Northern Prairie Wildlife Research Center
Bradley L. Weigle and Richard O. Flamm, Florida Department of Environmental Protection, Florida Marine Research Institute, 100 Eighth Avenue S.E., St. Petersburg, FL 33701 USA
Marine mammal biologists at the Florida Department of Environmental Protection (FDEP) conducted a six-year telemetry study of 59 Florida manatees (Trichechus manatus latirostris) along the west-central coast of Florida between 1991 and 1996. Primary objectives of the study were to determine habitat utilization and travel patterns of this endangered sirenian as they dispersed from and returned to a winter aggregation site in Tampa Bay. Satellite transmitter signals processed by Service Argos provided over 33,000 animal locations of location classes one to three. Field tracking using VHF signals allowed collection of an additional 3,000 visual observations. Satellite data were integrated into a geographic information system (GIS) as point locations using a set of programs that combined the monthly update from Service Argos with a database containing information on individual animals and tag attachment/detachment dates. Visual data were entered into the GIS using "heads-up" digitizing from field data sheets based on 1:40,000 scale nautical charts. The two data sets were combined.
Analyses to estimate movement paths of individuals between sequential points, while keeping the path in the water, used a customized program centered on the "cost-path" function of ARC/INFO's GRID module and required the creation of two raster grids. The vector marine shoreline was used to create a grid with 25-m by 25-m cells containing land and water values. A grid with cells containing the maximum depth at mean low tide was generated from the bathymetry coverage. These cell values were reclassified to reflect manatee depth preferences determined from an empirical analysis relating each satellite point with location class three to its corresponding bathymetry. Cells with depths preferred by manatees were assigned a smaller value (cost) than deeper cells that were rarely used. The cost-path function then calculated the path between two sequential points with the lowest associated total cost. Each path was converted to a vector line for storage with attribute data including manatee information; duration; and time, date, and location class at start and end points of the path.
The line data set can be queried to select specific subsets such as individual animals, reproductively mature males, females with calves, nocturnal periods, or specific months. Another set of programs then reconstructs a set of five raster surfaces for the selected line data. A total time grid that accumulates time data for each line is generated: the total time in minutes between the start and end points is apportioned equally into each cell along the path. Four other grids contain values for 1) the number of times a path crossed each cell (total visits); 2) maximum time spent in a cell by any one path; 3) minimum time spent; and 4) sum of squares of time from each path. A circular spatial filter applied to each grid smooths the data over an area corresponding to a circle of 175-m diameter. By dividing the total time grid by the total visits grid, a grid surface depicting the average time per visit is created. To further simplify the data, four cell classes are created: cells with few visits and low mean time; few visits and higher mean time (secondary habitat); many visits and low mean time (potential travel corridors); and many visits and higher mean time (primary habitat). Model results can be overlaid on marine habitat data layers (sea grasses, boating channels, dredging projects, etc.) to explore and evaluate anthropogenic effects on the manatee population.