Measurement of the volume-backscattering spectrum from an aggregation of Antarctic krill and inference of their length-frequency distribution Kazuo Amakasu a, * , Tohru Mukai b , Masato Moteki a, c a Tokyo University of Marine Science and Technology, 4-5-7 Konan, Minato-ku, Tokyo 108-8477, Japan b Faculty of Fisheries Sciences, Hokkaido University, 3-1-1 Minato, Hakodate, Hokkaido 041-8611, Japan c National Institute of Polar Research,10-3 Midori-cho, Tachikawa, Tokyo 190-8518, Japan article info Article history: Received 29 November 2016 Received in revised form 22 February 2017 Accepted 24 February 2017 Available online xxx Keywords: Least-squares inversion Prolate spheroid Signal-to-noise ratio Target strength Theoretical scattering model abstract Antarctic krill, Euphausia superba, were observed using a broadband echosounder and their length- density distribution was inferred from the echo data through a least square inversion. The observation was conducted in the Indian Ocean sector of the Southern Ocean in January 2014. The transmit signal was a 10-ms linear frequency-modulated signal with a frequency sweep of 20e200 kHz. A large aggregation of Antarctic krill was observed at a sampling station over the continental shelf. The volume- backscattering strengths in the frequency range of 85e187 kHz were determined from the observed echoes. In addition, the signal-to-noise ratio at each frequency was estimated and the measured volume- backscattering strengths were evaluated before inversion for accurate inferences. In this inversion, Antarctic krill were modeled as prolate spheroids and the target strengths were predicted by the distorted-wave Born approximation. The acoustically inferred mean length was in good agreement with the mode length determined by a net sampling performed immediately after the echo sampling. Also, the acoustically inferred length-frequency distribution was in reasonable agreement with the determined one from the net samples. © 2017 Elsevier B.V. and NIPR. All rights reserved. 1. Introduction Antarctic krill, Euphausia superba, is important to the Southern Ocean ecosystem (Everson, 2000) and a valuable fisheries resource. Krill are food for a number of species such as whales, seals, fishes, and penguins and other seabirds. The annual catch was 293,815 tonnes in the Austral summer of 2013/2014 and most of this was caught in the Scotia Sea near the Antarctic Peninsula (CCAMLR, 2015). To obtain information needed to protect the ecosystem and manage the krill fishery, acoustic and net surveys are con- ducted there periodically (Hewitt et al., 2004; Reiss et al., 2008; Fielding et al., 2014). Although there is less krill fishing in the In- dian Ocean sector of the Southern Ocean, acoustic and net surveys are also conducted there to understand better its critical role in that ecosystem (Pauly et al., 2000; Nicol et al., 2008; Jarvis et al., 2010; Amakasu et al., 2011). Although acoustic sampling generally requires less cost, time, and effort per unit survey area, and echo data typically have higher temporal and spatial resolutions than net samples, the interpreta- tion of the echo data requires biological information (e.g. size, numerical density, and species) characterizing the echo. However, net sampling for obtaining biological information is performed at discrete points and its temporal and spatial resolutions are very low. Although acoustic and net sampling from research vessels can be used to survey large areas, this approach is not suitable for ob- servations of krill under winter sea ice, or continuous monitoring of krill at one or more locations. In these and other situations, krill may be sampled using echosounders deployed on platforms such as autonomous underwater vehicles (Brierley et al., 2002), gliders (Guihen et al., 2014), drifters, and moorings (Brierley et al., 2006). However, these platforms have a considerable limitation that bio- logical information required for the interpretation of echo data is not available during acoustic sampling. Because net sampling is impossible on these platforms. As alternative tools, optical cameras useful for species identification and sizing are appropriate for the * Corresponding author. E-mail address: amakasu@kaiyodai.ac.jp (K. Amakasu). Contents lists available at ScienceDirect Polar Science journal homepage: https://www.evise.com/profile/#/JRNL_POLAR/login http://dx.doi.org/10.1016/j.polar.2017.02.007 1873-9652/© 2017 Elsevier B.V. and NIPR. All rights reserved. Polar Science xxx (2017) 1e9 Please cite this article in press as: Amakasu, K., et al., Measurement of the volume-backscattering spectrum from an aggregation of Antarctic krill and inference of theirlength-frequency distribution, Polar Science (2017), http://dx.doi.org/10.1016/j.polar.2017.02.007