Application of Machine Learning and Acoustic Predation Tags to Classify Migration Fate of Atlantic Salmon Smolts

Abstract

Mortality and predation of tagged fishes presents 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 pH sensor data, while clustering reduced estimates by 3.5-30%. The greatest changes in estimates were seen in years with large class imbalance 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 predator-prone 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.

Predation research with electronic tagging

“…Specific interest in resolving the fate of animals in aquatic ecosystems has advanced on mortality sensors in radio tags and yielded a predation sensor acoustic tag, a device with an integrated sensor that transmits the tagged animal's fate to receivers (Halfyard et al 2017, Lennox et al 2021a). The difficulties of validating data from such sensors will, however, require ongoing refinement that requires collaboration between technology developers and scientists (Notte et al 2022). Specialized sensors for aquatic species offer promise as a tool that could be developed and integrated into other tag types used on terrestrial animals, which has the potential to support further advancement of predation research using telemetry.…”

“…These analytical tools have potential to be increasingly refined and applied to the detection of predation for free‐ranging animals based on state changes in behaviour. Manual methods have been applied successfully by Gibson et al (2015) and Daniels et al (2019), including refinements by Notte et al (2022). Principles for identifying predation from movement can be extended from the accelerometer‐based state machines integrated into acoustic tags (Lennox et al 2021a) that identify sudden and persistent changes in behavioural states indicative of a state change in the carrier of the tag.…”