State of the Planet

News from the Columbia Climate School


Data Dive: NYC Traffic Trends, Street Safety and Public Health

Data Dives

Data Dives are conversations with Columbia Climate School researchers and affiliates to learn more about their work and explore trending topics through the lens of data science and visualization.

Peter MuennigPeter Muennig is a professor in Columbia University’s Department of Health Policy and Management. His research links social determinants of health with cost-effectiveness analyses to determine the best mix of social policies for optimizing population health. He has been the principal investigator on multiple NIH grants, has received $16 million in funding, and has published over 200 articles in leading journals.

Muennig’s team has conducted research on increasing the number of bike lanes In New York City, expanding bike share programs to low-income communities, Vision Zero and speed reduction programs, congestion pricing, and building parks over the Cross Bronx Expressway.

Chart: Traffic Volume on MTA and Port Authority Bridges & Tunnels

Traffic volume on New York City bridges and tunnels returned to pre-pandemic levels in 2021, and has continued to trend higher. How does vehicle traffic affect the health and well-being of communities? What can cities do to regulate traffic and ease congestion?

Traffic affects the health and well-being of communities in countless ways, but four of these are very important contributors to suffering and the loss of life.

By far the biggest factor in determining the impact of traffic on community health is air pollution. The most important form of air pollution is small particulate matter called PM 2.5, which may be the biggest contributor to death and disability worldwide. PM 2.5 is produced in large quantities when diesel fuel is burned and is a part of the plumes of black smoke you sometimes see billowing out of truck exhaust pipes. PM 2.5 is also produced in smaller quantities by gasoline-burning engines, tires against the pavement, and brake pads against disc rotors. These smaller sources matter a lot because there is no safe level of exposure to PM 2.5—even a little will still spread through the air, into and through the lungs, and then into cells where it damages DNA. The biological damage caused by PM 2.5 can cause cancer, heart disease, stroke, and premature cellular aging. At least 4.5 million people are killed as a result of exposure to PM 2.5 every year globally. In the US, the annual number of deaths is lower, roughly 50,000, thanks to air quality regulations from the Environmental Protection Agency. While 4.5 million deaths is a lot, a much bigger toll is exacted on those still alive—stroke and heart disease certainly hasten death, but in the meantime they exact years of suffering.

Map: Numbers of deaths attributable to PM2.5 in 2019.
Numbers of deaths attributable to PM2.5 in 2019. Source: State of Global Air

Traffic produces other kinds of deadly air pollution as well, and the closer you are to running cars and trucks, the more intense the exposure. The Environmental Protection Agency has made great progress in regulating tailpipe emissions from non-commercial vehicles. But there is a lot of room to rein in the biggest polluters—commercial vehicles. Regulations on diesel emissions need to be stricter, there is a need to get older vehicles off the road, and a need to enforce laws that prohibit commercial vehicles from idling.

Chart: Pedestrian Injuries

Traffic within the community is also a major source of injuries, of course, killing about 40,000 people nationwide annually and about 250 in New York City alone. This number does not include people who end up in wheelchairs or who suffer chronic pain as a result of non-fatal injuries. There are about 10 serious injuries for every death that occurs. As with PM 2.5, many of the deaths due to vehicle injuries are never counted. People who are in wheelchairs tend to die early, but these deaths are recorded as pneumonia or heart disease rather than the crash that put the person in the wheelchair in the first place.

Traffic also exacts a huge toll on health via quality-of-life issues. The honking and engine noise from traffic is a major contributor to insomnia and mental health issues within the community. I find it amazing that horn loudness is not regulated. Emergency vehicles stuck in traffic can blare their sirens at levels that may damage the hearing of nearby pedestrians. A truck horn can reach 130 decibels, easily enough to damage the hearing of a cyclist next to the truck. Another big quality-of-life issue is the impact of traffic on exercise. If you go to the places without cars, people are moving—jogging, biking, skating. You don’t see much of that in a car-filled city center for good reason. Without exercise, you have heart problems, obesity, diabetes, and a shorter life. Somehow, policymakers are not bothered by the massive amount of urban real estate given up to cars. New York City is rated as one of the most walkable areas of the United States, but that may seem like news to a pedestrian trying to cross a busy intersection. Maybe there have to be separate car places and people places. But there is still plenty of room to create more equitable streets by reclaiming space reserved for private vehicles (particularly street parking) for public use.

So, the obvious answer to air pollution, injuries, and quality-of-life issues (that also probably rack up tens of thousands of deaths annually) is to simply limit the number of vehicles on the road. The intuition that many people have is that if you have traffic congestion, you just need to build more roads. The opposite is true. The more roads you build, the easier it is to drive, and you end up with more congestion and pollution.

If you create pedestrian-only streets, people walk, bike, or jog to work or the store. If you remove street parking, people will have to park in expensive lots and may decide to take the train instead. The more people that take public transportation, the more the economics of running public transportation improves. So does the political economy—it is easier to advocate for bus-only lanes when more people are using the bus than are using their personal vehicle. These trends can be accelerated by taxing cars that enter the most congested areas of the city, or “congestion pricing.” Lawmakers are currently seeking to bring congestion pricing to New York City. Removing highways is feasible when there are a lot of non-commercial vehicles on them. Commercial traffic presents a challenge, however, as goods still need to be delivered. In cases where you have a lot of commercial traffic, burying a highway under a park, converting to electric vehicles, and using innovative solutions to the “last mile” problem of getting goods to the consumer can all help. Research from my team at Columbia University has shown that a groundbreaking plan to put a deck park over the Cross Bronx Expressway would save money and lives.

Chart: Crash Deaths

Crash deaths and pedestrian injuries have both trended higher in NYC and across the U.S. in recent years. Can you provide any insights into what may be behind these trends? What could be done to make streets safer?

City planners were somewhat perplexed by this question until recently. They were perplexed because the roads were getting safer, vehicles were getting safer (with self-driving features and guard rails to keep cyclists from getting trapped under trucks), and the population of the city seemed to be on the decline. But this decline in population was only due to bad projections of Census data. It turns out that the answer to your question is straightforward. The population of the city is actually increasing over time as the economy is growing. There are also more people driving and more trucks delivering goods. The more vehicles on the road, the more chance you have for a collision.

We certainly don’t want to reduce housing or slow business activity. Instead, the answer is to continue to make the cars safer and the roads safer. Often overlooked is the role of distracted driving in this equation. Injuries really started to increase beyond what we would normally expect for a given traffic volume with increasing cellphone use. Those using cellphones may be roughly four times as likely to be in a crash as those who do not use them. For this, there needs to be better enforcement and better regulation of technology in vehicles (e.g., increasing driver mode use on cellphones).

Chart: NYPD and Camera Issued Speeding Violations

According to the data, the number of violations issued by speeding cameras in NYC has steadily increased, while speeding tickets issued by the NYPD have declined. How and why is traffic enforcement—by police or automated cameras—important to keeping streets safe?

Traffic enforcement is critical for two reasons. First, when traffic volumes are low, cars speed. The city has carefully selected speed limits so that collisions with pedestrians tend not to kill them. Even exceeding the New York City speed limit by a small amount greatly increases the risk of debilitating injury and death. Second, when traffic volumes are high, enforcement is needed to prevent congestion. When frustrated drivers run red lights and block intersections, chaos ensues. This leads to idling cars and more pollution.

My research team conducted a study to attempt to optimize the pattern of traffic cameras in the city. We found that optimizing a pattern of traffic camera installations would save lives and prevent significant disability. If we took the remaining lifespan of the average New Yorker alive today (about 40 years) and asked what traffic cameras would do during the rest of their lives, the answer is that a thousand fellow New Yorkers wouldn’t die from being struck by a car, and roughly a billion dollars would be put into city coffers. Right around the time that we published this paper, lawmakers decided to remove traffic cameras around schools. Outraged parents protested. Our study was helpful in getting the number of cameras up, but they still are not optimized.

New York City has seen an explosion of different types of micromobility vehicles on its streets and bike lanes. What challenges and opportunities do vehicles like e-bikes, scooters or mopeds pose for cities? Will micromobility continue to play a growing role in the urban transportation landscape?

E-bikes, scooters, and electric motorcycles each produce their own challenges. All three forms of transport are clearly superior to driving from a public health standpoint. But the benefits for public health all come from improvements in air quality. We also see potential increases in serious injuries, and certainly annoyance, that these new forms of transit can cause. When pedestrians and drivers are annoyed, they tend to complain. These complaints can, in turn, lead to a backlash against light electric transportation. That is a shame because these vehicles are often used as a bridge from the home to the subway. So, the big research question is not so much whether these vehicles should be in the traffic mix, but how. How can we get the mix right so that they are better integrated into the transportation network? How can we better protect both the users of light electric transportation and the pedestrians that they buzz by?

Chart: Scooter and Moped Crashes

Vehicle-sharing programs introduce novice riders to powerful two-wheelers. They are also disproportionately used by young riders who are male. Add inexperience to the mix, and you potentially have problems like reckless driving. The companies that share these vehicles should do a better job of tracking which ones are driving against traffic, on sidewalks, or other areas they should not be. Already strides are being made toward fixing these problems, and helmets are now mandatory on faster vehicles, such as electric motorbikes. However, not enough users are banned from ride sharing as a result of irresponsible driving, and there is plenty of room to monitor and enforce violations.

Privately owned e-bikes have speed limiters on them, but the user can sometimes hack the limiter. Better enforcement would help keep illegal vehicles out of bike lanes. When a bike is moving at high speeds, be it an e-bike or an electric motorbike, it should be in the automobile traffic flow and the rider should have a motorcycle helmet on.

Shared electric scooters (the kind that look like skateboards with a steering wheel) seem like an intuitively bad idea because they have little wheels and users tend not to be familiar with riding them. Despite this outward appearance of danger, they appear to be an asset even in places like Lisbon, which has slippery cobblestone streets and steep hills. We need more data and we need to judge micromobility vehicles according to it, not by their appearance.

One final comment is that traffic cameras have reduced the need for human-to-human police enforcement. They also remove police bias (such as against people of color or against cyclists). But they cannot be a substitute for it. Map applications warn drivers of cameras, and there are plenty of reckless drivers who are not caught running red lights or speeding (or, for that matter, doing any number of other crazy things). The police still need to be part of the equation.

Be sure to leave your questions and ideas for future Data Dives in the comments section below! ↓

Notify of

Inline Feedbacks
View all comments