Theses and Dissertations

Defense Date

4-19-2016

Document Type

Thesis

Degree Name

M.S. Marine Biology

Second Degree Name

M.S. Coastal Zone Management

First Advisor

Brian Walker

Second Advisor

Steven Miller

Third Advisor

Richard Spieler

Abstract

Benthic habitat maps provide the spatial framework for many research science and management activities in coastal areas such as coral-reefs. Accuracy, the degree to which information on a map matches true or accepted values, of benthic habitat maps is important because often times the map will be used in decision-making processes about how we manage our marine resources. It is critical that some measure, such as the accuracy, of the map be known in order to give a sense of how the overall map portrays the seascape. This study compared the accuracy in the following map classes; major structure, major and detailed biological cover, and detailed coral cover, of the 2014 NOAA Florida Keys Coral Reef Ecosystem Habitat map using two separate quantitative, in situ, and qualitative, drop camera, data sets in order to assess how the data sets compare to one another. Benthic habitat map classes of the NOAA Florida Keys map were based on a NOAA peer-reviewed hierarchical coral reef habitat classification scheme. Accuracy assessment tests to see how often the NOAA Florida Keys map producer correctly classified the different habitats, included error matrix analyses (overall, user’s and producer’s accuracy), and the tau coefficient. Study areas in the Florida Keys reef tract included hard-bottom reef habitat from Key West to the northern end of Key Largo, and focuses on three regions of interest that encompass the eastern and western Lower Keys and Key Largo. The Qualitative, drop-camera, accuracy assessment (AA) analyses for all three regions of interest gave overall accuracies of 84.2%, ±16.9, at the major level of geomorphological structure, 85.4%, ±16.4, and 73.8%, ±18.7, at the major and detailed levels of biological cover and 70.4%, ±20.6, for detailed coral cover. The Quantitative, in situ, AA analyses for all three regions of interest gave overall accuracies of 86.1%, ±0, at the major level of geomorphological structure, 85.2%, ±1.9, and 50.7%, ±13.4, at the major and detailed levels of biological cover and 47.5%, ±13.4, for coral cover. Qualitative and quantitative accuracies were similar at the major geologic structure (hard vs. soft bottom) and major biological cover (i.e. seagrass, algae) however qualitative AA’s for detailed biological cover (i.e. percent of seagrass, algae) and detailed coral cover (percent of coral) were 23.1% and 22.9% higher than the quantitative AA’s. This trend was also found when analyzing the accuracies for the individual regions of interest. The results suggest that for performing an AA of broad map categories, a Qualitative AA compares well with an in situ Quantitative AA, but for more detailed map categories the in situ quantitative AA is more accurate. Marine resource managers should consider these accuracies when making decisions based on the 2014 NOAA Florida Keys Coral Reef Ecosystem Habitat map.

Creative Commons License

Creative Commons Attribution 4.0 License
This work is licensed under a Creative Commons Attribution 4.0 License.

Share

COinS