Changes in minke whale distributions in the Southern Ocean

No abstracts are to be cited without prior reference to the author.

This study explores the various relationships between distance from the sea ice edge and minke whale density distribution by means of a circumpolar spatial analysis. For almost 30 years, the International Whaling Commission (IWC) has conducted whale surveys in the Southern Ocean during austral summer months under the IWC/IDCR-SOWER programme. This has resulted in three circumpolar sets of surveys, with over 21,000 minke whale sightings. We used these sightings in spatial models of line transect data based on generalised additive models (GAMs). The GAMs assumed an overdispersed Poisson error structure and log-link. Model selection was based on maximisation of explained deviance and minimising the Generalised Cross Validation (GCV) score, while excluding GAMs that generated extreme and unlikely minke whale densities. The GAMs were fitted independently by survey area and year. Selected GAMs included combinations of the following covariates: closest distance of sighting to sea ice edge, closest distance of sighting to either the Southern Antarctic Circumpolar Current Front (SACCF) or the Southern Boundary of the Antarctic Circumpolar Current (SBACC), bathymetric depth and distance from the shelf edge, Optimally Interpolated Sea Surface Temperature (OISST) and latitude. Minke whale densities did not show a consistent relationship with distance from the sea ice edge over the years, suggesting variability in minke whale distribution between years. For most of the regions within the Southern Ocean, mean predicted Antarctic minke whale densities were lowest for the most recent surveys. We plan to investigate if the declining trend in Antarctic minke whale density was accompanied by changes in the sea ice environment, such as changes in total sea ice extent, melt area or quality