USGS

Simulation of Projected Water Demand and Ground-Water Levels in the Coffee Sand and Eutaw-McShan Aquifers in Union County, Mississippi, 2010 through 2050

INTRODUCTION

Ground water from aquifers in formations of Cretaceous and Paleozoic age in northeastern Mississippi counties supplies most of the water used for residential, commercial, and industrial purposes. Ground water is the sole source of water for residential, commercial, and industrial use in Union County, Mississippi. Through time, increased pumpage has resulted in large water-level declines at major pumping centers. In the late 1980's, water levels in the confined part of the Eutaw-McShan aquifer may have declined sufficiently to reach the upper part of the Eutaw Formation in the Tupelo, Mississippi area (Jennings and others, 1994). An investigation of aquifers in northeastern Mississippi was begun in 1990 to better understand the hydrogeology and the flow of water in and between the aquifers, and to provide information necessary for water managers to address ground-water resource problems. As part of the investigation, a model was developed in cooperation with the Office of Land and Water Resources (OLWR) of the Mississippi Department of Environmental Quality (DEQ) to simulate ground-water flow (Strom and Mallory, 1995). This model subsequently was refined to incorporate additional stratigraphic data collected by OLWR to simulate the Coffee Sand aquifer, to simulate additional aquifers in rocks of Paleozoic age, and to incorporate additional water-use data collected by OLWR's water-use program (Strom, 1998).

 

Accelerated growth in Union County, located in northeastern Mississippi, is reflected in the increased demand for water (Glenn Duckworth, Executive Director, Union County Development Association, oral commun., 1999). Population increased nearly 8 percent from 1990 (22,085 people) to 1998 (23,828 people). During the same period, employment increased nearly 17 percent (9,893 to 11,533 employees) and municipal water use increased nearly 39 percent [1.84 to 2.55 million gallons per day (Mgal/d)] (James H. Eblen, Ph.D., Economist, Economic Development, Tennessee Valley Authority, written commun., 1999; A.J. Warner, OLWR, written commun., 1999, respectively). About one-half of the increase (an increase of 19 percent) in water use occurred from 1995 (2.14 Mgal/d) to 1998 (2.55 Mgal/d). From 1990 to 1998, self-supplied commercial and industrial use declined 28 percent from 0.40 to 0.29 Mgal/d.

 

As the population continues to grow and the economy continues to expand, the need for additional ground water will increase. Long-term projections of water use are needed for water managers to determine whether the Eutaw-McShan and Coffee Sand aquifers can supply anticipated future water demand or whether alternative water sources should be considered. An investigation was completed in 1999, in cooperation with the Tennessee Valley Authority (TVA), to project water demand to the year 2050, and to determine the regional impact of increased local withdrawals on the Coffee Sand and Eutaw-McShan aquifer systems.

Purpose and Scope

This report provides estimates of water demand for Union County, Mississippi, to the year 2050 and describes the simulated ground-water drawdowns in the Coffee Sand and Eutaw-McShan aquifers in northeastern Mississippi from 2000 to 2050, which are a result of the projected increases in pumpage. Water-demand estimates are limited to municipal water (from a public-supply system) delivered to the residential, commercial, and industrial sectors (including conveyance losses in the distribution systems) and to nonmunicipal water (from a private well) for industrial and commercial use. Water withdrawn for domestic purposes from private wells was not investigated as part of the study.

 

Maps of projected water levels for the year 2000 (near current conditions) are presented, along with maps of subsequent projected water-level drawdowns for the year 2050 representing baseline-, normal-, and high-growth water-demand scenarios. The calibrated ground-water flow model (Strom, 1998) used in this investigation was applied over the entire area that was originally simulated (fig. 1) to account for boundary conditions and maintain calibration. Although the study area corresponds to the area of the calibrated ground-water flow model, the focus of the investigation is Union County, Mississippi, and discussions of the results are limited to Union County.

Approach

Water-use data were collected for each municipal and nonmunicipal facility in Union County for 1998. By applying regression analyses to the water-use data, an intercept and coefficients for a set of economic and climatic variables were determined. Future water demand was simulated using the Institute for Water Resources—Municipal and Industrial Needs (IWR-MAIN) system (Planning and Management Consultants, Ltd., 1999) with a linear-predictive format using the estimated model intercept and regression coefficients, and demographic, economic, and climatic data for 2010, 2020, 2030, 2040, and 2050. Water conservation was a part of the analysis. Nonresidential water use was simulated using a constant-rate model and with employment projections for the manufacturing and nonmanufacturing sectors. The water-demand model output was generated for normal- and high-growth scenarios.

 

The water-demand model output for the normal- and high-growth scenarios was used as input to the ground-water flow model to project water levels to the year 2050. Because the ground-water flow model originally simulated flow from 1900 to 1995, water-use data were added for 1996-2000 to start projection simulations at near-current conditions. This was accomplished by using 1996-1998 water-use data developed for the Coffee Sand and Eutaw-McShan aquifers for Union County and then increasing the 1998 Union County water-use data by 1.03 percent annually for 2 years to create the water-use data sets for the years 1999 and 2000. For areas outside of Union County and for two wells screened in the Gordo aquifer within Union County, the existing 1995 water-use data sets were increased 1.03 percent annually to create the 1996-2000 water-use data sets for all scenarios.

 

Stress periods representing each year from 2001 to 2050 were added to the model for the projection scenarios. For a baseline projection scenario, water use was increased by 1.03 percent each year from 2001 to 2050 for all wells simulated in the model. For the normal-growth water-use scenario, input from the water-use model representing normal growth was used for the Coffee Sand and Eutaw-McShan aquifers in Union County. In other areas, water use was increased by 1.03 percent each year from 2001 to 2050. For the high-growth water-use scenario, input from the water-use model representing high growth was used for the Coffee Sand and Eutaw-McShan aquifers in Union County. In other areas, water use was increased by 1.03 percent each year from 2001 to 2050.

 

In all of the scenarios simulated, one addition was made: pumpage from a likely mining operation in Choctaw County, Alabama, was added and simulated from 2001 to 2031 using withdrawal rates of 2 Mgal/d in the Coker aquifer and 4 Mgal/d in the massive sand aquifer. These rates were held constant for the projected 30 years of mining operations.

Hydrogeologic Setting

The study area for the ground-water flow model covers 34,960 square miles (mi2), primarily in northeastern Mississippi, but includes parts of northwestern Alabama, southwestern Tennessee, and eastern Arkansas (fig. 1). The area includes the extent of the aquifers that are a source of freshwater in sediments and rocks of Cretaceous and Paleozoic age (excluding the Cretaceous Ripley aquifer) and adjacent areas that affect ground-water flow and availability of water in northeastern Mississippi (fig. 2). A detailed description of the hydrogeology of the study area may be found in Strom (1998). The focus of this investigation is limited to the Coffee Sand and Eutaw-McShan aquifers.

Coffee Sand Aquifer

The Coffee Sand aquifer crops out predominantly in northeastern Mississippi and in central Tennessee (fig. 3). Although outcrops of the Coffee Sand occur as far north as southern Illinois, in Mississippi the unit appears to be continuous, extending northward to roughly an east-west line about 10 miles north of the Mississippi-Tennessee State line (E.F. Hollyday, U.S. Geological Survey, oral commun., 1997; W.S. Parks, U.S. Geological Survey, oral commun., 1997). To the west, in the downdip direction, the aquifer contains water with increasing dissolved-solids concentrations. To the south, the extent of the aquifer is limited by a facies change where the sand grades into chalk (Mellen, 1958). The aquifer dips about 35 feet per mile westward toward the axis of the Mississippi embayment (Boswell and others, 1965).

 

The Coffee Sand aquifer generally is composed of fine to medium quartz sand that is generally calcareous and glauconitic, with lenses of silt, sand, and clay (Boswell, 1963). Well-log data indicate the total sand thickness within the study area ranges from about 1 foot in the eastern part of the outcrop area to more than 200 feet in the downdip western part of the study area. Horizontal hydraulic conductivity values reported by Slack and Darden (1991) range from about 10 to 40 feet per day.

 

The Coffee Sand aquifer receives the majority of recharge from precipitation in the outcrop area. Water-level data indicate that discharge from the aquifer is to topographic lows in the outcrop area, to downdip areas of the Eutaw-McShan aquifer (Wasson, 1980a; Hoffmann and Hardin, 1994), and to wells screened in the Coffee Sand aquifer. The Coffee Sand aquifer is well confined from overlying aquifers by a thick sequence of chalk of the Demopolis Chalk formation (fig. 2).

Eutaw-McShan Aquifer

The Eutaw-McShan aquifer includes sediments of the Eutaw and McShan Formations. In Mississippi, these formations are considered to be a single aquifer because the sands are hydraulically connected; however, intervening beds of clay and silt may result in localized vertical head gradients.

The Eutaw-McShan aquifer crops out primarily in the northeastern part of Mississippi and northwestern part of Alabama within the study area (fig. 4). The northern and northwestern extent of the aquifer is the extent of the sediments. To the west, southwest, and south, in the downdip direction, the aquifer contains water with increasing dissolved-solids concentrations. The aquifer dips about 35 to 40 feet per mile westward toward the axis of the Mississippi embayment in the northern part and dips southwestward in the southern part.

 

The uppermost part of the Eutaw-McShan aquifer consists of the Tombigbee Sand Member, which is characterized by fine sand and silt that produces little water. The remainder of the Eutaw-McShan aquifer mainly consists of thin beds of fine- to medium-glauconitic sand (Boswell, 1963). Analysis of well-log data indicates the total sand thickness within the study area ranges from about 1 foot in the eastern part of the outcrop area to more than 300 feet in the southwestern and southern parts of the study area. An average horizontal hydraulic conductivity value of 12 feet per day, based on the results of 50 aquifer tests, was reported by Slack and Darden (1991).

 

The Eutaw-McShan aquifer receives recharge from precipitation in the outcrop area. Smaller amounts of recharge come from overlying and underlying aquifers (Mallory, 1993; Strom and Mallory, 1995). Water-level data indicate that discharge from the aquifer is to topographic lows in the outcrop area and to the Tombigbee and Black Warrior Rivers from upward leakage through units of the Selma Group (Wasson, 1980b; Gardner, 1981). The aquifer also may discharge water to the Gordo aquifer in parts of the updip area (J.H. Hoffmann, OLWR, oral commun., 1994) and to wells screened in the aquifer.

 

The Coffee Sand aquifer overlies the Eutaw-McShan aquifer, and is in turn overlain by the Ripley and lower Wilcox aquifers (fig. 2). The Eutaw-McShan aquifer is hydraulically separated from the Coffee Sand aquifer by the Mooreville Chalk south of an approximate east-west line at about the latitude of the Union and Pontotoc County boundary. North of this line, the Mooreville Chalk is absent, and the Eutaw-McShan aquifer is in contact with the Coffee Sand aquifer. Data indicate, however, that the Tombigbee Sand Member is very fine grained in this area and effectively acts as a confining unit, hydraulically separating the Eutaw-McShan and Coffee Sand aquifers (S.P. Jennings, Mississippi Office of Land and Water Resources, oral commun., 1994). The Eutaw-McShan aquifer is separated from the overlying Ripley and Lower Wilcox aquifers by thick sequences of clay and chalk in the Selma and Midway Groups (fig. 2).

Shallow Aquifers

The lower Wilcox aquifer and the Ripley aquifer occur at land surface in Union County (Boswell, 1977 and 1978). The lower Wilcox aquifer is present in the western part of Union County, and the Ripley aquifer occurs in the eastern part of the county (fig. 5). The shallow aquifers provide water to private domestic wells. The Ripley aquifer consists of sands of the Ripley Formation and the McNairy Sand Member (Boswell, 1978). The shallow aquifers in Union County, the lower Wilcox and the Ripley aquifers, are separated from the underlying Eutaw-McShan aquifer by thick clay and chalk in the Selma and Midway Groups (fig. 2).

Water Use

Municipal (public-supplied) and nonmunicipal (self-supplied commercial and industrial) water-use data were collected for 1998 for Union County. In addition to documenting the amount of water use by sector in Union County, the data were used as base-year data to calibrate the water-demand models. The ancillary information, such as number of wells and aquifer name, was used to prepare the water-use data input for the ground-water flow model. Conveyance losses and free water are referred to as public/unaccounted. For the purposes of this study, Union County is modeled as one water-service area (WSA), containing 12 public-supply systems (fig. 5). The self-supplied industrial and commercial facilities are modeled as one reporting unit, and the associated water use is included in the WSA total.

 

Municipal water withdrawals were either measured by meters at the wells (and the data provided by the public-supply system), estimated from usage metered at the connection (water sales and unaccounted for water), or estimated using an average household rate of use for the WSA and the number of connections. Nonmunicipal commercial and industrial rates of withdrawal were provided from internal reports by the facility. The public-supply systems provided billing records for 1998 for determining residential, commercial, and industrial water use and the public/unaccounted for water; the number of wells corresponding to each aquifer; price-rate structure; and number of connections. Municipal water withdrawals for Union County totaled 2.55 Mgal/d in 1998. The municipal water systems provided 1.50 Mgal/d (59 percent) for residential deliveries, 0.456 Mgal/d (18 percent) for commercial and industrial use, and 0.594 Mgal/d (23 percent) for public/unaccounted water.

 

Municipal and nonmunicipal water use for 1998 is summarized as follows: