Exposure to fine particulate matter in the New York City subway system during home-work commute

Shams Azad; Pau Ferrer-Cid; Masoud Ghandehari

doi:10.1371/journal.pone.0307096

Abstract

The New York City (NYC) subway system accommodates 5.5 million daily commuters, and the environment within the subway is known to have high concentrations of fine particulate matter (PM_2.5) pollution. Naturally, subway air pollution varies among individuals according to their mobility patterns, introducing the possibility of inequality in PM_2.5 exposure. This study aims to evaluate individual and community-level exposure to subway PM_2.5. We simulated the intracity home-to-work trip patterns using the Longitudinal Employer-Household Dynamics (LEHD) records of 3.1 million working commuters across 34,169 census blocks in four boroughs (Manhattan, Brooklyn, Queens, and the Bronx) of NYC. We incorporated the on-platform and on-train measured PM_2.5 concentration data for the entire subway system. The mean underground platform concentration in the city was 139 μg/m³ with a standard deviation of 25 μg/m³, while the on-train concentration when underground was 99 μg/m³ with a standard deviation of 21 μg/m³. Using a network model, we determined the exposure of individual commuters during their daily home-work trips. We quantified the mean per capita exposure at the census block level by considering the proportion of workers within the blocks who rely on the subway for their work commute. Results indicate statistically significant weak positive correlation between elevated subway PM_2.5 exposure and economically disadvantaged and racial minority groups.

Citation: Azad S, Ferrer-Cid P, Ghandehari M (2024) Exposure to fine particulate matter in the New York City subway system during home-work commute. PLoS ONE 19(8): e0307096. https://doi.org/10.1371/journal.pone.0307096

Editor: Qing-Chang Lu, Chang’an University, CHINA

Received: July 21, 2023; Accepted: June 28, 2024; Published: August 7, 2024

Copyright: © 2024 Azad et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Data Availability: All data related to this manuscript can be accessed on Zenodo via the following link: https://zenodo.org/records/10157151.

Funding: M.G. and S.A. received funding from National Science Foundation (award number 1856032), and C2SMART Center, with a grant from the U.S. Department of Transportation’s University Transportation Centers Program under Grant Number 69A3551747124. The U.S. Government assumes no liability for the contents or use thereof. The funders had no role in study design, data collection and analysis, decisions to publish, or preparation of the manuscript.

Competing interests: The authors have declared that no competing interests exist.

Introduction

Particulate matter (PM) is a complex mixture of solid and liquid particles comprising a range of inorganic and organic chemicals [1]. PM_2.5 refers to the PM with an aerodynamic diameter equal to or smaller than 2.5 μm [2]. Due to their small size, these particles can remain suspended in the air for long periods, and when inhaled, they can easily enter the bloodstream. This can cause short-term and long-term health complications, including cardiovascular, respiratory, metabolic, and neurological disorders [3–13]. Approximately 4.1 million people prematurely die worldwide each year due to exposure to PM_2.5 [14].

For the last few decades, cities worldwide have promoted public transportation systems to reduce traffic congestion and improve air quality. These measures have helped reduce emissions and improve city ambient outdoor air quality [15–18]. Although subway systems effectively reduce ambient air pollution by decreasing the number of fossil fuel-powered vehicles on the road, the air quality inside the subway system is poor. This is largely due to elevated concentrations of PM_2.5 with high concentrations of iron [19–28]. Metal-rich particles in subway systems are mostly generated by the wear and friction of brakes and between rails and wheels [29]. Contributing factors for the high PM_2.5 concentrations in subways include train frequency, station depth, ventilation, age of the subway system, piston effect, and others. [30–33].

When assessing health impacts, the total exposure and the inhaled dose are commonly used as representative measures of exposure to pollutants such as PM_2.5 in indoor and outdoor settings [34–42]. Several studies have been carried out on personal exposure in environments such as the subway and bus systems of a city, using the inhaled dose of PM_2.5 [34, 37, 43]. The total exposure and the inhaled dose depend on the mean inhalation rate during a trip, the mean concentration during a trip, and the trip duration [35, 36, 38]. In complex cases, exposure has been calculated as the sum of exposures in different microenvironments, such as inside a subway train, waiting on a platform, or walking [38]. To evaluate the exposure city residents face in transportation systems, network-based models are generally used to simulate mobility patterns between different parts of a city, following the most common requirement and the fastest route to estimate the air pollution to which commuters have been exposed when they have traveled using that route [41, 42, 44, 45]. For instance, [41] calculated the shortest routes for different locations to simulate mobility patterns and compute the total exposure associated with those trips.

Existing research on subway PM_2.5 has focused on a variety of topics, including measuring air pollution concentrations in subway systems [24, 46–54], identifying the factors that contribute to air pollution and mitigation strategies [55–58], and assessing the health effects of subway PM_2.5 exposure [59–63]. However, no studies to date have investigated inequality in subway PM_2.5 exposure at the community level while accounting for daily intracity subway mobility. It is important to note that not everyone is equally exposed to subway air pollution, even if the system as a whole is polluted. Individual exposure is influenced by the frequency and duration of subway use, which introduces the potential for inequality in subway-related exposure. As a result, we hypothesize that economically disadvantaged communities and racial minority groups living farther from the city center may take more frequent and longer subway trips, thus disproportionately exposed to subway PM_2.5. The main objective of this study is to assess individual and community level PM_2.5 exposure and uncover any disparities in subway-related exposure. To achieve this goal, we carry out our analysis in three distinct phases. In the first phase, we focus on measuring PM_2.5 concentrations across the entire subway system. This phase builds upon our prior investigation [26], which examined PM_2.5 concentrations on subway trains and platforms within nine subway lines. In this study, we expanded our research to include the assessment of particle concentrations on an additional ten subway lines, resulting in a comprehensive analysis of on-train concentrations for a total of 19 subway lines and on-platform concentrations for 608 subway station platforms located within 429 stations. The second phase involves quantifying PM_2.5 exposure associated with subway usage at the individual and community levels. To accomplish this, we utilized LEHD origin-destination (OD) data [64] to simulate the daily commutes of 3.1 million working individuals on 34,169 census blocks in New York City. We constructed a network model to analyze mobility patterns and calculate the corresponding subway PM_2.5 exposure for each origin and destination. In the third phase, we conducted correlative analyses to highlight any disparities in exposure resulting from the use of the subway system in NYC.

2. Materials and methods

2.1 Study region

New York City comprises five boroughs: Manhattan, Brooklyn, Queens, the Bronx, and Staten Island. Staten Island has a standalone subway line not connected to the rest of the city’s subway network. This study focused on the four boroughs with the connected subway system. The working population within these four boroughs is approximately 3.6 million, among which 3.1 million live and work in the city, while the remaining 0.5 million commute outside the city for work [64]. Our study focuses specifically on the mobility of work-home trips for these 3.1 million workers who reside and work within the study area.

Fig 1(a) shows the distribution of jobs across the city. Manhattan accommodates over 2.5 million jobs within its boundaries. Most jobs in this borough are concentrated in downtown and midtown. Approximately 71% of workers who live in Manhattan also work within Manhattan. However, Brooklyn, Queens, and the Bronx workers often have to commute to other boroughs for employment. Only 35% of workers in Brooklyn, 29% in Queens, and 22% in the Bronx work within the same borough in which they reside. Fig 1(b) shows the distribution of workers across the city based on their residence. Lastly, Fig 1(c) presents a detailed representation of the home-job dynamics at the borough level, categorizing workers into three distinct groups: those who live and work within the same borough, those who live in one borough and commute to work in another borough, and the total number of jobs available in each borough.

Download:

Fig 1. Home-job mobility dynamics.

(a) Heatmap showing the number of jobs, (b) Shows a heatmap of where workers live. The kernel density estimation is done with 1-km bandwidth and 100-meter pixel size. (c) Displays the distribution of workers’ residential and workplace locations at the borough level, along with the number of jobs in each borough.

https://doi.org/10.1371/journal.pone.0307096.g001

2.2 Workers’ intracity mobility

We used the 2019 LEHD Origin-Destination (OD) dataset [64] to map the mobility of workers traveling from their homes to their places of work within the city. The LEHD dataset provides OD data at the census block level. This data also provides information on the number of workers residing in each block and their workplaces. According to the 2019 census block boundary [65], there are 34,169 blocks in our study area. Of these, 25,694 census blocks have at least one worker in the city. On average, 119 workers live in each block, with a standard deviation of 135. We have used data for 2019 because, in our analysis, we wanted to exclude the changes in mobility patterns following the onset of the COVID-19 pandemic in 2020 [66, 67]

Workers living in one block often need to travel to other blocks for work. Those who live further away from their job take longer subway rides to get to work, increasing their exposure to high subway PM_2.5. Fig 2(a) and 2(b) shows the home-to-job mobility patterns of two examples (Block X Fig 2a and YFig 2b). The red star represents the origin block, with block ‘X’ in the Bronx accommodating 965 workers and block ‘Y’ in midtown Manhattan accommodating 732 workers. The heatmap in the figure shows where these workers go to work. For block ‘X’, where there are not enough jobs around, the heatmap indicates that most workers living in block ‘X’ travel to midtown Manhattan, where most jobs are concentrated. Some even travel to downtown Manhattan and Brooklyn, with very few working in the Bronx, which is their home borough. In contrast, block ‘Y’ is located in midtown Manhattan, and the workers who live in this block work in midtown and downtown Manhattan, which is very close to their residence. Unlike block ‘X’ workers, block ‘Y’ workers do not need to travel far. As a result, they are less vulnerable to subway exposure. We have made a web platform with this LEHD OD data which can be utilized to investigate the workers’ mobility at the census block level. This platform can be accessed through this link: https://bit.ly/3MK7B4r.

Download:

Fig 2. Home-job trips at the census block level.

(a) Heatmap Home to job mobility for block ‘X’ located in the Bronx, (b) Heatmap Home to job mobility for block ‘Y’ located in Manhattan, Kernel density estimation is done with 1-km bandwidth and 100-meter pixel size. (c) Number of destinations blocks from each origin block for 10 workers. (d) Percent of workers using public transportation for work.

https://doi.org/10.1371/journal.pone.0307096.g002

When a census block has a significant number of job opportunities, many workers who live there may not need to travel to other blocks for work. In such cases, some workers live and work in the same block, which can significantly reduce the necessity of subway travel for work. Fig 2(c) displays the mobility patterns of workers, which are determined by the number of blocks they travel to for work, measured at a ratio of ten workers per block. The green color-coded blocks in the figure indicate that, for every ten workers, at least three workers work within the same block they live in, reducing the need to travel to other blocks. In contrast, the red shaded blocks show that all ten workers need to travel to other blocks to work, increasing the need for travel and potential exposure to subway air pollution.

ACS provides data on the proportion of workers who use public transit for their commute at the census tract level (2057 census tracts in the study area); this includes subway and public buses [68]. While we do not have the exact number of subway users at census block level, we have approximated that the majority of workers who use public buses likely use them to get to the closest subway station, and therefore, most of them eventually use the subway. We disaggregated the census tract-level transit usage data to the census block level.

Fig 2(d) illustrates the percentage of workers who use the subway for their home-to-job commute. The map shows that workers living in midtown and downtown Manhattan use the subway less frequently compared to those residing in other areas. This indicates the people who live in these areas may use private transportation or carpool services or live in close proximity to their workplace. However, subway usage is also low in the outskirts of Queens. This is likely due to the lack of nearby subway stations in those places, which forces workers to rely on private transportation to get to work.

2.3 Measurement of PM_2.5 concentration in the subway system

This study used sampling data from our previous study [26], which offered information on PM_2.5 concentration on subway station platforms and inside train cabins for nine subway lines. However, to incorporate the entire NYC subway system, this study conducted additional measurements of PM_2.5 concentration for stations and train cabins, which were not covered by the previous study.

We measured PM_2.5 concentration on the station platforms and inside train cabins by taking a round trip of each subway line. First, we measured on-train concentration by co-locating real-time and gravimetric instruments while the train moved from the first station to the last station on its route. On the return journey, we got off at each station along the line and collected samples of air pollution concentrations on the platform until the next train arrived on the same line. We spent around 5–15 min on each platform before boarding the train to the next station. For real-time PM_2.5 measurements, we utilized Nephelometric-based DataRAM pDR 1500 units (pDR) manufactured by Thermo Fisher Scientific Inc. PDRs were equipped with a 2.5 μm diameter cut point inlet cyclone and underwent calibration with gravimetric PM_2.5 concentrations. The pDRs collected real-time measurements at 1-s intervals and were zeroed with HEPA-filtered air before the start of each sampling run. The mean concentration for each platform is determined by calculating the mean value of measurements taken at 1-s intervals over a duration of 5 to 15 min, which depends on how long the investigators remained on the platform for sampling. We also report the on-train concentration for each link, which represents the segment between two stations. This is calculated by determining the mean concentration inside the train car from the moment the doors close at one station until they open at the next station. During sampling, we recorded the times of door opening and closing at stations, as well as boarding and disembarking times for each station. During post-processing, we filtered the real-time data using these recorded times to calculate the mean on-platform concentration in each station and on-train concentrations for each link. We used MTA defined publicly available subway station and route data to visualize the concentration data on maps. The on-train and on-platform concentration for the NYC subway system is shown in Fig 4.

2.4 Network model

In order to evaluate the exposure, it is necessary to profile the mobility patterns from one block to another. Therefore, it is reasonable to assume that commuters use the fastest route to get from one subway station to another, so a model of the NYC subway system is needed to calculate the shortest paths [69]. Therefore, in order to estimate mobility patterns, the shortest paths between city blocks can be calculated [41, 42, 44]. A very common approach in the modeling of transport networks consists of modeling a transport system by means of a network [36, 45, 70–72]. A network G = {V, E, W} can be used to represent a subway system, where V = {v₁,…, v_N} is the set of graph nodes, where N is the number of network nodes. E = {(v_i, v_j): v_i, v_j ∈ V} is the set of edges, represented by a set of vertex tuples, where e_ij ∈ E represents an edge connecting nodes v_i and v_j, i.e., a path between two stations. Finally, is the matrix of costs, where W_ij is the cost associated to edge e_ij. There are different variants that can be used depending on the network, in the case of the subway, it makes sense to use an undirected graph, where the existence of a subway line in one direction implies the existence of the line in the opposite direction, i.e., e_ij implies the existence of e_ji. Moreover, there is an additional challenge with subway networks, which is the existence of multiple lines between different stations as well as the possibility of commuting from one line to another and transferring from one subway station to another.

In this study, we create a network G where the nodes represent line platforms within the different stations, i.e., each node represents the specific platform of subway station’s line. The set of edges E is formed by edges connecting stations along the same line E_l, edges connecting line platforms within the same station E_p, and edges connecting line platforms from different stations where there is the possibility of transferring E_t, so E = E_l ∪ E_p ∪ E_t. We have decided to use this network approach to penalize possible line changes within the same subway station. Fig 3 shows an example of how different stations are represented by the network as well as the different station transfers and line transfers.

Download:

Fig 3. Network model architecture.

(a) Stations are represented by their corresponding line platforms, e.g., the 86^th Street Station has three different lines, and edges e_p mark the cost of transferring from one line to another within the same station. (b) Shows four different line platforms from different stations that are reachable by a transfer, denoted by the edges e_t. As illustrated, there are three different stations in Canal Street where interstation transfers can be made.

https://doi.org/10.1371/journal.pone.0307096.g003

Finally, costs W_ij must be assigned to the different connections between nodes, so as to reflect the cost of going from one station to another. Usually, the most important metric for commuters is time, so we can define the cost W_ij as follows, (1) Where is the distance between stations i and j, is the mean train speed, is the time required to commute between lines within the same station, and is the transfer time. The mean train speed is set to 28 km/h [73]. The waiting times in the NYC subway system can range from 3–5 min [74]. Thus, for our calculations, we set the time required to change platforms to 5 min, while the time required to transfer from one station to another is set to 8 min, i.e., 3 min for transferring and 5 min for waiting.

With the network G created and the different weights W_ij assigned to the corresponding edges, Dijkstra’s algorithm is used to calculate the shortest path between two stations, thus mimicking the mobility patterns of commuters [36]. Dijkstra’s computational cost is often set to O (|E| + |V|log|V|) [37, 38, 75].

2.5 PM_2.5 exposure calculations

2.5.1 Mean per capita exposure at the census block level.

We assessed subway PM_2.5 exposure by considering both the concentration of PM_2.5 within the subway system and the duration of exposure. When individuals use the subway to commute to work, they typically spend time waiting on the station platform before boarding the train [74]. To calculate the exposure at the first station, we can multiply the on-platform PM_2.5 concentration by the waiting time at the station. Subsequently, workers embark on the train and travel for several minutes, which could potentially be the longest part of their journey, depending on the distance covered. The exposure experienced on the train can be determined by multiplying the time spent on the train by the mean PM_2.5 concentration measured onboard. During their commute, workers may also need to transfer trains. These transfers can involve waiting on the same platform as the initial train or walking to a different platform. To calculate the exposure during transfers, we consider the PM_2.5 concentration at the transferring station and the duration of the transfer. It is possible for workers to have multiple transfers or train changes throughout their commute. We considered a constant 5-min waiting time for boarding and 3 mins for the exit station. The total exposure E can be expressed using the following notation: (2) Where,

E_a,b = Exposure for moving from block ‘a’ to ‘b’

t₀ = Waiting time at station i. This is the boarding station in block ‘a’

C_i = On-platform PM_2.5 concentration on station i

C_i,j = On-train PM_2.5 concentration between station i and station j

L = Set of subway line segments traveled for a trip

P = Set of transfers within same station for a trip

TR = Set of interstation transfers for a trip

d_i,j = Distance between station i and station j

v_avg = Mean train speed

t_f = Exiting time at last station f

In this study, we employed the network model described in Section 2.4 to identify the travel routes encompassing boarding, transfers, and destination stations, as well as the specific train lines used for commuting between blocks. Each train route corresponds to a measured PM_2.5 concentration (as shown in Fig 4) and time, allowing us to calculate the overall exposure. The total travel time (TT) for a trip can be computed as: (3)

In order to determine the mean per capita PM_2.5 exposure per census block, we used the weighted average technique. As illustrated in Fig 2, workers residing in a particular block may commute to various other blocks for work purposes. However, not all workers use public transportation for their daily commute. The percentage of workers using the public transit system to travel to their workplace at the census block level is demonstrated in Fig 2(d). The calculation of the mean per capita exposure for each census block can be expressed mathematically as Eq 4. (4) Where, PE_(a) = Per capita workers’ exposure for a round work-home subway trip

W_a,bi = total workers move from ‘a’ block to ‘b_i’ block for work, (i = 1, 2, …, n).

E_a,bi = total PM_2.5 exposure for moving to block ‘a’ to block ‘b_i’

% subway_a = Percent worker in block ‘a’ uses subway to work

Here the E_(a,bi) is calculated by Eq 2, and W_a,bi comes from LEHD OD dataset described in section 2.1.

Download:

Fig 4. PM_2.5 concentration in the subway system.

(a) On-platform PM_2.5 concentration and (b) on-train PM_2.5 concentration.

https://doi.org/10.1371/journal.pone.0307096.g004

2.5.2 Methods for correlation analysis.

To understand the relation between the workers of different races as well as income groups and per capita mean exposure at census block level, we performed the bivariate correlation analysis with the Pearson correlation coefficient that can be determined with the following equation. (5) Where, ρ(x,y) is the Pearson coefficient, Cov(x,y) is the covariance of x and y. σ_x and σ_y are the standard deviation of x and y, respectively. There is a total of 25,694 observations to calculate this correlation where each observation is a census block.

Additionally, we computed the mean PM_2.5 exposure for different race and income groups. This is done with the equation below, (6)

Here, ER is the mean exposure for one race or income group, PE_(i) is the per capita exposure for block i measured with Eq (4), and R_(i) is the population of the race group in block i.

3. Results and discussion

3.1 PM_2.5 concentration in the NYC subway system

Results indicate on-platform underground mean and standard deviation concentration of 139 ± 25 μg/m³ versus aboveground values of 30 ± 11 μg/m³. On-train underground values were measured as 99 ± 21 μg/m³ versus aboveground 25 ± 18 μg/m³. These readings contrast with the 24-hr PM_2.5 exposure guideline value of 15 μg/m³ set by the World Health Organization (WHO), which means 24-hour mean exposures should not exceed 15 μg/m³ more than 3 to 4 days per year [76]. It should be noted that a subway ride is not an entire day; hence, results cannot be directly contrasted with WHO’s mean daily guideline. However, it does provide some context or baseline to understand the extent of exposure during a daily commute. Fig 4(a) shows the on-platform PM_2.5 concentration for 429 stations from 19 subway lines. Fig 4(b) shows the on-train concentration of 19 subway lines. We have created an interactive platform that can be used to calculate personal exposure for any origin and destination within NYC. This can be accessed at https://bit.ly/47kTcDh.

3.2 Quantifying personal exposure with the network model

We have implemented the network-based model above on a server with the measured air pollution concentrations. In this way, users can select two locations on the map (origin and destination) and get the exposure level for their subway trip. The output consists of the approximate travel time and total exposure associated with that subway trip. Fig 5 shows examples of two different trips.

Download:

Fig 5. Personal subway PM_2.5 exposure.

(a) Exposure for a sample trip from the Bronx to Downtown Manhattan (b) Exposure for a sample trip from Midtown to Downtown Manhattan.

https://doi.org/10.1371/journal.pone.0307096.g005

In the example of the trip from the Bronx to Downtown Manhattan, Fig 5 (a), a line change is needed, resulting in a travel time of approximately 48 min for a total exposure of 47.72 μg/m³-hr. On the other hand, the trip from Midtown to Downtown Manhattan, Fig 5(b), shows a case where no transfer is needed, where the travel time is short (17 min), resulting in a total exposure of 26.96 μg/m³-hr.

3.3 Quantifying per capita exposure at the census block level

In addition to personal exposure, we also computed the per-capita census block level exposure to understand the scenario at the community level. In Fig 6, a detailed representation is provided, illustrating the per capita exposure to PM_2.5 at the census block level. The exposure metric used in this context is expressed in units of micrograms per cubic meter per hour (μg/m³-hr). To provide a comprehensive understanding of per capita exposure, let us examine the northwestern part of Queens as an example. In this area, the per capita exposure to subway air pollution appears to be lower compared to other areas (Fig 6). However, workers residing in these outskirts and utilizing the subway system may encounter longer commuting times due to their distance from the job center, such as midtown. Consequently, these workers may experience high subway PM_2.5 exposure. It should be noted that the majority of workers residing in these areas do not rely on the subway for their daily commute (Fig 2d). Therefore, while those who use the subway may face elevated exposure, the overall mean per capita exposure remains lower due to the significant number of workers who do not utilize the subway system. So, it is important to note that per capita exposure is a metric to compare exposure at the community level and is not suitable for measuring individual or personal exposure. For accurately assessing personal exposure during a subway trip, we recommend employing the network model described in section 3.2.

Download:

Fig 6. Mean per capita exposure at census block level.

https://doi.org/10.1371/journal.pone.0307096.g006

3.4 Inequality in subway PM_2.5 exposure

In this section, we have explored the socioeconomic dimensions of subway PM_2.5 exposure by focusing on economic and racial disparities. The downtown and midtown Manhattan areas are mostly inhabited by White workers. These areas serve as the city’s central business districts and major job hubs. These residents live close to their workplace, so they have shorter commuting distances and a lower possibility of extensive subway PM_2.5 exposure. Conversely, upper Manhattan and the Bronx offer comparatively fewer job opportunities and have a higher proportion of Black and Hispanic workers. Southwest Queens and Brooklyn also have a sizable population of Black workers. Due to the limited job opportunities in these areas, many workers commute longer distances to their workplaces. Asian workers, although constituting only around 11% of the total workforce in the city, are concentrated in specific areas, including south-downtown Manhattan (Chinatown). West Brooklyn and East Queens also have significant Asian populations. Fig 7 shows the distribution of workers of different races.

Download:

Fig 7. Spatial distribution of different racial groups in the city at census block level.

(a) percent of White workers, (b) percent of Hispanic workers, (c) percent of Black workers, (d) percent of Asian workers.

https://doi.org/10.1371/journal.pone.0307096.g007

The Pearson correlation coefficients between the percentages of White and Asian workers and subway PM_2.5 exposure are -0.22 and -0.23, respectively. These are statistically significant, but weak negative correlation coefficients indicate that areas with a higher percentage of White and Asian workers may have slightly lower levels of subway-driven PM_2.5 exposure. Conversely, the correlation coefficients between the percentages of Hispanic and Black workers and subway exposure are +0.38 and +0.18, respectively. These weak but statistically significant ((p-value < 0.05) positive correlation coefficients suggest that areas with a higher percentage of Hispanic and Black workers may be associated with high subway PM_2.5 exposure.

We also quantified the mean PM_2.5 exposure for different race groups. Fig 8 shows the mean per capita exposure for different race groups. The numbers indicate that Asian and White workers are exposed to lower concentrations of subway PM_2.5 compared to Black and Hispanic workers. The mean exposure for Asian workers is 51 μg/m³-hr, while White workers have a mean exposure of 56 μg/m³-hr. However, both Black and Hispanic workers experience a mean exposure of 69 μg/m³-hr. This means that Black and Hispanic workers are exposed to 35% higher subway PM_2.5 exposure compared to Asian workers.

Download:

Fig 8. Mean exposure for different race groups.

https://doi.org/10.1371/journal.pone.0307096.g008

Moving on to the analysis of the exposure of different income groups, the correlation analysis reveals a weak positive Pearson correlation coefficient of +0.36 between subway PM_2.5 exposure and the size of the population below the poverty threshold at the census block level, a statistically significant positive relation. Looking at it from a different angle, a statistically significant negative Pearson correlation coefficient of -0.23 was found between subway exposure and median family income. This weak negative correlation indicates that higher income levels have a chance to associate with lower levels of subway PM_2.5 exposure. A note on the income data: the available poverty and income data from the U.S. Census Bureau is reported at the census tract level. In order to conduct our correlative analysis, we spatially disaggregated the census tract data to the census block level, allowing us to examine the relationship between subway exposure and socioeconomic factors at a more granular level.

The mean per capita subway PM_2.5 exposure within a community can be influenced by the percentage of workers relying on the subway system. Higher dependence on the subway corresponds to a higher mean per capita exposure to PM_2.5. Fig 9(a) shows the relationship between exposure and income and exposure and poverty in box plots. Here, first we divide the census block level income and poverty data into deciles (divide data into ten equal parts), and the distribution of exposure in each decile is shown in a box in the plot. The horizontal line inside the box represents the median exposure value for each group and the whiskers extending from the box represent the minimum and maximum values of the data, excluding any outliers. From the figure, it is evident that the lowest income group in the city, with a median family income between 12,000 and 39,000, experiences the highest levels of subway PM_2.5 exposure. As the income level increases, there is a gradual decline in exposure levels until reaching the highest income groups. Fig 9(b) reveals a clear relationship between the percentage of individuals living below the poverty line and subway PM_2.5 exposure. Areas with a low percentage of people below the poverty line (less than 4%) exhibit the lowest levels of subway PM_2.5 exposure. However, as the percentage of people living below the poverty line increases, the exposure levels also gradually rise. Fig 9(c) shows the spatial distribution of the median household income in the city. Fig 9(d) shows the spatial distribution of the percentage of poverty in the city.

Download:

Fig 9. Mean exposure in different income levels.

(left) Box plot shows the exposure level in different income groups, (right) relation of exposure level with poverty.

https://doi.org/10.1371/journal.pone.0307096.g009

To some extent, the discrepancy in subway usage between income groups can be attributed to high-income workers having access to alternative transportation options, such as private vehicles or carpool services. Fig 2(a) highlights that many workers residing in downtown and midtown Manhattan have the convenience of working in the same block where they live, eliminating the need for subway usage. This stands in contrast to workers residing in other parts of the city who must undertake longer commutes to reach their workplaces. However, there are some notable anomalies; for example, one of the highest household income groups experiences high levels of exposure when using the subway. This is because these census blocks are heavily populated with high-income workers and use highly polluted subway stations, as shown in Fig 4.

Limitations

While the reported high concentration of PM_2.5 is alarming, a health outcomes analysis of that exposure is complex since health outcomes guidelines (by US EPA and WHO) are largely based on particulate matter from fossil fuel combustion. Health impact analysis of inhalation of iron-based particles is needed to contextualize the results presented here.

This study relied on data related to workers’ mobility for their daily commutes from home to work to estimate community-level subway exposure. However, achieving more precise results would have been feasible if it had been possible to include all types of trips, not only home-to-job commutes, and considered all residents, not just the working population. Such enhancement could be achieved through multiagent simulation modeling and is an area to be explored in future research.

Conclusion

Through this study, it was possible to examine the population exposure to PM_2.5 in the NYC subway system. This was accomplished using a network model that incorporated the commuter’s origin-destination data, as well as measured fine particulate matter concentrations throughout the system (platforms and cars), resulting in the mean per capita PM_2.5 exposure of 3.1 million working commuters across 34,169 census blocks in NYC.

The study revealed the presence of exceptionally high concentrations of iron rich fine PM_2.5, where mean concentrations on platforms and train cars were measured to be 10 times and 7 times the 24-hr guideline set by WHO, respectively. We quantified the disparities in the subway PM_2.5 exposure across different racial and income groups. Results indicate weak positive correlation between elevated subway PM_2.5 exposure and racial minority as well as economically disadvantaged groups. Results show statistically significant weak positive correlation between high subway PM_2.5 exposure and lower-income working communities. We also found that Black and Hispanic workers experience 35% and 23% higher PM_2.5 exposure, respectively, compared to Asian and white workers.

One of the primary contributors to this disparity is the difference in commuting patterns observed among subway users. Different workers have distinct origins and destinations, resulting in variations in the stations they pass through during their home-job subway trips. As a consequence, exposure to PM_2.5 can significantly differ based on the specific route taken and the stations encountered along the way.

Additionally, Individuals with longer commutes or those who frequently transfer between subway lines may spend more time in the subway environment, potentially leading to increased exposure to air pollution. We found that certain subway stations exhibit higher concentrations of PM_2.5 than others. Consequently, commuters who spend time at these stations for transferring or boarding are subject to increased exposure.

We also found that the on-train concentration of PM_2.5 increases in certain subway lines when passing through specific underground tunnels. This implies that commuters whose subway routes include these lines and tracks may be exposed to higher concentrations within the train cabins.

The mean per capita subway PM_2.5 exposure within a community can be influenced by the proportion of workers who rely on the subway system for their daily commute. A higher dependence on the subway is associated with a higher mean per capita exposure to PM_2.5. We observed that workers residing in low-income communities tend to have a greater reliance on subways compared to workers in more affluent communities. Therefore, socioeconomic factors play a role in shaping the transportation choices of individuals and subsequently impact their exposure to subway driven PM_2.5.

References

1. Mohapatra K, Biswal SK. Effect of particulate matter (PM) on plants, climate, ecosystem and human Health. International Journal of Advanced Technology in Engineering and Science. 2014.
- View Article
- Google Scholar
2. U.S. EPA. Particulate Matter (PM) Basics. 2022 [cited 20 May 2023]. https://www.epa.gov/pm-pollution/particulate-matter-pm-basics
3. Braithwaite I, Zhang S, Kirkbride JB, Osborn DPJ, Hayes JF. Air pollution (Particulate matter) exposure and associations with depression, anxiety, bipolar, psychosis and suicide risk: A systematic review and meta-analysis. Environ Health Perspect. 2019;127. pmid:31850801
- View Article
- PubMed/NCBI
- Google Scholar
4. Xing YF, Xu YH, Shi MH, Lian YX. The impact of PM2.5 on the human respiratory system. J Thorac Dis. 2016;8: E69–E74. pmid:26904255
- View Article
- PubMed/NCBI
- Google Scholar
5. Castillo MD, Kinney PL, Southerland V, Arno CA, Crawford K, van Donkelaar A, et al. Estimating Intra-Urban Inequities in PM2.5-Attributable Health Impacts: A Case Study for Washington, DC. Geohealth. 2021;5. pmid:34765851
- View Article
- PubMed/NCBI
- Google Scholar
6. Bowe B, Xie Y, Yan YY, Al-Aly Z. Burden of Cause-Specific Mortality Associated with PM2.5 Air Pollution in the United States. JAMA Netw Open. 2019;2. pmid:31747037
- View Article
- PubMed/NCBI
- Google Scholar
7. He J, Deng Z, Saleem NF, Fathy Elsharkawy M, Azoz AM. Effect of Airborne Particulate Matter on Cardiovascular Diseases. 2022.
8. Jo EJ, Lee WS, Jo HY, Kim CH, Eom JS, Mok JH, et al. Effects of particulate matter on respiratory disease and the impact of meteorological factors in Busan, Korea. Respir Med. 2017;124: 79–87. pmid:28284326
- View Article
- PubMed/NCBI
- Google Scholar
9. Ning X, Ji X, Li G, Sang N. Ambient PM2.5 causes lung injuries and coupled energy metabolic disorder. Ecotoxicol Environ Saf. 2019;170: 620–626. pmid:30579162
- View Article
- PubMed/NCBI
- Google Scholar
10. Peng RD, Chang HH, Bell ML, McDermott A, Zeger SL, Samet JM, et al. Coarse Particulate Matter Air Pollution and Hospital Admissions for Cardiovascular and Respiratory Diseases Among Medicare Patients. JAMA. 2008;299: 2172–2179. pmid:18477784
- View Article
- PubMed/NCBI
- Google Scholar
11. Polezer G, Tadano YS, Siqueira H V., Godoi AFL, Yamamoto CI, de André PA, et al. Assessing the impact of PM2.5 on respiratory disease using artificial neural networks. Environmental Pollution. 2018;235: 394–403. pmid:29306807
- View Article
- PubMed/NCBI
- Google Scholar
12. Roberts S, Arseneault L, Barratt B, Beevers S, Danese A, Odgers CL, et al. Exploration of NO2 and PM2.5 air pollution and mental health problems using high-resolution data in London-based children from a UK longitudinal cohort study. Psychiatry Res. 2019;272: 8–17. pmid:30576995
- View Article
- PubMed/NCBI
- Google Scholar
13. Maciejczyk P, Chen LC, Thurston G. The Role of Fossil Fuel Combustion Metals in PM2.5 Air Pollution Health Associations. Atmosphere 2021, Vol 12, Page 1086. 2021;12: 1086.
- View Article
- Google Scholar
14. World Bank. The Global Health Cost of Ambient PM2.5 Air Pollution. 2020.
15. Xiao D, Li B, Cheng S. The effect of subway development on air pollution: Evidence from China. J Clean Prod. 2020;275.
- View Article
- Google Scholar
16. Li Z, Che W, Frey HC, Lau AKH. Factors affecting variability in PM2.5 exposure concentrations in a metro system. Environ Res. 2018;160: 20–26. pmid:28941800
- View Article
- PubMed/NCBI
- Google Scholar
17. Zheng S, Zhang X, Sun W, Wang J. The effect of a new subway line on local air quality: A case study in Changsha. Transp Res D Transp Environ. 2019;68: 26–38.
- View Article
- Google Scholar
18. Li S, Liu Y, Purevjav AO, Yang L. Does subway expansion improve air quality? J Environ Econ Manage. 2019;96: 213–235.
- View Article
- Google Scholar
19. Vilcassim MJR, Thurston GD, Peltier RE, Gordon T. Black carbon and particulate matter (PM2.5) concentrations in New York city’s subway stations. Environ Sci Technol. 2014;48: 14738–14745. pmid:25409007
- View Article
- PubMed/NCBI
- Google Scholar
20. Luglio DG, Katsigeorgis M, Hess J, Kim R, Adragna J, Raja A, et al. Pm2:5 concentration and composition in subway systems in the Northeastern United States. Environ Health Perspect. 2021;129. pmid:33565894
- View Article
- PubMed/NCBI
- Google Scholar
21. Smith JD, Barratt BM, Fuller GW, Kelly FJ, Loxham M, Nicolosi E, et al. PM2.5 on the London Underground. Environ Int. 2020;134. pmid:31787325
- View Article
- PubMed/NCBI
- Google Scholar
22. Pan S, Du S, Wang X, Zhang X, Xia L, Liu J, et al. Analysis and interpretation of the particulate matter (PM10 and PM2.5) concentrations at the subway stations in Beijing, China. Sustain Cities Soc. 2019;45: 366–377.
- View Article
- Google Scholar
23. Chang L, Chong WT, Wang X, Pei F, Zhang X, Wang T, et al. Recent progress in research on PM2.5in subways. Environ Sci Process Impacts. 2021;23: 642–663. pmid:33889885
- View Article
- PubMed/NCBI
- Google Scholar
24. Azad S, Luglio DG, Gordon T, Thurston G, Ghandehari M. Particulate matter concentration and composition in the New York City subway system. Atmos Pollut Res. 2023; 101767. pmid:37275568
- View Article
- PubMed/NCBI
- Google Scholar
25. Chillrud SN, Epstein D, Ross JM, Sax SN, Pederson D, Spengler JD, et al. Elevated Airborne Exposures of Teenagers to Manganese, Chromium, and Iron from Steel Dust and New York City’s Subway System. Environ Sci Technol. 2004;38: 732–737. pmid:14968857
- View Article
- PubMed/NCBI
- Google Scholar
26. Grass DS, Ross JM, Family F, Barbour J, James Simpson H, Coulibaly D, et al. Airborne particulate metals in the New York City subway: A pilot study to assess the potential for health impacts. Environ Res. 2010;110: 1–11. pmid:19926083
- View Article
- PubMed/NCBI
- Google Scholar
27. Cartenì A, Cascetta F. Particulate matter concentrations in a high-quality rubber-tyred metro system: the case study of Turin in Italy. International Journal of Environmental Science and Technology. 2018;15: 1921–1930.
- View Article
- Google Scholar
28. Johansson C, Johansson A. Particulate matter in the underground of Stockholm. Atmos Environ. 2003;37: 3–9.
- View Article
- Google Scholar
29. Jung HJ, Kim BW, Malek MA, Koo YS, Jung JH, Son YS, et al. Chemical speciation of size-segregated floor dusts and airborne magnetic particles collected at underground subway stations in Seoul, Korea. J Hazard Mater. 2012;213–214: 331–340. pmid:22381374
- View Article
- PubMed/NCBI
- Google Scholar
30. Martins V, Moreno T, Minguillón MC, Van Drooge BL, Reche C, Amato F, et al. Origin of inorganic and organic components of PM2.5 in subway stations of Barcelona, Spain. Environmental Pollution. 2016;208: 125–136. pmid:26189044
- View Article
- PubMed/NCBI
- Google Scholar
31. Wen Y, Leng J, Shen X, Han G, Sun L, Yu F. Environmental and Health Effects of Ventilation in Subway Stations: A Literature Review. International Journal of Environmental Research and Public Health 2020, Vol 17, Page 1084. 2020;17: 1084. pmid:32046319
- View Article
- PubMed/NCBI
- Google Scholar
32. Targino AC, Krecl P, Brimblecombe P, Oukawa GY, Danziger Filho JE, Moreno FL. Spatio-temporal variability of airborne particulate matter in the São Paulo subway. Build Environ. 2021;189: 107526.
- View Article
- Google Scholar
33. Delgado-Saborit JM, Lim S, Hickman A, Baker C, Barratt B, Cai X, et al. Factors affecting occupational black carbon exposure in enclosed railway stations. Atmos Environ. 2022;289: 119301.
- View Article
- Google Scholar
34. Zheng J, Qiu Z, Gao HO, Li B. Commuter PM exposure and estimated life-expectancy loss across multiple transportation modes in Xi’an, China. Ecotoxicol Environ Saf. 2021;214. pmid:33690005
- View Article
- PubMed/NCBI
- Google Scholar
35. Motlagh NH, Zaidan MA, Fung PL, Lagerspetz E, Aula K, Varjonen S, et al. Transit pollution exposure monitoring using low-cost wearable sensors. Transp Res D Transp Environ. 2021;98.
- View Article
- Google Scholar
36. Wang Y, Wu Y, Li Z, Liao K, Li C, Song G. Route planning for active travel considering air pollution exposure. Transp Res D Transp Environ. 2022;103.
- View Article
- Google Scholar
37. Liu Y, Zhong H, Liu K, Oliver Gao H, He L, Xu R, et al. Assessment of personal exposure to PM for multiple transportation modes. Transp Res D Transp Environ. 2021;101.
- View Article
- Google Scholar
38. Morales Betancourt R, Galvis B, Rincón-Riveros JM, Rincón-Caro MA, Rodriguez-Valencia A, Sarmiento OL. Personal exposure to air pollutants in a Bus Rapid Transit System: Impact of fleet age and emission standard. Atmos Environ. 2019;202: 117–127.
- View Article
- Google Scholar
39. Kolluru SSR, Patra AK. Personal exposures to PM during short distance highway travel in India. Transp Res D Transp Environ. 2020;81.
- View Article
- Google Scholar
40. Saeedi R, Jazani RK, Khaloo SS, Ardeh SA, Fouladi-Fard R, Nikukalam H. Risk assessment of occupational and public exposures to airborne particulate matter arising from a subway construction site in Tehran, Iran. Air Qual Atmos Health. 2021.
- View Article
- Google Scholar
41. Lu M, Schmitz O, Vaartjes I, Karssenberg D. Activity-based air pollution exposure assessment: Differences between homemakers and cycling commuters. Health Place. 2019;60. pmid:31675651
- View Article
- PubMed/NCBI
- Google Scholar
42. Li Q, Liang S, Xu Y, Liu L, Zhou S. Assessing personal travel exposure to on-road PM2.5 using cellphone positioning data and mobile sensors. Health Place. 2022;75. pmid:35443227
- View Article
- PubMed/NCBI
- Google Scholar
43. Roy D, Lyou ES, Kim J, Lee TK, Park J. Commuters health risk associated with particulate matter exposures in subway system—Globally. Build Environ. 2022;216.
- View Article
- Google Scholar
44. Ma X, Longley I, Gao J, Salmond J. Assessing schoolchildren’s exposure to air pollution during the daily commute—A systematic review. Science of the Total Environment. Elsevier B.V.; 2020. pmid:32783874
- View Article
- PubMed/NCBI
- Google Scholar
45. Mahajan S, Tang YS, Wu DY, Tsai TC, Chen LJ. CAR: The Clean Air Routing Algorithm for Path Navigation with Minimal PM2.5 Exposure on the Move. IEEE Access. 2019;7: 147373–147382.
- View Article
- Google Scholar
46. Luglio DG, Katsigeorgis M, Hess J, Kim R, Adragna J, Raja A, et al. PM2.5 Concentration and Composition in Subway Systems in the Northeastern United States. Environ Health Perspect. 2021;129. pmid:33565894
- View Article
- PubMed/NCBI
- Google Scholar
47. Adams HS, Nieuwenhuijsen MJ, Colvile RN, Mcmullen MAS, Khandelwal P. Fine particle PM personal exposure levels in 2.5 transport microenvironments, London, UK. Sci Total Environ. 2001. pmid:11712603
- View Article
- PubMed/NCBI
- Google Scholar
48. Martins V, Moreno T, Minguillón MC, Amato F, de Miguel E, Capdevila M, et al. Exposure to airborne particulate matter in the subway system. Science of The Total Environment. 2015;511: 711–722. pmid:25616190
- View Article
- PubMed/NCBI
- Google Scholar
49. Querol X, Moreno T, Karanasiou A, Reche C, Alastuey A, Viana M, et al. Variability of levels and composition of PM 10 and PM 2.5 in the Barcelona metro system. Atmos Chem Phys. 2012;12: 5055–5076.
- View Article
- Google Scholar
50. Chang L, Chong WT, Wang X, Pei F, Zhang X, Wang T, et al. Recent progress in research on PM2.5 in subways. Environ Sci Process Impacts. 2021;23: 642–663. pmid:33889885
- View Article
- PubMed/NCBI
- Google Scholar
51. Kamani H, Hoseini M, Seyedsalehi M, Mahdavi Y, Jaafari J, Safari GH. Concentration and characterization of airborne particles in Tehran’s subway system. Environmental Science and Pollution Research. 2014;21: 7319–7328. pmid:24573466
- View Article
- PubMed/NCBI
- Google Scholar
52. Lu S, Liu D, Zhang W, Liu P, Fei Y, Gu Y, et al. Physico-chemical characterization of PM2.5 in the microenvironment of Shanghai subway. Atmos Res. 2015;153: 543–552.
- View Article
- Google Scholar
53. Lee Y, Lee YC, Kim T, Choi JS, Park D. Sources and characteristics of particulate matter in subway tunnels in seoul, korea. Int J Environ Res Public Health. 2018;15. pmid:30424555
- View Article
- PubMed/NCBI
- Google Scholar
54. Roy D, Seo YC, Namgung HG, Kwon SB. Inhalation cancer risk from PM10 in the metropolitan subway stations in Korea. J Transp Health. 2019;14: 100580.
- View Article
- Google Scholar
55. Moreno T, Pérez N, Reche C, Martins V, de Miguel E, Capdevila M, et al. Subway platform air quality: Assessing the influences of tunnel ventilation, train piston effect and station design. Atmos Environ. 2014;92: 461–468.
- View Article
- Google Scholar
56. Tokarek S, Bernis A. An Exemple of Particle Concentration Reduction in Parisian Subway Stations by Electrostatic Precipitation. Environ Technol. 2006;27: 1279–1287. pmid:17203611
- View Article
- PubMed/NCBI
- Google Scholar
57. Martins V, Moreno T, Mendes L, Eleftheriadis K, Diapouli E, Alves CA, et al. Factors controlling air quality in different European subway systems. Environ Res. 2016;146: 35–46. pmid:26717078
- View Article
- PubMed/NCBI
- Google Scholar
58. Kim JB, Kim S, Lee GJ, Bae GN, Cho Y, Park D, et al. Status of PM in Seoul metropolitan subway cabins and effectiveness of subway cabin air purifier (SCAP). Clean Technol Environ Policy. 2014;16: 1193–1200.
- View Article
- Google Scholar
59. Moreno T, Kelly FJ, Dunster C, Oliete A, Martins V, Reche C, et al. Oxidative potential of subway PM2.5. Atmos Environ. 2017;148: 230–238.
- View Article
- Google Scholar
60. Karlsson HL, Nilsson L, Möller L. Subway Particles Are More Genotoxic than Street Particles and Induce Oxidative Stress in Cultured Human Lung Cells. Chem Res Toxicol. 2004;18: 19–23. pmid:15651844
- View Article
- PubMed/NCBI
- Google Scholar
61. Gustavsson P, Bigert C, Pollán M. Incidence of lung cancer among subway drivers in Stockholm. Am J Ind Med. 2008;51: 545–547. pmid:18493925
- View Article
- PubMed/NCBI
- Google Scholar
62. Klepczyńska Nyström A, Svartengren M, Grunewald J, Pousette C, Rödin I, Lundin A, et al. Health effects of a subway environment in healthy volunteers. European Respiratory Journal. 2010;36: 240–248. pmid:20032018
- View Article
- PubMed/NCBI
- Google Scholar
63. Lundström SL, Levänen B, Nording M, Klepczynska-Nyström A, Sköld M, Haeggström JZ, et al. Asthmatics Exhibit Altered Oxylipin Profiles Compared to Healthy Individuals after Subway Air Exposure. PLoS One. 2011;6: e23864. pmid:21897859
- View Article
- PubMed/NCBI
- Google Scholar
64. U.S. Census Bureau. LEHD Origin-Destination Employment Statistics Data (2002–2020). Washington, DC: U.S. Census Bureau, Longitudinal-Employer Household Dynamics Program. 2023 [cited 20 May 2023]. https://lehd.ces.census.gov/data/#qwi.
65. U.S. Census Bureau. 2019 TIGER/Line Shapefiles, prepared by the U.S. Census Bureau. 2019.
66. Azad S, Ghandehari M. Spatio-Temporal Study Of The Determinants Of Residential Satisfaction In New York City During Covid-19 Using Crowdsourced Data. ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences. 2021;VIII-4/W1-2021: 3–10.
- View Article
- Google Scholar
67. Azad S, Ghandehari M. Emissions of nitrogen dioxide in the northeast U.S. during the 2020 COVID-19 lockdown. J Environ Manage. 2022;312: 114902. pmid:35364514
- View Article
- PubMed/NCBI
- Google Scholar
68. U.S. Census Bureau. 2019 American Community Survey 5-year estimates. 2023 [cited 31 Jan 2023]. https://data.census.gov/
69. Kim J, Kwan MP. Beyond commuting: Ignoring individuals’ activity-travel patterns may lead to inaccurate assessments of their exposure to traffic congestion. Int J Environ Res Public Health. 2019;16. pmid:30598024
- View Article
- PubMed/NCBI
- Google Scholar
70. Derrible S. Network centrality of metro systems. PLoS One. 2012;7. pmid:22792373
- View Article
- PubMed/NCBI
- Google Scholar
71. Derrible S, Kennedy C. Network analysis of world subway systems using updated graph theory. Transp Res Rec. 2009; 17–25.
- View Article
- Google Scholar
72. Zou B, Li S, Zheng Z, Zhan BF, Yang Z, Wan N. Healthier routes planning: A new method and online implementation for minimizing air pollution exposure risk. Comput Environ Urban Syst. 2020;80.
- View Article
- Google Scholar
73. Matt Johnson. Average schedule speed: How does Metro compare? 2010.
74. Todd W. Schneider. Using Countdown Clock Data to Understand the New York City Subway. 2018.
75. Cao SJ, Kong XR, Li L, Zhang W, Ye ZP, Deng Y. An investigation of the PM2.5 and NO2 concentrations and their human health impacts in the metro subway system of Suzhou, China. Environ Sci Process Impacts. 2017;19: 666–675. pmid:28338137
- View Article
- PubMed/NCBI
- Google Scholar
76. WHO. WHO Air Quality Guidelines. 2021 [cited 3 May 2024]. https://www.c40knowledgehub.org/s/article/WHO-Air-Quality-Guidelines?language=en_US#:~:text=By%20reducing%20air%20pollution%20levels,3%20%2D%204%20days%20per%20year.

[ref1] 1. Mohapatra K, Biswal SK. Effect of particulate matter (PM) on plants, climate, ecosystem and human Health. International Journal of Advanced Technology in Engineering and Science. 2014.
View Article
Google Scholar

[2] View Article

[3] Google Scholar

[ref2] 2. U.S. EPA. Particulate Matter (PM) Basics. 2022 [cited 20 May 2023]. https://www.epa.gov/pm-pollution/particulate-matter-pm-basics

[ref3] 3. Braithwaite I, Zhang S, Kirkbride JB, Osborn DPJ, Hayes JF. Air pollution (Particulate matter) exposure and associations with depression, anxiety, bipolar, psychosis and suicide risk: A systematic review and meta-analysis. Environ Health Perspect. 2019;127. pmid:31850801
View Article
PubMed/NCBI
Google Scholar

[6] View Article

[7] PubMed/NCBI

[8] Google Scholar

[ref4] 4. Xing YF, Xu YH, Shi MH, Lian YX. The impact of PM2.5 on the human respiratory system. J Thorac Dis. 2016;8: E69–E74. pmid:26904255
View Article
PubMed/NCBI
Google Scholar

[10] View Article

[11] PubMed/NCBI

[12] Google Scholar

[ref5] 5. Castillo MD, Kinney PL, Southerland V, Arno CA, Crawford K, van Donkelaar A, et al. Estimating Intra-Urban Inequities in PM2.5-Attributable Health Impacts: A Case Study for Washington, DC. Geohealth. 2021;5. pmid:34765851
View Article
PubMed/NCBI
Google Scholar

[14] View Article

[15] PubMed/NCBI

[16] Google Scholar

[ref6] 6. Bowe B, Xie Y, Yan YY, Al-Aly Z. Burden of Cause-Specific Mortality Associated with PM2.5 Air Pollution in the United States. JAMA Netw Open. 2019;2. pmid:31747037
View Article
PubMed/NCBI
Google Scholar

[18] View Article

[19] PubMed/NCBI

[20] Google Scholar

[ref7] 7. He J, Deng Z, Saleem NF, Fathy Elsharkawy M, Azoz AM. Effect of Airborne Particulate Matter on Cardiovascular Diseases. 2022.

[ref8] 8. Jo EJ, Lee WS, Jo HY, Kim CH, Eom JS, Mok JH, et al. Effects of particulate matter on respiratory disease and the impact of meteorological factors in Busan, Korea. Respir Med. 2017;124: 79–87. pmid:28284326
View Article
PubMed/NCBI
Google Scholar

[23] View Article

[24] PubMed/NCBI

[25] Google Scholar

[ref9] 9. Ning X, Ji X, Li G, Sang N. Ambient PM2.5 causes lung injuries and coupled energy metabolic disorder. Ecotoxicol Environ Saf. 2019;170: 620–626. pmid:30579162
View Article
PubMed/NCBI
Google Scholar

[27] View Article

[28] PubMed/NCBI

[29] Google Scholar

[ref10] 10. Peng RD, Chang HH, Bell ML, McDermott A, Zeger SL, Samet JM, et al. Coarse Particulate Matter Air Pollution and Hospital Admissions for Cardiovascular and Respiratory Diseases Among Medicare Patients. JAMA. 2008;299: 2172–2179. pmid:18477784
View Article
PubMed/NCBI
Google Scholar

[31] View Article

[32] PubMed/NCBI

[33] Google Scholar

[ref11] 11. Polezer G, Tadano YS, Siqueira H V., Godoi AFL, Yamamoto CI, de André PA, et al. Assessing the impact of PM2.5 on respiratory disease using artificial neural networks. Environmental Pollution. 2018;235: 394–403. pmid:29306807
View Article
PubMed/NCBI
Google Scholar

[35] View Article

[36] PubMed/NCBI

[37] Google Scholar

[ref12] 12. Roberts S, Arseneault L, Barratt B, Beevers S, Danese A, Odgers CL, et al. Exploration of NO2 and PM2.5 air pollution and mental health problems using high-resolution data in London-based children from a UK longitudinal cohort study. Psychiatry Res. 2019;272: 8–17. pmid:30576995
View Article
PubMed/NCBI
Google Scholar

[39] View Article

[40] PubMed/NCBI

[41] Google Scholar

[ref13] 13. Maciejczyk P, Chen LC, Thurston G. The Role of Fossil Fuel Combustion Metals in PM2.5 Air Pollution Health Associations. Atmosphere 2021, Vol 12, Page 1086. 2021;12: 1086.
View Article
Google Scholar

[43] View Article

[44] Google Scholar

[ref14] 14. World Bank. The Global Health Cost of Ambient PM2.5 Air Pollution. 2020.

[ref15] 15. Xiao D, Li B, Cheng S. The effect of subway development on air pollution: Evidence from China. J Clean Prod. 2020;275.
View Article
Google Scholar

[47] View Article

[48] Google Scholar

[ref16] 16. Li Z, Che W, Frey HC, Lau AKH. Factors affecting variability in PM2.5 exposure concentrations in a metro system. Environ Res. 2018;160: 20–26. pmid:28941800
View Article
PubMed/NCBI
Google Scholar

[50] View Article

[51] PubMed/NCBI

[52] Google Scholar

[ref17] 17. Zheng S, Zhang X, Sun W, Wang J. The effect of a new subway line on local air quality: A case study in Changsha. Transp Res D Transp Environ. 2019;68: 26–38.
View Article
Google Scholar

[54] View Article

[55] Google Scholar

[ref18] 18. Li S, Liu Y, Purevjav AO, Yang L. Does subway expansion improve air quality? J Environ Econ Manage. 2019;96: 213–235.
View Article
Google Scholar

[57] View Article

[58] Google Scholar

[ref19] 19. Vilcassim MJR, Thurston GD, Peltier RE, Gordon T. Black carbon and particulate matter (PM2.5) concentrations in New York city’s subway stations. Environ Sci Technol. 2014;48: 14738–14745. pmid:25409007
View Article
PubMed/NCBI
Google Scholar

[60] View Article

[61] PubMed/NCBI

[62] Google Scholar

[ref20] 20. Luglio DG, Katsigeorgis M, Hess J, Kim R, Adragna J, Raja A, et al. Pm2:5 concentration and composition in subway systems in the Northeastern United States. Environ Health Perspect. 2021;129. pmid:33565894
View Article
PubMed/NCBI
Google Scholar

[64] View Article

[65] PubMed/NCBI

[66] Google Scholar

[ref21] 21. Smith JD, Barratt BM, Fuller GW, Kelly FJ, Loxham M, Nicolosi E, et al. PM2.5 on the London Underground. Environ Int. 2020;134. pmid:31787325
View Article
PubMed/NCBI
Google Scholar

[68] View Article

[69] PubMed/NCBI

[70] Google Scholar

[ref22] 22. Pan S, Du S, Wang X, Zhang X, Xia L, Liu J, et al. Analysis and interpretation of the particulate matter (PM10 and PM2.5) concentrations at the subway stations in Beijing, China. Sustain Cities Soc. 2019;45: 366–377.
View Article
Google Scholar

[72] View Article

[73] Google Scholar

[ref23] 23. Chang L, Chong WT, Wang X, Pei F, Zhang X, Wang T, et al. Recent progress in research on PM2.5in subways. Environ Sci Process Impacts. 2021;23: 642–663. pmid:33889885
View Article
PubMed/NCBI
Google Scholar

[75] View Article

[76] PubMed/NCBI

[77] Google Scholar

[ref24] 24. Azad S, Luglio DG, Gordon T, Thurston G, Ghandehari M. Particulate matter concentration and composition in the New York City subway system. Atmos Pollut Res. 2023; 101767. pmid:37275568
View Article
PubMed/NCBI
Google Scholar

[79] View Article

[80] PubMed/NCBI

[81] Google Scholar

[ref25] 25. Chillrud SN, Epstein D, Ross JM, Sax SN, Pederson D, Spengler JD, et al. Elevated Airborne Exposures of Teenagers to Manganese, Chromium, and Iron from Steel Dust and New York City’s Subway System. Environ Sci Technol. 2004;38: 732–737. pmid:14968857
View Article
PubMed/NCBI
Google Scholar

[83] View Article

[84] PubMed/NCBI

[85] Google Scholar

[ref26] 26. Grass DS, Ross JM, Family F, Barbour J, James Simpson H, Coulibaly D, et al. Airborne particulate metals in the New York City subway: A pilot study to assess the potential for health impacts. Environ Res. 2010;110: 1–11. pmid:19926083
View Article
PubMed/NCBI
Google Scholar

[87] View Article

[88] PubMed/NCBI

[89] Google Scholar

[ref27] 27. Cartenì A, Cascetta F. Particulate matter concentrations in a high-quality rubber-tyred metro system: the case study of Turin in Italy. International Journal of Environmental Science and Technology. 2018;15: 1921–1930.
View Article
Google Scholar

[91] View Article

[92] Google Scholar

[ref28] 28. Johansson C, Johansson A. Particulate matter in the underground of Stockholm. Atmos Environ. 2003;37: 3–9.
View Article
Google Scholar

[94] View Article

[95] Google Scholar

[ref29] 29. Jung HJ, Kim BW, Malek MA, Koo YS, Jung JH, Son YS, et al. Chemical speciation of size-segregated floor dusts and airborne magnetic particles collected at underground subway stations in Seoul, Korea. J Hazard Mater. 2012;213–214: 331–340. pmid:22381374
View Article
PubMed/NCBI
Google Scholar

[97] View Article

[98] PubMed/NCBI

[99] Google Scholar

[ref30] 30. Martins V, Moreno T, Minguillón MC, Van Drooge BL, Reche C, Amato F, et al. Origin of inorganic and organic components of PM2.5 in subway stations of Barcelona, Spain. Environmental Pollution. 2016;208: 125–136. pmid:26189044
View Article
PubMed/NCBI
Google Scholar

[101] View Article

[102] PubMed/NCBI

[103] Google Scholar

[ref31] 31. Wen Y, Leng J, Shen X, Han G, Sun L, Yu F. Environmental and Health Effects of Ventilation in Subway Stations: A Literature Review. International Journal of Environmental Research and Public Health 2020, Vol 17, Page 1084. 2020;17: 1084. pmid:32046319
View Article
PubMed/NCBI
Google Scholar

[105] View Article

[106] PubMed/NCBI

[107] Google Scholar

[ref32] 32. Targino AC, Krecl P, Brimblecombe P, Oukawa GY, Danziger Filho JE, Moreno FL. Spatio-temporal variability of airborne particulate matter in the São Paulo subway. Build Environ. 2021;189: 107526.
View Article
Google Scholar

[109] View Article

[110] Google Scholar

[ref33] 33. Delgado-Saborit JM, Lim S, Hickman A, Baker C, Barratt B, Cai X, et al. Factors affecting occupational black carbon exposure in enclosed railway stations. Atmos Environ. 2022;289: 119301.
View Article
Google Scholar

[112] View Article

[113] Google Scholar

[ref34] 34. Zheng J, Qiu Z, Gao HO, Li B. Commuter PM exposure and estimated life-expectancy loss across multiple transportation modes in Xi’an, China. Ecotoxicol Environ Saf. 2021;214. pmid:33690005
View Article
PubMed/NCBI
Google Scholar

[115] View Article

[116] PubMed/NCBI

[117] Google Scholar

[ref35] 35. Motlagh NH, Zaidan MA, Fung PL, Lagerspetz E, Aula K, Varjonen S, et al. Transit pollution exposure monitoring using low-cost wearable sensors. Transp Res D Transp Environ. 2021;98.
View Article
Google Scholar

[119] View Article

[120] Google Scholar

[ref36] 36. Wang Y, Wu Y, Li Z, Liao K, Li C, Song G. Route planning for active travel considering air pollution exposure. Transp Res D Transp Environ. 2022;103.
View Article
Google Scholar

[122] View Article

[123] Google Scholar

[ref37] 37. Liu Y, Zhong H, Liu K, Oliver Gao H, He L, Xu R, et al. Assessment of personal exposure to PM for multiple transportation modes. Transp Res D Transp Environ. 2021;101.
View Article
Google Scholar

[125] View Article

[126] Google Scholar

[ref38] 38. Morales Betancourt R, Galvis B, Rincón-Riveros JM, Rincón-Caro MA, Rodriguez-Valencia A, Sarmiento OL. Personal exposure to air pollutants in a Bus Rapid Transit System: Impact of fleet age and emission standard. Atmos Environ. 2019;202: 117–127.
View Article
Google Scholar

[128] View Article

[129] Google Scholar

[ref39] 39. Kolluru SSR, Patra AK. Personal exposures to PM during short distance highway travel in India. Transp Res D Transp Environ. 2020;81.
View Article
Google Scholar

[131] View Article

[132] Google Scholar

[ref40] 40. Saeedi R, Jazani RK, Khaloo SS, Ardeh SA, Fouladi-Fard R, Nikukalam H. Risk assessment of occupational and public exposures to airborne particulate matter arising from a subway construction site in Tehran, Iran. Air Qual Atmos Health. 2021.
View Article
Google Scholar

[134] View Article

[135] Google Scholar

[ref41] 41. Lu M, Schmitz O, Vaartjes I, Karssenberg D. Activity-based air pollution exposure assessment: Differences between homemakers and cycling commuters. Health Place. 2019;60. pmid:31675651
View Article
PubMed/NCBI
Google Scholar

[137] View Article

[138] PubMed/NCBI

[139] Google Scholar

[ref42] 42. Li Q, Liang S, Xu Y, Liu L, Zhou S. Assessing personal travel exposure to on-road PM2.5 using cellphone positioning data and mobile sensors. Health Place. 2022;75. pmid:35443227
View Article
PubMed/NCBI
Google Scholar

[141] View Article

[142] PubMed/NCBI

[143] Google Scholar

[ref43] 43. Roy D, Lyou ES, Kim J, Lee TK, Park J. Commuters health risk associated with particulate matter exposures in subway system—Globally. Build Environ. 2022;216.
View Article
Google Scholar

[145] View Article

[146] Google Scholar

[ref44] 44. Ma X, Longley I, Gao J, Salmond J. Assessing schoolchildren’s exposure to air pollution during the daily commute—A systematic review. Science of the Total Environment. Elsevier B.V.; 2020. pmid:32783874
View Article
PubMed/NCBI
Google Scholar

[148] View Article

[149] PubMed/NCBI

[150] Google Scholar

[ref45] 45. Mahajan S, Tang YS, Wu DY, Tsai TC, Chen LJ. CAR: The Clean Air Routing Algorithm for Path Navigation with Minimal PM2.5 Exposure on the Move. IEEE Access. 2019;7: 147373–147382.
View Article
Google Scholar

[152] View Article

[153] Google Scholar

[ref46] 46. Luglio DG, Katsigeorgis M, Hess J, Kim R, Adragna J, Raja A, et al. PM2.5 Concentration and Composition in Subway Systems in the Northeastern United States. Environ Health Perspect. 2021;129. pmid:33565894
View Article
PubMed/NCBI
Google Scholar

[155] View Article

[156] PubMed/NCBI

[157] Google Scholar

[ref47] 47. Adams HS, Nieuwenhuijsen MJ, Colvile RN, Mcmullen MAS, Khandelwal P. Fine particle PM personal exposure levels in 2.5 transport microenvironments, London, UK. Sci Total Environ. 2001. pmid:11712603
View Article
PubMed/NCBI
Google Scholar

[159] View Article

[160] PubMed/NCBI

[161] Google Scholar

[ref48] 48. Martins V, Moreno T, Minguillón MC, Amato F, de Miguel E, Capdevila M, et al. Exposure to airborne particulate matter in the subway system. Science of The Total Environment. 2015;511: 711–722. pmid:25616190
View Article
PubMed/NCBI
Google Scholar

[163] View Article

[164] PubMed/NCBI

[165] Google Scholar

[ref49] 49. Querol X, Moreno T, Karanasiou A, Reche C, Alastuey A, Viana M, et al. Variability of levels and composition of PM 10 and PM 2.5 in the Barcelona metro system. Atmos Chem Phys. 2012;12: 5055–5076.
View Article
Google Scholar

[167] View Article

[168] Google Scholar

[ref50] 50. Chang L, Chong WT, Wang X, Pei F, Zhang X, Wang T, et al. Recent progress in research on PM2.5 in subways. Environ Sci Process Impacts. 2021;23: 642–663. pmid:33889885
View Article
PubMed/NCBI
Google Scholar

[170] View Article

[171] PubMed/NCBI

[172] Google Scholar

[ref51] 51. Kamani H, Hoseini M, Seyedsalehi M, Mahdavi Y, Jaafari J, Safari GH. Concentration and characterization of airborne particles in Tehran’s subway system. Environmental Science and Pollution Research. 2014;21: 7319–7328. pmid:24573466
View Article
PubMed/NCBI
Google Scholar

[174] View Article

[175] PubMed/NCBI

[176] Google Scholar

[ref52] 52. Lu S, Liu D, Zhang W, Liu P, Fei Y, Gu Y, et al. Physico-chemical characterization of PM2.5 in the microenvironment of Shanghai subway. Atmos Res. 2015;153: 543–552.
View Article
Google Scholar

[178] View Article

[179] Google Scholar

[ref53] 53. Lee Y, Lee YC, Kim T, Choi JS, Park D. Sources and characteristics of particulate matter in subway tunnels in seoul, korea. Int J Environ Res Public Health. 2018;15. pmid:30424555
View Article
PubMed/NCBI
Google Scholar

[181] View Article

[182] PubMed/NCBI

[183] Google Scholar

[ref54] 54. Roy D, Seo YC, Namgung HG, Kwon SB. Inhalation cancer risk from PM10 in the metropolitan subway stations in Korea. J Transp Health. 2019;14: 100580.
View Article
Google Scholar

[185] View Article

[186] Google Scholar

[ref55] 55. Moreno T, Pérez N, Reche C, Martins V, de Miguel E, Capdevila M, et al. Subway platform air quality: Assessing the influences of tunnel ventilation, train piston effect and station design. Atmos Environ. 2014;92: 461–468.
View Article
Google Scholar

[188] View Article

[189] Google Scholar

[ref56] 56. Tokarek S, Bernis A. An Exemple of Particle Concentration Reduction in Parisian Subway Stations by Electrostatic Precipitation. Environ Technol. 2006;27: 1279–1287. pmid:17203611
View Article
PubMed/NCBI
Google Scholar

[191] View Article

[192] PubMed/NCBI

[193] Google Scholar

[ref57] 57. Martins V, Moreno T, Mendes L, Eleftheriadis K, Diapouli E, Alves CA, et al. Factors controlling air quality in different European subway systems. Environ Res. 2016;146: 35–46. pmid:26717078
View Article
PubMed/NCBI
Google Scholar

[195] View Article

[196] PubMed/NCBI

[197] Google Scholar

[ref58] 58. Kim JB, Kim S, Lee GJ, Bae GN, Cho Y, Park D, et al. Status of PM in Seoul metropolitan subway cabins and effectiveness of subway cabin air purifier (SCAP). Clean Technol Environ Policy. 2014;16: 1193–1200.
View Article
Google Scholar

[199] View Article

[200] Google Scholar

[ref59] 59. Moreno T, Kelly FJ, Dunster C, Oliete A, Martins V, Reche C, et al. Oxidative potential of subway PM2.5. Atmos Environ. 2017;148: 230–238.
View Article
Google Scholar

[202] View Article

[203] Google Scholar

[ref60] 60. Karlsson HL, Nilsson L, Möller L. Subway Particles Are More Genotoxic than Street Particles and Induce Oxidative Stress in Cultured Human Lung Cells. Chem Res Toxicol. 2004;18: 19–23. pmid:15651844
View Article
PubMed/NCBI
Google Scholar

[205] View Article

[206] PubMed/NCBI

[207] Google Scholar

[ref61] 61. Gustavsson P, Bigert C, Pollán M. Incidence of lung cancer among subway drivers in Stockholm. Am J Ind Med. 2008;51: 545–547. pmid:18493925
View Article
PubMed/NCBI
Google Scholar

[209] View Article

[210] PubMed/NCBI

[211] Google Scholar

[ref62] 62. Klepczyńska Nyström A, Svartengren M, Grunewald J, Pousette C, Rödin I, Lundin A, et al. Health effects of a subway environment in healthy volunteers. European Respiratory Journal. 2010;36: 240–248. pmid:20032018
View Article
PubMed/NCBI
Google Scholar

[213] View Article

[214] PubMed/NCBI

[215] Google Scholar

[ref63] 63. Lundström SL, Levänen B, Nording M, Klepczynska-Nyström A, Sköld M, Haeggström JZ, et al. Asthmatics Exhibit Altered Oxylipin Profiles Compared to Healthy Individuals after Subway Air Exposure. PLoS One. 2011;6: e23864. pmid:21897859
View Article
PubMed/NCBI
Google Scholar

[217] View Article

[218] PubMed/NCBI

[219] Google Scholar

[ref64] 64. U.S. Census Bureau. LEHD Origin-Destination Employment Statistics Data (2002–2020). Washington, DC: U.S. Census Bureau, Longitudinal-Employer Household Dynamics Program. 2023 [cited 20 May 2023]. https://lehd.ces.census.gov/data/#qwi.

[ref65] 65. U.S. Census Bureau. 2019 TIGER/Line Shapefiles, prepared by the U.S. Census Bureau. 2019.

[ref66] 66. Azad S, Ghandehari M. Spatio-Temporal Study Of The Determinants Of Residential Satisfaction In New York City During Covid-19 Using Crowdsourced Data. ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences. 2021;VIII-4/W1-2021: 3–10.
View Article
Google Scholar

[223] View Article

[224] Google Scholar

[ref67] 67. Azad S, Ghandehari M. Emissions of nitrogen dioxide in the northeast U.S. during the 2020 COVID-19 lockdown. J Environ Manage. 2022;312: 114902. pmid:35364514
View Article
PubMed/NCBI
Google Scholar

[226] View Article

[227] PubMed/NCBI

[228] Google Scholar

[ref68] 68. U.S. Census Bureau. 2019 American Community Survey 5-year estimates. 2023 [cited 31 Jan 2023]. https://data.census.gov/

[ref69] 69. Kim J, Kwan MP. Beyond commuting: Ignoring individuals’ activity-travel patterns may lead to inaccurate assessments of their exposure to traffic congestion. Int J Environ Res Public Health. 2019;16. pmid:30598024
View Article
PubMed/NCBI
Google Scholar

[231] View Article

[232] PubMed/NCBI

[233] Google Scholar

[ref70] 70. Derrible S. Network centrality of metro systems. PLoS One. 2012;7. pmid:22792373
View Article
PubMed/NCBI
Google Scholar

[235] View Article

[236] PubMed/NCBI

[237] Google Scholar

[ref71] 71. Derrible S, Kennedy C. Network analysis of world subway systems using updated graph theory. Transp Res Rec. 2009; 17–25.
View Article
Google Scholar

[239] View Article

[240] Google Scholar

[ref72] 72. Zou B, Li S, Zheng Z, Zhan BF, Yang Z, Wan N. Healthier routes planning: A new method and online implementation for minimizing air pollution exposure risk. Comput Environ Urban Syst. 2020;80.
View Article
Google Scholar

[242] View Article

[243] Google Scholar

[ref73] 73. Matt Johnson. Average schedule speed: How does Metro compare? 2010.

[ref74] 74. Todd W. Schneider. Using Countdown Clock Data to Understand the New York City Subway. 2018.

[ref75] 75. Cao SJ, Kong XR, Li L, Zhang W, Ye ZP, Deng Y. An investigation of the PM2.5 and NO2 concentrations and their human health impacts in the metro subway system of Suzhou, China. Environ Sci Process Impacts. 2017;19: 666–675. pmid:28338137
View Article
PubMed/NCBI
Google Scholar

[247] View Article

[248] PubMed/NCBI

[249] Google Scholar

[ref76] 76. WHO. WHO Air Quality Guidelines. 2021 [cited 3 May 2024]. https://www.c40knowledgehub.org/s/article/WHO-Air-Quality-Guidelines?language=en_US#:~:text=By%20reducing%20air%20pollution%20levels,3%20%2D%204%20days%20per%20year.

Figures

Abstract

Introduction

2. Materials and methods

2.1 Study region

2.2 Workers’ intracity mobility

2.3 Measurement of PM2.5 concentration in the subway system

2.4 Network model

2.5 PM2.5 exposure calculations

2.5.1 Mean per capita exposure at the census block level.

2.5.2 Methods for correlation analysis.

3. Results and discussion

3.1 PM2.5 concentration in the NYC subway system

3.2 Quantifying personal exposure with the network model

3.3 Quantifying per capita exposure at the census block level

3.4 Inequality in subway PM2.5 exposure

Limitations

Conclusion

References

2.3 Measurement of PM_2.5 concentration in the subway system

2.5 PM_2.5 exposure calculations

3.1 PM_2.5 concentration in the NYC subway system

3.4 Inequality in subway PM_2.5 exposure