In recent years, there has been a considerable amount of attention paid to transportation issues in climate change circles. This makes sense, given that the transportation sector is the second largest source of greenhouse gas (GHG) emissions in the United States. Mobile sources produced 1,806 million metric tons of CO2 equivalent (MMtCO2e) in 2013 (27%), trailing just electricity generation, which accounted for 21% of total emissions (2,077 MMtCO2e). Emissions from the transportation sector have also grown by 16.4% since 1990, making it the second fastest growing emissions source behind agriculture.
Accordingly, the Obama administration has taken a number of steps to address the issue. These include corporate average fuel economy (CAFE) standards for passenger vehicles, new investments in electric vehicles (EVs), proposed stricter rules for emissions from heavy-duty trucks, and the recent endangerment finding for GHGs from air travel. Each of these steps will be important if the US is to meet its goal to cut overall GHGs by 26-28% by 2025, as outlined in the administration’s pledge for the upcoming Paris Conference.
How climate change affects transportation
But the other side of this equation – how climate change will affect the US transportation sector – has garnered far less focus. The 2014 National Climate Assessment included a detailed chapter on the transportation sector, and the Federal Highway Administration (FHWA) manages a pilot program to help transportation agencies assess their systems’ vulnerability to a changing climate. We know, for instance, that more extreme rainfall could wash out roads, that sea level rise endangers coastal transportation infrastructure, and that accelerated freeze-thaw cycles may increase the costs of road maintenance. But much research in this area remains to be done.
A few weeks ago, Resources for the Future, a leading environmental economics think tank, released a report that examines one as yet unexplored issue – how climate change may influence traffic accident rates. I’ll admit that the idea that climate change could affect the number of car accidents in the US seemed a bit far fetched to me a first. People tend to jump through all sorts of hoops in order to connect everything to climate change these days. But this report provides a convincing case that, barring aggressive action both to cut carbon pollution and become more resilient, climate change could make our roads even more dangerous.
The connection between weather and traffic accidents
In order to explore the relationship between climate and traffic accidents, economists Benjamin Leard and Kevin Roth first examined existing evidence on how changes in weather patters affect accident rates. Using data from the National Highway Traffic Safety Administration (NHTSA) on the number of traffic accidents that result in property damage, injuries, and fatalities for 20 states, the authors identified the existing relationships between temperature and precipitation fluctuations and traffic accidents. When temperatures fall below 20ºF, accidents that result in property damage increase by 9.3%. The relationship between temperature and accidents that lead to injuries is weak, but it appears highly significant for fatal traffic accidents. In contrast to property damage accidents, fatal accidents are 9.5% more likely on days when temperatures climb above 80ºF.
The relationship between precipitation and traffic accidents is more complex. Both rainfall and snowfall increase the incidence of property damage accidents; when rain and snow totals exceed 3 centimeters, accidents increase by 18.8% and 43.3%, respectively. This effect changes when we consider accidents leading to injuries and fatalities. In the former category, rain and snow totals over 3 centimeters lead to 14.4% and 25.9% increases in accidents, a relative reduction of 23.4% and 40.2%, respectively, compared to property damage accidents. But Leard and Roth found that fatal traffic accidents are actually less common on days with rainfall. On days with 1.5-3 centimeters of rain, fatal accident rates fall by 8.6%; this result is highly statistically significant. In contrast, this same amount of snowfall leads to 15.5% more fatalities. According to the authors, these results indicate “that drivers behaviorally compensate for these conditions,” but these adjustments are not enough to reduce the elevated accident risk presented by snowfall.
Importantly, the study also finds a strong correlation between weather conditions and the number of trips people make by foot, bike, or motorcycle (the authors term these “ultralight duty vehicles,” or ULDs). Unsurprisingly, these ULD trips decrease significantly as temperatures dip below 40ºF and as the precipitation begins to fall. Put a different way, as the weather improves, an increasing number of people choose to walk, bike, or motorcycle. This increases their exposure to automobiles, elevating the risk that they may be the victim of an accident. Accordingly, when the authors removed pedestrians, cyclists, and motorcyclists from their models, fatality rates fell by roughly half.
Climate change will cause more traffic fatalities
The authors then used these observed relationships to project how climate change could affect traffic accident rates in the future. They utilize the IPCC’s A1B scenario – a middle of the road scenario that assumes global temperatures will rise by around 4ºC – to project changes in weather and traffic accidents through the end of the century. According to the Climate Action Tracker, we are currently on pace for 3.6-4.2ºC of warming in the absence of further action, making A1B a good model for this study.
As global temperatures increase, precipitation will gradually shift from snowfall to rain. The authors find that this change will decrease the number of annual traffic fatalities by roughly 253. However, the changing climate will also induce an increase in the number of trips people take by foot, bike, and motorcycle – leading to an additional 849 traffic fatalities per year – which brings the net change to 603 additional deaths per annum. This spike in traffic fatalities will carry an annual cost of $515.7 million. All told, by 2090 climate change will lead to an additional 27,388 traffic-related fatalities in the US, carrying total costs of approximately $61.7 billion.
Now, I should note that this study does not explicitly address a few issues.
Research shows that as the number of pedestrians and cyclists increases, the chance that they will be struck by a car declines. Each time that the number of pedestrians and cyclists doubles, the risk that they will be injured in an accident falls by a third. But this decline in the relative risk of injury does not overcome the increase in the absolute number of injuries, which actually rises by a similar percentage. Leard and Roth’s study finds similar results. Furthermore, their use of fixed effects should account for this safety-in-numbers effect.
Moreover, the study does not directly account for the fact that expanding bike and pedestrian-friendly infrastructure tends to make roads safer and reduce the number of accidents. Protected bike lanes, for instance, can cut the risk of injury by up to 90%. To be fair, Leard and Roth admit that this is a potential shortcoming of their study, noting that failing to control for this effect “overstates the long-run impacts of climate change.” They also explicitly point out the important role that these types of interventions can play in climate adaptation planning,
Our results do not indicate that reliance on walking, biking, and motorcycling imply large fatality rates, as other developed English speaking and western European nations have per-capita fatality rates that are often less than half that of United States. Some countries like Sweden with extraordinarily low fatality rates have pursued a variety of urban design and legislative changes to reduce fatalities with policies such as replacing intersections with roundabouts to slow vehicles where they are likely to encounter pedestrians. Relatively simple changes like these may prove to be effective, although unglamorous, adaptation strategies to climate change.
How can this study inform climate policy?
I have two main takeaways from this study.
1. Climate change will affect nearly every aspect of our lives, and we will never be able to fully anticipate and prepare for it. That’s what happens when humanity performs a global science experiment on the planetary systems that facilitated the development of human civilization.
2. It provides even more evidence of the benefits of investing in better infrastructure for cyclists and pedestrians, particularly when accounting for climate change. It emerges as a win-win-win.
Promoting active transportation is a vital component of any mitigation strategy, as every mile we don’t drive keeps roughly one pound of CO2 out of the atmosphere.
This type of people-centric infrastructure also represents an important step that local governments can take to enhance their resilience to the impacts of climate change. We know that it may help to offset potential increases in fatal accidents due to climate change. But, more than that, it can also serve as a key lifeline to supplement existing road networks, which may be endangered by a changing climate. When roads are washed away and subway tunnels flooded, being able to ride your bike or walk to access resources and social services becomes that much more important.
Lastly, these types of investments would be valuable even in the absence of climate change, as they improve quality of life. Active transportation benefits air quality and public health, which reduces premature mortality and health care costs. Complete streets can also raise property values, increase business activity, create jobs, and make neighborhoods safer. All of these things make communities more vibrant and better able to withstand external shocks, whether from economic or climatic forces. In this way, pedestrian and cyclist-friendly infrastructure is exactly the type of no-regrets investment that climate resilience experts say we should be making now, regardless of the inherent uncertainties.