| Summary: | Coral reef ecosystems are the most diverse on earth, and their subsistence is being
threatened by natural and adverse anthropogenic patterns and processes. In an effort to
understand and protect these marine environments, several programs have outlined
strategies and initiatives. For example, the United States Coral Reef Task Force���s
Mapping and Information Working Group has outlined a specific goal to map all coral
reefs below 30 m depth by 2009. This study contributes to achieving that goal for three
sites around the island of Tutuila, American Samoa, lying in the heart of the South
Pacific. American Samoa, a U.S. territory, is home to the Fagatele Bay National Marine
Sanctuary, the smallest and most remote in the United States, and to the National Park of
American Samoa. Extensive modern scientific surveys were implemented around the
territory in 2001 and have since continued and increased. The presence of protected
areas and the existence of scientific data collected with state of the art technology have
made the site a priority for the Coral Reef Task Force. In this study, methods for
classifying surficial seafloor characteristics as bathymetric position index (BPI) zones
and structures were developed and applied to the study sites. BPI zones and structures
were classified by using algorithms that combine high-resolution (1 m) multibeam
bathymetry and its derivatives: bathymetric position index at multiple scales and slope.
The development of algorithms and the classification scheme involved the use of
historical and current classification studies and three-dimensional visualization. In
addition, the BPI zones and structures were compared to limited biological, geological,
and physical attributes recorded during accuracy assessment surveys (photos) and towed
diver surveys (video). A rugosity (surface ratio) analysis was added to the study to give a
picture of the seafloor roughness. The BPI zone and structure classifications overlap and
extend existing classifications from Ikonos satellite imagery for water depths shallower
than 30 m. Methods, data and classifications developed and applied in this study will be
available to the public as a benthic habitat mapping tool (ArcGIS extension), in an online
GIS data archive, and on a compact disc attached to this thesis. They contribute to a
broader understanding of the marine and coastal environment and will serve as a baseline
of information for benthic habitat mapping and future biological, ecological, and
geological surveys. The baseline gives a good indication of characteristics that may
indicate areas of high biodiversity. The final maps presented here are especially useful to
managers, researchers and scientists that seek to establish and monitor a wider and more
effective network of marine and coastal protection. === Graduation date: 2005
|
|---|
