Evaluation of soil treatment techniques on remediated brine spill sites in semi-arid rangelands

Abstract

Unconventional oil and gas development (UOG) generates high volumes of flowback and produced water, byproducts of hydraulic fracturing operations, that are often released or spilled on the soil surface. Soil contamination with these wastewaters, commonly referred to as brine, has the potential to inhibit vegetation growth indefinitely. Natural attenuation of brine is not expedient in arid and semi-arid regions where most United States UOG developments are located, including the Bakken region of North Dakota. In situ (at-site) and ex situ (off-site) soil treatment techniques are commonly employed to remediate brine-contaminated soils in the Bakken. However, little is known regarding each technique's efficacy despite differences in application, cost, and efficiency. We selected 10 sites previously remediated with chemical amendments (in situ) and 11 sites with topsoil excavation (ex situ) in the United States Forest Service Little Missouri National Grasslands. We paired each remediated site with a reference to examine the ability of each strategy to return brine-contaminated sites to conditions reflective of the current state of the surrounding semi-arid rangeland ecosystem. At each site, we quantified soil electrical conductivity (ECe) as an indicator of soil salinity and measured vegetation cover, biomass production, bare ground, and litter. The difference between paired reference and remediated sites was used for analysis. Brine contamination was still evident as soil ECe was similarly increased on chemical amendment and topsoil excavation remediated sites over paired references at all soil depths tested. Due to the nature of the topsoil excavation treatment, elevated ECe in the 0–15 cm depth suggested resalinization of the new topsoil. Remediation techniques also resulted in similar plant community composition marked by an increase in exotic forb biomass, largely due to the invasion of kochia (Bassia scoparia) which was absent from reference sites. However, remediation techniques differed substantially in vegetation establishment. We found 15% more bare ground on sites remediated with chemical amendment treatment than paired references and 55% more with topsoil excavation. Our results indicate that in situ strategies may be more suitable than ex situ strategies for brine-spill remediation in semi-arid rangelands like the Bakken in North Dakota as they cause less disturbance and likely require less post-remediation management to establish adequate vegetation cover to protect the soil from further erosion.