Estimating Regions of Oceanographic Importance for Seabirds Using A-Spatial Data

Area of Science:

• Marine Ecology
• Ornithology
• Conservation Biology

Background:

• GPS tracking advances enable oceanographic region assessment for seabirds.
• Long-term population data may exist where GPS data is unavailable.
• Understanding seabird distribution and population success factors is crucial.

Purpose of the Study:

• To present a method for inferring important oceanographic regions for seabirds.
• To use breeding sooty shearwaters as a case study.
• To combine machine learning, GIS, ecological data, and oceanographic datasets.

Main Methods:

• Utilized generalized boosted regression modeling (a machine learning algorithm).
• Integrated geographic information systems (GIS) and open-access oceanographic data.
• Employed time series of chick size and harvest index from historical Maori 'muttonbirder' diaries as response variables in a spatial model.

Main Results:

• Identified sub-Antarctic water regions as key areas explaining chick size variation.
• Found wind speed and charnock (ocean surface roughness) to be top predictor variables.
• Confirmed these regions function as sooty shearwater breeding season flyways.

Conclusions:

• Wind speeds in identified flyways likely impact sooty shearwater chick provisioning due to altered flight dynamics.
• The presented approach can be implemented with various statistical algorithms.
• This method is applicable to any long-term population time series for identifying species-specific important regions.