Prev Page--Water Table || Next Page--Discharge, Recovery
Ground Water, continued
Ground-water Recharge
Recharge is the addition of water to the underground reservoir and may be accomplished in several different ways. All ground water within a practical drilling depth beneath Finney and Gray counties is derived from the water that falls as rain or snow either within the area or on nearby areas to the west. Once the water becomes a part of the ground-water body it moves down the slope of the water table, later to be discharged at some point farther downstream.
The underground reservoir beneath this area is recharged by local precipitation within the area, by seepage from streams and irrigation canals, and by subsurface inflow from areas to the west.
Recharge from local precipitation
The average annual precipitation in Finney and Gray counties is about 19 inches, but probably only a small part of this amount reaches the zone of saturation owing to several complicating factors. The depth of the water table below the surface determines in part the amount and frequency of this recharge. Where the water table stands comparatively far below the surface, as it does in parts of Finney and Gray counties, it fluctuates less in response to precipitation than it does in areas like the Arkansas valley where the water table is comparatively shallow.
The relation between the fluctuations of water levels and the depth to the water level in wells is shown by the hydrographs of wells 432, 531, and 541 in figure 11. Well 432 is a shallow well in the Arkansas valley. The water level in this well responds readily to the rains that fall on the valley surface. Wells 531 and 541 are upland wells no southern Gray County. The depth to water level in well 531 is about 60 feet and in well 541 about 138 feet. Because of the greater depth to water level in well 531, it fluctuates less in response to precipitation than does the water level in the shallow valley well (432). According to the hydrograph shown in figure 11, the water level in well 541 seems to be unaffected by precipitation. It is true that the water level in a well in which the depth to water is great will not readily respond to precipitation, but over a long period of years there should be some comparison between the water level in a deep well and the precipitation, even though it would not be as marked as in a shallow well.
Figure 11--Hydrographs showing the relation bewteen the monthly fluctuations of the water levels in four wells in Gray County and the monthly precipitation at Cimarron. Precipitation data from U.S. Weather Bureau.
Of the total precipitation, part is lost by evaporation into the air, part is lost, through runoff, and part is used by growing plants. Thus, only a part of the precipitation enters the interstices of the soil and only part of this reaches the zone of saturation. When the amount of water absorbed by the soil is greater than can be held against the pull of gravity, a part of the water will move downward to the zone of saturation. At the end of the growing season in the fall of the year, the moisture in the upper part of the soil zone is usually depleted. Evaporation and transpiration by plants remove most of the moisture in this zone, thereby causing a deficiency in soil moisture. Moisture in the depleted soil zone is replenished by fall and winter precipitation. Water that enters the interstices of the soil during the fall and winter remains, there until the moisture deficiency is satisfied or until the growing season begins. There is relatively little evaporation and transpiration during the fall and winter. If there is sufficient precipitation, the moisture deficiency is taken care of and recharge to the zone of saturation takes place. If, however, the precipitation is small, all the water may be retained in the soil, and there will be no ground-water recharge.
The amount of water lost through evaporation into the air varies from one season to the other, the rate of evaporation being greatest in the summer when temperatures are highest. In an average year, more than half the total precipitation in this area comes during the summer from May through August, when the rate of evaporation is greatest. The average rates of evaporation from a free water surface for the months of the growing season based on 30-year records at the Garden City Experimental Farm are as follows: April, 6.68 inches; May, 8.46 inches; June, 10.25 inches; July, 11.90 inches; August, 10.42 inches; September, 8.10 inches (Smith, 1940, p. 28). It seems likely, therefore, that a rather large percentage of the annual precipitation in this area is lost through evaporation.
The amount of water leaving the area by runoff in streams is probably comparatively small. The duration and intensity of the rainfall, the slope of the land surface, and the type of soil and vegetation principally determine the amount of local runoff from precipitation. The runoff from a gentle rain as a rule is much smaller than the runoff from a heavy downpour, hence the amount of ground-water recharge from a gentle rain of long duration generally is greater than the recharge from a heavy downpour of short duration, provided all other factors are equal.
The slope of the land is an important factor in determining the amount of runoff, and in general the steeper the slope the greater the runoff. Steep slopes occur along the major streams in this area; otherwise the slopes of the land surface are relatively gentle. Large areas in Finney and Gray counties lack any but local surface drainage. This is true of the area of sand hills south of the Arkansas valley, the Finney basin, and much of the upland plains.
Runoff is greater in places where the soil is tightly compacted and consists of fine relatively impermeable material than in places where the soil is sandy and loosely compacted. The latter type of soil allows a part of the water to percolate into the ground, thus decreasing the amount of surface runoff.
The loss of precipitation by runoff in the Arkansas River drainage basin is very small. According to records of the Division of Water Resources of the Kansas State Board of Agriculture, the average annual net runoff in the Arkansas River drainage area above Garden City for the period October 1, 1928, to September 30, 1939, was only 0.09 inch (excluding water diverted by irrigation canals west of Garden City). The average annual net runoff in the Pawnee River drainage basin above Lamed for the same period was 0.298 inch (Anon., 1937, pp. 401, 402, 490, 491; 1940, pp. 249, 303). Thus the loss by runoff amounts to a relatively small part of the total precipitation.
The most favorable areas in Finney and Gray counties for ground-water recharge from precipitation are the sand hills, the shallow-water areas in stream valleys, and the shallow surface depressions on the uplands and in the Finney depression.
Recharge in areas covered by dune sand--The large area covered by dune sand south of Arkansas River and the smaller areas covered by dune sand north of the river offer unusually good opportunities for recharge from precipitation. Much of the rain that falls on the dune-covered areas is absorbed by the sand and for that reason there is little runoff, as is evidenced by the almost total absence of drainage lines. The water that enters the interstices of the sand percolates downward rapidly, thus escaping evaporation.
Recharge in Arkansas valley--The bottom lands of the Arkansas valley also offer exceptional opportunities for ground-water recharge from precipitation. As stated above, the percentage of the precipitation that is lost by runoff is very small. According to Slichter (1906, p. 53), there is no runoff from precipitation in the area between Garden City and Deerfield. If so, the rainfall that is not lost through evaporation and transpiration must pass through the sandy porous soil and percolate downward to the water table.
Recharge by precipitation on the valley floor is evidenced by the fact that following rains there is a rapid rise of the water levels in the wells situated in the valley. The fluctuations of the water level in two observation wells (246 and 350) in Finney County, the cumulative departure from normal monthly precipitation at Garden City, and the monthly precipitation at Garden City are shown in figure 12. Both wells are shallow, unused irrigation wells in the Arkansas valley. Well 246 is about 1.75 miles north of Arkansas River and well 350 is about 0.75 mile south of the river. Figure 11 shows the fluctuations of the water levels in four observation wells in Gray County, the cumulative departure from normal monthly precipitation at Cimarron, and the monthly precipitation at Cimarron. Well 432 in figure 11 is in the Arkansas valley about a quarter of a mile north of the river.
Figure 12--Hydrographs showing the relation bewteen the monthly fluctuations of the water levels in wells 246 and 350 and the monthly precipitation at Garden City. Precipitation data from U.S. Weather Bureau.
During a normal year the water table in the Arkansas valley in this area reaches its highest stage in June or July, and its lowest stage in the fall--generally in September or October. The decline of the water table from about June until October is caused in part by the withdrawal of large quantities of water from irrigation wells in the valley and in part by the heavy draft created by the vegetation during the growing season. The water levels in all wells do not follow this routine behavior every year, for other factors may affect them. Many of the valley wells in Finney County deviated from their routine behavior in 1941 owing in part to the abnormally high rainfall during May, June, July, and August and in part to the resulting decrease in withdrawal of water for irrigation. The rainfall as recorded at Garden City for the 4-month period from May through August was 6.67 inches above normal. The result of this on the water levels in wells is shown by the hydrographs of wells 246 and 350 in figure 12.
Figure 13 shows the relation between the annual fluctuations of the water levels in eight irrigation wells in the Arkansas valley in Finney County and the cumulative departure from the average annual precipitation at Garden City for the period from 1934 to 1942. The depths to water level in the wells were measured each year in October, November, or December after the wells had ceased pumping. Kenneth McCall, engineer for the Division of Water Resources of the Kansas State Board of Agriculture, made all measurements from 1934 through 1940. Woodrow Wilson of the Federal Geological Survey measured the depths to water level in the wells in 1941 and 1942. The hydrographs in figure 13 seem to show a direct correlation between the fluctuations of the water levels in the irrigation wells and the cumulative departure from the average annual precipitation. The fluctuations of the water levels, however, have not been due entirely to precipitation, but have been due partly to the withdrawal of water for irrigation.
Figure 13--Hydrographs showing the relation bewteen the annual fluctuations of the water levels in eight irrigation wells in the Arkansas valley in Finney County and the cumulative departure from the annual precipitation at Garden City for the 9-year period from 1934 to 1942.
Wells 233, 255, 267, 274, and 287 are located in or near the heavily pumped areas surrounding Garden City and Holcomb. Wells 238, 241, and 342 are farther from the heavily pumped areas. The cumulative deficiency in precipitation for the period from 1934 to the end of 1939 was 18.15 inches. For this same period the five wells in the heavily pumped areas showed net declines in water level ranging from 0.9 foot to 2.13 feet, the average net decline being about 1.5 feet. Of the three wells outside the heavily pumped areas, well 238 showed a net decline of only 0.1 foot for the same period and wells 241 and 342 had net rises of 0.39 foot and 0.13 foot, respectively. In summary, the wells in the heavily pumped areas showed an average net decline in water level of about 1.5 feet, whereas those outside the heavily pumped areas showed an average rise in water level of 0.14 foot for the period of below average rainfall from 1934 to 1939. It is obvious from these figures that much of the decline in water levels in the wells in the heavily pumped areas was the result of the cumulative deficiency of precipitation and also of the heavy withdrawals of water for irrigation. In studying the factors producing a 9-year decline in the ground-water levels in the shallow-water basin in central Scott County, Waite (1941, p. 775) found that somewhat more than half of the decline resulted from the effects of pumping and the rest from the deficiency in precipitation.
The amount of water withdrawn annually by pumps in the Arkansas valley generally varies indirectly with the precipitation. In 1939, when the precipitation was 4.48 inches below the annual average for the 9-year period, approximately 27,120 acre-feet of water was pumped from wells in the Arkansas valley in Finney County for irrigation. As a result of the deficiency in precipitation and the heavy withdrawal of water for irrigation, all of the wells on the heavily pumped areas showed sharp declines in water level in 1939. In 1940, however, the precipitation was 6.57 inches above the annual average. As a result of this increase in precipitation and the accompanying decrease in pumpage, six of the eight irrigation wells (fig. 13) showed a rise in water level in 1940. The average net rise in water level for the eight irrigation wells was about 0.17 foot in 1940. In 1941, the precipitation was 11.57 inches above the annual average, and the average net rise in water level for the eight irrigation wells amounted to about 2.27 feet.
That the annual recharge from precipitation in the Arkansas valley is high is further substantiated by the following discussion. Slichter (1906, p. 24) calculated the underflow in a narrow part of the valley near Hartland and found it to be about 2,100 acre-feet a year. At this point the river has cut its valley into relatively impermeable Cretaceous rocks; the alluvium is only 2,250 feet wide and has an average thickness of about 33 feet. Slichter stated that the above figure for the underflow "represents the maximum that can be claimed in a high estimate." McLaughlin (1943, p. 77) estimated that the "underflow" near Hartland is much less than 2,100 acre-feet a year and probably is about 1,000 acre-feet a year. Nearly 20,000 acre-feet of water is withdrawn annually for irrigation in the Arkansas valley in eastern Kearny County (McLaughlin, 1943, p. 77), and approximately 30,000 acre-feet of water is withdrawn annually from wells in the Arkansas valley in western Finney County--a total of about 50,000 acre-feet. An annual "underflow" of 1,000 acre-feet of water a year obviously would be entirely inadequate to supply either the 50,000 acre-feet that is withdrawn annually from wells in the valley in eastern Kearny County and western Finney County or the large quantity of water that is transpired by plants.
The average annual loss in the flow of Arkansas River between Syracuse and Garden City amounts to about 25,000 acre-feet (p. 77, table 7). If it is assumed that most of this loss takes place east of Hartland, then Arkansas River would contribute about 25,000 acre-feet of water a year to the ground-water reservoir in eastern Kearny County and western Finney County. If this assumption is incorrect, then the river probably would contribute less than 25,000 acre-feet of water a year to the ground-water reservoir. The remainder of the water pumped from wells (about 24,000 acre-feet a year) and transpired by plants probably is derived largely from precipitation.
Recharge in depressions--Shallow depressions, or sinks, are common in the Finney basin and on the uplands in this area. Following heavy rains, water collects in these depressions and forms temporary ponds. The water in some of the ponds disappears in a short time, whereas in others it remains for several weeks or months. Whether or not such intermittent ponds can furnish water to the underground reservoir depends entirely on the character of the underlying deposits. If the underlying deposits are sufficiently permeable, recharge will take place and the water will soon disappear from the pond; otherwise, water will stand in the pond until it is all evaporated. In studying similar depressions in the High Plains of Texas, White, Broadhurst, and Lang (1940, p. 7) found that:
"The bottom of most of the depressions is covered with deposits of silt and soil. ... After the ponds become dry, fractures and crevices several feet in depth frequently develop in their beds. In some of the depressions small sinks, apparently developed by solution channeling in the underlying caliche deposits, are present. These crevices and solution channels may provide a pathway for the downward movement of water for a time after the ponds are filled, although they may become sealed after water has stood over them for several days."
Similar conditions probably exist in some of the depressions in this area. Mud cracks were noticed in the dry bed of some of the depressions, but to what depth they extend is not known.
The water table mound shown in sec. 2, T. 26 S., R. 27 W., on plate 1 was formed by water percolating to the water table from a shallow upland depression. Fluctuations of the water level in well 443 (fig. 11), which is situated in the depression, reflect this recharge. The water level in this well, even though it is about 112 feet below the surface, responds readily to precipitation. Normally the water level in a well in which the depth to water level is this great would not fluctuate as much in response to precipitation as does the water level in this well. After each heavy rain this depression becomes filled with water, and a part of the water that accumulates in the depression seeps through the soil and percolates downward to the water table. Upon reaching the water table the water at once starts to move from higher to lower altitudes, and the mound gradually becomes smoothed out. After each heavy rain the mound is again formed and the process is repeated. Other depressions on the uplands probably act as catchment areas for recharge in the same manner as the one just described, and the water table beneath them probably shows similar fluctuations.
Summary--From the foregoing discussion, it can be seen that the amount and frequency of recharge from precipitation varies greatly from one locality to another, depending entirely on local conditions. Theis, Burleigh, and Waite (1935, pp. 2-3) believe that:
"On the average over the High Plains only about half an inch of water a year escapes evaporation and absorption by the vegetation and percolates through the soil to the ground-water body."
In areas such as the sand hills (dune-covered areas) and the shallow-water area along the Arkansas valley, the amount of recharge from precipitation is undoubtedly much greater than half an inch of water a year, but on the upland surface where the water table lies at greater depth the recharge from precipitation probably is less than half an inch a year. Much of the upland surface in Finney and Gray counties is mantled by loess, which greatly impedes if not prohibits the downward movement of water from the surface.
In many places the descending water will not recharge the underground reservoir directly beneath the area of intake, for the water, upon reaching a bed of impervious clay, silt, or caliche, must take a lateral course until it comes to an opening or pervious zone before continuing its downward course. In all probability the water follows a very irregular course from the surface to the water table.
Recharge from streams
Two factors determine whether or not a stream is capable of supplying water to the underground reservoir. First, the water surface of the stream must be above the water table; and second, the material between the stream channel and the water table must be sufficiently permeable to permit water to percolate downward. If the water surface of the stream is lower than the water table and the materials forming the sides of the channel are permeable, the process is reversed; that is, the ground-water reservoir will discharge water into the stream. A stream that contributes water to the zone of saturation is said to be an influent, or losing, stream; a stream that receives water from the zone of saturation is said to be an effluent, or gaining, stream.
The greatest amount of recharge from streams in this area comes from Arkansas River. East of a point a few miles downstream from Garden City, Arkansas River is a perennial stream fed by ground water, and, therefore, cannot, under normal conditions, contribute to the zone of saturation. Upstream from this point, however, Arkansas River, because the water table is lowered by pumping, loses water to the underground reservoir during certain periods and receives water from the underground reservoir during other periods. The relationship between Arkansas River and the water table is shown diagrammatically in figure 14. During periods of below normal rainfall and heavy withdrawals of water for irrigation, the water table in Finney County declines below the level of the channel of the river, and water moves from the river into the ground-water reservoir (fig. 14B). During periods of above-normal rainfall and light withdrawals of water for irrigation, the water table rises above the level of the channel of Arkansas River and the river gains water from the underground reservoir (fig. 14C).
Figure 14--Diagrammatic section showing relationship between Arkansas River and the water table. A--Longitudinal section showing (right) how river gains waer and (left) how it loses water. B--Transverse section across influent part of river. C--Transverse section across effluent part of river.
The greatest amount of recharge probably occurs after the river has been dry for some time. While the river is dry, the water table beneath its bed and adjoining lands is lowered not only by withdrawals of water for irrigation but also by evaporation and transpiration. When the river starts to flow again, the surface of the stream is considerably higher than the water table, and water moves from the river into the zone of saturation. After the water table rises to the level of the stream, there is little loss from the river, and when the discharge of the river decreases, ground water moves back into the rover.
Water is transmitted slowly through the ground; therefore, there is somewhat of a lag between the fluctuations of the river and fluctuations of the adjacent water table. The longer the river maintains its high stage, the farther back from the river will the water table be affected. The monthly water levels in wells 266 and 361 are plotted in comparison with the monthly stage of Arkansas River in figure 15. Both wells are shallow and are situated in the valley a quarter of a mile or less from the river. The wells in figure 15 are close enough to the river so that the fluctuations in their water levels follow very closely fluctuations of the water surface in the river.
Figure 15--Hydrographs showing the monthly water levels in wells 266 and 361 and the monthly stages of Arkansas River, in 1940 and 1941. The stages of the river shown are for the same days the water levels in the wells were measured.
The high stages caused by floods are usually of short duration and, therefore, do not affect the ground-water levels very far back from the river. Slichter (1906, p. 39) made water-level observations in test wells adjacent to the river during a flood on July 27, 1904. He found that the water levels in wells within 900 feet of the river fluctuated quite accurately with the changing level in the river, but that the water level in a test well half a mile from the river showed no effect of the flood.
Further evidence of recharge from Arkansas River is furnished by comparing the discharge measurements of the river at Syracuse and Garden City and computing the loss or gain in flow between these two stations (table 7). Between October 1, 1922, and September 30, 1942, the average annual net loss of water between Syracuse and Garden City was about 25,300 acre-feet. Most of this water was lost between Hartland and Garden City. Transpiration by plants and evaporation into the air account for part of the loss in stream flow, but a large part of the loss must be attributed to ground-water recharge. The area between Hartland and Garden City offers excellent opportunities for recharge from the river. The valley in this area is very wide and is underlain by 33 to 40 feet of alluvial sand and coarse gravel, beneath which there is 225 to 320 feet of Pleistocene and Pliocene silt, sand, and gravel. The large draft on the ground-water supply made by the great concentration of irrigation wells in this part of the valley causes the water table to decline each year during the pumping season. If there was no pumping in this area, the water table probably would be built up to the level of the river and there would be little or no loss from the river.
Arkansas River is not the only stream in this area that contributes water to the zone of saturation. The northeastward flexure of the water-table contours in T. 22 S., R. 34 W., Finney County (pl. 1), indicates that the two ephemeral streams ending in that area and the Great Eastern irrigation ditch (p. 58) contribute water to the ground-water reservoir. Throughout most of each year these streams are dry, but during times of heavy rains in areas to the west they carry a large volume of water. The channels of these streams lie above the water table, and the deposits beneath the channels are sufficiently permeable to allow water to percolate downward. A large part of the water carried by these streams is emptied into shallow surface depressions, whence it disappears partly by evaporation and partly by seepage into the ground.
No evidence for recharge from the other streams in this area is apparent on the water-table contour map, but Pawnee River, Crooked Creek, and other smaller streams undoubtedly supply some water to the underground reservoir.
Recharge from irrigation water
Ground-water recharge from irrigation water is accomplished in two ways--by seepage from canals and ditches and by downward percolation after the water has been spread on fields.
The irrigation canals and ditches in this area are above the water table, and seepage from them probably takes place whenever they are filled with water. The amount of ground-water recharge from irrigation water that has been spread on fields probably is small except in areas where the water table is shallow. Water pumped from irrigation wells generally is carried in ditches and spread over fields that lie at considerable distances from the well. In the immediate vicinity of the pumping well the water table declines, but in the area where the water returns to the ground-water reservoir the water table rises temporarily.
The hydrograph of well 269 illustrates the effect of returned irrigation water, precipitation, and pumping on the water level in a well (fig. 16). A continuous automatic water-stage recorder has been maintained on well 269 since July 17, 1936, by the Division of Water Resources of the Kansas State Board of Agriculture. Complete water-level records of this well have been made available through the courtesy of Mr. G. S. Knapp, chief engineer. Well 269 is a shallow well in the Arkansas valley in the vicinity of several irrigation wells (pl. 2). The highest water level in this well is reached each year during the pumping season in July or August as a result of recharge from returned irrigation water and from spring precipitation. The small peaks occurring in July or August are the result of irrigation water returning to the underground reservoir in the vicinity of the well. The rise in water level caused by returned irrigation water, however, is only temporary, for the water soon starts moving from the area of recharge to the cones of depression produced by the pumping wells. Water moving into the cones of depression while and after the wells are being pumped causes a decline in the water levels of all wells in the pumping area. This decline was especially noticeable in 1939, which was characterized by abnormally low precipitation and unusually heavy pumpage. As a result of the light pumpage and nearly normal precipitation in 1938, the water level in well 269, after reaching its peak the latter part of July, declined only until about October 1, at which time it again started rising. There was not sufficient precipitation in 1939 to recharge the underground reservoir and, as a result of the heavy withdrawals for irrigation, the water level declined until the supply was replenished by precipitation the following summer (1940). In 1941, the water level rose rapidly in response to the heavy precipitation. Pumpage in 1941 was comparatively light; therefore, the temporary rise in water level caused by the return of irrigation water was small and the decline that followed was small.
Figure 16--Hydrograph of well 269 showing changes in water level caused by returned irrigation water, by precipitation, and by pumping. Water-level data supplied by the Division of Water Resources, Kansas State Board of Agriculture.
In an unpublished report* by the Bureau of Agricultural Economics it was estimated that 10 percent, or 7,500 acre-feet, of surface irrigation ditch water annually returns to the ground-water reservoir by deep percolation from irrigation canals between the Colorado-Kansas state line and Garden City.
(* Water facilities area plan for the Arkansas valley in western Kansas, Water Utilization Section, Division of Land Economics, Bureau of Agricultural Economics, U.S. Dept. of Agriculture, 1939.)
There is probably little recharge from the irrigation water spread on fields on the uplands where the water table is deeper, except where the water collects in depressions or where crops are excessively watered. Irrigation water may increase recharge from precipitation by supplying a part of the moisture needed to overcome the deficit that may exist in the belt of soil moisture.
Recharge by subsurface inflow
The movement of the ground water in this area, as indicated by the slope of the water table (pl. 1), is in an easterly direction; hence recharge from precipitation or stream flow that occurs in areas to the west eventually moves into this area and contributes to the available supply of ground water.
Much of the water contained in the Dakota formation probably entered the formation in areas of outcrop west of Finney County. Sandstone of the Dakota formation is exposed over wide areas on western Las Animas County, Colorado, and adjacent areas and undoubtedly absorbs water directly from rainfall and from streams that cross the outcrops. It is likely that a part of the water thus absorbed by the sandstone travels down the dip into Kansas, and probably migrates into the overlying Ogallala formation at places where the two formations are in contact and especially where the Dakota is thin (Latta, 1941, p. 41). The small rainfall in southeastern Colorado and the capacity of the sandstone to transmit water laterally limit the amount of recharge received in this manner.
Summary of ground-water recharge
Most of the annual recharge to the ground-water reservoir beneath this area is derived from precipitation that falls on the surface on these two counties and from seepage from streams and irrigation ditches, especially Arkansas River at times when it is influent. The available data indicate that the recharge for the entire area is only a small percentage of the total precipitation. Areas such as the Arkansas valley, the sand bills, and the Finney basin receive much more recharge than the upland areas.
Prev Page--Water Table || Next Page--Discharge, Recovery
Kansas Geological Survey, Geology
Web version April 2002. Original publication date Dec. 1944.
Comments to webadmin@kgs.ku.edu
The URL for this page is http://www.kgs.ku.edu/General/Geology/Finney/06_gw5.html