Mapping critical Australian sea lion habitat to assess ecological value and risks to population recovery

The seafloor habitats critical to sea lions, their vulnerability to human impacts and the role of existing management measures (marine protected areas) in protecting them, are poorly understood. Potential threats include bycatch in fisheries, entanglement in marine debris, disease, pollution, and climate change. Although measures to mitigate bycatch impacts in demersal gillnet and lobster fisheries were introduced a decade ago, declines in some populations cannot be explained by fishery interactions.

Monitoring the population status of Australian sea lions is challenging due to their unique, 18 month breeding cycle that is asynchronous across breeding colonies. Previous satellite tracking and time-depth recorders have provided a wealth of data, but the instruments were too large and unreliable to explore the habitat use of populations. This pilot study assessed the reliability and feasibility of state-of-the-art animal-borne cameras, combined with high-resolution GPS and tri-axial accelerometer/magnetometer devices, to map and identify benthic habitats used by the endangered Australian sea lion. The results provide an improved understanding of sea lion habitat use and behaviour to support future conservation actions to recover the species.

Approach and findings

Cameras and tracking instruments were deployed on eight adult female sea lions during a single foraging trip (2–6 days), four at Seal Bay (Kangaroo Island) and four at Olive Island (off Streaky Bay, western Eyre Peninsula, South Australia). Each deployment recorded approximately eight hours of footage (range 6.7 to 12.8 hours), capturing nearly 80 hours in total and covering 560 km of seabed at depths between 5 and 110 metres.

The footage provided new information on sea lion habitat use, foraging behaviours, and prey species. The sea lions utilised six benthic habitat types: macroalgae reef, macroalgae meadow, bare sand, sponge/sand, invertebrate reef, and invertebrate boulder habitats. Sea lions from Olive Island primarily used macroalgae reef (37%), bare sand (36%), and sponge/sand (21%), while those from Seal Bay favoured invertebrate reef (38%), bare sand (16%), sponge/sand (15%), and invertebrate boulder habitats (13%). Using machine learning-based Random Forest models, habitat maps were developed over a broad area of the shelf adjacent to each colony, spanning more than 5,000 km².

Each deployment yielded valuable and unique information, refining our understanding of sea lion habitat use and behaviour. The animal-borne instruments facilitated the discovery and mapping of hundreds of kilometres of previously unmapped seabed, including previously unknown rocky reefs and kelp forests. This innovative and cost-effective approach allows for the exploration of large areas of unmapped shelf waters, locating crucial sea lion habitats both inside and outside marine reserves, while complementing traditional benthic survey methods.

Although some sea lions foraged within marine reserves, most of the habitat used in this study was outside protected areas. Moving forward, sea lion video data will be invaluable for managing populations, marine park planning, and broader benthic habitat surveys. The cameras provide robust observational data to inform the locations of Biologically Important Areas, assisting in the delineation of protection zones critical to sea lion survival. Feeding rate data will help distinguish between transitory and foraging habitats and enable the ecological value of different habitats to be estimated. This will enhance the quantitative assessment of the effectiveness of marine park zoning in safeguarding key foraging habitats.

Sea lion video data can be used to validate existing benthic habitat models and provide valuable georeferenced observations that would otherwise be expensive to obtain. Repeat surveys will allow for the evaluation of ecological and habitat changes over time. Improved information on diet and consumption rates will aid in developing ecosystem models to assess the impacts from ecological change and prey depletion due to fisheries and climate change, and the potential impacts of emerging industries, such as offshore renewables.

As part of the project’s engagement and participation initiatives, six Far West Coast Aboriginal rangers contributed to the fieldwork on Olive Island. Presentations about the project were also delivered to students at Yalata Anangu School and Ceduna Area School. The project team also took part in the NAIDOC Family Day community event held at Koonibba Oval in Ceduna.

Ongoing hub research

Ongoing research under Project 4.14 is continuing camera deployments at Seal Bay, a major breeding site that is relatively easy to access, and at other sites off the western Eyre Peninsula. Further deployments at Seal Bay will provide a greater understanding of the level of within-colony, and inter-individual variability in habitat-use and foraging strategies and allow validation and further development of habitat models developed for this location. Deployments planned at other sites off the western Eyre Peninsula will progress knowledge on the variability of habitat-use and foraging strategies among colonies, as well as provide new data and maps of the distribution of benthic habitats elsewhere in the sea lion’s range.

Outcomes

New information on the movement and habitat-use of Australian sea lions provided by this project supports:

• revision of the Recovery Plan for the Australian Sea Lion;
• evaluation of risks to populations from fisheries interactions (bycatch and prey depletion) and other threats; and
• implementation of the Parks Australia South-west network management plan which recognises the Australian sea lion as both a core natural value, and as being directly impacted by pressures (climate change, extraction of living resources, human presence and marine pollution).

The mapping and tracking method developed in this project can be used to:

• record sea lion habitat and foraging strategies in detail;
• map the distribution of different habitats across large areas of the continental shelf; and
• evaluate the importance of different habitats to Australian sea lions, including inside and outside marine protected areas.

The method provides a cost-effective way to explore large areas of unmapped shelf and locate valuable sea lion habitat both within and outside of marine reserves, complementing existing survey methods. In the future, these data will be extremely valuable for directing management initiatives for Australian sea lions and other marine species, marine park planning, and benthic surveying more broadly.

The sea lion camera data also provides robust observational data to inform Biologically Important Areas (BIAs) for the species and delineate zones that would be valuable for the protection of Australian sea lions, including protection of habitat types critical to their survival. Additional analyses of feeding rate data will facilitate estimation the ecological value of different habitats to the species, and to evaluate the value and efficiency of marine park zoning for protecting critical sea lion foraging habitats.

Sea lion camera data could be used to validate existing benthic habitat models and improve foraging models for the Australian sea lion. Camera data also provide valuable georeferenced observational records of habitat, providing capacity for future repeat surveys to evaluate ecological and habitat quality changes over time, especially those critical to the survival of the Australian sea lion.

Further evaluation of dual camera and satellite telemetry data can offer insight to the degree of habitat protection provided by marine reserves and the level of sampling required for individual breeding sites to obtain representative data. Data from this project will be made available to other mapping projects to improve the seafloor habitat maps of our southern shelf.

Project location

Australian sea lion breeding sites off South Australia including Olive Island, Pearson Island and Seal Bay, Kangaroo Island