The term "impervious" is commonly used in urban settings to describe the permeability of buildings, roads, and parking lots. When estimating recharge to an aquifer underlying an urbanized area, impervious cover becomes a prime consideration. It is commonly assumed that an increase in impervious cover leads to a decrease in precipitation recharge. However, even a cursory glance at most roads, sidewalks, or parking lots reveals that, far from being impervious, there are abundant fractures that may provide avenues of infiltration. For this study, method was developed to determine the secondary permeability of pavements using a double ring infiltrometer to measure the infiltration rate of water into fractured pavements. Linear extrapolation is employed to determine the infiltration rate as the water depth approaches zero, which is used as a proxy for hydraulic conductivity by assuming that the gradient is unity. Data were collected on concrete and asphalt pavements located in Austin, Texas, at each point a fracture or expansion joint intersected along 30-meter scanlines. By dividing the sum of the discharges for each fracture by the area represented by the scanline we are able to determine the equivalent-porous-media hydraulic conductivity. The equivalent hydraulic conductivities for discrete fractures were found to range at least three orders of magnitude, from >10⁻² to 10⁻⁵ cm/sec; scanline hydraulic conductivities range two orders of magnitude from >10⁻⁴ to 10⁻⁶ cm/sec; permeability along the scanlines tends to be dominated by one or two highly conductive fractures; and the hydraulic conductivity of the entire paved surface is 5.9·10⁻⁵ cm/s. Both apertures and point hydraulic conductivities were found to have logarithmic distributions but cross plots demonstrated no correlation, which indicated that a combination of the fill material and sub grade, not the fractures and expansion joints themselves, limit infiltration. By multiplying the paved surface hydraulic conductivity by the time the surface can be expected to be saturated, we find that 170 mm or 21 percent of mean annual rainfall is available as potential recharge. When coupled with an enhanced subsurface permeability structure resulting from the installation of utilities and the reduction of evapotranspiration from the reduction of vegetation, the net effect of roads and parking lots could be an increase in precipitation recharge. / text

Not available / text

The Leona aquifer is an important, but overlooked, water resource in Central Texas. The Quaternary Leona Formation occurs as several isolated alluvial deposits at the margins of the Edwards Plateau. Each of these deposits forms an aquifer. One of these aquifers is located near Lockhart, Texas. This aquifer is recharged by infiltration of precipitation and is discharged by numerous springs and seeps. Additional sources of discharge are evapotranspiration and cross-formational flow into the Wilcox aquifer. The saturated thickness at this location varies seasonally but is rarely greater than 3 m (10 ft). Groundwater flow in an aquifer of this scale is influenced by its heterogeneous nature. This research identified seven different facies in the Leona Formation and the underlying Wilcox Group. These divisions were based on sediment classification, lithology, and sedimentary structures. The Leona Formation is covered by sandy and silty clay soil and caliche. Each of these facies has different hydraulic properties. Many empirical relationships between grain size distribution and hydraulic conductivity (K) have been discussed in the literature. Equations developed by Hazen, Slichter, Terzaghi, Beyer, Saurbrei, and Kozeny were used to estimate hydraulic conductivity. Hydraulic conductivity was also measured in the laboratory with constant and falling head permeameters. Hydraulic conductivity of the Leona aquifer varies seven orders of magnitude. Hydraulic conductivity varies up to four orders of magnitude within a single facies due to small-scale differences in grain size distribution and degree of cementation. The arithmetic mean of hydraulic conductivity in vertical profiles through the Leona aquifer ranges from 0.013 cm/sec (37 ft/day) to 0.14 cm/sec (397 ft/day). Water quality is a concern for many unconfined shallow alluvial aquifers, including the Leona aquifer. Elevated nitrate levels indicate contamination resulting from agricultural land use. Nitrate concentration in the Leona aquifer ranges from 4 ppm nitrate as NO₃ to greater than 70 ppm nitrate as NO₃. These concentrations are significantly greater than those observed in the Wilcox aquifer. The U.S. Geologic Survey computer code MODFLOW was used to create a groundwater model of the Leona aquifer. In the best simulation, specific yield was 0.1 and horizontal hydraulic conductivity was 0.058 cm/sec (164 ft/day). The simulated hydraulic conductivity is an order of magnitude less than observed in gravel pit outcrops. Modeled recharge was 9 percent of annual precipitation in 2003 and 20 percent of precipitation in the first six months of 2004. Five hypothetical wells were placed in the model to examine the effects of pumping on the aquifer. Wells pumped for 61 days at 0.04 l/sec (0.6 gpm) cause insignificant drawdown while wells pumped at a rate of 3.5 l/sec (55 gpm) cause up to 0.55 m (1.8 ft) of drawdown. Natural drainage of the aquifer caused the water table to decline 0.8 m (2.6 ft) over this same period. MODPATH simulations using this groundwater model indicate an average residence time in the aquifer of 13 years and a maximum residence time of 70 years. / text