National Park Service

South Florida/Caribbean I&M Network (SFCN)

Vegetation Communities Extent & Distribution

View of Everglades National Park from a helicopter
View of Everglades National Park from a helicopter.

Vegetation Mapping Reports

For more information contact:
Kevin R. T. Whelan, Ph D.

Importance/Issues

The spatial patterns of vegetation in wet prairies and marshes, forests, tree islands, mangroves, beaches and tidal wetlands are expected to change due to management regimes (regional hydrology changes by Everglades restoration efforts; fire), natural succession processes, sea level rise, and invasive species. A baseline and sound monitoring program should be established to track impacts of these changes at a regional scale. The mosaic and diversity of vegetation communities across the landscape strongly influences animal communities, food web-structure and distribution of rare plants. Vegetation patterns are also useful in planning for management, monitoring and visitors.

Approach

SFCN is working with cooperators and with funding from the National Park Service Vegetation Inventory Program, the U.S. Army Corps of Engineers, and other sources to map the extent and distribution of vegetation communities in SFCN parks. The Vegetation Inventory Program is an effort by the National Park Service to classify, describe, and map detailed vegetation communities in more than 270 national park units across the United States and produce high-quality, standardized maps and associated data sets of vegetation occurring within parks. Approach and funding source varies with each park and, while guidance from the Vegetation Inventory Program is followed, no single multi-park protocol is in development. Maps will only be updated as additional funding becomes available.

Park-specific Vegetation Communities Extent & Distribution Mapping Projects

For more information contact Kevin Whelan, Ph. D.

Project Summary

An aerial view the mangroves and tidal channels of the Rubicon Keys in Biscayne National Park.
An aerial view the mangroves and tidal channels of the Rubicon Keys in Biscayne National Park.

The draft Biscayne National Park (BISC) vegetation map was created by Pablo L. Ruiz, Patricia A. Houle, and Michael S. Ross of Florida International University (Cooperative agreement H500 06 5040 Task agreement J2117062272) with the National Park Service South Florida / Caribbean Inventory and Monitoring Network conducting the accuracy assessment and assembling the final joint report and deliverables.

Biscayne National Park's 3,096 hectares of terrestrial vegetation, including the wetlands along the western shore of Biscayne Bay, mangrove islands in the bay, and barrier islands that parallel the mainland, were mapped with a vector-based approach using photo-interpretation of Florida Fish and Wildlife Conservation Commission Fish and Wildlife Research Institute (FWRI) 2005 aerial imagery (30cm, 5-band) as well as 2002 Light Detection and Ranging (LiDAR) data (mainland only). An additional 429 hectares were mapped in areas immediately adjacent to the park boundary. The map was made in UTM, NAD 83, Zone 17N coordinates with a minimum mapping unit of 400 m² with notable smaller objects mapped at the photo-interpreter's discretion. The vegetation classification system used is a 6-level hierarchical vegetation classification system developed by Rutchey et al. (2006, ver. 5/22/2007) with 29 new map classes added in this project. Digitized polygons were classified to the highest feasible level of resolution. The map consists of a total of 3,524 hectares delineated into 4,672 polygons representing 100 map classes and using 1,081 training field points. A high resolution shoreline layer was also created as part of this project.

An accuracy assessment of the draft map was conducted by the National Park Service South Florida/Caribbean Inventory and Monitoring Network. The assessment included the imagery positional accuracy, the map positional accuracy, and vegetation classification accuracy. The horizontal accuracy of the aerial imagery was checked using 12 ground control locations and measured ±2.7m at the 95% confidence level. The positional accuracy of well-defined map features was checked with 25 positional accuracy assessment locations and measured ±2.5m at the 95% confidence level. The vegetation classification accuracy was checked using a total of 390 locations to determine if the vegetation community observed in the field matched the annotated map classification at the most detailed level. The draft Biscayne National Park vegetation map's overall accuracy is 83.3% (325 of 390 accuracy assessment locations were acceptable) with a lower 90% confidence level of 80.7% accuracy (Kappa index 82.6%) when assessed at the most detailed level of the classification. The draft Biscayne National Park vegetation map was updated to include corrections to polygons based on findings from the 390 accuracy assessment locations. The final map product for this project is this updated map and is referred to as the Biscayne National Park vegetation map.

View Project Products

For more information contact Kevin Whelan, Ph. D.

Project Summary

2007 U.S. Army Corps of Engineers 30cm resolution orthoimage of Buck Island within Buck Island Reef National Monument.
2007 U.S. Army Corps of Engineers 30cm resolution orthoimage of Buck Island within Buck Island Reef National Monument.

The National Park Service Vegetation Mapping Inventory Program (NPS-VMP) funded the NPS South Florida/Caribbean Network (SFCN) to map Buck Island of Buck Island Reef National Monument in 2007. The project was completed in 2009. SFCN established a cooperative agreement with Florida International University for the Resource Management Intern Program with one of the deliverables being development of a vegetation map of Buck Island Reef National Monument.

The Buck Island vegetation map was made in UTM, NAD 83, zone 20N coordinates with a minimum mapping unit of 400m2. Photointerpretation was based on aerial imagery of a single U.S. Army Corps of Engineers natural color GeoTIFF orthophoto acquired as part of a larger collection in 2006–2007. SFCN field tests indicate that the horizontal accuracy (RMSE) of the aerial image to be approximately 2.2m.

The final vegetation map has a total of 26 mapping classes and 51 polygons. A vegetation key and classification were generated as part of the project which included four proposed new Alliances, and 11 new Associations. The vegetation classification system was created based on vegetation physiognomic class, Caribbean Formations from Areces-Mallea et al. (1999), U.S. Virgin Islands Sub-formations from Gibney et al. (2000), and Alliances/Associations defined by the SFCN based on data from vegetation plots and field notes.

Also included are: a photo-interpretation key, a list of vegetation species observed in the field during this project, a vegetation species photo guide for Buck Island Reef National Monument, field data sheets, and reference literature. Ninety-two percent of the polygons either had vegetation data collection or were directly visited and confirmed; consequently SFCN concluded the map meets the NPS-VMP requirement of 80% accuracy with 90% confidence. The final product meets the FGDC metadata standard.

View Project Products

For more information contact Kevin Whelan, Ph. D.

Project Summary

Mangroves are the most common vegetation community at De Soto National Memorial. Photo Credit NPS-SFCN Robert Brooke Shamblin.
Mangroves are the most common vegetation community at De Soto National Memorial.

De Soto National Memorial (DESO) was created in 1948 to commemorate the 1539 landing of Hernando de Soto. De Soto National Memorial and the adjacent Riverview Pointe Preserve are part of the Shaw's Point Archaeological District which comprises a large prehistoric coastal village site that was inhabited by Florida Indians from about 356 B.C. to A.D. 110. The native vegetation consists primarily of mangroves, beaches, hammock, and pines. The majority of the park is comprised of wetlands (mangroves and associated shoreline mostly) that are intimately tied to the Manatee River and Tampa Bay. De Soto National Memorial provided $5,900 funded by the Natural Resources Preservation Program (NRPP), Small Parks (Project Management Information System [PMIS] project number 105225), to the NPS South Florida / Caribbean Network (SFCN) to create a vegetation map of De Soto National Memorial (26 acres) and the adjacent Riverview Pointe Preserve (11 acres) in 2007. This report documents the steps involved in creating the De Soto National Memorial vegetation map.

The De Soto National Memorial vegetation map was made in UTM, NAD 83, zone 17N coordinates with a minimum mapping unit of 400 m². Aerial imagery from January 2007 and LIDAR imagery from 2003 were used for initial polygon development. These polygons were further refined with field data from December 2007 and June 2009. The final vegetation map has a total of 21 mapping classes and 56 polygons. Map class descriptions are provided in Appendix B with some modifications to the Rutchey et al. (2006, v. 5/22/2007) vegetation key and classification provided in Appendix F. Photos of map classes are provided in Appendix C. A list of species encountered during the field assessment is provided in Appendix D. Maps are provided in Appendix E. Data were collected for 70 field points as well as additional points delineating 0.57 miles of drainages. Nineteen (19) shell mound polygons were created from data collected during this project and a previous effort by Daniel Stephens (DESO NPS). Eighty-four percent (84%) of the polygons had vegetation plot data collection and the remainder were visually confirmed.

View Project Products

For more information contact Kevin Whelan, Ph. D.

Loggerhead Key Lighthouse. Photo credited to Jose Luciani (NPS).
Loggerhead Key Lighthouse.

Project Summary

The National Park Service Vegetation Mapping Inventory Program (NPS-VMP) funded the NPS South Florida/Caribbean Network (SFCN) to map the seven islands of Dry Tortugas National Park in 2009. The vegetation map of Dry Tortugas National Park was created by in field polygon delineation instead of the more common photo-interpretation of aerial imagery. Using the Trimble GeoXT GPS device in the field allowed the production of a finely detailed vegetation map (vegetation census) for Loggerhead Key, Garden Key, Bush Key, East Key, Long Key, Middle Key, and Hospital Key. The Dry Tortugas vegetation maps were made using the Universal Transverse Mercator (UTM) zone 17 North coordinate system and the North American Datum of 1983 (NAD 83) with a minimum mapping unit of 25m². The SFCN developed a vegetation classification in congruence with the vegetation map of Dry Tortugas National Park. The Dry Tortugas vegetation classification conforms to the Rutchey et al. (2006) South Florida Vegetation Classification System and those vegetation communities unique to the Dry Tortugas.

The final vegetation map has a total of 41 mapping classes and 438 polygons with a total land area of 39.4 hectares. At the physiognomic level, Dry Tortugas vegetation map included 3 Woodland classes, 8 Shrubland classes, 5 Scrub classes, 20 Dune classes, 3 Sparse Vegetation classes, and 2 Non-vegetative classes.

Class/
Island
Bush East Garde Hospital Loggerhead Long Middle Total
Woodland 0% 0% 3.3% 0% 0% 11.5% 0% 1.1%
Shrubland 6.6% 0% 0.1% 0% 19.6% 1.5% 0% 11%
Scrub 0.2% 0% 0% 0% 0.2% 0.6% 0% 0.1%
Dune 54.6% 0.4% 12.9% 0% 38.4% 19.2% 0% 31.7%
Sparse 38% 99.6% 7.2% 100% 34.4% 67.2% 100% 34.8%
Non- vegetative 0.6% 0% 76.5% 0% 7.4% 0% 0% 21.3%
Total Area (ha) 6.4 ha 2.3ha 9 ha 0.4 ha 19.8 h 1.1 ha 0.4 ha 39.4 ha

Table Summary. Area percentage of the physiognomic classes and total area (ha) for each island and for all the islands combined.

Sparse vegetation accounts for 34.8% of the entire land area in the park consisting of low to zero cover of vegetation. Non-vegetative areas are places in the park where permanent structures have been introduced such as Fort Jefferson and Loggerhead Lighthouse that have a mixture of native and exotic vegetation on the disturbed grounds surrounding the structures. Dune associative communities are the main physiognomic class pertaining to native communities which have significant cover on four of the seven islands. This is the most diverse physiognomic class with 20 individual associations that are common to rare in the park. Shrubland classes constitute only 11% of the total land area which are more commonly found on Loggerhead Key and less so on Bush Key. Woodland and Scrub classes are even less common in the park. Woodland classes are restricted to Garden Key on the grounds outside Fort Jefferson and the mangroves of Long Key. The vegetation communities as they are represented on these islands differ in distribution and commonality. Some classes are only found on one island, whereas, many can be found on more than one with differences in abundance and composition.

SFCN established a cooperative agreement with Florida International University for the Resource Management Intern Program with one deliverable being the development of a vegetation map of Dry Tortugas National Park. This report is a deliverable under task agreement #J2117072808 and cooperative agreement #H5000060104.

Also included are: detailed maps of close-up sections of Loggerhead Key, Bush Key, and Long Key, historical maps from reference literature, and the field data sheet.

The polygon delineation occurred in the field with full visitation (vegetation census) of the entire area, so the classification and map accuracy is assumed to be extremely high and we feel it easily meets the National Vegetation Inventory Program standard of 80% classification accuracy with 90% confidence. The final product meets the FGDC metadata standard.

View Project Products

For more information contact Kevin Whelan, Ph. D.

Project Summary

2007 U.S. Army Corps of Engineers 30cm resolution orthoimage of Salt River Bay within Salt River Bay National Historical Park & Ecological Preserve.
2007 U.S. Army Corps of Engineers 30cm resolution orthoimage of Salt River Bay within Salt River Bay National Historical Park & Ecological Preserve.

The National Park Service Vegetation Mapping Inventory Program (NPS-VMP) funded the NPS South Florida / Caribbean Network (SFCN) to map the terrestrial portions of Salt River Bay National Historical Park & Ecological Preserve in 2007. The mapping project was completed in 2009. SFCN established a cooperative agreement with Florida International University for the Resource Management Intern Program with one of the deliverables being development of a vegetation map of Salt River Bay National Historical Park & Ecological Preserve.

The Salt River vegetation map was made in UTM, NAD 83, zone 20N coordinates with a minimum mapping unit of 400 m2. Photo-interpretation was based on aerial imagery of a single U.S. Army Corps of Engineers natural color GeoTIFF orthophoto acquired as part of a larger collection in 2006–2007. SFCN field tests indicate that the horizontal accuracy (RMSE) of the aerial image to be 3.5 m.

The final vegetation map has a total of 58 mapping classes and 247 polygons. A vegetation classification was generated as part of the project which included 32 proposed new Alliances, and two new Associations. A total of 482 terrestrial acres (195 hectares) were mapped in this project including some areas outside of the park boundaries as boundaries were not well established at the time of this project. Four hundred and nine acres (166 hectares) were mapped within the park and include 100% of the terrestrial acreage within the park boundaries. The vegetation classification system was created based on vegetation physiognomic class, Caribbean Formations from Areces-Mallea et al. (1999), U.S. Virgin Islands Sub-formations from Gibney et al. (2000), and Alliances/Associations defined by the SFCN based on data from vegetation plots and field notes. All proposed new classifications are intended for use within Salt River Bay National Historical Park & Ecological Preserve and not meant to represent the greater geographic region of Saint Croix or the Virgin Islands.

Also included are photos of map classes, a list of vegetation species observed in the field during this project, and field data sheets. Eighty percent of the polygons either had vegetation data collection or were directly visited and confirmed; consequently SFCN concluded the map meets the NPS-VMP requirement of 80% accuracy with 90% confidence. The final product meets the FGDC (Federal Geographic Data Committee) metadata standard.

View Project Products

Last Updated: March 28, 2017 Contact Webmaster