Analysis of backscatter properties and application of classification procedures for the identification of small pelagic fish species in the Central Mediterranean

Abstract The pelagic realm of the Central Mediterranean Sea is populated by four main species of fish: sardine, anchovy, horse mackerel and a mix of other pelagic fish species. In this study we employed a multifrequency acoustics approach to detect and classify fish schools of these groups. Monospecific trawl catches were selected from eight acoustic surveys and examined in relation to the coincident acoustic data. The backscattering properties of the three main species were determined using the decibel difference (Sv 120  − Sv 38 ) and the frequency response (NASC 120 /NASC 38 ). The results indicate that schools of these species cannot be distinguished on the basis of energetic properties alone, because they are very similar in physiology and scattering is dominated by the swimbladder, which is similar in shape and size. However, the use of classification models (classification tree, random forest), using energetic features, as well as bathymetric and morphometric parameters, allowed for some discrimination among the groups. According to the classification tree, school depth was found to play an important role in the identification of these fish groups, especially for anchovy and horse mackerel, for which the contribution to the overall performance of the tree was about 20%. The tree models, with only energetic or morphometric parameters, were able to classify sardine schools reasonably well, but not so well for anchovy and horse mackerel. Using a random forest method, which accounted for the variability in the learning sample, an accuracy of 85% in the overall classification rate was reached with a greater power of discrimination for sardine and anchovy schools.

[1]  G. Tutz,et al.  An introduction to recursive partitioning: rationale, application, and characteristics of classification and regression trees, bagging, and random forests. , 2009, Psychological methods.

[2]  Interannual fluctuations in acoustic biomass estimates and in landings of small pelagic fish populations in relation to hydrology in the Strait of Sicily , 2004 .

[3]  R Core Team,et al.  R: A language and environment for statistical computing. , 2014 .

[4]  John K. Horne,et al.  Acoustic approaches to remote species identification: a review , 2000 .

[5]  Jacques Masse,et al.  Acoustic observations in the Bay of Biscay: Schooling, vertical distribution, species assemblages and behaviour , 1996 .

[6]  A. Prasad,et al.  Newer Classification and Regression Tree Techniques: Bagging and Random Forests for Ecological Prediction , 2006, Ecosystems.

[7]  Ole Arve Misund,et al.  Underwater acoustics in marine fisheries and fisheries research , 1997, Reviews in Fish Biology and Fisheries.

[8]  H. Svedäng,et al.  Life under pressure: insights from electronic data-storage tags into cod swimbladder function , 2007 .

[9]  Paul G. Fernandes,et al.  A consistent approach to definitions and symbols in fisheries acoustics , 2002 .

[10]  Andy Liaw,et al.  Classification and Regression by randomForest , 2007 .

[11]  D. R. Cutler,et al.  Utah State University From the SelectedWorks of , 2017 .

[12]  C. Lingen,et al.  Laboratory investigations on the effect of prey size and concentration on the feeding behaviour of Sardina pilchardus , 2007 .

[13]  Geir Odd Johansen,et al.  Acoustic species identification of schooling fish , 2009 .

[14]  Leo Breiman,et al.  Bagging Predictors , 1996, Machine Learning.

[15]  K. Foote.  Calibration of acoustic instruments for fish density estimation : a practical guide , 1987 .

[16]  Jacques Masse,et al.  Acoustic detection of the spatial and temporal distribution of fish shoals in the Bay of Biscay , 1993 .

[17]  P. Wiebe,et al.  Sound scattering by several zooplankton groups. II. Scattering models. , 1998, The Journal of the Acoustical Society of America.

[18]  E. Ona,et al.  Synthetic echograms generated from the relative frequency response , 2003 .

[19]  E. Ona,et al.  An operational system for processing and visualizing multi-frequency acoustic data , 2002 .

[20]  R. Nero,et al.  In situ acoustic estimates of the swimbladder volume of Atlantic herring (Clupea harengus) , 2002 .

[21]  M. Barangé,et al.  Trends in the abundance and distribution of anchovy and sardine on the South African continental shelf in the 1990s, deduced from acoustic surveys , 1999 .

[22]  S. Brandt,et al.  Diet of Engraulis encrasicolus in the northern Adriatic Sea (Mediterranean): ontogenetic changes and feeding selectivity , 2009 .

[23]  Noël Diner,et al.  Correction on school geometry and density: approach based on acoustic image simulation , 2001 .

[24]  Paul G. Fernandes,et al.  Classification trees for species identification of fish-school echotraces , 2009 .

[25]  E. Murphy,et al.  Interpretation of acoustic data at two frequencies to discriminate between Antarctic krill (Euphausia superba Dana) and other scatterers , 1993 .

[26]  L. Breiman.  Arcing classifier (with discussion and a rejoinder by the author) , 1998 .

[27]  A. Pallaoro,et al.  Feeding habits of horse mackerel, Trachurus trachurus (Linneaus, 1758), from the central Adriatic Sea , 2005 .

[28]  J. Magnuson,et al.  Characterization of Patches Along Transects Using High-Resolution 70-kHz Integrated Acoustic Data , 1989 .

[29]  P. Wiebe,et al.  Sound scattering by several zooplankton groups. I. Experimental determination of dominant scattering mechanisms. , 1998, The Journal of the Acoustical Society of America.

[30]  Noël Diner,et al.  MOVIES-B: an acoustic detection description software. Application to shoal species' classification , 1993 .

[31]  Paul G. Fernandes,et al.  Multifrequency backscattering properties of Atlantic herring (Clupea harengus) and Norway pout (Trisopterus esmarkii) , 2007 .

[32]  I. Aoki,et al.  Acoustic identification of isada krill, Euphausia pacifica Hansen, off the Sanriku coast, north‐eastern Japan , 1998 .

[33]  Ariel G. Cabreira,et al.  Artificial neural networks for fish-species identification , 2009 .

[34]  D. Eggers.  Theoretical Effect of Schooling by Planktivorous Fish Predators on Rate of Prey Consumption , 1976 .

[35]  C. V. Lingen.  Effect of particle size and concentration on the feeding behaviour of adult pilchard Sardinops sagax , 1994 .

[36]  A. Cuttitta,et al.  Spawning site selection by European anchovy (Engraulis encrasicolus) in relation to oceanographic conditions in the Strait of Sicily , 2013 .

[37]  M. Iglesias,et al.  Spatio-temporal patterns and morphological characterisation of multispecies pelagic fish schools in the North-Western Mediterranean Sea , 2003 .

[38]  Wei-Yin Loh,et al.  Classification and regression trees , 2011, WIREs Data Mining Knowl. Discov..

[39]  M. Barangé,et al.  Strategies of space occupation by anchovy and sardine in the southern Benguela: the role of stock size and intra-species competition , 2005 .

[40]  E. John Simmonds,et al.  Fisheries Acoustics: Theory and Practice , 2005 .

[41]  G. Champalbert,et al.  Diet, feeding behaviour and trophic activity of the anchovy (Engraulis encrasicolus L.) in the Gulf of Lions (Mediterranean Sea). , 2000 .

[42]  Manuel Barange,et al.  Acoustic identification, classification and structure of biological patchiness on the edge of the Agulhas Bank and its relation to frontal features , 1994 .

[43]  A. Pallaoro,et al.  Diet composition and feeding intensity of horse mackerel, Trachurus trachurus (Osteichthyes: Carangidae) in the eastern Adriatic , 2004 .

[44]  H. Cabral,et al.  The diet of blue whiting, hake, horse mackerel and mackerel off Portugal , 2002 .

[45]  R. E. Thorne,et al.  Ground truth and target identification for fisheries acoustics , 2000 .

[46]  M. Barangé,et al.  Species identification of pelagic fish schools on the South African continental shelf using acoustic descriptors and ancillary information , 2001 .

[47]  Robert Tibshirani,et al.  The Elements of Statistical Learning: Data Mining, Inference, and Prediction, 2nd Edition , 2001, Springer Series in Statistics.