UNIVERSITY OF CALIFORNIA PUBLICATIONS COLLEGE OF AGRICULTURE AGRICULTURAL EXPERIMENT STATION BERKELEY, CALIFORNIA PUMPING FOR DRAINAGE IN THE SAN JOAQUIN VALLEY, CALIFORNIA BY WALTER W. WEIR BULLETIN No. 382 January, 1925 UNIVERSITY OF CALIFORNIA PRINTING OFFICE BERKELEY, CALIFORNIA 1925 PUMPING FOR DRAINAGE IN THE SAN JOAQUIN VALLEY, CALIFORNIA By WALTER W. WEIR INTRODUCTION The first comprehensive system of drainage by pumping from deep wells was installed in the Salt River Valley of Arizona in 1919. This method of drainage promises, under favorable conditions, to be more successful than any previously undertaken and the apparent ease with which these pumps lowered the water table has led western engineers to take an unusual interest in it. The fundamental principles involved in lowering a water table by pumping were not new, but until this time they had not been demonstrated as applicable to the drainage of waterlogged and alkali lands of the irrigated regions. As the result of notable success in pumping in Arizona, there has been considerable activity along similar lines in the San Joaquin Valley of California. The lack of published information on this subject and the varying conditions encountered in the different sections has led to considerable experimental work both by the Agri- cultural Experiment Station and the several irrigation districts in California. In this paper an attempt has been made to gather together and correlate some of the important data which have been accumulated by these agencies. I am indebted to the engineers of the San Joaquin Valley irrigation districts for access to their files for much of the detail of this report; to the pump manufacturers for diagrams, cuts and other information on pumps; and to Mr. J. C. Marr, Drainage Engineer, of the U. S. Department of Agriculture, for recent infor- mation on the Salt River Valley Project. The drawings were made by Mr. Stanley W. Cosby, Research Associate in Soil Technology. HISTORY OF DRAINAGE OF IRRIGATED LANDS It must be recognized that although in recent years there has been considerable progress, the drainage of irrigated lands has not been completely successful. In the early attempts, failures resulted from inexperience with the conditions which are peculiar to the waterlog- 4 UNIVERSITY OF CALIFORNIA EXPERIMENT STATION ging of soils of arid regions. The presence of alkali and its concen- tration at or near the surface as the result of a high water table brought about by irrigation practices adds considerably to the com- plexity of the drainage problem in an arid region. It was only after an extended period of experimentation that the principles now recognized as fundamental were utilized by engineers and that the drainage of irrigated lands became fairly successful. There are still a great many problems connected with the removal of alkali from the root zone of plants which have not been satisfac- torily solved.* A better understanding of the fact that the water table must be maintained at sufficient depth below the surface to prevent further deposition of alkali by capillary rise and evaporation within the root zone of plants has led to greater success in more recent drainage enter- prises. However, the utilization of this idea is often difficult, first, because of insufficient knowledge as to the proper minimum depth for any given soil, and second because of the necessity of accomplishing the desired result at a cost commensurate with the value of the im- proved land. The high " first cost" of efficient tile or open ditch drainage for alkali lands together with the frequent necessity for additional ex- penditures in removing alkali from the surface of the soil after the lowering of the water table has resulted in rather meager attempts at drainage for this type of land in California. Where black alkali (sodium carbonate) exists in any considerable quantity the difficulty of removing it at a cost within reason further complicates the problem. As a matter of fact, very little land in which black alkali has accumulated in quantities has been fulty reclaimed in California. It is better, though difficult in practice, to prevent land from becoming waterlogged than to reclaim it after it has reached that condition. The main difficulty in this respect lies in the refusal of those concerned to admit that all lands irrigated from gravity systems are potentially subject to drainage difficulties. With these facts in mind, drainage engineers must attempt to provide cures for the trouble at the same time that they recommend prevent ive measures. * K.llrv, W. P., The Present Status of Alkali. Agr. Exp. Sta. Cir. 219: 1-10, ] 020. BULLETIN 382] PUMPING FOR DRAINAGE IN THE SAN JOAQUIN VALLEY 5 PREVIOUS DRAINAGE IN THE SAN JOAQUIN VALLEY In each of the irrigation districts in the San Joaquin Valley where drainage pumps are installed at the present time (1924), a con- siderable amount of drainage work of the open ditch type had been installed prior to the pumping. Before the organization of the Merced Irrigation District, two drainage districts were organized in a portion of the same territory, one at Livingston and the other at Atwater. Both of these had com- pleted a rather extensive system of open drains discharging into the Fig. 1. — A typical drainage pump installation in the San Joaquin Valley. Note the metal pump house, short discharge line and concrete lined irrigation canal. San Joaquin River. The rather rough topography of these areas necessitated deep cuts through the ridges in order to drain the depres- sions to a satisfactory depth. Although aiding materially in the removal of surplus water from these areas, more was needed before entirely satisfactory conditions were reached. The Merced Irrigation District has now absorbed these drainage districts and taken over the drains as a part of its outlet system. Similar but more extensive drainage works have been completed in the Turlock Irrigation District. More than 75 miles of drainage ditches and tile lines have been constructed. These drains which were constructed according to the most advanced engineering ideas and average from 8 to 12 feet in depth, normally discharge about 100 cubic feet per second or approximately l 1 /^ cubic feet per second per mile of drain. This system cost $300,000. 6 UNIVERSITY OF CALIFORNIA EXPERIMENT STATION Aside from these drains which also act as spillways for the irri- gation ditches, the district has within the past few years prevented much seepage from the larger irrigation laterals by lining them with concrete. In 1923, there were 45 miles of concrete-lined canals. Figure 1 shows such a canal. Drains have also been constructed in both the Modesto and South San Joaquin Irrigation Districts. Those in the South San Joaquin have been less beneficial than those in the other districts, largely because the outlet was not such as would permit a satisfactory depth of drainage. Furthermore, a shortage of late season irrigation water led to the impression among the water users that deep drainage was undesirable. Many farms in the various districts are not levelled or prepared for irrigation and have depended upon a high water table to provide moisture for their crops. This has been a more common practice in the vicinity of Manteca than farther south in the valley. The com- parative absence of drainage of any type south of the Merced district has undoubtedly been due to the lack of any organization which could satisfactorily undertake work of a nature which requires strict cooper- ation among the land owners both for planning and financing. EFFECTS OF PUMPING FOR IRRIGATION It has long been recognized that in irrigated areas where the water supply is obtained by pumping from underground sources within the area irrigated the drainage problem is reduced to a minimum. Con- tinued pumping in such areas or a material increase in the number of pumps has frequently resulted in a water table receding to such an extent that pumps have been lowered and restrictions placed upon their operation. This condition is well brought out in certain areas along the Kaweah River delta in Tulare County, in the Santa Clara Valley and in several areas in the southern coast counties of the state. A particularly interesting example, because at one time drainage was a problem, has occurred near Chino in San Bernardino County. Prior to 1914, a large acreage operated by the American Beet Sugar Company near Chino was wet and strongly alkaline. In places during the winter months water stood on the land and a few permanent cat- tail swamps provided duck hunting sites. Because of this condition, beet planting was delayed until late in the spring with the result that inferior crops were; produced. The area was later tile drained, the tile having an average depth of six feet and an average spacing of 660 feet. As a result of these BULLETIN 382] PUMPING FOR DRAINAGE IN THE SAN JOAQUIN VALLEY 7 drains the water table was lowered, the ponds disappeared and crop returns enhanced. During the past few years, this land has become more thickly settled and more intensively farmed, the water for irri- gation being pumped from wells on the property. These wells have so lowered the water table that the tile now lies above the level of the ground water and has ceased to flow. Drainage is no longer considered necessary and the pumps raise the water from about 30 feet below the surface. On portions of the Rancho La Sierra near Arlington, Riverside County, where irrigation is supplied from paimps located in a low, poorly drained depression, the water table is lowered to about 6 feet below the surface while the pumps are in operation, but during the non-irrigating season, it again rises to the surface. At Fresno, drainage conditions have been incidentally improved by pumps. The city of Fresno, which is located in the heart of a large irrigated area, was in 1914 seriously affected by poor drainage and high water table. At that time the average depth to water under the main business portion of the city was about 6 feet. Most of the larger buildings having basements and the subway under the Southern Pacific tracks were provided with pumps in order to keep them dry. Twenty-seven pumping plants supplying domestic water are now in operation by the Fresno City Water Corporation. These cover an area of about 8000 acres. The demand for water does not require the continuous operation of all of these pumps, but during brief periods of high consumption when they are all in concurrent operation the total water pumped approximates 13,000,000 gallons per day. The total water pumped during 1923 was 19,500 acre feet. This has lowered the water table about 13 feet, making a total depth to water of about 20 feet. During the construction of a large building in Fresno in 1923 excavations to a depth of 23 feet below the street level did not encounter water. APPLICATION OF PUMPING TO DRAINAGE The principle of lowering the water table by pumping was not applied, with that object in view, to poorly drained alkali areas of irrigated land until 1919, when the present plans in the Salt River Valley were undertaken. In fact, it was only after exhaustive studies carried on in previous years had shown that the water so pumped was needed and could be economically used for irrigation that the Board of Governors of the Salt River Vallcv Water Users Association was authorized to proceed with the work. The results so far have proved 8 UNIVERSITY OF CALIFORNIA EXPERIMENT STATION beyond question the soundness of the pumping idea and farmers are more generally coming to realize that it is just as economically feasible, where conditions are not unfavorable, to pump water for drainage as it is to pump water for irrigation. The almost immediate lowering of the water table following the operation of a limited number of pumps in the Salt River Valley led to unusual and wide- spread interest in this method. The success of pumping was doubly striking in this valley because in the nearby Tempe Drainage District, where conditions were similar, open ditch drains had not met with marked success after several years of operation. FUNDAMENTAL REQUIREMENTS It must not be assumed that at the present time (1924) the best procedure regarding location of wells, type of pump and many other details have been fully worked out even for any one locality, and it is doubtful if the details for a widely applicable design can ever be worked out. It would appear, however, from the data which have been collected from various sources that there is justification in making a few broad and general statements regarding the fundamental requirements to be met in order to be successful in drainage by pumping. 1. There must be a direct connection between the ground water table near the surface and the deeper lying pervious water bearing strata from which the water is pumped. In other words, the water table as first encountered must not be artificial and have a layer of dry or impervious material between it and the normal ground water. 2. The underlying water bearing strata must be porous enough to give up their water freely under pumping. 3. There must be sufficient water pumped from the lower strata to cause that which is near the surface to move downward by gravity to replace that which has been pumped. The general effect on the water table should be the same whether the water is pumped from the bottom of the reservoir or taken from near the top by means of tile or open drains. Drainage is accom- plished by lowering the water table below the point where it can either directly or indirectly cause damage to growing plants. GROWTH OF THE IDEA IN SAN JOAQUIN VALLEY In California and especially in the San Joaquin Valley, pumping for drainage has increased even more rapidly than in Arizona. In the spring of 1924, there were ready for operation, for drainage pur- BULLETIN 382] PUMPING FOR DRAINAGE IN THE SAN JOAQUIN VALLEY 9 poses, approximately 30 pumping plants in the South San Joaquin Irrigation District, 32 in Modesto, 45 in Turlock, and 30 in Merced, and several in the Fresno Irrigation District. In addition to these, there are 9 pumps on the Fresno Sewer Farm, whose primary object is sewage disposal but which are similar in design to the drainage pumps of the valley and which have had a material effect on drainage conditions in that vicinity. The several districts which have installed drainage pumps have not followed the same plan in all details and two somewhat divergent ideas are represented. In order to bring out these differences, each will be discussed and the facts given which Fig. 2. — Sandy land at Turlock cultivated for the first time in several years. The depression shown in the foreground was rapidly drained of standing water by the operation of a drainage pump located nearly a half mile distant. might justify each design. The plan adopted by the Merced Irriga- tion District is typical of the ideas of those who advocate the larger type of well and pump and a general lowering of the water table over the area, while the Turlock Irrigation District has followed out the idea that smaller pumps which will have only a local effect on the water table will be the more satisfactory. SOILS AND TOPOGRAPHY The character of the poorly drained soils on the east side of the San Joaquin Valley is apparently of such a nature that this type of drainage will prove satisfactory. Throughout the northern end of the valley in the South San Joaquin, Modesto, Turlock and Merced Dis- tricts, the principal soils belong to the Oakley, Fresno and Madera 10 UNIVERSITY OF CALIFORNIA EXPERIMENT STATION series, and are usually sandy in nature. Practically all of the wind deposited Oakley series and much of the soils of other series have a more or less uneven topography, while the hardpan layer of the Madera, and the intermittent and fragmentary lenses of hardpan at varying depths in the Fresno series, have not been found to be a serious interference with the vertical movement of water. Figure 2 gives a general idea of the local topography near Turlock and the rapidity with which these sandy soils can be drained. Occasionally, in the Madera and Fresno soils, saucer-shaped pockets underlaid by hardpan have held water above the surrounding water table; but this condition is unusual. In a general way, the topography is more even and flat toward the southern part of the valley, and the water table is more nearly parallel to the ground surface. Extensive ground water studies covering a number of years in the Turlock District show that the water table is approximately parallel to the general ground surface, but not necessarily parallel to the local topography. In this area the water table has a gradual southwest slope toward the San Joaquin River and parallel to the Tuolumne and Merced Rivers. The more or less uneven local topography of this region results in local areas of poor drainage or "pot holes" surrounded by areas where the water table is at a safe distance below the surface. The ponds thus formed are not so much the result of water having flowed into depressions from surrounding higher areas as of the fact that the bottoms of these depressions are in actual elevation lower than the ground water table. Differences of elevation of 15 to 20 feet within a quarter of a mile are not uncommon in some of the rougher areas. In the Merced district, the topography is generally a little more even than at Turlock, while in Fresno, Kings and Kern counties, the waterlogged areas are generally flat and the water table at a more uniform distance from the surface over larger areas. The slope of the water table and general trend of water movement at Merced is similar to that at Turlock, namely toward the San Joaquin River. The logs of the drainage wells in the valley show considerable variation in profile, but it has been found that these variations are of minor importance as regards the suitability of the well for drainage purposes. Similar variations were noted in the wells of the Salt River Valley. In the latter place considerable depths of a hardpan known locally as "caliche," did not interfere with drainage. In many cases, however, the log of the wells was a deter- mining factot in deciding upon the proper depth to which it should be drilled. An attempt was made to continue the well to a porous water bearing stratum of considerable Ihiekness. BULLETIN 382] PUMPING FOR DRAINAGE IN THE SAN JOAQUIN VALLEY 11 Figure 3 shows a theoretical cross-section of the area in which wells were drilled on the Salt River Valley project, while figures 4 and 5 show the plotted well logs on two areas in the San Joaquin Valley. The particular significance of these logs is that they show no uniformity of profile even for wells within short distances of each other. Wells 1 and 1a on the Fresno Sewer Farm are about 250 feet apart. S; C/?OSd-S£Cr/OA/ THROUGH: ^ rv£ A/frr /?/i/f# apea.\ \ s r 5! •-er.cf '03 340 C/POJJ- SECT/ON THPOUtfH THE .- C/ay. ^So/9 C/ay. Fig. 4. — Logs of four Fresno Sewer Farm wells. These logs indicate a soil profile having considerable variation within short distances. Such variations, how- ever, have not had a material effect upon the efficiency of the well for drainage purposes. 14 UNIVERSITY OF CALIFORNIA EXPERIMENT STATION over the disposal area prevented such filtering. Pumping from wells not only provided the necessary filter beds by lowering the water table, but delivered relatively pure water to the outlet drains. When pumps are provided for the drainage of wet areas south of Merced, it will be necessary to effect a general lowering of the water table over considerable areas because of the flat topography and the large areas under which the water stands at a nearly uniform depth. o — 5 6ZZS : .? m :oy*ro, %i?: : :?:'a}~ji V;d : . : ob? 46- .O;0 .''•0.0. o?..'?/o:ov° y rf ••*:••..•.¥ J-Q 6 - "• -0 IIIIIIIIIIIIIIIIIIIIIIIIIIJIIIIIIII L£G£ND^ 5anc/. 1 P&ckec/ 3an \ + f>A j«> » -*/1c> !«/«/ r/i/ct i/e*t/c *"» i *3> <^^" /or/'oi ■* f"o . Ot/mjo, s>p 1 J 1 6 \ 7 $ K^^ v° e / Y >••" ■—•—... ~--— / ^ei 'sv c 't/r-re \> /o / &**f ■for- c »/yO»r«- V /0 I ^ // I* I* v* 23 AFT£-/? S£COMG WA& fttnta r gg< eerf* S*//. /ke /7/im J/>#. A?/»r t/M. id*. sft/A £*/? 0cr Mm Zfec. Fig. 7. — A theoretical consideration of the effect of pumping on the water table. When a general lowering of the water table is desired, it is only by some similar consideration that drainage systems can be designed to meet a predeter- mined ideal. In this design, it was considered necessarj' to give the pumps sufficient capacity to overcome the lateral movement of water into this comparatively small area from a much larger area where the water table never falls below 7 or 8 feet from the surface. The total annual pumping from this area was estimated at 17,000 acre feet or over three acre feet per acre of the area actually drained. 18 UNIVERSITY OF CALIFORNIA EXPERIMENT STATION On the Fresno Sewer Farm on an area of 500 acres, nine pumping plants maintain a water table five feet below the surface, although sewage water is added at the rate of 1.5 acre feet per acre per month or 18 acre feet per acre per year. The nine pumping plants operating continuously discharge 19,700 acre feet per year or about twice the amount applied. The total water pumped by the Fresno city water system was 19,500 acre feet or about two and one-half acre feet per acre. As the amount of water pumped for domestic purposes varies with the demand there is considerable difference between maximum and minimum requirements. If the twenty-seven pumps were con- tinuously operated to their full capacity, they would have more effect on the water table than is shown here. In drainage by pumping, as with other systems, the amount of water which it is necessary to remove should be determined in spite of the fact that insufficient capacity can be more easily remedied with pumps than with ditches. When the amount of necessary discharge has been determined, the number of pumps required can be decided upon according to the supply that can be secured from each well, the draw-down, and comparative cost of operating pumps of different sizes. LOCATION OF WELLS Different methods of locating the wells have been followed in the various districts, as will be noted in studying the maps, figures 8, 9 and 10. In the Modesto and Turlock areas, for instance, where only local lowering of the water table was desired the wells were located in or near the local wet areas. The wells are scattered irregularly over the district but their accessibility from the highways and the convenience of an outlet for the pumped water were taken into con- sideration. It is advisable, unless some other consideration is more important, to locate a well so that it can be conveniently reached by truck or repair outfit and as near as possible to the point of discharge. (See figure 1.) In locating the wells of the Merced district one other condition was considered, namely, that they be arranged in approximately parallel lines across the line of flow of the underground waters. The map, figure 10, shows that there are roughly three parallel lines of wells, one jnsl south of the Santa Fe railroad, one south of the South- ern Pacific railroad, and a third still farther to the southwest. The lines are from two to three miles apart and the wells average some- BULLETIN 382] PUMPING FOR DRAINAGE IN THE SAN JOAQUIN VALLEY 19 V) CO P8E yA P9E W£>£Sro PiOt PllE Pl2E zzr}^ — ^> i r ' II A^ -*rO «u ^T& V.J r~" R/y/ or* '/«/v j fc /y< '(S/v s of ' i Wl c ?*>£ s I ! V J «r^ ***L ■ 5 1 ms _.j i \ Vv • ^ ■ — \* % ^D£/VAA L i t • V: • \ bi" o • • • » •s ^ • V • ( 7^ >co • \c» < ' • >^ -V i l c * CK ■ • Vi ^ ^^ * ■ • >j " 1 "\T- • 4 "1 SCOTCH, — .^ r: >* • • \; b ^! ( >P« /v(J ^\ » *£*? Af/A/(/7-£. /6 /700 /e /SOO 20 Fig. 11. — Typical test well chart. For a drainage well, a minimum drawdown is essential. For this particular well in order to get a drawdown of 30 feet, it will be necessary to pump about 1450 gallons per minute. Trouble of this nature can undoubtedly be overcome by installing a fully cased or closed bottom well instead of an open bottom well. This has been done on the Fresno Sewer Farm and is the usual prac- tice for irrigation wells in most parts of the state. With a fully cased well, it is necessary to penetrate far enough into the water bearing strata for the required flow to be obtained through the perforations in the casing. On the Fresno Sewer Farm, the wells are cased their entire depth, the lower 100 feet being perforated. In many wells of similar type the casings are perforated or slit after being placed. In such instances perforations are made only at points where the well logs show porous strata. 24 UNIVERSITY OF CALIFORNIA EXPERIMENT STATION DEVELOPMENT AND TESTING OF WELL As soon as a well is drilled, it should be developed and tested in order first, to see if it is satisfactory, and second, to determine the most economical pump and motor with which to equip it. Although the testing of a well is frequently neglected, especially by persons requiring only one or two wells, the practice has been quite generally followed by those interested in drainage pumps. The testing consists of temporarily installing a pump whose maximum capacity is in excess of the anticipated capacity of the well SOO 6 700 POO /O //OO /2 /300 14 /SOO /6 /7"00 /B /9O0 ZO MLLOHJ PS/? rt/#C/T£ Fig. 12. — Typical pump efficiency test curves. When pump makers have the exact requirements to be met, they can design a pump to fulfil them. The above chart shows that this particular pump has an efficiency of 72 per cent when delivering 1450 gallons per minute against a head of 35 feet and 18 h.p. of energy is required. and of operating this at varying speeds in order to determine the maximum capacity of the well, the relation of power consumed to the different discharges, and the relation between drawdown in the well and the effect upon the surrounding ground water table, the power and the discharge. Figure 11 shows a typical test chart. With these data at hand, it is possible to specify the quantity of water to be pumped and head against which it is to be pumped. From these data, the manufacturer can design a pump with a motor of such size and speed that the plant will have the highest efficiency. (See fig. 12.) Incidentally to the testing of the well, it is developed. This estab- lishes the normal flow and removes considerable sand, materially lessening the danger of "sanding up" after final installation. BULLETIN 382] PUMPING FOR DRAINAGE IN THE SAN JOAQUIN VALLEY 25 Occasionally upon being tested, a well proves to be unsatisfactory. It can then be either deepened or abandoned as the conditions warrant. In the Salt River area, drainage wells were abandoned and new ones drilled when upon test they did not deliver 1000 gallons per minute with a drawdown of 40 feet. In order to be most successful as a drainage well its maximum capacity should be reached at the required drawdown. It is easily possible to obtain a well having a maximum drawdown which would have almost no effect upon the water table because the quantity of water removed was so small. On the other hand, the quantity of water pumped might be large, but the drawdown so small that the water table would be likewise unaffected. CAPACITY OF PUMPING UNIT As has already been indicated, the amount of discharge for which any given well or series of wells should be designed depends largely upon the results desired. Obviously, with an equal quantity of water to be drawn from, the greater the quantity pumped, the greater will be the effect upon the surrounding water table. It has been this con- sideration more than any other which has determined the size and capacity of the pumps in the various areas drained by the pumping method. In the Salt River Valley, for instance, it was determined by experiment that wells having a minimum capacity of about 1000 gallons per minute and a drawdown of 20 to 40 feet had a certain effect on the water table. Other wells having a greater capacity for a similar drawdown had a similar effect. As it was desirable to create a drawdown of about 40 feet, pumps were designed to meet this requirement. The average drainage pump in this area discharges about 1400 gallons per minute and materially affects the water table for a radius of about half a mile. It should be repeated here that the Salt River Valley plants are expected to function collectively and not as individual pumps. There is no ideal unit suitable for all cases. As a matter of fact a recently installed pump in the Salt River Valley has a discharge of 5600 gallons per minute, while pumping against a total head of 32% feet. In the Merced district, the smallest plant discharges about 1000 gallons per minute and the largest between 1800 and 2000 gallons per minute. Some of the wells at Merced developed under test a capacity of more than 2500 gallons per minute with a drawdown not incon- 26 UNIVERSITY OF CALIFORNIA EXPERIMENT STATION sistent with economical pumping. One pump is being operated under a maximum drawdown of about 70 feet. The average delivery, how- ever, is about 1400 gallons per minute with a drawdown of 40 feet, and the effect on the water table is noticeable almost a mile away. Here again, however, it is not possible to judge properly the effect of an individual • pump by the combined effect of a number of plants working in unison. Z60 2SO Z+O ->v£sr &9ST- ^^> *???/&> www 777777 &777T. 77777 G/?oi>*>o iSi/trr. l H v bringing power to the point of consumption. BULLETIN 382] PUMPING FOR DRAINAGE IN THE SAN JOAQUIN VALLEY 35 ten to twenty per cent that of tile or open drains. Upkeep and depre- ciation will run approximately the same percentage of first cost, 5 per cent to 7 per cent per annum, in each case. In other words, at the end of a twenty year period amounts about equal to the original cost will have been spent in replacements and maintenance. In the case of pumping, however, there is the operation cost which does not occur with other types of drainage and this may run from 30 per cent to 50 per cent of the first cost per annum. It will be seen therefore that at the end of a certain period which can be reasonably well deter- mined for any particular system, pumping will have cost as much per acre as other types of drainage. This, however, does not consider the difference in interest charges on a large and comparatively small initial investment. Operation and maintenance charges should in either case be paid from current funds and interest on such money should not be charged against the cost of operation. USE OF DRAINAGE WATER When drainage is secured by tile or open drains the water is usually not considered valuable and is allowed to waste into the rivers. With pumping, however, there is a tendency to put as much of this water to beneficial use as possible. This is undoubtedly due to two factors, first, that it is usually in a more convenient location and more readily diverted to land needing irrigation, and second, that putting the water to beneficial use offers a means of reducing the ever present operation cost. In many cases it is possible that all drainage water pumped during the irrigating season can be disposed of at a cost at least equal to the pumping charge. In this way, drainage can be secured at the cost of off season pumping plus the construction charge. One of the defects in the present gravity systems of irrigation where storage is not available, is the possibility of a shortage of late season water. Drainage water pumped after the exhaustion of the normal irrigation suppty then becomes of considerable economic im- portance. There is still much land in the San Joaquin Valley for which there is no available gravity supply of irrigation water. Pumped drainage water can therefore be used to extend the irrigated area and the cost can be equalized between drainage and irrigation to the advantage of both. Figure 1 shows drainage water being pumped directly into an irrigation canal. A prominent engineer in the San Joaquin Valley 36 UNIVERSITY OF CALIFORNIA EXPERIMENT STATION is credited with, the statement that "Irrigation in the San Joaquin Valley is dependent for its ultimate solution upon the maximum use of all underground waters and that no irrigation district has the right to allow a water table to rise to a height greater than that from which it can economically pump. ' ' Such a use of underground waters would, of course, automatically solve the drainage problem. According to another engineer storage water will eventually become so valuable that it can be used as a source of gravity irrigation supply only after it has been used to the limit for power and that the power so developed can be used for pumping water, which will have the two-fold purpose of supplying irrigation and preventing a rise in the water table. Pumped drainage water is as suitable for irrigation as any pumped water because it is taken from the same strata and under the same conditions as pumped irrigation water. It has the advantage in that the ground water is being constantly replenished by seepage and irri- gation losses from gravity supplies and that it is nearly always diluted by gravity water in the canals before delivery to the irrigator. ORGANIZATION Pumping for drainage, or in fact drainage of irrigated lands in general, is a problem the solution of which can seldom be undertaken by the individual farmer. Drainage is a matter which not only inter- ests the individual who is unfortunate enough to have wet or alkali land, but the whole communitv in which he lives and therefore the community should undertake the solution of the problem. The indi- vidual farmer can seldom solve his drainage problem without assist- ance. The necessary construction is costly, particularly with regard to an outlet and often the drainage of an individual farm, unless it is very large, will not justify the expense. With few exceptions, successful drainage in irrigated areas has been confined to that undertaken by organized drainage districts or irrigation districts. The California laws make specific provision for irrigation districts doing drainage work. Throughout the San Joaquin Valley the irrigation district, where such exists, is the logical organ- ization for the administration of drainage. The number of new irri- gation districts which have been formed within the last few years and are still in the process of formation will be a great impetus to drainage activities whether by pumping or otherwise. Bulletin 382] PUMPING FOR DRAINAGE IN THE SAN JOAQUIN VALLEY 37 ADVANTAGES AND DISADVANTAGES OF PUMPING One of the advantages of pumping for drainage is the flexibility of the system. Additional pumps can be installed when and where they are most needed without any loss in efficiency in those already installed, while individual pumps may be discontinued or removed without great loss in investment. The greatest advantage, however, is in the ability of pumping plants to lower the water table to greater depths than can usually be done economically by any other method. On the other hand, the flexibility of drainage pumps, that is, the ease with which they can be stopped or operated, may, when misused, militate against the suc- cess of drainage by this method. False economy may cause the pumps to be stopped before drainage is fully accomplished. It has already been shown that in certain localities where the water table has been lowered rapidly trees and other crops have suffered. This occurs where plants have developed a shallow rooting system in order to adjust themselves to high water table conditions. A sudden lowering of the water table results in the plants suffering from drought before they can adjust themselves to the altered condition or be supplied with surface applications of water. A careful, gradual lowering of the water table by drainage will permit the plants to develop deep root systems and then if they are properly irrigated from the surface it is immaterial how far below the roots the water table is lowered. The drainage pumps in the San Joaquin Valley, as at present installed and as they are likely to be installed under the present conditions, will unquestionably aid materially in reducing the drain- age problem of the valley and this type of drainage bids fair to supersede open ditch and tile drains wherever soil, geological and topographical conditions made it workable. The economic situation, however, may retard a full realization of a condition of adequate drainage throughout the valley until there is such an urgent need for more water for irrigation that pumping plants will be established in sufficient numbers to produce it, and drainage will then be only incidental. CONCLUSIONS 1. Drainage of irrigated lands by means of pumping from deep wells, first undertaken in a comprehensive way in the Salt River Valley of Arizona in 1918, met with such success in that region that it is being undertaken rather extensively in the San Joaquin Valley of California. 38 UNIVERSITY OF CALIFORNIA EXPERIMENT STATION 2. The South San Joaquin, Modesto, Turlock and Merced Irriga- tion Districts, have each installed a series of deep well pumps which will be operated for the primary purpose of lowering the water table and improving drainage conditions. 3. No standard design has yet been adopted, nor is it necessarily desirable that there should be, as local conditions of soil, geology and related subjects may vary sufficiently to preclude a general plan being applicable. 4. Success may be expected where the surface water is directly connected with and a part of the normal ground water and pervious strata can be found from which relatively large quantities of water can be pumped. The only other essential is the installation of pumps of large enough capacity to deplete the underground supply sufficiently to cause the water table to recede to the desired level. 5. The feasibility of this method of lowering the water table has been repeatedly shown in those parts of California where the irriga- tion supply is obtained by pumping. 6. Two general plans are found in the San Joaquin Valley, one exemplified by the installations at Merced and the other by those at Turlock. The Merced installations are generally of a larger type with deeper and larger wells and greater drawdown than those at Turlock. At Merced the desire is to lower the general water table, while at Turlock a local lowering only is desired. 7. The cost of drainage by this method compares favorably with that of any other method yet tried, though the cost of operation is more. Over a period of years the cost of the two methods may not be very different. 8. More effective drainage may be accomplished by pumping than by tile or open drains because of the flexibility of the pumping system, and the greater depth to which it is economically feasible to 1 lower the water table. 9. The pumped water is readily available for irrigation and com- pensates in part for the cost of drainage by this method. To utilize water from tile or open drains entails additional cost and is not common. 10. The ultimate and complete solution of the drainage problem in the San Joaquin Valley probably lies in the use of all the under- ground waters which can be economically pumped for irrigation and used on areas not now irrigated. STATION PUBLICATIONS AVAILABLE FOR FREE DISTRIBUTION BULLETINS No. No. 253. Irrigation and Soil Conditions in the 346. Sierra Nevada Foothills, California. 347. 261. Melaxnma of the Walnut, "Juglans regia." 348. 262. Citrus Diseases of Florida and Cuba 349. Compared with Those of California. 263. Size Grades for Ripe Olives. 350. 268. Growing and Grafting Olive Seedlings. 351. 273. Preliminary Report on Kearney Vine- 352. yard Experimental Drain. 275. The Cultivation of Belladonna in Cali- 353. fornia. 354. 276. The Pomegranate. 357. 277. Sudan Grass 278. Grain Sorghums. 279. Irrigation of Rice in California. 358. 280. Irrigation of Alfalfa in the Sacramento Valley. 359. 283. The Olive Insects of California. 360. 285. The Milk Goat in California. 286. Commercial Fertilizers. 361. 287. Vinegar from Waste Fruits. 294. Bean Culture in California. 362. 298. Seedless Raisin Grapes. 363. 304. A Study of the Effects of Freezes on Citrus in California. 364. 310. Plum Pollination. 312. Mariout Barley. 366. 313. Pruning Young Deciduous Fruit Trees. 317. Selections of Stocks in Citrus Propa- 367. gation. 319. Caprifigs and Caprification. 368. 321. Commercial Production of Grape Syrup. 324. Storage of Perishable Fruit at Freezing 369. Temperatures. 370. 325. Rice Irrigation Measurements and Ex- 371. periments in Sacramento Valley, 1914-1919. 372. 328. Prune Growing in California. 331. Phylloxera-Resistant Stocks. 373. 334. Preliminary Volume Tables for Second- 374. Growth Redwood. 335. Cocoanut Meal as a Feed for Dairy Cows and Other Livestock. 375. 336. The Preparation of Nicotine Dust as an Insecticide. 376. 339. The Relative Cost of Making Logs from Small and Large Timber. 377. 340. Control of the Pocket Gopher in Cali- 378. fornia. 343. Cheese Pests and Their Control. 344. Cold Storage as an Aid to the Market- ing of Plums. Almond Pollination. The Control of Red Spiders in Decidu- ous Orchards. Pruning Young Olive Trees. A Study of Sidedraft and Tractor Hitches. Agriculture in Cut-over Redwood Lands. California State Dairy Cow Competition. Further Experiments in Plum Pollina- tion. Bovine Infectious Abortion. Results of Rice Experiments in 1922. A Self-mixing Dusting Machine for Applying Dry Insecticides and Fungicides. Black Measles, Water Berries, and Related Vine Troubles. Fruit Beverage Investigations. Gum Diseases of Citrus Trees in Cali- fornia. Preliminary Yield Tables for Second Growth Redwood. Dust and the Tractor Engine. The Pruning of Citrus Trees in Cali- fornia. Fungicidal Dusts for the Control of Bunt. Turkish Tobacco Culture, Curing and Marketing. Methods of Harvesting and Irrigation in Relation to Mouldy Walnuts. Bacterial Decomposition of Olives dur- ing Pickling. Comparison of Woods for Butter Boxes. Browning of Yellow Newtown Apples. The Relative Cost of Yarding Small and Large Timber. The Cost of Producing Market Milk and Butterfat on 246 California Dairies. Pear Pollination. A Survey of Orchard Practices in the Citrus Industry of Southern Cali- fornia. Results of Rice Experiments at Cor- tena, 1923. Sun-Drying and Dehydration of Wal- nuts. The Cold Storage of Pears. Studies on the Nutritional Disease of Poultry Caused by Vitamin A De- ficiency. CIRCULARS No. No. 70. Observations on the Status of Corn 155. Growing in California. 157. 87. Alfalfa. 160. 111. The Use of Lime and Gypsum on Cali- 161. fornia Soils. 164. 113. Correspondence Courses in Agriculture. 165. 117. The Selection and Cost of a Small Pumping Plant. 166. 127. House Fumigation. 167. 129. The Control of Citrus Insects. 170. 136. Melilotus indica as a Green-Manure Crop for California. 172. 144. Oidium or Powdery Mildew of the Vine. 173. 151. Feeding and Management of Hogs. 152. Some Observations on the Bulk Hand- 174. ling of Grain in California. 178. 154. Irrigation Practice in Growing Small 179. Fruit in California. Bovine Tuberculosis. Control of the Pear Scab. Lettuce Growing in California. Potatoes in California. Small Fruit Culture in California. Fundamentals of Sugar Beet Culture under California Conditions. The County Farm Bureau. Feeding Stuffs of Minor Importance. Fertilizing California Soils for the 1918 Crop. Wheat Culture. The Construction of the Wood-Hoop Silo. Farm Drainage Methods. The Packing of Apples in California. Factors of Importance in Producing Milk of Low Bacterial Count. CIRCULARS — (Continued) No. 184. 190. 193. 198. 199. 202. 203. 205. 208. 209. 210. 212. 214. 215. 217. 219. 220. 228. 230. 231. 232. 233. 234. 235. 236. 237. 238. 239. 240. 241. 242. 243. 244. A Flock of Sheep on the Farm. Agriculture Clubs in California. A Study of Farm Labor in California. Syrup from Sweet Sorghum. Onion Growing in California. County Organizations for Rural Fire Control. Peat as a Manure Substitute. Blackleg. Summary of the Annual Reports of the Farm Advisors of California. The Function of the Farm Bureau. Suggestions to the Settler in California. Salvaging Rain-Damaged Prunes. Seed Treatment for the Prevention of Cereal Smuts. Feeding Dairy Cows in California. Methods for Marketing Vegetables in California. The Present Status of Alkali. Unfermented Fruit Juices. Vineyard Irrigation in Arid Climates. Testing Milk, Cream, and Skim Milk for Butterfat. The Home Vineyard. Harvesting and Handling California Cherries for Eastern Shipment. Artificial Incubation. Winter Injury to Young Walnut Trees during 1921-22. Soil Analysis and Soil and Plant Inter- relations. The Common Hawks and Owls of Cali- fornia from the Standpoint of the Rancher. Directions for the Tanning and Dress- of Furs. The Apricot in California. Harvesting and Handling Apricots and Plums for Eastern Shipment. Harvesting and Handling Pears for Eastern Shipment. Harvesting and Handling Peaches for Eastern Shipment. Poultry Feeding. Marmalade Juice and Jelly Juice from Citrus Fruits. Central Wire Bracing for Fruit Trees. No. 245. 247. 248. 249. 250. 251. 252. 253. 254. 255. 256. 257. 258. 259. 260. 261. 262. 263. 264. 265. 266. 267. 268. 269. 270. 271. 272 273. 275. 276. 277. 278. Vine Pruning Systems. Colonization and Rural Development. Some Common Errors in Vine Pruning and Their Remedies. Replacing Missing Vines. Measurement of Irrigation Water on the Farm. Recommendations Concerning the Com- mon Diseases and Parasites of Poultry in California. Supports for Vines. Vineyard Plans. The Use of Artificial Light to Increase Winter Egg Production. Leguminous Plants as Organic Fertil- izer in California Agriculture. The Control of Wild Morning Glory. The Small-Seeded Horse Bean. Thinning Deciduous Fruits. Pear By-products. A Selected List of References Relating to Irrigation in California. Sewing Grain Sacks. Cabbage Growing in California. Tomato Production in California. Preliminary Essentials to Bovine Tuber- culosis Control. Plant Disease and Pest Control. Analyzing the Citrus Orchard by Means of Simple Tree Records. The Tendency of Tractors to Rise in Front; Causes and Remedies. Inexpensive Lavor-saving Poultry Ap- pliances. An Orchard Brush Burner. A Farm Septic Tank. Brooding Chicks Artificially. California Farm Tenancy and Methods of Leasing. Saving the Gophered Citrus Tree. Marketable California Decorative Greens. Home Canning. Head, Cane, and Cordon Pruning of Vines. Olive Pickling in Mediterranean Coun- tries. 10m-l,'25 The publications listed above may be had by addressing College of Agriculture, University of California, Berkeley, California.