A statistical assessment of the density of Antarctic krill based on “chaotic” acoustic data collected by a commercial fishing vessel

• Published in: Frontiers in Marine Science
• Main Authors: Yunxia Zhao, Xinliang Wang, Xianyong Zhao, Yiping Ying
• Format: Article in Journal/Newspaper
• Language: English
• Published: Frontiers Media S.A. 2022
• Subjects: Antarctic krill; acoustic data; Regional Gridding; Extended delta-distribution model; fishing vessel; Science; Q; General. Including nature conservation; geographical distribution; QH1-199.5; Antarctic; The Antarctic; Bransfield Strait; Fishing Ground; Antarc*; Antarctic Krill
• Online Access: https://doi.org/10.3389/fmars.2022.934504; https://doaj.org/article/5d1b9a77795449b69fa1c87051a2b646

With the development of acoustic data processing technology, it is possible to make full use of the “chaotic” acoustic data obtained by fishing vessels. The purpose of this study is to explore a feasible statistical approach to assess the Antarctic krill density rationally and scientifically based on the acoustic data collected during routine fishing operations. The acoustic data used in this work were collected from the surveys conducted by the Chinese krill fishing vessel F/V Fu Rong Hai since the 2015/16 fishing season in the Bransfield Strait. We first processed acoustic data into small units of 0.1 nm, then selected the location of the central fishing ground for grid processing. Because of many zero and low values, we established a Regional Gridding and Extended Delta-distribution (RGED) model to evaluate the acoustic density of the krill. We defined the selection coefficient of grid size by using the coefficient of variation (CV) of the mean density and the weight of the effective covered area of the grids. Through the comparison of selection indexes, cells of 5′S × 10′W were selected as a computational grid and applied to the hotspot in the Bransfield Strait. Acoustic data reveal the distribution of krill density to be spatially heterogeneous. The CV of the mean density for 4 months converges at ~15% for cells of 5′S × 10′W. Simulations estimate krill resource densities in February to be ~1990 m2 nm−2 and to increase to ~8760 m2 nm−2 in May (4.4 times higher). We deem the RGED model to be useful to explore dynamic changes in krill resources in the hotspot. It is not only of great significance for guiding krill fishery, but it also provides krill density data for studying the formation mechanism of the resource hotspots.