A practical method to account for variation in detection range in acoustic telemetry arrays to accurately quantify the spatial ecology of aquatic animals. Issue 1 (22nd November 2019)

• Record Type: Journal Article
• Authors: Brownscombe, Jacob W.; Griffin, Lucas P.; Chapman, Jacqueline M.; Morley, Danielle; Acosta, Alejandro; Crossin, Glenn T.; Iverson, Sara J.; Adams, Aaron J.; Cooke, Steven J.; Danylchuk, Andy J.
• Editors: Codling, Edward
• Is Part Of: Methods in ecology and evolution. Volume 11:Issue 1(2020)
• Page Start: 82
• Page End: 94
• Publication Date: 2019-11-22
• Subjects: acoustic telemetry arrays -- animal tracking -- bioacoustics -- biotelemetry -- data analysis -- detection range -- machine learning -- spatial ecology; Ecology -- Periodicals; Evolution -- Periodicals; 577
• DOI: 10.1111/2041-210X.13322
• Languages: English
• ISSNs: 2041-210X

Abstract

Abstract: Acoustic telemetry is a popular tool for long‐term tracking of aquatic animals to describe and quantify patterns of movement, space use, and diverse ecological interactions. Acoustic receivers are imperfect sampling instruments, and their detection range ( DR ; the area surrounding the receiver in which tag transmissions can be detected) often varies dramatically over space and time due to dynamic environmental conditions. Therefore, it is prudent to quantify and account for variation in DR to prevent telemetry system performance from confounding the understanding of real patterns in animal space use. However, acoustic receiver DR consists of a complex, dynamic, three‐dimensional area that is challenging to quantify. Although quantifying the absolute DR of all receivers is infeasible in the context of most acoustic telemetry studies, we outline a practical approach to quantify relative variation among receiver DR over space and time. This approach involves selecting a set of sentinel receivers to monitor drivers of variation in detection range. Each sentinel receiver is subject to a range testing procedure to estimate detection efficiency ( DE ; the proportion of total transmissions detected by the receiver), at a range of distances from the receiver, to derive the maximum range ( MR ; distance from the receiver where DE is 5%) and Midpoint (distance from the receiver where DE is 50%). A reference transmitter is then placed at the Midpoint, providing a standardized measure of long‐term variation in DE, with each station having similar freedom of variance. Variation in reference tag DE is then combined with MR to calculate a DR correction factor ( DRc ). A modelling approach is then used to estimate DRc for all receivers in the array at spatial and temporal scales of ecological interest, which can be used to correct animal detection data in various ways. We demonstrate this method with a hypothetical dataset, as well as empirical data from an acoustic telemetry array to delineate spatio‐temporal patterns of fish habitat use. This is a flexible and practical approach to account for variation in acoustic receiver performance, allowing more accurate spatial and temporal patterns in aquatic animal spatial ecology to be revealed.