102 804 'THS A STUDY OF SOIL WATER MOVEMENT AS AFFECTED BY DRAINAGE Thesis Ivar tho Degree of M. S. MICHIGAN STATE COLLEGE Joseph Barnsfein I949 This is to certify that the thesis entitled "A Study of Soil Water Movement as Affected by Drainage" presented by Joseph Bornstein has been accepted towards fulfillment of the requirements for ".3. degree in Agr. Eng. F. I. Peike rt Major professor Date wrCh 10.1949 _ . . - -w".. , . V f .. .r 2' c . ,: 2.',-l“,’_"vff.t.;..’ . Mr ”Ll:- . ;° gig-44.1 5-1;; ----- ...... 'o n L...- I V . '. - P... r . "7' 5’ t’! . . _ . ' 0 ~ M . ’- al.’-‘f. (. ).. s’r . .‘» ,I ‘ I . If]. V A STUDY 01' SOIL WATER MOVEMENT AS AFFECTED BY DRAINAGE 3! Joseph Bornstein w 1 meeie Submitted to the School of Graduate Studies of Michigan State College of Agriculture and Applied Science In partial fulfillment of the requirements for the degree of MASTER or some: Department of Agricultural Engineering 1949 THESIS ACKNOWLEDGEMENTS The author wishes to express his apprecia- tion to the following: Professor Frank I. Peikert of the Agricul- tural Engineering Department for his continued assistance during the entire construction, oper- ation, and reporting of this project: 3. Am Meyers, District Conservationist, Soil Conservation Service, Saginaw, Michigan, Dennis Mcguire, Farm Service Adviser, Consumers Power 006, Owosso, Michigan. and Paul Albosta, farmer, Alicia, Michigan, for assistance and equipment: Professor J. N. Winburne of the Written and Spoken.lnglish Department for his constructive criticism of this thesis. 13 pa F“ CA L! 11 TABLE OF CONTENTS Page INTRODUCTION’———— -- w - - - — -— A - 1 REVIEW 0! LITERATURE -----------—------ ~—- ——-- - — 3 Permeability and Drainage ~—- -—___ —= — - - -— 3 Coefficient of’Permeability —~--—— ~——- —— 5 Application of Coefficient of Permeability —--------- 7 Piezameters and Soil Water Flow ----_ ----------------- 8 Other Draw Down Investigations -------------------- 9 Electrical Analysis in Drainage Studies ------------- 10 Summary of Previous Investigations ----------------- 12 EXPERIMENTAL WORK ~ - -- ——-—— -— --r‘— -— ‘—= 14 Soil Description =—- --— - - — 14 Project Construction — — — —- — - -——- - 15 Draw Down Curve Analysis ‘_ ——- -—‘—— — 20 Observation Wells between Tile Lines -- — r 34 CONCLUSIONS ————- ——- — —— --—-——- — w — 37 RECOMMENDATIONS — —- --* — — — - — —— A -——— ——=— 37 BIBLIOGRAPHY —— - - - - _ - — 39 APPENDIX.- ———- —————_—_ —_ — - —_ 41 Table 1 Table 2 Fig. rig. Fig. Pig. rig. Fig. Fig. Fig. rig. Pig. rig. rigs Ollfiblwt" 10 11 12 Graph 1 Graph 2 Graph 3 Graph.4 Graph 5 iii LIST OF FIGURES, TABLES, AND GRAPES Changes in Permeability — —— -—— _ _ — _.. Recommended Depth and Spacing of Tile --------- TTP1Ca1 Draw Down Curve Cross Section ------------ Tools Used for Project Construction — - — — -— ‘Observation Well Electrical Depth Measurer ------- Electric Poles Wired to sump Location ---—------ 'Sump, Showing.Pump Outlet, Supporting Hope, and Sump Box —— — —— —— — — _ __—-— __ _ 2 Pump Shelter, Showing Power Line Lead-in and Hater Discharge Hose ——- _ — _ - _ w _ __ _ _ Flint and walling g 35?. Sump Pump with Autmmatic Float Switch - — - - - ~~ —_ _ —_‘— —__ Uncompleted Observation Well —-—— — -4 Completed Observation Well -— — — -=___ __ Project Area Looking Southwest —- - __— _ Project Area Looking Northeast —- —- —— — Rain Guage in Observation Well.Area -------------- Draw Down Curves from Observation Wells 41-97 —- —— _— _—-_=_—- — _ Draw Down Curves from Observation Wells tie-#14 -- — —- __— -_- u Draw Down Curves from Observation Valle #15 - #21 —- — — - — —_ Draw Down Curves from Observation Wells m-m _-aa=‘— z — — Comparitive Draw Down Curves -‘ —— -——— — —_ Page 16 16 18 18 19 19 21 21 22 23 26 27 28 Graph 6 Graph 7 Graph 8 iv Draw Down Curves from Observation Wells between Tile Lines — — —— _ _— - __ Plan of Observation Wells — - — — — -_ _ Profile of Observation Wells --------------------- Sump and Observation Wells between Tile Lines ---- Page 29 31 32 36 A Study of Soil Water Movement as Affected.by Drainage Introduction , Preper spacing and depth of tile and moles for land drainage was the primary objective.of this project. If wider spacing would give ade- quate drainage considerable time and money might be saved in future tiling of fanm land. More effective drainage would result if it was found that closer spacing was needed. Also of great importance was the determination of the practicability for further research of the sump and observation well method which was used. This is its first trial in Michigan and one of the few in the United States. The soil of Saginaw County, Michigan, includes areas of Clyde clay loam which are considered to be poorly drained soils. without artificial drainage this soil has definitely limited farming use. Properly drained, however, the Clyde loams are very productive. When the water table is lowered sufficiently it produces abundant yields of sugar beets, nary beans, onions, and forage crops. Therefore, it is to the advantage of owners of Clyde soils to be sure that their subsurface drainage systems are properly designed. During the last two years there has been considerable interest in tiling and in plowing of mole drains on the 7500 acres of land known as the Prairie Farms in Saginaw County. Much of this area is classed as Clyde soil. In 1948 alone several miles of tile plus a limited number of mole drains were placed. Before this investigation no attempt had been made to find the optimum spacing and depth for tile and moles for these Prairie Farms' soils. This sump and observation, or test well method was tried -3... during May and June, 1948. The project, set up cooperatively by the Soil Conservation Service and the Agricultural Engineering Department of Michigan State College, consisted of a central sump or well from which water was pumped and observation wells. The latter were placed in two lines at right angles to each other, and the sump was located at the intersection of the lines. - 3 - REVIEW OF LITERATURE Permeability and Drainage Most drainage researchers have approached the problem of depth and spacing of tile and mole drains through a study of soil permeability. If there is a relatively pervicus layer of soil resting on an impervious substratum, the incidence of rain causes a body of ground water to build up, bounded above by the water table. The height of the water table can be controlled by antificial drainage for given conditions of rainfall, soil permeability, and drainage layouts The investigation of the shape and location of this free water surface and the nature of ground water flow to the drains is primarily a study of Ivdraulic head. Design of drainage structures also involves the quantity of water to be handled and the dimensions of the aquifer. The aquifer is defined as 'those strata that provide a medium through which water mey flow at a satisfact- ory rate toward a drainage facility.“1 Although there is no generally applicable procedure for mathematical analysis of the problem, Darcy's Law of flow through porous media has been called the fundamental equation of ground water movement. This law, which follows in equation form, states that flow porous media to drains is directly preportional to the hydraulic gradient and the cross sect- icnal area and inversely prepcrtional to the length of flow. nag-n1) I: g: “Q" is equal to the quantity of flow (cfs) '1' is equal to the constant of proportionality and depends upon permeability of the medium, viscosity and density of the fluid, and the gravitational constant. -4... ha - hl is equal to the hydraulic gradient (equal to the vertical distance from the crest of the water table to the water in tile). g is equal to the cross sectional area of flow 1 is equal to the length of flow There are other equations for pemeability calculations. Some of these attemts to simplify the procedure have given inaccurate results. These have assumed that flow takes place below the level of the drain, that above the drain all streamlines are horizontal, and that the water table must intersect the drain lines.‘ Good drainage, according to J. H. Neal, is primarily the removal ofexcess or injurious water. Insurance against drouth is an added characteristic. By lowering the water table early in the growing season roots can get down to a permanently moist soil. The water table for grain crops can be one foot below the surface all the time or two feet below three-fourths of the time without injury." Drainage results in improvement of the physical properties of, the topsoil, but the effect decreases with depth. With proper drainage bacterial action can take place which causes the decomposition of organic matter, and increases the available plant food. Also the improved ventilation aids the stabili- sation of colloidal material in the top layers of the soil. Porosity increases when drain spacing is decreased. Cracks, root cavities, and worm bores all develop with time. During early years of a tile system four feet deep in heavy clay, water will not drain as fast as is required: however, soils will become more open in time and much valuable plant food will become available as the subsoil is loosened.20 It has been shown, too, that deeper tile in fairly permeable soil start to drain first when rain has saturated the subsoil and raised the water table. Deeper tile will also carry more'water longer after a storm than shallow tile. There - 5 - is, however, the danger of tile or moles being too deep in a tight subsoil. In which case until the system begins to take effect craps may be drowned before the injurious water is removed.20 To allow for this improvement with time, J. R. Haswell of Pennsylvania State College recanmends that an adequate system of mains be planned first.n H. B. Roe's studies of drainage improvement in Minnesota are shown in Table (1)3-9 Some aver- age depth and spacings of tile drains which have been suggested are shown in Table (2). Table 1. Changes in Permeability Date Installed 1915 1915 1925 Topsoil very fine sandy loam clay loam ~ 12 - 18" of traprock fragments peat over subsoil muck Depth 3 - 3%" 3%” at - 4' Spacing 220' 66 - 110' 125' When wet sticky, impervious ditto, lac- ustrine clay Effect of little drainage little till Slowly in- time (1921 but increas- after 1919, in- creasing in- ing influence to creasing fluence, 1925 present time, 1988) to 1938 Coefficient of Permeability Host attempts to solve the depth and spacing problem by mathem- atical formulae have made use of a coefficient of permeability while 2,23 other equations have been based on moisture equivalent, which is a function of permeability. The former drainability factor varies with 23 Moisture equivalent is the amount of moisture remaining in capillary pores of soil after centrifuging a saturated seaple in perforated cups at a speed equivalent to a force 1000 times that of gravity. l""" 5011 Clay Clay Clay Clay Clay Hardpan Clay loam Clay loam Clay loam Peat topsoil Medium sand Clay loam *Note: Ge »-4 Of “1 f. seasoé heavi— zafiygroscopici temperature' -7- the moisture content of the soil, but is not a simple relationship, although it does diminish rapidly with moisture content .4 Don Kircham refers to this constant as governing the rate of flow through the ”11.13 Its determination for application to drainage design requires measure- ment of permeability of disturbed soil samples and measurement and correlation of permeability of undisturbed samples taken parallel and normal to bedding planes. Comparisons must be made between laboratory and field measurements. Also an indirect method of rapidly evaluating the coefficient so it can be applied directly to drainage planning must he develo'ped.1 In drainage work permeability includes rate of infil- trat ion as well as percolation below ground surface. Application of Coefficient of Permeability The use of a coefficient of permeability has been made by Aron- ovici and Donnan for design of a sump and observation well systan and for tile spacing. A Mole drain spacing may also be computed from the same form- ula. lith the permeability of the aquifer known they have predicted draw down curves for wells and sums. ror sandy loam soil: 8 equal to 4P(b - a)Q where 8 equals spacing, P the coefficient of permeability, a and b equal the limits of the draw down curve (figure 1), and Q is the amount of water ’18- 0)- 19mm mm -3- drained from a soil profile per unit of time.1 When applying this formula 9. strata survey is first made to deter- mine the extent of the aquifer through which percolation takes place. With piesometers the hydraulic gradient of flow is established, includ- ing pressure and direction of water movement. The piezometers will also be useful in determining the efficiency of various drainage methods tried in the future.6 The experimental evaluation of P, the coefficient of permeability, completes the data necessary for percolation quantity computations.1 Darcy's Law is assumed correct, the hydraulic potential in a vertical direction is considered constant, and horizontal is assimed not to occur in the capillary stratum of sandy soil. Aronovici and Donnan emphasise the danger of trying to simplify the solution of field problems with laboratory procedure.’ Piesometers and Soil Water flow Discussing the piezometers as applied to ground water studies in relation to drainage, Christiansen.7 as quoted below, finds that the path of a drop of soil water and the “shape of the equipotential surfaces which are everywhere normal to the direction of flow will depend to a very large extent upon how permeability of a soil varies with depth. Ground water piesometers offer a means of studying the soil water flow conditions in the field. With pisse- meters one can determine ground water potential at any point in the soil mass. These ground water potentials are everywhere proportional to hydraulic head or height to which water will stand in an Open manometer tube. By making simultaneous measurements of twdraulic head at a number of points in the soil mass in the same vertical plane one can determine the components of the hydraulic gradient and direction of flow. Plotting the equipotent- ial lines on a vertical section the ground water flow pattern is obtained. It should be recognised that these equipotential lines are the intersections of the equi- potential surfaces and an arbitrary vertical plane and ’1 -9- that unless this vertical plane is selected to coincide with the direction of flow the hydraulic gradients as indicated on the flow pattern are not the true gradients but only the components along this plane. Thus piezometers are satisfactory for determining general flow patterns and show promise of being aids to determining the actual soil permeability. Other Draw Down Investigations Sven Hurling,17 reporting on his depth and spacing experiments at Albert Lea, Minnesota in 1921-23 feels that the soil survey should go to a depth of seven or eight feet for all drainage work. His project was similar to this Saginaw County drainage study with observation wells spaced between tile lines. Wells were dug six to ten feet deep, lined with five inch tile, and covered with a wooden block. During the three seasons of investigation complete rainfall records were kept. Accurate soil profiles were taken between each tile line. On heavier soils deep drainage was needed to increase percolation, to limit the runoff, and to protect the lower lands from flooding. Neal's field studies, similar to this investigation of soilwater movement as affected by drainage, and Aronovici and Donnan's mrk show that flow toward drains occurred only when the hydraulic head was greater than five feet per 100 feet, varying somewhat with soil and tile spacing. Pluctuations in test well level and tile flow were attributed to precip- itation, transpiration, evaporation, runoff, and deep seepage. Neal'used six foot observation wells of four and five inch tile spaced five and ten feet apart.16 Moisture Equivalent and Soil Water Movement Based on the field work described above and the formulas of- Rothe,16 Neal developed equations for rate of water table decline and tile depth and spacing. Rothe's equation E : mtg ' '9 based on soil hygroscOpicity was not a readily adaptable index for engin- eers working with farmers. In Rothe's formula I equals spacing: in meters. ' equals hygrosc0picity, and '5 equals percent of particles less than 0.002 mm. minimum’diameter at 1.25 meters. Instead Neal worked from moisture equivalent, which though not easily determined in the field, could be related to both soil plasticity and clay content. He found that the plasticity and clay content of soil have definite percentages of moisture equivalent when scaled together on the nomograph. The percent of clay was determined by the hydrometer. Also, Mr. R. L. Patty of South Dakota State College states that soil drain- ability depends largely on particle size and clay colloid content. .A soil having 35% colloidal clay has average drainage qualities, while, at 45% it has poor drainability.18 According to Neal the average rate of drop of ground water midway between tile is found from the equation 14 = 0.165 8 where Rd is the rate of drop of ground water midway between tile, in feet. 8 is hydraulic slope. Neal also reports that Ed: K(T.)°'7 I is a drainage factor depending on soil and hydraulic slepe. It is measured by moisture equivalent when a definite slope is considered. T. equals tile spacing in feet. - 11 - Graphs in Neal's report relate K to lower and upper plastic limit, moisture equivalent, and clay content. In equation T. 12000 : and ma: 17.5 “—15.5 0 . I. is moisture equivalent, and Ta equals tile depth in feet. The reasoning behind the above formulae is that the larger the pores the faster the soil water movement will be; therefore the spacing and the depth of drain lines is a function of nonrcapillary porosity. Pore space is, of course, a function of moisture equivalent: therefore, Neal says, it is reasonable to expect drain spacing and depth to be dir- ectly influenced by the moisture equivalent, a factor more easily deter- mined than pore space. Combining the known or desired values it is possible to vary drain placement to give different rates of drOp of water table, within the range of 0.2 to 2.0 feet per day.16 Results of the application of this method can be checked with results from other areas with accuracy and the method is readily applicable by field engineers. Electrical Analysis in Drainage Studies The difficulties of mathematical analysis of field drainage prob- lems are at present unsolvable. according to Mr. E. C. Childs5 of the Cam- bridge University Agricultural Experiment Station, Cambridge, England. Progress has been.made with electrical analogy, since the equation of ground water flow to parallel drain lines is also the equation of two- dimensional flow of electricity in a sheet conductor cut to the shape of the soil cross section. Three sets of experiments are presented by Mr. -13... Childs. In one he shows how the water table is lowered by increasing drain diameter, and by decreasing tile spacing. A second investigation gives the relation of water table height to rate of rainfall. The water table midway between tile lines is at higher elevations for greater rain- fall intensities. The third study determines the effect the impermeable soil layer below the drains has on tile efficiency. The influence of the hardpan on drain lines varies with its depth below the lines. It is shown that the well hown formula giving an elliptical water table sec- tion is not in agreement with the requirements of the theory of flow through soil from higher to lower elevation. In later studies of ground water movement affected by drainage using electrical analogues, Mr. Childs found that observation wells may cause significant disturbance to the water table and to soil moisture seepage, particularly at the point of examination. He also found that lowering the water level in drains alters the shape of the water table and lowers its height at the midpoint by an approximately equal amount, that an open ditch is more efficient than the same ditch piped and filled in, and that neglecting the cap- illary fringe causes little error in estimating the position of the water table. The capillary fringe is the distance above the water table of a given soil that water will move upward by capillary action.5 Summary of Previous Investigations Depth and spacing research and subsurface drain design require con- sideration of several factors. Some of those considered are outlined below. 1. The texture of the soil is one of the most important factors.” 2. leither intense nor long-continued rainfall are themselves good in- deces of needed capacity of a subsurface drainage system. -13.. 3. Heavy subsurface runoff, soon after rainfall, is not always required but its necessity depends on soil and subsoil texture, previous soil moisture content, and the season relative to plant growth. 4. Proper determination of the maximum required effectiveness of a tile system should generally be based on soil moisture and runoff conditions present during the early weeks of the growing season. 5. Effectiveness of a tile system as crop protection and growth stim- ulator is dependent on rate of drop of the watertable at the midpoint between lines. This rate depends upon texture and moisture condition of soil when drained and on depth and spacing of tile. Previous dis- cussion has shown that the rate of drop of the water table and the depth and spacing of tile are definite functions of moisture equivalent, plas- ticity, and clay content of soil considered. 6. Methods of design require checking on different soilsvunder various climates. Research is needed to determine rate of water table lowering for other areas. 7. Results of depth and spacing fomulee should be comparable to recom- mendations of American drainage engineers and investigators.16 - 14 .- EHERIMENTAL WORK The project outlined called for determination of the ”gradient of the water table in a given soil.'25 The central sump, sump pump, and radiating observation wells were installed in a typical soil at a suit— able location in the area known as the Big Prairie in Saginaw County, Michigan. iieasurements of the water level in the wplls were made period- ically while the pump was in operation. from the above measurements it was proposed that prOper depth and spacing of tile or moles could be determined for the particular soil. The bases for the sump and test well system for water table measurements were the recommendations of Mr. L. A. Jones, Chief, Division of Drainage and tater Control, Soil Con- servation Service, Research Division, as outlined by him in late 1945. Soil Description On May 13, 1948, the topsoil was found to have 93-11% moisture by dry weight. The tight yellow clay subsoil also appeared saturated al- though the moisture content was only 39-38% of. dry weight. Results of the size particle analysis by the Department of Soils, Michigan State College showed the clay subsoil was 12.7% sand, 21.6% clay, and 65.6fi clay. In the soil profile the clay was located at 15 inches, overlain by a black, mucklike topsoil, and the change from muck to clay was quite marked. The description of Clyde clay loam as. given by J'. A. Bonsteel3 in the United States Department of Agriculture Department'Bulletin #141 and by the soil survey rwort of Saginaw County is verified by the experimen- tal conditions.15 According to these publications Clyde clay loam and Clyde clay loam, much phase, were formed through redeposition of fine 25 Appendix, P. 42. -15.. grained glacial materials in the beds of extinct glacial lakes and particularly in depressed areas where natural drainage conditions were very poor and where partially decomposed organic matter accumulated abundantly.3 The mucky phase differs from the typical soil in having a surface covering that is as much as 12" in thickness. Where the exper- iment was located had 15' of muck. This is close enough to specifica- tions to be included in the Clyde series. As mapped in Saginaw County the Clydes are restricted to the natural flood plain of the rivers, where adverse natural drainage and water conditions necessitate large expendi- tures to reclaim the land for crops. Nearly 90 percent of the total area of this soil supported a cover of marsh grass and sedge after it was un- covered by glacial lake waters.15 lhere substantial amounts of organic matter have accumulated the clay loam is classed as mucky phase. Such a condition is closely related to the typical soil. The topsoil is fri- able and easily worked when not too wet, for such fine textured soil. Lime Carbonate content is generally between one and two percent. The sub- soil, a depth of 36 inches, is play of light gray to yellow mottled color with some gravel. It is no exaggeration to add that it is a dense sticky mat erial when wet. Project Construction Previous to setting up the project in the field, numbered marker stakes and' observation well covers were made. A pump shelter was con- structed and a battery operated depth measuring device was designed and assanbled. Tools used at the Prairie terms are shown in figure 2. A tile puller, the second item from the right, was simply a threaded bent piece of pipe about 4" long that could be attached to the handle of the post hole auger, with the blade removed. lhen disassembling the project -16- figure 2. Tools Used for Project Construction 1 x x l *. ‘ ,. , 7 '0 -e~." ‘ —- " .v'f'x‘. . , fit."- s. . ' In H" :’ " 't'. t‘ " - sz-‘~ MA.“ _ ---~-- —~ ~—- .77. 1!.- .V l I. ‘ . . . ',. ‘. _ . ' ' Le 'e’ J4 . “‘3- 1‘ s’.‘_ Lq-p‘u L~ ‘ _ . 4 f, z .2 ., 3- ‘r fay}. A . . ‘2‘ a x“? .. .7 ., , WSW... ,;: , , Q Q : .', If ,. e._e' ‘ , ‘ e , . _ (._’ .' .L ' O ‘s _ -. ‘_v ‘ $" ._A' V“.£“"4fih ' "H;L 0L. ‘1‘ 19““; .1 ‘; Lg ' ' ‘ . O ‘ k I l‘igure 3. Observation Iell Electrical Depth leasurer -17.. it was lowered into an observation well, hooked under the bottan tile, and withdrawn with all the tile. . . The depth measuring device shown in figure 3 depended on the electro- lytic-action of the soil water to complete the circuit. 'A 2% volt dry cell supplied the current to a Model T ford Spark coil and an electric fence light. When the measuring rod was lowered into a well the wires at the lower end of the rod, making contact with the water, completed the circuit. The light would then be illuminated. A six foot scale attached to the rod was read directly against the side of the tile nearest the marker stains. (An ordinary six foot steel tape was also tried and found practical for this purpose. As the water level receded, however, it be- came more and more difficult to see Just when the lower end of the taie reached the water surface.) Preliminary work at the Prairie farms was done during the second half of April, 1948, Just after the flood waters had receded enough to leave the land of the project area visible. his poles were erected and powerlines were strung to the pumpsite (figure 4). A wooden box open at both ends was constructed in the 2% x 2% x 6% foot sunp to prevent cav- ing of the sides (figure 5). The pump shelter (figure 6) was placed over this central well and the * LP. flint and falling sump (figure '7), equipped with an autmnatic float switch, was hung from its roof. Three sections of hose totaling 150 feet were attached to carry the water away from the immediate drainage sons. By having the '9qu suspended from the roof of the shelter there was no possibility of it digging itself into the ground (figure 5). The pump was lowered into the sump enough to reduce sump water level to approximately five feet below the gseund level. At the same time the twenty-eight test wells were sunk and tile placed in them up to the ground surface (figures 7 through 11). Ends '— 18- Hom ham and oemom wautommnm eveapso 95% mat—ohm .mfism on enema 9.3m ”83.33 on» 3 can. .38 3.383 2.. 8sz ll £33m soon ”.3333. fie. .ooom outdone the: use 5.33 E gm a w M533 6 95.3 .N. 0.3th 33 aerom msgoafi 332$ E .m earn I c ‘. _.fiM c u . \4. h. IS- “an \. 1.. elf. K . .II aeu". - _ ..... . . . .. O.‘ . -19 -' -20- of each tile were chipped slightly to insure easy flow of water into the wells. The topsoil was packed Just tightly enough around the tile to hold them in place. The depth of the test wells varied frm three to five feet. Going outward from the sump the first three wells were five feet deep, the next three were four feet, and the last one was three feet in depth. The lines of wells were run arbitrarily northeast, southeast, northwest, and south- west fran the sump, and were. starting from the sump, spaced at 5, 5, 10, 15, 25, 40, and 32 feet apart (figures 12 and 13). As the wells were dug and tiled the. numbered stakes were placed next to them and cover was placed over each one. Covers later proved to be of no particular value. Two additional wells were dug and tiled, one at 830' north and one at 330' east of the sump to serve as control wells. It was hoped that these would show a different rate of drainage from that recorded in the obser- vation wells. In both of these outer wells, however, the water level receded at a rate equal to the drainage in the sump area. A rain guage set into the ground within the radius of the observation wells (figure 4) showed no precipitation during Operations. _ The levels of the tops of all the wells, except the two at 330 feet, were taken in relation to a bench mark. headings of water level in the observation wells were made at least three times a day from May 14 through June 3rd, except Hey 30. when only one measurement was recorded. The depth readings, in inches below the top of the tile, were converted to feet and tenths and substracted from the elevation of each tile in order \ to plot the draw down curves. Draw Down Curve Analysis Pumping began on May 14, 1948. The draw down curves that resulted I'"\ . ,‘a figure 9. Completed Observation Iell Tigers 11. Project Area Looking Northeast. '1.er 130 m W 1. “IOfltti-u Iell Area. Mgrw eel \, , ._ 6 IOO T l 1 I I l -_.__..- e - -QE. -; TEST WELL SIAN 5340.4”- rviflftu. . oH—g—o *Md_._~'v_._--_‘ I ' 0 ...... O. . B. H . 6.2 LI .... -.- .. 77». 7.. I- _ . . fin (Onto ...IO‘ 'lel filt'lllt. ». .. . elydlll n .. ii» . 9‘ a v- ..... 582. ZO_._.<>w.._w VAJJ Fxo Q T... a .. half ZAUNPUR. ,1 v. ...... .,..a {I}. e . ' . ’ . e _ .-. .+..- true .. . e ,9 a I e e ‘1 . .0... L 10'. Is A t 901:. 7--. .. é TEQT _ ‘ - - I . . 4 . . , r . . r I . .-.,I:.. . .. ,oIol I- I q I II II I? . H.... to . 5 ._ H n . H L n k, . F 2 . g .. a I .7111 .....-7.-:-I v . v . . . t I v ..y . v.-. t .v.. . I‘M . I I v I..I+II bl IIOIIIcIIOIIIhIf . Ctr. A .0 Y 1 w ltloII . I 33 4 . .3IH . . . . . . . , _ .- - _r - pm I 1 . -I.. ‘11.. - ---. II..-.!_T.--.I.II 7 .I. I I14 III I. 'l .‘r’ n .1 “I. .dIII I4- .I o J. . I- .III... III IJT I I? v JIITIO .9 VII II.I. -YI_IIIIII JTII III.:-.I.-II.IIO III+III.I f . VI . . . w 6 e . u e t . F r w | . I! no. I. o. .I .II III. I». v III.I -1»1.1. ..I.¢. .II. . 5 5555 .._.6 .. - .. .. .a. _ L - I: -ILI.-- ,2 42gv. .. nv .w . + -11 .. l . w . a. o I l .. II t . w . I 34 law... . . . ~ . . P v ..... .II. . . a . I. I‘ . I v .MIII 061- IqI VIW: I | I I 9 JT 4 . I. - . . .. . . A i .. I. u. I .LI. In .1. .. . 1| u u I .oI . a . n .. .I .. . . w T _ E . . II J. a I I .m. ‘ Ilmvhvy c .. I a . . 4 . w . . . fl . I . A 1 .- . t I QI. ‘ an 4' +n. IO.L .I _ II x .n ,. . . _ ~ . . .. It I .. . I II V V 9 0| (ILV.M)WIIIM . hw- , fnlolo 0 I16 to I l . .3 . . I .o, . U u I ,I.It|. II II Ii I I._ l I I oInFIJIIVIlIQIrI ? I IIYI a .W..: - - .. . . . . a . a . ., r . . IIIII a H . _ . , aI .. a . . + . A . _ L , .. . - . I . . ,_ v _. q . . val... “I..- r ' I O HHW‘ . f I II'I'II'IAfiI . . I ? . O. I . .. n . I.fi,v . H . m - .. I. - mi I .u- _ q 1-..- - .Ir 1 LIMIT 1. _ . _I o ..:.-. hmflfl- 3 . -1I-I - V . l0 -_ . . . . - . I I . . n 4 u _ 6. . I “ SE 1, a _ M . wc: . . _ .r - Tm, w... . _ . v . . ST” 5. . _. . ,. S V . 4 * u Lllf Ell” . , _ .. I T DrIII LI-I-I r: I I I -I- LI I I: I - I a _ r . . o. o. o. ._ .3 2 9 5 5 .5 5 . a a . , . hm. J. any _.......4 h. Ellsrbzarfl' I Is .I \‘L n.1,»: I .- ....,- .... . I , ,..I. ”a”. 4 .n. . - ....... - .. . an? 7.2...- - a. ......... .5 I. ,. . 55 6 ._ a. . 2' H jji ..... . . n2 A _ ' I u . . . . s .7...~ H-4._.....__...4._..._..- a“..- ”Hg’ —- ‘o . . . .'I ‘c‘ \ TEST WELL pus TANCE 534.0; rest as 0.0. ' 589.0061; *1. 29233 .d IIIITT ...I..I. .. - . .. . .Y._TI 3.--.IsI. ILI. .. .I q IAIT. _ . . . . .. Ir.\w.I1l4-J I. . . . . _ I l I I I I A .4 f w I A 9 I ., Y . may L A v e 28 I32 27 M'I be‘. DISTANCE 2 2 q Mfifi— I . I ‘ I I I t "'—T TEST WEL I I 58L ZOC.<>MIE 5840 ‘58&0€.v‘1. 582 0”” ”“- ....... I32 I I h I I I 0—,- “ - I . h". ' VIIIQOIIIOII' fi . .II .II.I .__,.._ - _.Y - I I I --+ I “I I I t 1 I I . - r . I J . . - a .._...i “j: 1 r L I YI‘IITI . e I. o-- I I I I- ’. .‘. I I I I I t I I I .II.II.I I. . III+ I..- II III..II.I . . u . . . . . . _ . . a -__. _...- . .. i-.. -. I -c— n. I I I I I I 1 I o... -~-_._._._-.—-.- , . I . . n [00 Sn _. a i . a ._ ._ a " EM . C“ . ~ . N. a H x n: _, . T~ k a S. . maxi- - LII I - I 584.0 583 O 20:.<>m4m r. I 5" p l‘ - I ‘.-'-' 4"- I I _ - , _ . 9 _ I . . . A I . II I . I . 4 I I. . 4 . I T I I . I 4 I 2 .;I I H: .9 .. - IIPII .II I_II I w.I . ..I 9* . .I fl I . I II I. J vbQIII III — I ,r. _ ._ . _ III f. .I . I_wII.- T. I I 'lIIII. ‘IIOIIIO. III¢ III I v I 0 u I I I o. I . + I I I q \I L 0 III II VI I I. II w I I 0 I I fi I , III I III I I I I I I IIIA I III I” I I. I I .h. a. h I I + I IIL. I III- JIIIIIIIIII III III , I I I$ I. II II. . , . OIIOI IIQIII III! II II Y I a I . H . I “I v. ~ 9 I .I I n I I I v A IIIII+I It I _ ., .I . I .I I.I...-. I I I p I I I III--- IIL. I . I up . . . w . I. IIIIIIIIIIIIII: I _ _ . . . I II _ I I W . * b . I _ I I I . A I , I I I + . . a ; . . I. , M . W . I A x . I. III IoII I . ” ... . I . . I _ a I . I . a h I .I IZI IIII II II * , . I . g . . h ¢ H I u. I I I II ' I I I . o I . I III I I . u . . I III-L W I I . h u . . . . . . . . I . I . III “I. I _ _ I _ II . * + . . . w I fl . . . I _ I _ .* II - I I . I I IIIIwTI I I . . .I . . I _ . . I I . . v A I IIII IITIIII I M ._. __ ..... .. . I .I. III. IIIIIIIIIII . I I o I v _ . . c o I 6 . H III IIOIIII IIIIOII‘IIIOI'ILII O .I w I h I . a I . I I I * . . v I. o . Old. . _ . ”I. k .I I . I .Iowlf O VII II‘I. v o * . . I m I! * I H I . I I. II I, I. . . I # , .— . * - lb v n .I . I ‘I III 4 V A v & _ , . a - I . M - . 0 . . T. __ . .I I ...L 70 IIIIIII. I4 I IO M . . . . I II . . b III I+IIII I LI . fl . w % _ I . .. . I I - I II I ._ I , w I I .IH . I II I I I . “ II II I. I III. I I h . . _ I . y In *1 H . I I . . _ II III I 9 |% +LIQIMI+O IIIIIIIII II I+I III I Q! I I I LII II I III fiIIII Inflld I! II, . ll . I .I .. I. . W F u I II ¢ _ I g . III..IIIIMIIIWII IIhIIII..I II I . I . EFL . , I _ IIIIIIIFII ~ W 0.. . . M _ _ ‘ , # m . . H . _ . , I I _ I , _ . M I _ - r . I a _ o — . I. _ . . _ I . _ . M . I r . . . . I . 1 o .w l.|l I _I .. M _ . .. III? 'IIIIII III | . ,IIII I , A. “II o I J I III. Iqlo 0. 5 I o ,I I O I III II 0 9 9 5 O 5 I . II I O 9 8 . I I I IIIIIFI I 9 9 8 I . 5. O 9 6 5 a 9 ZOEI<>me -30.. from the readings (Appendix) show the usual trend, though not the snooth elliptical curves usually shown in texts (Graphs 1 through 5). Within one hour after pumping began there was rapid drainage close to the sump. At ten feet from the sump the curve had leveled off. Beyond the .ten foot well the curve was practically level even out to the last observation well. At the end of three weeks of Operation similar leveling off was noticeable (Graphs 1 through 5) although the entire water table had declined an average of about one and one half feet. This decline was equal to a little more than 'A" horizon thickness. There was good drainage at five feet from the sump. a condition which changed rapidly between zero and ten feet out, giving curves similar to those plotted by the Soil Conservation Service at Blacksburg. Virginia.26 The removal of sump water had a great effect on the water movement at 5, lO, and 20 feet so that the curve to the last point is steep. lore exactly it might be said that in three. weeks the water in the surrounding area had a chance to percolate downward to replace that removed by the pump. It should not be inferred that water necessarily flowed along the draw down curves to the sump. Instead the effect seemed to be that the ludraulic head pushes the water down by the shortest. most easily accessible route. It is generally stated by experienced investigators that flow is horizon- tal at the water table to the tile or sump. Percolation flow is consid- ered as being as vertical as the soil structure will allow, down to the water table.21 Although the original or natural condition of this soil was one of frequent flooding, high water table, and very poor drainage. about three- 26 Personal Communication from T. I. Edminster, Project Supervisor (Nov- anber, 1948). -32- :0 .CN II..— maqom ”I. 9.63 mnqmg zo_.r<>mmmmo ...._o Minoan. .m radio 0 o o o o o Tlhn lYTllbe IYTII .3 IQTJB IvT..o_.+.m O O diam +1 -33- fourths of its total area has been protected from overflow by the Prairie farms‘ dikes and drainage pumps. Most of the area has also been ditched and about one half of it has been tiled. In spite of such drainage facilities three to five feet of water covered the project site for sev- eral weeks in March and April. 1948, previous to setting up the sump and obserVation well aystan. The Panns‘ drainage ditches were pumped down before the drainage investigation was set up and again on Sunday. May 23rd. .Lfter the second draining, water movement from the land was too slow to fill the drain- age channels more than.ha1f full before the project was terminated on June 3. . The first impression gained from a survey of the draw down curves (Graphs 1 through.5) is that the muck topsoil is drained.mnch.more rapid- ly than the tight clay underneath. There is'a noticeable decrease in the rate of soil water movement after the water table is down to one and a half feet below the surface. The curves are more closely spaced, verti- cally, after May 23rd than up to that time. Drainage at the same time tends to be toward the sump only from a distance of 60 feet. Earlier the curve gradient extended. downward from 132 and then 100 feet from the pump. The total drop in water table from.May 14 to 31 showed some variance as seen from the Graphs 1 through 4, both from well to well in.the same line, as expected, in addition to the variation between comparable wells in each of the four lines (Graph 5). A.greater total decline in water level was found in wells #15 to #21 (except I16 and.§21) than any of the others. This row was in the direction of the main drainage channel and was on land 0.75 feet above #7. 0.15' above #28, and 0.25 feet below #14. Moisture along the #15 to I21 row of wells had further to go to reach a level with the surrounding area. Then, too the ditch south of -u- the project could have some draw down effect at least on the outer two or three wells. What still seems unexplainable is that well {’7 drained to a lower depth than am of the other outer observation wells. It was out in the field toward some buildings northwest of the sump. but much further from the ditches than either {'28 or #14. A part of the drain- age was, of course, due to the natural percolation downward through the soil. Possibly the area around I” had the most pervicus subsoil of the area under investigation. This well {'7 was in a temporary 40 x 75 foot ellipse-shaped puddle of water until seven days after pumping began. The percolation of water in the 330' wells progressed at a faster rate than the observation wells during the last week and a half of pumping. The two 330' ones were not put in until that time. Neal cementing on simi- lar action on other soils finds that the drop of ground water table due to deep percolation and transpiration is sometinws greater then move- ment in tile lines. It is difficult to separate water movements as they 6 all work simultaneously} Observation Wells between Tile Lines In an effort to see how this Clyde soil reacts under long time drainage a line of test wells was also placed at right angles to a five inch tile line which was at least ten years old (l'igure' 15). he sump was dug at the tile line and wells similar to those at the main project were spaced 3, 5, 10, 20. 35. 64. and 100 feet from the tile line. Water level readings were taken once every other day from neyzz through May 31. 1948. In graph #6 the draw down curves plotted from the data show the improvement in drainage with time. The curves have fairly uni- form slopes unlike those of graphs 1 through 5. This shows that drain- age in the tiled land has been more rapid than in the sump and observation well area. The soil profile at these tile line observation wells was some- what different from.that in the untiled land. The tapsoil blended into the substrate. Here the top material was more completely decomposed. ialthough the top one foot closely resembled the muck of the uncultivated land. Nb crops had recently been taken from the area in the vicinity of the sump and test wells. Between the FA“ horizon and the clay was 0.8 feet of loamy clay and then the tight pale yellow clay. The water table at the tile line was four feet below the surface. -35- rigure 13. Sump and Observation Fells Between Tile Lines \ .3 f!) f» ‘37. CONCLUSIONS It is probable that draw down curves of soil water movement can be plotted from the sump and observation well method of investigation. From this study it seems clear that a.properly plotted draw down curve_ will indicate the proper depth and spacing of tile to give the desired lowering of the water table midway between the drain lines. The results of this experiment do not, of course, give a final answer even for Clyde clay loam. .L number of variables have not been contidered. This is rathe er a trial of one method which shows possibilities for mmre complete study. If such a.project were run for a period of years as is being done at Blacknburg, Virginia and at Clemson. South Carolina accurate depth and spacing data for tile and moles should result. .Allowances should be made for a margin of improvement when using the data from this project, since drainage on tiled land and. on land,drained.by the sump method should hmprove with time. It is known that when tile are still working, drainage and therefore tilth will continue to develop even after ten 19 or more years. RECOMMENDATIONS To give more reliable results several improvements also might be applied to the method used here, in addition to increasing the time of operation. The project could be set up at several points in an area needing tile. More complete soil tests should be made. particularly as to permeability and to soil profile at each observation well. The Soil Conservation Service has outlined a standard soil permeability procedure that could beusedf‘:2 More detailed size particle analyses should be made. By plotting a detailed soil profile along with the draw down curves the effect of the profile on drainage characteristics could be -38- studied. Additional equipment that would prove useful includes a water meter to record total flow from the sump and automatic water level record- ing guages. The cost of the latter item makes it impractical for a short time, graduate subject. Thus results from this or similar investigations could be applied only to the project area or to land having comparable soil conditions. Comparisons should. however, be made with studies from other soil types in an effort to reach the final objective -- the equation that can be used for determining depth and spacing of tile and mole drains on em agricultural land. In conclusion it should be emphasised that this has been a study of method as well as an attupt to get closer to solving the problem of proper depth and spacing of tile and; mole drains. .. 39 .. BIBLIOGRAPHY 1. Aronovici, V. S. and Donnan, W. 11., "Soil Permeability as Criterion for Drainage Design," America Geophylical Union Trgsac- M, vol. 2'7, (1946). 2. Baver, L. D., §p_i_1_ thsicg, John Wiley 8: Sons, New York, (1940). 3. Bonsteel, J. A., Egg 91-119. m g; 5.9.119 U. S. Department of Agriculture, Departmental Bulletin #141, (1914). 4. Childs. I. C., "The Water Table, Equipotentials, and Streamlines in Drained Land," 10;; Sciencg, vol. 56, (1943). 5. Childs. E. C., "Water Table in Drained Land," “1 m, vol. 59, (1945). 6. Christiansen, J. D. and Pillsubry, A. 3., "Installing Ground Water Piesometers by Jetting for Drainage Investigations," Aggi- MW. volo 28. (September. 1947). 7. Christiansen, J. 1., ''Ground Water Studies in Relation to Drainage," Agziculturgl Engineeripg, vol. 24, (October, 1943). 8. Diserens, 1., "Means for Determining of Mode of Action of Drainage Systems," International 191131335 9_f_ _S_g_i_1_ W, Transagtionl, Oxford England, vol. 3, pp. 45-65 (1935). 9. Donnan, I. L, "Model Tests for 9. Tile Spacing Formula," §_9_i_; m Society 9; America, Prpceedingp, vol. 11 (1946). 10. Gardner, I., "Influence of Soil Characteristics on Drainage and Irrigation Practices," figi_1_ §cieg9g Society 91 £323.95: Proceedings, vol. 1, (1936). ll. Baswell, J. R., "Principles of Tile Drainage," ricultur 1 11333- eering, vol. 21, (August, 1940). -40.. 12. Israelsen, 0. IL, "Application of Hydrodynamics to Irrigation and Drainage Problems," (abstract), mm En ineeri , vol. 8, (November, 1927). 18. Kirkham, Don, "Preposed Method for Field Measurement of Permeability Below the Water Table," §_o_il_ Science Society 9_f_ Amerigg, Eroceedingg, vol. 10. pp. 58-68 (1940). 14. Lyon, T. L. and Buckman, H. 0., M We. mPrgpertieg 2g _S_9_L, The Macmillan 00., p. 153 (1943). 15. Moon, J. W. and others, Soil grvez p; Sgginaw Coun y, Michigg, Bureau of Chemistry on Soils, U.S.D.A., (1938). 16- Neal. J. H” tam was as Death 2: Lilo 1......eteminod 1: Banana Prgpertiel, Minnesota Station Technical Bulletin 4.101, (1934). 17. Norling, Sven A., "mbsoil as a Factor in Drainage Design," Agi- cultural W, vol. 8, pp. 311-319 (November, 1927). 18. Patty, R. D., "How to Determine the Drainability of Soils," Agg- cultural Epgineering, vol. 25, p. 221 (June, 1944). 19. 1106, H. 3., "Some Soil Changes Resulting from Drainage," .52}; Sciencg Society _o_f_ America, Proceedingg, vol. 4, p. 402 (1939). 20. Schliclc, I. J., flagging m 29913.9. Q1 M 13 L915, Engineering Experiment Station Bulletin #52, (October 16, 1918). 21. Schlick, W. J., m 91 W, Iowa State College Engineer- ing Experiment Station Bulletin '['50, (September 25, 1918). 22. Uhland, R. D., S9}; Eemggbilitz, United States Department of Agri- culture, Soil Conservation Service, Washington, D.C., mimeographed (October 13, 1948). APPENDIX -42- Proposed.Drainage Investigations in Saginaw County Soil Conservation Service and Agricultural Engineering Department Michigan State College COOperating The subjects to be investigated on a cooperative basis are the following: I. MOLE DRAINAGE II. DETERMINE GRADIENT 0! WATER TABLE IN.A GIVEN SOIL AS IT AFFECTS TILE AND MOLE SPACING. At some suitable location in the Big’Prairie or Little Prairie, install sump with electric pump and observation wells radiating out in all directions. Obtain a site having typical soil and near source of electric power. Measure the water level in the observation wells at different distances from sump when the pump is in operation. Prom above measurements, determine preper spacing for tile or moles in that particular soil. DATA -43- SHEET T _ .9 ts _75’4-13i54131511471 5-15 l 5.15 15—12-- 5-15] ._ __ -- - 1 -_-__'" 11908 819084.1940811111908. 111908181994- 91994111 0388:" ~‘. nn 3).. :1 \ ‘I‘m'mtion of ‘siqter in Hells :-- -—- - --.L_-- h- :j '1--.)- --:--.._ 158294__1582.15_J582185 582-60 582.211 ..-._ __.----_-10._ __ 583.14 9.93.4.1 989.121.. MSW-88.1.9931]? 532.83 82.55 1.1. -’ .--:___.._-._--;.0.- .----. 9992.97. 5992.57. 599-31 59-3.. 99-99.3199 99-3-90 992-4.; - __ -1-'-_-;~15 -. 15.99-93..-999194 .999235 889a 492....999-9999-284 182.19. .._. _- i‘ --.---1-.1__.9.'J_-....-.-15.9.9.9961991'5490T989 23.989.901.552 .88..582.921552.88. ____ -. __ .-_..lQQ_-_.-- 583.47 588.481.:583.29 583.08_4__5~8-2‘.A9#QF.5_8189_09+4588.881 ”“13.- _1 15; __1 1993.. 53 _- _ 1583.50 583.44 583.37 583.311583AL .._.--_9..-_-..,_._ .51 _1 -1: "1114,”. 582.9 582.58 582.4 582.59-582.15 _______E_:__ _ __1 1 _19__~_ _ 888.84 583.48 5838081582.,75 1.992115 582.79 582.54. ___-E'- _ 39 _ 88882 999.99.989.95 99319.9-..5-93499 592.94.992.98 __ 13.1...- 11138188 583. _72 599-9994533333 583.02 582,188+1588.9_81588.81j F“ _1_8_M_1__gg_ _ 583 53 588.59 583.31 583.04 5382.92158380 582.85 1.3.3;- 1 « 583-52 583.45 9.89.211.599.991999951999999.992.91- ._---i.‘*_-- --,- 3.13---- ,9992. 39 519.3459 1599-311991'1199159319999919.0 99.39.79... 583. 86 583315 582.9_Q_+r§‘}?.__5_2__1r_5‘83_._44m__ 582 .......~.44 581.94 ‘ 5955.- 99.99.5153_. 59.93 1919.93.95.19}917.321.999.271 99319.21 993-29311999199'599-42 583.04 582.88 582.92 582.18 99319-919993.” 583.73 583.23 583.00 583318 3.9.2. :88 _fiflflflSfifi 583.75 583.40 583.21 583.25 582.98 f583.83 583.21r‘688171383..58 583.31 58314413851984 _518_§_.__8__14}_583.81 583.77 583.59 539.45 553 58.3.1.9 -_’ ,---. _. .1..- 1593.999... 599-99 9931.40 59.31591993219- 583. 8.6 £83.39 583.08 582.77 582.59 5811.73 582.48 999-95 fl9.-71-.-.99.9.99.1_158-2_ 49.91.99-2.o8.-1..._ 582 4.9.10 116392.411. ‘1 583.76 583.63 583.33 593.499.- M58231 -59..3o 88 593 85 (“-39. -_ 168 .1985..H158.§199.583.46 5.8.3.111 582 81 582_,_98 588. 73 . 5. 7.._ . i .---_l-‘.39: 599.919.59.9410155505; E99917... 595:.9-91).59921-9 93.3-39.9. “'45 )1_15"_ _ __1 _j g;- ~~+583. .21T582. 9.9. F5542. 9.5 582. 98 582. 77 1211-821“ ;____ -0---- 393° 814L581)"w j 518 .35 #157935 578.35 :579. 351 579.44 -44- DATA SHEET 37:71:“ ”-7293. 36:5: 6:561:21 542: 5:93 :5: 6428?" ~—-——-~--—1» ”-7139 _..12: 0017113104 09 9:4.5458811112;:p_cg17_1121_