Using artificial neural networks to model the suitability of coastline for breeding by New Zealand fur seals (Arctocephalus forsteri)

Corey J.A. Bradshaw, Lloyd S Davis, Martin Purvis, Qingqing Zhou, George L Benwell

Research output: Contribution to journalArticlepeer-review

29 Citations (Scopus)

Abstract

New Zealand fur seal (Arctocephalus forsteri) numbers and distribution were reduced by human exploitation but the species is now re-colonizing much of its former range. Pinnipeds occupy two different habitat media: the marine (feeding) and terrestrial (breeding) environments. Measures of geographic variation in both these environments can be modelled together to predict coastline suitability for colonization (i.e. potential availability of breeding sites). To avoid problems of non-linear modelling, we used an artificial neural network (ANN) to: (1) predict the suitability of coastline in South Island, New Zealand to support breeding A. forsteri colonies by creating a model using pup condition (measured from 20 breeding colonies during 1996–98), prey distribution and abundance, bathymetry, and the type of coastal substrate; (2) compare the predicted distribution of suitable coastline for colonization from the model to the current distribution of A. forsteri colonies (n=198 colonies); and (3) using ANN inference rule extraction, determine which factors are the most influential in predicting coastline suitability. ANN model predictions overlapped current distributions of A. forsteri colonies in South Island. Inference rule extraction gave good predictions of colony performance (i.e. the ability to predict observed pup condition); however, they were not consistent among years in terms of the prey species constituting the rules or in the direction of the relationships. Arrow squid and octopus were important model terms in 1996 and 1997, but the direction of their coefficients in the inference rules were opposite between years. Hoki was an important term in 1997 and 1998, but it also varied in direction between years. Terms of secondary importance include the distance from sample colonies to 250 m-, 500 m- and 1000 m-isobaths. Variation in model predictions may result from climatic variation, the constant index of prey availability that was used and the potential for A. forsteri to switch main prey species among year. Resource availability appears to be a good predictor of the potential distribution of A. forsteri colonies, but future models should attempt to incorporate indices of temporal variation in resource availability as well as population density to better predict the colonization process and understand the ecological mechanisms operating within.
Original languageEnglish
Pages (from-to)111-131
Number of pages21
JournalEcological Modelling
Volume148
Issue number2
DOIs
Publication statusPublished - 15 Feb 2002
Externally publishedYes

Fingerprint

Dive into the research topics of 'Using artificial neural networks to model the suitability of coastline for breeding by New Zealand fur seals (<i>Arctocephalus forsteri</i>)'. Together they form a unique fingerprint.

Cite this