Abstract
Mortality and predation of tagged fishes present a serious challenge to interpreting results of acoustic telemetry studies. There is a need for standardized methods to identify predated individuals and reduce the impacts of “predation bias” on results and conclusions. Here, we use emerging approaches in machine learning and acoustic tag technology to classify out-migrating Atlantic salmon (Salmo salar) smolts into different fate categories. We compared three methods of fate classification: predation tag pH sensors and detection data, unsupervised k-means clustering, and supervised random forest combined with tag pH sensor data. Random forest models increased predation estimates by 9–32% compared to relying solely on pH sensor data, while clustering reduced estimates by 3.5–30%. The greatest changes in fate class estimates were seen in years with large class imbalance (one or more fate classes underrepresented compared to the others) or low model accuracy. Both supervised and unsupervised approaches were able to classify smolt fate; however, in-sample model accuracy improved when using tag sensor data to train models, emphasizing the value of incorporating such sensors when studying small fish. Sensor data may not be sufficient to identify predation in isolation due to Type I and Type II error in predation sensor triggering. Combining sensor data with machine learning approaches should be standard practice to more accurately classify fate of tagged fish.
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs00442-022-05138-3/MediaObjects/442_2022_5138_Fig1_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs00442-022-05138-3/MediaObjects/442_2022_5138_Fig2_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs00442-022-05138-3/MediaObjects/442_2022_5138_Fig3_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs00442-022-05138-3/MediaObjects/442_2022_5138_Fig4_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs00442-022-05138-3/MediaObjects/442_2022_5138_Fig5_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs00442-022-05138-3/MediaObjects/442_2022_5138_Fig6_HTML.png)
Similar content being viewed by others
Data availability
The data that support the findings of this study will be made publicly available on the Ocean Tracking Network database following publication of the data.
Code availability
Associated R code for analyses is publicly available on GitHub: https://github.com/danielanotte
References
Beland KF, Kocik JF, VandeSande J, Sheehan TF (2001) Striped bass predation upon Atlantic salmon smolts in maine. Northeast Nat 8(3):267–274
Bendall B, Moore A (2008) Temperature-sensing telemetry – possibilities for assessing the feeding ecology of marine mammals and their potential impacts on returning salmonid populations. Fisheries Manag Ecol 15:339–345
Binder TR, Dini A (2019) glatos: An R package for the Great Lakes Acoustic Telemetry Observation System. R package version 0.5.1. https://rdrr.io/github/jsta/glatos/man/glatos.html
Bland LM, Collen B, Orme CDL, Bielby J (2014) Predicting the conservation status of data-deficient species. Conserv Biol 29(1):250–259
Bradford RG, Halfyard EA, Hayman T, LeBlanc P (2015) Overview of the 2013 Bay of Fundy striped bass biology and general status. DFO Can Sci Advis Sec Res Doc 2015/024, pp.iv + 36
Brownscombe JW, Griffin LP, Gagne TO, Haak CR, Cooke SJ, Finn JT, Danylchuk AJ (2019) Environmental drivers of habitat use by a marine fish on a heterogenous and dynamic reef flat. Mar Biol. https://doi.org/10.1007/s00227-018-3464-2
Brownscombe JW, Griffin LP, Morley D, Acosta A, Hunt J, Lowerre-Barbieri SK, Adams AJ, Danylchuk AJ, Cooke SJ (2020) Application of machine learning algorithms to identify cryptic reproductive habitats using diverse information sources. Oecologia 194:283–298
Buchanan RA, Skalski JR, Brandes PL, Fuller A (2013) Route use and survival of juvenile chinook salmon through the San Joaquin River Delta. N Am J Fish Manage 33(1):216–229
Chen C, Liaw A, Breiman L (2004) Using random forest to learn imbalanced data. Report number 666, Journal University of California, Berkeley, vol 110, pp 1–12
Clark JS, Carpenter SR, Barber M, Collins S, Dobson A, Foley JA, Lodge DM, Pascual M, Pielke R, Pizer W et al (2001) Ecological forecasts: an emerging Imperative. Science 293(5530):657–660
Cordier T, Esling P, Lejzerowicz F, Visco J, Ouadahi A, Martins C, Cedhagen T, Pawlowski J (2017) Predicting the ecological quality status of marine environments from eDNA metabarcoding data using supervised machine learning. Environ Sci Technol 51(16):9118–9126
Coreau A, Pinay G, Thompson JD, Cheptou PO, Mermet L (2009) The rise of research on futures in ecology: rebalancing scenarios and predictions. Ecol Lett 12:1277–1286
Cutler DR, Edwards TC, Beard KH, Cutler A, Hess KT, Gibson J, Lawler JJ (2007) Random forests for classification in ecology. Ecology 88(11):2783–2792
Daniels J, Chaput G, Carr J (2018) Estimating consumption rate of Atlantic salmon smolts (Salmo salar) by striped bass (Morone Saxatilis) in the Miramichi River estuary using acoustic telemetry. Can J Fish Aquat Sci 75:1811–1822
Daniels J, Sutton S, Webber D, Carr J (2019) Extent of predation bias present in migration survival and timing of Atlantic salmon smolt (Salmo salar) as suggested by a novel acoustic tag. Anim Biotelemetry. https://doi.org/10.1186/s40317-019-0178-2
DFO (2019a) Atlantic Salmon (Inner Bay of Fundy Population). http://www.dfo-mpo.gc.ca/species-especes/profiles-profils/salmon-atl-saumon-eng.html
DFO (2019b) Atlantic Salmon Marine Threats Research. https://www.bio.gc.ca/science/research-recherche/fisheries-pecheries/managed-gere/smtr-rfms-en.php
Evans AF, Hostetter NJ, Roby DD, Collis K, Lyons DE, Sandford BP, Ledgerwood RD, Sebring S (2012) Systemwide evaluation of avian predation on juvenile salmonids from the Columbia River based on recoveries of passive integrated transponder tags. T Am Fish Soc 141(4):975–989
Gibson AJF, Halfyard EA, Bradford RG, Stokesbury MJW, Redden AM (2015) Effects of predation on telemetry-based survival estimates: insights from a study on endangered Atlantic salmon smolts. Can J Fish Aquat Sci 72:728–741
Halfyard EA, Gibson AJF, Ruzzante DE, Stokesbury MJW, Whoriskey FG (2012) Estuarine survival and migratory behaviour of Atlantic salmon Salmo salar smolts. J Fish Biol 81:1626–1645
Halfyard EA, Gibson AJF, Stokesbury MJW, Ruzzante DE, Whoriskey FG (2013) Correlates of estuarine survival of Atlantic salmon postsmolts from the Southern Upland, Nova Scotia, Canada. Can J Fish Aquat Sci 70:452–460
Halfyard EA, Webber D, Del Papa J, Leadley T, Kessel ST, Colborne SF, Fisk AT (2017) Evaluation of an acoustic telemetry transmitter designed to identify predation events. Methods Ecol Evol 8:1063–1071
Hanssen EM (2020) Novel telemetry predation sensors and mechanistic models reveal the tribulations of Atlantic salmon (Salmo salar) smolts migrating through lakes. MSc thesis, Department of Biological Sciences, University of Bergen, Norway, vol 1, pp 1–59
Indira V, Vasanthakumari R, Sugumaran V (2010) Minimum sample size determination of vibration signals in machine learning approach to fault diagnosis using power analysis. Expert Syst Appl 37(12):8650–8658
Jain AK (2010) Data clustering: 50 years beyond k-means. Pattern Recogn Lett. https://doi.org/10.1016/j.patrec.2009.09.011
Kassambara A, Mundt F (2019) factoextra: Extract and Visualise the Results of Multivariate Data Analyses. R package version 1.0.6. https://CRAN.R-project.org/package=factoextra
Klinard NV, Matley JK (2020) Living until proven dead: addressing mortality in acoustic telemetry research. Rev Fish Biol Fisheries 30:485–499
Klinard NV, Matley JK, Fish AT, Johnson TB (2019) Long-term retention of acoustic telemetry transmitters in temperate predators revealed by predation tags implanted in wild prey fish. J Fish Biol. https://doi.org/10.1111/jfb.14156
Klinard NV, Matley JK, Ivanova SV, Larocque SM, At F, Johnson TB (2021) Application of machine learning to identify predators of stocked fish in Lake Ontario: using acoustic telemetry predation tags to inform management. J Fish Biol. https://doi.org/10.1111/jfb.14574
LaCroix GL (2008) Influence of origin on migration and survival of Atlantic salmon (Salmo salar) in the Bay of Fundy, Canada. Can J Fish Aquat Sci 65:2063–2079
Lennox RJ, Nilsen CI, Nash A, Hanssen EM, Johannesen HL, Berhe S, Barlaup B, Wiik VK (2021) Laboratory and field experimental validation of two different predation sensors for instrumenting acoustic transmitters in fisheries research. Fisheries 46(11):565–573
Liaw A, Wiener M (2002) randomForest: Breiman and Cutler's Random Forests for Classification and Regression. R package version 4.6–14. https://cran.r-project.org/web/packages/randomForest/index.html
Maxwell AE, Warner TA, Fang F (2018) Implementation of machine-learning classification in remote sensing an applied review. Int J Remote Sens 39(9):2784–2817
Moghaddam DD, Rahmati O, Panahi M, Tiefenbacher J, Darabi H, Haghizadeh A, Haghighi AT, Nalivan OA, Bui DT (2020) The effect of sample size on different machine learning models for groundwater potential map** mountain bedrock aquifers. CATENA 187:104421
Moxam EJ, Cowley PD, Bennett RH, von Brandis RG (2019) Movement and predation: a catch-and-release study on the acoustic tracking of bonefish in the Indian Ocean. Environ Biol Fish 102:365–381
Olden JD, Lawler JJ, Poff NL (2008) Machine learning methods without tears: a primer for ecologists. Q Rev Biol 83(2):171–193
Perry RW, Skalski JR, Brandes PL, Sandstrom PT, Klimley P, Ammann A, MacFarlane B (2010) Estimating survival and migration route probabilities of juvenile chinook salmon in the Sacramento-San Joaquin River Delta. N Am J Fish Manage 30:142–156
Pincock DG (2012) False detections: what they are and how to remove them from detection data. Amirix Document DOC-004691 Version 3, pp 1–11
Romine JG, Perry RW, Johnston SV, Fitzer CW, Pagliughi SW, Blake AR (2014) Identifying when tagged fishes have been consumed by piscivorous predators: application of multivariate mixture models to movement parameters of telemetered fishes. Anim Biotelemetry. https://doi.org/10.1186/2050-3385-2-3
Schultz AA, Kumagai KK, Bridges BB (2015) Methods to evaluate gut evacuation rates and predation using acoustic telemetry in the tracy fish collection facility primary channel. Anim Biotelemetry. https://doi.org/10.1186/s40317-015-0034-y
Simpfendorfer CA, Huveneers C, Steckenreuter A, Tattersall K, Hoenner X, Harcourt R, Heupel MR (2015) Ghosts in the data: false detections in VEMCO pulse position modulation acoustic telemetry monitoring equipment. Anim Biotelemetry 3:55. https://doi.org/10.1186/s40317-015-0094-z
Tabak MA, Norouzzadeh MS, Wolfson DW, Sweeny SJ, Vercauteren KC, Snow NP, Halseth JM, Di Salvo PA, Lewis JS, White MD et al (2018) Machine learning to classify animal species in camera trap images: applications in ecology. Methods Ecol Evol 10(4):585–590
Thessen AE (2016) Adoption of machine learning techniques in ecology and earth science. One Ecosystem. https://doi.org/10.3897/oneeco.1.e8621
Thorstad EB, Whoriskey FG, Rickardsen AH, Aarestrup K (2011) Aquatic nomads: the life and migrations of the Atlantic salmon. In: Aas O, Einum S, Klemetsen A, Skurdal J (eds) Atlantic salmon ecology Chichester, West Sussex, vol 1. Wiley-Blackwell, UK, pp 1–33
Thorstad EB, Whoriskey FG, Uglem I, Moore A, Rikardsen AH, Finstad B (2012) A critical life stage of the Atlantic salmon Salmo salar: behaviour and survival during the smolt and initial post-smolt migration. J Fish Biol 81:500–542
Weinz AA, Matley JK, Klinard NV, Fisk AT, Colborne SF (2020) Identification of predation events in wild fish using novel acoustic transmitters. Anim Biotelemetry. https://doi.org/10.1186/s40317-020-00215-x
Whittingham H, Ashenden SK (2021) Hit discovery. In: Ashenden SK (ed) The era of artificial intelligence, machine learning, and data science in the pharmaceutical industry. Elsevier, Amsterdam, Netherlands, pp 81–102
Acknowledgements
We would like to acknowledge Cindy Hawthorne, Jeff Reader, Alana Ransome, George Nau, and other members of Fisheries and Oceans Canada and the Mi’kmaw Conservation Group for assistance in the field. As well as Darren and Erica Porter for receiver deployment and retrieval in the Minas Basin. We also thank Jake Brownscombe for answering questions about machine learning model tuning and members of the Ocean Tracking Network Data Centre for data acquisition and coding help.
Funding
Funding provided by NSERC Strategic Partnership Grant No. 521256. R.J. Lennox was supported by the NFR project LaKES (#320726).
Author information
Authors and Affiliations
Contributions
DVN: performed field work, contributed to concept development, performed analyses, and wrote initial manuscript drafts. RJL: contributed to concept development and manuscript drafts. DCH: performed field work, developed study design, and contributed to manuscript drafts. GTC: contributed to manuscript drafts.
Corresponding author
Ethics declarations
Conflict of interest
The authors declare no competing interests.
Ethical approval
All animal experiments were approved by the Canadian Committee on Animal Care, via permits issued by Fisheries and Oceans Canada (Maritimes Region Animal Care Committee Animal Utilization Protocols 17–16, 18–13, 19–10) and by Dalhousie University (University Committee on Lab Animals permit 18–126).
Consent to participate
Not applicable.
Consent for publication
Not applicable.
Additional information
Communicated by Yannis Papastamatiou.
Supplementary Information
Below is the link to the electronic supplementary material.
Rights and permissions
About this article
Cite this article
Notte, D.V., Lennox, R.J., Hardie, D.C. et al. Application of machine learning and acoustic predation tags to classify migration fate of Atlantic salmon smolts. Oecologia 198, 605–618 (2022). https://doi.org/10.1007/s00442-022-05138-3
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00442-022-05138-3