Circumpolar patterns in Antarctic krill larval recruitment: an environmentally-driven model

Larval recruitment in Antarctic krill is known to be episodic and regional. We consider the importance of a range of recruitment factors using an environmentally driven model of larval development from spawning to post-larvae. Our model examines the timing of spawning, interaction with bathymetry, s...

Full description

Bibliographic Details
Published in:Marine Ecology Progress Series
Main Authors: Thorpe, Sally E., Tarling, Geraint A., Murphy, Eugene J.
Format: Article in Journal/Newspaper
Language:English
Published: Inter-Research 2019
Subjects:
Online Access:http://nora.nerc.ac.uk/id/eprint/521045/
https://nora.nerc.ac.uk/id/eprint/521045/1/Thorpe.pdf
https://www.int-res.com/articles/meps_oa/m613p077.pdf
Description
Summary:Larval recruitment in Antarctic krill is known to be episodic and regional. We consider the importance of a range of recruitment factors using an environmentally driven model of larval development from spawning to post-larvae. Our model examines the timing of spawning, interaction with bathymetry, susceptibility to cold temperatures, temperature-driven development and the seasonal cycle of sea ice to identify those factors with the greatest impact. The model predicts that the seasonal location of sea ice is the main limiting factor for successful larval recruitment. Spawning in January leads to the greatest area of viable larval recruitment habitat. Dense sea ice cover, which we assume that adult krill do not spawn under, prevents spawning in large areas early in the breeding season (December). Nevertheless, later spawning in February, when sea ice is at a minimum, means there is often insufficient time for the larvae to reach a viable developmental stage before the sea ice advances. Meanwhile, although spawning is possible in more northerly areas throughout the breeding season, these are generally remote from winter sea ice, which is assumed to be necessary for larvae to overwinter. Interaction with bathymetry before hatching further limits suitable habitat. Over a 12 yr period, the model predicted larval re - cruitment from January spawning in all years in the Cooperation, Ross and Weddell Seas, with episodic larval recruitment in the Bransfield Strait in 9 of the 12 years. Additional understanding of the overwintering requirements of larvae, together with regional studies at higher spatial resolution, particularly in shelf regions, will better constrain the uncertainties in the model.