C56 REGIONAL AQUIFER-SYSTEM ANALYSIS—MIDWESTERN BASINS AND ARCHES consistent with the hypothesis that regional recharge to this Midwestern Basins and Arches aquifer system — aquifers part of the aquifer system could be limited by the inability of within glacial deposits and the carbonate-rock aquifer — and the aquifer system to carry ground water away from the area. were obtained from records in the U.S. Geological Survey’s This area is largely coincident with the area of weak regional National Water Information System (NWIS) data base; files discharge (fig. 28). of the Indiana Department of Natural Resources, the Ohio Simulated discharge vectors indicate high magnitudes of Department of Natural Resources, and the Ohio Environmen- horizontal regional flow in the carbonate-rock aquifer in the tal Protection Agency; various published reports; and samples areas around the regional potentiometric highs (fig. 31B). collected as part of this investigation. The data were compiled High magnitudes of horizontal regional flow are also associ- and analyzed to investigate the ground-water chemistry of the ated with the downstream end of the Wabash and White Riv- aquifer system on a regional scale. Ground-water chemistry ers, the margin of the Illinois (structural) Basin, the Ohio of subregional areas of the Midwestern Basins and Arches River, an area west of the Scioto River, and the area east of aquifer system is described in the following reports: in Ohio, the Sandusky River. Discharge vectors along part of the Lake by Ohio Department of Natural Resources, Division of Water Erie shore indicate that the magnitude of horizontal regional (1970), Norris and Fidler (1973), Norris (1974), Deering and flow in the carbonate-rock aquifer in this area is fairly small. others (1983), Breen and Dumouchelle (1991); and in Indi- Ground-water flow may be predominantly vertical in this area ana, by Geosciences Research Associates, Inc. and Purdue because it is an area of regional ground-water discharge. Sim- University, Water Resources Research Center (1980) and ulated discharge vectors were computed for the upper weath- Indiana Department of Natural Resources (1988, 1990). ered zone water-bearing unit, but the relative magnitudes of Analyses of brines from rocks of Silurian and Devonian age flow in this poorly permeable unit are so small that the vec- are found in Stout and others (1932), Lamborn (1952), tors do not show up at the scale of figure 31B. Walker (1959), Stith (1979), Keller (1983) and Wilson and It should be noted that the discharge vectors show only Long (1993a, b). relative magnitudes of horizontal regional ground-water flow Data compiled from the literature and the available data and do not indicate flow velocities. Additional information on bases were selected on the basis of the following criteria: (1) the effective porosity of the aquifers would be necessary to major-ion concentrations (Ca, Mg, Na, Cl, SO4, and HCO3) compute flow velocities. Appropriate effective-porosity data were determined, (2) the analyses balanced electrochemically for fractured carbonate rock are difficult to obtain and were within 10 percent and, (3) lithologies of the water-producing not available for this investigation. Ground-water ages pre- units were determined. In cases where multiple analyses were sented in the following section, however, provide insight into available for a well, the most recent analysis that met the ground-water residence times. above criteria was selected. The dissolved-solids data for The calibrated final model was not used to simulate most of the analyses that were used in this report were calcu- potential effects of future pumpage on regional ground-water lated by summing the concentrations of all major constituents flow in the aquifer system. Data on future pumpage needs at according to the method described in Fishman and Friedman the regional scale are not available, and any simulations of (1989). Dissolved-solids concentrations for waters in the Illi- future pumpage at this time would be contrived. It is notewor- nois and Michigan Basins were estimated from borehole geo- thy, however, that only a small percentage of current pump- physical data where available laboratory determinations were age is associated with the regional flow systems explicitly sparse (D.J. Schnoebelen, U.S. Geological Survey, written simulated with this model. Therefore, more water associated commun., 1993). with such regional flow systems almost certainly could be New data that were collected during this investigation used. The quality of the ground water associated with some include detailed chemical and isotopic analyses of ground parts of the aquifer system, however, may limit its use. water from the aquifer system along general directions of regional ground-water flow, as determined from the map of the potentiometric surface of the carbonate-rock aquifer (fig. GEOCHEMISTRY 12), and isotopic analyses of aquifer material collected from cores of glacial deposits and carbonate rock. The locations of Geochemical data were collected from the Midwestern the ground-water and aquifer-material samples are shown in Basins and Arches aquifer system to investigate the relations figure 32. At each sampling location along four transects among ground-water chemistry, aquifer mineralogy, and across the aquifer system, ground-water samples were col- present and past patterns of regional flow. The data include a lected from the carbonate-rock aquifer, and, where possible, synthesis of basic data from more than 1,300 ground-water from a glacial aquifer. Sampling was restricted to existing analyses of water samples from the aquifer system, as well as domestic wells or test wells; wells with short open intervals detailed chemical and isotopic analyses of ground water and in the deep parts of the aquifer were generally not available. aquifer material along general directions of regional flow. The At each sampling location, an attempt was made to sample analyses represent two hydrologic units (table 1) within the the deepest available well in the carbonate-rock aquifer in GEOCHEMISTRY C57 86° 84° Kalamazoo Lake Lake Detroit St. Clair RiverMICHIGAN Michigan ONTARIO River UNITED STATES 42° Chicago CANADA Joseph Lake Erie St. Toledo D' South 557 18D Bend River 550 17D River C 14S 17G C2 Maumee 14D 546 River G2 Kankakee 13G 13S Fort 22 13D 16D 16S 21 Wayne 12S 16G 20 12D 15S Sandusky G4 C4 11S 15D 15G 9D 9S 11G 11D RiverB9G 8D G3 8G 7S 10S 7D 7G 6D 10D 10G Scioto 6G D ILLINOIS C' 6S 5G C1 5S 4S C3 G1 Wabash INDIANA 5D River B' OHIO 4D A' 40° 4G Columbus River 2D 3S 2S 19 2G 3D G5 Miami Indianapolis River A 1D Dayton Great White Cincinnati River White Ohio Fork River East Kentucky Ohio Licking River Huntington KENTUCKY River Frankfort WEST River Louisville River VIRGINIA Evansville Ohio 38° Lexington Base from U.S. Geological Survey digital data, 0 20 40 60 MILES 1:2,000,000, 1972 0 20 40 60 KILOMETERS EXPLANATION Boundary of study area 3D Carbonate-rock aquifer, deep D D' Location of geochemical section Core sampled for isotopic analysis— Number refers to table 9 Well sampled for chemical analysis during this investigation— G2 Number refers to tables 6, 7, and 8 Core samples collected during this investigation 2G Glacial aquifers 557 Mineral samples collected by Botomon and Fauer (1976) 11S Carbonate-rock aquifer, shallow FIGURE 32.—Locations of wells sampled during this investigation, carbonate-rock and glacial cores sampled for isotopic analysis, and geochemical sections A–A', B–B', C–C', and D–D'. C58 REGIONAL AQUIFER-SYSTEM ANALYSIS—MIDWESTERN BASINS AND ARCHES order to intersect the dominant regional ground-water flow used to delimit zones of potable water in the carbonate-rock paths in the aquifer system. The deep bedrock wells are com- aquifer, to examine the relation between the glacial aquifers pleted from 100 to nearly 450 ft into the carbonate-rock aqui- and the carbonate-rock aquifer, and to evaluate possible fer. To evaluate the effects of sampling the deep wells with geochemical and hydrological processes that control the dis- long open intervals, RASA investigators also sampled a tribution of major dissolved solutes. nearby well with a short open interval. These wells were shal- Generally, dissolved-solids concentrations in ground low and completed in the top 5 to 50 ft of the carbonate-rock water increase along flow paths — from the surface to the sat- aquifer. Well depth and length of open interval for each well urated zone and through the aquifer — because of dissolution sampled are listed in table 6. Deep bedrock wells that are of minerals (Freeze and Cherry, 1979). Additional processes cased through the top part of the carbonate-rock aquifer are such as evaporation or evapotranspiration can increase con- also noted in table 6. Where possible, a glacial well, a shallow centrations of dissolved solutes. Some ground waters, notably bedrock well, and a deep bedrock well within several miles of brines, contain extremely high concentrations of dissolved each other were identified and grouped as a sample site. solids because of combined effects of mineral dissolution and Ground-water samples were collected during 1991-92 and evaporation. Thus, dissolved-solids concentrations generally analyzed for major and minor constituents according to the can be used as an indicator of the degree of chemical evolu- methods described in Fishman and Friedman (1989). Field tion of ground water in an aquifer system. determinations were made for pH, temperature, and alkalinity The predominant major ions in ground water can also be according to methods described in Wood (1976). Dissolved used as indicators of the important chemical and hydrologic sulfide was determined in the field with a Hach DR-2000 processes in an aquifer system. Chebotarev (1955) character- spectrophotometer according to the procedure described in ized regional changes in the dominant anion species in the instrument manual (Hach Chemical Company, 1989).