No evidence for a decline in the density of Antarctic krill Euphausia superba Dana, 1850, in the Southwest Atlantic sector between 1976 and 2016

Martin James Cox,Nicholas Gales,Steven Candy,William K De La Mare,So Kawaguchi,Stephen Nicol

doi:10.1093/jcbiol/ruy072

Martin James Cox, Nicholas Gales + Show 4 more

Open Access

https://doi.org/10.1093/jcbiol/ruy072

Copy DOI

Abstract

The Antarctic marine environment is changing, and changes in the Southwest Atlantic sector have included decreases in sea ice and increases in water temperature. Associated with these changes is a reported 38% and 81% per decade decline in the numerical density (hereafter density) of Antarctic krill Euphausia superba Dana, 1850 , between 1976 and 2003. Few changes in other components of the ecosystem that could be attributed to such a change, such as a mass decline in krill-dependent predators, have been detected. In an ecosystem so dependent on this keystone species, a massive population decline in krill ought to have had an obvious effect. In the absence of such an effect, it is timely to revisit the issue of the purported decline in krill density. The original analysis that indicated a decline in krill density was based on the 2004 version of KRILLBASE, a database of net samples. We analysed the publicly available and updated version (version 1, accessed 30 November 2017) and our analyses did not suggest a significant decline in krill density. Rather, after accounting for sampling heterogeneity and habitat variables, average krill density appears to have been stable but with considerable inter-annual variability. Since our results were unable to find any evidence for a decline in krill density we recommend a re-appraisal of many of the paradigms that underlie much of the recent thinking about ecosystem change Antarctic waters. Such a revision is necessary to provide a firmer foundation for predictions of the effects of climate change and resource extraction on the Southern Ocean ecosystem.

Highlights

A long-term sustained decline in krill would have management consequences
We investigated variability in krill density over time using the data contained in the database originally used in Atkinson et al, (2004), KRILLBASE, which has been recently updated (Atkinson et al, 2017)
We isolated a subset of 7,670 KRILLBASE records that covered the Southwest Atlantic sector (–70o to –30o and latitude range –69o to –51o), and the ‘modern era’, from 1976 to 2016, in order to be consistent with the data used in Atkinson et al (2004)

Summary

Introduction

A long-term sustained decline in krill would have management consequences. The krill fishery in the Southwest Atlantic sector is managed using a single biomass estimate derived from an acoustic survey conducted in 2000 (Hewitt et al, 2004). We explored the inter-annual variability in krill density using data from individual records, rather than by cell means, using a two-component statistical hurdle model. Using a hurdle model avoided the need to add an arbitrary constant to krill densities before log transforming highly skewed net density data. Both modelling stages used mixed effects models that include both fixed and random (population level) effects. Models were fitted using R (version 3.4.3; R Core Team, 2017) and the gamm (version 1.8.23; Wood, 2006) and lme (version 0.2.5; Bates et al, 2015) packages

Methods

Results

Conclusion