CCAMLR

Having accurate surface area estimates of toothfish (Dissostichus spp) habitats is important for CCAMLR in estimating the fishable biomass of these species in data-poor areas, developing scientific survey designs and models of habitat use.  Planimetric seabed areas (i.e. a two-dimensional surface) within the Convention area are commonly calculated, but surface area (i.e. a three-dimensional surface) is rarely estimated. The aims of this paper are to examine differences in: (1) planimetric seabed area based on different datasets (GEBCO 20014 and GEBCO 2008) and; (2) planimetric and surface area estimates for the same areas using the most up-to-date global bathymetry dataset (GEBCO 2014). Comparisons were performed at four different scales. At the coarsest scale areas were calculated within research blocks (all depths included), then estimates were limited to the fishable depth range of each research block (600-1800m included), the region were scaled down further to look at differences in eight depth classes and at the finest resolution areas were compared in individual (~500 m x 500m) grid cells within research blocks. Differences between the datasets varied between 0-62% depending on the research block and we considered the GEBCO 2014 dataset to be the most accurate so we used this in comparing different area metrics.  Results from the comparison of total surface and planimetric areas across all depths and the fishable depth range within research blocks showed differences of less than 2%. However, at the smaller grid cell scale, larger differences, of up to 137%, between the area metrics emerged. Given differences between surface and planimetric areas within fishable depth ranges at the scale of a research block were no greater than 2% this is not sufficiently different to impact toothfish biomass estimates based on the current CPUE-analogy calculations. However, in many of the research blocks a higher proportion of the grid cells within the fishable depth range (i.e. -600 - -800, -800 - -1000 etc.) had a greater surface to planimetric area difference than those cells that were outside of the fishable depth ranges. In those research blocks that showed larger difference at the grid cell level, it may be worth examining whether fishing occurred more frequently in cells with a large differences and if so whether deriving a more specific measure of surface area in fished locations (as opposed to the fishable depth range) impacts estimates of biomass. Spatial habitat modeling of toothfish species’ could also benefit from the use of surface area measurement at the grid cell level.

Single frequency (38 and 120 kHz) acoustic and associated net data were extracted from US AMLR datasets and used to mimic data that might be collected by the commercial fishery during the fishing seasons. From these data we develop semi-empirical indices of acoustic biomass in a Generalized Linear Modeling framework. We correlate these estimates of biomass with acoustic estimates derived from the 3-frequency method using the CCAMLR protocol for the period between 1996 and 2011 in both the Elephant Island and West Shelf Areas of the US AMLR survey grid. Cross-validation of acoustic estimates between these two areas and the 3-frequency biomass showed that models developed using a wide length frequency distribution (Elephant Island) could be used to estimate biomass from other areas where the length frequency of animals is skewed towards larger animals. We show that it is possible to develop semi-empirical models of krill biomass, at 120 kHz frequency that can be used to augment research acoustic surveys if proper survey design and calibration of transducers is maintained and if time series are sufficiently long to average out differences among years.

The use of fishing-vessel-based acoustic data has been recognized by SC-CAMLR as an important way to estimate the distribution and relative abundance of Antarctic krill (Euphausia superba), yet the quality and even the utility of the data may be seriously degraded by interferences due to the lack of synchronization device for the acoustic instruments equipped on some of the vessels. A simple algorithm to remove noise and significant interference from other acoustic instruments was introduced. The algorithm was built on relevant modules in the Echoview acoustic data post-processing software. The utility of the method was demonstrated by comparing the appearances of krill swarm echograms, the dB differences and the echo integrations. Results showed that the interference were effectively removed while the retained krill swarms were maintained in good geometrical shape and volume backscatter strength.
 

Increasing volumes of acoustic data are being collected which necessitates a reappraisal of many of the current methods for data processing and analysis.  For some acoustic data sets manual processing is no longer an option; the large size of the data set is preventing analysis.  This document provides an overview of an automated procedure for processing acoustic data that is illustrated using Echoview, EK60 data and the Echoview^R package available in the statistical language R.  Whilst the scripting of Echoview through EchoviewR has been successful, there are several challenges remaining before the acoustic processing can be truly automated, such as robust seabed and false bottom detection.