An injection test was performed within the NASCENT project which consisted of the injection of a gas mixture into a fault structure and the subsequent monitoring of soil gas and groundwater chemistry in order to outline permeable pathways and flow rates in the subsurface. Such types of gas injection tests have been used by various researchers to better understand the dynamics of gas migration in the shallow environment (Ciotoli et al., in submission; Gascoyne and Wuschke, 1997; Lineham et al., 1996).
The gas injection test was conducted within the area of the Latera study site, at a location which has numerous gas vents and which, based on geophysical surveys and the alignment of the vents, has a fault crossing the site at depth. During this experiment a total of 8000 L of gas (40% Ar, 40% He and 20% CO2) was injected into a faulted interval between 9 and 12 m below ground surface. Soil gas samples were collected at regular intervals around the injection borehole from a grid of fixed sample points (64) and analysed both in the field and the laboratory for the injected gas species. Ground water samples were also collected from 6 peizometers drilled to 5m depth into the shallow aquifer in order to monitor for breakthrough of the highly soluble CO2 in the dissolved phase.
Results from the gas injection test highlight the importance of understanding the chemical and physical characteristics of the gases being monitored, as aqueous solubility, gas density and both water- and gas-phase diffusivities play a critical role in the travel times and mass attenuation features of migrating gases. The gas mixture injected at the Latera test site consisted of three rather different species, ranging from the highly insoluble, highly mobile low density helium to the very soluble, reactive and dense gas carbon dioxide, while argon has characteristics which lie between these two extremes. These differences are mirrored in the results obtained during the test, with helium arriving very early at surface within a very small area around the injection well and at high concentrations. In contrast CO2 arrived much later, had a more diffuse distribution and was observed at much lower relative concentrations than those seen for He. In addition injected helium returned to background values within a very brief period of time, on the order of weeks, whereas the injected CO2 appears to continue to slowly seep from the system even a month after the experiment had ceased.
A mechanism to explain these results is proposed whereby at early times the more soluble gases are stripped from the gas bubbles, enriching them in helium, and that this gas phase, upon reaching the water table, migrates rapidly through the unsaturated zone to surface due to its low density and high diffusivity. In contrast at later times the groundwater in the bubble flow path becomes progressively more saturated in CO2, resulting in less transfer to the dissolved phase and more of this gas reaching the water table. Once in the unsaturated zone the more dense CO2 moves laterally on top of the water table, resulting in the observed lateral dispersion, temporal attenuation and lower concentrations (due to both dilution by soil gas and removal into the aqueous phase) observed in the soil gas surveys. Lateral migration of CO2 saturated groundwater, and eventual release of this gas to the unsaturated zone via vent activity is another possible mechanism, however this can only explain anomalies observed down gradient from the injection point. In any case, dissolved gas results confirm the importance of CO2 mass transfer into the aqueous phase, as measured concentrations of dissolved CO2 in observation piezometers increased by 2 to 3 times background levels in wells located down-gradient from the injection well
Was this article helpful?