key: cord-0491437-c91lpqpf
authors: Basu, Bidroha; Murphy, Enda; Molter, Anna; Basu, Arunima Sarkar; Sannigrahi, Srikanta; Belmonte, Miguel; Pilla, Francesco
title: Effect of COVID-19 on noise pollution change in Dublin, Ireland
date: 2020-08-20
journal: nan
DOI: nan
sha: 3badfd8ceb114efd6632f015bd3a52c21c69c808
doc_id: 491437
cord_uid: c91lpqpf

Noise pollution is considered to be the third most hazardous pollution after air and water pollution by the World Health Organization (WHO). Short as well as long-term exposure to noise pollution has several adverse effects on humans, ranging from psychiatric disorders such as anxiety and depression, hypertension, hormonal dysfunction, and blood pressure rise leading to cardiovascular disease. One of the major sources of noise pollution is road traffic. The WHO reports that around 40% of Europe's population are currently exposed to high noise levels. This study investigates noise pollution in Dublin, Ireland before and after the lockdown imposed as a result of the COVID-19 pandemic. The analysis was performed using 2020 hourly data from 12 noise monitoring stations. More than 80% of stations recorded high noise levels for more that 60% of the time before the lockdown in Dublin. However, a significant reduction in average and minimum noise levels was observed at all stations during the lockdown period and this can be attributed to reductions in both road and air traffic movements.

The growth in urban populations around the world has resulted in the expansion of cities which are important for providings jobs, housing, and sustainable livelihoods (He et al., 2018; Forman and Wu, 2016; Zhou et al., 2016; Sun et al., 2016; Brueckner, 2000; Peng, 1997) . Given the evolution of cities and particularly those that are more heavily dependant on car-based travel, pollution levels within cities has shown an increasing trend over time (Hien et al., 2020; Lagonigro et al., 2018; Rocha et al., 2017) . Amongst all pollutants, the World Health Organisation (WHO) has reported that noise pollution is the third most hazardous type of pollutant after air and water pollution (WHO, 2005) . Health impacts due to an increase in environmental noise are a concern worldwide (Alves et al., 2015; Merchan et al., 2014; Nedic et al., 2014; Ongel and Sezgin, 2016; Pathak et al., 2008) . For example, noise sensitivity can be an important contributor to psychiatric disorders such as anxiety and depression (Belojević et al., 1997; Fyhri and Aasvang, 2010; Ongel and Sezgin, 2016) . Recent studies suggest that an increase of 5dB roadside noise can raise the chance of hypertension by 3.4% Oh et al., 2019; Eriksson et al., 2012) . Further studies indicate that exposure to a high level of noise can result in hormonal dysfunction and can also contribute to the rise in blood pressure which can severely impact the cardiovascular system in the body (Said and Goharty, 2016; Münzel and Sørensen, 2017) . Other studies have reported that pregnant women might be at a greater risk of being affected to noises since they are more sensitive to environmental stress factors (Selander et al., 2019; He et al., 2019; Poulsen et al., 2018; Sears et al., 2018; Murphy and Faulkner, 2018) . Ashin et al. (2018) noted that road traffic noise can increase gestational diabetes mellitus, which leads to glucose intolerance that occurrs during the beginning of pregnancy. Based on 29 case study analysis where pregnant women were exposed to 80dB or higher noise, Dzhambov et al. (2014) documented that the risk for having gestational hypertension, small for gestational age and babies with congenital malformations increases significantly. The European Union documented that more than 40 percent of the total European population is exposed to a Day-Evening-Night ( ) noise level of 55dB or greater, while 30 percent of the population is exposed to the same noise level during night-time (WHO, 2017; Foraster et al., 2017; Pitchika et al., 2017; Basner and Mcguire, 2018) . A city soundscape study (Lebiedowska, 2005; Léobon, 1995) suggested that urban noise can be broadly classified in the following four categories: i) background noise: classified as unpleasant due to the presence of high pitched, piercing, strong, continuous, irregular, or intermittent noises that cause humming of the peripheral environment; ii) mechanical noise: noise caused by mechanical equipments such as vehicles, railway, and aircraft as well as large industrial plants generating noises; iii) human activity related noise: can generate from demonstrations, gatherings, sirens, trades, household noise due to usage of vacuum cleaners or drills etc.; iv) other environmental noise:

can be attributed to the presence of storms, thunders, winds, and creaking.

The Noise Observation and Information Service for Europe (NOISE) suggest that the majority of the noise affecting the exposure of the population are being generated by road vehicle traffic, compared to other sources of transportation (King and Murphy, 2016; Śliwińska-Kowalska and Zaborowski, 2017; Nedic et al., 2014; Voltes and Martin, 2016; Yuan et al., 2019; Das et al., 2019) . Detailed studies and literature reviews have shown that the noise pollution from urban road traffic has the highest level of exposure as the roads are located in close proximity to built infrastructures such as schools, offices and residential buildings (Khan et al., 2018; Paiva et al., 2019; von Graevenitz, 2018; Rey and Barrigón, 2016; Cai et al., 2019; González et al., 2018; Ruiz-Padillo et al., 2016; Zannin et al., 2003) . Moreover, research has shown that not only the noise pollution, but the duration of exposure to traffic-related noises has impacted the health conditions of people severely (Tonne et al., 2016; Buxton et al., 2017; Khan et al., 2018; Cai et al., 2019) . The European Commission has reported that the noise from rail and road costs the European Union around 40 billion euros per year in terms of socioeconomic damages (Vaitkus et al., 2016; Andriejauskas et al., 2018) . Apart from socioeconomic cost, studies have also revealed that noise pollution caused due to human-generated factors can potentially alter the biodiversity by impacting the distribution and behavior of species and their habitat quality (Siemers and Schaub, 2011; Estabrook et al., 2016; Guttal and Jayaprakash, 2007) . Although noise levels are generally highest around the roadways and near the transportation terminals, the exposure of the large majority of city inhabitants is determined by a city's overall background noise (Mehdi et al., 2011; Stansfeld and Matheson, 2003; Khan et al., 2018; Raimbault et al., 2003; Morillas et al., 2002) . Other studies have further revealed that several factors such as morphology of the city (aspect ratios of buildings), current and projected population as well as household density, noise regulations laws and policies, traffic network design, availability of affordable public transport nearby of the area, frequency of traffic jams, the public vehicles to private vehicles on road, the average insulation of the homes and the noises escaping from the building, and the motorized driving behavior, possibility of construction of new buildings, etc. can determine the noise levels at different scales (Ariza-Villaverde et al., 2014; Bouzir and Zemmouri, 2017; Piccolo et al., 2005; Abbaspour et al., 2015) . Noise distribution assessment in a city should also take into the local behavior of traffic flows, street, and urban maintenance or expansion initiatives, if being undertaken (Mato and Mufuruki, 1999; Yu and Kim, 2011; Barros and Dieke, 2008) . Thus, noise mapping studies need to be conducted at the microscale as well as at a broader scale (Margaritis and Kang, 2016; Dutta et al., 2017; Tiwari et al., 2019) . Other studies reveal that the material building blocks used for the construction of building infrastructure as well as building and street design at the nearby vicinity and neighborhood quality are correlative factors responsible for determining the level of noise (Onaga and Rindel, 2007; Picaut and Simon, 2001; Salomons and Pont, 2012) .

Similar studies suggest that the closed building blocks with low population density lead to lower

noise levels than open building block constructions with higher population density (Silva et al., 2018; Zhou et al., 2017; de Souza et al., 2011; Costa et al., 2011; Nega et al., 2012) . The percentage of quiet areas is correlated to the nature of the street and the influx of traffic and population present in the area (Liu et al., 2019; Shepherd et al., 2013) .

Within the foregoing context, this study investigates the effect of the COVID-19

lockdown on changes in noise pollution in Dublin, Ireland. Since the lockdown, the traffic volume has been substantially reduced and social events have been restricted considerably; therefore, a reduction in noise pollution across the city would be expected. This study performs a regression based trend analysis to investigate the changes in noise pollution before and after the lockdown. Furthermore, a change-point analysis was performed to identify the exact date on which the change in the pattern of noise pollution occurred in the noise monitoring stations;

change-point analysis is useful for estimating the point at which the statistical properties of a sequence of observations change.

An analysis of changes in noise pollution in Dublin, Ireland due to the lockdown is performed based on noise data obtained from 12 monitoring stations in Dublin. This section provides a detailed description of the data and the methodology used in the analysis.

Equivalent Continuous Sound Pressure Level ( ) and maximum noise ( ) data recorded at 5 minute intervals at 12 locations in Dublin ( Figure 1 and Table 1) 

where { , = 1, ⋯ , ℎ} is the ℎ number of noises during the chosen hour at a 5 min interval. Equation (1) ensures that the hourly average noise produces the same amount of total energy produced over the one-hour period.

Relating high noise with traffic variable is a challenging task. The major difficulty lies in acquiring reliable traffic variable data for Dublin at an hourly scale. In this study, annual average hourly traffic flow data at every road in Dublin were obtained from the Dublin City

Council's Sydney Coordinated Adaptive Traffic System (SCATS) for the year 2016. It needs to be noted that 2016 is the latest available traffic data for Dublin, which can be assumed to be a representative traffic flow for the city before lockdown, while considerably lower traffic flow can be assumed during the lockdown period.

In order to estimate the changes in noise pollution over the entire time period , two approaches have been used in this study. A linear regression is fitted to estimate the slope of the noise pollution time series. Considering there are number of hourly noise data points Identification of can be performed by minimization of the following objective function.

where is a cost function and is a penalty term to stop overfitting.

there is no sudden change in the time series.

The penalty term considered in the study is the Schwarz/Bayesian information criteria and is given by,

where d is the number of additionally introduced model parameters for estimating the change point.

The cost function is provided as,

where ( | ) is the density function of the time series data , ⋯ , and is the set of distribution parameters.

A linear model (linear regression) can be used to model the properties of a time series in situations where the underlying system is linear in nature. A system can be assumed to be linear if the response function ( ) from the system follows the following criteria (Billings and Voon,

Considering ( ) = ( ) − ( ) , if the cross-correlation function between ′( ) and ′( ) for any lag are zero (equation 6), the system is assumed to be linear.

For large data (considerable high value of ≥ 30), the 95% confidence interval can be chosen as ± 1.96 √ ⁄ , indicating values of , within this bound can be assumed to follow a linear system.

Hourly average noise obtained from the 12 noise monitoring stations for the pre-lockdown period ranging from 1 st January 2020 to 24 th March 2020 was compared to the hourly average Since the noise pattern varies considerably on the time of the day due to changes in traffic conditions and other human activities, the 24-hour period is subdivided into three timeframes, night time (11 PM to 7 AM), day time (7 AM to 7 PM) and evening time (7 PM to 11 PM).

Average, maximum, and minimum noise for each of those three timeframes for each station were estimated based on the hourly average noise data. Subsequently, a regression analysis was performed for each station corresponding to every timeframe to identify the trend in changes in the noise. Figure 5 shows the time series plots (shown in black) and the trend (shown in red), of the average noise, for brevity. The regression slope values for average, maximum, and minimum noise for each station and for every timeframe are provided in ( ) corresponding to lags ranging from 1 to 50 for each station corresponding to night, day and evening time was plotted ( Figure 6 ). The figures indicated that majority of the time series data can be considered to be linear in nature.

Traffic is considered to be one of the major sources of noise in a city. The average traffic volume at each street in Dublin for night, day, and evening time is shown in Figure 7 . Average traffic counts surrounding each noise monitoring station within a radius of 500 meters for night time, day time and evening time were estimated (Table 3) the noise in the four stations (ID #3, 4, 7, 12) located near the city center is highly related to the traffic, while the other 6 stations is somewhat dependend on traffic. It should be noted that Ballymun Library (ID # 2) is closer to the Dublin airport and DCC Rowing Club (ID #5) is located adjacent to the river Liffey, hence noise sources such as aeroplanes and waterbodies, respectively, can have a major impact on the noise pollution in those two stations.

According to WHO guidelines (WHO, 2009), any noise above a threshold value of 55 dB is considered to be harmful for the human population. Long term exposure to noise level above the threshold value leads to frequent occurrence of health adversity to a large part of the exposed population. The total number and the percentage of times the hourly average noise pollution exceeded the threshold 55dB for every monitoring station before and during lockdown is shown in Gardens (ID #12), every monitoring stations had recorded noise pollution greater than the threshold for more than 60% of the time, while for Ballymun Library (ID #2) and Chancery

Park (ID #4), more than 95% of the time the hourly average noise exceeded the threshold value.

During lockdown, the number and the percentage of times the noise exceeded the threshold reduced considerably for each monitoring station. During the lockdown period, hourly average noise pollution at only 3 stations (Ballymun Library, ID #2; Chancery Park, ID #4; Dolphin's Barn, ID #6) exceeded the threshold for more than 60% of the time.

A higher value of noise pollution in Ballymun Library (ID #2) can be attributed to the fact that the monitoring station is located in close proximity to Dublin airport and the M50 highway, which is the busiest street in Dublin (Figure 1 ). The other two locations (Chancery Park, ID #4

and Dolphin's Barn, ID #6) exhibiting higher noise pollution are located close to the city center area. Furthermore, Chancery Park is adjacent to the Liffey river, where waterbody contributes to noise pollution. On the other hand, Drumcondra Library (ID #7) and Woodstock Gardens (ID #12) are located away from major roads (Figure 7) and have the lowest noise pollution.

During the lockdown, DCC Rowing Club (ID #5), Navan Road (ID #9) and Woodstock Gardens (ID #12) had the highest reduction in noise (Figure 2 and Figure 3a) , which can be attributed to the fact that DCC Rowing Club was shut down during the lockdown period, and the major source of noise in the station is the water flow in the river Liffey. Navan Road is located adjacent to the Phoenix Park, which is the biggest park in Dublin and was closed. Also, people living in the Lucan area located in the northwest part of the city mainly use this road to travel to the city center for jobs and other activities. Woodstock Gardens, on the other hand, is located in close proximity to University College Dublin that has the biggest university campus in Dublin, which was closed during the lockdown. Reduction in noise at Mellows Park (ID #8) and Walkinstown Library (ID #11) were found to be least (Figure 3a) Table 1 ). For precautionary measures, the majority of the primary and secondary schools in Dublin were closed from 14 th March 2020. The number of schools within 500 meter radius distance from each monitoring station are shown in Table 1 .

Except DCC Rowing Club (ID #5) and Walkinstown Library (ID #11), all the other stations are surrounded by several primary and secondary schools (Table 1) Since the traffic information is unavailable for Dublin during the lockdown period, the traffic data was linked with noise pollution for the pre-lockdown period. Furthermore, since the traffic data was not available for 2020, the best available data were used as an approximation. As the Ballymun Library (ID #2) is close to the M50 but it is more than 500 meters in distance (~950 meters), the traffic count does not get reflected. Also, one of the major noise sources for this station is the flights since it is in close proximity to the Dublin airport, which can be the reason for different behavior of the station compared to the other stations. However, even though the scatterplot denotes the presence of a relation between noise and traffic for all the other stations, it needs to be noted that traffic alone cannot be considered to be the sole factor generating noise, and other factors need to be explored to model noise pollution, which is evident from the amount of variability present in those scatterplots (Figure 8 ). The correlation between traffic count and noise pollution is considerably high only for the four stations located surrounding the city center (ID #3, 4, 7, 12).

As Drumcondra Library (ID #7) and Woodstock Gardens (ID #12) are located in regions with low traffic, the number of times the noise position exceeded the threshold of 55 dB were low, even before the lockdown. Since Ballymun Library (ID #2) is located close to the airport and M50, and Chancery Park (ID #4) is located in the city center, both the locations receive considerably high noise pollution, which reduced from 97-98% to 82-83% during lockdown.

Ballyfermot Civic Office (ID #1), Rowing Club (ID #5) and Navan Road (ID #9) had the highest reduction of high hourly average noise (47-63%) during locadown. It can be noted that those three stations are located in the west Dublin and close to the Phoenix park.

Based on the analysis it can be noted that the lockdown had indeed reduced the noise pollution considerably in all the 12 monitoring stations. The relation between noise pollution and variables such as traffic volume, location of river, time of the day etc. are related to each other,

where they have an impact in the noise pollution.

This study investigated the effect of the COVID-19 lockdown on noise pollution in Dublin,

Ireland. The results show a reduction in noise pollution at all 12 noise monitoring stations in Dublin. It is likely that this reduction in noise is due to a decrease in road and air traffic during the lockdown; however, these are not the only factors determing noise levels, and noise due to factors such as weather and natural phenoma do impact measurements at monitoring stations.

This study has several important implications: it demonstrates that a significant (p<0.05) reduction in average and minimum noise can be achieved in a major urban area through a reduction in traffic. Although it may not be feasible to completely eliminate traffic from cities, it shows that alternative modes of transport, such as walking and cycling, should be promoted to improve sustainability of the transport infrastructure. It is well established that noise pollution is associated with ill health; therefore, city-wide reductions in noise could provide important public health benefits. Further analyses of potential health impacts associated with the reduction in noise in Dublin will be carried out in the future. Finally, this study demonstrates the value of a city-wide permanent noise monitoring network and suggests that noise monitoring networks should be implemented in urban areas around the world. 

Hierarchal assessment of noise pollution in urban areas-A case study

The influence of low-frequency noise pollution on the quality of life and place in sustainable cities: A case study from

Tyre/road noise spectrum analysis of ageing low noise pavements

986: Is exposure to traffic-noise associated with the development of gestational diabetes mellitus

Choice valuation of traffic restrictions: Noise, pollution, and congestion preferences. A note

WHO environmental noise guidelines for the European region: a systematic review on environmental noise and effects on sleep

Subjective reactions to traffic noise with regard to some personality traits

Effect of urban morphology on road noise distribution

Urban sprawl: diagnosis and remedies. International regional science review

Noise pollution is pervasive in US protected areas

Evaluation of road traffic noise exposure based on high-resolution population distribution and grid-level noise data

Geoprocessing applied to the assessment of environmental noise: a case study in the city of Sorocaba

Noise mapping and assessing vulnerability in meso level urban environment of Eastern India

Urban indices as environmental noise indicators. Computers, Environment and Urban Systems

Short-term effects of traffic noise on suicides and emergency hospital admissions due to anxiety and depression in Madrid (Spain)

Economic value of traffic noise reduction depending on residents' annoyance level

Environmental noise -'Forgotten' or 'Ignored' pollutant?

Environmental noise inequity in the city of Barcelona

Acoustic background and transport noise in urbanised areas: A note on the relative classification of the city soundscape

La qualification des ambiances sonores urbaines

Noise in cities: urban and transport planning determinants and health in cities

Evaluating public transportation health benefits. Victoria Transport Policy Institute

Factors associated with soundscape experiences in urban green spaces: A case study in Rostock

Relationship between urban green spaces and other features of urban morphology with traffic noise distribution. Urban forestry & urban greening

Noise pollution associated with the operation of the Dar es Salaam International Airport

Spatio-temporal patterns of road traffic noise pollution in Karachi

Noise pollution in national parks: Soundscape and economic valuation

An environmental noise study in the city of Cáceres

Noise pollution and arterial hypertension

Transportation noise and public health outcomes: biological markers and pathologies

Environmental noise pollution: Noise mapping, public health, and policy

An analysis of landscape penetration by road infrastructure and traffic noise. Computers, environment and urban systems

Comparison of classical statistical methods and artificial neural network in traffic noise prediction

Influence of noise exposure on cardiocerebrovascular disease in Korea

Acoustic characteristics of urban streets in relation to scattering caused by building facades

Assessing the effects of noise abatement measures on health risks: A case study in Istanbul

Exposure to road traffic noise: Annoyance, perception and associated factors among Brazil's adult population

Evaluation of traffic noise pollution and attitudes of exposed individuals in working place

The jobs-housing balance and urban commuting

A scale model experiment for the study of sound propagation in urban areas

Evaluation and analysis of the environmental noise of Messina

Long-term associations of modeled and self-reported measures of exposure to air pollution and noise at residence on prevalent hypertension and blood pressure

Pregnancy exposure to wind turbine noise and adverse birth outcomes: a nationwide cohort study

Ambient sound assessment of urban environments: field studies in two French cities

Analysis of sampling methodologies for noise pollution assessment and the impact on the population

Environmental quality assessment in areas used for physical activity and recreation in a city affected by intense urban expansion (Fortaleza-CE, Brazil): Implications for public health policy

Selection of suitable alternatives to reduce the environmental impact of road traffic noise using a fuzzy multi-criteria decision model

Effect of noise stress on cardiovascular system in adult male albino rat: implication of stress hormones, endothelial dysfunction and oxidative stress

Urban traffic noise and the relation to urban density, form, and traffic elasticity

The association of traffic-related air and noise pollution with maternal blood pressure and hypertensive disorders of pregnancy in the HOME study cohort

Fulltime exposure to occupational noise during pregnancy was associated with reduced birth weight in a nationwide cohort study of Swedish women

Do quiet areas afford greater health-related quality of life than noisy areas?

Hunting at the highway: traffic noise reduces foraging efficiency in acoustic predators

Assessing the influence of urban geometry on noise propagation by using the sky view factor

WHO environmental noise guidelines for the European region: a systematic review on environmental noise and permanent hearing loss and tinnitus

Noise pollution: non-auditory effects on health

Urban spatial structure and commute duration: An empirical study of China

Considerations for evaluating green infrastructure impacts in microscale and macroscale air pollution dispersion models

Long-term traffic air and noise pollution in relation to mortality and hospital readmission among myocardial infarction survivors

Traffic/road noise mitigation under modified asphalt pavements

The association between noise exposure and blood pressure and ischemic heart disease: a meta-analysis

Benchmarking the noise-oriented efficiency of major European airports: A directional distance function approach

The amenity cost of road noise

United Nations road safety collaboration: a handbook of partner profiles

Night noise guide lines for Europe

Night noise guide lines for Europe

Building environmental assessment schemes for rating of IAQ in sustainable buildings

Examining the associations between urban built environment and noise pollution in high-density high-rise urban areas: A case study in Wuhan

A survey of urban noise annoyance in a large Brazilian city: the importance of a subjective analysis in conjunction with an objective analysis

Impact of megacity jobs-housing spatial mismatch on commuting behaviors: A case study on central districts of Shanghai

Analysis of traffic noise distribution and influence factors in Chinese urban residential blocks

Average hourly traffic count within 500 meters radius surrounding each station in the night, day, and evening time obtained from DCC SCATS