Behavioural models of penguins and krill in the Southern Ocean

Cresswell, Katherine A. (2006). Behavioural models of penguins and krill in the Southern Ocean. PhD thesis The Open University.



The thesis applies state of the art ecological modelling methods to predict optimal behavioural patterns in two key Southern Ocean species, macaroni penguins, Eudyptes chrysolophus, and Antarctic krill, Euphausia superba. The work is divided into three main parts. The first considers how female macaroni penguins allocate their time between searching, foraging and feeding their chick during the guard stage. The model is forced by the availability of their main prey, Antarctic krill, and the increasing demands of the chick. The second part focuses on the behaviour of Antarctic krill at South Georgia, a hotspot in the Southern Ocean. This model predicts the most likely distribution of krill between 3 main environments that relate to bathymetry, resulting from an individual's choice of depth, density of swarm and swimming behaviour, which has some influence on their advection. The model is forced by the availability of phytoplankton and by the prevailing advection regime. The third part takes a similar approach, determining the optimal behaviour of krill as they are transported from the Antarctic Peninsula to South Georgia, except in this model, krill do not have any influence on their horizontal location. The first main finding of the thesis is that macaroni penguins will travel further to obtain a more reliable meal of krill, even if the krill reward does not change with distance from nest. A second important finding is that krill are able to increase their overall concentration in favourable areas simply by altering their swimming speed and tum rate. The third major finding describes the likely existence of a threshold availability of krill to penguins, below which the chick dies and above which there is no change to chick growth. The thesis discusses the suitability of the modelling technique and proposes future fieldwork for better model parameterisation and validation. The models provide a framework for predicting the responses of these organisms to future changes in their environment.

Viewing alternatives

Download history


Public Attention

Altmetrics from Altmetric

Number of Citations

Citations from Dimensions

Item Actions