Hydrogeology and simulation of ground-water flow in the Silurian-Devonian aquifer system, Johnson County, Iowa

Bedrock of Silurian and Devonian age (termed the “Silurian-Devonian aquifer system”) is the primary source of ground water for Johnson County in east-central Iowa. Population growth within municipal and suburban areas of the county has resulted in increased amounts of water withdrawn from this aquifer and water-level declines in some areas. A 3-year study of the hydrogeology of the Silurian-Devonian aquifer system in Johnson County was undertaken to provide a quantitative assessment of ground water resources and to construct a ground-water flow model that can be used by local governmental agencies as a management tool.

Johnson County is underlain by unconsolidated deposits of Quaternary age and Paleozoic-age bedrock units. The bulk of the Quaternary deposits consists of weathered and unweathered glacial till; however, shallow alluvium and buried sand and gravel deposits also are present. Six bedrock hydrogeologic units are present in Johnson County (oldest to youngest): Maquoketa confining unit, Silurian aquifer, Wapsipinicon Group (aquifer and confining unit), Cedar Valley aquifer, Upper Devonian shale confining unit, and Cherokee confining unit. Although separate aquifers and confining units are described, the Silurian- and Devonian-age units are considered as a single aquifer system. The top of the Silurian-Devonian aquifer system is considered as the top of the Cedar Valley aquifer, where present, and the base of the aquifer system is considered as the top of the Maquoketa confining unit.

The hydraulic properties of the rocks that comprise the Silurian-Devonian aquifer system are highly variable as a result of the variable composition of the rocks and the presence of solution features in some of the carbonate-rock units. For the combined Silurian-Devonian aquifer system, specific capacity averages 2.1 gallons per minute per foot of drawdown, transmissivity averages about 580 feet squared per day, and hydraulic conductivity averages 8.3 feet per day.

Recharge to the Silurian-Devonian aquifer system in Johnson County is predominantly from infiltration of precipitation to the bedrock. Discharge from the aquifer is primarily to municipal, industrial, and private-development wells. Reliable measurements of the amount of recharge to or discharge from the ground-water system in Johnson County, however, are not available.

Altitude of the 1996 potentiometric surface ranged from more than 750 feet above the North American Vertical Datum of 1988 (NAVD88) in northern Johnson County to less than 575 feet above NAVD88 in the central part of the county. A large cone of depression within the potentiometric surface is present in the central part of the county, between Coralville and Iowa City. A large limestone quarry is located near the center of this cone of depression. Ground water generally flows from the northern and western parts of Johnson County either toward the cone of depression in the center of the county or south out of the county. Ground water also flows toward the Cedar River in the northeastern part of the county. A ground-water divide in the northeastern part of the county roughly approximates the surface-water divide between the Iowa River and Cedar River drainages.

A numerical ground-water-flow model of the Silurian-Devonian aquifer system in Johnson County was used to test concepts of ground-water flow, to assess the need for additional data, and to evaluate the potential effects of anticipated increased ground-water development and drought. The 1-layer model was calibrated to average 1996 ground-water conditions, which were assumed to approximate steady-state flow conditions. The model also was used to simulate steady-state conditions for 2004, steady-state conditions using anticipated pumping rates for 2025, and potential future drought conditions.

The simulated potentiometric surface generally replicated the potentiometric surface for 1996 and 2004 conditions. The calculated root mean squared error values for the 1996 and 2004 simulations were 13.6 and 18.6 feet, respectively. The mean absolute differences between measured and simulated water levels for the 1996 and 2004 simulations were about 11 and 14 feet, respectively.

Total model-calculated inflow to the ground-water system for the 1996 simulation was 19.6 million gallons per day (Mgal/d), and the largest model-calculated inflow component was areal recharge (15.1 Mgal/d). Total model-calculated outflow from the ground-water system was 19.7 Mgal/d, and the largest outflow component was discharge to wells (10.5 Mgal/d). Model-calculated water-budget components for the 2004 simulation were similar to the 1996 components.

Potential future steady-state conditions were simulated using anticipated 2025 pumping rates. Pumpage both for existing wells and for assumed new wells, based on anticipated population growth in the northern part of the county and for the nearby municipalities, was included in the model. Simulated 2025 pumpage was about 1.5 Mgal/d greater than simulated 2004 pumpage. Simulated steady-state ground-water levels, using anticipated 2025 pumping rates, were lower than 2004 simulated levels throughout the county, and simulated water-level declines ranged from less than 1 foot near the county boundaries to about 11 feet.

Potential future drought conditions were simulated by assuming that recharge to the Silurian-Devonian aquifer system is reduced by a factor of 0.75 and that water-supply pumpage is increased by a factor of 1.25 over the anticipated 2025 pumping rates. Overall, simulated water levels for future drought conditions were greater than 5 feet lower than simulated 2004 conditions and were a maximum of about 30 feet lower in the northeastern part of the county.

The greatest limitation to the model is the lack of measured or estimated water-budget components for comparison to simulated water-budget components. Because the model is only calibrated to measured water levels, and not to water-budget components, the model results are nonunique. Other model limitations include the relatively coarse grid scale, lack of detailed information on pumpage from the quarry and from private developments and domestic wells, and the lack of separate water-level data for the Silurian- and Devonian-age rocks.