Fine-scale habitat partitioning of Chilean and Peale’s dolphins and their overlap with aquaculture

Abstract

1. Predictive species distribution models (SDMs) have become powerful tools to determine habitat use patterns of mobile marine predators and their spatial overlap with potentially impacting anthropogenic activities. 2. This study used SDMs to investigate fine‐scale habitat use patterns of two poorly known and broadly sympatric coastal delphinids, Chilean dolphins (Cephalorhynchus eutropia) and Peale's dolphins (Lagenorhynchus australis), and their spatial interactions with intense aquaculture farming activities in the Chiloé archipelago, southern Chile. 3. A long‐term dataset (2002–2012) of boat‐based dolphin sightings and concurrently in situ collected environmental and anthropogenic variables was analysed using binomial Generalized Additive Models to investigate ecological drivers of each species' fine‐scale distribution and to predict dolphin occurrence spatially. 4. Chilean dolphins preferred shallow (<30 m deep), turbid waters, close to shore (<500 m) and river mouths which often placed them in sheltered bays and channels used intensively by shellfish farms. Peale's dolphins were also found in shallow waters but occurred over a wider range of conditions along more open or exposed coastlines. Both species had to navigate extensive salmon and shellfish farming sites to transit between areas of important habitat. 5. Sightings and predicted occurrence maps showed a clear pattern of spatial habitat partitioning between species, which remained stable across the 11 year study period. The identification of important habitat for Chilean dolphins warrants the consideration of spatially explicit conservation measures to limit the potential effects of overlapping salmon and shellfish farming. 6. The observed differences in ecological plasticity of the two sympatric species should be considered when evaluating and mitigating the effects of environmental change and ongoing anthropogenic pressures on their nearshore habitat. The estimated species–environment relationships could also be used to predict where dolphin habitat and anthropogenic activities are most likely to overlap in other parts of the species' ranges.