Inter-Research > ESR > v53 > p227-245  
Endangered Species Research

via Mailchimp

ESR 53:227-245 (2024)  -  DOI:

Sea turtle density surface models along the United States Atlantic coast

A. DiMatteo1,2,*, J. J. Roberts3, D. Jones4, L. Garrison5, K. M. Hart6, R. D. Kenney7, W. A. McLellan8, K. Lomac-MacNair9,10, D. Palka11, M. E. Rickard12, K. E. Roberts13, A. M. Zoidis9,10, L. Sparks14

1McLaughin Research Corporation, Middletown, RI 02842, USA
2CheloniData LLC, Berthoud, CO 80513, USA
3Marine Geospatial Ecology Lab, Duke University, Durham, NC 27708, USA
4Naval Facilities Engineering Systems Command, Atlantic, Norfolk, VA 23508, USA
5National Marine Fisheries Service, Southeast Fisheries Science Center, Miami, FL 33149, USA
6U.S. Geological Survey, Wetland and Aquatic Research Center, Davie, FL 33314, USA
7University of Rhode Island Graduate School of Oceanography, Narragansett, RI 02882, USA
8Biology & Marine Biology, University of North Carolina Wilmington, Wilmington, NC 28403, USA
9Tetra Tech, Oakland, CA 94612, USA
10Cetos Research Organization, Berkeley, CA 94709, USA
11National Marine Fisheries Service, Northeast Fisheries Science Center, Woods Hole, MA 02543, USA
12New York Natural Heritage Program, College of Environmental Science and Forestry, State University of New York, Kings Park, NY 11754, USA
13Department of Oceanography and Coastal Sciences, Louisiana State University, Baton Rouge, LA 70803, USA
14Naval Undersea Warfare Center, Newport, Newport, RI 02841, USA
*Corresponding author:

ABSTRACT: Spatially explicit estimates of marine species distribution and abundance are required to quantify potential impacts from human activities such as military training and testing, fisheries interactions, and offshore energy development. There are 4 protected species of sea turtle (loggerhead, green, Kemp’s ridley, and leatherback) commonly found along the east coast of the USA, our study area, and which require impact assessments. Data from 7 different survey organizations were used to create density surface models for the 4 sea turtle species utilizing 1.2 million km of line-transect surveys. A substantial portion (29.7%) of available sightings were not identified to the species level. Not including these sightings would underestimate density, so a conditional random forest model was used to assign unidentified sightings to species. Higher densities of loggerhead, green, and Kemp’s ridley sea turtles were predicted south of the Outer Banks in cool months, transitioning northwards in late spring to occupy seasonal neritic habitats. The highest leatherback densities were predicted off the coasts of Georgia and Florida. Leatherbacks were also predicted throughout offshore areas. The predicted distribution patterns generally matched satellite tracking and strandings data, indicating the models reproduced established seasonal movements. Surveys rarely detect sea turtles smaller than 40 cm, so these age classes are not represented. The models are the first for the study area to apply availability bias estimates developed in or near the study area and attempt to classify unidentified sightings to the species level, providing an updated, critical tool for conservation management along the eastern seaboard.

KEY WORDS: Density · Abundance · Sea turtle · Classification · Availability bias

Full text in pdf format
Supplementary material
Cite this article as: DiMatteo A, Roberts JJ, Jones D, Garrison L and others (2024) Sea turtle density surface models along the United States Atlantic coast. Endang Species Res 53:227-245.

Export citation
Share:    Facebook - - linkedIn

 Previous article Next article