Sleep actigraphy time-synchronized with wind turbine output

• Published in: Sleep (New York, N.Y.) Vol. 44; no. 9; p. 1
• Main Authors: Michaud, David S, Keith, Stephen E, Guay, Mireille, Voicescu, Sonia, Denning, Allison, McNamee, James P
• Format: Journal Article
• Language: English
• Published: US Oxford University Press 01-09-2021
• Subjects: Air-turbines; Analysis; Motility; Sleep; Turbines; Community Noise and Health Study; awakenings; wind turbine; motility; sleep actigraphy; Actiwatch2; SCADA
• ISSN: 0161-8105; 1550-9109
• DOI: 10.1093/sleep/zsab070

Abstract
Studies have yielded inconsistent evidence for an association between long-term average wind turbine sound pressure level (SPL) and disturbed sleep. Transient changes in sleep may be more susceptible to short-term variations in wind turbine SPL throughout the sleep period time. We analyzed sleep actigraphy data (participant sleep nights = 2,094, males = 151, females = 192) in 10 min intervals time-synchronized to wind turbine supervisory control and data acquisition. Calculated indoor wind turbine SPL was considered after adjusting for turbine rotor speed and closed/open bedroom windows. Maximum calculated nightly average wind turbine SPL reached 44.7 dBA (mean = 32.9, SD = 6.4) outdoors and 31.4 dBA (mean = 12.5, SD = 8.3) indoors. Wind turbine SPL in 10 min intervals, and nightly averages, was not statistically associated with actigraphy outcomes. However, the variability in wind turbine SPL due to changes in wind turbine operation across the sleep period time, as measured by the difference between the 10 min SPL and the nightly average SPL (∆SPL), was statistically related to awakenings (p = 0.028) and motility (p = 0.015) rates. These diminutive differences translate to less than 1 min of additional awake and motility time for a 5 dBA increase over a 450 min sleep period time. Overall results showed that wind turbine SPL below 45 dBA was not associated with any consequential changes in actigraphy-measured sleep. Observations based on ∆SPL provided some indication that a more sensitive assessment of sleep may be one that considers variations in wind turbine SPL throughout the sleep period time.