Ecological Archives E084-042-A1

Mark D. Scheuerell and Daniel E. Schindler. 2003. Diel vertical migration by juvenile sockeye salmon: empirical evidence for the antipredation window. Ecology 84:1713–1720.

Appendix A. Details of the hydroacoustic methodology for sampling the spatial distributions of predators and prey.

We used a BioSonics DT4000 split-beam echosounder (BioSonics, Seattle, Washington, USA) operating at 200 kHz with an elliptical transducer consisting of -3dB beam angles of 6.7 and 13.5 degrees. We used a 0.4 ms pulse at a rate of 2 pulses/s with the lower threshold set at -70 dB. The boat speed was 10 km/h. All incoming signals were digitized at the transducer and then stored on a computer hard drive for later analyses. Only echoes from depths <80 m were analyzed to eliminate potential electrical noise problems. The size and depth of individual echoes were obtained using the BioSonics DT Analyzer 4.0 software. We classified all targets less than -46 dB as juvenile sockeye and all fish greater than -37 dB as predators. According to the equation of Burczynski and Johnson (1986) , these sound-intensity thresholds correspond to fish lengths of <100 mm and >282 mm respectively. In our case, fish densities are low (~0.01 fish/m3) and therefore classifying individual fish based on single target detections is generally not a problem (MacLennon and Simmonds 1992) .

The split-beam hydroacoustics provided us with estimates of the size and depth of individual fish, but we required a correction for the relative oversampling of deeper depths due to the cone-shaped geometry of the sound beam. Therefore, we calculated a maximum-likelihood estimate of the weighted mean fish depth for juvenile sockeye based on targets grouped into 5-min intervals for each survey. We calculated the weights as

                                           w = [ z2 tan(/2) tan(/2)]-1,

where z is the depth of an individual target in meters, and and are the hydroacoustic beam angles in degrees. Our transducer is elliptical, such that = 6.7° and = 13.5°; a transducer with a circular beam would have equal parameters. These weighted mean depths were based on a mean of 46 ± 2.4 (means ± 1 SE) individual fish per 5-min sampling interval.

Literature Cited

Burczynski, J. J., and R. L. Johnson. 1986. Application of dual-beam acoustic survey techniques to limnetic populations of juvenile sockeye salmon (Oncorhynchus nerka). Canadian Journal of Fisheries and Aquatic Sciences 43:1776–1788.

MacLennon, D. N., and E. J. Simmonds. 1992. Fisheries Acoustics. Chapman and Hall, London, UK.



[Back to E084-042]