Acoustic classification of coexisting taxa in a coastal ecosystem

作者：Sato Mei^*; Horne John K; Parker Stetter Sandra L; Keister Julie E

来源：Fisheries Research, 2015, 172: 130-136.

DOI：10.1016/j.fishres.2015.06.019

• Acoustics
• Classification
• Pacific Herring
• Pacific Hake
• Zooplankton

Classifying coexisting taxa in a coastal ecosystem remains an analytic challenge due to the difficulty in verifying species compositions within backscatter data. Multifrequency measurements (38, 70, 120, 200 kHz) were combined with midwater trawls and zooplankton MultiNet tows in Hood Canal, WA, to classify backscatter dominated by single fish species (Pacific Herring, Pacific Hake) or major zooplankton taxa (euphausiids, copepods). Backscatter was categorized into aggregations, single targets, and layers based on morphology. Aggregations and single targets were identified in raw volume backscattering strength (S-v), while layers were classified using differences in mean volume backscattering strength (Delta MVBSi-j = MVBSi-MVBSj, where i and j denote frequency in kHz). Based on a subset of trawl-validated in situ acoustic measurements, backscatter with -16 dB < Delta MVBS200-38 <= 2 dB were classified as fish, and 2 dB < Delta MVBS200-38 < 30 dB as zooplankton. Backscatter identified as fish were further classified to hake when Delta MVBS120-38 < -4.8 dB, and herring when Delta MVBS120-38 >= -4.8 dB. The classification method was evaluated using a second set of trawl-validated acoustic data, resulting in classification accuracy of fish or zooplankton ranging from 95% to 100%. At the species level, misclassifications of herring and hake were both similar to 13%. Removal of aggregations and single targets before calculating Delta MVBS values minimized the possibility of mixed species backscatter within layers. This classification technique provides an approach to separate coexisting aggregations of dominant taxa which are common in mid- and low-latitude coastal ecosystems.

• 出版日期2015-12