A comparison of net and acoustic estimates of krill density in the Scotia Sea during the CCAMLR 2000 Survey

A multi-ship, multi-national survey to assess the acoustic biomass of Antarctic krill across the Scotia Sea was undertaken in January and February 2000. In addition, a total of 135 Rectangular Midwater Trawls (RMT8) were undertaken to determine the structure of the krill population and to validate acoustic target detection techniques. This paper reports a comparison of the density estimates derived from net and acoustic sampling for a range of spatial scales; from individual net hauls to regional estimates of krill biomass. The different sources of error and the different characteristics of the density estimates from net and acoustic techniques are also defined. Direct quantitative comparisons of net and acoustic densities are shown to be inappropriate at both the small scale (i.e. individual net tows; the typical net sampling unit for census surveys) and the large scale (i.e. regional surveys). Therefore, a direct comparison of density estimates from net and acoustic surveys is not practicable for retrospective analysis of krill abundance. However, the results of net and acoustic surveys do appear comparable in terms of trends in krill distribution at the large scale. Therefore, the combined use of net and acoustic data can be useful in the analysis of interannual trends in the variability of krill distribution at the regional level. At the local level, data from trawl surveys using comparable nets can be used to examine interannual variability in krill distribution as there has been little change in the methodology used for net surveys over the last 25 years.