Convective transport of dissolved oxygen (DO) from shallow to deeper parts of wells was observed as the shallow water in wells in South Carolina became cooler than the deeper water in the wells due to seasonal changes. Wells having a relatively small depth to water were more susceptible to thermally induced convection than wells where the depth to water was greater because the shallower water levels were more influenced by air temperature. The potential for convective transport of DO to maintain oxygenated conditions in a well was diminished as ground-water exchange through the well screen increased and as oxygen demand increased. Convective flow did not transport oxygen to the screened interval when the screened interval was deeper than the range of the convective cell.
The convective movement of water in wells has potential implications for passive, or no-purge, and low-flow sampling approaches. Transport of DO to the screened interval can adversely affect the ability of passive samplers to produce accurate concentrations of oxygen-sensitive solutes, such as iron. Other potential consequences include mixing the screened-interval water with casing water and potentially allowing volatilization loss at the water surface. A field test of diffusion samplers in a convecting well during the winter, however, showed good agreement of chlorinated solvent concentrations with pumped samples, indicating that there was no negative impact of the convection on the utility of the samplers to collect volatile organic compound concentrations in that well. In the cases of low-flow sampling, convective circulation can cause the pumped sample to be a mixture of casing water and aquifer water. This can substantially increase the equilibration time of oxygen as an indicator parameter and can give false indications of the redox state.
Data from this investigation show that simple in-well devices can effectively mitigate convective transport of oxygen. The devices can range from inflatable packers to simple baffle systems.