Air Pollution NYC
Kevin Huynh and Andres Moreno
This is air pollution and hospital data collected from individual sites all around the New York City area from 2005-2023 [1]. The data set provides info on levels of particulate matter of 2.5 micrometers or less (PM2.5), Ozone (O3) levels, hospital visit and death information related to PM2.5 and Ozone, nitrogen dioxide (NO2) levels, boiler emissions by chemical (PM, NOx, SO2), and other outdoor air toxic chemical levels of Benzene and Formaldehyde.
PM2.5 particles are about 28 times smaller than the diameter of an average human hair, come directly from sources like construction sites, and have adverse health impacts when inhaled [2].
Ground level ozone, which causes “smog”, is known to be caused primary via the reaction of man made volatile organic compounds and nitrogen oxides. When inhaled, Ozone reacts with the respiratory tract, causing adverse effects [3].
NO2 is primary generated from the burning of fuel (like cars), and long term inhalation exposure causes asthma and a higher susceptibility to respiratory infections [4].
SO2 is produced via the combustion of fossils fuels in power plants and industrial facilities and contributes to harmful particulate matter [5].
Benzene is a cancerous chemical used in motor fuels and can also be produced from burning coal or oil [6].
Formaldehyde is currently under research as possible health hazard and is produced via combustion, but also found in many other products such as composite wood, glue, paint, medicines, fertilizers and cosmetic products [7].
This graph was created by looking at all of the data that had a time label starting with “Summer” or “Winter” and thus is a collection of data from across NYC. The box plot sizes are roughly the same even though they they are shifted from each other, meaning that the all of the quartiles have been moved a similar amount from one season to the other with their sizes retaining almost the same size and outlines distributions appearing similar. This indicates that the production and therein measurable levels of NO2 in the NYC area are greater in winter than the summer for all areas of NYC.
This shows the mean and max of ground level ozone collected excluding any data that was classified as city wide. This was done to highlight specific collection that expired high spikes of Ozone readings because only looking at citywide data does not allow for the ability to view locations that had significant peaks. By also taking the mean of this same filtered data, a pattern is exposed between max levels and the mean of all collection sites by year. All but one of the peaks from the max levels graph appear in the mean graph. This indicates that there is a high correlation (if you ignore the outlier peak and dip of 2019-2020) between the max levels of Ozone measured and what the city experiences on average.
This timeline shows that the trajectory of Ozone concentrations in Queens across consecutive summer seasons from 2009 to 2023 were particularly volatile, spiking and plunging from one summer to the next. It shows that Queens hit a historical summer minimum around 2020, only for Ozone levels to aggressively rebound over the next three years, eventually leading to a sharp surge in summer 2023.
Because Ozone is produced/formed when sunlight reacts with industrial and vehicle pollutants, this means that winds are likely pushing the chemical deposits away from Manhattan’s heavy city traffic towards Queens, where it’s then left to bake under the summer heatwaves resulting in the higher Ozone concentrations in Queens compared to Manhattan [8][9].
Some of the Boroughs’ in New York seem to have marginally higher pollutant levels particularly in Manhattan where we see a disproportionately high amount of Nitrogen dioxide (NO2) and Fine particles (PM 2.5), while other boroughs peak heavily on Ozone.
Different neighborhoods suffer from fundamentally different types of pollution. Manhattan has a much larger level of NO2 because of the city’s extreme traffic density and commercial infrastructure. Meanwhile, Ozone peaks in outer boroughs because it’s a secondary pollutant produced when sunlight reacts with chemical emissions over time, drifting away from the city center.
Concentration of fine particles (PM 2.5) appears to have steadily decreased throughout the years in Manhattan. There’s a downward trend showing that concentration of fine particles dropped from the double digits (above 10µg/m3) down to an annual average hovering near 6-7µg/m3, which was cut roughly in half over this period.
This shows that public policy actually works. Decades of passing stricter environmental laws, the phasing out of dirty heating oils (like No.6 in New York buildings), and updating vehicle emissions standards are all paying off in the long-term for public health [10]. The air in NYC is objectively cleaner today than it was a generation ago. But does everyone in NY breathe the same clean air?
Some neighborhoods in New York may receive more effective measures for pollutant control, while other neighborhoods such as Chelsea and Lower Manhattan may not receive as much measures and are therefore highly impacted.
Even though the previous timeline shows that air quality is overall improving, this new plot shows that clean air is still a luxury that’s distributed unevenly. The longer bars represent neighborhoods that have dense commercial boiler usage and perhaps heavy industry [11][12].
This is a graph of individual data collection locations throughout NYC that had info under PM2.5 levels, cardio hospitalizations, respiratory hospitalizations, deaths from PM2.5, and Asthma Emergency related to PM2.5. This data was scaled so that the z-scores could provide the ability to search for patterns in the map related to specific locations. There are no locations that share especially high and low z-scores from the same location (the same row). There are a few locations that share a few columns with significantly high and low z scores within the same location, yet, it’s not across the board. The locations with the highest and lowest PM2.5 levels do not actually have the highest scores for their respective category (column). One reason to explain this trend is that ill individuals could be traveling to other areas in NYC to receive treatment when they have very severe symptoms.
This graph focuses in solely on data collected in Manhattan. Again this data is scaled allowing z-scores to paint the relationships in a consumable way. Here there are two large areas of note: the peak and subsequent fall of PM2.5 deaths from 2005 until the dip around 2014. The high death count in 2005 is accompanied by high levels of outdoor air toxins, and they seem to decrease at the same time, suggesting a correlation. The deaths due to Ozone seem to be flipped, lower in 2005, but highest in 2012 providing a contradiction to the idea that they are all always correlated. The other important aspect of the graph is the spike and drop of asthma hospital admission z-scores that move in tandem with the boiler emission data from 2013 to 2016, also suggesting a correlation.
This graph of Brooklyn was made in the same was as the previous one of Manhattan (see above) and shares some of the same general trends as well. The biggest difference in the data is from this area is that the PM2.5 and Ozone deaths now share the same peaks and valleys in 2005 and 2014. Outdoor air toxins still follow suit with the trends of PM2.5 and Ozone deaths. Asthma incidents and boiler emissions still seem disconnected from the rest of the data, yet, they do get overshadowed by the size of the peak z-scores of deaths and outdoor air toxins, producing more substantial evidence to argue that there is a correlation over time between the two.
Source emissions lead to → ambient air pollutants which lead to → severe health consequences across the five boroughs. The colored bands track how emissions cascade through the atmospheric ecosystem, highlighting the boroughs carrying the heaviest pollutant volumes feed into massive, thick blocks of Asthma impacts and Respiratory deaths.
Emissions aren’t just invisible gas, they flow into specific types of chemical pollution, which funnels down into vulnerable communities.
Sources:
-1 Data.gov Dataset -2 PM EPA -3 Ozone EPA -4 NO2 EPA -5 SO2 EPA -6 Benzene EPA -7 Formaldehyde EPA -8 Sea Breeze Converegence -9 Cornell Summertime Air -10 Energo Oil -11 Air Quality Snapshots -12 Cleaner Heating Fuels in New York City