land cover. The base map classes consist of 4,718 polygons in the project area. A field-based accuracy assessment of the base map classes showed the overall accuracy to be 50.8% The group map classes represent aggregations of the base map classes, approximating the group level of the National Vegetation Classification Standard, Version 2 (Federal Geographic Data Committee 2008). Terrestrial ecological systems, as described by NatureServe (Comer et al. 2003), were used as a first approximation of the group level. The project team identified 16 group map classes in this project. The overall accuracy of the group map classes was determined using the same accuracy assessment data as for the base map classes. The overall accuracy of the group representation of vegetation was 79.9%.

The files linked to this reference are the geospatial data created as part of the completion of the baseline vegetation inventory project for the NPS park unit. Current format is ArcGIS file geodatabase but older formats may exist as shapefiles. The project team also developed a spatial vegetation map database representing CACH, with three different map-class schemas: base, group, and management map classes. The base map classes represented the finest level of spatial detail. Photointerpreters delineated initial polygons through manual interpretation of 2003/2004 1:12,000-scale true color aerial photography supplemented by occasional computer screen digitizing on a mosaic of digitized aerial photos. These polygons were labeled with base map classes during photointerpretation. Field visits verified interpretation concepts. The vegetation map database includes, 53 base map classes, which consist of associations and park specials classified with the quantitative analysis, additional associations noted during photointerpretation, non-vegetated land cover, such as infrastructure, land use, and geological land cover. The base map classes consist of 4,718 polygons in the project area. A field-based accuracy assessment of the base map classes showed the overall accuracy to be 50.8% The group map classes represent aggregations of the base map classes, approximating the group level of the National Vegetation Classification Standard, Version 2 (Federal Geographic Data Committee 2008). Terrestrial ecological systems, as described by NatureServe (Comer et al. 2003), were used as a first approximation of the group level. The project team identified 16 group map classes in this project. The overall accuracy of the group map classes was determined using the same accuracy assessment data as for the base map classes. The overall accuracy of the group representation of vegetation was 79.9%.

The files linked to this reference are the geospatial data created as part of the completion of the baseline vegetation inventory project for the NPS park unit. Current format is ArcGIS file geodatabase but older formats may exist as shapefiles. Fifty-one map classes were developed to describe the COLM vegetation mapping project area. Of these, 26 are NVC-based vegetation map classes, four are geology map classes, seven are vegetated land use map classes and 14 are non-vegetated land-use map classes. Of the 26 vegetation map classes, 16 represent single NVC plant associations; the other 10 map classes contain multiple plant associations. One map class consists of point data representing seep and spring vegetation. It is contained in a separate coverage from the polygon map classes.