Gas-Tracer Experiment for Evaluating the Fate of Methane in a Coastal Plain Stream: Degassing versus in-Stream Oxidation.

Methane emissions from streams and rivers have recently been recognized as an important component of global greenhouse budgets. Stream methane is lost as evasion to the atmosphere or in-stream methane oxidation. Previous studies have quantified evasion and oxidation with point-scale measurements. In this study, dissolved gases (methane, krypton) were injected into a coastal plain stream in North Carolina to quantify stream CH4 losses at the watershed scale. Stream-reach modeling yielded gas transfer and oxidation rate constants of 3.2 ± 0.5 and 0.5 ± 1.5 d-1, respectively, indicating a ratio of about 6:1. The resulting evasion and oxidation rates of 2.9 mmol m-2 d-1 and 1,140 nmol L-1 d-1, respectively, lie within ranges of published values. Similarly, the gas transfer velocity (K600) of 2.1 m d-1 is consistent with other gas tracer studies. This study illustrates the utility of dissolved-gas tracers for evaluating stream methane fluxes. In contrast to point measurements, this approach provides a larger watershed-scale perspective. Further work is needed to quantify the magnitude of these fluxes under varying conditions (e.g., stream temperature, nutrient load, gradient, flow rate) at regional and global scales before reliable bottom-up estimates of methane evasion can be determined at global scales.

Stream measurements locate thermogenic methane fluxes in groundwater discharge in an area of shale-gas development.

The environmental impacts of shale-gas development on water resources, including methane migration to shallow groundwater, have been difficult to assess. Monitoring around gas wells is generally limited to domestic water-supply wells, which often are not situated along predominant groundwater flow paths. A new concept is tested here: combining stream hydrocarbon and noble-gas measurements with reach mass-balance modeling to estimate thermogenic methane concentrations and fluxes in groundwater discharging to streams and to constrain methane sources. In the Marcellus Formation shale-gas play of northern Pennsylvania (U.S.A.), we sampled methane in 15 streams as a reconnaissance tool to locate methane-laden groundwater discharge: concentrations up to 69 µg L(-1) were observed, with four streams ≥ 5 µg L(-1). Geochemical analyses of water from one stream with high methane (Sugar Run, Lycoming County) were consistent with Middle Devonian gases. After sampling was completed, we learned of a state regulator investigation of stray-gas migration from a nearby Marcellus Formation gas well. Modeling indicates a groundwater thermogenic methane flux of about 0.5 kg d(-1) discharging into Sugar Run, possibly from this fugitive gas source. Since flow paths often coalesce into gaining streams, stream methane monitoring provides the first watershed-scale method to assess groundwater contamination from shale-gas development.

Innovative environmental tracer techniques for evaluating sources of spring discharge from a carbonate aquifer bisected by a river.

Littlefield Springs discharge about 1.6 m³ /s along a 10-km reach of the Virgin River in northwestern Arizona. Understanding their source is important for salinity control in the Colorado River Basin. Environmental tracers suggest that Littlefield Springs are a mixture of older groundwater from the regional Great Basin carbonate aquifer and modern (post-1950s) seepage from the Virgin River. While corrected ¹4C apparent ages range from 1 to 9 ka, large amounts of nucleogenic 4He and low ³He/4He ratios suggest that the carbonate aquifer component is likely even older Pleistocene recharge. Modeled infiltration of precipitation, hydrogeologic cross sections, and hydraulic gradients all indicate recharge to the carbonate aquifer likely occurs in the Clover and Bull Valley Mountains along the northern part of the watershed, rather than in the nearby Virgin Mountains. This high-altitude recharge is supported by relatively cool noble-gas recharge temperatures and isotopically depleted δ²H and δ¹8O. Excess (crustal) SF6 and 4He precluded dating of the modern component of water from Littlefield Springs using SF6 and ³H/³He methods. Assuming a lumped-parameter model with a binary mixture of two piston-flow components, Cl⁻ /Br⁻, Cl⁻ /F⁻, δ²H, and CFCs indicate the mixture is about 60% Virgin River water and 40% groundwater from the carbonate aquifer, with an approximately 30-year groundwater travel time for Virgin River seepage to re-emerge at Littlefield Springs. This suggests that removal of high-salinity sources upstream of the Virgin River Gorge would reduce the salinity of water discharging from Littlefield Springs into the Virgin River within a few decades.

A stream-based methane monitoring approach for evaluating groundwater impacts associated with unconventional gas development.

Gaining streams can provide an integrated signal of relatively large groundwater capture areas. In contrast to the point-specific nature of monitoring wells, gaining streams coalesce multiple flow paths. Impacts on groundwater quality from unconventional gas development may be evaluated at the watershed scale by the sampling of dissolved methane (CH4 ) along such streams. This paper describes a method for using stream CH4 concentrations, along with measurements of groundwater inflow and gas transfer velocity interpreted by 1-D stream transport modeling, to determine groundwater methane fluxes. While dissolved ionic tracers remain in the stream for long distances, the persistence of methane is not well documented. To test this method and evaluate CH4 persistence in a stream, a combined bromide (Br) and CH4 tracer injection was conducted on Nine-Mile Creek, a gaining stream in a gas development area in central Utah. A 35% gain in streamflow was determined from dilution of the Br tracer. The injected CH4 resulted in a fivefold increase in stream CH4 immediately below the injection site. CH4 and δ(13) CCH4 sampling showed it was not immediately lost to the atmosphere, but remained in the stream for more than 2000 m. A 1-D stream transport model simulating the decline in CH4 yielded an apparent gas transfer velocity of 4.5 m/d, describing the rate of loss to the atmosphere (possibly including some microbial consumption). The transport model was then calibrated to background stream CH4 in Nine-Mile Creek (prior to CH4 injection) in order to evaluate groundwater CH4 contributions. The total estimated CH4 load discharging to the stream along the study reach was 190 g/d, although using geochemical fingerprinting to determine its source was beyond the scope of the current study. This demonstrates the utility of stream-gas sampling as a reconnaissance tool for evaluating both natural and anthropogenic CH4 leakage from gas reservoirs into groundwater and surface water.

Determining anisotropic transmissivity using a simplified Papadopulos method.

The straight-line method presented by Papadopulos requires a minimum of three observation wells for determining the transmissivity tensor of a homogeneous and anisotropic aquifer. A simplification of this method was developed for fractured aquifers where the principal directions of the transmissivity tensor are known prior to implementation, such as when fracture patterns on outcropping portions of the aquifer may be used to infer the principal directions. This new method assumes that observation wells are drilled along the two principal directions from the pumped well, thus reducing the required number of observation wells to two. This method was applied for an aquifer test in the fractured Navajo Sandstone of southwestern Utah and yielded minimum and maximum principal transmissivity values of 70 and 1800 m(2)/d, respectively, indicating an anisotropy ratio of approximately 24 to 1.

Gas-partitioning tracer test to quantify trapped gas during recharge.

Dissolved helium and bromide tracers were used to evaluate trapped gas during an infiltration pond experiment. Dissolved helium preferentially partitioned into trapped gas bubbles, or other pore air, because of its low solubility in water. This produced observed helium retardation factors of as much as 12 relative to bromide. Numerical simulations of helium breakthrough with both equilibrium and kinetically limited advection/dispersion/retardation did not match observed helium concentrations. However, better fits were obtained by including a decay term representing the diffusive loss of helium through interconnected, gas-filled pores. Calculations indicate that 7% to more than 26% of the porosity beneath the pond was filled with gas. Measurements of laboratory hydraulic properties indicate that a 10% decrease in saturation would reduce the hydraulic conductivity by at least one order of magnitude in the well-sorted sandstone, but less in the overlying soils. This is consistent with in situ measurements during the experiment, which show steeper hydraulic gradients in sandstone than in soil. Intrinsic permeability of the soil doubled during the first six months of the experiment, likely caused by a combination of dissolution and thermal contraction of trapped gas. Managers of artificial recharge basins may consider minimizing the amount of trapped gas by using wet, rather than dry, tilling to optimize infiltration rates, particularly in well-sorted porous media in which reintroduced trapped gas may cause substantial reductions in permeability. Trapped gas may also inhibit the amount of focused infiltration that occurs naturally during ephemeral flood events along washes and playas.