Seasonal productivity drives aggregations of killer whales and other cetaceans over submarine canyons of the Bremer Sub-Basin, south-western Australia

• Published in: Australian Mammalogy
• Main Authors: Salgado Kent, Chandra, Bouchet, Phil, Wellard, Rebecca, Parnum, Iain, Fouda, Leila, Erbe, Christine
• Format: Article in Journal/Newspaper
• Language: English
• Published: CSIRO PUBLISHING 2020
• Subjects: Science & Technology; Life Sciences & Biomedicine; Zoology; generalised estimating equations; habitat modelling; submarine canyons; temporal autocorrelation; whale watching; ESTIMATING EQUATIONS; FORAGING BEHAVIOR; HABITAT; MODELS; POPULATION; ABUNDANCE; PREFERENCES; SELECTION; POD; BAY; Killer Whale; Orca; Orcinus orca; Killer whale
• Online Access: https://hdl.handle.net/20.500.11937/80261; https://doi.org/10.1071/AM19058

Cetaceans are iconic predators that serve as important indicators of marine ecosystem health. The Bremer Sub-Basin, south-western Australia, supports a diverse cetacean community including the largest documented aggregation of killer whales (Orcinus orca) in Australian waters. Knowledge of cetacean distributions is critical for managing the area’s thriving ecotourism industry, yet is largely sporadic. Here we combined aerial with opportunistic ship-borne surveys during 2015–2017 to describe the occurrence of multiple cetacean species on a regional scale. We used generalised estimating equations to model variation in killer whale relative density as a function of both static and dynamic covariates, including seabed depth, slope, and chlorophyll a concentration, while accounting for autocorrelation. Encountered cetacean groups included: killer (n ¼ 177), sperm (n ¼ 69), long-finned pilot (n ¼ 29), false killer (n ¼ 2), and straptoothed beaked (n ¼ 1) whales, as well as bottlenose (n ¼ 12) and common (n ¼ 5) dolphins. Killer whale numbers peaked in areas of low temperatures and high primary productivity, likely due to seasonal upwelling of nutrient-rich waters supporting high prey biomass. The best predictive model highlighted potential killer whale ‘hotspots’ in the Henry, Hood, Pallinup and Bremer Canyons. This study demonstrates the value of abundance data from platforms of opportunity for marine planning and wildlife management in the open ocean.