Dynamic ensemble models to predict distributions and anthropogenic risk exposure for highly mobile species

• DOI: 10.1111/ddi.12940
• Publication Date: 2019-06-04T10:15:21Z
• Authors: Briana Abrahms; Heather Welch; Stephanie Brodie; Michael G. Jacox; Elizabeth A. Becker; Steven J. Bograd; Ladd M. Irvine; Daniel M. Palacios; Bruce R. Mate; Elliott L. Hazen
• Keywords: Species distribution; Environmental niche modelling

Abstract Aim Advances in ecological and environmental modelling offer new opportunities for estimating dynamic habitat suitability highly mobile species supporting management strategies at relevant spatiotemporal scales. We used an ensemble approach to predict daily, year‐round a migratory species, the blue whale ( Balaenoptera musculus ), demonstrate application evaluating dynamics of their exposure ship strike risk. Location The California Current Ecosystem (CCE) Southern Bight (SCB), USA. Methods integrated long‐term (1994–2008) satellite tracking dataset on 104 whales with data‐assimilative ocean model output assess suitability. evaluated relative utility ensembling multiple types compared using single models, selected validated candidate models cross‐validation metrics independent observer data. quantified spatial temporal distribution risk within shipping lanes SCB. Results Multi‐model ensembles outperformed single‐model approaches. final had high predictive skill (AUC = 0.95), resulting predictions CCE that accurately captured behaviour. Risk was variable among years as function conditions (e.g., marine heatwave). Main conclusions Daily information three‐dimensional oceanic habitats daily power indicated could benefit by incorporating information. This is readily transferable other species. Dynamic, high‐resolution are valuable tools assessing targeting needs.