Advanced

The utility of relative environmental suitability (RES) modelling for predicting distributions of seabirds in the North Atlantic

Watson, Hannah LU ; Hiddink, Jan G.; Hobbs, Matthew J.; Brereton, Tom M. and Tetley, Michael J. (2013) In Marine Ecology - Progress Series 485. p.259-273
Abstract
Understanding spatial and temporal variability in the distribution of seabirds is fundamental for the conservation and management of marine ecosystems. In the absence of large-scale systematic survey data, the application of standard habitat modelling techniques to predict the at-sea distributions of seabirds at large spatial scales has been limited. In this study, we examine the utility of relative environmental suitability (RES) modelling to predict large-scale distributions and habitat suitability for 6 seabirds in the North Atlantic. An index of habitat suitability was derived by relating niche characteristics to environmental attributes. Predictive performance of models was evaluated with Receiver Operating Characteristic plots, using... (More)
Understanding spatial and temporal variability in the distribution of seabirds is fundamental for the conservation and management of marine ecosystems. In the absence of large-scale systematic survey data, the application of standard habitat modelling techniques to predict the at-sea distributions of seabirds at large spatial scales has been limited. In this study, we examine the utility of relative environmental suitability (RES) modelling to predict large-scale distributions and habitat suitability for 6 seabirds in the North Atlantic. An index of habitat suitability was derived by relating niche characteristics to environmental attributes. Predictive performance of models was evaluated with Receiver Operating Characteristic plots, using independent survey data from the Bay of Biscay. RES models performed significantly better than null models at predicting relative likelihood of occurrence for 5 out of 6 species. Qualitative assessment showed that model outputs corresponded well with published range maps, though a common discrepancy was the inclusion of enclosed seas in which species are not known to regularly occur. This study demonstrates that RES modelling can be used to predict large-scale habitat suitability for wide-ranging marine animals for which occurrence data are limited and biased in geographical extent. RES predictions represent simple, testable hypotheses concerning a species’ potential niche in respect of a few environmental predictors. RES modelling can help to identify biodiversity hotspots, predict effects of climate change and develop criteria for designating marine protected areas. (Less)
Please use this url to cite or link to this publication:
author
publishing date
type
Contribution to journal
publication status
published
subject
in
Marine Ecology - Progress Series
volume
485
pages
259 - 273
publisher
Inter-Research
external identifiers
  • Scopus:84879624284
ISSN
1616-1599
DOI
10.3354/meps10334
language
English
LU publication?
no
id
965681a5-16a1-49df-834b-03c162545ea5 (old id 4693425)
alternative location
http://www.int-res.com/abstracts/meps/v485/p259-273/
date added to LUP
2014-10-03 11:38:59
date last changed
2016-11-18 13:45:59
@misc{965681a5-16a1-49df-834b-03c162545ea5,
  abstract     = {Understanding spatial and temporal variability in the distribution of seabirds is fundamental for the conservation and management of marine ecosystems. In the absence of large-scale systematic survey data, the application of standard habitat modelling techniques to predict the at-sea distributions of seabirds at large spatial scales has been limited. In this study, we examine the utility of relative environmental suitability (RES) modelling to predict large-scale distributions and habitat suitability for 6 seabirds in the North Atlantic. An index of habitat suitability was derived by relating niche characteristics to environmental attributes. Predictive performance of models was evaluated with Receiver Operating Characteristic plots, using independent survey data from the Bay of Biscay. RES models performed significantly better than null models at predicting relative likelihood of occurrence for 5 out of 6 species. Qualitative assessment showed that model outputs corresponded well with published range maps, though a common discrepancy was the inclusion of enclosed seas in which species are not known to regularly occur. This study demonstrates that RES modelling can be used to predict large-scale habitat suitability for wide-ranging marine animals for which occurrence data are limited and biased in geographical extent. RES predictions represent simple, testable hypotheses concerning a species’ potential niche in respect of a few environmental predictors. RES modelling can help to identify biodiversity hotspots, predict effects of climate change and develop criteria for designating marine protected areas.},
  author       = {Watson, Hannah and Hiddink, Jan G. and Hobbs, Matthew J. and Brereton, Tom M. and Tetley, Michael J.},
  issn         = {1616-1599},
  language     = {eng},
  pages        = {259--273},
  publisher    = {ARRAY(0x7f52bb0)},
  series       = {Marine Ecology - Progress Series},
  title        = {The utility of relative environmental suitability (RES) modelling for predicting distributions of seabirds in the North Atlantic},
  url          = {http://dx.doi.org/10.3354/meps10334},
  volume       = {485},
  year         = {2013},
}