Passive acoustic monitoring complements traditional methods for assessing marine habitat enhancement outcomes

Abstract

acoustic monitoring complements traditional methods for assessing marine habitat enhancement outcomes. Ecosphere 12(11):e03840.

Automated acoustic detection of river herring (Alewife and Blueback Herring) spawning activity

“…Many fish produce species‐specific aquatic acoustic signals, which can be passively monitored to study their behavior and ecology (reviewed by Gannon 2008, Luczkovich et al 2008; recent examples include Ibrahim et al 2018, Picciulin et al 2019, Van Hoeck et al 2021). Similar to the monitoring of aquatic signals, terrestrial acoustic monitoring of spawning splashes can be an efficient method for measuring species presence as well as spawning activity (e.g., Lebel et al 2001; Baglinière et al 2003; Chanseau et al 2004).…”

Introduction: Honoring the contribution of Charles H. Peterson (1946–2020) to the field of marine ecology

“…Pete was one of the first researchers to consider the role of landscape processes in driving marine population and community dynamics. To celebrate that research, Van Hoeck et al (2021) develop a new technique to improve coastal management, specifically the population enhancement of target species, a model system that Pete often used to test both basic ecological theory and as a means to improve management outcomes. Linking land to sea, Gehman et al (2021) uncover the influence that land use patterns have on estuarine ecosystems by modifying parasite–host interactions.…”

Unlocking the soundscape of coral reefs with artificial intelligence: pretrained networks and unsupervised learning win out

“…However, these diver-led surveys incur high expertise, logistical and financial costs (1,2). Passive acoustic monitoring (PAM) presents an alternative means of gathering monitoring data which can be collected with greater ease and over extended periods (3,4). However, in contrast to ecosystems where PAM is well established, the species identity for any given sound in reef PAM recordings is usually unknown (5,6).…”

“…Remaining plots can be found in Supplementary 1. (C) Chi-squared scores (x 10 3 ) generated from contingency tables of each recordings’ true class and the cluster that each recording was assigned to by affinity propagation clustering. Affinity propagation was performed on UMAP embeddings generated using each of the three methods (compound index, pretrained CNN and trained CNNs).…”