Evaluation of potential gas clogging associated with managed aquifer recharge from a spreading basin, southwestern Utah, U.S.A.

Abstract

Sand Hollow Reservoir in southwestern Utah, USA, is operated for both surface-water storage and managed aquifer recharge via infiltration from surface basin spreading to the underlying Navajo Sandstone. The total volume of estimated recharge from 2002 through 2011 was 131 Mm 3 ., resulting in groundwater levels rising as much as 40 m. Hydraulic and hydrochemical data from the reservoir and various monitoring wells in Sand Hollow were used to evaluate the timing and location or reservoir recharge moving through the aquifer, along either potential clogging from trapped gases in pore throats, siltation, or algal mats. Several hyrdochemical tracers indicated this recharge had arrived at four monitoring wells located within about 300 m of the reservoir by 2012. At these wells, peak total dissolved-gas pressures exceeded two atmospheres (>1,500 mm mercury) and dissolved oxygen approached three times atmospherically equilibrated concentrations (>25 mg/L). these field parameters indicate that large amounts of gas trapped in pore spaces beneath the water table have dissolved. Lesser but notable increases in these dissolved-gas parameters (without increases in other indicators such as chloride-to-bromide ratios) at monitoring wells farther away (>300 m) indicate moderate amounts of in-situ sir entrapment and dissolution caused by the rise in regional groundwater levels. This is confirmed by hydrochemical difference between these sites and wells closer to the reservoir where recharge had already arrived. As the reservoir was being filled by 2002, managed aquifer recharge rates were initially very high (1.5 x 10 -4 cm/s) with the vadose zone becoming saturated beneath and surrounding the reservoir. These rates declined to less than 3.5 x 10 -6 cm/s during 2008. The 2002-08 decrease was likely associated with a declining regional hydraulic gradient and clogging. Increasing recharge rates during mid-2009 through 2010 may have been partly caused by dissolution of air bubbles initially entrapped in the aquifer matrix. Theoretical gas dissolution rates, coupled with field evidence of a decline iin total dissolved-gas pressure and dissolved oxygen from nearby monitoring wells, support the timing of this gas dissipation.