Estimation of unconfined aquifer hydraulic properties using gravity and drawdown data

Abstract

An unconfined aquifer test using temporal gravity measurements was conducted in shallow alluvium near Fort Collins, Colorado on September 26-27, 2009. Drawdown was recorded in four monitoring wells at distances of 6.34, 15.4, 30.7, and 60.2 m from the pumping well. Continuous gravity measurements were recorded with a Scintrex® CG-5 gravimeter near the closest well, at 6.3 m, over several multi-hour intervals during the 27 hour pumping test. Type-curve matching of the drawdown data performed assuming Neuman's solution yields transmissivity T, specific yield Sy, and elastic component of storativity S estimates of 0.018 m2s-1, 0.041, and 0.0093. The gravitational response to dewatering was modeled assuming drawdown cone geometries described by the Neuman drawdown solution using combinations of T, Sy, and S. The best fitting gravity model based on minimization of the root mean square error between the modeled and observed gravity change during drawdown resulted from the parameters T =0.0033 m2s-1, Sy =0.45, and S =0.0052. Conservative precision estimates in the gravity data widen these estimates to T =0.002-0.006 m2s-1, Sy =0.25-0.65, and S =0-0.2. Drawdown conforming to the Neuman solution was forward modeled using combinations of T, Sy, and S. Minimization of the root mean square misfit between these forward models and observed drawdown in the monitoring wells results in T =0.0080 m2s-1, Sy =0.26, and S =0.000004. Discrepancy between type-curve matching results, gravity analysis results, and drawdown modeling is attributed to heterogeneity and anisotropy within the aquifer, and a relatively large amount drawdown compared to initial saturated thickness, conditions which fail the Neuman solution assumptions. In this aquifer test, gravity was most sensitive to transmissivity, less sensitive to specific yield, and insensitive to the specific storage-saturated thickness quotient. Simultaneous deployment of multiple gravity stations during similar tests should better constrain gravity analysis aquifer property estimates of transmissivity and specific yield.