Climate variables are not the dominant predictor of Arctic shorebird distributions

RESEARCH ARTICLE Climate variables are not the dominant predictor of Arctic shorebird distributions Christine M. AndersonID1*, Lenore Fahrig1, Jennie Rausch2, Paul A. Smith3 1 Department of Biology, Geomatics and Landscape Ecology Laboratory, Carleton University, Ottawa, ON, Canada, 2 Canadian Wildlife Service, Environment and Climate Change Canada, Yellowknife, NT, Canada, 3 Wildlife Research Division, Environment and Climate Change Canada, Ottawa, ON, Canada * christineanderson3@carleton.ca Abstract Competing theoretical perspectives about whether or not climate is the dominant factor influ- encing species’ distributions at large spatial scales have important consequences when habitat suitability models are used to address conservation problems. In this study, we tested how much variables in addition to climate help to explain habitat suitability for Arctic- breeding shorebirds. To do this we model species occupancy using path analyses, which allow us to estimate the indirect effects of climate on other predictor variables, such as land cover. We also use deviance partitioning to quantify the total relative importance of climate versus additional predictors in explaining species occupancy. We found that individual land cover variables are often stronger predictors than the direct and indirect effects of climate combined. In models with both climate and additional variables, on average the additional variables accounted for 57% of the explained deviance, independent of shared effects with

the climate variables. Our results support the idea that climate-only models may offer incom- plete descriptions of current and future habitat suitability and can lead to incorrect conclu- sions about the size and location of suitable habitat. These conclusions could have important management implications for designating protected areas and assessing threats like climate change and human development. Introduction Habitat suitability models are used to describe the relationship between species’ presence in geographic space and the characteristics of the environment, and to estimate how likely a spe- cies is to occupy unsampled locations. Along with closely related species distribution models, environmental niche models, resource selection functions and bioclimatic envelope models [but see 1], these analyses are increasingly popular for addressing a wide range of questions, such as how species will respond to climate change [e.g. 2], identifying priority areas for threat- ened species [e.g. 3], managing landscapes [e.g. 4], and understanding the spread of invasive species [e.g. 5]. However, different theoretical perspectives about how these relationships between species and their environment are structured have important consequences for the variables included in these models, and therefore how we use them to address conservation problems. PLOS ONE PLOS ONE | https://doi.org/10.1371/journal.pone.0285115 May 17, 2023 1 / 14 a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 OPEN ACCESS Citation: Anderson CM, Fahrig L, Rausch J, Smith

PA (2023) Climate variables are not the dominant predictor of Arctic shorebird distributions. PLoS ONE 18(5): e0285115. https://doi.org/10.1371/ journal.pone.0285115 Editor: Tunira Bhadauria, Feroze Gandhi Degree College, INDIA Received: February 9, 2023 Accepted: April 16, 2023 Published: May 17, 2023 Peer Review History: PLOS recognizes the benefits of transparency in the peer review process; therefore, we enable the publication of all of the content of peer review and author responses alongside final, published articles. The editorial history of this article is available here: https://doi.org/10.1371/journal.pone.0285115 Copyright: © 2023 Anderson et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Data Availability Statement: All data and code files are available from the Zenodo database (https://doi. org/10.5281/zenodo.7613304). Funding: PAS - P57 - ArcticNet - https://arcticnet. ulaval.ca PAS - 2017-06391 - National Science and