Kansas Geological Survey, Open-File Rept. 91-1
Annual Report, FY91--Page 7 of 8

Geochemistry

Field Survey and Pumping-Test Sites

Water samples were collected from the wells at the pumping-test sites in Cloud and Lincoln counties before the pumping tests. Field measurements were made and samples also taken during the pumping tests. The waters obtained from both the deep observation (183 ft) and pumping (138 ft) wells at the Cloud County site before the pumping test were fresh, with the deeper well water containing approximately 70% more total dissolved solids based on the specific conductances than the shallower water. Both of the waters were fresher than the waters from domestic and stock wells at a farm about one-half mile to the west. The water in the deep observation well (145 ft) at the Lincoln County site was saline, whereas the water from the pumping well (96 ft) was fresh. Rattlesnake Creek to the south of the site area contains saline water that is derived from the Dakota Formation. The water collected at the end of the Cloud County pumping test had the same specific conductance and chloride concentration (within analytical error) as before the test. The specific conductance was monitored during the Lincoln County pumping test and was found to change little, with the final reading lower than the reading taken before the test. This indicates that for both sites waters were being pumped from within the more permeable strata horizontally and that no detectable amounts were being drawn from the depths of the deep observation wells. This suggests that shale layers within the Dakota can provide sufficient protection of overlying fresh ground water from deeper saline water in some locations of the aquifer system.

Water-Quality Distribution Maps

A series of maps is being prepared to display the concentration distribution of total dissolved solids (TDS), hardness, sodium-adsorption ratio, the major constituents chloride and sulfate, and the minor constituent fluoride. The TDS distribution shows where waters are fresh or saline and thus allows a general determination of the usability of waters for municipal, agricultural, and industrial supplies. Hardness represents the total calcium and magnesium expressed as calcium carbonate and is used for consideration of water uses or treatment for domestic and industrial water supplies. The sodium-adsorption ratio is calculated from the calcium, magnesium, and sodium concentrations and indicates the relative sodium hazard of an irrigation water to soil. Recommended standards for both drinking waters and irrigation waters exist for chloride contents of waters. The sulfate concentration standard presently used in Kansas is only a recommended limit. However, the federal government is proposing that a higher value be adopted for a maximum contaminant level or primary standard for sulfate; therefore this constituent will be examined to a greater extent for drinking waters in public supplies in the future. Although the standard for fluoride in public water supplies in Kansas has been increased from the former standard of 1.8 mg/L to 4 mg/L to fit the current federal standard, there are still areas in west-central Kansas where waters in the Dakota aquifer are fresh but exceed the fluoride standard.

The maps are being prepared using the geographic information system ARCINFO. Concentration contours have been digitized and are being modified as new data are added or revisions are suggested after review. The contoured values are shown either by lines or by colored intervals. An example of one of the maps is the chloride concentration distribution shown in Figure 45. Draft copies of most of the maps, including Figure 45, were shown during the Dakota Aquifer Symposium in October 1991. The maps will be included as a set in an atlas to be printed during FY92.

Figure 45. Spatial variation of chloride concentration in ground waters from the upper Dakota aquifer. The hatched area represents parts of the aquifer that outcrop or are overlain by unconsolidated sediments such as alluvium and the Ogallala aquifer. The map was prepared using the ARC/INFO geographic information system. A larger version of this figure is available; the larger version is 183k.

Water-Quality Use Assessment for Water Supplies

Assessment of the water-quality data for uses as drinking and agricultural supplies was continued as new data became available. The new data were primarily the results for samples collected and analyzed at the KGS as part of the Dakota Aquifer Program. The overall assessment based on all data available for the Dakota is summarized in Table 3. The criteria listed are those currently used or suggested by the Kansas Department of Health and Environment. The three sets of criteria for chloride and TDS concentrations are based on drinking water limits (the lowest values) and the divisions between fresh (<500 mg/L chloride and <1,000 mg/L TDS), usable (500-5,000 mg/L chloride and 1,000-10,000 mg/L TDS), and mineralized waters (>5,000 mg/L chloride and >10,000 mg/L TDS) as defined in Kansas.

Table 3. Assessment of Water-Quality Data for the Dakota Aquifer Based on Drinking water and Water-Classification Limits.^a

Constituent	Number analyzed		Concentration (mg/L)				Percent above criterion		Limit of detection	Number of < values
	Sites	Samples	Minimum	Median	Maximum	Criterion	Sites	Samples
Alkalinity	761	1208	24	254	1600	300	32.3	31.0	1	0
Ammonia-N	60	110	<0.01	0.078	5.82	0.1	53.3	46.8	0.01	19
Arsenic	130	154	<0.001	0.001	0.096	0.05	1.5	1.3	0.001	96
Barium	94	103	<0.001	0.033	0.366	1	0.0	0.0	0.001	1
Cadmium	60	66	<0.001	<0.001	0.010	0.01	0.0	0.0	0.001	49
Calcium	813	1263	1.6	71.9	2130	200	9.7	8.5	0.1	0
Chloride	866	1310	2.7	65	36500	250	22.9	24.7	0.1	0
Chloride	866	1310	2.7	65	36500	500	15.0	16.3	0.1	0
Chloride	866	1310	2.7	65	36500	5000	4.8	5.6	0.1	0
Chromium	98	101	<0.001	<0.001	0.013	0.05	0.0	0.9	0.001	24
Copper	58	69	<0.001	0.002	0.333	1	0.0	0.0	0.001	12
Diss. solids	749	1191	58	595	63800	500	58.7	55.8	1	0
Diss. solids	749	1191	58	595	63800	1000	29.2	30.1	1	0
Diss. solids	749	1191	58	595	63800	10000	5.1	6.0	1	0
Fluoride	696	1118	0.1	0.6	9	4	8.6	6.0	0.01	0
Hardness	804	1264	6.1	238	9000	400	25.0	24.8	0.1	0
Iron	638	917	<0.005	0.4	66	0.3	61.4	54.5	0.005	3
Lead	58	64	<0.001	0.003	0.040	0.05	0.0	0.0	0.001	17
Magnesium	815	1271	0.5	14	1090	150	4.8	5.4	0.1	0
Manganese	281	421	<0.005	0.06	4.3	0.05	44.1	50.6	0.005	66
Mercury	82	105	<0.0002	<0.0002	0.0056	0.002	7.3	5.7	0.0002	77
Nitrate-N	189	392	0.012	0.85	150	10	8.5	8.7	0.01	0
Phosphorus	102	190	<0.01	0.063	2.32	5	0.0	0.0	0.01	18
Potassium	452	732	0.4	4.4	156	100	0.9	0.5	0.1	0
Selenium	132	160	<0.001	0.001	0.042	0.01	6.8	6.9	0.001	78
Silica	501	887	2	16.4	60	50	0.4	0.2	0.1	0
Silver	98	101	<0.001	<0.001	0.010	0.05	0.0	0.9	0.001	41
Sodium	789	1245	3.2	73.0	22000	100	43.9	44.1	0.1	0
Sulfate	787	1239	<1	97.6	6200	250	22.2	20.7	1	1
Zinc	121	140	<0.005	0.021	2.0	5	0.0	0.0	0.005	27

^aCriteria are current primary or secondary drinking-water standards or, in the absence of a promulgated standard, suggested upper limits of the Kansas Department of Health and Environment for drinking water. The two higher limits for chloride and total dissolved solids are classification values for fresh and usable water.

A comparison of two assessments, one based on data gathered from sources other than the Dakota Aquifer Program sampling with one based on all program sampling, was made to determine similarities and differences in the percentage of standards exceeded for each constituent and property considered in the two sampling sets. The comparison assessments for the two data sets were relatively similar. However, although the percentage exceeding the recommended limits for TDS, alkalinity, and hardness in drinking waters were about the same for the two data sets, the percentage exceeding the criteria for individual major cations and chloride, sulfate, and fluoride were less for the KGS Dakota program sample set, probably because of a bias toward wells used for water supply in comparison with the previous data set, which includes many test wells. Also, earlier KGS data collected in the subcrop area in central Kansas during the study for the Kansas Corporation Commission are included in the previous data set, whereas recent KGS data have more of a bias toward the outcrop area and fringes of the subcrop area. The other main differences observed in the assessments were the higher percentage of samples exceeding selenium and arsenic limits in the recent KGS data set. However, the new federal standard to be adopted in 1992 for selenium in drinking water is 0.05 mg/L. None of the well waters collected during the Dakota Aquifer Program sampling exceed this new standard. The mercury criterion was exceeded by several percentage points in both data sets; mercury is probably the heavy metal of most concern for future examination in Dakota aquifer waters.

Table 4 lists an assessment of water quality based on radiochemical parameters for well waters collected as part of Dakota aquifer program sampling. The criteria are those currently applicable in Kansas. However, the federal government is proposing to increase the maximum contaminant level for radium-226 from 5 pCi/L to 20 PCVL. None of the well waters in Table 4 has radium-226 contents that exceed the proposed standard. The criterion for gross alpha radiation is proposed to stay the same; thus the percentage of samples exceeding will remain the same. Based on a proposed standard of 20 lig/L (0.02 mg/L) for uranium, none of the waters analyzed for this constituent will be above the limit. The maximum contaminant level for dissolved radon proposed by the U.S. Environmental Protection Agency (EPA) is quite low (300 PCVL) and corresponds to less than 10% of the natural radon content in outdoor air. If this value is accepted, over half of the Dakota waters analyzed for this parameter will be above the criterion. There has been much debate about an acceptable level for radon; the EPA is currently considering comments received from public hearings and by mail. The final radiochemical standards will be promulgated in 1993 and will become effective in 1994.

Table 4. Assessment of Water-Quality Data for the Dakota Aquifer Based on Drinking Water Limits for Radiochemical Constituents and Properties.

Property or constituent	Number of sites	Concentration or radioactivity			Criterion	Percent exceeding criterion	Limit of detection	Number of < values
		Minimum	Mean Maximum
Gross alpha radioactivity (pCi/L as Sr-90)a	26	<0.4	6.5	29	15	7.7	0.4	4
Gross beta radioactivity (pCi/L as Sr-90)	26	2.4	7.7	18	50	0.0	0.4	0
Radium-226 (pCi/L)	26	0.06	1.4	6.6	5b	3.8	0.02	0
Uranium (mg/L)	26	0.03	5.3	19.5	(20)c	(O.O)d	0.01	0
Radon-222	28	<40	428	1210	(300)c	(54.6)d	4	1

Criteria are current limits used by the Kansas Department of Health and Environment. Proposed federal standards are described in the text

1. Adjusted gross alpha radioactivity = measured gross alpha in pCi/L-uranium in pCi/L. The conversion for uranium concentration units is 0.7 pCi/microg.
2. The current standard is 5 PCi/L for combined radium-226 and radium-228.
3. The standard is proposed by the federal government and is undergoing review after public comment.
4. Based on the proposed federal standard.

Factors Controlling Water Chemistry

1. Meteoric recharge in the outcrop zone dissolves carbonate minerals to produce calcium bicarbonate waters that are fresh.

2. Pyrite in some Dakota strata is oxidized by meteoric recharge, producing solutions that add sulfate and increase calcium by dissolving more calcite to give calcium sulfate waters.

3. Recharge passing through and dissolving calcite and gypsum in overlying Upper Cretaceous rocks brings in calcium bicarbonate waters, calcium sulfate waters, or mixtures of the two waters mainly near the top of the Dakota aquifer where the confining strata of the Upper Cretaceous are thin.

4. Recharge from the overlying High Plains aquifer brings in calcium bicarbonate waters, calcium sulfate waters, or mixtures of the two waters.

5. Recharge from underlying Permian rocks, primarily the Cedar Hills Sandstone, is the main source of saltwater and produces sodium chloride water with high concentrations of calcium, magnesium, and sulfate. The saltwater source has been geochemically identified as primarily from the solution of halite (rock salt) in Permian rocks.

6. Cation exchange and calcite equilibria adjust the cation and bicarbonate contents of Dakota aquifer waters. Resultant sodium bicarbonate waters are common in the subcrop areas near the eastern outcrop band, and within the subcrop zone in west-central Kansas. Sodium sulfate waters are probably derived from the combined effects of cation exchange and mixing.

7. The final water in the aquifer at a location reflects the composition of the aquifer materials, the relative proportion of the different sources of recharge waters entering the aquifer, the order in which the recharge waters enter, the current or recent flow rate of the recharge sources, and the degree of mixing of the different recharge sources as affected by the relative vertical and horizontal permeabilities of the aquifer system. Stream-aquifer interactions and discharge zones of the aquifer are important controls on the rates and proportions of mixing of different water sources.

Coupled Geochemical and Mass Transport Model

Several mathematical models have been examined for components appropriate for simulating the coupled chemistry and flow in the saline transition zone and stream-aquifer system. One approach considered was to couple separate flow and geochemical models. In this approach the geochemical model SOLMINEQ.88 is believed to be the most suitable based on its capabilities for handling ion exchange and chemical reactions under high- and low-salinity conditions. Other programs either are too complex to be linked or run within reasonable computing times or do not have the capability needed. Both one-dimensional and two-dimensional models were considered for simulation of physical flow and chemical transport. The program SATRA-CHEM was examined for possible selection as the physical part of the coupled geochemical and flow model. An advantage of SATRA-CHEM is that it can simulate the movement of three or four solutes simultaneously. The program is also designed to be able to handle simple ion complexation and ion exchange. The procedure would be to use both the flow and simple chemical parts of SATRA-CHEM in conjunction with the more comprehensive chemical simulation by SOLMINEQ.88. Other programs to be examined include HST3D and SUTRA. However, both of these programs can simulate the transport of only one solute at a time.

The second approach to the coupled flow and chemistry simulation is to use a model in which the flow and chemistry calculations are integrated rather than passed back and forth between two linked programs. This type of model is currently favored for use in this study because making the two computer codes compatible presents substantial difficulties. After reviewing the literature, a program called HYDROGEOCHEM has been selected as the most suitable of the coupled hydrochemical models. This two-dimensional program has been under development since the early 1980's. The source code (in Fortran) of the 1989 version of HYDROGEOCHEM has been obtained from the Oak Ridge National Laboratory. An initial test indicated that there is a problem with program convergence on the Survey's computer system. One of the authors of the program is willing to provide a more recent version of the model that may solve the convergence problem.

Input data for the models are being assembled. Data for the cross-section locations in north-central Kansas are being examined to determine the major factors controlling the flow and ground-water chemistry. The top configurations of the Dakota Formation and underlying Permian System were mapped for north-central Kansas. These maps will be used to determine the upper and lower boundaries of the Dakota aquifer in the cross sections. The complicated configuration of the top of Dakota Formation in the outcrop area generally reflects the surface topography. The regional dip of the Lower Cretaceous system is to the north and northwest in the study area. Little geologic data were found for the western and northwestern parts of the mapped area, meaning that some assumptions will have to be made based on results from the overall Dakota aquifer program. Preliminary studies of the geochemical equilibria in the Dakota aquifer have been conducted to examine the controlling chemical equations that will be appropriate for later use in the model. The program SOLMINEQ.88 is being used for this purpose in conjunction with the water-quality information in the KWATCHEM data base.