Case Study: Determining in situ target strength and biomass of Atlantic mackerel

Location

Shetland Islands, Scotland

Background

The abundance and distribution of many aquatic organisms can be effectively quantified using acoustic surveying methods, and are particularly useful for the assessment of pelagic fish. In order to convert acoustic backscatter into estimates of abundance, it is essential to know the target strength (TS) for the species and size of interest.

In situ TS estimates of Atlantic mackerel (Scomber scombrus) were obtained using a multi-frequency split-beam echosounder and estimates of abundance from a dedicated mackerel acoustic survey around the Shetland Islands, Scotland, in 2014 were made.

A series of arithmetic and logical variables were used to construct virtual echograms using Echoview. Firstly, suitable sample regions (defined as a volume which contains an aggregation with a surrounding margin of open water) were created. Next high density regions (i.e. schools) were identified and removed using a multifrequency thresholding filtering algorithm (adapted from Fernandes, 2009, see Figure 1) in order to minimize multiple-target errors. A single-target detection algorithm (Soule et al., 1997) in Echoview (spilt-beam method 2) was then applied to isolate in situ TS at each frequency within the selected sample regions. Furthermore, the alpha-beta tracking algorithm (Blackman, 1986), implemented by Echoview, was applied to all accepted single-target detections. All single-targets and fish tracks were exported to a .csv format and post-processed outside of Echoview.

To determine biomass of mackerel, schools were firstly detected using the multifrequency detection algorithm described by Fernandes (2009). Alternatively, the SHAPES algorithm in Echoview (Shoal Analysis and Patch Estimation System, Barange, 1994), whereby schools or shoals are selected based on contiguous features between data points determined from the acoustic data works well. Target classification was then based on decibel difference (ΔdB) between the echo strengths recorded simultaneously at several frequencies (Figure 2). Schools classified as mackerel where integrated and then subsequently exported and analyzed outside of Echoview.

Echoview atlantic mackerel biomass survey target strength

Figure 1. Multifrequency, single-target, filter algorithm. The top four panels illustrate Sv echograms at 18, 38, 120 and 200 kHz (-70 dB display threshold). (1) The middle left panel shows the summation of Sv samples across the four frequencies with a threshold value (minimum and maximum in-range values of -999 and -270 dB, respectively). (2) The middle right panel shows a 200 kHz masked echogram with the school removed. (3) Regions of high density identified according to methods described in Sawada et al (1993). (4) Bottom panels show single target echograms. The horizontal black lines indicate the range interval (20 to 60 m). Taken from Scoulding (2016).

Echoview Atlantic mackerel biomass survey species identification

Figure 2. Example of identification algorithm which calculates dB difference. a) Sv echogram at 38 and b) 200 kHz (-70 dB display threshold). c) region mean Sv at 38 kHz and d) 200 kHz. e) a dB difference virtual echogram (200 kHz – 38 kHz), mackerel are assumed to be represented by values on this echogram for which ΔdB>2; a mask is generated accordingly and applied to f) and g) which are echograms at 38 and 200 kHz, respectively, masked to reveal only schools identified as mackerel. Taken from Scoulding (2016).  

References

  • Barange, M. 1994. Acoustic identification, classification and structure of biological patchiness on the edge of the Agulhas bank and its relation to frontal features. South African Journal of Marine Science, 14: 333-347.
  • Blackman, S. S. 1986. Multiple-target tracking with radar applications, Artech House.
  • Fernandes, P. G. 2009. Classification trees for species identification of fish-school echotraces. ICES Journal of Marine Science, 66: 1073–1080.
  • Sawada, K., Furusawa, M., and Williamson, N. J. 1993. Conditions for the precise measurement of fish target strength in situ. Fisheries Science, 20: 15-21.
  • Scoulding, B. 2016. In situ target strength of pelagic fish (Unpublished doctoral thesis). University of Aberdeen, Aberdeen, Scotland.
  • Soule, M., Barange, M., Solli, H., and Hampton, I. 1997. Performance of a new phase algorithm for discriminating between single and overlapping echoes in a split-beam echosounder. ICES Journal of Marine Science, 54: 934-938.