MEPS 295:105-111 (2005)  -  doi:10.3354/meps295105

Towards the acoustic estimation of jellyfish abundance

Andrew S. Brierley1,*, David C. Boyer2,6, Bjørn E. Axelsen3, Christopher P. Lynam1, Conrad A. J. Sparks4, Helen J. Boyer2, Mark J. Gibbons5

1Gatty Marine Laboratory, University of St. Andrews, Fife KY16 8LB, UK
2National Marine Information and Research Centre, PO Box 912, Swakopmund, Namibia
3Institute of Marine Research, PO Box 1870 Nordnes, 5817 Bergen, Norway
4Faculty of Applied Sciences, Cape Technikon, PO Box 652, Cape Town 8000, South Africa
5Zoology Department, University of Western Cape, Private Bag X 17, Bellville 7535, South Africa
6Present address: Fisheries & Environmental Research Support, Orchard Farm, Cockhill, Castle Cary, Somerset BA7 7NY, UK

ABSTRACT: Acoustic target strengths (TSs) of the 2 most common large medusae, Chrysaora hysoscella and Aequorea aequorea, in the northern Benguela (off Namibia) have previously been estimated (at 18, 38 and 120 kHz) from acoustic data collected in conjunction with trawl samples, using the ‘comparison method’. These TS values may have been biased because the method took no account of acoustic backscatter from mesozooplankton. Here we report our efforts to improve upon these estimates, and to determine TS additionally at 200 kHz, by conducting additional sampling for mesozooplankton and fish larvae, and accounting for their likely contribution to the total backscatter. Published sound scattering models were used to predict the acoustic backscatter due to the observed numerical densities of mesozooplankton and fish larvae (solving the forward problem). Mean volume backscattering due to jellyfish alone was then inferred by subtracting the model-predicted values from the observed water-column total associated with jellyfish net samples. Zooplankton-corrected echo intensity/jellyfish density data pairs were in close agreement with linear relationships determined previously from uncorrected data. Small sample sizes precluded recalculation of TS, but non-parametric pair-wise tests failed to detect any significant differences between echo intensities for jellyfish densities observed in the present study and echo intensities predicted for those densities by density–intensity relationships arising from the previous study. Previous linear density–intensity relationships had y-axis intercepts greater than zero. On the assumption that the positive intercepts were due to backscatter from unsampled mesozooplankton, new TS relationships were calculated from downward-adjusted density–intensity relationships. New values agreed closely with TS estimates determined elsewhere using single-target echo detection techniques. Given that estimates of jellyfish TS appear robust, it should now be feasible to identify jellyfish acoustically at sea and to assess their abundance, even in the presence of mixed mesozooplankton assemblages.


KEY WORDS: Aequorea aequorea · Chrysaora hysoscella · Forward problem · Jellyfish · Multi-frequency · Target strength · Zooplankton


Full text in pdf format