Indoor radon levels in Pennsylvania have been slowly rising since 2004, around the time that unconventional natural gas development using hydraulic fracturing ("fracking") began in the state. In our new study published in Environmental Health Perspectives, we found three pieces of evidence that there may be a link between unconventional natural gas development and indoor radon levels across the state.
Historically, Pennsylvania has had one of the biggest indoor radon problems in the country. Why? Much of the bedrock in Pennsylvania contains high levels of uranium, which is radioactive and eventually decays to radium and radon gas. Radon gas can then enter buildings by diffusing through cracks in the foundation or by dissolving in water. It can get trapped in the basement and other areas of the home and can lead to health effects. Radon is the second-leading cause of lung cancer in the United States.
Due to high levels in Pennsylvania, indoor measurements are often taken when buildings are bought and sold with results reported to the state. This created a huge database, with over 1.5 million measurements from 1989 to 2013. The first unconventional natural gas development in Pennsylvania was in 2005, and by 2013 the industry had drilled 7,469 unconventional wells in the Marcellus shale.
Fracking is just one step in the process of unconventional natural gas development. The sheer scale of development in Pennsylvania made us, and others, wonder if the industry might be influencing those indoor radon measurements collected by the state. Once fully developed, some have estimated that there will be over 50,000 wells in the state.
Big data, small signal
To develop an unconventional natural gas well, several steps are completed in turn. First, an area the size of about five football fields is cleared. Then hundreds of truck trips bring thousands of feet of piping, water, chemical additives, sand and equipment to the site. Next, millions of gallons of water are injected deep underground, fracturing the shale rock and releasing natural gas.
Much of the water -- known as flow back or produced waste waters -- and drilling solids -- or well cuttings -- returns to the surface. That flow back sometimes contains radium, a precursor to radon gas. Also, the actual natural gas from shale formations can contain radon, and the Marcellus shale can be very high in this regard.
With prior scientific findings in mind, we took a "big data" approach, linking several sources of already-collected data on radon, wells and communities to see if there was any evidence that this new industry may be creating new pathways for radon or its precursors to surface and enter buildings.
We started with an enormous amount of information: addresses, radon levels, test types, and building types, and detailed data on wells and communities, and got to work. First, we had to put all the addresses on a map. This gave us a quick look as to whether levels looked higher near fracking sites. Because of our other ongoing health studies, we already knew all about the wells -- locations, dates, phases, drilling depth and amount of natural gas produced. We also had information on community factors that could influence indoor radon levels, including temperature, rainfall, well water use, community type and community socioeconomic status.
Although we completed a detailed statistical analysis, we thought the chances of finding a relationship between unconventional natural gas development and radon were low because we were missing other important information about the buildings, such as age, whether they used gas or had radon-resistant construction, the integrity of the foundation (cracks or not), and others.
Different statistical approach than state
We made three observations that together made us concerned that unconventional natural gas development might affect radon levels. We also observed, all else being equal, that buildings with more exposure to producing Marcellus wells had higher levels of indoor radon. This rating was derived from a combination of the volume of gas produced and the distance between the building and the well.
Since the amount of natural gas produced increased dramatically from 2005 to 2013, another factor that also changed during that time could explain the finding. For example, if homes near producing wells became more tightly sealed over time, trapping more radon gas, we could observe the same relationship. In other words, we can't be sure that nothing else affected radon levels, but we do know that there have been thousands of new wells -- a seemingly larger change.
Our data showed that summer radon levels on the first floor of buildings within 20 kilometres of drilled wells had higher levels the closer they were located to unconventional drilling. We also did see radon levels begin to rise statewide in 2004, near the time the fracking began, and that levels rose more quickly in counties with more drilling.
Of course, in the way that dropping divorce rates in Maine have paralleled, but not caused, lower consumption of margarine nationwide, it is possible that radon levels could have increased due to a factor other than the unconventional gas industry. As it is always said: correlation does not mean causation! Although we don't have all the pieces, our results are quite suggestive of a link between the spread of fracking and higher indoor radon levels.
Exceeding EPA-recommended threshold
A recent study by the Pennsylvania Department of Environmental Protection took a polar opposite approach from us. While we used a huge amount of data across the state over many years, the agency took very detailed measurements at a small number of wells. What it found was reassuring: "little potential for additional radon exposure to the public" from Marcellus shale development.
We do worry about false reassurance since they only took measurements at 34 wells, less than 1% of the wells statewide, and did not consider how two or more wells close together could be additive. In our study we sought to take into account the cumulative effect of thousands of wells drilled statewide, and at a location more relevant to health -- in buildings where people live, work and play. By contrast, the state study evaluated a few point sources of radiation pollution.
But what happens if a building is surrounded by hundreds of wells, which are each a potential point source? Is there a cumulative impact that the state missed? We believe our study is better suited for that possibility.
As public health professionals, our goal is to protect public health. We discovered that 42% of basement radon levels exceeded the level at which the EPA recommends people take action. Homes using well water had 21% higher radon levels than homes using municipal water.
With upward trends in radon levels, Pennsylvania's long-standing radon problem certainly hasn't gone away. We now leave it up to others -- health professionals, economists, politicians and community members debating together -- to weigh the evidence regarding the risks and benefits of unconventional natural gas development. For now we suggest homeowners continue to measure radon in their homes, the state continue to be vigilant about possible impacts of this industry on pathways and levels, and that future studies move beyond this first look to better understand the relationships that we may have uncovered.
Brian S. Schwartz receives research or education funding from the National Institutes of Health, the National Institute of Occupational Safety and Health, and the Department of Energy. Dr. Schwartz is a Fellow of the Post Carbon Institute. The details of that relationship are in his profile.
Joan A. Casey does not work for, consult to, own shares in or receive funding from any company or organisation that would benefit from this article, and has no relevant affiliations.
Picture: Daniel Foster