U.S. Geological Survey, Hydrologic atlas 730-D James A. Miller and Cynthia L. Appel, 1997 REGIONAL SUMMARY INTRODUCTION The three States‹Kansas, Missouri, and Nebraska‹that comprise Segment 3 of this Atlas are in the central part of the United States. The major rivers that drain these States are the Niobrara, the Platte, the Kansas, the Arkansas, and the Missouri; the Mississippi River is the eastern boundary of the area. These rivers supply water for many uses but ground water is the source of slightly more than one-half of the total water withdrawn for all uses within the three-State area. The aquifers that contain the water consist of consolidated sedimentary rocks and unconsolidated deposits that range in age from Cambrian through Quaternary. This chapter describes the geology and hydrology of each of the principal aquifers throughout the three-State area. Some water enters Segment 3 as inflow from rivers and aquifers that cross the segment boundaries, but precipitation, as rain and snow, is the primary source of water within the area. Average annual precipitation (1951­80) increases from west to east and ranges from about 16 to 48 inches (fig. 1). The climate of the western one-third of Kansas and Nebraska, where the average annual precipitation generally is less than 20 inches per year, is considered to be semiarid. This area receives little precipitation chiefly because it is distant from the Gulf of Mexico, which is the principal source of moisture-laden air for the entire segment, but partly because it is located in the rain shadow of the Rocky Mountains. Average annual precipitation is greatest in southeastern Missouri. Much of the precipitation is returned to the atmosphere by evapotranspiration, which is the combination of evaporation from the land surface and surface-water bodies, and transpiration from plants. Some of the precipitation either flows directly into streams as overland runoff or percolates into the soil and then moves downward into aquifers where it is stored for a time and subsequently released as base flow to streams. Average annual runoff, which is the total discharge into a stream from surface- and ground-water sources, ranges from about 0.2 inch in the western part of the area to about 20 inches in southeastern Missouri (fig. 2). Average annual runoff generally reflects the distribution of average annual precipitation during the same period. However, runoff is less than precipitation everywhere and ranges from less than 5 to about 35 percent of the average annual precipitation. Evapotranspiration rates are high, especially in the western one-half of the area; thus, only a small percentage of the precipitation is available to recharge aquifers in most places. Locally, however, runoff might be significantly less than shown in figure 2, and ground-water recharge, greater, especially where highly permeable rocks or deposits at the land surface allow precipitation to rapidly infiltrate. Examples of such places are the Sand Hills area of Nebraska, which is blanketed by permeable windblown sands, and parts of southern Missouri, where permeable limestone is at or near the land surface. The land surface of Segment 3 generally slopes gradually from west to east. In the Great Plains Physiographic Province (fig. 3), the altitude of the flat land surface locally is about 5,000 feet above sea level in westernmost Nebraska. By contrast, in the flat Coastal Plain Physiographic Province of eastern Missouri, the altitude is about 500 feet above sea level. The land surface is gently rolling in the Central Lowland Province except where major rivers and their tributaries are deeply incised. In the Ozark Plateaus Physiographic Province, rugged topography has developed where the underlying rocks have been uplifted and deeply eroded. AREAL DISTRIBUTION OF AQUIFERS The numerous aquifers within Segment 3 vary in composition. Some of the aquifers are unconsolidated sand and gravel, some are semiconsolidated sediments, and some are consolidated sandstone, limestone, or dolomite. The aquifers have been defined primarily on the basis of differences in their rock types and ground-water flow systems and secondarily by the chemical quality of water they contain. Some of the aquifers are grouped into aquifer systems. An aquifer system consists of two or more aquifers that are hydraulically connected. The flow systems of the connected aquifers function similarly, and a change in conditions in one of the aquifers affects the other aquifer or aquifers. Seven principal aquifers or aquifer systems are at the land surface in the three-State area. The extent of those aquifers which primarily consist of unconsolidated deposits of late Quaternary age and are collectively called the surficial aquifer system is shown in figure 4. The remaining six aquifers and aquifer systems primarily consist of semiconsolidated to consolidated sedimentary rocks; figure 5 shows where these aquifers are exposed or covered with only a thin blanket of soil and unconsolidated material. Some of the aquifers extend into the subsurface far beyond the areas where they are mapped in figure 5. One additional aquifer system, the Western Interior Plains, is present only in the subsurface and, therefore, is not shown in the figure. This aquifer system contains saline water or brine and is not as well known as the aquifers that primarily contain freshwater. In this report, the dissolved-solids concentration in ground water is used to classify the water as fresh, saline, or brine. The concentrations used to categorize the water are as follows: Dissolved-solids concentration, in milligrams per liter Freshwater less than 1,000 Slightly saline water 1,000 to 3,000 Moderately saline water 3,000 to 10,000 Very saline water 10,000 to 35,000 Brine Greater than 35,000 Aquifers that are part of the surficial aquifer system are in all three States of Segment 3 (fig. 4). In this report, the surficial aquifer system is divided into the following parts: stream-valley aquifers, the Mississippi River Valley alluvial aquifer, and glacial-drift aquifers. The stream-valley aquifers are the most extensive part of the system and consist of sand and gravel deposited as alluvium in and adjacent to the channels of the larger streams in the segment. The Mississippi River Valley alluvial aquifer in southeastern Missouri also consists of alluvial sand and gravel, but these materials have been deposited as a thick, wide blanket as the channel of the Mississippi River changed its position over time. The glacial-drift aquifers consist of sand and gravel that were deposited during multiple advances of continental ice sheets from the north and northwest, primarily during the Pleistocene Epoch. Rock and soil particles were planed from the land surface as the massive sheets of ice advanced and were transported by the ice. Some of these materials were redistributed by meltwater and were deposited in preglacial channels as stratified sand and gravel that formed productive aquifers. In contrast, poorly sorted unstratified glacial deposits of clay, silt, sand, gravel, and boulders (called till) and stratified clay and silt deposited in glacial lakes have minimal permeability. Some of the glacial-drift aquifers in eastern Nebraska, northeastern Kansas, and northern Missouri are buried beneath till or glacial-lake deposits. The High Plains aquifer (fig. 5), which is at the land surface in most of Nebraska and a large part of Kansas, is the most productive aquifer in the segment. This aquifer mostly consists of unconsolidated to consolidated sand and gravel of Quaternary and Tertiary to age which were deposited as a broad, thick sheet of alluvium on a wide, gentle plain by a network of branching streams whose channels migrated across the plain. Dune sand that covers an area of about 20,000 square miles in Nebraska is part of the High Plains aquifer where the sand is saturated. Where the stream-valley aquifers overlie the High Plains aquifer, they are connected hydraulically to the aquifer and are considered to be part of it. The Mississippi embayment aquifer system in southeastern Missouri underlies and is in hydraulic connection with the Mississippi River Valley alluvial aquifer. Semiconsolidated sands of Tertiary and Cretaceous age compose the Mississippi embayment aquifer system. The Great Plains aquifer system is exposed at the land surface in a band that extends from south-central Kansas to northeastern Nebraska (fig. 5). This aquifer system consists of two sandstone aquifers in Cretaceous rocks, separated by a shale confining unit. Although the Great Plains aquifer system extends in the subsurface throughout Kansas and Nebraska, it contains saline water in many places northward and westward from the area where it is exposed. A thick confining unit composed of Cretaceous shale, chalk, and limestone formations overlies the Great Plains aquifer system (figs. 6, 7) and separates it from the High Plains aquifer in most places. The Ozark Plateaus aquifer system is exposed at the land surface in most of southern Missouri and in a small part of southeasternmost Kansas (fig. 5). This aquifer system consists of three aquifers that are separated by two confining units, all in Paleozoic rocks. The upper two aquifers are predominantly carbonate rocks, whereas the lower aquifer is predominantly sandstone. The Ozark Plateaus aquifer system extends northwestward for more than 50 miles beneath a thick confining unit called the Western Interior Plains confining unit (fig. 6). This confining unit extends throughout Kansas and Nebraska and consists of poorly permeable sedimentary rocks of variable composition that range in age from Jurassic through late Mississippian. Permeable carbonate rocks that are the subsurface equi-valents of the aquifers of the Ozark Plateaus aquifer system are called the Western Interior Plains aquifer system (fig. 6). Because this system is deeply buried everywhere, it contains saline water or brine and its hydrology, therefore, is not well known. The Mississippian aquifer in northeastern Missouri (fig. 5) is in carbonate rocks that are stratigraphically equivalent to those that compose the uppermost aquifer of the Ozark Plateaus aquifer system. However, the ground-water flow systems of the two aquifers are not connected east of Boone County, Missouri, and the Mississippian aquifer is considered to be separate from the Ozark Plateaus aquifer system in this report. The Cambrian­Ordovician aquifer is exposed in a small part of northeastern Missouri (fig. 5). This aquifer mostly consists of carbonate rocks and contains freshwater only in a band about 50 miles wide, which is parallel to and north of the Missouri River from Boone County eastward to the Mississippi River. The rocks that contain the Cambrian­Ordovician aquifer are stratigraphically equivalent to those that form part of the middle aquifer of the Ozark Plateaus aquifer system. The degree to which these aquifers are hydraulically connected is not precisely known, but the two aquifers are considered to be partly continuous in this report. GEOLOGY The geologic and hydrogeologic nomenclature used in this report differs from State to State because of independent geologic interpretations and varied distribution and lithology of rock units. A fairly consistent set of nomenclature, however, can be derived from the most commonly used rock names. Therefore, the nomenclature used in this report is a synthesis of that of the U.S. Geological Survey, the Kansas Geological Survey, the Missouri Department of Natural Resources, Division of Geology and Land Survey, and the Nebraska Conservation and Survey Division of the University of Nebraska. Individual sources for nomenclature are listed with each correlation chart prepared for this report. Kansas, Missouri, and Nebraska are in part of the North American craton, which is an area that has been tectonically stable throughout most of geologic time. The area has undergone some deformation, however, as shown by faults and by upwarps and downwarps on the surface of the crystalline Precambrian rocks (fig. 8) that underlie Paleozoic and younger sedimentary rocks everywhere. Precambrian rocks are exposed only in the St. Francois Mountains of southeastern Missouri, where they are locally more than 1,000 feet above sea level; these rocks are buried to depths of as much as 6,000 feet below sea level in southwestern Kansas on the northern flank of the Anadarko Basin. Because the crystalline-rock surface slopes outward in all directions from the Ozark Uplift and northward or southwestward from high areas along the Chadron Arch and the Central Kansas Uplift (fig. 8), the overlying sequence of sedimentary rocks slopes and thickens away from all these high areas. The greatest sedimentary rock accumulations are in the Salina Basin in south-central Nebraska and north-central Kansas and in the parts of the Mississippi Embayment and the Anadarko, the Denver, the Kennedy, and the Forest City Basins that are in Segment 3. For example, total sedimentary rock thickness in the southwestern part of the Nebraska panhandle and in southwestern Kansas along the Oklahoma State line is about 9,000 feet. Numerous faults in the crystalline rocks of the three-State area are grouped mostly in or adjacent to the Central Kansas Uplift, the Nemaha Uplift, and the St. Francois Mountains. Vertical displacement across the faults varies from less than 100 to more than 2,500 feet. Displacement generally is greatest across some of the faults in the north-trending fault zone just east of the Nemaha Uplift. This fault zone and a second zone (not shown in fig. 8) that trends northeastward across the Mississippi Embayment are thought to represent zones of continental rifting that formed during Precambrian time. Postdepositional erosion of the Paleozoic sedimentary-rock sequence from eastern Missouri to central Kansas and eastern Nebraska has beveled off some of the rocks. As a consequence, progressively younger rocks are exposed to the west and northwest of the Precambrian core of the St. Francois Mountains in southeastern Missouri (fig. 9). The glacial sediments that cover bedrock strata in eastern Nebraska, northeastern Kansas, and northern Missouri are not shown in figure 9, and stream-valley deposits are shown only along the major streams; the total extent of these deposits in Segment 3 is the same as that shown in figure 4. The widespread areas of Tertiary and Quaternary sediments in western Kansas and Nebraska are not related to erosion or beveling of rocks away from the St. Francois Mountains and the Ozark Uplift. These Tertiary and Quaternary sediments are mostly alluvium that was derived from erosion of the Rocky Mountains to the west of the segment. Cambrian rocks are exposed in southeastern Missouri in an area that encircles the Precambrian core of the St. Francois Mountains. The basal Cambrian rocks are sandstones, and the upper parts of the Cambrian sequence are mostly dolomite. Ordovician rocks are exposed in a large area in southern Missouri and in smaller areas in northeastern and southeast-ern Missouri. The thick sequence of Ordovician strata mostly consists of dolomite and limestone interbedded with minor sandstone and shale and has been divided into a large number of geologic formations. Silurian rocks consist of a thin sequence of dolomite and limestone and are exposed only locally in southeastern and northeastern Missouri near the Mississippi River. Some studies have postulated that the Silurian Period was characterized by extensive uplift and erosion in the area of Segment 3. Devonian rocks are exposed in scattered areas of southern, southeastern, and northern Missouri. Like the Silurian strata, Devonian rocks are thin. The lower and middle parts of the Devonian sequence are mostly limestone interbedded with minor sandstone and chert, whereas the upper part is mostly widespread shale. Mississippian rocks crop out in a wide to narrow band that extends from southwestern Missouri to just north of the Missouri River in central Missouri and as a second, less extensive band in northeastern Missouri parallel to the Mississippi River. Mississippian strata in Segment 3 are mostly limestone (commonly cherty) but include some beds of sandstone and shale. Outliers of Mississippian rocks in southern Missouri show that these beds extended over a much larger area before most were removed by erosion. Pennsylvanian strata crop out in large areas of eastern Kansas and western Missouri. These rocks are covered with glacial drift to the west and north of the Missouri River where they are mapped in figure 9 as the shallowest bedrock. Pennsylvanian rocks consist of shale, sandstone, limestone, and some coal beds and were deposited in a series of sedimentary cycles, each of which represents a transgression and regression of the Pennsylvanian sea. Each cycle, known as a cyclo-them (fig. 10), begins and ends with nonmarine shale deposits; intervening marine limestone and shale were deposited in shallow to deep water. The thick Pennsylvanian section has been divided into a large number of geologic formations, especially in Kansas where 49 formations are recognized in exposed Pennsylvanian strata and several additional formations are delineated in the subsurface. Outliers of Pennsylvanian rocks in east-central Missouri show that before they were partly eroded, these strata covered a much greater area than at present. Permian rocks are exposed in a wide to narrow band that extends from south-central Kansas to southeastern Nebraska. Permian rocks primarily consist of shale and sandstone but also contain beds of halite (rock salt), gypsum, anhydrite, and minor limestone. Cyclic deposition is characteristic of Permian strata, but the cycles are not as numerous as those of the Pennsylvanian rocks. Triassic and Jurassic rocks are present in the subsurface of western Kansas and Nebraska. These rocks mostly are shale, siltstone, and dolomite, but some Jurassic sandstone beds locally yield small amounts of water. Triassic and Jurassic rocks are not shown in figure 9. Cretaceous rocks are exposed in large areas of central Kansas and eastern Nebraska and smaller areas in southeastern Missouri and western Kansas and Nebraska. Cretaceous strata in Nebraska and Kansas consist largely of shale, but prominent, widespread sandstones are in the lower part of the Cretaceous section, and equally widespread limestone and chalk units are in the upper part. Semiconsolidated sand and clay form the Cretaceous beds of Missouri. Tertiary and Quaternary deposits are the most widespread geologic unit in Segment 3 and are especially prominent in Kansas and Nebraska. They are characterized mainly by unconsolidated sand and gravel, but locally include beds of sandstone, siltstone, silt, and clay. The Quaternary and Tertiary deposits mapped along the major stream courses in the segment consist primarily of unconsolidated sand and gravel. In the Missouri bootheel, Tertiary beds consist of unconsolidated to semiconsolidated clay and sand overlain by unconsolidated Quaternary sand and gravel. VERTICAL SEQUENCE OF AQUIFERS Some of the principal aquifers and aquifer systems in Segment 3 are stacked atop others. For example, in parts of Kansas and Nebraska, the High Plains aquifer overlies the Great Plains aquifer system, which, in turn, overlies the Western Interior Plains aquifer system (fig. 7); the aquifers and aquifer systems, however, are separated by thick shale confining units in most places. Although the confining units are poorly permeable, some water is able to move vertically through them, from one aquifer to another. Movement is in the direction of decreasing hydraulic head and is easiest where the confining units are thin, leaky, or both. Where confining units are absent, water moves readily between aquifers. As an example, where stream-valley aquifers of the surficial aquifer system overlie the High Plains aquifer, no confining unit separates the aquifers, both of which consist of unconsolidated sand and gravel. The aquifers cannot be hydraulically distinguished from each other, and the stream-valley aquifers are considered to be part of the High Plains aquifer where the two are in contact. The sequence of maps (figures 11 through 15) shows the extent of each aquifer or aquifer system. Comparison of the maps shows the places where aquifers are stacked upon each other. The vertical sequence is different from area to area, and no single location contains all the aquifers. The uppermost aquifers in the segment are in unconsolidated sand and gravel of the surficial aquifer system (fig. 11). This system has been subdivided into three parts (stream-valley alluvial aquifers, Mississippi River Valley alluvial aquifer, glacial-drift aquifers), primarily on the basis of differences in the origin and geometry of the permeable material that composes the aquifers. The aquifers primarily contain water under unconfined (water-table) conditions and mostly consist of sediments of Quaternary age. The High Plains aquifer (fig. 12) underlies part of the surficial aquifer in Nebraska and Kansas. No confining unit separates the two aquifers. The principal geologic unit in the High Plains aquifer in Nebraska and western Kansas is the Ogallala Formation, which mostly consists of unconsolidated sand and gravel; locally, the High Plains aquifer is called the Ogallala aquifer. In south-central Kansas, the aquifer comprises mostly Quaternary sediments. Although clay beds create local confined conditions, most of the water in the aquifer is unconfined. The High Plains aquifer is an extremely important source of water, primarily for irrigated agriculture, in Segment 3. The Mississippi embayment aquifer system (fig. 13) underlies part of the surficial aquifer system in the bootheel and adjacent counties of Missouri. No confining unit separates the two aquifer systems. Unconsolidated to semiconsolidated sand aquifers, separated by clayey confining units, compose the Mississippi embayment aquifer system. The Great Plains aquifer system extends from its southern and eastern limits continuously northward and westward through Kansas and Nebraska (fig. 13) except for a small area in northwestern Nebraska where the system is missing. Two sandstone aquifers in Lower Cretaceous rocks, separated by a shale confining unit, compose the aquifer system. An extremely thick shale confining unit underlies the aquifer system almost everywhere. Water in the Great Plains aquifer system is under confined conditions in most places. Exceptions are where the aquifer system is exposed at the land surface or is directly overlain by the High Plains aquifer; in these places, water-table conditions exist in much of the aquifer. The Ozark Plateaus aquifer system extends over most of southern Missouri (fig. 14) and consists of three aquifers that are separated by two confining units, all in consolidated rocks of Paleozoic age. The uppermost aquifer is in Mississippian carbonate rocks; stratigraphically equivalent carbonate rocks in northern Missouri are called the Mississippian aquifer (fig. 14). The middle aquifer of the Ozark Plateaus aquifer system is in carbonate rocks of Cambrian and Ordovician age, and the lowermost aquifer in the system is in Cambrian sandstones. The confining units that separate the aquifers are dolomite and shale. Water in the aquifers of the Ozark Plateaus aquifer system and the Mississippian aquifer is unconfined in and just downgradient from aquifer outcrop areas but is confined elsewhere. Water is confined in the Mississippian aquifer in most places by poorly permeable Pennsylvanian strata that overlie the aquifer. A thick shale confining unit overlies the Ozark Plateaus aquifer system westward from western Missouri and southeasternmost Kansas. This confining unit is exceedingly thick and poorly permeable. Water-yielding rocks beneath the confining unit compose the Western Interior Plains aquifer system (fig. 14) in Paleozoic rocks that are lateral equivalents of the aquifers of the Ozark Plateaus aquifer system. The Western Interior Plains aquifer system is entirely in the subsurface and contains slightly saline water or brine that is under confined conditions everywhere. The Ozark Plateaus and the Western Interior Plains aquifer systems directly overlie poorly permeable crystalline rocks, whereas the Mississippian aquifer is underlain by a confining unit of shale and carbonate rocks. The Cambrian­Ordovician aquifer in northern Missouri (fig. 15) consists of several water-yielding beds of sandstone and dolomite. Some of the permeable strata in this aquifer are equivalent to parts of the middle aquifer of the Ozark Plateaus aquifer system, but the Missouri River is a discharge area for both aquifers and, thus, hydraulically separates them in some places. Water in the Cambrian­Ordovician aquifer is confined in most places. The aquifer is underlain by a confining unit of Cambrian shale, dolomite, and limestone. FRESH GROUND-WATER WITHDRAWALS Ground water is the source of water supply for more than 5 million people, or almost 70 percent of the population in the three-State area (table 1). Public water-supply systems provide more than twice as much water as private (domestic) sys-tems. Ground water supplies nearly 100 percent of the population in rural areas and is the source for many water-supply systems in small cities. About 86 percent of the population of Nebraska depends on ground water for supply. Nearly 10 billion gallons per day was withdrawn from all the aquifers in Segment 3 during 1990 (fig. 16). About 90 percent of the total water withdrawn was used for agricultural, primarily irrigation, purposes. Withdrawals for public supply were about 6 percent of the total water withdrawn. Total fresh ground-water withdrawals, by county, during 1990 in Kansas, Missouri, and Nebraska are shown in figure 17. Counties with the largest withdrawals are those in which agricultural irrigation is most intense. Some large cities located adjacent to major rivers (for example, St. Louis, Missouri) withdraw surface water for public supply, and their effect is accordingly not indicated on the map. The total freshwater withdrawn from each principal aquifer and aquifer system in the three-State area is shown in figure 18. About 8,191 million gallons per day was withdrawn from the High Plains aquifer; this was about 8 times as much water as was withdrawn from the surficial aquifer system, which is the second most used source of ground water (1,037 million gallons per day). The Ozark Plateaus aquifer system supplied water at the rate of about 330 million gallons per day and is the third largest producer. Withdrawals from the Great Plains aquifer, which is the fourth largest producer in the segment, were about 133 million gallons per day. Withdrawal rates from the Mississippi embayment aquifer system were small (95 million gallons per day) because the aquifer is limited in areal extent in Segment 3 and is overlain by the productive Mississippi River Valley alluvial aquifer. SURFICIAL AQUIFER SYSTEM INTRODUCTION The surficial aquifer system in Segment 3 consists of unconsolidated sand and gravel and is divided into three parts: stream-valley aquifers, the Mississippi River Valley alluvial aquifer, and glacial-drift aquifers. These aquifers are hydraulically connected in some places. For example, many of the glacial-drift aquifers in northern Missouri, northeastern Kansas, and eastern Nebraska occupy ancient stream channels that have been eroded into bedrock. At locations where modern streams follow the ancient drainage patterns, the alluvial deposits of sand and gravel that compose a stream-valley aquifer may lie directly on glacial outwash that also consists of sand and gravel. Much of the sand and gravel of the stream-valley aquifers in Missouri and eastern Kansas and Nebraska has been reworked from older glacial-drift deposits and, therefore, may be difficult to distinguish from glacial outwash. Most of the water in the surficial aquifer system is under unconfined conditions. STREAM-VALLEY AQUIFERS The stream-valley aquifers of Segment 3 consist of narrow bands of fluvial and alluvial sediments which fill or partly fill the valleys of meandering to braided streams that have eroded shallow channels into glacial deposits, older unconsolidated alluvium, or bedrock. Where these streams cross the High Plains aquifer, the stream-valley aquifers are hydraulically connected to and are considered to be part of the underlying High Plains aquifer. Locally, the stream-valley aquifers are hydraulically connected to bedrock aquifers, but, in most places, they are separated from the bedrock aquifers by poorly permeable beds of clay or shale. The extent of the stream-valley aquifers is shown in figure 19. The unconsolidated sand and gravel deposits that compose the stream-valley aquifers are thicker, more widespread, and more productive in the valleys of the larger rivers than those of smaller streams. In Kansas, stream-valley aquifers are along the courses of the Republican, the Kansas, the Missouri, the Solomon, the Saline, the Neosho, the Smoky Hill, the Marais des Cygnes, the Arkansas, and the Cimarron Rivers. In Missouri, the stream-valley aquifers along the Missouri and the Mississippi Rivers and their tributaries are important sources of freshwater for many communities and industries. Stream-valley aquifers occur along the Missouri, the Niobrara, the Loup, the Platte, the Republican, and the Blue Rivers in Nebraska. No comprehensive, unified study has been done for the stream-valley aquifers; accordingly, local investigations have been selected to show their hydrology. The designated boundaries D8 and D9 in figure 19 are the locations of detailed studies of the stream-valley aquifers, which are described in the following sections of this report. The stream-valley aquifers consist mostly of sand and gravel of Holocene age but locally include sediments of Pleistocene age. The average thickness of the aquifers is about 90 to 100 feet, but locally they are as much as 160 feet thick. However, the average thickness of saturated alluvial material is less and generally ranges from 50 to 80 feet; the thicker saturated sections yield more water to wells. Most of the water in the stream-valley aquifers is under unconfined, or water-table, conditions. Locally, where coarse-grained aquifer sediments are capped by poorly permeable silt or clay, confined (artesian) conditions exist. The stream-valley aquifers are in direct hydraulic connection with the adjacent streams and water levels in the aquifers are, therefore, closely related to river levels (fig. 20). Aquifer and river water levels rise following precipitation events; the rise in the wa-ter level of the river precedes that in the aquifer. Recharge to a typical stream-valley aquifer is by precipitation that falls directly on the aquifer, seepage through the beds of streams and of reservoirs and canals constructed in the stream valleys, downward percolation of applied irrigation water (fig. 21), and ground-water inflow from underlying, permeable bedrock. The aquifer discharges by leakage to streams and canals, pumpage from wells, and evapotranspiration (evaporation plus transpiration by plants, especially crops during the growing season). A small amount of water is consumed by crops. Along reaches of some streams, such as the Arkansas, the Smoky Hill, and the Solomon, some of the water potentially available to recharge the aquifer from the stream is diverted by networks of canals and irrigation ditches, from which evaporation occurs. Such diversions, coupled with intense irrigation pumpage from the aquifer and a resulting decrease in base flow to the stream, have severely reduced streamflow. In some cases, streams that were formerly perennial are now dry most of the year. The stream-valley aquifers are reliable sources of ground water because of the coarse-grained nature and high permeability of the aquifer material. Yields that range from 100 to 1,000 gallons per minute commonly are reported for wells completed in these aquifers; maximum yields of more than 1,500 to 2,500 gallons per minute are reported locally in Nebraska and Missouri, respectively, and yields of as much as 3,000 gallons per minute are reported from stream-valley aquifers in Kansas. Reported transmissivity values for these aquifers, as calculated from aquifer tests, range from 8,000 to 80,000 square feet per day. The chemical quality of the water in the stream-valley aquifers generally is suitable for most uses. Typically, the water is hard and a calcium bicarbonate type. Dissolved-solids concentrations generally are less than 500 milligrams per liter but locally are as much as 7,000 milligrams per liter; the larger concentrations reflect an influx of water with large chloride or sulfate concentrations from underlying aquifers or from irrigation return flow. Large iron concentrations are common. Arkansas River Valley, Southwestern Kansas The stream-valley aquifer that borders the Arkansas River in southwestern Kansas (area D8 in fig. 19) has been studied along a 48-mile reach of the river between the Colorado-Kansas State line and a geologic structure called the Bear Creek Fault Zone (fig. 22). This zone is thought to have been formed by the collapse of bedrock following dissolution of underlying salt beds. East of this zone the stream-valley aquifer is hydraulically connected to the High Plains aquifer and is, therefore, considered to be part of it. The stream-valley aquifer shown in figure 22 primarily comprises alluvial sand and gravel that fill a valley which is as much as 5 miles wide and has been eroded into poorly permeable Cretaceous bedrock. The sand and gravel range in thickness from 0 to 125 feet and generally are thickest near the center of the valley; however, the saturated thickness of this permeable material is less than 75 feet except in small areas (fig. 23). Well yields are directly proportional to saturated thickness. Where the saturated thickness is 25 feet or less, well yields are between 100 and 500 gallons per minute, but where saturated thickness is 75 feet or more, yields that range from 1,000 to 3,000 gallons per minute have been reported. The Arkansas River and the stream-valley aquifer are hydraulically connected and water moves freely from the aquifer into the stream or from the stream into the aquifer, depending upon the relative position of the water levels in the stream and aquifer. The flow of the Arkansas River is impounded by the John Martin Reservoir in Colorado, about 50 miles upstream from the Colorado-Kansas State line, and water is released from the reservoir during the growing season to supply downstream irrigation ditches and canals. Beginning about 1970, the flow of the river was insufficient to meet irrigation needs near the river in southwestern Kansas (fig. 24), partly because of increased diversions upstream. The river was a perennial stream throughout the studied 48-mile reach until 1975 but was dry most of the year east of (downstream from) Kendall from 1975 through 1980. Accordingly, farmers who used ditch irrigation changed from surface water to ground water as a source of irrigation supply; this change, coupled with increased development, caused withdrawals from the stream-valley aquifer to more than triple between 1970, when about 20,000 acre-feet per year was pumped, and 1979, when withdrawals were about 65,000 acre-feet per year. (One acre-foot is the volume of water that will cover 1 acre of land to a depth of 1 foot, or about 43,560 cubic feet of water.) The combination of decreased streamflow from Colorado, less-than-normal precipitation, and increased pumpage caused water levels in the aquifer to decline 4 feet in the western part of Hamilton County (fig. 25) and more than 25 feet in Kearny County. Water in the stream-valley aquifer moves from west to east (fig. 26) and follows the eastward slope of the topography of the valley floor. Movement of the water in the aquifer is nearly parallel to streamflow. In places, the altitude of the water table is below the altitude of the stream, and the contours show a slight ³v² that points in a downstream direction (for example, the area near Medway). In other places, the contours ³v² slightly upstream (for example, near Mayline), which indicates the aquifer is discharging to the stream. The lack of sharp, well-defined ³v² points on the contours indicates good hydraulic connection between the stream and the aquifer. The depth to the water table also is shown in figure 26. The water table is nearest to land surface near the river and is deepest near the edge of the valley. Water levels in the stream-valley aquifer and in the streams to which it is connected vary seasonally in response to pumping from the aquifer. The effects of pumping are summarized in figure 27. Before withdrawals began, the water table in the aquifer was above the water level in the stream, and the aquifer discharged to the stream as shown in figure 27A. Wells installed to provide irrigation supplies can greatly lower the water table in the aquifer, especially during the summer, when crop growth rate is highest and water demands are greatest. At such times, ground-water movement can be reversed from that before pumping began (fig. 27B), and water can move from the stream to recharge the aquifer. When irrigation pumping ceases during the winter months, movement of water from the aquifer to the stream is reestablished; however, water levels may not recover fully to those before pumping began. If pumping is greatly increased (fig. 27C), however, then the regional water table may decline to a level below the streambed; the stream will flow only after periods of heavy rainfall and will lose water to the aquifer during all seasons. This condition exists along some reaches of the Arkansas River in Kansas. The hydrographs in figure 28 show the response of the water level in a well 100 feet from the Arkansas River near Kendall to changes in the river stage. From 1979 through 1981, the water level in the aquifer near the river was generally below or at the same level as the river stage. Thus, the river is a losing stream here and contributes water to the aquifer. The hydrograph shows that the river ceased to flow for several months in late 1979; thus, the aquifer received no recharge from the stream. Smoky Hill, Saline, And Solomon River Valleys, Central Kansas The stream-valley aquifer along parts of the Smoky Hill, the Saline, and the Solomon Rivers (area D9 in fig. 19) ranges in width from about 3 to 5 miles (fig. 29). The lower two-thirds of the aquifer generally consists of coarse-grained alluvial deposits, whereas the upper third is finer grained material. The uplands adjacent to the rivers are underlain by sandstone, shale, and limestone of Cretaceous to Permian age (fig. 29), all of which are less permeable than the alluvium. Locally, the stream-valley alluvium is bordered by Quaternary dune sands and poorly sorted terrace deposits that are mostly unsaturated and not part of the stream-valley aquifer. Shale beds underlie the stream-valley aquifer and form a confining unit that separates the aquifer from permeable beds in the Hutchinson Salt Member of the Wellington Formation of Permian age. Wells completed in the coarse-grained, lower part of the stream-valley aquifer commonly yield from 200 to 900 gallons per minute. Transmissivity values determined from aquifer tests in this part of the alluvium range from 8,000 to 13,000 feet squared per day. The water in the aquifer generally is unconfined. The stream-valley aquifer is in direct hydraulic contact with the three rivers and discharges water to them. The stream-valley aquifer mostly contains freshwater in its upper 35 to 50 feet in the area shown in figure 29. Below the freshwater from Salina eastward, the aquifer contains saline water, some of which discharges to some reaches of the rivers. During a period of stable base flow in 1976­77, the chloride concentration in water from the Smoky Hill River increased about 800 milligrams per liter in the reach of the river between New Cambria and Sand Springs. In the same period, an increase in chloride concentration of about 550 milligrams per liter was observed in water from the Solomon River in the reach between Niles and Solomon. Chloride concentrations as large as 73,000 milligrams per liter have been reported in water from the lower part of the stream-valley aquifer. Withdrawals from the aquifer in this part of central Kansas are small because of the poor quality of the water. The source of the saline water and brine in the stream-valley aquifer is the Hutchinson Salt Member of the Wellington Formation (fig. 30). Fresh ground water from the alluvial aquifer has circulated downward and dissolved some of the salt and evaporite beds in the Hutchinson Salt Member and the unnamed lower member of the Wellington Formation. Where this dissolution locally has increased the porosity and permeability of the Wellington, the formation is called the Wellington aquifer. Collapse structures have formed in the shale confin-ing unit that separates the Wellington and the stream-valley aquifers and allow brine to move upward from the Wellington aquifer (fig. 30) to the stream-valley aquifer. The brine is diluted as it mixes with freshwater and moves through the stream-valley aquifer to discharge into the Smoky Hill and the Solomon Rivers. Water in the stream-valley and the Wellington aquifers generally moves from west to east, parallel to the direction of flow of the Smoky Hill, the Solomon, and the Saline Rivers (fig. 31). From Salina westward, the water table in the stream-valley aquifer is higher than the potentiometric surface of the Wellington aquifer, which indicates that water moves downward from the alluvial material into the Wellington aquifer. As the water moves eastward through the Wellington aquifer, it partially dissolves rock salt and evaporite minerals in the Wellington Formation; as a result, the chloride and the dissolved-solids concentrations in the water increase. East of the confluence of the Saline River and Mulberry Creek, the potentiometric surface of the Wellington aquifer is higher than the water table in the stream-valley aquifer, and the poor-quality water moves upward into the stream-valley aquifer. This water subsequently discharges into the Smoky Hill and the Solomon Rivers as base flow. During periods of low rainfall, when most of the flow in these rivers is derived from ground-water discharge, the water in the rivers is unusable. Concentrations of chloride and dissolved solids in the river water decrease as rainfall increases and the water is diluted by surface runoff. Saline springs and seeps occur to the south in Oklahoma and Texas where circulating ground water brings to the surface saline water derived from the partial dissolution of salt beds in the manner described above. The saline water there contains tritium, a radioactive isotope of hydrogen that originates chiefly as a product of hydrogen bomb explosions. The concentrations of tritium in water from the salt springs and seeps show that the saline water has moved from recharge to discharge areas in less than 40 years. Ground-water circulation in the Smoky Hill-Solomon-Saline River Valley area may be equally rapid. Missouri River Valley, Missouri Alluvial deposits along the Missouri River form an important stream-valley aquifer from the Iowa-Missouri State line to the junction of the Missouri and the Mississippi Rivers (fig. 32); small areas of similar deposits in eastern Nebraska compose local aquifers. The deposits partly fill an entrenched bedrock valley that ranges from about 2 to 10 miles wide. In many places in northern Missouri, the bedrock contains slightly saline to saline water, and the stream-valley aquifers, along with aquifers in glacial drift, are the only sources of fresh ground water. The part of the stream-valley aquifer along the Missouri River between St. Charles and Jefferson City, Missouri (area D9 in fig. 19) is described below. The stream-valley aquifer consists of clay, silt, sand, and gravel. Gravel and sand generally are most common in the lower parts of the aquifer (fig. 33). Poorly permeable silt and clay are prominent in the upper part of the aquifer and locally create confined conditions. Sandstone, limestone, dolomite, and shale of Pennsylvanian and Mississippian age mostly compose the bedrock that underlies the stream-valley aquifer in western Missouri. From the Howard­Boone County line eastward, the bedrock consists of Ordovician limestone and dolomite. In upland areas, glacial deposits overlie the bedrock and locally are hydraulically connected to the stream-valley aquifer. The alluvial material of the stream-valley aquifer averages about 90 feet in thickness but is locally as much as 160 feet thick. The saturated thickness of the aquifer averages about 80 feet. Reported yields of wells completed in the aquifer range from less than 100 to about 3,000 gallons per minute. Recharge to the stream-valley aquifer is by infiltration of precipitation, seepage of water from the Missouri River to the aquifer during periods of high streamflow, and inflow from bedrock aquifers. Discharge from the aquifer is by evapotranspiration, withdrawals by wells, and seepage to the Missouri River during periods of low streamflow. The general direction of water movement in the stream-valley aquifer is downstream and toward the river (fig. 34). Water in the stream-valley aquifer is a calcium bicarbonate type and is characterized by excessive iron content and hardness; in many places, the water is softened before use. Dissolved-solids concentrations in water from the aquifer range from about 250 to 1,500 milligrams per liter and are largest in areas where saline water leaks upward from bedrock and is diluted by mixing with freshwater. During 1990, an average of about 147 million gallons of water per day was withdrawn from the stream-valley aquifer (fig. 35). About 45 percent of this amount, or about 66 million gallons per day, was used for public supply. Industrial, mining, and thermoelectric power withdrawals amounted to about 48 million gallons per day, and agricultural withdrawals were about 24 million gallons per day. The remainder of the water withdrawn (about 9 million gallons per day) was used for domestic and commercial purposes. MISSISSIPPI RIVER VALLEY ALLUVIAL AQUIFER Alluvial material adjacent to the Mississippi River forms an important aquifer in the northern part of the Mississippi Embayment. This aquifer, which is called the Mississippi River Valley alluvial aquifer, is in parts of Arkansas, Illinois, Kentucky, Louisiana, Mississippi, Missouri, and Tennessee (fig. 36). It is most extensive, and is described in more detail, in Segment 5 of this Atlas. The part of the aquifer that is located in the bootheel of Missouri is the principal source of irrigation water there. The Mississippi River Valley alluvial aquifer in Missouri consists of sand and gravel with minor amounts of silt and clay. The alluvium that composes the aquifer was deposited by the ancestral Mississippi and Ohio Rivers during Quaternary time and is bounded to the north by Paleozoic rocks that are exposed along the Ozark Escarpment (fig. 37). The thickness of the alluvium ranges from a featheredge along the escarpment to more than 250 feet locally near the Mississippi River and generally increases southward and southeastward (figs. 37, 38). The thickest areas probably represent infilling of ancient stream channels. The alluvium is thin or absent from Crowleys Ridge, Hickory Ridge, and the Benton Hills. These ridges and hills rise as much as 250 feet above the surface of the surrounding alluvial plain and are underlain by rocks of early Tertiary, Cretaceous, and Paleozoic age. The coarse-grained sediments of the Mississippi River Valley alluvial aquifer are highly permeable. Wells completed in the aquifer will yield 1,000 gallons per minute in most places, and, locally, yields of 3,000 gallons per minute have been reported. Well depths commonly are 100 feet or less. Transmissivity values for the alluvial aquifer, which were calculated from aquifer test data, range from 15,000 to 54,000 square feet per day and average about 40,000 square feet per day. This productive aquifer is the major source of water for the intense agricultural development in the area. The water in the Mississippi River Valley alluvial aquifer is mostly unconfined and water levels rise rapidly in response to rainfall. Near major streams, aquifer water levels rise and fall in response to changes in stream water levels. Water in the aquifer moves generally southward from topographically high areas near the Ozark Escarpment (fig. 39). Local high areas on the potentiometric surface represent sandy ridges where the aquifer receives larger amounts of recharge. The poorly permeable rocks that underlie Crowleys Ridge and the Benton Hills form local barriers to ground-water flow. An extensive network of agricultural drainage ditches (fig. 40) has been constructed, and the aquifer discharges to these ditches and to major streams. The chemical quality of the water in the Mississippi River Valley alluvial aquifer generally meets the standards recom-mended for public water supplies by the U.S. Environmental Protection Agency; locally, excessive concentrations of iron and manganese have been reported. Iron concentrations in water from the aquifer locally are as much as 35 milligrams per liter and average about 4.3 milligrams per liter; manganese concentrations locally are as much as 2.0 milligrams per liter and average 0.46 milligram per liter. The water is a calcium-magnesium bicarbonate type, generally hard, and has small dissolved-solids concentrations (averaging 240 milligrams per liter). Locally, the water in the aquifer contains traces of pesticides and nutrients as a result of downward leakance of irrigation water from fields that have been treated with chemicals for insect control or with fertilizer. Withdrawals of freshwater from the Mississippi River Valley alluvial aquifer totaled 129 million gallons per day during 1990 (fig. 41). About 92 million gallons per day was withdrawn for agricultural purposes, which is the principal water use. About 22 million gallons per day was withdrawn for public supply, and about 10 million gallons per day was pumped for industrial, mining, and thermoelectric power uses. Withdrawals for domestic and commercial uses were about 5 million gallons per day. GLACIAL-DRIFT AQUIFERS The maximum southern extent of glacial ice and glacial-drift deposits was about the present location of the Missouri River in Missouri and just south of the Kansas River in northeastern Kansas. The glacial deposits in Segment 3 are pre-Illinoian and, thus, are older than deposits in States to the north and east of the segment. Some of the drift in Segment 3 might be of late Pliocene age, whereas most glacial deposits in North America are considered to be Pleistocene. Although deposits of glacial drift extend over wide areas, most were laid down directly by the ice; are fine grained, poorly sorted, or both; and, therefore, yield only small amounts of water to wells. The thickness of glacial drift generally is 100 to 200 feet but locally is greater than 300 feet in eastern Missouri and 400 feet in western Missouri and northeastern Kansas. In southeastern Nebraska, local drift thicknesses of more than 350 feet have been reported. Meltwater created an extensive stream network in front of the advancing ice (fig. 42), and the streams deposited gravel, sand, and finer sediments as alluvium along the courses of preglacial bedrock valleys. Complex interbedding of fine- and coarse-grained material is characteristic of the glacial deposits (fig. 43). The lenslike shape of some of the beds is the result of meandering of the meltwater streams across their valley floors and of periodic changes in stream-channel locations. However, in parts of Missouri, the glacial-drift aquifers are not complexly interbedded. For example, in the Grand River Valley of Daviess County, Missouri, the basal part of the deposits that fill glacial stream channels is coarse grained, and the upper part generally consists of poorly permeable silt, clay, or till (fig. 44). Such aquifers are called buried channel or buried valley aquifers and contain water under confined or semiconfined conditions. Not all the drift consists of sand and gravel, (shown in figures 43 and 44), and not all is saturated. Water generally is obtained from sand beds that range from 20 to 40 feet in thickness. Yields of wells completed in the glacial-drift aquifers are highly variable and range from less than 10 to about 1,000 gallons per minute. Large diameter wells that penetrate several thick, saturated, highly permeable sand beds yield the most water. Even in places where wells penetrate only one thin sand bed in the glacial-drift deposits, yields are generally larger than those of wells completed in the underlying bedrock. Transmissivity values that range from 200 to 13,000 feet squared per day have been reported from aquifer tests in glacial-drift aquifers in Kansas. The larger transmissivity values represent places where several thick sand beds were encountered by wells; the smaller values indicate that thin sand beds with low permeability were penetrated. Movement of water in the glacial-drift aquifers is from recharge areas to discharge areas along major modern streams. Much of the water moves along short flow paths to the nearest surface-water body, where it discharges. Some water follows longer flow paths and discharges to regional drains. A small amount of the water percolates downward and enters un-derlying bedrock aquifers. The complex interbedding of permeable and poorly permeable sediments in the glacial- drift aquifers results in a large number of local confining units. Accordingly, water in these aquifers is under unconfined conditions in some places and confined conditions in other places. Where several sand and clay beds are stacked, water levels in each of the stacked sand beds may be different. The potentiometric surface of the glacial-drift aquifers is, therefore, a composite surface and shows only the general configuration of water levels. The influence of topography on water levels in these aquifers is shown by a map of their potentiometric surface in Missouri (fig. 45). Topographically high areas in Clinton and Sullivan Counties, for example, stand out clearly. The low water levels parallel to the courses of the Missouri and the Mississippi Rivers and some of their tributaries show that the aquifer discharges water to these streams. The chemical quality of the water in the glacial-drift aquifers generally is suitable for most uses. The water is hard and commonly is a calcium bicarbonate type although in many places in Missouri and locally in Kansas, it is a sodium sulfate type. Dissolved-solids concentrations in water from these aquifers usually are less than 500 milligrams per liter but exceed 3,900 milligrams per liter in places. Sulfate concentrations ordinarily are 250 milligrams per liter or less except locally in Kansas and in Missouri; concentrations of sulfate as great as 2,150 milligrams per liter have been reported in Missouri. The source of the sulfate is dissolution of gypsum in the underlying bedrock in areas where the hydraulic head in the bedrock is greater than that in the glacial-drift aquifers; this condition allows the high-sulfate water to leak upward. Locally, concentrations of as much as 30 milligrams per liter of iron have been reported in Missouri. Nitrate concentrations that are greater than 45 milligrams per liter have been reported in water from these aquifers in Kansas, which probably reflects local contamination from agricultural sources or human wastes. HIGH PLAINS AQUIFER INTRODUCTION The High Plains aquifer underlies an area of about 174,000 square miles in parts of Colorado, Kansas, Nebraska, New Mexico, Oklahoma, South Dakota, Texas, and Wyoming (fig. 46). Parts of the aquifer are in Segments 2, 3, 4, 8, and 9 of this Atlas; for the most part, the aquifer is within Segments 3 and 4 and is discussed in detail only in the chapters that describe those segments. The High Plains aquifer is the principal source of ground water for the High Plains region, which is one of the Nation¹s most important agricultural areas. In Nebraska, the aquifer underlies an area of about 63,650 square miles, and in Kansas, it underlies an area of about 30,500 square miles; only these parts of the aquifer are discussed in this chapter. The High Plains aquifer is named from the High Plains Physiographic Province, an area of flat to gently rolling topography (fig. 47) which is part of a vast plain that slopes gently eastward from the Rocky Mountains. The plain was formed by the deposition of sediments that were transported eastward from the Rocky Mountains by a network of streams. Subse-quent uplift and erosion have partly dissected the plain. In places, extensive areas of windblown silt and sand that were derived from channel deposits of the streams are at the land surface. The windblown sand deposits form dunes that cover an area of about 20,000 square miles in central Nebraska. Local dune sands also are common in parts of southern Kansas. The economy of the High Plains area, which provides a major part of the food supply of the Nation, is dependent on the successful growing of crops. The prevailing method of farming in the High Plains before the drought of 1930 through 1939 was dryland farming. By the 1930¹s, continuous cropping, primarily by repeatedly planting the cropland in wheat, had depleted the humus that bound the soil. During the drought, the wind blew away much of the remaining pulverized soil as huge clouds of dust, and the area was known as the ³Dust Bowl.² Ground-water irrigation, which had begun in the late 1800¹s, was greatly intensified in the 1940¹s in response to the drought. A second surge in irrigation development followed a severe drought during the 1950¹s. During 1985, about 10.5 million acres were irrigated on the High Plains in Nebraska and Kansas. Most of the water that supplies these irrigation needs was withdrawn from the High Plains aquifer. HYDROGEOLOGIC UNITS The High Plains aquifer consists of all or parts of several geologic units of Quaternary and Tertiary age. The stratigraphic column in figure 48 shows the formation name, generalized rock type, thickness, and age of the geologic units that compose the aquifer. The Brule Formation of Oligocene age is the oldest geologic unit included in the aquifer. The Brule Formation is the upper unit of the White River Group and is primarily massive siltstone with beds and channel deposits of sandstone. Locally, the Brule includes lenticular beds of volcanic ash, clay, and fine sand. The Brule underlies much of western Nebraska and is included in the aquifer only where it has been fractured or where the formation contains solution openings. Such secondary porosity and permeability are developed only where the Brule crops out or is near the land surface (fig. 49). The Arikaree Group of Miocene and Oligocene age overlies the Brule Formation and consists primarily of massive, very fine to fine-grained sandstone. Locally, the Arikaree includes beds of volcanic ash, siltstone, claystone, and marl. The Arikaree Group crops out in western Nebraska and pinches out to the south and east as does the White River Group, which includes the Brule Formation (fig. 50). The maximum thickness of the Arikaree is about 1,000 feet in western Nebraska. The Ogallala Formation of Miocene age is the principal geologic unit included in the High Plains aquifer and is at the land surface throughout most of the extent of the aquifer (fig. 49). The Ogallala consists of unconsolidated gravel, sand, silt, and clay. Locally, it also includes caliche, which is a hard deposit of calcium carbonate that precipitated when part of the ground water that moved through the formation evaporated. The Ogallala Formation was deposited by an extensive eastward-flowing system of braided streams that drained the eastern slopes of the Rocky Mountains during late Tertiary time. The location of the stream system migrated during a long period of time, and the Ogallala Formation was deposited over about 134,000 square miles in eastern Colorado, Kansas, Nebraska, New Mexico, Oklahoma, South Dakota, Texas, and Wyoming. Unconsolidated deposits of Quaternary age overlie the Ogallala Formation. These Quaternary deposits consist of gravel, sand, silt, and clay, much of which is reworked material that was derived from the Ogallala Formation. Where these unconsolidated deposits are saturated, such as in southeastern Nebraska and south-central Kansas, they compose part of the High Plains aquifer (fig. 49). Deposits of loess (fig. 51) overlie the Ogallala Formation or the unconsolidated Quaternary sediments in some locations. The loess was deposited as windblown material and consists mostly of silt with small quantities of very fine-grained sand and clay. Where the loess is thick, it forms the upper confining unit of the High Plains aquifer. Dune sands of Quaternary age compose part of the aquifer where they are saturated. The dune sands are most extensive in west-central Nebraska where they cover about 20,000 square miles (fig. 49) and attain a maximum thickness of about 300 feet. Saturated dune sands also are part of the High Plains aquifer south of the Arkansas River in southwest and south-central Kansas. The dune sands are highly porous and, therefore, quickly absorb rainfall that recharges the High Plains aquifer. Valley-fill deposits along the channels of streams, such as the Platte and the Arkansas Rivers, also are considered to be part of the aquifer where they are hydraulically connected to it. In such places, the valley-fill deposits directly link the streams to the High Plains aquifer and allow water to move freely between the aquifer and the streams. The High Plains aquifer is underlain by rocks that range in age from Tertiary to Permian. Rocks of Permian age directly underlie parts of the aquifer in southern Kansas (fig. 52). These rocks are predominantly red shale, siltstone, sandstone, gypsum, anhydrite, and dolomite and locally include limestone and halite (rock salt) as beds or disseminated grains. Partial dissolution of salt and evaporite minerals by circulating ground water has adversely affected the chemical quality of water in the High Plains aquifer where Permian rocks that contain salt beds or saline water are in hydraulic connection with the aquifer. The dissolution of salt beds also has resulted in collapse structures and faulting in the overlying deposits. The Bear Creek and the Crooked Creek Fault Zones in southwestern Kansas (fig. 52) are collapse structures that have formed as a result of partial dissolution of salt beds. Rocks of Jurassic and Triassic age directly underlie the High Plains aquifer in small parts of southwestern Kansas and western Nebraska. These rocks consist primarily of shale and sandstone, and some of the sandstone beds are permeable enough to yield water to wells. Some irrigation wells in southwestern Kansas withdraw water from the High Plains aquifer and the rocks of Jurassic and Triassic age. For the most part, however, the Jurassic and Triassic rocks have low permeability. Lower Cretaceous rocks directly underlie the High Plains aquifer in parts of southern Kansas and eastern Nebraska (fig. 52). These rocks are primarily shale and sandstone. The hydraulic properties of the sandstones are highly variable, but Lower Cretaceous rocks provide water for irrigation and other uses in parts of Kansas and Nebraska. Upper Cretaceous rocks directly underlie the High Plains aquifer in large parts of Nebraska and Kansas. These rocks consist primarily of shale, chalk, limestone, and sandstone of which only the chalk (where it is fractured, contains solution openings, or both) yields quantities of water large enough for irrigation purposes. Elsewhere, Upper Cretaceous rocks have little permeability. The Chadron Formation that is part of the White River Group of Tertiary age (fig. 48) directly underlies the High Plains aquifer in most of western Nebraska (fig. 52). The Chadron Formation is predominantly clay and silt, both with minimal permeability. The Brule Formation, which also is part of the White River Group, is predominantly siltstone but locally is fractured. Where it contains fracture or solution permeability, the Brule Formation is considered to be part of the High Plains aquifer. GROUND-WATER HYDROLOGY Depth to Water The depth to water in a particular area is the difference between the altitude of land surface and the altitude of the water table. The generalized depth to water in the High Plains aquifer in 1980 is shown in figure 53. In most places, the water levels shown are lower than those that existed before widespread irrigation withdrawals began. The depth to water in the High Plains aquifer is less than 100 feet in about one-half of the area of the aquifer and less than 200 feet in most of Nebraska and Kansas. The depth to water generally is less near the Platte and the Arkansas Rivers than in areas farther from the rivers because the rivers are hydraulically connected to the aquifer through the stream valley aquifers that parallel the rivers. The water table is between 200 and 300 feet below the land surface in parts of western and southwestern Nebraska and in parts of southwestern Kansas. The depth to water is as much as 400 feet below the surface in a small area in southwestern Kansas where development of the aquifer began earlier than in most parts of Kansas; consequently, water-level declines are greater. Ground-Water Flow Water in the High Plains aquifer generally is under unconfined, or water-table, conditions. Locally, water levels in wells completed in some parts of the aquifer may rise slightly above the regional water table because of artesian pressure created by local confining beds. The altitude and configuration of the water table of the High Plains aquifer are shown in figure 54. The configuration and slope of the water table are similar to the configuration and slope of the land surface. Water in the aquifer generally moves from west to east, or perpendicular to the contours and in the direction of the arrows shown in figure 54. Water moves in response to the slope of the water table, which typically averages between 10 and 15 feet per mile. On the basis of this average slope and aquifer hydraulic properties, the velocity of water that moves through the aquifer is estimated to average about 1 foot per day. Where the water-table contours cross streams, the configuration of the contours indicates the relation of the water in the aquifer to the water in the stream. For example, where the contours from 3,200 to 4,000 feet in figure 54 cross the North Platte River in western Nebraska, the contours bend upstream. This upstream flexture indicates that water moves from the aquifer to the stream, and the North Platte River is a gaining stream in this area. By contrast, where the 2,000-foot contour crosses the Platte River in west-central Nebraska, a slight downstream bend in the contour indicates that water is moving from the stream to the aquifer; the Platte River is a losing stream in this area, and the water from the river recharges the aquifer. In southwestern Kansas, the Bear Creek and the Crooked Creek Fault Zones (fig. 52) have displaced the High Plains aquifer and little or no saturated thickness of the aquifer exists on the upthrown side of the faults. In these areas, the water-table contours shown in figure 54 end abruptly at the faults. The spacing of the water-table contours is affected by different hydrologic conditions. For example, where contours are widely separated, such as in western Nebraska, the slope of the water table is gentler than where the contours are more closely spaced. Widespread recharge to the aquifer by infiltration of precipitation through dune sands occurs in western Nebraska, and, thus, the slope of the water table is relatively gentle. Recharge and Discharge In an undisturbed ground-water flow system, the amount of water that moves into an aquifer (recharge) and the amount of water that moves out of the aquifer (discharge) are equal, and the flow system is in equilibrium. Before development in an unconfined aquifer, such as the High Plains aquifer, the water table of the aquifer and the quantity of water stored in the aquifer vary little in response to changes in precipitation, streamflow, and the amount and types of vegetation. A ground-water flow system is no longer in equilibrium when the long-term discharge is not equal to the long-term recharge. The altitude of the water table rises when the recharge rate exceeds the discharge rate and declines when the discharge rate exceeds the recharge rate. Withdrawal of large quantities of ground water by wells and redistribution of surface water in ditches and canals, all for irrigation purposes, have changed the natural recharge and discharge of the High Plains aquifer. Recharge to the High Plains aquifer is primarily by infiltration of precipitation and locally is by infiltration from streams and canals. Some surface water that is applied to crops for irrigation also percolates downward and recharges the aquifer. A small quantity of water from the underlying bedrock moves upward and mixes with water in the High Plains aquifer; this water is also considered to be recharge. The aquifer is recharged at total rates of between 0.05 and 6 inches per year in Nebraska and Kansas. The rates of recharge are highly variable and range from about 0.3 to 20 percent of the average annual precipitation in the dry and wet parts of these States. The greatest rates of recharge by precipitation are in areas where dune sand or other highly permeable material is at the land surface. Recharge by infiltration of streamflow usually is greatest when streamflow is high and, thus, provides a large difference between stream and aquifer water levels. Natural discharge from the High Plains aquifer is to springs, seeps, and streams and by evapotranspiration. Where the water table is near the land surface, ground water can evaporate directly. Transpiration rates are greatest along stream valleys where deep-rooted salt cedar, willows, cottonwoods, and sedges grow. Where the High Plains aquifer locally is underlain by permeable bedrock and the water table in the aquifer is higher than that in the bedrock, small amounts of water move downward from the aquifer into the bedrock. Large quantities of water are withdrawn from the aquifer by wells, and in some areas large amounts of water discharge from the aquifer to streams. For example, a study was done during 1975 to determine how much of the flow of the Platte River in Nebraska was derived from ground water. The gain in streamflow within Nebraska was about 3 million acre-feet, most of which was ground-water discharge from the High Plains aquifer to the river. Large quantities of surface water are diverted from the Platte River and used for irrigation; thus, the amount of ground-water discharge to the river probably was significantly greater than the measured gain in streamflow. Most of the discharge from the High Plains aquifer is by withdrawals from wells, and practically all of the water withdrawn is used for irrigation purposes. Total withdrawals of water from the entire aquifer for irrigation increased from about 4 million acre-feet during 1949 to about 23 million acre-feet during 1978 (fig. 55) then decreased to about 16 million acre-feet during 1990. During 1978, more than 4 million acre-feet of irrigation water was withdrawn in Kansas and about 8 million acre-feet was withdrawn in Nebraska. During 1990, irrigation withdrawals in Kansas were about the same as those during 1978, whereas withdrawals in Nebraska were only about 5 million acre-feet per day. Saturated Thickness and Well Yield The saturated thickness of an aquifer is the vertical distance between the water table and the base of the aquifer and is one of the factors that determines the quantity of water that can be pumped from a well. Other factors that affect well yield include well construction and the hydraulic properties of the aquifer. The saturated thickness of the High Plains aquifer in 1980 (fig. 56) ranged from 0 (where the sediments that compose the aquifer were unsaturated) to about 1,000 feet. The greatest saturated thickness is in north-central Nebraska where the aquifer consists of the Ogallala Formation and overlying dune sands. Locally in southwestern Kansas, dissolution of salt in the Permian bedrock that underlies the High Plains aquifer has resulted in collapse features that were filled with younger sediments. These anomalously thick accumulations of sediments coincide with thick sequences of saturated aquifer materials. The average saturated thickness of the entire aquifer in 1992 was about 190 feet. In Nebraska, the average saturated thickness was 340 feet, but in Kansas, it was only about 90 feet. Changes in the saturated thickness of the High Plains aquifer have resulted from ground-water development. Saturated thickness has decreased in most places (fig. 57), but in two areas in south-central Nebraska, recharge to the aquifer from surface-water irrigation combined with downward leakage of water from canals and reservoirs has increased saturated thickness. In large areas in southwestern Kansas, large-scale irrigation development has decreased the saturated thickness of the aquifer more than 25 percent. Decreases of more than 10 percent in saturated thickness result in a decrease in well yields and an increase in pumping costs because of the increased depth at which the pump must be set in order to lift the water. The entire High Plains aquifer contained about 21.7 billion acre-feet of saturated material in 1992. The quantity of drainable water in storage in the aquifer can be estimated by multiplying the volume of saturated material by the average specific yield (15 percent). The specific yield of an aquifer is the volume of water that will drain from a unit volume of rock under the influence of gravity alone. Therefore, about 3.25 billion acre-feet of drainable water was in storage in the High Plains aquifer in 1992. About 65 percent of the drainable water in storage in the entire aquifer is in Nebraska, and about 10 percent is in Kansas. The remaining drainable water is in storage in Colorado, New Mexico, Texas, South Dakota, Oklahoma, and Wyoming. Not all drainable water in storage within the aquifer can be recovered for use. The quantity of water that can be recovered varies with location and depends on the lithology, saturated thickness, hydraulic conductivity, and specific yield of the aquifer at that location and on well construction. Water has been almost completely removed from about 8 percent of the formerly saturated aquifer material in Kansas, whereas the quantity of material dewatered in Nebraska is negligible. The greatest yields of water generally are obtained from wells that are completed in coarse-grained aquifer material in places where the saturated thickness of the High Plains aquifer is great. A generalized map of the potential yield of properly constructed wells completed in the High Plains aquifer is shown in figure 58. The potential yield of wells is greater than 750 gallons per minute in most of Nebraska and large parts of Kansas. A well capable of producing 750 gallons per minute can irrigate 125 acres and effectively supply one center-pivot irrigation system. Well yields from different formations that compose the aquifer vary. Yields from the Brule Formation typically are less than 300 gallons per minute. Wells completed in the Arikaree Group generally do not yield large quantities of water but might yield as much as 350 gallons per minute in western Nebraska where the saturated thickness is about 200 feet. Well yields from the Brule Formation and the Arikaree Group are greatest where secondary porosity, such as fractures or solution openings, has been developed in the rocks. Well yields from the Ogallala Formation are 1,000 gallons per minute from 100 feet of saturated sand and gravel in many parts of Kansas and Nebraska but are only 100 gallons per minute from 20 feet of saturated sand and gravel in western Kansas. GROUND-WATER DEVELOPMENT AND WATER-LEVEL FLUCTUATIONS Development of the High Plains aquifer began in the late 1800¹s, when windmills were used as a source of power to pump water from scattered irrigation wells. Spurred by the drought of the 1930¹s, ground-water irrigation expanded rapidly. Development of the High Plains aquifer generally began in Texas, adjacent parts of New Mexico, and in major stream valleys in other States in the 1930¹s. Widespread development progressed to Oklahoma and Kansas in the 1940¹s and extended to Colorado, Nebraska, and Wyoming in the 1950¹s; the aquifer has undergone little development in South Dakota. In 1949, the total acreage irrigated by ground water in the High Plains was slightly more than 2 million acres (fig. 59), most of which was in Texas. The number of acres irrigated in Kansas and Nebraska expanded greatly in the late 1950¹s in response to a drought. The amount of irrigated acreage in those States continued to increase until 1978, by which time nearly 170,000 irrigation wells had been completed in the High Plains aquifer; about 23,000 of these wells were in Kansas, and about 59,300 were in Nebraska. Collectively, irrigation wells in eight States pumped about 23 million acre-feet of water from the High Plains aquifer in 1978 to irrigate about 13 million acres. In 1978, water from the High Plains aquifer was used to irrigate about 4.5 million acres in Nebraska and more than 2 million acres in Kansas. The large increase in the number of wells drilled for irrigation between 1952 and 1978 was partially a result of the development of center-pivot irrigation systems during the 1960¹s. Center-pivot systems, such as those shown from an aerial view in figure 60, are supplied by irrigation wells and have the water-distribution pipes mounted on a wheeled boom that rotates in a circle around the center of the irrigated area. Such systems make it possible to irrigate the rolling terrain of the High Plains. In Nebraska alone, the number of center-pivot irrigation systems increased from about 2,800 in 1972 to more than 27,000 by the end of 1984 (fig. 61). As the number of irrigation wells increased, the percentage of land that was irrigated also increased. In 1949, the 5 percent or less of land that was irrigated in most of Nebraska and Kansas (fig. 62A), was mostly along river valleys. By 1964, the percentage of irrigated land had increased, and much of the irrigated acreage was away from the rivers (fig. 62B). By 1978, the percentage of irrigated acreage had greatly increased (fig. 62C) as more upland areas were irrigated. As development of the High Plains aquifer became more extensive, water levels in the aquifer began to decline in some locations. Well yields decreased in some places as a result of the water-level declines. The cost of pumping increased as water levels declined because pumps were set deeper and more energy was required to lift the water an increased distance. The cost of the energy used to pump the water also increased. This increased cost for obtaining irrigation water decreased the profit in growing crops that require irrigation in parts of the High Plains area. Average annual withdrawals of water from the High Plains aquifer generally are much larger than recharge to the aquifer from precipitation. In places where recharge rates are high, the demand for irrigation water could be as low as twice the average annual recharge; where recharge rates are low, the demand could be more than 100 times the recharge. The quantity of water removed from storage from the entire aquifer between predevelopment conditions and 1980 is estimated to be about 166 million acre-feet, of which about 27 million acre-feet was withdrawn in Kansas. No significant quantity of water has been removed from storage in Nebraska. By 1980, withdrawals of water from the High Plains aquifer had resulted in water-level declines of more than 100 feet in parts of southwestern Kansas (fig. 63). Development of the aquifer in this area began in the 1940¹s, which is earlier than in most other places in Kansas. Generally, the later development of the aquifer began in an area and the less intense the development has been, the smaller the water-level decline in that area. Water-level rises shown in parts of southern Nebraska are in response to increased recharge of the aquifer by infiltration of surface water applied for irrigation. GROUND-WATER QUALITY The chemical quality of water in the High Plains aquifer is affected by many factors. These factors include the chemical composition and solubility of aquifer materials, the increase in dissolved-solids concentrations in ground water in areas where the water discharges by evapotranspiration, and the chemical composition of water that recharges the aquifer. Ground water generally contains smaller concentrations of dissolved minerals near recharge areas where the residence time of the water in the aquifer has been short, and, thus, dissolution of aquifer minerals has been less. The water generally is more mineralized near discharge areas because residence time has been longer and more dissolution of minerals has taken place. The dissolved-solids concentration in ground water is a general indicator of the chemical quality of the water. Dissolved-solids concentrations in water from the High Plains aquifer are less than 500 milligrams per liter in most of Kansas and Nebraska (fig. 64) but locally exceed 1,000 milligrams per liter in both States. The limit of dissolved solids recommended by the U.S. Environmental Protection Agency for drinking water is 500 milligrams per liter. Most crops can tolerate water in which the dissolved-solids concentration is 500 milligrams per liter or less. In places with well-drained soils, many types of crops can tolerate water with a dissolved-solids concentration of between 500 and 1,500 milligrams per liter. In southwestern and south-central Kansas, the High Plains aquifer overlies Permian bedrock that contains bedded salt. Where circulating ground water has dissolved some of this salt and the mineralized water has subsequently moved upward into the High Plains aquifer, the dissolved-solids concentration of the water in the High Plains aquifer is greatly increased. Also, dissolved-solids concentrations generally are greater near streams where water from the High Plains aquifer discharges. Ground water near the streams is shallow enough to be transpired by plants or to be evaporated directly from the soil. Concentrations of dissolved solids in the ground water are increased by the evapotranspiration process. Rates of transpiration are greatest where deep-rooted phreatophytes, such as sedges, cottonwood, willows, and salt cedar, grow. Excessive concentrations of sodium in water adversely affect plant growth and soil properties, and constitute salinity and sodium hazards that may limit irrigation development. Sodium that has been concentrated in the soil by evapotranspiration and ion exchange decreases soil tillability and permeability. Areas of high or very high sodium hazard occur in parts of Kansas. Sodium hazard is evaluated by the sodium adsorption ratio, which relates the concentration of sodium to calcium plus magnesium; if this ratio is high, then the sodium can destroy any clay in the soil and thus affect soil structure. Sodium concentrations in water from the High Plains aquifer in Kansas and Nebraska are shown in figure 65. Concentrations are less than 25 milligrams per liter in most of Nebraska and northern Kansas. Concentrations are greatest in southwestern Kansas where evapotranspiration rates are high and in south-central Kansas where the High Plains aquifer overlies Permian bedrock that contains saline water derived from partial dissolution of salt beds. Sodium concentrations are increased along the Platte and the Republican Rivers where evapotranspiration rates also are high. Salinity and sodium hazards generally are low in Nebraska where the High Plains aquifer primarily consists of sand and gravel, which contain few sodium-bearing minerals. Excessive fluoride concentrations are a widespread problem in water from the High Plains aquifer. Some of the fluoride is derived from dissolution of fluoride-bearing minerals in parts of the aquifer that contain sand and gravel, such as the Ogallala Formation. Extremely large concentrations (2­8 milligrams per liter) of fluoride are reported where the aquifer contains volcanic ash deposits or where it is underlain by rocks of Cretaceous age. Large concentrations of fluoride in drinking water cause staining of teeth, but fluoride is not a concern in irrigation water. The generally shallow depth of the water table in the High Plains aquifer makes water in the aquifer susceptible to contamination. Application of fertilizers and organic pesticides to cropland has greatly increased since the 1960¹s, thus increasing the availability and the amount of potential contaminants available. Increased concentrations of sodium, alkalinity, nitrate, and triazine (a herbicide) have been found in water that underlies small areas of irrigated cropland in Nebraska and Kansas. Of 132 wells sampled during 1984­85 in Nebraska, 43 had measurable concentrations (greater than 0.04 microgram per liter) of the herbicide atrazine. Increased concentrations of 2,4-Dichlorophen-oxyacetic acid (2,4-D, a pesticide) were found in water that underlies rangeland in a small part of the Great Bend area of the Arkansas River in Kansas. FRESH GROUND-WATER WITHDRAWALS Withdrawals of fresh ground water from the High Plains aquifer in Segment 3 during 1990 totaled 8,181 million gallons per day (fig. 66). Of this amount, 4,556 million gallons per day was withdrawn in Nebraska. About 97 percent of the total withdrawals, or about 7,900 million gallons per day, was used for agricultural, primarily irrigation, purposes. About 200 million gallons per day was pumped for public supply. Domestic and commercial withdrawals were about 40 million gallons per day, and industrial, mining, and thermoelectric power withdrawals also were about 40 million gallons per day. MISSISSIPPI EMBAYMENT AQUIFER SYSTEM INTRODUCTION The Mississippi embayment aquifer system in Segment 3 is in southeastern Missouri (fig. 67) on the western side of the Mississippi Embayment section of the Coastal Plain Physiographic Province. The aquifers that comprise the aquifer system are unconsolidated to semiconsolidated sands that range in age from Eocene to Late Cretaceous. The Mississippi embayment aquifer system extends over large areas in Arkansas, Louisiana, and Mississippi and smaller areas in Alabama, Florida, Illinois, Kentucky, Missouri, and Tennessee (fig. 67). The aquifer system is areally extensive in Segment 5 of this Atlas and is discussed in greater detail in chapter F which describes that segment. Little freshwater is withdrawn from the Mississippi embayment aquifer system in Missouri because it is overlain in most places by the productive Mississippi River Valley alluvial aquifer. HYDROGEOLOGIC UNITS The six aquifers and two confining units that compose the Mississippi embayment aquifer system in Missouri subcrop as narrow bands (fig. 68) beneath the Mississippi River Valley alluvial aquifer. Five of the aquifers of the Mississippi embayment aquifer system are in Tertiary rocks (fig. 69). In descending order, these are the upper Claiborne aquifer, the middle Claiborne aquifer, the lower Claiborne-upper Wilcox aquifer, the middle Wilcox aquifer, and the lower Wilcox aquifer. The clayey middle Claiborne confining unit separates the upper Claiborne and middle Claiborne aquifers. The McNairy­Nacatoch aquifer, deepest aquifer in the system, is in sands of Cretaceous age and underlies thick clay of the Midway confining unit. Clayey silt of the Vicksburg­Jackson confining unit overlies the aquifer system locally in an area adjacent to Tennessee. Although some of the aquifers in the Mississippi embay-ment aquifer system are not separated by regional confining units, they can be defined on the basis of changes in lithology and hydraulic head (water level) between aquifers. The vertical movement of water between the middle Claiborne through lower Wilcox aquifers is restricted somewhat by interbedded fine-grained sediments within the aquifers. In contrast, the middle Claiborne and Midway confining units more effectively retard the vertical movement of water between aquifers. The extensive, massive water-yielding sands of the aquifer system slope and thicken in Missouri toward the axis of the Mississippi Embayment. The lower Wilcox aquifer has been chosen to illustrate the aquifer system because it extends over a wide area and has been penetrated by numerous wells. The top of the lower Wilcox aquifer is about 200 feet above sea level at its updip limit but slopes to more than 1,000 feet below sea level in southeastern Pemiscot County (fig. 70). On the opposite (Tennessee) side of the Embayment, the top of the aquifer is shallower. The aquifer, thus, has a troughlike shape, as do the aquifers that overlie and underlie it. The lower Wilcox aquifer thickens from a featheredge at its northwestern limit to more than 300 feet in southeastern Pemiscot County (fig. 71). The troughlike shape of the aquifer is shown by the curvature of the lines of equal thickness near the Mississippi River. East of the river, the aquifer thins toward its outcrop area in Tennessee. Shallower and deeper aquifers in the Mississippi embayment aquifer system show the same thickening and thinning trends as those of the lower Wilcox aquifer. GROUND-WATER FLOW Because the Mississippi embayment aquifer system in Missouri is covered by the Mississippi River Valley alluvial aquifer in most places, the aquifer system is recharged mostly by downward leakage of water from the alluvial aquifer. Water enters the lower Wilcox aquifer in a band where sands of the aquifer are in hydraulic contact with those of the overlying Mississippi River Valley alluvial aquifer (fig. 72). The water then moves southeastward down the dip of the sand beds in the lower Wilcox aquifer. Water in the aquifer also moves westward and southwestward from aquifer outcrop areas in Tennessee and Kentucky. Because the lower Wilcox aquifer is deeply buried, this water moves under the Mississippi River without discharging to the river. Water discharges from the aquifer by upward leakage to shallower aquifers in Arkansas. GROUND-WATER QUALITY The chemical quality of freshwater in the aquifers of the Mississippi embayment aquifer system in Missouri is suitable for most uses. The water is fresh except locally in the McNairy­Nacatoch aquifer, which is the deepest aquifer of the system (fig. 73). Dissolved-solids concentrations in water from this aquifer locally exceed 2,000 milligrams per liter in New Madrid and Stoddard Counties and exceed 1,000 milligrams per liter in an area of about 450 square miles. Water in this area is a sodium chloride type and probably has entered the McNairy­Nacatoch aquifer by upward leakage from underlying Paleozoic rocks that locally contain saline water. Water in shallower aquifers of the Mississippi embayment aquifer system is fresher than that in the McNairy­Nacatoch aquifer. Water in the aquifers of the Mississippi embayment aquifer system is a calcium magnesium bicarbonate type in aquifer recharge areas. As the water moves down the hydraulic gradient, it changes to a sodium bicarbonate type and locally changes to a sodium chloride type in deeply buried, down-gradient parts of the aquifers. FRESH GROUND-WATER WITHDRAWALS About 95 million gallons per day of freshwater was pump-ed from all the aquifers of the Mississippi embayment aquifer system in Missouri during 1990. Most of the water was withdrawn for agricultural, primarily irrigation, use. Because the overlying Mississippi River Valley alluvial aquifer is a thick, productive aquifer, larger amounts of water are withdrawn from it than from the deeper Mississippi embayment aquifer system. GREAT PLAINS AQUIFER SYSTEM INTRODUCTION The Great Plains aquifer system underlies most of Nebraska, about one-half of Kansas, the eastern one-third of Colorado, and small parts of New Mexico, Oklahoma, Texas, South Dakota, and Wyoming (fig. 74). The rocks that compose the aquifer system extend northward into Segment 8 of this Atlas, where they mostly contain saline water, and equivalent rocks extend still further northward into western Canada. The aquifer system has been studied in detail, however, only in the area shown in figure 74, where it extends for about 170,000 square miles. Although the Great Plains aquifer system is extensive in Segment 2 of this Atlas, it contains primarily saline water there, along with some brine, oil, and gas, and is accordingly not discussed in detail in the chapter which describes that segment. The maps and descriptions presented in this chapter apply to the parts of the aquifer system that are in Kansas and Nebraska. The Great Plains aquifer system is named for the Great Plains Physiographic Province, which is a vast, rolling plain that slopes eastward for several hundred miles from the front of the Rocky Mountains. The water-yielding rocks of the aquifer system are sandstone; confining units in the system consist of siltstone and shale. Water has been produced from the uppermost part of the aquifer system for many years; more than 1,000 flowing wells were reported to produce water from the aquifer system in 1905. In the early 1900¹s, the uppermost aquifer of the system was named the Dakota aquifer and was described as a classic artesian system. More recent studies have shown that the flow system of this aquifer is more complex than was originally thought. Likewise, the stratigraphy of the rocks that form the aquifer is complex. It is now known to be an oversimplification to refer to the aquifer simply as the ³Dakota aquifer,² and it has accordingly been renamed. HYDROGEOLOGIC UNITS The rocks that contain the Great Plains aquifer system were deposited during the early part of the Cretaceous Period and reflect transgressions and regressions of the Cretaceous sea. The thickness of the basal part of the aquifer system varies because of the relief of the pre-Cretaceous erosional surface. Local stratigraphy is complex because numerous oscillations of the sea resulted in repeated, rapid shifts from marine to marginal marine to nonmarine depositional environments. These environmental changes are reflected in abrupt changes in the texture of the sediments. Sand bodies are typically linear, lenticular, or sinuous, which indicates that they were deposited in deltaic, shoreline, or fluvial environments. Regionally, the Great Plains aquifer system is characterized by two sandstone aquifers separated by a shale confining unit (fig. 75). The aquifer system contains more geologic formations and is more complex in southwestern Nebraska, where it is extremely thick, than to the east in Nebraska and Kansas, where it is thinner. In places, both aquifers are subdivided by shale beds that form local confining units. The Great Plains confining system, which consists mostly of thick layers of Upper Cretaceous shale, overlies the Great Plains aquifer system in most places. The aquifer system is confined below by a thick sequence of shale, limestone, sandstone, and anhydrite beds of Jurassic to Late Mississippian age. This underlying sequence of confining beds is called the Western Interior Plains confining system. The upper aquifer of the Great Plains aquifer system is called the Maha aquifer. This aquifer was formerly called the Dakota aquifer from the Dakota Sandstone, which is a prominent part of the aquifer. The Maha is the best known and most used part of the aquifer system. The correlation chart (fig. 75) shows the different usage of the term ³Dakota² in different places throughout the extent of the aquifer system. In eastern Nebraska and Kansas, the name ³Dakota Formation² applies only to the upper aquifer of the system. In southwestern Nebraska, the Dakota is considered to be a stratigraphic group that includes all Lower Cretaceous rocks and the lower part of Upper Cretaceous rocks, and consists of four sandstone aquifers and three confining units. In Kansas, the term ³Dakota aquifer² refers to the entire Great Plains aquifer system. Because of the multiple meanings of the term ³Dakota² as applied to either geologic units or hydrogeologic units, confusion is avoided by assigning a new name, the ³Maha aquifer,² to the upper aquifer. Likewise, it is inexact to apply the formerly used name ³Cheyenne Sandstone aquifer² to the lower aquifer of the Great Plains aquifer system. Accordingly, this lower aquifer has been renamed the ³Apishapa aquifer.² The Maha aquifer is separated from the Apishapa aquifer in most places by the Apishapa confining unit, which consists of the Kiowa and the Skull Creek Shales. The Maha and the Apishapa aquifers consist of loosely cemented, medium- to fine-grained sandstone. The Apishapa confining unit that separates the aquifers consists of slightly permeable shale. The Maha aquifer is more extensive than the Apishapa aquifer (fig. 76); where the Apishapa confining unit pinches out, the two aquifers are in direct contact and are considered to be part of the Maha aquifer (fig. 77). In western Nebraska, both aquifers are downwarped to depths of more than 5,000 feet below land surface in part of the Denver Basin (fig. 78), where parts of the Great Plains aquifer system contain brine, oil, and gas. In eastern Nebraska and most of Kansas, the aquifer system is buried to depths of 1,000 feet or less below land surface. The Maha aquifer is much thicker than the Apishapa aquifer in most places. The thickness of the Maha aquifer generally is about 200 to 300 feet in western Nebraska and most of Kansas, but locally the aquifer is more than 800 feet thick in north-central Nebraska (fig. 79). The greatest thicknesses of the Maha aquifer are just east of where the Apishapa confining unit pinches out and the sandstones of the underlying Apishapa aquifer merge with those of the Maha aquifer. The Apishapa aquifer typically is between 100 and 200 feet thick (fig. 80) but locally is more than 400 feet thick in west-central Nebraska. The thickness of the Apishapa aquifer varies because the sedimentary rocks that compose the aquifer were deposited on an irregular erosional surface. GROUND-WATER FLOW The regional movement of water in the Great Plains aquifer system in Segment 3 can be inferred from a map of the potentiometric surface of the aquifer system (fig. 81). Water levels in the aquifer system are highest in southwestern Kansas and northwestern Nebraska. Water moves generally eastward and northeastward from recharge areas in southeastern Colorado toward discharge areas in central Kansas, eastern Nebraska, and along the Missouri River in northeastern Nebraska. In its central parts, the aquifer system typically is overlain by a thick confining system, and the hydraulic gradient is flat. Water in the aquifer system in these places moves sluggishly. Near the eastern limit of the aquifer system, the hydraulic gradient becomes steeper, which indicates a more dynamic ground-water flow system. Much of the recharge to the aquifer system is from precipitation that falls directly on aquifer outcrop areas in southern and southeastern Colorado and east-central Kansas. Some recharge, however, enters the aquifer system as downward leakage through the overlying Great Plains confining system. The High Plains aquifer overlies this confining system, and the hydraulic gradient is downward from the High Plains aquifer to the Great Plains aquifer system in most places (fig. 82). For example, in the southwestern part of the Nebraska panhandle, the water table in the High Plains aquifer is more than 2,500 feet higher than the hydraulic head in the Great Plains aquifer system. In many places near the eastern limit of the Great Plains aquifer system, however, water levels in the High Plains aquifer and the Great Plains aquifer system are about equal. Locally, the hydraulic gradient is reversed, and discharge takes place by upward leakage from the Great Plains aquifer system to the High Plains aquifer. Most discharge from the Great Plains aquifer system is by upward leakage, but some discharge is as base flow to streams in aquifer outcrop areas. The Great Plains aquifer system is mostly confined above by the Great Plains confining system and below by the Western Interior Plains confining system. Movement of water through the confined parts of the aquifer system is very slow and is estimated to be 10 feet per year or less. Flow is more rapid in places where the aquifer system crops out or the overlying confining system is thin. The presence of oil, gas, and brine in deeply buried parts of the aquifer system, such as the Denver Basin, indicates that the water in such places is virtually stagnant. Most of the data available for the water-yielding capability of the Great Plains aquifer system are from the Maha aquifer. Sparse data from the deeper Apishapa aquifer indicate that the two aquifers have similar hydraulic properties, which show similar trends. Most of the porosity in the aquifer system is intergranular; that is, it consists of pore spaces between individual sand grains. Joints, fractures, and bedding planes exist locally in the sandstones, but most of the water moves through the intergranular pore spaces. Sandstone porosity in the Great Plains aquifer system generally decreases as the depth of burial of the aquifer system increases because the sandstone has compacted where it is buried beneath thousands of feet of overlying rocks. The compaction has reduced the percentage of pore space in the sandstone from more than 30 percent where overlying rocks are thin to less than 10 percent where the aquifer system is deeply buried. This reduction in pore space not only reduces the capacity of the aquifer to store water, but also its capability to transmit water. Transmissivity, or the capacity of an aquifer or aquifer system to transmit water, is one way to measure the ease with which ground water moves. The greater the transmissivity of an aquifer, the more readily water moves through it, and the greater the chances of obtaining large well yields from the aquifer. The distribution of the estimated transmissivity of the Maha aquifer is shown in figure 83. The transmissivity of the aquifer is greater in its eastern parts, and the larger transmissivity values (1,000 to more than 10,000 feet squared per day) coincide with places where the aquifer is thickest. The transmissivity values in western Nebraska locally are less than 100 feet squared per day where the aquifer is deeply buried and compacted in part of the Denver Basin. Reported yields of wells completed in the Maha aquifer in eastern Nebraska and central Kansas commonly exceed 50 gallons per minute and locally are as much as 1,000 gallons per minute. These large-yield areas coincide with places where the transmissivity of the aquifer is high. GROUND-WATER QUALITY The chemical character of water in the Great Plains aquifer system is determined by many factors. Some of or all the following factors determine the ground-water chemistry: the mineral content of the soil or aquifer material through which the water has passed; the rate of movement of the water; the length of time the water remains in the aquifer; the chemistry of water trapped during sediment deposition; diagenesis, or the chemical and physical changes that take place in aquifer sediments after they are deposited; and mixing of the water with water in adjacent hydrogeologic units or from the land surface. Concentrations of dissolved solids in water from the Maha aquifer are mapped in figure 84. The aquifer contains freshwater only near its southern and eastern margins and in a small area in northwestern Nebraska. These are places where the overlying confining unit is thick and represent areas of aquifer recharge or discharge; the flow system is accordingly dynamic, and mineralized water can be readily flushed from the aquifer. Also, the quartz sand that composes most of the aquifer is not readily dissolved; this condition leads to small dissolved-solids concentrations in the ground water. Water that contains dissolved-solids concentrations of between 1,000 and 10,000 milligrams per liter is considered to be slightly to moderately saline. Such water is characteristic of much of the Great Plains aquifer system and results from incomplete flushing of highly mineralized water by a sluggish flow system. Some of the mineralized water has leaked upward from underlying Permian rocks that contain halite and evaporite minerals. In places where the Maha aquifer is deeply buried, it locally contains water with dissolved-solids concentrations of greater than 125,000 milligrams per liter (fig. 84). Such excessively large concentrations result from increased mineralization of saline water, which was trapped when the sandstones were deposited, by mixing with brine. The brine might have formed in barred basins into which seawater periodically spilled and became concentrated by evaporation, or it might have migrated into the aquifer in water that moved through and partially dissolved nearby salt or evaporite deposits. Concen-trations of dissolved solids that range from 10,000 to about20,000 milligrams per liter are common in water from the central and east-central parts of the aquifer. Concentrations of this magnitude can result from the combination of incomplete flushing and the upward migration of highly mineralized water from underlying Permian rocks. Ground water can be classified into hydrochemical facies on the basis of the dominant cations and anions in the water. To demonstrate the classification used, a sodium chloride water is one in which sodium ions account for more than 50 percent of the total cations in the water, and chloride ions account for more than 50 percent of the total anions. The distribution of hydrochemical facies in water from the Great Plains aquifer system is shown in figure 85. In many artesian flow systems, the water changes progressively from a calcium bicarbonate type in upgradient recharge areas to a sodium chloride type in deep, confined parts of the flow system. This is not the case with the Great Plains aquifer system because the distribution of hydrochemical facies is more complex (fig. 85). Calcium bicarbonate type water mostly is in a narrow band along the eastern and southern limits of the aquifer system. Where the aquifer system is confined, it mostly contains sodium bicarbonate or sodium chloride type waters. These hydrochemical facies are in places where mineralized water in the aquifer system has not been completely flushed by circulating freshwater and where underlying saline water leaks upward faster than it can be flushed. Some of the sodium bicarbonate water may be the result of ion exchange of calcium for sodium on the surface of clay or other sodium-rich minerals in the rocks of the aquifer system. The calcium sulfate type water in northeastern Nebraska results from the upward leakage of mineralized water from underlying rocks that contain anhydrite and gypsum. The leakage is thought to take place along fractures and faults. In summary, incomplete flushing, slow circulation through most of the aquifer system, rock-water interactions in the aquifer system, and mixing of highly mineralized waters from adjacent rocks are the processes that produce the observed distribution of hydrochemical facies. FRESH GROUND-WATER WITHDRAWALS Freshwater is withdrawn from the Great Plains aquifer system in Kansas and Nebraska mostly along the southern and eastern margins of the aquifer. These are the parts of the aquifer system that are nearest to land surface and contain most of the freshwater. Although development of the aquifer system began in the early 1900¹s and moderate amounts of water were withdrawn from the aquifer system during the 1940¹s and the 1950¹s, it was not until the 1960¹s that withdrawals were significant (fig. 86). Estimated withdrawals from the aquifer system in Kansas and Nebraska during the 1970¹s were at a rate of about 390,000 acre-feet per year, which is almost eight times the rate of withdrawal during the 1950¹s. The distribution of withdrawals has changed with time. During the 1950¹s, withdrawals in Kansas and Nebraska were about equal; during the 1960¹s, however, withdrawals increased greatly in Kansas, while withdrawals in Nebraska remained about the same as in the 1950¹s. During the 1970¹s, withdrawals increased greatly in both States. Total fresh ground-water withdrawals from the Great Plains aquifer in Kansas and Nebraska were about 133 million gallons per day during 1990 (fig. 87). About 73 percent of the water withdrawn, or about 97 million gallons per day, was used for agricultural purposes, primarily irrigation. About 17 million gallons per day was withdrawn for public supply purposes, and the same amount was withdrawn for domestic and commercial purposes. About 2 million gallons per day was pumped for industrial, mining, and thermoelectric power uses. OZARK PLATEAUS AQUIFER SYSTEM INTRODUCTION The Ozark Plateaus aquifer system contains most of the freshwater in the aquifers that consist of Mississippian and older rocks in Segment 3. The aquifer system underlies most of southern Missouri and a small part of extreme southeastern Kansas in this segment; it also underlies a large area in northwestern Arkansas and a small part of northeastern Oklahoma (fig. 88). The Arkansas part of the aquifer system is discussed in detail in the chapter of this Atlas that describes Segment 5, and the Oklahoma part is discussed briefly in the chapter that describes Segment 4. Rocks equivalent to parts of the Ozark Plateaus aquifer system locally contain freshwater in parts of northeastern Missouri and are called the Mississippian and the Cambrian-Ordovician aquifers. Equivalent carbonate rocks to the west and northwest that contain saline water or brine have been named the ³Western Interior Plains aquifer system² (fig. 89). The water-yielding rocks in the Ozark Plateaus aquifer system and equivalent beds are mostly limestones and dolomites, but some sandstones are productive aquifers. Confining units within the aquifer system and its equivalents are shale or dolomite. The lithology of the individual aquifers and confining units and their hydraulic character are consistent over large areas. Ground water in the aquifer system locally moves from topographically high recharge areas to surface streams. Regional movement is northwestward, eastward, and southward from the St. Francois Mountains and other topographically high areas in southern Missouri. The water moves upward at the transition zone between the Ozark Plateaus and the Western Interior Plains aquifer systems and discharges either to streams as base flow or to shallow stream valley alluvial aquifers. RELATION TO ADJACENT AQUIFERS AND AQUIFER SYSTEMS The Ozark Plateaus aquifer system consists of three aquifers separated by two confining units (fig. 89), all of which grade laterally westward into equivalent hydrogeologic units that are part of the Western Interior Plains aquifer system. Sparse information indicates that the rocks of the St. Francois confining unit, which is the lowermost confining unit within the Ozark Plateaus aquifer system, are thin or absent in the area of the Western Interior Plains aquifer system. The combined Ozark and St. Francois aquifers of the Ozark Plateaus aquifer system are thus considered to be equivalent to the unnamed lower aquifer units of the Western Interior Plains aquifer system. The Ozark confining unit and the Springfield Plateau aquifer of the Ozark Plateaus aquifer system are equivalent to the confining unit and upper aquifer unit (both unnamed) of the Western Interior Plains aquifer system, respectively. Hydrogeologic units of the Ozark Plateaus aquifer system also are equivalent to aquifers and confining units north of the Missouri River in Missouri (fig. 89). The Springfield Plateau aquifer is stratigraphically equivalent to the Mississippian aquifer of northern Missouri, but limited information indicates that these aquifers have little or no hydraulic connection. Part of the Ozark aquifer is equivalent to the Cambrian­Ordovician aquifer of northern Missouri, and these two aquifers are considered to be hydraulically connected in places. Upper Devonian and Lower Mississippian rocks compose a confining unit in the Ozark Plateaus aquifer system in southern Missouri, and equivalent rocks are part of a thick confining unit in northern Missouri. The lowermost aquifer of the Ozark Plateaus aquifer system is equivalent to a poorly known, unnamed minor aquifer in northern Missouri. The general direction of ground-water movement in the lower part of the Ozark aquifer and its equivalents can be inferred from a map of the potentiometric surface of these aquifers (fig. 90). Water in Cambrian and Ordovician rocks in southern Missouri moves northward, southward, and northwestward from a ground-water divide in south-central Missouri. Water in the Western Interior Plains aquifer system generally moves southeastward and eastward; locally, in southwestern Kansas, the water moves northward. Little or no water leaks upward across the thick, effective Western Interior Plains confining system that overlies most of the Western Interior Plains aquifer system. Lateral flow in the two aquifer systems merges in a northeast-trending transition zone in eastern Kansas and west-central Missouri. This transition zone, which separates the regional ground-water flow systems (fig. 91), coincides with a low area on the potentiometric surface and with a topographically low area at the land surface. Ground water in the transition zone moves mostly upward, and discharges either to streams as base flow or to shallow, unconfined, stream-valley alluvial aquifers. Water that moves northward in the lower part of the Ozark aquifer discharges mostly to the Missouri River along a low area of the potentiometric surface parallel to the river (fig. 90). The Missouri River is likewise a discharge area for water that moves southward in the Cambrian­Ordovician aquifer, which is located north of the river. Locally, water in the Ozark and Cambrian­Ordovician aquifer system moves eastward toward the Mississippi River, which is the largest and most deeply incised regional drain in the area. Water in the lower part of the Ozark aquifer is fresh in most of the southern one-half of Missouri and adjacent areas (fig. 92). Water in the equivalent Cambrian­Ordovician aquifer is fresh in a large area in east-central Missouri but is slightly to moderately saline in northern and northwestern Missouri where the aquifer is overlain by a thick confining unit and ground-water flow in the aquifer is sluggish. Water in the Western Interior Plains aquifer system is slightly saline or a brine (fig. 92) and locally contains dissolved-solids concentrations of more than 200,000 milligrams per liter in deeply buried parts of the aquifer system, such as those along the Kansas-Oklahoma State line. Compaction due to deep burial reduces the porosity and permeability of the aquifer system, and the hydraulic gradient of the system is small. Movement of water in the deep parts of the aquifer system is, therefore, very slow, and highly mineralized water has not been flushed from these parts of the system. The transition zone between the freshwater of the Ozark Plateaus aquifer system and the more mineralized water of the Western Interior Plains aquifer system is narrow and is reflected by an abrupt westward increase in dissolved solids in the ground water. Some of the mineralized water discharges upward to saline springs, such as those in Saline County, Missouri, and some locally discharges to the Osage River in Henry County, Missouri. This discharge of saline water indicates that water moves eastward in the Western Interior Plains aquifer system. HYDROGEOLOGIC UNITS Several geologic formations compose each of the three aquifers and two intervening confining units of the Ozark Plateaus aquifer system. These formations are grouped into the hydrogeologic units that make up the aquifer system in Missouri and Kansas (fig. 93). The top and bottom of each hy-drogeologic unit coincide with the top or bottom of a sequence of geologic units, although the number of geologic units in each aquifer and confining unit varies from place to place. The water-yielding formations are mostly limestone and dolomite but locally include sandstone and chert. The confining units that separate the aquifers primarily are shale but also consist of limestone, dolomite, and sandstone, all of which have minimal permeability. For example, the St. Francois confining unit consists mostly of dolomite. Locally, shale beds within the aquifers form confining units of limited extent. The aquifers in the Ozark Plateaus aquifer system have been named for geographic or physiographic features (fig. 94). From shallowest to deepest, the three aquifers are the Springfield Plateau aquifer, which was named for a physiographic feature in western Missouri and adjacent areas; the Ozark aquifer, which was named for the rolling uplands that compose most of the Ozark Plateaus Physiographic Province in central Missouri; and the St. Francois aquifer, which was named for the St. Francois Mountains in eastern Missouri. Confining units in the system are named the same as the aquifers they overlie; for example, the St. Francois aquifer is overlain by the St. Francois confining unit. The aquifers and confining units of the Ozark Plateaus aquifer system are exposed as a sequence of concentric bands that are centered around the Precambrian rocks that are exposed in the St. Francois Mountains (fig. 95). These Precambrian igneous and metamorphic rocks form the basement confining unit, which is the lower confining unit of the Ozark Plateaus aquifer system. Exposures of the St. Francois aquifer surround this confining unit and are, in turn, surrounded by a band of the overlying St. Francois confining unit. The rocks that compose the Ozark aquifer crop out over more than one-half of southern Missouri and in a large part of northern Arkansas. The thick, widespread Ozark aquifer is by far the most important aquifer of the Ozark Plateaus aquifer system; the equivalent Cambrian­Ordovician aquifer north of the Missouri River is also an important source of water. The thin Ozark confining unit overlies the Ozark aquifer and crops out as a narrow band that separates the Ozark aquifer from the overlying Springfield Plateau aquifer. A thick sequence of rocks with minimal permeability, which is called the Western Interior Plains confining system, overlies and effectively confines the Springfield Plateau aquifer west of the outcrop area of the aquifer. The Ozark Plateaus aquifer system is covered in southeastern Missouri by Mesozoic and younger rocks and deposits that are part of the Mississippi embayment aquifer system or the Mississippi River Valley alluvial aquifer. Variations in the thickness and extent of the aquifers and confining units of the Ozark Plateaus aquifer system and their equivalents are shown in figure 96. The thickness of the Springfield Plateau aquifer is uniform slightly to the west of its outcrop area, but the equivalent upper aquifer unit of the Western Interior Plains aquifer system thins westward where it is covered by the Western Interior Plains confining system. The Ozark aquifer thickens gradually westward and is more than 1,000 feet thick in central Missouri; the aquifer thickens more rapidly to the east of the St. Francois Mountains. The thickness of the St. Francois aquifer varies greatly because the rocks that compose this aquifer were deposited on an irregular erosional surface that was developed on Precambrian rocks of the basement confining unit. The thickness of the St. Francois and the Ozark confining units does not vary as much as that of the aquifers that they separate. SPRINGFIELD PLATEAU AQUIFER The Springfield Plateau aquifer is the uppermost aquifer of the Ozark Plateaus aquifer system and consists almost entirely of limestone of Mississippian age. The thickest and most productive water-yielding geologic formations included in the aquifer are the Burlington and the Keokuk Limestones (fig. 93). North of the Missouri River, the Burlington and the Keokuk Limestones also yield water but are considered to be a separate aquifer, the Mississippian aquifer. The thickness of the Springfield Plateau aquifer ranges from less than 200 to more than 400 feet (fig. 97) and averages about 200 feet. Locally, the aquifer is absent in the subsurface. The aquifer thins abruptly at the eastern edge of the Salem Plateau. Equivalent rocks in two small areas in St. Louis and Ste. Genevieve Counties, Missouri, are not considered to be part of the aquifer. Most of the water in the Springfield Plateau aquifer occurs in and moves through secondary openings, such as fractures and bedding planes. The slightly acidic ground water that moves through these openings has dissolved part of the limestone and has resulted in a network of solution channels. This dissolution activity is reflected at the land surface by springs, caves, and sinkholes, and by sparse surface drainage. These features are characteristic of a type of topography called karst topography, which commonly is developed in areas underlain by limestone. Recharge to the Springfield Plateau aquifer is mostly from precipitation on outcrop areas of the aquifer. After the recharge water percolates downward to the water table, most of it moves laterally along short flow paths to discharge as base flow to nearby streams. Some of the water follows flow paths of intermediate length and discharges to large streams, and a small part of the recharge moves laterally into deep, confined parts of the aquifer. A small amount of recharge to the Springfield Plateau aquifer is by upward leakage of water from the deeper Ozark aquifer in places where the hydraulic head of the Ozark aquifer is higher than that of the Springfield Plateau aquifer. In much of the outcrop area of the Springfield Plateau aquifer, however, water levels in this aquifer are higher than those in the Ozark aquifer, and water leaks downward to recharge the Ozark aquifer. Upward and downward leakage take place through the Ozark confining unit that separates the Springfield Plateau and the Ozark aquifers. A map of the estimated potentiometric surface of the Springfield Plateau aquifer before development (fig. 98) shows that water in the aquifer moved mostly from local recharge areas to nearby surface drains. The configuration of the potentiometric surface contours is irregular because of the influ-ence of streams on the ground-water flow system. Where the aquifer is confined above by the Western Interior Plains confining system, the contours are smoother and only reflect the influence of large streams. The regional movement of water in the aquifer is westward. The chemical quality of water in the Springfield Plateau aquifer generally is suitable for most uses where the aquifer is unconfined or where the confining unit that overlies the aquifer is thin. The water commonly is a calcium bicarbonate type and is moderately hard. Dissolved-solids concentrations in water from the aquifer generally are less than 1,000 milligrams per liter except where the aquifer is confined (fig. 99). Dissolved-solids concentrations increase rapidly downgradient where the aquifer becomes confined. Concentrations of sulfate generally are small in water from the aquifer except in the Tri-State lead-zinc mining district of southwestern Missouri, southwestern Kansas, and northeastern Oklahoma where concentrations of more than 500 milligrams per liter are reported near some mining areas. These large concentrations result from leaching of the sulfide minerals that contain the lead and zinc. Most of the water withdrawn from the Springfield Plateau aquifer is used for domestic and stock-watering supplies. Yields of wells completed in the aquifer generally are less than 20 gallons per minute. OZARK CONFINING UNIT The Ozark confining unit underlies the Springfield Plateau aquifer and hydraulically separates this aquifer from the deeper Ozark aquifer. The Ozark confining unit consists mostly of shale but locally includes limestone of minimal permeability. The confining unit generally is less than 100 feet thick except in small areas. Where the shale content of the confining unit is greater, the confining unit can more effectively retard the vertical movement of water between the Springfield Plateau and the Ozark aquifers. North of the Missouri River, rocks equivalent to the Ozark confining unit separate the Mississippian and the Cambrian-Ordovician aquifers. OZARK AQUIFER The Ozark aquifer is the middle aquifer of the Ozark Plateaus aquifer system and consists of numerous geologic formations that range in age from Devonian to Cambrian (fig. 93). The rocks of the aquifer are mostly dolomite and limestone, but some beds of sandstone, chert, and shale are included in it. The Ozark aquifer is the primary source of water in the Ozark Plateaus Physiographic Province from which it is named. The aquifer provides water for municipal, industrial, and domestic supplies. The main water-yielding formations in the Ozark aquifer are the Upper Cambrian Potosi Dolomite, the Lower Ordovician Gasconade Dolomite, and the Roubidoux Formation. The Potosi Dolomite is the most permeable of these three formations. North of the Missouri River, rocks that are equivalent to the Ozark aquifer are called the Cambrian­Ordovician aquifer. Like the Ozark aquifer, the Cambrian­Ordovician aquifer consists mostly of dolomite and limestone; however, it also includes beds of sandstone and shale. The Upper Cambrian Potosi and the Eminence Dolomites are the main water-yielding formations in the Cambrian­Ordovician aquifer, but the Lower Ordovician Gasconade Dolomite and locally the Middle Ordovician St. Peter Sandstone are important sources of water. Most wells completed in the Cambrian­Ordovician aquifer are open to more than one water-yielding unit. The Ozark and the Cambrian­Ordovician aquifers are mapped together in this report. The Ozark aquifer underlies most of Missouri south of the Missouri River, and the Cambrian­Ordovician aquifer underlies the eastern one-half of Missouri north of that river (fig. 100). The Ozark aquifer is less than 1,000 feet thick throughout the Salem Plateau but thickens to more than 3,000 feet in southeastern Missouri just north and east of the bootheel. The Ozark aquifer pinches out against the flanks of the St. Francois Mountains, and its thickness is irregular where it has been eroded in outcrop areas. Caves, sinkholes, and other types of solution features characteristic of karst topography have developed on the carbonate-rock units that compose the aquifer. North of the Missouri River, the thickness of the Cambrian­Ordovician aquifer averages about 1,200 feet but is locally greater than 1,800 feet. Carbonate rocks equivalent to the Cambrian-Ordovician aquifer in northwestern Missouri are deeply buried and contain saline water. Recharge to the Ozark aquifer is mostly from precipitation on aquifer outcrop areas. Small volumes of water recharge the aquifer by downward leakage from the shallower Springfield Plateau aquifer. Most ground-water flow in the shallow part of the Ozark aquifer moves from topographically high recharge areas along short flow paths to discharge as base flow to nearby streams. The shallow flow system is accordingly con-trolled mostly by topography. The ground water mostly occurs in and moves through fractures and bedding planes in carbonate rocks. These openings have been enlarged by dissolution of the carbonate rocks and have been reported at depths of as great as 1,500 feet below land surface. Where sinkholes have formed from dissolution of the carbonate rocks, water that runs over the land surface may enter the sinkholes and large volumes of recharge can enter the aquifer in this manner. Recharge to the equivalent Cambrian­Ordovician aquifer likewise is mostly from precipitation on aquifer outcrop areas, but small amounts of recharge enter this aquifer by downward leakage of water from the overlying Mississippian aquifer. Discharge from the Ozark and the Cambrian­Ordovician aquifers is mostly to streams in aquifer outcrop areas. Some water follows flow paths of intermediate length and discharges to regional drains, such as the Missouri and the Mississippi Rivers. A small volume of water leaks upward from the Ozark aquifer and locally discharges to the overlying Springfield Plateau aquifer. In southeastern Missouri, a small volume of water discharges from the Ozark aquifer to the Mississippi River Valley alluvial aquifer by upward leakage. A map of the potentiometric surface of the Ozark and the Cambrian­Ordovician aquifers before development of the aquifers began (fig. 101) can be used to show the regional direction of ground-water movement in the aquifers. Wate