Derivation and application of a composite annoyance reaction construct based on multiple wind turbine features

OBJECTIVES: Noise emissions from wind turbines are one of multiple wind turbine features capable of generating annoyance that ranges in magnitude from not at all annoyed to extremely annoyed. No analysis to date can simultaneously reflect the change in all magnitudes of annoyance toward multiple wind turbine features. The primary objective in this study was to use principal component analysis (PCA) to provide a single construct for overall annoyance to wind turbines based on reactions to noise, blinking lights, shadow flicker, visual impacts, and vibrations evaluated as a function of proximity to wind turbines. METHODS: The analysis was based on data originally collected as part of Health Canada’s cross-sectional Community Noise & Health Study (CNHS). One adult participant (18–79 years), randomly selected from dwellings in Ontario (ON) (n = 1011) and Prince Edward Island (PEI) (n = 227), completed an in-person questionnaire. Content relevant to the current analysis included the annoyance responses to wind turbines. RESULTS: The first construct tested in the PCA explained 58–69% of the variability in total annoyance. Reduced distance to turbines was associated with elevated aggregate annoyance scores among ON and PEI participants. In the ON sample, aggregate annoyance was effectively absent in areas beyond 5 km (mean 0.12; 95% CI 0.00, 1.19), increasing significantly between (2 and 5] km (mean 2.13; 95% CI 0.92, 3.33), remaining elevated, but with no further increase until (0.550–1] km (mean 3.37; 95% CI 3.02, 3.72). At ≤ 0.550 km, the average overall annoyance was 3.36 (95% CI 2.03, 4.69). In PEI, aggregate annoyance was essentially absent beyond 1 km; i.e., (1–2] km (mean 0.21; 95% CI 0.00, 0.88); (2–5] km (mean 0.00; 95% CI 0.00, 1.37); > 5 km (mean 0.00; 95% CI 0.00, 1.58). Annoyance significantly increased in areas between (0.550 and 1] km (mean 1.59; 95% CI 1.02, 2.15) and was highest within 550 m (mean 4.25; 95% CI 3.34, 5.16). CONCLUSION: The advantages and disadvantages to an aggregated annoyance analysis, including how it should not yet be considered a substitute for relationships based on changes in high annoyance, are discussed. ELECTRONIC SUPPLEMENTARY MATERIAL: The online version of this article (10.17269/s41997-018-0040-y) contains supplementary material, which is available to authorized users.