Welcome to the new IOA website! Please reset your password to access your account.

Volume : 44

Part : 2

Proceedings of the Institute of Acoustics

 

 

Noise in restorative environments – perceptions, positive and negative environmental components in Tyrolean children

 

Angel Dzhambov1, Medical University of Plovdiv, Plovdiv, Bulgaria

Peter Lercher2, Graz University of Technology, Graz, Austria

 

ABSTRACT

 

Annoyance is the most common adverse reaction to environmental noise, but relatively little is known of noise annoyance in children, even though children find themselves in a vulnerable developmental period. Moreover, noise perception is usually studied as a unidimensional phenomenon with an emphasis on annoyance, disregarding positive appraisals of the acoustic environment and the contribution of further physical environment attributes. Here, we explored the relationships between traffic noise and greenspace and children’s perception of their neighbourhood’s quality and sonic environment. For this purpose, we used data from a cross-section survey among 1251 schoolchildren (8–12 years old) in the Tyrol area (Austria/Italy). Children and their mothers completed questionnaires on socio-demographics, living conditions, perceived neighbourhood quality, and noise disturbance in different situations. Traffic noise (Lden), air pollution (NO2), and greenness (NDVI) estimates were assigned to children’s home address. Disturbance was associated with higher Lden and NO2 and lower NDVI. Lden and NDVI had direct and indirect associations with disturbance though perceived neighbourhood quality. To gain more insights into children’s reactions to their sonic environment, both negative and positive environmental characteristics should be studied in conjunction to provide more differentiated input to planning.

 

Keywords: air pollution; greenness; noise annoyance

 

1. INTRODUCTION

 

Although annoyance is regarded as the most common adverse reaction to environmental noise, only few studies have investigated it in children [1-8]. This is surprising, as, compared with adults, children spend more time at home, are in a vulnerable developmental period, have fewer coping optionsр and are thought to exhibit less flexibility in meeting various adaptive challenges and threats from the environment [9]. Furthermore, the major established non-auditory long-term effects (e.g., hypertension, angina pectoris, diabetes) observed in adults [10] take years to unfold. Therefore, annoyance, quality of life, sleep disturbance, and cognitive effects remain the central areas of concern in children [11], but the pathways and the relative importance of these adverse outcomes are not yet established in this age group.

 

Another critical question relates to the measurement of annoyance in children. Hitherto, annoyance has predominantly been measured with single items, adopted from adult studies [12]. Such an approach disregards other potentially positive appraisals and coping opportunities provided by the physical environment, which are also important for preventive planning. Furthermore, mostly noise has been in the focus of earlier studies, while the impact of transportation is much broader and includes air pollution, vibration, and safety issues [2]. Whether this narrow focus on noise annoyance is sufficient and valid to cover the broad perceptual and behavioural impacts of traffic noise on children remains an open question.

 

An earlier alpine study [2], using a broader “environmental list” in addition to standard annoyance ratings, found a lower mean annoyance response in children compared with their mothers. However, stronger positive and negative effect modification was observed by contextual factors related to the surrounding built environment and behavioural options related to playing outdoors. In the meantime, a few studies in adults have observed lower annoyance response in the presence of more greenness [13-16].

 

The present study uses an augmented “environmental list”, to explore the joint relationships between traffic noise, greenness, and children’s perception of their environment. More specifically, we theorize that traffic noise and greenness are associated with noise annoyance (or disturbance as we refer to it later in the paper) both directly and indirectly, through perceived environmental quality.

 

2. METHODS

 

2.1. Questionnaire Data Collection

 

We used data from the Brenner Base Tunnel Study collected in 2004-2005 from 1251 8-12 years old schoolchildren sampled from 49 schools in the Tyrol region of Austria and Italy (Lower Inn, Wipp, and side valleys). Ethical approval was obtained from the Ethics committee of the Medical University Innsbruck (Ethics commission number 2105/2004).

 

Mother’s questionnaire provided information on sociodemographic and housing factors. Covariates considered in this study were child’s sex and age, maternal education (basic, skilled labour, vocational, and A-level), dwelling type (single family detached house, row house, or multiple dwellings), and crowding in the home (people/rooms ratio).

 

Children answered questions on disturbance by different sounds from which we created a summary disturbance scale by summing up nine items (each measured on a scale from 1 to 4) asking about disturbance by car, truck, and railway noise, general noisiness of the neighbourhood, as well as disturbance by traffic noise during specific activities (doing homework, relaxation, TV watching, being outdoors, and trying to fall asleep). This disturbance scale had high internal consistency reliability (McDonald's ω = 0.84).

 

Children answered seven items on perceived neighbourhood qualities (a lot of space to play, meadows and trees, clean air, quietness, children being allowed to run around, people being helpful to children, and children enjoying living there). These items conceptually overlapped the Perceived Restorativeness Scale for Children [17]. We treated perceived neighbourhood quality as a mediator between environmental exposures and disturbance.

 

2.2. Exposure Variables

 

Modelled day-evening-night sound (Lden) and nitrogen dioxide (NO2) levels were assigned to children’s home address. The Lden indicator reflected combined freeway, main road, and railway sound levels with background level. In addition to traffic noise, we considered an indicator of air pollution. These exposures were calculated from bespoke models and validated against field measurements [18, 19]. The normalized difference vegetation index (NDVI 100-m) was used as a measure of general vegetation level in a 100 m circular buffer radius around the child’s home, where values closer to +1 indicate high greenness [20]. NDVI was calculated at a 30-m2resolution using satellite data (cloud free images from July-August 2003) from the Landsat 4-5 Thematic Mapper. Mothers also reported whether their house had a garden.

 

2.3. Statistical Analysis

 

Following descriptive analyses, we fitted linear regression models to examine the effects on disturbance of Lden, NO2, NDVI 100-m, and having a garden. The models were adjusted for child’s sex, and age, maternal education, dwelling type, and garden. The models with Lden and NO2 were additionally adjusted for NDVI 100-m, and the model with garden, for NDVI 100-m. Models did not suffer from multicollinearity (tolerance > 0.2 and Variance Inflation Factor < 5.0).

 

Next, we tested the functional form between disturbance and Lden and NDVI 100-m, using nonparametric series regressions. In this model, the data supplies the model structure as well as the model estimates, which enables the flexible modelling of non-linear associations.

 

Finally, we fitted a structural equation model (SEM) to test the associations between exposures (Lden, NO2, NDVI 100-m, and garden) and disturbance, as mediated by perceived neighbourhood quality. As control variables, we also included child’s sex and age, maternal education, dwelling type, and crowding. A priori, we assumed the following covariances: NDVI 100-m ↔ Lden, NDVI 100- m ↔ NO2, NDVI 100-m ↔ garden, NO2 ↔ Lden, between dummy variables defining dwelling type, and between maternal education and crowding. We employed the diagonally weighted least squares estimation method with robust standard errors. Coefficients reported in the SEM were probit regression estimates. Indirect effects were computed as the product of the regression weights associated with their constituent paths, and standard errors for these defined parameters were computed using the Delta method. We evaluated goodness-of-fit using indices of acceptable model fit suggested by Hu and Bentler [21]: non-significant χ2(p > 0.05); CFI ≥ 0.95; RMSEA ≤ 0.06 with a 90% CI ≤ 0.06; and SRMSR ≤ 0.08. Structural equation modelling was conducted with JASP v. 0.16.1.0. All other analyses were conducted with Stata MP v. 17. A p-value of < 0.05 was considered statistically significant.

 

3. RESULTS AND DISCUSSION

 

In total, 1251 children were included in the sample, of which 49.8% (n = 623) were boys. The mean age of the children was 9.36 (SD = 0.65) years. Descriptive statistics for the exposure variables are shown in Table 1. While greenness levels were generally high, most children were exposed to relatively low Lden and NO2 levels.

 

As expected, Lden and NO2 were highly positively correlated (ρ = 0.77) and inversely correlated with NDVI 100-m (ρ = -0.31 and -0.39) and presence of a garden (ρ = -0.15 and -0.14).

 

Table 1: Descriptive statistics for the exposure variables in the study.

 

 

In a linear regression model, Lden (β per 1 dBA = 0.14; 95% CI: 0.10, 0.17) and NO2 (β per 1 μg/m3 = 0.11; 95% CI: 0.08, 0.15) were associated with higher disturbance. On the other hand, higher NDVI 100-m (β per 1 dBA = -8.30; 95% CI: -11.17, -5.43) was associated with lower disturbance. Living in a home with a garden was not associated with disturbance (β = -0.52; 95% CI: -1.41, 0.37). When we explored the functional form of the associations between Lden and NDVI 100-m and disturbance, we found no evidence of deviation from linearity (Figure 1). There was no interaction between Lden and NDVI 100-m (p = 0.540).

 

 

Figure 1: Associations between disturbance and Lden and NDVI 100-m (postestimation results from nonparametric series regressions)

 

The path model we constructed next is shown in Figure 2. It had a good fit to the data: χ2(25) = 35.28, p = 0.083; CFI = 1.00; RMSEA = 0.02 (90% CI: 0.00, 0.03); SRMR = 0.02. The model explained 27% of the variance in sound disturbance and 14% in perceived neighbourhood quality.

 

Table 2 shows selected parameter estimates for the SEM. Children exposed to higher noise levels and reporting lower neighbourhood quality were more likely to be disturbed by sounds. Other predictors of higher disturbance were younger age (-0.06; 95% CI: -0.11, <-0.001), female sex (-0.07; 95% CI: -0.12, -0.01), and less crowding in the home (-0.06; 95% CI: -0.11, -0.01).

 

Perceived neighbourhood quality was negatively associated with traffic noise (-0.11; 95% CI: - 0.19, -0.03) and positively associated with greenness (0.23; 95% CI: 0.16, 0.30) and maternal education (0.07; 95% CI: 0.01, 0.13).

 

 

Figure 2: Structural equation model showing estimated paths linking environmental exposures to sound disturbance (N = 1027). Notes: *Estimate is statistically significant at p < 0.05. Percentages indicate the variance explained in endogenous variables. Control variables (child’s sex, age, maternal education, house type, and crowding), covariances, and errors terms are not displayed to enhance readability.

 

Table 2: Parameter estimates for the structural equation model linking environmental exposures to sound disturbance (N = 1027).

 

 

Lden and NDVI 100-m had significant total effects on disturbance, where 21% of the effect of Lden was mediated by lower neighbourhood quality, and 69% of the effect of NDVI 100-m was mediated by higher neighbourhood quality. The effect of garden was in the same direction as that of NDVI 100-m, but borderline significant.

 

A comparison of these results with the few earlier studies [2-8] is difficult as those studies analysed annoyance as dichotomized outcome and did not include air pollution and greenspace indicators. Eventually, positive and restorative items or scores were not considered in their models.

 

4. CONCLUSIONS

 

In this study, we investigated a pathway through perceived neighbourhood qualities, reflecting the potential of children’s living environment to provide opportunities for distancing from stressors and positive engagements with the environment. To gain more insights into children’s reactions to their sonic environment and possibly ways in which soundscape quality can be enhanced, both negative and positive environmental characteristics should be studied in conjunction to provide more differentiated input to planning.

 

5. ACKNOWLEDGEMENTS

 

We want to thank first the inhabitants of the Lower Inn and Wipp valleys. Our thanks also go to the Austrian Ministry of Science and Transportation for funding the framework of the Environmental Health Impact Assessment (EHIA), the government of the Tyrol region for providing GIS data and informational support from the BEG (Brenner Eisenbahn Gesellschaft). The BBT survey got support from the BBT company within a legally required EHIA through EU-support. The noise mapping was done by INTEC, Ghent, and the air pollution assessment by an Italian-Austrian consortium. We also thank the large EHIA-teams in both studies who did the fieldwork. Finally, we thank Iana Markevych and Matthew Browning for calculating the normalized difference vegetation index.

 

6. REFERENCES

 

  1. Lercher, P. Annoyance, disturbance and severances in children exposed to transportation noise. Proceedings of the 8th International Congress on Noise as a Public Health Problem, pp. 241- 248. Rotterdam, the Netherlands, 29 June - 3 July 2003.

  2. Lercher, P., Brauchle, G., Kofler, W., Widman, U. & Meis. The assessment of noise annoyance in schoolchildren and their mothers. Proceedings of the 29th International Congress and Exhibition on Noise Control Engineering., pp. 2318–2322. Société Française d’Acoustique, Nice 2000.

  3. Haines, M. M., Brentnall, S. L., Stansfeld, S. A. & Klineberg E. Qualitative responses of children to environmental noise. Noise & Health, 5(19), 19-30 (2003).

  4. van Kempen, E. E., van Kamp, I., Stellato, R. K., Lopez-Barrio, I., Haines, M. M., Nilsson, M. E., Clark, C., Houthuijs, D., Brunekreef, B., Berglund, B. & Stansfeld S. A. Children's annoyance reactions to aircraft and road traffic noise. The Journal of the Acoustical Society of America, 125(2), 895-904 (2009).

  5. Persson Waye, K., van Kamp, I., Nielsen, K. & Dellve, L. Reliability and validity of instruments to measure pre-school children’s reaction to and coping with noise. Proceedings of the 10th International Congress on Noise as a Public Health Problem, London, 24-28 July, (2011).

  6. Babisch, W., Schulz, C., Seiwert, M., & Conrad, A. Noise annoyance as reported by 8-to 14-year old children. Environment and Behavior, 44(1), 68-86 (2012).

  7. Grelat, N., Houot, H., Pujol, S., Levain, J. P., Defrance, J., Mariet, A. S. & Mauny, F. Noise Annoyance in Urban Children: A Cross-Sectional Population-Based Study. International Journal of Environmental Research and Public Health, 13(11), 1056 (2016).

  8. Quehl, J., Bartels, S., Fimmers, R. & Aeschbach, D. Effects of Nocturnal Aircraft Noise and Non Acoustical Factors on Short-Term Annoyance in Primary School Children. International Journal of Environmental Research and Public Health, 18, 6959 (2021).

  9. Evans, G. W. Child development and the physical environment. Annual Reviews of Psychology, 57, 423-51 (2006).

  10. WHO. Environmental Noise Guidelines for the European Region. Copenhagen (2018).

  11. Clark, C. & Paunovic, K. WHO Environmental Noise Guidelines for the European Region: A Systematic Review on Environmental Noise and Cognition. International Journal of Environmental Research and Public Health, 15, 285 (2018).

  12. Fields, J.M.; De Jong, R.G.; Gjestland, T.; Flindell, I.H.; Job, R.F.S.; Kurra, S.; Lercher, P.; Vallet, M.; Yano, T.; Guski, R.; et al. Standardized general-purpose noise reaction questions for community noise surveys: Research and a recommendation. J. Sound Vib. 2001, 242, 641–679.

  13. Van Renterghem, T. & Botteldooren D. View on outdoor vegetation reduces noise annoyance for dwellers near busy roads. Landscape and Urban Planning, 148, 203–215 (2016).

  14. Dzhambov, A. More residential greenspace is associated with lower noise annoyance: Results from a quantitative synthesis of the literature. In: Lucjan C, Gérard D (eds.) Traffic Noise: Exposure, Health Effects and Mitigation. Nova Science Publishers, New York, pp. 77–104 (2017).

  15. Dzhambov, A. M., Markevych, I., Tilov, B., Arabadzhiev, Z., Stoyanov, D., Gatseva, P. & Dimitrova, D. D. Lower Noise Annoyance Associated with GIS-Derived Greenspace: Pathways through Perceived Greenspace and Residential Noise. International Journal of Environmental Research and Public Health, 15(7), 1533 (2018).

  16. Schäffer, B., Brink, M., Schlatter, F., Vienneau, D. & Wunderli, J. M. Residential green is associated with reduced annoyance to road traffic and railway noise but increased annoyance to aircraft noise exposure. Environment International, 143, 105885 (2020).

  17. Berto, R., Pasini, M. & Barbiero, G. How does Psychological Restoration Work in Children? An Exploratory Study. Journal of Child and Adolescent Behavior, 3(3), 200 (2015).

  18. Öttl, D., Sturm, P-J., Anfossi, D., Trini Castelli, S., Lercher, P., Tinarelli, G. & Pittini, T. Chapter 7.2 Lagrangian particle model simulation to assess air quality along the Brenner transit corridor through the Alps. Air Pollution Modeling and Its Application XVIII (1 ed., Vol. Volume 6, pp. 689-697). (Developments in Environmental Science). Elsevier B.V.. (2007).

  19. Jonasson, H.G. Acoustical source modelling of road vehicles. Acta Acustica united with Acustica, 93(2), 173-184 (2007).

  20. Gascon, M., Cirach, M., Martínez, D., Dadvand, P., Valentín, A., Plas`encia, A. & Nieuwenhuijsen, M. Normalized difference vegetation index (NDVI) as a marker of surrounding greenness in epidemiological studies: the case of Barcelona city. Urban Forestry and Urban Greening, 19, 88–94 (2016).

  21. Hu, L. T. & Bentler, P. M. Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling, 6, 1–55 (1999).

 

1 angelleloti@gmail.com; angel.dzhambov@mu-plovdiv.bg

2 peter.lercher.at@gmail.com