The acoustic multifrequency classification of two sympatric euphausiid species (Meganyctiphanes norvegica and Thysanoessa raschii), with empirical and SDWBA model validation

Abstract

McQuinn, I. H., Dion, M., and St. Pierre, J.-F. 2013. The acoustic multifrequency classification of two sympatric euphausiid species (Meganyctiphanes norvegica and Thysanoessa raschii), with empirical and SDWBA model validation. – ICES Journal of Marine Science, 70: 636–649. The ecosystem approach to fishery management requires monitoring capabilities at all trophic levels, including pelagic organisms. However, the usefulness of active acoustics for ecosystem monitoring has been limited by ambiguities in the identification of scattering layers. Increasingly, multifrequency acoustic methods are being developed for the classification of scattering layers into species or species groups. We describe a method for distinguishing between sympatric northern and Arctic krill (Meganyctiphanes norvegica and Thysanoessa raschii) using sv amplitude ratios from 38, 120, and 200 kHz data which were pre-processed through a self-noise removal algorithm. Acoustic frequency responses of both euphausiid species were predicted from species-specific parameterizations of a SDWBA physical model using specific body forms (shape, volume, and length) for Arctic and northern krill. Classification and model validation were achieved using macrozooplankton samples collected from multiple-sampler (BIONESS) and ringnet (JackNet) hauls, both equipped with a strobe light to reduce avoidance by euphausiids. SDWBA frequency responses were calculated for a range of orientations (± 45°) and compared with observed frequency responses, solving for orientation by least squares. A tilt angle distribution of N[9°,4°] and N[12°,6°] for T. raschii and M. norvegica, respectively resulted in best fits. The models also provided species-specific TS–length relationships.

Spatial and temporal variations in the abundance, distribution, and aggregation of krill (Thysanoessa raschii and Meganyctiphanes norvegica) in the lower estuary and Gulf of St. Lawrence

“…Acoustic S v data was pre-processed as described by McQuinn et al (2013) by a series of automated and manual procedures. System noise and surface reverberation were removed automatically.…”

“…below the photic zone, most were well within the range of the acoustic beam (>4.5-220 m) and therefore not in the uppersurface layer above the transducers. Also, it was found that the two dominant krill species had little vertical overlap in the daytime (McQuinn et al, 2013), thereby minimizing mixed scattering layers. Finally, we avoided potential changes of krill target strength associated with varying orientation between day and night.…”

“…Other euphausiid species, mainly T. inermis and T. longicaudata are also found in the GSL and although they can be locally abundant, historically they have not been major contributors to the total abundance in the SLE and NWG (Berkes, 1976;Harvey and Devine, 2009). McQuinn et al (2013) showed that the two principle krill species had distinctive frequency responses at 38, 120 and 200 kHz, due in large part to their non-overlapping length distribution as adults, which allowed for effective discrimination of krill scattering layers. Similar to the dB-difference method, the classification technique uses the difference in pairs of S v binned values from 38, 120 and 200 kHz data:…”

“…However, the combination of the two methods takes advantage of the high horizontal and vertical resolution of the acoustics with the detailed biological information of the net sampler. Further, advances in the use of multifrequency acoustic techniques for the accurate determination of species-specific euphausiid target strength through modeling (Demer and Conti, 2004a, 2004b and the classification of krill species in the north Atlantic (McQuinn et al, 2013) and elsewhere (Lawson et al, 2008) represent significant progress toward describing krill distribution and augmenting our understanding of krill behavior and species interactions.…”

Diel variation in the vertical distribution of deep-water scattering layers in the Gulf of Mexico

“…The acoustic backscattering spectral characteristics are a function of an organism's physical properties, orientation, morphology, size, and behavior Martin et al, 1996;Stanton and Chu, 2000). Therefore, organisms with different biophysical characteristics will often produce distinctive frequency-dependent responses, facilitating acoustically-based taxonomic differentiation (McQuinn et al, 2013).…”