Assessment of environmental noise in high-density cities : simulation, measurement, and data analytics

Abstract

Environmental noise is listed by the World Health Organization as the number two cause of ill health after air pollution. Residents of high-density cities are especially vulnerable to noise. In order to advance the knowledge of environmental noise and understand its impacts on urban residents in high density cities, there is a need to link individual noise exposure with personal characters, health conditions, and noise perceptions, that are captured by the urban big data. There is a call for the assessment of noise exposure, which is highly variable from household to household, at household or individual level. Existing noise assessment methods are challenged by either feasibility or reliability issues in the applications of assessing large quantity of household or individual noise exposure in complex high-density cities. The opportunities provided by urban big data were yet sufficiently explored for investigating the urban noise environment. The objectives of this research are to 1) assess the exposure and responses to environmental noise in high-density cities at household or individual level, 2) demonstrate a 3D geo-tagged database composed of household road traffic noise exposure and individual health records that is capable of advancing researches on health impacts of road traffic noise in high-density cities; 3) explore potentials in urban big data for proactive monitoring and management of urban noise environment. This research advanced the knowledge and assessment techniques of environmental noise in high-density cities with the three studies described below: Firstly, a computer simulation-based method, that combines the CadnaA software with an “adjusted CRTN” algorithm, was developed and evaluated to assess point-based household road traffic noise exposure in high-density cities. The model can be “scaled-up” to estimate road traffic noise exposure for large population. Secondly, a database consisted of household road traffic noise exposure and 3D built environment attributes was developed for over 17,000 residents in Hong Kong in collaboration with the FAMILY Cohort and “Children of 1997”, a Hong Kong Chinese birth cohort (Birth Cohort) that are originally developed by the Faculty of Medicine, HKU. This 3D geo-tagged database is the first of its kind and is of value to advance studies on health impacts of road traffic noise as well as other population-based researches on built environment and health. Lastly, a method that combines data mining, machine learning, spatial analysis and data analytics was developed to measure human responses to noise incidents in the Greater Taipei Area, using geo-tagged Twitter data and Public Nuisance Petitions (PNPs). The two databases, when combined, provide new means to study human perceptions of the urban noise environment, and show potentials for proactive monitoring and management of urban noise environment. Original contributions of this research are threefold: 1) a computer simulation-based method is developed to assess household road traffic noise exposure in high-density cities; 2) a 3D database of household road traffic noise exposure and built environment attributes is developed for 8,158 households in the FAMILY Cohort and the Birth Cohort databases; 3) a social-media-based method is developed to assess human perceptions of urban noise environment.