Improving near real-time acoustic monitoring of two whale species that make similar calls

Abstract

Passive acoustic monitoring can be used to detect and classify whales. In the Gulf of St. Lawrence, endangered blue (Balaenoptera musculus) and sei (Balaenoptera borealis) whales are acoustically monitored in near real-time; however, their similar vocalizations can hinder species discrimination. To determine whether blue whale misclassification could be reduced in a region with irregular sei whale presence, a new blue whale call library for automated detection and classification and revised detection protocols were implemented during the 2025 monitoring season. Comparing detections from 2025 to previous monitoring seasons (2022-2024), before both advancements, showed that days with sei whale acoustic presence decreased, and days with blue whale acoustic presence increased. Modelling analyses revealed blue whale acoustic detections can vary depending on season and location. Overall, misclassifications were reduced, and blue whale call discrimination was improved, which is important as detections are used to monitor these species and must accurately reflect their presence.