MEPS 533:277-290 (2015)  -  DOI: https://doi.org/10.3354/meps11316

Key seabird areas in southern New England identified using a community occupancy model

Nicholas P. Flanders1,2,*, Beth Gardner1, Kristopher J. Winiarski3, Peter W. C. Paton4, Taber Allison5,7, Allan F. O’Connell6

1Department of Forestry and Environmental Resources, North Carolina State University, Raleigh, NC 27695, USA
2Department of Biological Sciences, Old Dominion University, Norfolk, VA 23529, USA
3Department of Environmental Conservation, University of Massachusetts, 160 Holdsworth Way, Amherst, MA 01003, USA
4Department of Natural Resources Science, University of Rhode Island, 1 Greenhouse Road, Kingston, RI 02881, USA
5Mass Audubon, 208 South Great Road, Lincoln, MA 01773, USA
6US Geological Survey, Patuxent Wildlife Research Center-Beltsville Lab, Beltsville, MD 20705, USA
7Present address: American Wind Wildlife Institute, 1110 Vermont Ave., NW, Washington, DC 20005, USA
*Corresponding author:

ABSTRACT: Seabirds are of conservation concern, and as new potential risks to seabirds are arising, the need to provide unbiased estimates of species’ distributions is growing. We applied community occupancy models to detection/non-detection data collected from repeated aerial strip-transect surveys conducted in 2 large study plots off southern New England, USA; one off the coast of Rhode Island and the other in Nantucket Sound. A total of 17 seabird species were observed at least once in each study plot. We found that detection varied by survey date and effort for most species and the average detection probability across species was less than 0.4. We estimated the influence of water depth, sea surface temperature, and sea surface chl a concentration on species-specific occupancy. Diving species showed large differences between the 2 study plots in their predicted winter distributions, which were largely explained by water depth acting as a stronger predictor of occupancy in Rhode Island than in Nantucket Sound. Conversely, similarities between the 2 study plots in predicted winter distributions of surface-feeding species were explained by sea surface temperature or chlorophyll a concentration acting as predictors of these species’ occupancy in both study plots. We predicted the number of species at each site using the observed data in order to detect ‘hot-spots’ of seabird diversity and use in the 2 study plots. These results provide new information on detection of species, areas of use, and relationships with environmental variables that will be valuable for biologists and planners interested in seabird conservation in the region.


KEY WORDS: Aerial strip-transect survey · Community occupancy model · Imperfect detection · Seabird habitat relationships · Species distribution models


Full text in pdf format
Supplementary material 
Cite this article as: Flanders NP, Gardner B, Winiarski KJ, Paton PWC, Allison T, O’Connell AF (2015) Key seabird areas in southern New England identified using a community occupancy model. Mar Ecol Prog Ser 533:277-290. https://doi.org/10.3354/meps11316

Export citation
Mail this link - Contents Mailing Lists - RSS
- -