Electronic Field Guide » Restoration & Goals » Salt-Contaminated Soils

Prepared by Patrick Drohan (Ecosystem Science and Management)

Since Pennsylvania’s shale-gas development began in about 2005, there have been several incidences in which salt-contaminated waters from a drilling operation have affected Pennsylvania’s landscapes.  In some cases these incidents have been isolated to the pad, areas adjacent to the pad, or areas along Pennsylvania’s roads where vehicles have had an accident. In addition, salt contamination can develop via the transportation of gas development chemicals, improper site storage leading to a container leak, waterway transport of flowback waters, and the use of salt-laden liquids for dust suppression. Due to the high salinity composition of such waters, degradation of soils, vegetation, and adjacent waterways that the fluids spill into can occur.
 
Although shale gas development is relatively new to Pennsylvania, similar extraction operations around the world have prompted the exploration of how fluids high in salt affect waterways and landscapes.  One of the best publications on this topic was developed by the British Columbia Ministry of Water, Land, and Air Protection; Ministry of Transportation and Highways; British Columbia Buildings Corporation; and the Canadian Association of Petroleum Producers (Bright and Addison, 2002). This report outlines the chemistry of salts, their interaction with many facets of the environment, and the health risk of spills to a variety of organisms, and proposes soil matrix standards and regulations for spill cleanup. 
 
In addition, there has been some research in Pennsylvania on the topic of brine spills from oil and gas operations in existence prior to shale gas and their effect on adjacent ecosystems (Walters and Auchmoody, 1989), and on purposeful application of brine fluids as a strategy to renovate wastewaters (DeWalle and Galeone, 1990). 

Walters and Auchmoody (1989) studied a leaking impoundment on an Allegheny National Forest oil operation where all vegetation was killed. The vegetation recovered by year four into a young forest (deer fencing was used to inhibit browsing).  

DeWalle and Galeone (1990) evaluated three application rates of gas well brine (1.52, 0.69, and 0.17 kg chloride/m2) applied one time in November 1985.  They found that little overstory mortality occurred, but that understory red maple and flowering dogwood saplings were reduced by 30 and 40%, respectively, on the plot receiving the highest application rate.  Several species of herbaceous plants were killed in all plots, but the majority of understory woody plants survived.  Chloride (Cl), barium (Ba), lead (Pb), arsenic (As), selenium (Se), and cadmium (Cd) were found in soil water to a depth of 70 cm, and levels exceeded the Safe Drinking Water Act standards right after brine application.  Within three months, concentrations in soil macropores had declined to safe levels, but concentrations in micropores remained at or slightly above Safe Drinking Water Act levels for 8 months after treatment (for As, Se, Ba, and Cd).  Lead and Cd appeared to be mobilized in the soil due to complexation with Cl from the brine. 

Finally, Adams (2011) documented the purposeful application of 303,000 L of shale gas flowback water on 0.20 ha of forest at the USDA Fernow Experimental Forest in West Virginia. Adams (2011) noted that understory and some overstory vegetation were outright killed or experiencing leaf drop within days of application, and 56% of trees were dead within two years; soil sodium and chloride increased 50-fold, but declined over time.
 
From these few studies, and the review from British Columbia, we can assume that spills
of flowback water or production chemicals high in sodium and chloride across
Appalachian shale gas operations will result in vegetation damage and/or death. Soil organisms will also likely be negatively affected.  How long an ecosystem will take to recover will depend on the concentration of salts, the amount of fluid spilled, and the cleanup effort conducted.  Monitoring of areas where a spill has occurred should certainly be conducted for several years, and involve soil and water testing.
 

  • Adams, M.B. 2011. Land application of hydrofracturing fluids damages a deciduous forest stand in West Virginia. Journal of Environmental Quality 40:1340-1344. 
  • DeWalle, D.R. and D.G. Galeone. 1990. One-time dormant-season application of gas well brine on forest land. Journal of Environmental Quality. 19:288-295. 
  • Walters, R.S. and l.R. Auchmoody. 1989. Vegetation re-establishment on a hardwood forest site denuded by brine. Landscape and Urban Planning 17:127-133.