Framework for Examining Stream Ecosystem Health in Areas of Shale Gas Development—A Multi-Parameter Watershed-Based Case Study in Pennsylvania

Science Center Objects

In a case study of 25 headwater streams in Pennsylvania, no statistically significant associations were determined between shale gas development and geochemical tracers of produced waters or measures of microbial and macroinvertebrate community composition.  Although the results are specific to the region studied, the integrated biological and geochemical framework provides a tool for examining effects of energy development in other watersheds with differing size, land cover, or best management practices.

Image: Hydraulic Fracturing Operation Underway

A hydraulic fracturing operation is underway at this drilling pad in the Marcellus Shale gas play of southwestern Pennsylvania.

(Credit: Doug Duncan, USGS. Public domain.)

The development of horizontal drilling and high-volume hydraulic fracturing techniques to extract natural gas from shale formations has matured over the past decade, leading to a rapid expansion in the development of this resource. This increase in shale gas development has been especially notable for the Marcellus Formation in Pennsylvania, where gas production has increased from 193 billion cubic feet in 2006 to 5,500 billion cubic feet by 2017, accounting for nearly 20 percent of total natural gas production in the United States according to the Energy Information System Administration.  As the activities associated with shale gas development have expanded rapidly, so have questions about the potential for changes in water chemistry and health risks to biota in nearby stream ecosystems.

Previous research has indicated the potential for various aspects of oil and gas production (unintentional releases of hydraulic fracturing fluid and wastewater, erosion and sedimentation from the construction and operation of well pads, and installation of pipelines and roads) to affect nearby headwater stream ecosystems and other water resources. However, the ability to generalize the results from these studies is limited by the small geographic scope as well as limited breadth and integration of measured chemical and biological parameters.

Adam Mumford sampling a stream in PA

Adam Mumford sampling a stream in PA

(Credit: Adam Benthem, USGS. Public domain.)

Scientists from the U.S. Geological Survey Environmental Health Energy Science Team collaborated with the Pennsylvania Department of Conservation of Natural Resources and the Pennsylvania Department of Environmental Protection  to understand if the intensity of shale gas development in a watershed was associated with increased stream-water concentrations of known geochemical tracers of gas extraction (for example sodium, chloride, lithium, barium, and bromide), and changes in the composition of benthic macroinvertebrate and sediment microbial communities.  

The team sampled 25 headwater streams in Pennsylvania State forests that drain lands with varying degrees of shale gas development intensity. Recreation and timber harvesting have been the primary land uses since the early to mid-1900s thus providing an opportunity to study the effect  of shale gas development in an environment with few other potential stressors.

Image of numerous aquatic macroinvertebrate taxa in a tray of water

Benthic macroinvertebrate taxa collected from Straight Run, Tioga County, Pennsyvania, USA. 

(Credit: Kelly Maloney, USGS. Public domain.)

To select sites, watersheds within the Pennsylvania State forest system were scored and categorized based on a gradient of shale gas development intensity, then five sites were randomly selected from each assigned category. The development of intensity scores was based on a previously designed geospatial index with 17 measures that incorporate all steps in oil and gas development, including infrastructure, well drilling and stimulation, and production. The sampling strategy included measurements of multiple hydrologic, geochemical, and biological parameters across multiple seasons of sampling in 2016 and 2017 to account for temporal variability. 

This study reported that geochemistry of the streams varied seasonally and corresponded to changes in streamflow; however, there were no definitive geochemical indications of hydraulic fracturing fluid or produced water in any of the study streams.  Streambed microbial diversity and the number of sensitive macroinvertebrate taxa (Ephemeroptera, Plecoptera, and Trichoptera), overall invertebrate diversity, and Index of Biotic Integrity scores did not differ between years or differ based on shale gas intensity index.

Overall, no statistically significant associations could be determined between shale gas development intensity and geochemical tracers of produced waters or measures of microbial and macroinvertebrate community composition in the watersheds studied.  Although these results are specific to the watersheds studied and cannot be extrapolated elsewhere, the integrated biological and geochemical approach provides a framework to examine whether there are effects of energy development in other watersheds with differing size, land cover, or best management practices. This integrated, multiparameter study is part of the USGS Environmental Health Energy Science Team’s overall approach to assess if there are health effects to humans and wildlife  because of exposures to energy production materials or activities in the environment.

The USGS Environmental Health Program (Toxic Substances Hydrology  and Contaminant Biology Programs combined), the Pennsylvania Department of Conservation and Natural Resources Shale Gas Monitoring Program, and the Pennsylvania Department of Environmental Protection provided funding for this work.

 

Reference

Mumford, A., Maloney, K.O., Akob, D.M., Nettemann, S., Proctor, A., Ditty, J., Ulsamer, L., Lookenbill, J., and Cozzarelli, I., 2020, Shale gas development has limited effects on stream biology and geochemistry in a gradient-based, multi-parameter study in Pennsylvania: PNAS, v. 117, no. 7, p. 3670-3677, https://doi.org/10.1073/pnas.1911458117.