Predicting current and future global distributions of whale sharks

Authors

  • Ana M. M. Sequeira,

    Corresponding author
    1. The Environment Institute and School of Earth and Environmental Sciences, The University of Adelaide, South Australia, Australia
    • Correspondence: Ana M. M. Sequeira, tel. +61 0 8 6488 2219, fax +61 0 8 8313 4347, e-mail: ana.sequeira@uwa.edu.au

    Search for more papers by this author
  • Camille Mellin,

    1. The Environment Institute and School of Earth and Environmental Sciences, The University of Adelaide, South Australia, Australia
    2. Australian Institute of Marine Science, Townsville, Queensland, Australia
    Search for more papers by this author
  • Damien A. Fordham,

    1. The Environment Institute and School of Earth and Environmental Sciences, The University of Adelaide, South Australia, Australia
    Search for more papers by this author
  • Mark G. Meekan,

    1. Australian Institute of Marine Science, UWA Oceans Institute (MO96), Crawley, Western Australia, Australia
    Search for more papers by this author
  • Corey J. A. Bradshaw

    1. The Environment Institute and School of Earth and Environmental Sciences, The University of Adelaide, South Australia, Australia
    2. South Australian Research and Development Institute, Henley Beach, South Australia, Australia
    Search for more papers by this author

Abstract

The Vulnerable (IUCN) whale shark spans warm and temperate waters around the globe. However, their present-day and possible future global distribution has never been predicted. Using 30 years (1980–2010) of whale shark observations recorded by tuna purse-seiners fishing in the Atlantic, Indian and Pacific Oceans, we applied generalized linear mixed-effects models to test the hypothesis that similar environmental covariates predict whale shark occurrence in all major ocean basins. We derived global predictors from satellite images for chlorophyll a and sea surface temperature, and bathymetric charts for depth, bottom slope and distance to shore. We randomly generated pseudo-absences within the area covered by the fisheries, and included fishing effort as an offset to account for potential sampling bias. We predicted sea surface temperatures for 2070 using an ensemble of five global circulation models under a no climate-policy reference scenario, and used these to predict changes in distribution. The full model (excluding standard deviation of sea surface temperature) had the highest relative statistical support (wAICc = 0.99) and explained ca. 60% of the deviance. Habitat suitability was mainly driven by spatial variation in bathymetry and sea surface temperature among oceans, although these effects differed slightly among oceans. Predicted changes in sea surface temperature resulted in a slight shift of suitable habitat towards the poles in both the Atlantic and Indian Oceans (ca. 5°N and 3–8°S, respectively) accompanied by an overall range contraction (2.5–7.4% and 1.1–6.3%, respectively). Predicted changes in the Pacific Ocean were small. Assuming that whale shark environmental requirements and human disturbances (i.e. no stabilization of greenhouse gas emissions) remain similar, we show that warming sea surface temperatures might promote a net retreat from current aggregation areas and an overall redistribution of the species.

Ancillary