Harbour porpoise distribution can vary at small spatiotemporal scales in energetic habitats

• Published in: Deep Sea Research Part II: Topical Studies in Oceanography
• Main Authors: Benjamins, Steven, van Geel, Nienke, Hastie, Gordon, Elliott, Jim, Wilson, Ben
• Format: Article in Journal/Newspaper
• Language: English
• Published: 2017
• Subjects: Monitoring; Arrays - Marine mammals; Tidal currents; Water waves - Marine renewable energy development (MRED) - Passive acoustics; Gam; Harbour porpoise; Phocoena phocoena
• Online Access: https://risweb.st-andrews.ac.uk/portal/en/researchoutput/harbour-porpoise-distribution-can-vary-at-small-spatiotemporal-scales-in-energetic-habitats(84692776-4f98-4e09-b844-d8c0602bcdc9).html; https://doi.org/10.1016/j.dsr2.2016.07.002; https://research-repository.st-andrews.ac.uk/bitstream/10023/11068/1/Hastie_2017_DSRII_SpatiotemporalScales_CC.pdf; http://www.sciencedirect.com/science/article/pii/S0967064516301849?_rdoc=1&_fmt=high&_origin=gateway&_docanchor=&md5=b8429449ccfc9c30159a5f9aeaa92ffb#s0065

Marine habitat heterogeneity underpins species distribution and can be generated through interactions between physical and biological drivers at multiple spatiotemporal scales. Passive acoustic monitoring (PAM) is used worldwide to study potential impacts of marine industrial activities on cetaceans, but understanding of animals’ site use at small spatiotemporal scales (<1 km, <1 day) remains limited. Small-scale variability in vocalising harbour porpoise ( Phocoena phocoena ) distribution within two Scottish marine renewable energy development (MRED) sites was investigated by deploying dense arrays of C-POD passive acoustic detectors at a wave energy test site (the European Marine Energy Centre [Billia Croo, Orkney]) and by a minor tidal-stream site (Scarba [Inner Hebrides]). Respective arrays consisted of 7 & 11 moorings containing two C-PODs each and were deployed for up to 55 days. Minimum inter-mooring distances varied between ~300–600 m. All C-POD data were analysed at a temporal resolution of whole minutes, with each minute classified as 1 or 0 on the basis of presence/absence of porpoise click trains (Porpoise-Positive Minutes/PPMs). Porpoise detection rates were analysed using Generalised Additive Models (GAMs) with Generalised Estimation Equations (GEEs). Although there were many porpoise detections (wave test site: N = 3,432; tidal-stream site: N = 17,366), daily detection rates varied significantly within both arrays. Within the wave site array (<1 km diameter), average daily detection rates varied from 4.3–14.8 PPMs/day. Within the tidal-stream array (<2 km diameter), average daily detection rates varied from 10.3–49.7 PPMs/day. GAM-GEE model results for individual moorings within both arrays indicated linkages between porpoise presence and small-scale heterogeneity among different environmental covariates (e.g. tidal phase, time of day). Porpoise detection rates varied considerably but with coherent patterns between moorings only several hundred metres apart and within hours. These ...