Concern for future water supplies in Tennessee has grown in recent years as a result of increased awareness of competing needs, the impact of droughts, and the need for more water to support growing populations. The U.S. Geological Survey conducts investigations to improve the knowledge about interactions of geology, climate, humans, and ecosystems with the water cycle, which is critical to understanding and optimizing water availability. The Hixson Utility District in Hamilton County, Tennessee, uses groundwater resources in the Cave Springs area as a water supply, withdrawing water from two well fields located at Cave Springs and Walkers Corner. Historically, Hixson Utility District has withdrawn about 5 million gallons per day (Mgal/d) at the Cave Springs well field and between 2 and 3 Mgal/d at the Walkers Corner well field. To assess the capacity of the groundwater resources in the Cave Springs area to meet future demands, four different scenarios of increased groundwater withdrawals were analyzed using computer model simulations.

In the study area, groundwater is present in both regolith and bedrock. Groundwater flow in the regolith occurs as diffuse flow as recharge from precipitation moves through the regolith to discharge to streams and springs or to the underlying bedrock. Most of the bedrock in the study area has low primary porosity and permeability; however, fracturing and dissolution have produced substantial secondary porosity and permeability. Groundwater flow through the bedrock occurs as both diffuse and conduit flow. Recharge to the aquifer is from two distinct sources: direct infiltration of precipitation and losing streams. A major source of recharge to the aquifer that supplies Cave Springs is surface water that is lost from North Chickamauga Creek as it flows from the Cumberland Plateau onto the Newman Limestone. Average annual streamflow loss (groundwater recharge) from this reach of North Chickamauga Creek for the period November 2000 through June 2006 is about 18 cubic feet per second (ft³/s). Groundwater leaves the aquifer as either discharge to North Chickamauga Creek, Poe Branch, and Lick Branch; discharge to Chickamauga Lake; spring flow to Cave Springs or Rogers Spring; or withdrawals at the Cave Springs or Walkers Corner well fields.

Using computer model simulations, four scenarios of increased groundwater withdrawals were analyzed. Each of these four scenarios are compared to a base-case simulation that uses groundwater withdrawal rates from 2012 of 5.1 Mgal/d from the Cave Springs well field and 2.7 Mgal/d from the Walkers Corner well field. Under scenarios A and B, pumpage is increased at Cave Springs by 2 Mgal/d and 5 Mgal/d, respectively, while pumpage at Walkers Corner remains unchanged. Under scenarios C and D, pumpage is increased at Walkers Corner by 2.6 Mgal/d and 4.5 Mgal/d, respectively, while pumpage at Cave Springs remains unchanged. The effects of the increased withdrawals were analyzed by comparing water budget changes of the groundwater discharges to Chickamauga Lake, North Chickamauga Creek, Cave Springs, Poe Branch, and Lick Branch/Rogers Spring for each of the scenarios and evaluating changes in groundwater levels at the well fields.

Under scenarios A and B, the largest change in the water budget occurs for flow to Cave Springs with decreases of 1.9 and 4.7 ft³/s, respectively. Similarly, groundwater discharge to North Chickamauga Creek decreases by 1.0 ft³/s and 2.6 ft³3/s, respectively. Under scenarios C and D, the largest change in the water budget occurs for flow to Chickamauga Lake with decreases of 1.3 ft³/s and 2.3 ft³/s, respectively. Similarly, groundwater discharge to North Chickamauga Creek decreases by 1.1 ft³/s and 2.1 ft³/s, respectively. Changes in groundwater levels at the well fields were also analyzed. At the Cave Springs well field, maximum declines in groundwater levels due to additional pumpage are less than 1 foot for all scenarios. Groundwater level changes at the Cave Springs well field are small due to the highly transmissive nature of the aquifer in this location. Maximum groundwater-level declines at Walkers Corner are less than 1 foot for scenarios A and B and about 52 feet and 82 feet for scenarios C and D, respectively. Under scenarios C and D, the regional potentiometric surface shows a large cone of depression centered on the Walkers Corner well field and elongated along geologic strike.