CALIFORNIA AGRIClftTURAL EXPERIMENT STATION BULLETIN 779 CONTENTS Introduction 6 Water Supply Features of the Klamath Project 6 Geology of the Tulelake Basin 8 Soils and Soil Charac- teristics 10 Drainage-Salinity Investigations, 1955 and 1956 17 Drainage Investigation ]J Salinity Investigation 18 Results 23 Effect of Drains on Salinity 26 Relationship Between Salinity and Water Table Levels 29 Discussion 30 Conclusions 31 Study of Deep and Shallow Open Drains, 1957 31 Construction Problems 31 Effect of Deep Drains on Water Tables 33 Shallow Drains 34 Conclusions 34 Drainage-Salinity Investigation, 1958 35 Drainage Investigation 35 Procedures and Materials 35 Results 37 Discussion 41 Salinity Investigation 43 Procedures 43 Results 4 Literature Cited 56 APRIL, 1961 Drainage- Salinity Investigation of the Tulelake Lease Lands This bulletin reports the results of four years of studies in the Tulelake lease lands after farmers of that area noticed a build-up of salinity in their fields causing an appreciable decline of their potato and grain yields. The problem was brought to the attention of the Irrigation Department of the University of California by the Tulelake Farm Advisor. An initial survey indicated a correlation between a severe salinity problem and the levels of the water tables. It was tentatively concluded that the build-up of salinity in the soils was caused by evaporation from the shallow water table prevalent throughout the lease lands, and that present drainage systems were inefficient in controlling water table levels. To determine the extent of the problem and find possible remedies, the U. S. Bureau of Reclamation agreed to conduct a joint investigation with the University of California. The studies began in 1954 with preliminary work, and continued through the summer of 1958 when a detailed experiment was conducted on a 78-acre tract of public land. This bulletin summarizes the objectives, procedures, and conclusions reached during each year of the investigation. 1 Submitted for publication, June 16, 1960. THE AUTHORS L. G. WILSON is Assistant Specialist in the Experiment Station, Department of Irrigation, Davis; J. N. LUTHIN is Associate Professor of Irrigation and Associate Irrigationist in the Experiment Station, Davis; and J. W. BIGGAR is Assistant Irrigationist in the Experi- ment Station, Department of Irrigation, Davis. UPPER KLAMATH LAKE OREGON KLAMATH FALLS LINK RIVER DAM LOST RIVER DIVERSION CHANNEL Olene OST RIVER DIVERSION WORKS U.S. B.R. KLAMATH PROJECT EXPLANATION ♦ PUMPING PLANT L.N. LEAGUE OF NATIONS FR SWS FROG POND 2 2 4 6 SOUTHWEST SUMP SCALE F MILES Yanno ■e FOLD DAM. ■J Bonanza i DRY LAKE'^-Z'' tw ^ ^ ' my M lift of three parts containing a total of 62,280 irrigable acres. Parts 1 and 2 contain 44,081 irrigable acres, and part 3, con- sisting of the Tulelake lease lands and Southwest Sump, contain 18,199 irrigable acres. The bulk of the water is obtained from the Lost River at the Lost River Diversion Dam and delivered through the J Canal. Pumping plants 1 to 6 inclusive are drainage pumps which deliver return flow water from land in Tulelake Division part 1 into the restricted sump or into N Canal for use on the lease lands. The drainage water pumped into the N Canal is diluted with water from the restricted sump. Pumping plants S and C pump drainage water from the Tulelake lease lands into the interconnected restricted sump. Pumping plant D, in turn, delivers water from the restricted sump through a 6,600 ft. concrete lined tunnel (250 cfs capacity) to Lower Klamath Lake. Addi- tional pumps lift the water from Lower Klamath Lake back into the Klamath River, completing the circuit. Prior to 1957 the operation and main- tenance of irrigation and drainage works of the Tulelake Division were under the jurisdiction of the Bureau of Reclama- tion. In January 1957, the dikes, distri- bution and drainage systems of this di- vision were transferred to the newly formed Tulelake Irrigation District. Geology of the Tulelake Basin The Tulelake Basin is located on the Modoc Plateau, the California portion of the Columbia Plateau, a volcanic field occupying a high semiarid area between the Warner Mountains and the Cascade Range, and encompassing a vast area in Washington, Oregon, and southern Idaho. Because of the geologic and his- toric interest of the Modoc Plateau, an area of about 75 square miles has been set aside as the Lava Beds National Monument. The geology of this area has not been investigated extensively, although the salient features have been discussed by Peacock (1931), and Hinds (1952). , Peacock considers the main characteris- tics of the region to be little moded pla- teau surface, mountains of volcanic ac- cumulation, and mountainous and hilly regions whose configuration was caused by dislocation and erosion. "In the Modoc area the Trans-Cascade Plateau has a general elevation of about 3,500 < feet. Rising to a maximum height of 3,500 feet above this level are a number of wide mountainous and hilly areas, and entrenched to a depth of more than 2,500 feet below the general plateau surface, the Pit River marks the lowest level in the region." The oldest feature in the Modoc Lava Field are tilted tuff and trap hills in the south, southeast, and north, probably developed during early Tertiary times. They form wide tracts of height varying from 500 to 1,500 feet above the general < plateau. The ridge dominating the west side of Tulelake is of this nature. The rocks composing these hills are the oldest in the region and consist of "a series of dense dark basalts of exclusive and in- clusive character, massive accumulation of basic, intermediate, and acidic tuffs and breccias, minor bodies of intrusive acidic rocks, and water-laid sediments of various kinds, the relative proportions of these components varying from place to place" (Peacock) . The original forma- tions were probably continuous, and be- cause the present hills and highlands are detached, the action of erosion and/or fracturing of the original formation fol- lowed by differential subsidence of dis- located blocks is indicated. Peacock and Hinds consider that the first phase of vulcanism in the Modoc Plateau was ex- plosive and related to central vents about which volcanoes developed. Volcanic ac- tivity probably was intermittent and during the quiescent periods lacustrine and other sediments were laid down, [8] only to be covered by subsequent out- bursts of the same cycle. A second type of highland called the Shasta Lava Highlands, as defined by Peacock, consists of a broad, irregular, mountainous belt occupying the mid- western and southwestern parts of the region extending east to the Central Volcanic Dome. This area merges, to the west, with the base of Mt. Shasta. The material of the highland differs from that * of the older formation and consists of "a massive, deeply dissected series of gray, loose textured lavas, mainly ande- sites and basalts with some subordinate obsidians, very different from the dense, dark traps occurring in the older forma- tion. The massive gray lavas were appar- ently delivered from central volcanoes of the Shasta type." The relief developed during this period of time is the greatest in the area with several mountains over 6,000 feet and with the Pit River cutting the formation below the 2,000 foot level. The main factor developing the present configuration appears to be erosion in conjunction with initial irregularities. The Shasta lavas follow the general rule that basic igneous material weathers more readily than acid igneous material and have given rise to abundant gray soils that support heavy timber growth. The third and youngest type of high- land in the area is called the Modoc Pla- teau, which occupies more than one-half of the whole sheet and whose elevations vary consistently between 4,500 and 5,000 feet. Most of the plateau was con- structed during Pleistocene time, al- though evidence is available (from dead trees, for example) that the youngest flows were deposited only a few centuries ago. Peacock divides the lavas of the pla- teau into three groups based on age, mode of eruption, and respective influ- ence on appearance of the plateau sur- face. The three groups are described by Hinds as follows: . . . the oldest (are) by far the thickest and most extensive. These lavas were apparently highly liquid, forming thin, rather even surfaced flows, were erupted from fissures rather than from central openings, and flooded the re- gion to build up a plain of gently undulating surface. The second group is comprised of generally much rougher surfaced flows, mostly erupted from central vents about which were constructed broad, low shield volca- noes. Explosive eruptions also oc- curred forming a considerable number of small cinder cones. The third group includes the most recent flows such as the Callahan on the northern edge of the Medicine Lake Highland, an eastward-projecting promontory of the Cascade Range and the Burnt Lava at the southern margin of the same area. The flow surfaces are chaotic jumbles of great blocks. Ris- ing from their surfaces are cinder cones whose craters and outer slopes are almost perfectly preserved, testi- fying to the recency of the activity; in fact, this suggests the definite pos- sibility of further eruptions . . . Several lakes in past and recent geo- logic times have deposited lacustrine ma- terials on the lava flows. The major recent lakes are Medicine Lake, Clear Lake, Klamath and Lower Klamath lakes, and Tulelake. (Tulelake has dried up in recent time and similar shrinkage has occurred in Lower Klamath Lake) . Peacock summarizes the events bear- ing on the origin and history of Tulelake as follows: Towards the end of the volcanic period in which the older lavas originated, probably towards the close of the Mio- cene, the country was traversed by a series of fractures trending north and south in accordance with the main trend of the Basin Range faulting. Irregular differential subsidence of the dislocated blocks produced hoist and graben structure and initiated vig- orous degradation. The Modoc Plateau lavas were then poured out on the comparatively level sedimentary plains lying between the horsts, and renewed crustal movements along the north- south fractures produced the horst and graben structure of the plateau . . . Again erosion and sedimentation were promoted, and the wide areas of lake sediments . . . were deposited. [9] The present Tulelake sediments lie on the lavas of the Modoc Plateau, indicat- ing that this lake formed after the forma- tion of the plateau, and according to Peacock in late Pleistocene or recent time. Lacustrine sediments visible under the older lavas of the plateau on the east side of the Tulelake Basin also indicate that more extensive lakes existed before Modoc lavas were deposited. In fact, be- cause of the geologic construction of the area it is possible to uncover old lake sediments at great depths below the pres- ent soil surface. This is indicated by ex- amination of the two well logs given in tables 1 and 2. When Tulelake was initially surveyed in 1884 it occupied an area of 150 square miles. Surveyed again in 1924, it occu- pied an area of only half this size. Today the lake bed is entirely dry. The control of the inflow of water by project structures to Tulelake and the construction of pump- ing plant D has permitted the reduction of the former Tulelake to the present open water sumps. An unpublished study of the so-called underground passages, made many years ago by the Bureau of Reclamation, showed that the amount of water removed through these passages was small. About 36 inches of water a year is the average evaporation in the Tulelake area. Soils and Soil Characteristics To date there has been only one de- tailed soil survey (U. S. Department of Agriculture, 1908) conducted in the Tulelake area ; at the time of this survey the lake was still quite extensive so that the data are valueless for modern inter- pretation. The Bureau of Reclamation has conducted a land classification sur- vey (U. S. Bureau of Reclamation, 1950) , but its major purpose was to investigate the lands for irrigation feasibility. Dach- nowski-Stokes (1936) made a reconnais- sance survey of the lake bed and gives the only detailed profile available in the literature. This profile was obtained in the NW% Sec. 5, T. 47N, R4E and showed the following characteristics: 1 to 12 inches, sedimentary muck. The upper 9 inches of the cultivated organic soil consist of dark-gray mel- low diatomaceous material which is loamy in character. Below this is a brownish-gray diatomaceous organic sediment containing a few flattened root stocks of water lily, rhizomes, and rootlets of tule in an advanced stage of decomposition, and possibly an ad- mixture of clay. The underlying mate- rial is a thin band of white pumice which resembles coarse sand. 12 to 36 inches, very compact green- ish-brown or olive-green organic sedi- ments, largely diatomaceous and more or less banded. 36 to 72 inches, gray siliceous pre- dominantly diatomaceous sediments. This material includes thin bands of coarse pumice from volcanic eruptions and varying quantities of clayey resi- due brought in, probably, by flood waters. The material grades rather abruptly into a. light-gray accumulation of sili- ceous remains of diatoms which was not probed below a depth of 8 feet from the present surface. The most significant feature of the soils of Tulelake, as observed in this pro- file, is the high amount of diatomaceous earth they contain. This material, briefly * described here, is responsible for many of the unusual characteristics of these soils. Diatomaceous earth is composed of the remains of countless numbers of one- celled organisms called diatoms. Taxo- nomically, diatoms belong to the Phylum Chyrsophycophyta class of Bacillarieae, one of the seven phyla of algae. It has been estimated (Leppla, 1953) that there are about 10,000 varieties of diatoms. Dia- tom cells contain a nucleus, vacuoles, cytoplasm and golden-brown chromo- plasts. Chlorophyll a and c, fucoxanthin and carotenoids are present, rendering the diatoms autotrophic. The cell wall is composed of silica-impregnated pectin 10] and consists of two halves fitting to- gether much like the two parts of a petrie dish. During reproduction by cell divi- sion the two halves separate, each daugh- ter cell receiving one-half of the old cell wall. The daughter cells then build a new second half of the cell wall fitting inside the original half (Robbins and Weier, 1950). Under proper environmental condi- tions cell division is quite rapid and it has been calculated that one frustule would become a billion diatoms within a month, if the time of a single subdivision were 24 hours (Leppla, 1953). Deposits of diatomaceous earth are formed by set- tling of the large masses of dead diatoms on the bed of the lake in which they have lived. Diatomaceous earth chemically con- sists mainly of silica and has been de- scribed as deposits of opaline silica in a classification of the minerals of Califor- nia (Murdock and Webb, 1956). There are no known published analyses of diat- omite from the Tulelake area, but an analysis has been determined for samples from Klamath Lake and will probably serve to approximate Tulelake deposits. The analysis is given in table 3. Table 3 ANALYSIS OF DIATOMITE 4.5 MILES SOUTHWEST OF KLAMATH FALLS (MOORE, 1937) Table 1 WELL LOG DRILLING, AUGUST 1950 WELL LOG NO. 3, CITY OF TULELAKE Chemical Per cent Si0 2 75.56 Ti0 2 0.64 Fe 2 O s 2.66 A1 2 0, 8.64 CaO 1.20 MgO 0.37 Na 2 1.08 K 2 0.26 S0 3 0.06 CI None C0 2 0.11 The remainder is water and organic matter. The source of silica apparently is the volcanic deposits common in areas where diatomaceous earth develops (Taliaferro, Material Depth (ft.) Thickness (ft.) Mud Pumice Mud 170 171.5 185 186 190 191 247 247.5 283 290 291.5 593 603 850 170 1.5 13.5 1 Mud Pumice Mud 4 1 56 0.5 Mud Peat brown 35.5 7 1.5 Mud Chalk Mud 301.5 10 247 Table 2 WELL LOG DRILLING, NOVEMBER 1953 WELL LOG NO. 4, CITY OF TULELAKE Material Lake bottom mud Lake bottom mud Pumice and shells Lake bottom mud Mud, shale streaks, and sand . . Shale Hard sand Mud Hard shale Soft shale Sand streaks and shale Shale Rock Clay and shale Rock abrasive Lava rock hard Shale tough Rock Shale tough Green shale and sand streaks . . Soft lava rock and clay streaks Soft clay Rock-sand streaks Rock Lava rock soft Lava rock Lava rock, lime deposits Soft lava rock Broken lava rock Hard rock Sandy lava rock Soft lava rock, shale streaks. . . Firm lava rock Firm rock Firm rock Lava cinders Hard black rock Depth (ft.) (Ground level)-150 150- 283 283- 291 291- 825 825-1105 1105-1255 1255-1259 1259-1325 1325-1336 1336-1567 1567-1597 1597-1610 1610-1673 1673-1679 1679-1725 1725-1865 1865-1870 1870-1917 1917-1922 1922-1981 1981-2129 2129-2172 2172-2200 2200-2207 2207-2248 2248-2391 2391-2444 2444-2454 2454-2480 2480-2503 2503-2548 2548-2566 2566-2582 2582-2600 2600-2607 2607-2665 2665-2676 [11] 1935). These deposits create large amounts of hydrous silicates and silicic acid in the water in which the organisms live. The most outstanding characteristic of diatomaceous earth, and which makes this material industrially important, is its low bulk density. Some values for Oregon deposits are given in table 4. Table 4 APPARENT DENSITY OF DRIED SAMPLES OF DIATOMITE (MOORE, 1937) District Apparent density (lbs/ft 3 ) Torrebonne 21.8 55 28.7 Harper 27.1 " 33.8 55 26.7 55 29.2 Austin 29.1 Otis Basin 33.8 55 30.1 »» 26.6 Klamath 34.6 55 32.1 Soil texture. One of the characteristics of the Tulelake soils which make them useful for field research is their uniform- ity of texture throughout large areas. Data from the mechanical analysis of a typical profile is given in table 5. This analysis shows the predominance of clay- and silt-sized particles through- out the profile. The composition of this material is, of course, largely diatoma- ceous earth. The large amount of sand at the 66-inch depth is probably pumice ma- terial. In addition, it is not uncommon to find large quantities of fresh-water shells in these soils. Because of the large amount of organic material in the surface of these soils, not shown in the analysis, they would probably be best classified as organic clay loams, or more specifically, peaty or mucky clay loam depending on the degree of decomposition of the or- ganic material. In addition to the vol- canic origin of the soils, some material was brought in by the Lost River. Soil structure (the groupings of pri- mary soil particles into aggregated units separated from adjacent units by sur- faces of weakness) has two important effects on soil as a medium for plant growth. First, structure determines to a great extent the ability of the soil to transmit water. Secondly, as a corollary to the first, structure contributes to aera- tion of the soil. Tulelake soil structure was probably formed in the following manner: At the beginning, when the area was submerged, the dead remains of countless numbers of diatoms slowly settled to the bottom of the lake, forming the typical horizontal laminar structure of lacustrine deposits. Over a vast period of time a considerable depth of diatomaceous soil material was Table 5 MECHANICAL ANALYSIS OF TULELAKE SAMPLES 4 Depth Sand Silt Clay Fine clay Texture inches per cent 0-12 26.5 43.2 25.4 4.9 Clay loam 12-18 17.0 35.3 40.2 7.5 Clay 18-24 15.5 31.9 43.3 9.3 Clay 24-30 16.2 41.0 35.7 7.1 Silty clay 30-42 18.1 39.1 35.3 7.5 Clay 42-48 18.0 46.8 30.9 6.3 Silty clay loam 48-54 36.0 32.4 26.8 4.8 Clay loam 66+ 54. 5t 18.3 23.0 4.2 Sandy clay loam * Sand 0.05-2.00 mm; Silt 0.002-0.05 mm; Coarse clay 0.2-2.0 microns; Fine clay 0.2 microns. Determined at Ohio State University. t Variation in sand ±13. [12 My; ; '-MM ^MMMMMfi Exposed diatomaceous soil in bottom of shallow drainage ditch. Note the large cracks due to drying. deposited along with pumice material de- rived from the volcanic outbursts. As mentioned previously, volcanic material served as the source of silica used in the cell walls of the diatoms. When the water in the lake became suf- ficiently shallow to permit plant growth, the peat deposits of the area began to be developed. Their shallow depths indicate that the time of peat formation was very short compared to the time of the forma- tion of the diatomaceous deposits. As the water began to disappear from the surface of the former lake, the soil material was probably essentially struc- tureless except for the laminar nature of the deposits. When the water table began to recede even further as a result of natu- ral or man-made drainage, the action of alternate wetting and drying began to develop vertical cracks in the soil. Over the period of time that these soils have been drained and cultivated, a fairly definite structure has been built up which may best be described as prismatic. The cracks play an important role in drain- age because they increase the permea- bility. Although no data are available, it appears from observations of water flow into newly opened ditches, that a large proportion of the water flow occurs through these cracks. The structure of Tulelake soils appears susceptible to deterioration through com- paction. For example, one road con- stantly used by heavy equipment was found to be compacted to such an extent that it formed a dam preventing drain- age from an adjacent field into an open ditch on the other side of the road. Porosity. The high macroporosity of Tulelake soils, to which the structural cracks contribute, is based on the in- trinsic nature of the material itself. The pore space can be determined from bulk- density-particle-density calculation — or [13 Table 6 MOISTURE EQUIVALENT (ME) AND 15 ATMOSPHERE DATA FOR TULELAKE SOILS* Sample 1 Sample 2 Sam pie 3 Sample 4 Sample 5 Depth (in.) ME (per cent) 15 atm. (per cent by wt.) ME (per cent) 15 atm. (per cent by wt.) ME (per cent) 15 atm. (per cent by wt.) ME (per cent) 15 atm. (per cent by wt.) ME (per cent) 15 atm. (per cent by wt.) 0-12. . . 67.89 84.39 83.80 75.33 75.48 71.90 55.04 51.41 66.43 84.30 80.80 76.78 79.35 73.73 52.47 48.54 70.69 86.96 78.50 78.48 91.67 56.49 52.79 66.96 87.54 85.44 77.27 76.50 53.51 47.61 64.56 89.71 88.32 79.48 76.84 47 85 12-24 24-36 36-48 50 25 48-60 60-72 Average 76.46 76.90 81.26 78.74 79.78 * Determined by Irrigation Department, University of California, Davis. can be estimated from per cent by vol- ume water held at saturation. Such values obtained by the latter approach are given in table 8, where the per cent by volume held at saturation ranged from 71.41 per cent to 76.68 per cent with an average of 74.85 per cent. These values are approxi- mately in the 75 to 88 per cent range given by Leppla (1953) for porosity of diatomaceous earth. Values for the water- holding capacity of this soil at saturation, expressed on a per cent by weight basis, are also given in table 8. These values range from 140.27 to 172.0 per cent by weight with an average of 154.6 per cent for the entire profile. This compares to values normally obtained for organic soils rather than those for mineral soils. Field capacity, as defined by Veih- meyer and Hendrickson (1948), is "the amount of water held in the soil after the excess gravitational water has drained away and after the rate of downward movement of water has materially de- creased." This soil property is obtained in the laboratory by two techniques, the moisture equivalent method or the one- third atmosphere percentage method. The moisture equivalent is obtained by centrifuging saturated soils in perforated cups at a rate equivalent to a force of 1,000 times gravity for 30 minutes. The one-third atmosphere percentage is ob- tained by bringing the soil moisture in a sieved and saturated soil sample into equilibrium at one-third atmosphere. These two values have been found to correlate closely with each other. They are generally lower than field capacity in coarse soils, equal to field capacity at moisture values at 20 per cent, and higher than field capacity in fine textured soils. Moisture equivalent and one-third at- mosphere percentage values for a num- ber of Tulelake samples are given in tables 6 and 7. The one-third atmosphere values are slightly higher than those for moisture equivalent and range from 81.05 to 107.12 per cent with an average Table 7 MOISTURE EQUIVALENT DATA FOR TULELAKE* Soil Sample No. Depth (inches) ME (per cent ) 2-D 0-10 10-30 30-60 0-12 12-30 30-60 0-12 12-30 69 4 11-D 10-E 84.8 73.8 70.0 77.1 80.6 67 8 75.2 Average 74.84 * Determined by U.S.B.R. [14 Table 8 TULELAKE SOIL SAMPLES- PHYSICAL DATA< Sample No. Depth (inches) H 2 Oat satur. (per cent by wt.) H 2 Oat satur. (per cent by vol.) H 2 Oat 1/3 atmos. (per cent by wt.) H 2 Oat 1/3 atmos. (per cent by vol.) 15 atmos. (per cent by wt.) 15 atmos. (per cent by vol.) Dp| (gm/cm 3 ) Dpt (lba ft :i , 1 0-12 12-18 18-24 24-30 30-42 42-48 48-54 66+ 140.27 152.90 157.03 162.80 172.04 122.10 157.79 171.61 71.41 76.24 75.16 76.68 76.56 72.56 75.22 74.97 81.05 97.32 91.51 103.23 101.28 79.51 97.87 107.12 41.26 48.50 43.80 48.63 45.07 47.29 46.66 46.78 51.12 62.81 69.35 65.22 66.45 55.56 71.81 62.62 26.02 31.28 33.22 30.72 29.57 33.00 34.53 27.36 2.259 2.398 2.449 2.441 2 445 2.482 2 414 2.312 153.34 2 149.54 3 152.72 4 152.22 5 152.47 6 154.78 7 150.54 8 144 IK Average 154.6 74.85 94.86 46.00 63.12 30.71 2.425 151.22 * Determined at Ohio State University, t Particle density. of 94.86 per cent. Moisture equivalent values for all samples range from 64.56 per cent to 91.67 per cent with an average of 78.00 per cent. The data point out that the field capacity in the first foot of soil is lower than elsewhere in the soil. This seems paradoxical because of the high organic content of the surface which should raise the moisture equivalent above the deeper soils. The high values of moisture equivalent for these soils are even more pronounced when compared to values for mineral soils; clays are about 30 per cent, loams about 20 per cent, and sand 8 to 10 per cent. In fact, the moisture equivalents of Tulelake soils are comparable to those for organic soils rather than mineral soils. Feustel and Byers (1936), for ex- ample, report values for moss peat, sedge peat, and reed peat of 166, 112, and 100 per cent respectively. The lower limit of the so-called "avail- able moisture range" is represented in this report by the 15-atmosphere per- centage. The technique consists of deter- mining the amount of water retained in a previously saturated soil sample which has been placed on a porous membrane and which has been permitted to come to equilibrium with 15-atmosphere pres- sure in the gas phase. Fifteen-atmosphere percentage values for Tulelake soils are given in tables 6 and 8. The range of values in these tables is from 47.61 to 71.81 per cent by weight. The average value for the profile given in table 8 is 63.12 per cent. These values are more comparable to those obtained for organic soils than for mineral soils. Typical values for mineral soils are: clay 15 per cent, loam 10 per cent, and sand 4 per cent. The three organic materials tested by Feustel and Byers (1936) , moss peat, sedge peat, and reed peat had wilt- ing percentages of 82.3, 60.8, and 70.7 respectively. The moisture characteristics of the dia- tomaceous soils of Tulelake may be sum- marized by saying that the saturation percentage, field capacity, and 15-atmos- phere percentage values are considerably higher than values expected for typical minerals soils and compare closely to values for organic soils. Permeability or hydraulic conductivity of a soil is an important characteristic because it gives an idea of the ease with which drainage can occur. It may be defined as the rate of water transmission through a porous medium under a given potential gradient. Permeability can be determined in the field in a number of ways but the two most common are the auger hole and piezometer techniques (Luthin. 1957). Briefly, the auger hole technique consists of augering a hole into the soil below the water table. The [15] water in the hole is pumped out and the rate of rise of water table determined. Knowing the shape of the hole together with the depth to water before pumping and rate of rise of water, a curve may be used to determine permeability in inches per hour. The piezometer technique uses a pipe driven into the ground below the water table with the soil augered out of the pipe. A cavity is augered below the end of the pipe and water pumped out of the pipe. The rate of rise of water in the pipe together with the geometry of the cavity can be used to determine per- meability from a chart or by calculation. The auger hole technique gives an aver- age permeability throughout the depth that the water table is lowered. The pie- zometer method, on the other hand, gives the soil permeability directly at the end of the pipe. Because of the high permeability of the surface 5 feet of Tulelake soils it is difficult to determine permeability by these methods. The hole fills so rapidly after pumping that one does not always have time to lower a measuring device into the hole. Successful tests gave per- meability values of 7 to 14 inches per hour. The permeability of the material below about 5 feet (that is, below the zone of drying and shrinking; best described as diatomaceous ooze) is extremely differ- ent from that of the surface. For example, measurements of permeability at 6 feet and deeper by the piezometer method gave values of about 0.001 inch per hour. The vast permeability differences between surface and subsurface soils ap- pears to be caused mainly by the differ- ence in structure. The surface soil, in which alternate wetting and drying, freezing and thawing, and root action have contributed to formation of a pro- nounced structure, contains a large num- ber of cracks which are able to transmit water rapidly. Below the depth of these activities the soil is essentially structure- less and, in spite of the large microporos- ity of the material, water transmission is very slow. Measuring the soil hydraulic conductivity by the auger hole method. Drainage— Salinity Investigations, 1955 and 1956 The objectives of the 1955 and 1956 investigations were fundamentally the same and consisted of two phases: Drainage Investigation To determine water table levels throughout the lease lands. To study the efficiency of present drainage methods in controlling water table levels. To determine sources of excess water creating drainage problems and factors limiting efficiency of present drainage techniques. Salinity Investigation To determine salinity levels through- out the lease lands and delineate specifi- cally affected areas. To determine the nature of the saline soils. To determine relationship between sa- linity and depth to water table. To examine specific areas to determine relationship between salinity and drain spacing. To determine relationship between sa- linity and crop yields. Drainage Investigation To determine the depths to water throughout the lease lands, 31 observa- tion wells were installed during the spring of 1955. Nine of these wells were installed in the League of Nations, 15 in the Frog Pond, and seven in the South- west Sump. These wells consisted of standard pipe in about 5-foot lengths driven into the ground and augered out. They were read bimonthly during the summers of 1955 and 1956. In addition to these observation wells, 10 Stevens Type F automatic stage re- corders were installed in 1955 to give a better picture of water table fluctuations than can be obtained from observation wells. Four of the automatic recorders were re-installed in 1956. These instru- ments made it possible to obtain an esti- mate of the efficiency of drains in con- trolling water levels. Average depths to water table for the summers of 1955 and 1956, in the three areas of the lease lands studied are given in table 9. Table 9 AVERAGE DEPTHS TO WATER TABLE IN THE TULELAKE LEASE LANDS 1955-1956 (FT.) Area League of Nations Frog Pond Southwest Sump 1955 1956 2.92 2.92 2.84 1.62 2.29 1.83 Average depth to water table values are only approximately valid for such large areas and serve only as a rough estimate of actual conditions because of the limited number of observations, and also because of the possibility that some of the observation wells may not have been operating effectively. Two primary factors contribute to the maintenance of high water table condi- tions in the leases. The first factor, lim- iting efficient control of water table levels, [17] is the lack of adequate drainage. The present drains in the lease land are of inadequate depth and of insufficient ca- pacity to maintain the water table at sat- isfactory levels. Not only are the drains of inadequate depth, but the common practice of blocking the drain to obtain some benefit from subirrigation is detri- mental to the entire area since it removes the drain from operation. The inadequate drainage in the area is evident from the graphs of the automatic recorder which show that after an irrigation the rate of drop of the water table is very slow. A contributing factor to the low capacity of the drains in the lease land is the great amount of algae found in the drain ditches. The Bureau of Reclamation and the present operators, the Tulelake Irri- gation District, have spent a considerable amount of money and time to control the algae, but present chemicals and eradi- cation techniques appear to be ineffec- tive. The second factor, which is unfortu- nately common to most irrigated areas, is low irrigation efficiency. Responsible for this are the high infiltration rates of the soil necessitating large flows, and the ex- cessive application of water in those areas where it is cheap and plentiful. The primary reason for the higher water table levels in 1956 compared to 1955 was the above-normal amount of rainfall during the early part of 1956, shown in table 10. The 21 -year average mean precipitation for the Tulelake weather station was 9.97 inches per year. Precipitation from July 1955 to June 1956 was 16.67 inches compared to 5.02 inches for a similar period for 1954 to 1955. Salinity Investigation The presence of salt in the soil in quantities above allowable tolerances manifests itself to the farmer mainly by the reduction of crop yields. Excess salt in the soil solution influences crop growth in two ways. First, the osmotic SoP'w.^ Water stage recorder used to measure fluctuations of the water table. pressure is increased and this, in turn, increases soil moisture stress — which is the sum of the soil moisture tension and osmotic pressure of the soil solution. In- creasing the soil moisture stress limits the water available for growth, and the plant wilts at a higher moisture content than if the osmotic component were negligible. The second factor influencing crop growth is known as the specific ion effect. This implies that specific ions pro- duce toxic effects when taken up in other than tolerable quantities. It was not the objective of this study to determine whether osmotic pressure or toxicity effects were causing growth depression. Growth is depressed as salts accumulate in the soil, and there are wide differences in the tolerance of species to various ions. The salt content of soils can be evalu- ated by determining the electrical con- ductivity of the solution extracted from a saturated soil paste (Richards, 1954). The conductivity and the osmotic pres- [18] m Ph EH <1 CD CM 1 ~ CO ■g t- >o -r — O lO oc CD OO 1 - -r eq oo O co — -.". 73 _, o — o co — -f o PH _ O f» CI PH tf-j O r- ci 37 + + 1 1 + ' + 1 1 1 + 1 + 1 + 1 + + o CO t^ o O TP o N _ I_ CO «o CO CM -r »o CO oa < -r ~r -p "* -r J CM OS lO co -r CO ~P t^ CO ■* o pH OO rH ■* 00 o CM •o CO t ^ CS OO 00 CO OS -f >o o -p t -1 •H T— 1 CD O I- o -f C7j OS »o — < A H t^ CM -*i CM l~ t— 1 CO PH OS -f co ,_l o — pj CO CM CM CO CM CO CM CO OO CD CO OS ^ o 00 — OS CS "# OO CM 1—1 OO "+ 1 CO lO CO »o 00 o O OO l-H lO _, sc CO CM -P "* -r " T U3 00 to -P CS -P -P «H T- ' o CD 00 l-H o o CO CM O0 CM OJ CO © os OS ,_l CD m w "3 lO "0 •O iO >C co >o CM CO 00 ■*! CO « — ' CO -H OO o »o CM **■ T— CO O ^r co CD >> t^ lO •« t ^ © T-< o t- t— CS O t^ o -tf 1 C-) t-H co o t^ CO CM CO »-s CD O CD pP CO O CM T- CD CC CD CD CD co CD CD iO CM ^_ CD CM CD O C2 00 G OS CO O CO I s " * - ' CM 3 ,_, ph CM _ lO ,—, CD •-S 'C CD >C «5 »o lO «5 IQ co & § co co OO CM CM CM CO OO © CO t>- OS OS O0 CO CM CM CM co r^ co T— CO l^ CM OO CO ^H O '-i O CO CM lO © -H o o CM CM CM — t~ >o IC ^ lO co „ ^ CM «0 i_ CO lO CM -P CO T-H t"- o OS CO OO t-^ kO o o t^ OO «3 OS << CM -r O OO CD — -r O CM -p "* -r co "* CM _^ US >o CO cs 03 3 CD «o CO OS •~ ' CO CM »C OS t^- (^ T-H »# PH CO PP rt- o t^ © t^ CM OS O O © OS CO co CO OO lO CM 05 CO 00 -D CM CO OO "5 CO O0 o t"~ OS CO CO CM Tf CM O0 OS — - CD CO rH -r O CO T— T— PH Ui o t^ O OS CM CO — lO — © co CO CO w -,, CM (.M "5fl co CD OS t^ co c OS 00 t~- 'f T-H T-H >o CM c= CM "3 T-* OO OO O •*rt" CM X~- 00 03 ,_, -r T-H Ifl p_ *o> „ O O "* CM CM O ^f >-H CO CM co CM CO CO CM CO CM CO 03 0) ;- D. C ■ & a c • C c Q, o, ■ a H g H s H £ H 6 H 0) Ph Ph CO — X Ph H Ph H Ph H Ph H Ph H Ph Eh Ph H Ph Eh - h - H CM CO ~r in CD t^ OO OS lO 1/0 >o OS OS cs cs OS Oa 02 OV 2 cs sure of a solution are both dependent upon the number of ions in solution and the concentration of the solution. An increase in the concentration of salts in solution results in an increase in the electrical conductivity and osmotic pres- sure of the solution. The relationship be- tween concentrations and conductivity varies somewhat, depending on the type two quantities for any given combination of salts. This relationship has been deter- mined from a large number of samples taken in the Tulelake area and is pre- sented in figure 5. Determinations of the conductivity of the saturated extract provide a means of appraising soil salinity in relation to plant growth because the saturation per- of salts in solution. It is possible to de- centage can be directly related to the termine the relationships between these field moisture range. It has been found Figure 2. 1 BUREAU OF RECLAMATION, USOI, " ^ KLAMATH FALLS, ORE. TULE LAKE DRAINAGE STUDY « i for a large number of mineral soils that the saturation percentage is about twice the value for field capacity and about four times the value for the 15-atmos- phere percentage. Peats and diatoma- ceous soils do not necessarily follow this relationship. For example, on the basis of the analysis in table 8, the ^-atmos- phere percentage is about 60 per cent and the 15-atmosphere percentage about 40 per cent of the saturation extract. This means that the soil solution is 1.67 times more concentrated at field capacity and 2.5 times more concentrated at PWP, than for the saturated soil. On the basis of studies (Richards. 1954) on the response of plant growth in the presence of soluble salts in many soils it has been possible to obtain a guide, presented in table 11, which re- lates the conductivity of the saturation extract to the yields of crops which may Figure 3. -LCCCNO- — C«nol« — Lou. on - - Dit.i ■"« ROM o . 0M"> « »«lr "noMlli »••• (7-U-M) tout atari f.c.a tO*-fO-|/t) m ••'••«• toMii r-ze-M) tan atari I C x io'-<0-i/V) !• •••*• • (7-M-gSi »'«ot antra E.c xio*-(o-i/*l n * m ••• ••! tnt»» tr-ta-ss). «OTC TH.I not * topi.o Iron o USDI Bor.o. of »mo..i.». >ii ...i.ll.d y> TULt LAKt DIVISION- Kloaork Projocl- Ot.ro. t.W«»t, ..-».'.« II-0-2JI, ...,..« S-«-»4. IRRIGATION DEPARTMENT UNIVERSITY OF CALIFORNIA DAVIS, CALIFORNIA in cooc«rolion ailn Ita BUREAU OF RECLAMATION, USDl, KLAMATN FALLS, ORE TULE LAKE DRAINAGE STUDY be expected when grown at the indicated levels of salinity. The table indicates that above 4 minhos per cm, the yield of many crops may be restricted. Soils which have a conductivity above 4 mmhos are often re- ferred to as saline soils. The presence of excess exchangeable sodium in the soil often-times leads to problems of water and root penetration. Exchangeable sodium percentage (E.S. P.) is defined as the degree of saturation of the soil base exchange complex with sodium, expressed on a percentage basis. The E.S. P. of a soil which occurs when equilibrium between the soil solution and the exchange complex is reached has been estimated from the concentration of ions in the soil solution using the sodium adsorption ratio (S.A.R.). For many soils the correlation between E.S.P. and S.A.R. is good. For others it is not Figure 4. IRRIGATION DEPARTMENT UNIVERSITY OF CALIFORNIA DAVIS, CALIFORNIA BUREAU OF RECLAMATION, USOI. KLAMATH FALLS, ORC. TULE LAKE DRAINAGE STUDY ».»! I*. !•» O.-o.n d, R v WeriKll 10 20 30 •0 70 SO 90 100 IK) CONCENTRATION (m«/l) 120 ISO MO ISO 160 110 ISO Figure 5. Concentration of saturated soil extracts (me /I) related to electrical conduc- tivity (mmhos/cm) for Tulelake, California. satisfactory. Although the S.A.R. has been used in the present report for esti- mating the E.S.P., further investigation is needed to determine the accuracy of such a procedure. The sodium adsorp- tion ratio is defined as Na + /[(Ca ++ + Mg ++ )/2] 1/2 where Na, Ca, Mg refer to the concentrations of the designated ca- tions expressed in milliequivalents per liter. Soil samples were taken in the vicinity of each observation well and automatic recorder in one-half foot increments down to three feet, three times during the summer of 1956, to estimate changes in salinity throughout each season. In addition detailed studies were conducted on specific areas to determine the rela- tionship between the action of drains and salt concentration. An investigation of the effect of salinity on yields and vegeta- tive growth of crops grown in the Tule- lake lease lands was also conducted. Finally, the over-all relationship between salinity and water table levels was deter- mined. Irrigation water. Although the bulk of the irrigation water is obtained from the Lost River diversion through the J Canal, considerable mixing takes place in the areas under investigation with other waters, mainly drainage waters. Not only do these vary from place to place, but also during the season and from year to year. For this reason it is difficult to give analyses which would indicate the situ- ation for the entire area. A few analyses are given in table 12 to indicate what has been found. Results Average conductivity values of satu- Table 11 SCALE OF CONDUCTIVITY (MILLIMHOS PER CM AT 25°C) Salinity effects mostly negligible Yields of very sensitive crops may be restricted Yields of many crops may be restricted Only tolerant crops yield satisfactorily Only a few very tolerant crops yield satisfactorily 2 4 8 16 23] Table 12 ANALYSIS OF SOME IRRIGATION AND DRAINAGE WATERS IN THE TULELAKE AREA EC X 10 3 mmhos/cm milliequivalents per liter Water sample No. Ca Mg Na HCO3 CI SO4 1 0.35 0.42 0.52 0.89 1.46 0.78 1.00 1.28 2.60 5.68 1.26 1.20 1.88 2.92 4.56 1.37 1.60 2.39 4.14 6.61 2.28 2.00 4.11 5.37 6.45 0.28 0.30 0.34 0.45 0.96 0.96 2 1.10 3 1.17 4 3.91 5 9.48 Sample description: No. 1— J Canal, 6/28/56 No. 2— J Canal, 9/4/54 U.S.B.R. No. 3— Pumping Plant 5, 9/21/58 No. 4— Pumping Plant C, 7/27/56 No. 5— Pumping Plant C, 9/8/58 Table 13 CATION ANALYSIS OF SAMPLES TAKEN 5/25/55^ Soil depth (ft.) EC X 10 3 milliequivalents per liter Na Ca Mg Total Per cent Na 57.2 51.3 48.0 47 43.0 41.5 38.6 41.5 43.0 48.2 43.0 40 38.0 40.0 55.3 48.0 45.0 38.0 55.0 49.2 46.2 45.8 45.3 47.4 48.2 49.5 49.7 35.7 50.8 54.5 ESPf Lease No. 25 0-H V2~\ 1-1H 1^-2 2-2^ 2^-3 Lease No. 36 (south end) o-y 2 H-i i-i^ 1^-2 2-2^ 2^-3 Lease No. 78 0-^ Vr-1 1-1H 1M-2 2-2^ 2H-3 Lease No. 98 o-H Vr-l 1-1H 1^-2. 2-2M 2^-3 Lease No. Id (SW sump) 0- l A l A-l 1-1^ 1^-2 2-2^ 2^-3 9.74 7.86 7.30 6.98 5.85 5.25 3.29 4.15 3.72 2.48 2.75 2.91 1.54 1.56 1.07 1.49 1.19 1.18 8.98 7.05 6.28 5.62 5.93 6.18 2.25 2.11 1.99 3.56 1.88 1.56 85.22 62.61 54.78 48.26 36.96 32.61 17.52 24.13 22.26 15.50 15.50 15.50 7.79 8.00 7.30 8.70 5.98 5.23 75.65 50.78 40.00 35.65 35.91 38.96 14.24 13.04 12.52 15.39 11.39 28.50 26.60 27.50 26.60 26.60 25 00 14.25 18.15 16.70 9.58 11.80 13.80 5.90 5.70 2.85 5.14 4.36 5.14 27.50 26.60 25.70 23.60 25.70 25.70 15.75 6.10 4.56 32.65 31.65 30.49 27.15 21.32 19.58 10.67 14.17 12.08 6.50 8.16 8.79 6.10 5.83 2.71 3.86 2.60 3.06 31.65 24.16 19.32 17.00 16.16 16.16 6.50 5.83 5.00 10.67 4.26 2.76 2.87 1.28 1.28 0.87 0.87 1.02 2.92 1.23 0.64 0.51 51 0.51 0.67 0.49 0.37 0.41 0.34 0.34 2.46 1.69 1.69 1.43 1.28 1.28 1.02 0.82 0.82 1.23 0.67 0.67 149.24 122.14 114.05 102.88 85.75 78.21 45.36 57.68 51.68 32.09 35.97 38.60 20.46 20.02 13.23 18.11 13.28 13.77 137.26 103.23 86.71 77.68 79.05 82.10 29.64 26.37 25.22 43.04 22.42 17.64 15.5 11 9.5 9.5 9 10 5 5 4.5 6 4.5 * Obtained from saturated extracts. t Estimated from figure 27, Agricultural Handbook 60 (Richards, 1954). rated extracts from samples obtained in 1955 and 1956 are given in table 14. Cation analyses of the chemical con- stituents in the saturated extracts of five representative samples are given in table 13. In addition, the profiles of salinity obtained from three areas known to be badly affected are shown in table 15. Monthly conductivity values for the sum- mer of 1955 are shown in figures 2, 3, and 4. Using the criterion that 4 mmhos con- ductivity of saturated extract marks the division between saline and nonsaline Table 14 AVERAGE VALUES OF CONDUCTIVITY OF SATURATED EXTRACT (ECxlO 3 ) AND WATER TABLE LEVELS (FT.) FOR TULELAKE DRAINAGE STUDY 1955-56 Soil depth (ft.) Value of conductivity 5/25/55 7/26/55 9/15/55 Ave. 1955 Ave. depth to water 1955 (ft.) Value of conductivity 6/5/56 7/2/56 8/6/56 9/17/56 Ave. 1956 Ave. depth to water 1956 (ft.) League of Nations 0-0.5 0.5-1 1-2 2-3 4.18 5.87 6.94 5.66 3.96 5.08 5.56 4.86 3.98 4.60 4.73 4.43 4.15 4.37 4.09 4.20 2.92 5.25 5.38 5 15 4.77 5.13 5.24 4.71 4.81 4.25 4.75 4.51 3.81 4.28 4.04 4.16 3.58 2.84 3.78 3.63 3.45 1.62 Frog Pond 0-0.5 0.5-1 1-2 2-3 3.95 4.10 3.94 3.99 3.47 3.60 3.07 3.38 3.31 3.47 2.99 3.25 3.29 3.21 2.89 3.13 2.92 2.49 1.92 1.96 1.96 2.39 2.26 1.78 1.78 2.02 1.98 2.25 1.83 1.83 2.35 2.48 2.02 1.70 1.70 2.32 2.43 2.29 Southwest Sump 0-0.5 2.88 2.98 3.68 3.18 2.50 2.80 3.03 3.38 3.14 3.08 1.59 0.5-1 2.67 2.45 2.62 2.58 2.10 1.98 3.04 2.70 2.45 1-2 2.84 2.72 2.64 2.73 1.80 2.09 2.55 2.79 2.30 2-3 2.94 3.02 2.95 2.97 1.94 2.20 2.05 2.83 2.25 Total Area 0-0.5 0.5-1 1-2 2-3 3.80 4.43 4.82 5.35 3.45 3.84 3.76 3.68 3.39 3.66 3.46 3.50 3.46 3.53 3.25 3.41 2.84 3.51 3.44 3.49 3.43 3.46 3.20 2.82 3.29 2.97 3.07 2.18 2.57 3.06 3.10 2.72 2.51 2.24 2.71 2.96 2.60 1.83 Table 15 DISTRIBUTION OF SALINITY IN SOIL PROFILES OF TULELAKE LEASE LANDS Area of sampling Lease No. Date Conductivity of saturated extracts mmhos/cm 0-H" WW lH"-3" 3 "-6" 6 "-12" 12"-18" 18 "-24" 24 "-30" 30 "-36" League of Nations . Frog Pond Southwest Sump.. 5 94 10 9/15/55 9/15/55 9/15/55 14.80 7.09 10.98 14.80 7.51 12.45 9.12 4.86 6.90 8.48 4.26 6.54 6.90 4.64 6.81 6.23 5.21 7.40 6.01 5.68 7.74 5.32 6.23 7.29 4.64 5.43 7 74 [25 soils, examination of table 14 indicates that apparently a saline condition exists in the lease lands, especially in the League of Nations. The monthly maps of salinity show that a large area of the League of Nations contains an unfavorable amount of salt. The leases most affected are 5, 7, 23, 25, and 36. The Frog Pond area appears less badly affected than the League of Nations but nevertheless has specific areas which have high conductivity values. The par- ticular leases affected are numbers 59, 63, 73, 93, and 101. The Southwest Sump also does not appear to be badly affected but because of the small number of sam- ples taken a fairly large sampling error exists. The leases most affected with salt are numbers 10 and 19. Salinity was lower in 1956 compared to 1955 because, as mentioned above, of the unseasonable amount of rainfall (16.67 inches July 1955-June 1956, com- pared to 9.97 inches mean for a 24-year period). The salinity maps also point out that the areas affected with salinity expand progressively throughout the summer. This is especially shown in the League of Nations. The distribution of salinity in the soil profiles at the end of the 1955 season (table 15) shows that the high- est concentration of salts is in the first 1% inches of the surface. In the sam- ples for lease 5, the salt grades con- tinuously down from the surface to the water table but in those from lease 94 salinity again becomes high in the 24--30- inch zone, near the water table. The pro- file of salinity for lease 10 in the South- west Sump shows a salinity reduction for the l ] /2 to 12 inch zones but the concen- tration again becomes high from 12 to 36 inches. In general, these profiles point out that evaporation from the shallow water tables prevailing throughout causes a build-up of salinity at the sur- face. If this surface concentration were maintained throughout the winter until spring planting, germination of the young seedlings would be retarded. For example, if the salinity were maintained at 12 mmhos, germination of barley would be reduced 40 per cent (Richards, 1954). Fortunately winter rainfall re- duces the salt content below the fall con- centration but, nevertheless, over a pe- riod of years, unless immediate preven- tive measures are instigated, the salinity will probably build up to the point where even winter leaching will not be effective. The soluble sodium percentage and exchangeable sodium percentage appear to indicate that in general in these areas the hazard of excess exchangeable so- dium is not great. For some soils an E.S.P. of 15 per cent is excessive. Few samples indicate a value of this magni- tude. It should be emphasized, however, that this value may be unsatisfactory for these soils. Additional field and labora- tory investigation would aid in establish- ing safe values as well as suggesting the particular type of reclamation needed. Effect of drains on salinity In order to determine more specifically the effectiveness of the present types of drainage systems for salinity control, studies were made on two locations in the lease lands. The first area investi- gated was in the Southwest Sump and was studied to determine the effectiveness of the surface drains servicing leases 4, 5, 10, and 11 in salt removal by flushing. The second area of investigation was lease 25 in the League of Nations, studied to determine the ability of the open ditches in the area to remove salt by leaching. Drains in the Southwest Sump are pri- marily surface drains designed to re- move surface water from the area after pre-irrigation and winter flooding. Since these drains are shallow no drainage is provided for ground water control. Con- sequently salt removal in the Southwest Sump is mainly accomplished by surface flushing. The area provides an example of the effect on the salt regime of long- term application of water with only sur- 26 face runoff and consumptive use as means of removing water from the area. Surface water from leases 4, 5, 10, and 11 in the Southwest Sump is removed by an open ditch bordering the eastern edge of leases 5 and 11, and by a parallel ditch bordering the western edge of leases 4 and 10. In addition, a third open ditch of somewhat larger capacity cuts the southwest corner of lease 10. Considerable winter leaching of salt took place in the Southwest Sump dur- ing the winter of 1955-56 because pre- cipitation during this period was un- usually high, and also because the area was flooded as a flood protection meas- ure. It was possible, consequently, to make a study of the effectiveness of the drains servicing leases 4, 5, 10, and 11 in removing the salt. Three parallel transects were made across the field between the drains, which are one mile apart. The center transect was made on the border between leases 4, 5, 10, and 11. The northern transect was made across fields 4 and 5, and the southern transect across fields 10 and 11 (see figure 6). Soil samples were taken at 0.166 mile intervals along each tran- sect for salt determinations. The depths of samples were 0-6 inches, 0-12 inches, 12-24 inches, and 24-36 inches. The con- ductivity values of the saturated extracts of these samples are shown in Table 16, and are shown graphically in figure 7. The results indicate that the average salt concentration was above 4 mmhos/ cm at every point across the field except immediately adjacent to the east side drain. However, at 0.33 mile from this drain the highest concentration of salt was found. The conductivity decreased as the west side drain was approached but even adjacent to this drain the salt was still above the allowable limit for sensitive crops. Study of the salt distribu- Figure 6. Sampling locations for detailed salinity study in Southwest Sump, 1956. K G 2 F 2 E 2 D 2 C 2 B 2 A. — X X X X X X X- I G F E D C B A -X X X X- X X X- O RECORDER I -X X X X X X X- G, F, E, D ( C, B, A, 10 Nk T~ O.I MILE -4- 0.1 MILE [27 tion throughout individual profiles shows that in most cases surface flushing was effective in removing salt only from the first six inches of soil. The drain cutting the southwest part of lease 10 has evidently contributed somewhat to ground water as well as sur- face water control; salinity was low in the vicinity of this drain, that is, the salt values at locations E 1? F x and G t were considerably lower at each depth than values for corresponding depths of samples to the north. In addition, the salt values for samples from the first 12 inches at all of these locations except E x are below the critical value of 4 mmhos per cm. These observations indicate that drains in the Tulelake lease lands designed to remove only surface water when coupled with surface flushing are relatively in- effective as a means of maintaining salinity within allowable limits for plant growth. However, where drains provide ground water control in addition to sur- face water control, the salinity can be maintained at a favorable level. In the League of Nations and Frog Pond, drains were designed to provide ground water control and, therefore, a study similar to the one in the Southwest Sump was made to determine the effi- ciency of these drains for lowering the water table after irrigation and for trans- porting leached salt from the area. The lease selected was number 25 in the League of Nations. An automatic stage recorder was placed at the south center of this lease and continuous records were obtained from June 4 to September 3, 1956. Soil samples were taken each month and salinity determined from the conductivity of the saturated extract. On June 23, the leasee irrigated this lease seeded only to barley. This irriga- tion raised the. water table more than 2 feet in three days. About three months were required for the water table to drop to the original level. The conductivity values of the soil samples showed that immediately after irrigation, salts in soil depth of 6-12 inches, 12-24 inches, and 24-36 inches were lowered, but not as much in the 0-6 inch depth. Only at the 24^36 inch depth was the conduc- tivity value lowered below the critical value of 4 mmhos/cm. The values ob- tained a month after the irrigation were Table 16 DKAINAGE-SALT INVESTIGATION IN THE SOUTHWEST SUMP Soil depth (inches) Location Depth A Ai A 2 Average 0-6 3.20 1.55 1.49 2.40 1.60 1.42 1.75 2.20 2.00 1.21 1.15 1.75 2.26 6-12 1.39 12-24 1.46 24-36 2.12 Depth B Bi B 2 Average 0-6 6.00 7.50 7.60 6.90 4.50 5.50 5.50 5.60 4.10 5.00 5.50 5.50 4 86 6-12. . . 6 00 12-24 . . . 6 20 24-36 . . 6 00 Depth C Ci C 2 Average 0-6 8.00 9.10 10.00 8.00 5.00 7.00 6.70 6.00 5.00 7.00 8.00 7.00 6 00 6-12 7 70 12-24 8 23 24-36 7 00 Depth D D! D 2 Average 0-6 . . 6.00 7.50 8.10 6.50 3.90 5.00 4.90 4.50 4.50 4.60 5.60 5.50 4 80 6-12 5 70 12-24 6 20 24-36 5 50 Depth E Ei E 2 Average 0-6 4.10 5.00 8.00 8.00 3.90 4.90 5.50 4.10 8.00 9.00 10.00 8.60 5 33 6-12 6 30 12-24 7 83 24-36 6 90 Depth F Fi F 2 Average 0-6 5.50 5.70 7.50 7.00 1.65 3.10 4.50 4.00 7.00 9.00 6.40 7.00 4 72 6-12 5 93 12-24 6 13 24-36 6 00 Depth G Gi G 2 Average 0-6 6.00 5.50 6.00 4.90 2.00 2.10 4.90 2.90 5.50 4.70 6.50 6.00 4 50 6-12 4 10 12-24 5 80 24-36 4.60 [28] 05 D DISTANCE - MILES T6\J Figure 7. Conductivity of saturated extracts (average of three transects, 0-36 inch depth) as a function of distance. Southwest Sump, 1956. as high as or higher than those before the irrigation, the greatest increase being in the 0-6 inch depth. To be beneficial for leaching, the water table must be lowered rapidly after water application, so the salt can be quickly transported away from the area. If this is not accomplished, and the water table stays close to the ground, surface evapotranspiration will result in reaccu- mulation of salt in the surface zones of soil. Any beneficial effects of leaching will be negated. The above situation ex- isted on this lease and shows that present drains are inefficient. Although the water table rose over 2 feet in three days as a result of irrigation, it took almost three months to lower it to the original level by drainage. The salt concentration was not decreased to an acceptable value at any depth and again reached a high value within a month or so. The inefficiency of the drains in the League of Nations and Frog Pond is not a result of low capacity; the cross-sec- tional areas are large enough to handle the volume of drainage effluent. The fault lies mostly with the fact that drains have not been pumped low enough to permit adequate ground water control because of three limiting factors : excessive length of ditches, flat gradient, and algae and moss growth in the ditches restricting flow. To remedy the situation these fac- tors would have to be eliminated but that cannot be easily accomplished. The ex- perimental work of 1957 and 1958 was aimed at finding satisfactory alternatives. The study also indicated the difficulty that will be experienced in salt removal by leaching of these soils. Although a fairly large amount of water passed through the 0-6 inch depth and although the irrigation water is of low salt content, the salinity in this depth was not appre- ciably reduced. Even though drainage is not adequate, the salinity should have been reduced further below the original levels than indicated by the data. It ap- pears, therefore, that to leach these soils effectively large volumes of water must be applied. This conclusion was also reached during the 1958 investigation (see page 54). Relationship between salinity and water table levels Where the surface layers of soil are badly affected with salt, one of the first factors considered is the position of the water table. The combined actions of evaporation and plant transpiration re- sult in a suction gradient which causes the movement of water and salts to the soil surface where the salts become con- centrated because of the evaporation of [29 the water. This is shown by examination of conductivity values of saturated ex- tracts of the first 6 inches of soil. The process of evaporation and concentration of salts at the soil surface is carried on most rapidly in the presence of a high water table. The relationship between salinity and water tables on the Tulelake lease lands for the summers of 1955 and 1956 is summarized in table 14, and the monthly salt-water table maps. The average values for the individual areas for the summer of 1956 are given in table 17. The League of Nations is developing into a problem area in respect to high water table levels and high salt concen- trations. As the maps indicate the re- gions of extremely high conductivity values coincide closely with the regions of high water table levels. Two apparent reasons for the high water table levels have been discussed before (see page 00) — inadequate drainage because of in- sufficient depths and capacity of the present drains; and poor irrigation effi- ciency. The new leasing program of the Bu- reau of Reclamation whereby the farm- ers lease the land for five years rather than one year should make the operators more conscious of the developing situa- tion but education on the proper use of irrigation water is still imperative. The situation is better in the Frog Pond area. There are certain leases and areas, indicated on the maps, where con- ductivities are in the harmful region. The correlation between salt and water table levels is not borne out so well on specific sites in this area as in the League of Nations. The average values, on the other hand, show that for the area as a whole the conductivity values are lower and the water tables deeper than in the League of Nations. The situation in the Southwest Sump is different from that on the other lease lands. The drain ditches in this area are designed primarily for removing surface water; consequently, water table lower- ^ ing is provided mainly by evapotrans- piration. Water is applied only in the i spring for pre-irrigation, so that the water table is originally high and de- clines slowly through the summer. The surface drains provide for removal of surface salt by flushing but this technique has not proven effective. Consequently, salt is being shuttled up and down in the profile year after year and a gradual build-up is developing. There were two areas (see maps on pages 20, 21, 22) with high levels of salinity during 1955 and 1956, and undoubtedly these areas will grow. Improving drainage by means of ditches or tile designed to control ground water levels and remove leaching water are necessary to avoid a serious salt problem in future years. Discussion The average values for conductivity of the saturated extract of soil samples and average values for depth to water for the three main areas of the lease lands, 1955 and 1956 seasons, are shown in table 14. The values for the total area are also indicated. Study of this table shows that for all three areas a slight decrease in conductivity values occurred at every depth in 1956 compared to 1955. On the other hand, the water table levels for each area were high during the sum- mer of 1956 compared to summer 1955. The differences can be largely accounted for by reviewing precipitation records for the Tulelake area (table 10). For the summer of 1955 the salt values could be expected to be fairly high be- cause winter leaching had been low. Sim- ilarly, the water table levels were fairly low because of small accretion from pre- cipitation. On the other hand, because of the excess precipitation in 1955-1956, almost Yz foot above the long-time mean, considerable winter leaching of salt took • place, and the water table levels were high most of the season. < [30] Contributing to low salt values on the Southwest Sump was the fact that during the winter of 1956 dikes protecting the area from the water in the restricted sump broke, and flood water covered the area. As much as 5 feet of water stood over some areas. Several auxiliary pumps were brought in and the water was quickly removed, so that rapid removal of salts resulted in a general lowering of salt values for the area. Conclusions The first two summers of drainage- salinity investigation on the Tulelake lease lands have revealed a saline type condition throughout much of the area. In addition, a sodium hazard appears to be developing at some specific locations. The League of Nations in particular is affected, and local areas in the Frog Pond and Southwest Sump are also developing salinity problems. Drainage studies have shown that a high water table prevails throughout Table 17 AVERAGE VALUES OF SALINITY AND WATER DEPTH, SUMMER 1956 Area Conductivity EC X 10' (0-6 inches) Depth to water (ft.) League of Nations Frog Pond Southwest Sump 5.13 2.19 3.08 1.62 2.29 1.59 most of the growing season, responsible for the developing salinity problem in the surface soil. Present drains appear to be, on the whole, ineffective in con- trolling the water table. Studies on yield and vegetative growth of barley have shown that while vegeta- tive growth is prohibited by moderate quantities of salt, yields are not restricted greatly. Nevertheless, if the salinity is allowed to build up to high levels even the barley yield will be reduced consid- erably. Potatoes, the other crop widely grown, are restricted in growth by even moderate salt concentrations. Study of Deep and Shallow Open Drains, 1957 To make a more detailed study of drain- age methods to be used on the lease lands, the Bureau of Reclamation decided to dig one large deep open ditch and two shallower open ditches. The objective of the study was a comparison of the deep ditch with the shallow ditches to find out which was the more effective in lowering the water table. Construction Problems The large open drain was dug on the north-south boundary between leases 3, 4 and 9, 10 in the Southwest Sump. The two shallow ditches were about 2,400 feet long and ran east and west on lease 5. Observation wells were installed in a grid pattern about both deep and shallow drains. Eight automatic water stage re- corders were installed on a transect per- pendicular to the large ditch. Because the soil of the lease lands be- comes semi-fluid when saturated with water, construction techniques become [31 very important to insure a stable ditch of uniform cross section. The large ditch was carefully dug to a depth of 8 feet during the winter months and remained full of water until the following spring. A sump pump was placed at one end of the ditch and the water was pumped out of the ditch and into a higher-level ditch from which it was pumped into the lake. The ditch was constructed with a dragline and had a berm of 15 feet. In the spring of 1957, the pump was turned on at a low rate of speed in order to slowly lower the water in the ditch. The water level gradually receded until it reached the bottom of the ditch. At that time the ditch began to fail. The sides of the ditch slumped ver- tically, leaving large cracks in the ground and simultaneously, the bottom of the ditch rose vertically (see figure 8). The soil movement was so gradual that it appeared to be imperceptible to the eye. The entire movement took place over a period of several hours. It resembled the flow of molasses. The failure of the ditch was caused, not by the slumping of the sides into the ditch, but by the movement of the semi-fluid soil material according to well-established laws that govern such movement. The problem arose because the water was removed from the ditch while the water table in the adjacent soil was close to the soil surface. This meant that the water in the soil adjacent to the ditch stood at a level of about 7 feet above the bottom of the ditch. The weight of the water was exerted downward in such a way that it caused the bottom of the ditch, which lacks stability, to rise. Observations in the Tulelake area indi- cate that this is a common reason for the failure of ditches. It also indicates the need for a great deal of care when ditches are dug 4 or 5 feet or more below the ground surface. A possible alternative method of con- struction would be a repeated deepening of the ditch. The ditch is initially dug to, Drain in Southwest Sump (1957). Experimental section is beyond the sump and pump. Drain is full of water from winter floods. ,-*,-, ■ .-..: SPOIL Figure 8. Diagram of ditch failure following removal of water from ditch. Top diagram shows ditch before water is removed. Bottom diagram illustrates movement of soil following removal of water. Sides of ditch move vertically downward and bottom of ditch moves upward. say, 3 or 4 feet below the ground surface. After the water table has drained to the bottom of the ditch, equipment is brought in to deepen the ditch another foot or so. The water table is again allowed to drop below the bottom of the ditch, and equip- ment is again brought in to deepen the ditch still further. Other field observations have indi- cated that ditches as shallow as 4 or 5 feet will fail in a similar manner if water is allowed to pond close to the edge of the ditch. The situation could occur where an irrigation canal runs adjacent to a drainage ditch. The pressure of the water in the canal will cause the drainage ditch to fail. After the failure of the open drain, the equipment was brought back and the ditch was deepened once again, so that at the end of the construction period the ditch stood about 6 feet deep below the ground surface. Effect of Deep Drains on Water Tables Perpendicular to the ditch at three different locations a line of observation wells were installed. On the center line of these observation wells automatic re- corders were installed at various loca- tions to provide a continuous record of water table fluctuations adjacent to the ditch. Records obtained from the observation wells and automatic recorders indicated the position of the water table on differ- ent dates during the growing season. The most noticeable feature of these charts was the flatness of the water table. Drawdown near the ditch occurred within 10 to 15 feet from the ditch and there was very little curvature to the water table. Two conclusions can be drawn from the observations adjacent to the single deep drain. 1. The surface soil layers have a high hydraulic conductivity. This was confirmed by other tests made using the piezometer-method of determin- ing the hydraulic conductivity. 2. A single drain does not provide the information needed for the design of a drainage system in such soils. [33] 2HI mm I mMM Collapse of experimental drain. Note how bottom of ditch has risen. While the experiment did not indicate anything about the drain spacing, it did point out the difficulties of constructing and maintaining open drains in the Tule- lake area, and it also pointed out some techniques that can be used to establish a system of open drainage ditches on the lease lands. Shallow Drains The shallow drains which were placed about 4 to 5 feet below the ground sur- face did not encounter the same type of difficulty encountered with the deep open drains. According to the records, the shallow drains were quite effective in lowering the water table rather quickly in the spring. While the instrumentation on the shallow drains was not as com- plete as on the large deep drains, the results indicate that the shallow drains lowered the water table two to three times as fast as the adjacent large deep drain. Observations made in the fall showed that the water table was well below the bottom of the shallow drain, indicating that the large deep drain may have had some influence on lowering the water table over the entire area. Another factor influencing the drop of the water table would be crop use of the ground water. The crops extracting water from the soil will cause the water table to drop as this water-table water is used by the plants. Conclusions The studies of 1957 showed that the use of deep drains (6 feet or deeper) is limited due to the unstable soil condi- tions that prevail in the Tulelake area. The use of shallow drains (4 to 5 feet deep) is probably feasible for the lease lands. In addition, the results indicated that conclusions regarding drain spac- ings cannot be based on the observa- tions adjacent to a single drain. [34] Drainage— Salinity Investigation, 1958 The drainage investigation in 1957 showed that deep drains are difficult to maintain because of the inherent insta- bility of the soil. Techniques other than those used in 1957 will have to be devel- oped if open drains are to be used on the lease land. Because of the failure of the open drains in 1957, the 1958 studies investi- gated the use of tile drains as a means of controlling the high water table and of removing leaching water used for re- duction in salinity. In addition, it was felt that results of a drainage-salinity study on an individ- ual lease would lead to some general recommendations for reclamation of other affected areas in the lease lands. For these reasons, a 78-acre lease unit with a high level of soil salinity and high water table was obtained by the U. S. Bureau of Reclamation for experimental use in 1958. The objectives of the drainage investi- gation were to determine for the Tulelake lease lands : The most efficient spacing of tile drains. The most efficient and practical tile size. The most efficient and practical type of backfill material. To achieve these three objectives it was necessary to determine: The water table conditions throughout the field by means of observation wells and automatic recorders. The rate of drawdown of water table and the distance water table is lowered with drainage. The quantity of effluent handled by each drain. The objective of the salinity investiga- tion was to relate removal of salt from the soil to volume of water applied, vol- ume of water removed by the tiles, and methods of leaching, without cropping. To achieve this objective it was necessary to determine: The volume of leaching water applied and volume removed. Salinity conditions throughout field prior to leaching. Salinity at approximately midpoint of leaching treatment. Salinity at end of leaching treatment. Drainage Investigation Procedures and materials. The effect of tile drains as a means of controlling the water table was studied at the 78-acre lease. A system was de- signed to obtain information on tile sizes, spacing, and backfill material. Because the area available for this study was lim- ited, it was not possible to conduct the experiment with the controls usually established in an investigation of this kind. In addition, two of the tile lines developed physical difficulties during the 35 season so that the measure of control attempted was upset. The results should, therefore, be studied from a qualitative rather than a quantitative viewpoint — as a demonstration of the applicability of tile lines as a means of drainage rather than one giving exact information that could be statistically analyzed. Six tile lines were installed at a depth of about 5 feet in the test lease in March 1958 with the following variables (see figure 9) : Spacing. Three different spacings were included : 220 feet, 440 feet, and 660 feet. Lines A and F were intended to act as guard lines cutting off flow into the area from the south and north, respectively, and flow out of the area during leaching trials. Tile size and length. Three variable tile sizes were used: 4-inch, 6-inch, and 8-inch. The length of each line was 1,200 feet, but two lines were split into two 600 feet tile sizes. Tile materials. Two different materials were tested for use as drain tubes: bi- tuminous fiber and nonreinforced con- crete. The bituminous fiber pipe con- tained two parallel rows of %-inch holes 100 degrees apart to permit water entry into the line. Eight-foot lengths of fiber pipe were used with adjacent lengths joined by means of couplers. The fiber pipe used to make road crossings was heavy duty and unperforated. The concrete pipe was of tongue and groove construction. This type of pipe provides a better alignment between ad- jacent tile than with square-cut end tile. Water enters the line through the gap between the tongue and groove of adjoin- ing tubes. The length of individual pipe in this experiment was 3 feet. Backfill material. Three types of trench backfill material were tested: Type A (a) Topsoil placed over tile to a mini- mum depth of 24 inches above bottom of ditch, obtained adjacent to trench or separated from unclassified ditch excavated material, (b) Unclassified excavated material placed on top of topsoil and used to fill remainder of ditch. Type B (a) Pit-run volcanic cinder placed in trench 4 inches below and 4 inches above the tile. (b) Excavated material used to fill re- mainder of trench. Type C (a) Excavated material used to fill en- tire trench. Filter material. Twelve hundred feet of type "A" or "C" drain contained a 12- inch-wide strip of fiberglass filter placed above and an 8-inch strip placed below the tile. Slope. The slope of all lines was the same, 0.001. The specifications of the individual tile lines are given in table 18. Tile outflow measurement. In an inves- tigation of the effect of tile drains on water table levels and salt removal by leaching it is important to obtain a meas- urement of the quantity of effluent. The simplest method of measurement consists of using a bucket and a stopwatch ; more complex arrangements make use of weirs or slotted tubes in conjunction with auto- matic recorders. For the tile drainage in- vestigation at Tulelake slotted tubes were used. The slotted-tube device was devel- oped and tested at the Minnesota Agri- cultural Experiment Station and reported by Larson and Hermsmeier (1958). It consists of a tube in which is milled a slot. The tube is mounted vertically on the end of the tile line with the slot directed upstream. A hose connects the bottom of the tube to a stilling well on which is mounted an automatic recorder. The water level in the well measured from the pipe invert is the index of discharge. The head measured in this manner is [36] XL" - if-4— tto'— f-i © [® © I - — h I ®d e I I Til ® o ^ I N-« l»Tl*AL 1 Ifll •-• « WUN ♦• ► J® 0' t M-T LATCRAL 1 101 t-6-C MUM H4ff $ ftCCOftOCR STATKM MANHOU Figure 9. Plot layout for Tulelake study, 1958. North is to the right. approximately equal to the depth plus velocity head of water flowing in the pipe which, in turn, corresponds to the total flow energy. Calibration curves may be obtained by mounting each slotted tube on a test pipe in a hydraulic lab- oratory and plotting head values corre- sponding to a range of flow values. If automatic recorders are used with the tubes mounted on the tile line in the field, a continuous record of heads is obtained which can be converted into continuous flow values by means of the calibration curve. A slotted-tube device with stilling well and Stevens Type F automatic recorder was mounted on each tile line except line F during the period of the investigation. The recorders were equipped with weekly clocks. The relation between head in the stilling well and the pipe invert was de- termined by surveying with level and rod. The weekly records of head values were converted to flow values by means of the calibration curve obtained for each slot- ted tube. Results Tile spacing. From July 29 to August 18, leaching of the test plots was discon- tinued to permit determination of the effect of tile spacing on drawdown of the water table. The water table profile for this period is shown in figure 10 and tile outflow values shown in figures 11 to 15. On July 31, the water table was still at the surface on leases 2, 3, 6A, 6B, 6C, 6D, 6E, and 7A as a result of irrigation. The water table shows a slope toward tile D from both directions, indicating that it was receiving flow from both the north and south irrigated areas. The tile flow values at this time show that line A was handling the largest volume of water and was flowing full and under pressure. Line C was essentially inoperative. On August 1, the water table was still at the surface on plots 6A and 7A and sloped to the south toward line D where the depth to water was about 3.25 feet. The water table on plots 2 and 3, on the other hand, had dropped to about 2.5 feet below the surface and sloped from observation well 5 to the approximate water table midpoint between lines A and D north toward line D and south toward line A. The flow data indicate that lines A and D were handling the largest volumes of water. The flow for line A at this time was 37 gpm lower [37] Table 18 SPECIFICATIONS OF TILE LINES IN TULELAKE TEST PLOT- -1958 Tile line Length (ft.) Tile size (in.) Tile material Backfill material Slope A B 1,200 600 600 1,200 600 600 1,200 1,200 4 6 4 4 8 6 6 8 Bit. fiber Concrete Bit. fiber Bit. fiber Concrete Concrete Concrete Concrete TypeB TypeC TypeC Type A TypeB TypeB Type A Type A 0.001 0.001 C 0.001 0.001 D 0.001 E 0.001 0.001 F 0.001 than on the previous day because of the drop in head. On August 3, the water table in plots 6A and 6B had dropped to 2.00 feet below the surface and the water table gradient was still toward line D. Near lines A and D the water table was about 3.50 feet below ground surface, and about 3.00 feet below the surface at the midpoint. The midpoint between the lines had now moved toward line D. Lines D and A were handling the largest volume of flow but the quantities were less than on the previous date commensurate with the lower head. Line B was still removing a fair quantity but by no means the amount that could be removed if the line were functioning properly. Discharge of drain pipe during leaching. A slotted tube is used to measure the discharge. (Because the tile line is carrying water under pressure, the slotted tube shown here is not oper- ating properly.) §J 5 0© 500 600 700 800 900 1000 1100 DISTANCE FROM TILE LINE A (FEET) 1200 1300 1400 1500 1600 @ TILE LINE © OBSERVATION WELL Figure 10. Drawdown of water table by lines A, B,C, D, and E during period July 31 to August 18, 1958, after cessation of irrigation on test plot. By August 5, the water table in plots 6A and 6B had dropped an additional 0.65 feet to a depth of 2.65 feet. The slope was still toward line D. The water table was about 3.60 feet below ground surface near lines A and D and about 3.10 feet below the surface near the midpoint. Flow had decreased 75 and 100 gpm in lines A and D, respectively, from the previous date. As figure 10 indicates the water table fell quite slowly after August 5. On Au- gust 18, the day before resumption of irrigation, the water table on plots 6A and 6B was about 3.35 feet below the surface. The water table in the vicinity of line A, according to the reading in well 3, was about 4.15 feet below the surface. Near Line D the level was about 3.75 feet and level at midpoint about 3.45 feet. The flow in all lines was very low, with line A removing the largest volume. The rise in flow in line A at this time was caused by irrigation in the lease to the south. Tile outflow data. The total quantity of water applied to the plots and the total quantity removed by the individual tile lines (except line F) are given in tables 19 and 20. In addition, the outflow hy- drographs for each line are shown in figures 11, 12, 13, 14 and 15. The flow hydrograph for line F was not obtained and, therefore, the total shown in table 20 is short the amount needed for this line. Nevertheless, the quantity of ef- fluent from the individual lines can be compared with the amount of water ap- plied to the individual plots, to obtain an estimate of the efficiency of the var- ious lines. Thus, on the plots between lines A and C, 75.8 acre-feet were ap- plied. Line C was inoperative, and lines A and B removed 38.70 and 16.80 acre- feet respectively. The total for these two lines was 55.0 acre-feet. The remainder was probably removed by line D or by side losses. Line B was partially plugged so that line A removed the greater quan- tity. Between line C and about halfway between lines D and E, 65.3 acre-feet were applied. Line D removed 73 acre- feet, indicating that it probably handled part of the flow that should have been removed by lines C and E. From the ap- proximate midpoint between lines D and E and lines E and F, 56.4 acre-feet were applied. Line E removed only 20.4 acre- feet, showing that either this line was partially plugged or that the type of backfill did not permit effective drainage. Automatic recorder data. The graphs of fluctuation of water table obtained from the automatic recorders are pre- sented in figures 16, 17, 18, 19 and 20. Figure 16, for recorder No. 1, situated midway between lines A and B. shows that prior to the beginning of irrigation. the water table had receded to a depth of 4.0 feet below ground surface. During [39] Table 19 QUANTITY OF WATER APPLIED, TULELAKE EXPERIMENT PLOT 1958 Plot No. J A IB Flume Date of irrigation July 22. July 25. July 28. August 19 . August 25. August 28. . . September 1 Total. July 7.. July 8.. July 21. July 25. July 28..., August 18. August 25 . . . August 28 . . . September 1 Total . July 7.. July 8.. July 18. August 29. Total. June 30. July 3.. July 17. August 27. Total . Acre-ft. applied 1.373 4.100 1.624 0.067 1.084 2.718 0.787 0.941 5.785 1.678 2.378 4.075 26.610 0.696 0.501 4.998 5.824 3.465 3.575 1.465 1.300 2.530 2.204 4.201 3.606 2.510 3.174 3.629 3.093 2.401 49.172 0.251 0.339 2.433 2.941 1.276 1.816 2.875 2.383 14.304 0.951 2.933 4.350 2.672 3.249 4.147 3.869 23.390 Plot No. 5A 5B 6A 6B 6C 6D 6E 7 A Flume Date of irrigation June 30... July 16..., August 26. Total June 27 July 15 August 22... September 2. Total July 10 July 29 August 20. . . September 2. Total. July 11... July 29... August 20. Total. July 11.. July 29.. August \i Total.. July 11 ... . July 25.... July 28.... August 19. Total.... July 12... July 14... July 25... July 29... August 19. Total . July 12. July 14. July 29. Total . Total water applied, 229.733 acre-feet. Table 20 QUANTITY OF EFFLUENT FROM TILE LINES— TULELAKE EXPERI- MENTAL PLOT— 1958 Tile line Quantity of effluent (acre-feet) Tile size (inches) Tile length (ft.) A 38.20 16.80 4.46 72.66 20.40 4 6 and 4 4 8 and 6 6 8 1,200 B C D E F 1,200 1,200 1,200 1,200 1,200 Total 152.52 the shut-off period, August 1-18, the water table reached an equilibrium depth of about 3.50 feet. The difference in equi- librium values indicates the possibility of flow into the area from outside areas during the irrigation season. The average slope on the drop-off curves was about 0.9 feet per day to a depth of 3.0 feet showing that it would require about three days for the water table to drop to a depth usually considered safe for crop growth, and prevention of salt transport by evaporation. The shape of the curves obtained from recorders 2, 3, and 4, located between lines C and D are fundamentally the same as that for recorder 1. However, it is possible to determine the direction of slope of the water table from the differ- ences in heights of the peaks. For ex- ample, on August 18 and 19, after the shut-off period, an irrigation was applied to plots 2 and 3. The peaks produced on the recorders were as follows: No. 1, 0.80 feet; No. 2, 1.58 feet; No. 3, 2.35 feet; No. 4, 2.75 feet. Since the depth to water increases from south to north, it is evi- dent that a slope occurs toward line D and that the area of influence of this line is great. The average slope of the drop- off curves of these recorders was about 1.10 feet per day, again showing that the water table is rapidly lowered to a safe level. Figure 11. Discharge (gpm) from tile line A, 1958. The average drop-off slope for re- corder No. 5, located midway between lines E and F, was 0.60 feet per day. Five days would be required for the water table to drop to a safe level. The slow rate of drop may have been caused in part by partial plugging of line E. Discussion Installation. The failure or partial fail- ure of some of the tile lines in the 1958 drainage investigation points out the need for selection of the most favorable installation time. The test lines were in- stalled in March — the time of a high water table. Since the tile was installed below the water table, it was difficult for the workers to be certain of proper joints between adjacent tile. It appears, there- fore, that the best time of installation would be in the late summer or early fall, immediately after harvest, when the water table is naturally low. In addition, it appears advisable to dig all the outlets first, particularly if they are to cross roadways which are compacted to the extent of that of the test plot. This will [41 Figure 12. Discharge (gpm) from tile line B, 1958. Table 21 AVEEAGE RATES OF DROP OF WATER TABLE AUTOMATIC RECORDERS AND OBSERVATION WELLS No. Average rates of drop (ft. per day) No. Average rates of drop (ft. per day) Recorders Observation 1 0.9 2 1.0 14 0.5 3 1.6 15 0.4 4 0.85 16 0.6 5 0.6 17 0.2 18 0.25 Observation 19 2 wells 20 0.3 1 1 1 21 0.45 0.7 2 22 3 1 2 23 24 0.8 8 1 1 2 1 8 25 1 3 6. . . 8 26 0.9 0.8 7 1.1 27 8. . 9 28 0.8 8 g 0.8 97 29 10 30 1.4 0.85 0.9 II 1 8 31 12 0.8 0.8 32 13 i — ■ — i — ■ — i — ' — r 1 i — I — r 20 2S 30 5 10 15 20 25 31 5 10 15 20 25 31 5 10 15 Figure 13. Discharge (gpm) from tile Hue C, 1958. provide lowering of the water table prior to installation. Tile line spacing. The partial plugging of lines B and C limits evaluation of range of influence of the various sized lines and effectiveness of tile size and backfill. The irrigation shut-off period, August 1-18, however, permitted obser- vations of a general nature. Line A, the 4-inch bituminous fiber line with Type A backfill, flowed full and under pressure during the time that adjacent plots were leached. After water application ceased, the water table fell at a fairly rapid rate, and the midpoint of the water table be- tween lines A and D, if lines B and C are considered practically inoperative, was about 450 feet north of line A. This would imply that if 4-inch lines were in- stalled, a spacing of about 900 feet would provide effective drainage. To permit a measure of safety, however, it would [42] TOOi 1 r 600- soo- ■>! i ■ i ' 1 ' 1 ' 1 400- i 1 • 300- 1 i| ■ 200- II iiy - 100- \ 50- 1 I V u 1 1 1 1 1 — - , i Figure 14. Discharge fgpm) from tile line D, 1958. probably be better to install lines at about 600 feet spacing. Line D, the 6- and 8-inch concrete line with type B (volcanic cinders and exca- vated material) backfill appeared to be the most effective line in operation. This is evident from the tile effluent data which show that it removed the largest volume of water, and from the draw- down curves during the shut-off period. The volume removed by this line was almost twice as much as line A, the next most efficient line. The midpoint of the water table between this line and line A was about 450 feet to the north of A; however, the water table slopes indicated a movement toward line D for about 600 feet. This would indicate that if 6- and 8- inch combination tile lines were installed, a spacing of 1,200 feet would permit ade- quate water table control. Again, for safety, however, a spacing of 600-800 feet would appear advisable. i — • i ■ i Figure 15. Discharge (gpm) from tile line E, 1958. Rate of drop of water table. The effec- tiveness of the tile lines in lowering the water table after irrigation is best seen by comparing them with open ditches. For example, the study on lease 23 for 1956 showed that, where open ditches were used, almost three months were re- quired for the water table to return to the original level after an irrigation. The experimental tile lines, on the other hand, were capable of lowering the water table to 3 feet in three to five days. Salinity Investigation Procedures Plot division. The 78-acre lease was di- vided into 16 small plots for ease of irrigation and sampling. The size and arrangement of the plots are shown in figure 9. Because of the high infiltration rate of the soils each plot was divided into two halves by a border at midlength. to obtain complete coverage. In this way the upper half of the plot was allowed to fill before the water ran onto the bottom half. Irrigation supply and equipment. The irrigation water for the investigation was obtained from the N canal at a head gate: [43] JUNE JULY AUG. SEPT. Figure 16. Water table fluctuations at Recorder No. i, 1958. Figure 17. Water table fluctuations at Recorder No. 2, 1958. Figure 18. Water table fluctuations at Recorder No. 3, 1958. I it provided water to the supply lateral serving only the test lease and lease 32 to the south. Water was obtained on de- mand, except when water was desired for irrigation of lease 32. A head ditch was constructed on the east side of the lease and water diverted into it from the adjacent parallel supply lateral through a turnout structure at the northeast cor- ner. A coarse adjustment of head was obtained at the N canal head gate, and fine control at the gate in the turnout structure. Because seepage from the head ditch into the plot would have prevented ac- curate measurement of volume of water applied, a plastic film ditch liner was installed in 1,200 feet of the head ditch above the portion of the field to be heav- ily irrigated. A special liner was con- structed consisting of a 54-inch wide, 2.85 mil thick green vinyl film edge-glued to a 54-inch wide, 1.85 mil thick white vinyl film. The final width of the liner was about 9 feet. The thicker green por- tion of the liner was laid down on the west side of the ditch where the Parshall measuring flumes were later installed. Eleven Parshall flumes were installed in the ditch bank so that accurate meas- urements could be made of the quantity of water applied to each test plot (see photo on page 46). The throat width of these flumes was 9 inches. For measurement of head, one staff gauge was mounted on the side of each flume at the up- stream end, and another on the side at the bottom of the inclined portion. With these measuring devices only meas- urement of the upstream gauge was needed with "free flow," that is, when elevation of water in the downstream end of the throat section was not high enough to cause restriction of flow be- cause of the development of backwater. During "free flow" head values were converted to flow values by use of table 8 in University of California Circular 473 (Scott and Houston, 1959). However, 44 when the flow was great enough to cause submergence (when the ratio of the down- stream to upstream heads exceeded 0.7 j , a correction was applied, using the cor- rection factor from Circular 473. To ob- tain a tight fit between the flumes and the vinyl ditch lining, the ditch bank was cut to form-fit the flume and the lining cut and folded back into the cut. The flume was then placed in the cut, and sand bags and soil tamped into the space between the outside of the flume and the cut. Water flow into the flumes was con- trolled by gates in the flumes themselves and also by canvas dams in the head ditch. During the time each plot was leached, frequent measurements were taken of the heads in the flumes so that a fairly accurate determination of the volumes of water applied could be made. Leaching operation. Each of the test plots was leached several times through- out the season; however, most of the water was applied to the southern plots where the salinity was the highest. It was not possible to keep plots continually flooded for a long period of time, al- though several plots were leached for a period of two or three days. Soil sampling. Soil samples for salt de- termination were taken three times dur- ing the period of the investigation. The first samples were taken in the vicinity of each observation well in March 1958, immediately before installation of the tile lines. The second set of samples was taken near the observation wells and at other locations in each plot in mid-July, after several plots had been irrigated. The third set of samples was taken in each plot near the wells and at additional locations in September, after all irriga- tion had ceased. A fourth and fifth samp- ling was made in April and September 1959. The soil samples were analyzed for salt content by the conductivity of the saturated extract method. Water sampling. Twenty-five times during the period of the investigation Figure 19. Water table fluctuations at Recorder No. 4, 1958, Figure 20. Water table fluctuations at Recorder No. 5, 1958. water samples were obtained from the tile outlets. A complete chemical analysis was made on some of the samples: for the others only conductivity was de- termined. In addition to the tile water samples others were obtained at N canal at the turnout gate. Determination of effectiveness of leach- ing. The data on quantity of water ap- plied, together with salinity data at be- ginning, midseason, and end of the [45 Venturi-type flume used to measure the amount of leaching water applied to the field. leaching trials, provide a means of de- termining the efficiency of the leaching technique. The ratio of salt concentra- tion in the soil solution after a period of leaching, to initial salt concentration (C/C ) is made for each sampling site. This ratio is plotted against a second ratio, depth of leaching water used to reduce salinity from C to C per unit depth of soil. The resulting curve pro- vides an indication of the depth of water per unit depth of soil required to reduce the salinity to any given level. To obtain concentration units (meq per liter) for conductivity units (mmhos per cm) it is necessary to plot the cor- responding values for a number of sam- ples on which complete analyses have been made. Knowing the conductivity values for saturated samples it is then possible to determine the concentration values by consulting the graph. The ratio of the depth of leaching water to the depth of soil is obtained [ from a knowledge of the amount of water applied, area of the plots, and depth of leaching. Effect of rainfall. Some leaching results from rainfall during the winter months. In some areas this leaching represents an important contribution in maintain- ing a favorable salt balance in the root zone. To evaluate the leaching by rain- fall, soil samples were collected in the spring of 1959 for analysis at the sites of previous sampling. During the 1959 crop year, barley and potatoes were grown in the area of in- vestigation. Soil samples were again taken in the fall of 1959 to estimate changes in soil salinity following crop- ping. Results Relationship between volume of water applied and volume removed. The total amount of water applied to each plot 46] during the 1958 investigation is given in table 19, and the total volume removed by the tile lines in table 20. A greater quantity was applied to the southern plots because of the higher salt content in this area. As mentioned before, the total volume of tile effluent is short by the amount removed by line F. If it is assumed that this line removed an amount equal to that of line E (20 acre- feet), the total removed is still about 60 acre-feet less than the total applied. Reasons for the difference may have been difficulties in measuring tile outflow (see Table 22 TULELAKE DRAINAGE-SALINITY STUDY CONDUCTIVITIES OF SATURATED EXTRACTS— TULELAKE, CALIFORNIA, MARCH, 1958 Sample No. Depth (ft.) mmhos/cm Sample No. Depth (ft.) mmhos/cm Sample No. Depth (ft.) mmhos/cm 1 0-1 7.73 12 0-1 3.51 23 0-1 1.24 1-2 7.19 1-2 5.25 1-2 4.08 2-3 6.88 2-3 5.32 2-3 3-5 6.72 3-5 5.11 3-5 2^85 Ave. 7.13 Ave. 4.80 Ave. 2.73 2 0-1 13 0-1 2.39 24 0-1 4.96 1-2 7.93 1-2 4.50 1-2 4.43 2-3 7.73 2-3 4.64 2-3 4.49 3-5 7.42 3-5 5.13 3-5 4.64 Ave. 7.69 Ave. 4.16 Ave. 4.63 3 0-1 1-2 10.41 9.71 14 0-1 1-2 2.06 4.95 25 0-1 1-2 3 '23 2-3 8.09 2-3 4.61 2-3 3.65 3-5 8.25 3-5 5.10 3-5 4.18 Ave. 9.11 Ave. 4.18 Ave. 3.69 4 0-1 1-2 4.99 10.45 15 0-1 1-2 1.97 3.36 26 0-1 1-2 2 03 3.45 2-3 7.58 2-3 4.33 2-3 3.37 3-5 8.86 3-5 5.10 3-5 4.05 Ave. 7.97 Ave. 3.69 Ave. 3.22 5 0-1 1-2 7.31 8.40 16 0-1 1-2 2.04 4.83 27 0-1 1-2 4 50 4.56 2-3 6.49 2-3 4.00 2-3 4.61 3-5 7.10 3-5 4.86 3-5 4.74 Ave. 7.32 Ave. 3.93 Ave. 4.60 6 0-1 7 29 17 0-1 5 58 28 0-1 4 47 1-2 6.97 1-2 3.75 1-2 3.96 2-3 7.35 2-3 4.88 2-3 5.06 3-5 6.34 3-5 5.00 3-5 Ave. 6.99 Ave. 4.80 Ave. 4'o0 7 0-1 6 49 18 0-1 2 14 29 0-1 5 50 1-2 7.21 1-2 2.62 1-2 3.61 2-3 5.69 2-3 2.64 2-3 4.30 3-5 6.27 3-5 2.92 3-5 5.24 Ave. 6.41 Ave. 2.58 Ave. 4.66 8 0-1 3 92 19 0-1 1 03 30 0-1 5 19 1-2 7.19 1-2 2^06 1-2 477 2-3 5.91 2-3 2.58 2-3 4.18 3-5 6.11 3-5 3.51 3-5 5.05 Ave. 5.78 Ave. 2.29 Ave. 4.80 9 0-1 2 88 20 0-1 1 33 31 0-1 5 54 1-2 5.10 1-2 2^32 1-2 5.94 2-3 5.37 2-3 2.55 2-3 4.88 3-5 6.02 3-5 3.61 3-5 5.10 Ave. 4.84 Ave. 2.50 Ave. 5.36 10 0-1 2.34 21 0-1 1 06 32 0-1 6 84 1-2 1-2 2.81 1-2 6>0 2-3 5.12 2-3 1.89 2-3 5.34 3-5 5.92 3-5 3.75 3-5 5.00 Ave. 4.46 Ave. 2.38 Ave. 5.97 11 0-1 2 62 22 0-1 3 17 1-2 4.88 1-2 4^54 2-3 5.06 2-3 3.15 3-5 5.37 3-5 3.18 Ave. 4.48 Ave. 3.51 [47 Table 23 CONDUCTIVITIES OF SATURATED EXTRACTS OF SAMPLES COLLECTED SEPT. 1958 Sample No. Depth (ft.) mmhos/cm Sample No. Depth (ft.) mmhos/cm Sample No. Depth (ft.) mmhos cm SI 0-1 1-2 2-3 3-4 Ave. 7.69 6.53 6.04 6.68 S9 0-1 1-2 2-3 3-4 Ave. 3.19 4.54 4.47 S17 0-1 1-2 2-3 3-4 Ave. 6.54 5.81 4.15 4.91 6.73 4.07 5.35 S2 0-1 1-2 2-3 3-4 Ave. 12.75 11.59 8.46 6.45 5.66 S10 0-1 1-2 2-3 3-4 Ave. 1.25 2.83 4.21 3.95 S18 0-1 1-2 2-3 3-4 Ave. 2.00 2.09 1.96 1.84 3.06 1.97 8.98 Sll 0-1 1-2 2-3 3-4 Ave. 1.32 3.14 3.73 4.85 S19 0-1 1-2 2-3 3-4 Ave. 1 01 S3 0-1 1-2 2-3 3-4 Ave. 3.16 4.53 3.36 4.76 1.25 1.76 1.69 3.26 1.43 3.95 S12 0-1 1-2 2-3 3-4 Ave. 5.88 4.13 4.51 5.26 0-1 1-2 2-3 3-4 Ave. 0.85 1.11 1.59 1.99 S4 0-1 1-2 2-3 3-4 Ave. 3.91 5.76 5.20 5.00 4.94 1.38 4.97 S13 0-1 1-2 2-3 3-4 Ave. 2.02 3.51 5.00 4.49 4.40 0-1 1-2 2-3 3-4 Ave. 0.874 1 00 S5 0-1 1-2 2-3 3-4 Ave. 3.38 4 38 4.53 5.43 1.59 2.01 1.37 4.43 3.88 S22 0-1 1-2 2-3 3-4 Ave. 1.62 1.91 1.44 2.36 S6 0-1 1-2 2-3 3-4 Ave. 4.76 6.25 5.10 5.27 S14 0-1 1-2 2-3 3-4 Ave. 0.97 2.56 3.31 4.88 1.83 5.34 2.93 S23 0-1 1-2 2-3 3-4 Ave. 0.836 0.99 1.85 2.33 S7 0-1 1-2 2-3 3-4 Ave. 3.52 4.07 5.21 5.42 S15 0-1 1-2 2-3 3-4 Ave. 1.18 2.41 3.10 3.35 1.50 4.55 2.51 S24 0-1 1-2 2-3 3-4 Ave. 3.19 3.06 2.46 2.39 S8 0-1 1-2 2-3 3-4 Ave. 1.52 4.54 4.70 4.88 S16 0-1 1-2 2-3 3-4 Ave. 1.87 4.40 4.04 4.60 2.77 4.55 3.73 [48 Table 23 — (continued) Sample No. Depth (ft.) mmhos/cm Sample No. Depth (ft.) mmhos/cm Sample No. Depth (ft.) mmhos/cm S25 0-1 1-2 2-3 3-4 Ave. 3.65 2.24 2.20 2.56 S28 0-1 1-2 2-3 3-4 Ave. 4.55 5.10 3.58 3.61 831 0-1 1-2 2-3 3-4 Ave. 5.72 3.86 4.00 3.21 2.66 4.21 4.20 S26 0-1 1-2 2-3 3-4 Ave. 1.48 1.27 2.12 3.10 S29 0-1 1-2 2-3 3-4 Ave. 1.28 3.57 3.20 3.40 S32 0-1 1-2 2-3 3-4 Ave. 6.99 5.02 3.29 3.42 1.99 2.86 4.68 S27 0-1 1-2 2-3 3-4 Ave. 3.43 4.23 2.98 S30 0-1 1-2 2-3 3-4 Ave. 0.896 1.00 1.87 3.32 3.55 1.77 page 36) ; and side losses of water, par- ticularly south of line A which was flow- ing full and under pressure. Salinity distribution. Tables 22 and 23 list the conductivity values for saturated extracts of samples obtained in March and in September 1958. The value for each depth at each site is listed to show the distribution throughout the profile. Table 24 gives the average salinity at each site on each sampling date for the test plots. These results have been used to construct maps showing lines of equal salt concentration for March and Sep- tember 1958 and April 1959. The distribution of salinity through- out the profile in March 1958 from table 22 shows that the salinity distribution is rather uniform throughout the profile. This is to be expected since the winter water table is probably within 1 foot of the surface and little leaching below 1 foot occurs as a result of winter rainfall. The distribution at the end of leaching APRIL 1958 Figure 21. Lines of equal average conductivity of saturation extracts for test plot. April 1958. [49 SEPTEMBER 1958 — «• - 3 **^*. LINES OF EQUAL CONDUCTIVITY (mmhos/cm) < < OPEN DRAIN Figure 22. Lines of equal average conductivity of saturation extracts for test plot, September 1958. APRIL 1959 Figure 23. Lines of equal average conductivity of saturation extracts for test plot, April 1959. 1.0 ' ' 1 1 I — 1 1 I 1 1 1 1 1 1 1 I - 09 _ 08 *xV „ „ x - 0.7 o° 06 O ° 5 0.4 x -i X X » x X V 0.600* 0.0156 X - 03 r»-o - 0.2 - 0.1 i -1 1 1 1 i. I l , . I I I 0.2 03 04 0.5 06 07 0.8 09 10 I.I 12 1.3 1.4 1.5 16 1.7 1.8 1.9 2.0 Olw/Ot Figure 24. Depth of water per unit depth of soil required to leach Tulelake lease land soils. [50 period in September shows that the salin- ity was reduced below the March values throughout the profile. This is also shown in table 25 where the ratios of average final to initial salt expressed as concen- trations are listed for each plot. The high C/C value for plot 1 existed because no leaching was given in this plot. The results of this table are discussed in greater detail below. The average salinity distribution throughout the entire 78-acre field in March and September 1953 is shown as lines of equal salt concentration in figures 21 and 22. The figure for March indicates that prior to leaching, salinity increased in a northeast to southwest direction. The largest concentration ap- peared in the southwest corner. If the 4 mmhos per cm contour is selected as the boundary between saline and non- saline areas, it is seen that more than half the field contained a saline condi- tion. The September salinity map shows that the salt contours have decreased in a northeast-southwest direction as a re- sult of leaching and have been com- pressed into the southwest corner. The 4 mmhos per cm boundary has also been shifted and at this time less than half the area may be considered as saline. Effectiveness of leaching. The average conductivity values for each plot were calculated from table 24 and the con- centration values estimated from figure 5; the results are tabulated in table 25. The C values refer to salinity found in the March 1958 samples, the Cj values to midseason samples, and the C> values to September 1958. Table 25 also lists the C/C ratio and the depth of leaching water to depth of soil (Dlw/Ds) ratio corresponding to the C 2 and C 2 values. Using these results a graph of C/C vs Dlw/Ds was constructed and given in figure 24. Figure 24 indicates that salinity is re- duced fairly rapidly until a C/C value of 0.6 is reached. Beyond this point the Table 24 SUMMARY OF AVERAGE CONDUCTIVI- TIES AND CONCENTRATIONS OF SAT- [JEATED SOIL EXTRACTS FOR TEST PLOTS— MARCH, JULY, A X I > SEPTEMBER, 1958 Plot No. Sampling date Average m mhos/cm Average meq 1 1 March July 6.54 6.51 6.15 97.8 97.2 91 4 2 March July September 7.29 5.49 3.31 110.6 80.4 45.0 3 March July September 5.49 6.16 4.07 80.7 91.5 56.1 4A March July 6.00 4.84 4.66 88.93 70.0 66 9 4B 4.50 1.12 64 4 50 6 5A 4.39 4.49 3.53 62 3 64.2 48 5 5B March 3.59 2.28 49 4 28 6 6A March 4.64 3.28 67 44 7 6B 3.51 1.83 48 4 21 3 6C 3.27 2.62 44 6 34 6 6D 4.18 2.16 58 9 26 7 6E 3.69 1.86 50 9 22 3 7A 2.50 1.38 31 4 15 2 7B 3.11 1.43 41 5 15 7 7C 2.58 1.97 32 9 23 2 7D 4 80 5.35 69 78 [51] Table 25 DETERMINATION OF C/Co AND Dlw/Ds EATIOS Plot Co* meq/C Ci* meq/l d/Co Dlw (acre-ft.) Plot area (acres) Soil depth (ft.) Ds (acre-ft.) Dlw/Ds 1 97.8 110.6 80.7 88.9 64.4 62.3 97.2 80.4 91.5 70.0 64.2 0.994 0.727 1.134 0.787 1.030 12.019 5.964 8.234 7.200 6.05 6.15 6.46 5.44 7.30 4 4 4 4 4 4 24.2 24.6 25.8 21.8 29.2 2 3 0.488 4A 0.231 4B 0.378 5A 0.246 Plot Co* meq/C c 2 * meq/l Ci/Co Dlw (acre-ft.) Plot area (acres) Soil depth (ft.) Ds (acre-ft.) Dlw/Ds 1 97.8 110.6 80r7 88.9 91.4 45.0 56.1 66.9 0.934 0.407 0.695 0.752 26.610 49.172 14.304 6.05 6.15 6.46 4 4 4 4 24.2 24.6 25.8 2 1.100 3 1.999 4A 0.554 4B 64.4 50.6 0.786 23.390 5.44 4 21.8 1.073 5A 62.3 48.5 0.778 27.587 7.30 4 29.2 0.945 5B 49.4 28.6 0.579 25.100 4.65 4 18.6 1.349 6A 67.0 48.4 44.6 44.7 21.3 34.6 0.667 0.440 0.777 23.390 7.837 9.742 5.42 2.64 3.67 4 4 4 21.7 10.56 14.68 1.078 6B 0.742 6C 0.664 6D 58.9 26.7 0.453 9.383 2.96 4 11.84 0.794 6E 50.9 22.3 0.438 11.364 2.96 4 11.84 0.960 7A 31.4 41.5 15.2 15.7 0.484 0.378 3 074 2.34 4 9.36 0.328 7B 7C 32.9 23.2 0.705 7D 69.0 78.0 1.130 Co = March samples. Ci = Midseason samples. Co = End-season samples Table 26 CONDUCTIVITY OF SATURATED SOIL EXTRACTS— SPRING 1959 Sample Number in sample mmhos/cm Sample Number in sample mmhos/cm Sample Number in sample mmhos/cm 1 0-6 4.76 5.82 5.69 5.75 5.50 80.5 3 0-6 6-1 1-2 2-3 1.66 1.79 3.72 4.07 2.81 36.7 5 0-6 1.03 6-1 1-2 6-1 1-2 2-3 3.40 955 2-3 1.36 Average meq/l Average meq/l Average meq/l 1.70 19.5 2 0-6 6-1 9.50 8.10 6.60 5.92 7.53 114.0 4 0-6 1.28 1.49 2.77 3.31 2.21 26.7 6 0-6 1.22 6-1 6-1 1-2 2-3 1 17 1-2 2-3 1-2 2-3 0.975 1 40 Average meq/1 Average meq/l Average meq/l 1.19 12.7 [52] Table 26 — (continued) Sample Number in sample mmhos/cm Sample Number in sample mmhos/cm Sample Number in sample mmhos/cm 7 0-6 6-1 1-2 1.00 3.41 1.33 3.55 2.32 28.5 16 0-6 1.22 1.65 3.37 3.92 2.54 32.0 25 0-6 1.15 6-1 1-2 2-3 6-1 1.11 1-2 1.27 2-3 2-3 1.59 Average meq/1 Average meq/1 Average meq/1 1.28 18.0 8 0-6 6-1 1-2 2-3 Average meq/1 3.29 2.15 4.65 4.63 3.67 50.8 17 0-6 1.46 1.38 2.07 2.76 1.92 22.5 26 0-6 1.22 6-1 6-1 1.01 1-2 1-2 0.965 2-3 2-3 1.37 Average meq/1 Average meq/1 1.14 12.1 9 0-6 1.18 1.25 3.47 2.43 2.08 24.8 18 0-6 1.38 1.71 3.13 2.55 2.19 26.3 27 0-6 1.21 6-1 6-1 6-1 1.13 1-2 1-2 1-2 1.59 2-3 2-3 .. 2-3 1.81 Average meq/1 Average meq/1 Average meq/1 1.44 15.7 10 0-6 4.20 3.70 2.42 4.03 3.59 49.4 19 0-6 1.98 4.00 5.06 3.31 3.59 49.4 28 0-6 1.38 6-1 6-1 1-2 2-3 Average meq/1 6-1 1.54 1-2 .. . 1-2 1.56 2-3 2-3 2.19 Average meq/1 Average meq/1 1.68 19.2 11 0-6 3.34 1.79 4.73 4.65 3.63 50.0 20 0-6 1.51 1.41 3.13 2.42 2.12 25.4 29 0-6 6-1 1-2 2-3 0.930 6-1 1-2 2-3 6-1 1.86 1-2 3.08 2-3 2.11 Average meq/1 Average meq/1 Average meq/1 2.00 23.5 12 0-6 6-1 1-2 1.22 4.49 3.01 4.03 3.19 42.8 21 0-6 2.83 3.54 4.07 2.57 3.25 43.6 30 0-6 1.05 6-1 1-2 2-3 6-1 2.86 1-2 3.46 2-3 2-3 2.54 Average meq/1 Average meq/1 Average meq/1 2.48 31.1 13 0-6 1.34 1.70 2.80 3.36 2.30 27.0 22 0-6 1.54 2.09 2.29 1.88 1.95 22.9 31 0-6 2.40 6-1 6-1 6-1 3.62 1-2 1-2 1-2 4.21 2-3 2-3 2-3 2.87 Average meq/1 Average meq/1 Average meq/1 3.28 49.3 14 0-6 3.32 4.07 4.77 3.28 3.86 53.9 23 0-6 2.13 1.38 1.94 1.90 1.84 22.6 32 0-6 6.65 6-1 6-1 6-1 6.17 1-2 1-2 1-2 2-3 Average meq/1 5.32 2-3 2-3 4.25 Average meq/1 Average meq/1 5.59 82.2 15 0-6 6-1 1-2 3.38 3.92 3.87 2.46 3.41 41.5 24 0-6 6-1 1-2 7.48 5.91 5.11 3.65 5.54 81.3 2-3 2-3 Average meq/1 Average meq/1 [53] curve flattens out to a straight line. Ap- plying linear regression to this line the equation Y = 0.600 + 0.01 56X was cal- culated. It was found, upon further analy- sis, that the regression coefficient, 0.0156, was not significantly different from zero at the 1 per cent level. That is, beyond a C/C value of 0.600 and Dlw/Ds value of about 0.5 the equation Y = 0.600 is valid showing that Y is no longer a func- tion of X. This means that salinity is not statistically reduced (at least within the confines of the sampling location) with addition of quantities of leaching water greater than about 0.5 feet per foot depth of soil. The statistical analysis used here should not be interpreted to mean that no additional removal of salts occurs with further leaching. Rather it is pre- sented to indicate that these soils differ markedly in their leaching characteristics from mineral soils. A much more com- plete sampling of the test area would probably have indicated some additional salt removal with the large quantities of water applied. However, the efficiency of salt removal per unit depth of water mov- ing through the profile is much lower than for other soils. These results seem to correlate well with the extremely high water transmission rates of this soil, as well as the salinity data obtained from tile outflow. The nature of the soil structure appears to be the reason for the high flow rates and poor leaching. It will be recalled that the structure of the lease soils is such that large cracks are present be- tween adjacent soil blocks and that these cracks provide the main avenues for water and salt transmission. During leaching, therefore, water passes rapidly through the cracks to the tile lines and much more slowly through the soil blocks. As a result, the main soil body is not effectively leached and salt diffusion from the interior of the blocks to the cracks is too slow to result in an efficient leaching. It would appear that the differ- ential flow rate of water through this soil j* results in hydrodynamic dispersion dur- ing miscible displacement on a much ? more exaggerated scale than occurs in normal soils (Nielsen and Biggar, 1961) . Continuous flooding would represenl no improvement in the leaching efficiency. If the tile lines were purposely plugged so that the water table could be raised < with a minimum of water removed, some benefits from diffusion might be ob- tained with time. An alternative to the flooding proce- dure which would provide much greater 4 efficiencies would be the application of the leaching water to the soil surface at a rate comparable to flow through the soil blocks. This would reduce, to a mini- mum, water movement through the large cracks resulting in more efficient leach- , ing of the soil blocks. Effect of leaching by rainfall. It is be- lieved that the leaching by rainfall ob- tained from September 1958 to April 1959 may have been successful because of the effect mentioned above. A com- parison of the samplings in September 1958 and April 1959 indicates (see tables 23 and 26 and figures 22 and 23) considerable benefit with a small amount of water applied. The total rain- fall between samplings amounted to 4.01 inches during several rainfall periods. Since it is doubtful that water became ponded on the surface, much of the rain- J fall probably was transmitted through the soil blocks rather than the cracks. The leaching of a soil in the unsaturated state would seem to agree with these find- ings (Nielsen and Biggar, 1961). Irrigation water and tile effluent analy- sis. Since much of the salt found in irri- gated areas is deposited in the soil from the irrigation water, it is useful to know the quality of the irrigation water. The irrigation water in the lease lands varies somewhat over the season because it is mixed from the various sources. For the lease under investigation the irrigation water was quite uniform during the [54] Table 27 ANALYSIS OF WATER FROM TILE LINES TAKEN TX JUNE AND 8EPTEMBEE 1958 Tile Milliequivalents per liter Na Ca Mg K Total HC0 3 CI SO4 Total 1 2 A 36.18 21.65 28.00 18.70 21.66 10.58 0.67 82 86.51 51 75 7.08 6.62 9.30 3.66 68.68 40.64 85.06 50 94 1 2 B 30.78 24.61 23.20 21.60 18.66 13.58 67 1.02 73.31 60.81 7 00 7.19 6.48 3.29 57.18 49.55 70.66 60.11 1 2 C 36.18 23.91 28.00 18.37 22.66 14.58 1.02 0.82 87.86 57.68 8.45 7.65 8 79 3 04 70.00 46.33 87.24 57 12 1 2 D 22.17 19.35 20.60 19.20 16.17 11.00 0.51 0.82 59.45 50.37 4.91 6 51 3 10 2 31 49.62 40.75 57.63 49 71 1 2 E 21.74 18.78 21.10 18.70 16.17 9.67 0.67 0.82 59.68 47.97 5.94 7 54 3.27 2.03 48.76 37.63 57.97 47 44 1 2 F 17.17 16.87 20.15 16.35 13.83 9.67 0.51 0.61 51.66 43 50 5 94 8.34 4 20 1.97 41.38 32.91 51.52 43 34 1 = Samples collected June 10, 1958. 2 = Samples collected September 16, 1958. Analysis of Irrigation Water Milliequivalents per liter Na Ca Mg K T HC0 3 CI SO4 Total SAB 2.07 1.12 1.84 0.16 5.19 3.54 0.39 1.26 5.19 1.73 leaching period. The analysis given in table 27 indicates that it is a medium- salt low-sodium water. The bicarbonate content is somewhat high and shows that there are 0.53 me/1 of residual sodium carbonate. The presence of excessive bi- carbonate is undesirable because of the low solubility of calcium carbonate. Re- moval of calcium from the soil solution by precipitation of the Ca with the bi- carbonate increases the proportion of sodium to calcium and magnesium in the soil solution. This increase in sodium results in an increase of exchangeable sodium which over a period of time under accumulative conditions may lead to problems associated with excessive ex- changeable sodium. Tile outflow analysis. The quantity of outflow from the tile lines has been shown previously to fluctuate drastically during the leaching tests. Several conduc- tivity measurements during the leaching period indicate that the salinity fluctuates also but to a much lesser extent. In fact the analysis presented in table 27 of waters taken near the beginning and end of the leaching trials indicates a rather low salt load initially and a correspond- ing high salt load after the application of many feet of water. In general, the salt load in the tile effluent remained some- what constant throughout the season. The low initial salt loads and more or less constant salinity throughout the sea- son indicates that considerable dilution of the salty effluent by irrigation water has occurred. These results compare fa- vorably with the rather poor leaching of the soil and the fact that considerable flow occurred through the cracks rather than the soil blocks. [55 Literature Cited Baver, L. D. 1956. Soil Physics. John Wiley and Sons, Inc. Dachnowski-Stokes, A. P. n 1936. Peat land in the Pacific Coast states in relation to land and water resources. U. &. Dept. of Agric, Misc. Pub. 248. Feustel, I. C. and H. C. Byers. m 1936. The comparative moisture-absorbing and moisture-retaining capacities of peat and soil mixtures. U. S. Dept. of Agric, Tech. Bui. 532. Hinds, Norman, E. A. . 1952. Evolution of the California landscape. State of California, Dept. of Nat. Resources, Divi- sion of Mines, Bui. 158. Larson, C. L. and L. F. Hermsmeier. 1958. Device for measuring pipe effluent. Agric. Engr., 39 (5) : 282-284. Leppla P. W. 1953.' Diatomite. Mineral Information Service, Dept. of Nat. Resources, Division of Mines, 6 (11). Luthin, J. N. Ed. 1957. Drainage of Agricultural Lands. Amer. Soc. of Agronomy, Madison, Wisconsin. Moore, Bernard N. 1937. Nonmetallic mineral resources of Eastern Oregon. U. S. Dept. of Interior, Geolog. Sur- vey, Bui. 875. Murdock, Joseph and Robert W. Webb. 1956. Minerals of California. State of California, Dept. of Nat. Resources, Division of Minos, Bui. 173. Neal, J. H. 1934. Proper spacing and depth of tile drains determined by the physical properties of the soil. Univ. of Minn., Agr. Exp. Sta., Tech. Bui. 101. Nielsen, D. R. and J. W. Biggar. 1961. Miscible displacement in soils. I. Experimental information. Soil Sci. Soc. Amer. Proc. 25(1). Peacock, M. A. 1931. The Modoc lava field, North California. Geog. Review, 21 : 259-275. Reeve, R. C, A. F. Pillsbury and L. V. Wilcox. 1955. Reclamation of a saline and high boron soil in the Coachella Valley of California. Hil- gardia, 24 (4) : 69-91. Reeve, Ronald C. 1957. The relation of salinity to irrigation and drainage requirements. International Commis- sion on Irrigation and Drainage, Third Congress, San Francisco, Question 10, pp. 175- 187. Richards, L. A., Ed. 1954. Diagnosis and improvement of saline and alkali soils. Soil and Water Conservation Re- search Branch, Agric. Res. Serv., U. S. Salinity Lab. Staff, Agr. Handbook 60. Robbins, W. W. and T. E. Weier. 1950. Botany, an introduction to plant science. John Wiley and Sons, Inc., New York. Scott, V. H. and C. E. Houston. 1959. Measuring irrigation water. Calif. Agric. Expt. Sta., Extension Serv., Cir. 473. Taliaferro, N. L. 1935. The relation of volcanism to diatomaceous and associated siliceous sediments. Univ. of Calif. Publications in Geol. Sciences, 23: 1-55. U.S.D.I., Bureau of Reclamation. 1948. Reclamation project data. U. S. Government Printing Office. 1950. Land classification, Tule Lake Division, Klamath Project. 1957. Annual project history Klamath Project Oregon-California. Region 2, Report 46. 1957a. A half century of progress on the Klamath reclamation project. U. S. Government Print- ing Office. Veihmeyer, F. J. and A. H. Hendrickson. 1948. The permanent wilting percentage as a reference for the measure of soil moisture. Trans. Am. Geophysical Union 29: 887-896. 5m-3,'61(B5866)J.K.