Shale Gas Development Linked to Traffic Accidents in Pennsylvania
Pennsylvania has seen the development of more than
5,000 7,000 hydraulically fracked shale gas wells since 2004. The fracking process itself requires water and other liquids to work, not to mention rigs, other equipment, and labor, to fully develop the well. The water used in hydraulic fracturing is primarily brought to and from a well via tanker trucks, sometimes requiring more than a thousand trips per well—and much of these trips are along rural roads or through small towns.
Until now, there have been no statistical estimates of the causal effects these tanker trucks have on accidents, and the consequences of such accidents on human morbidity and mortality, and property damage. Such impacts would constitute external damages from shale gas development—damages that the companies (and regulators) may not take into account when they make decisions about developing the resource.
Now we are able provide preliminary analysis of the data that link well development activities to traffic—related accidents at the county level in Pennsylvania. On one hand, we expect to find such links because any increase in traffic will lead to increases in accidents, other things equal. This is a social cost of economic development and growing wealth. However, these links with shale gas development can be made—at least tentatively—because of the large degree of spatial and temporal variation of shale gas well development and knowing whether a heavy truck was involved in the accidents.
We used data on accidents provided by the Crash Reporting System (CRS) maintained by the Pennsylvania Department of Transportation. These data provide detailed information on all motor vehicle crashes in Pennsylvania between 1997 and 2011, such as the type of vehicles involved, exact location and time of the accident, and severity of accident (including the number of major injuries and fatalities). In the CRS dataset there are more than 2 million crashes, 20,392 of which resulted in one or more fatality. The data on shale gas wells are from the Pennsylvania Department of Environmental Protection and the Department of Conservation and Natural Resources.
In our preliminary analysis of the CRS data, we find a significant increase in the number of total accidents and accidents involving a heavy truck in counties with a relatively large degree of shale gas development as compared to counties with less (or no) development. Figure 1 compares the average number of accidents involving a heavy—duty truck in counties with 20 or more shale gas wells, the same information for counties that have less than 20 shale gas wells, and the number of shale gas wells drilled over the same time period throughout the state.
There is a high degree of variation in accident rates from year to year, and the lines for both groups of counties move together and intersect after shale gas development takes off in the 2009—10 period. At this point the two accident rates diverge, with far higher accident rates in the counties with more than 20 wells.
We find that one additional well drilled per month raises the frequency of accidents involving a heavy truck by more than 2 percent (on average there are 9 of these crashes per county per month). These results control for changes in county population, county—characteristics, and state—level trends in accident rates over time. We also find that with one additional well drilled in a county, the number of accidents involving a fatality increases by 0.6 percent (on average there are 1.8 fatalities per county per month).
This is just the beginning of our research in this area. We have also begun collaborating with Geisinger Health Care System, a Pennsylvania—based health care provider, to obtain detailed records on trauma cases seen in emergency rooms, follow—up doctor visits, and the cost of care.
Estimating how shale gas development impacts traffic is important for policy makers. A report by the Centers for Disease Control and Prevention finds that 27% of fatalities among oil and gas extraction workers in the United States from 2003 to 2006 were from highway motor vehicle crashes. One might expect to see an increase in the number of accidents in areas developing shale gas resources simply because there are more people on the road, not necessarily because the probability of an accident increases. However, there are reasons why increased shale gas truck traffic could increase the probability of an accident; there are oil field exemptions from highway safety rules created in the 1960s that allow truckers in the oil and gas industry to work longer hours than drivers in other industries. Furthermore, it has been shown that an increase in the number of light—trucks in the vehicle fleet increases annual traffic fatalities. A better understanding of the extreme costs (such as accidents and fatalities) associated with shale gas development activity could promote attention and resources to preventing such accidents in the future.