Using Advanced Mapping to Measure Changes in Mangrove and Seagrass Habitat over Time - NERRS/NSC(NERRS Science Collaborative)Office for Coastal Management (OCM)
Data Set (DS) | ID: 54589 | Published / External
|Title:||Using Advanced Mapping to Measure Changes in Mangrove and Seagrass Habitat over Time - NERRS/NSC(NERRS Science Collaborative)|
This project evaluated ecosystem damage and recovery by developing a time series of habitat maps for the Rookery Bay National Estuarine Research Reserve. Habitat maps were created based on WorldView-2 and Landsat-8 satellite imagery from 2010-2018 using an automated technique and validated with a field campaign. Landsat images were mapped using the Support Vector Machine machine learning method in ENVI, and WorldView images were mapped using a preliminary version of the SOALCHI decision tree algorithm. Habitats mapped include healthy mangrove, degraded mangrove, marsh, upland vegetation, soil, and water. Habitat change maps document the damage caused by Hurricane Irma in September of 2017 as it made landfall in the reserve as a Category 4 storm. Project outputs include field-survey results with GPS points, a baseline habitat map, annual habitat seasons for 2016-2018, and habitat change assessments. Outcomes include the development of new research collaborations, quantitative characterization of reserve habitat change, improved understanding of critical habitat change dynamics, and assessment of chronic and extreme-event disturbance and recovery.
In recent years, seagrass and mangrove deaths have accelerated in the Rookery Bay National Estuarine Research Reserve and other parts of southern Florida. These losses are a result of chronic stresses from watershed alterations, sea level rise and invasive species, along with acute impacts from hurricanes. Reserve managers seeking to build the resiliency of these habitats need to map the extent of storm-related damage and distinguish it from ongoing chronic degradation. The goal of this project was to identify the location, extent, and severity of damage to terrestrial habitats throughout the reserve. In addition, the team was able to test and refine a new high-resolution, automated mapping technique based on satellite imagery.
Contact information: Matt McCarthy, University of South Florida, Email: firstname.lastname@example.org, Phone: 727-553-1186
Maps of reserve habitat types from WorldView and Landsat satellite sensors were created for multiple time periods from 2010 to 2018. Maps of habitat change were also developed. Habitat maps and metadata are available for download from DataOne: https://search.dataone.org/view/10.24431/rw1k43j
Field surveys were conducted to document habitat types and vegetation characteristics at 2,700 discrete points throughout the study area in fall of 2018. These point observations and GPS coordinates were used to develop habitat classifications and then a separate set of points were used to perform an accuracy check of the maps. Field survey data have been archived by the Rookery Bay Reserve and can be requested by contacting: Brita Jessen, Research Coordinator, Rookery Bay National Estuarine Research Reserve, Naples, Florida, Email: Brita.Jessen@dep.state.fl.us, Phone: 239-530-5964
More information and links to other products generated by project are available through the Science Collaborative: www.nerrssciencecollaborative.org/project/MullerKarger18
|Global Change Master Directory (GCMD) Science Keywords||Earth Science > Biosphere > Ecosystems > Marine Ecosystems > Estuary|
|Global Change Master Directory (GCMD) Location Keywords||Continent > North America > United States Of America > Florida|
|None||Rookery Bay NERR, FL|
Data Set Information
|Data Set Scope Code:||Data Set|
|Maintenance Frequency:||None Planned|
The distributor does not assume liability.
|Date Effective From:||2017-10-31|
|Date Effective To:|
|Contact (Person):||Cothran, Jeremy|
|Date Effective From:||2018-09|
|Date Effective To:||2019-08|
|Contact (Person):||Muller-Karger, Frank|
|Currentness Reference:||Ground Condition|
Extent Group 1
Extent Group 1 / Geographic Area 1
Rookery Bay, FL NERR. The focal study area was Rookery Bay Reserve, in southwest Florida, just south of Naples. An adjacent property, the Ten Thousand Island National Wildlife Refuge, was partially included in maps.
Extent Group 1 / Time Frame 1
|Time Frame Type:||Range|
|Data Access Constraints:||
|Data Use Constraints:||
Cite this dataset when used as a source: NOAA retains the right to analyze, synthesize and publish summaries of the NERRS/NSC data. The NERRS/NSC retains the right to be fully credited for having collected and process the data. Following academic courtesy standards, the NERR site where the data were collected should be contacted and fully acknowledged in any subsequent publications in which any part of the data are used. The data enclosed within this package/transmission are only as accurate as the quality assurance and quality control procedures that are described in the associated metadata reporting statement allow. The user bears all responsibility for its subsequent use/misuse in any further analyses or comparisons. The Federal government does not assume liability to the Recipient or third persons, nor will the Federal government reimburse or indemnify the Recipient for its liability due to any losses resulting in any way from the use of this data. Requested citation format: NOAA National Estuarine Research Reserve System (NERRS) Science Collaborative(NSC).
This site provides a project overview and links to all associated products, including data.
Habitat maps are available for download from DataOne
|Quality Control Procedures Employed:||
1)Habitat maps of mangrove, marsh, and vegetative change
Habitat maps were evaluated for accuracy using the GPS points and found to be 82% accurate overall, which is above the 80% benchmark for suitability. Degraded mangrove was the least-accurate class (58% accuracy). Preliminary and final maps were sent out or displayed via webinar to end users with local knowledge so that feedback could be received. Where end users identified errors in the maps, the images were remapped iteratively until end users were confident in their identification of habitats.
2)Habitat type field surveys at Rookery Bay
The raw points were quality-controlled by eliminating all points that were within 20 meters of each other to avoid spatial autocorrelation, and were visually evaluated with independent imagery to ensure that the point overlaid correctly with the documented habitat.
|Catalog Item ID:||54589|
|Metadata Record Created By:||Jeremy Cothran|
|Metadata Record Created:||2018-11-01 11:34+0000|
|Metadata Record Last Modified By:||Tyler Christensen|
|Metadata Record Last Modified:||2020-04-10 10:30+0000|
|Metadata Record Published:||2020-04-10|
|Metadata Publication Status:||Published Externally|
|Do Not Publish?:||N|
|Metadata Next Review Date:||2021-04-11|