The trenches were excavated to explore both the dam foundation and the spillway foundation. These trenches were shallow, however, penetrating only to the compact members of the Ogallala formation.

16. Geology of the Dam Site. The geology of the dam site involves principally a consideration of the overburden and the Ogallala formation. Along the axis of the dam, the eroded surface of the Ogallala ranges from elevation 3165 along the abutments to as low as elevation 2983 at station 46+00. This material in the floor of the valley is for the most part mantled by alluvial deposits up to 35 feet in depth. Permeability tests were performed in several drill holes. Early attempts to correlate the various strata of the Ogallala on the basis of observation of drill cores proved infeasible. The great irregularity in cementation between holes made it impossible to correlate the predominantly sandy or predominantly silty zones from one hole to the next hole.

It was believed that the lenses of Ogallala sediment were discontinuous for short distances and that the permeability of the Ogallala formation was low. Foundation design was based originally on this concept. It was also believed that the Ogallala formation would be sufficiently compact to bear the weight of the structures without detrimental settlement. This was later confirmed by the lack of any serious settlement in the embankment that could be attributed to compression of the foundation materials.

The foundation for the dike is about 40 to 65 feet of loess overlying the more pervious Ogallala sediments. This loess varied from porous silty-fine sand to sandy silt or clay. In places it is damp to dry, firm to friable, and often compact. Colors are grey, green, yellow, or brown. Clayey material is stiff in place. Texture is predominantly smooth. The ground-water surface was not encountered during foundation exploration. Shells or shell fragments were common in some strata. Average properties of the undisturbed loess comprising the dike foundation were as follows:

17. Construction Geology. - Much of the geology at Enders Dam was determined during construction. The various ́members of the Ogallala, as observed during construction of the cutoff trench, tended to be somewhat stratified; however, the stratification was based primarily on a predominance of sand or silt in a certain layer. One section would be predominantly sandy but would contain lenses of silt and sandy silt; and another section which was predominantly silty would contain lenses of sand and gravel. A tentative correlation for predominantly sandy and silty layers was attempted as shown in figure 4.

The predominantly sandy zones were classified as the A, B, C, and D zones on the left abutment. A series of sand zones were similarly identified on the right abutment, although owing to the lack of drill holes such identification was based on surface observation and exposures in two trenches. Three other zones, the E, F, and G zones, are believed to exist underneath the present cutoff trench across the dam but are mantled by a somewhat impervious silt layer.

When the reservoir started to fill, considerable seepage from some of the sandy zones required special treatment. Because of this seepage, it was necessary in 1952 to perform additional exploratory drilling to further outline these sandy zones. This drill

ing confirmed earlier conclusions that the sand zones bifurcate and merge irregularly and vary from fine-grained to coarse in degree of cementation and that the top and bottom elevations at any point can be considerably different from those established by drawing correlative sections. Generally, these zones are separated by finer siltier beds that are quite impervious, but it could not be established with complete certainty that separate zones did not merge locally or possess some other kind of interconnection. The lowermost zones, E, F, and G, are overlain by compacted silty sand and silt into which the cutoff trench is bedded. These lower zones are probably extensive but are not exposed in the downstream portion of the reservoir nor in the valley downstream from the dam. The continuity of the covering bed of silty sand and silt is not known.

The sand zones consist of continuous layers of sand and partially to completely cemented sandstone beds. Sands are brown to grey, calcareous, clean to dirty, well to poorly graded, and crossbedded. The particle sizes vary from fine sand to fine gravel with sizes up to a maximum of about 1 inch. Particles are round to subround. Quartz and chert are the dominant minerals. Sandstone layers are coarse to fine grained, grey, cemented with calcium carbonate, hard to soft, 6 inches to 4 feet thick, and discontinuous. The sand zones are mappable horizons. The sand zones are separated by silty sand, silt, or clay strata. Silty sands are light brown to grey, calcareous, and somewhat crossbedded. The sand component varies from fine to medium. Silts are chalky and brown, and contain thin sand seams. Clays are moderately plastic, brown, red or grey, sandy, and slightly calcareous. The position of sand zones is shown in the following tabulation:

The foundation for the spillway is Ogallala formation, except for the outlet channel. Because of the extreme irregularity of materials, additional investigations were performed prior to the final design of the spillway structure. These investigations included penetration tests, experimental Vibroflotation tests, and Denison sampling. The experimental Vibroflotation tests were conducted in the sand strata of borrow area B-1. The sands in the borrow area were generally similar in gradation and particle shape to those in the spillway foundation area. Good consolidation of the free draining material in the test area was accomplished; however, when the excavation for the spillway reached grade, it was decided that consolidation by Vibroflotation was not advisable.

In addition to the above spillway foundation investigations, carefully controlled permeability tests were performed in the drill holes. The purpose of this testing was to determine the possibility of percolation underneath the spillway and subsequent hydrostatic uplift, and the possibility of settlement of the sandier materials by vibration of the water in the spillway stilling basin. It was concluded from the tests that, because of slow drainage, overburden, and in many cases partial cementation, the partly cemented and uncemented impervious fine-grained soils would not be subject to settle

ment or instability under the anticipated dynamic or static loads of the spillway. The uncemented pervious sands were overlain for the most part by highly cemented or partly cemented sands, impervious silts, or combinations thereof. Where the uncemented sands extended to the grade line, they were considered to be sufficiently dense that differential settlement would not take place. Artesian aquifers were not located; however, the grade of the water table was such that some severe losses might have occurred as a result of percolation beneath and around the spillway if adequate precautions had not been taken in the design.

After filling of the reservoir had started, observations were made on the groundwater table. It was found that the table indicated waters moving through the right abutment parallel to the valley with only slight contribution toward the valley itself, and surface water appeared in the draw about one-half mile downstream. A small amount of seepage developed in the vicinity of the outlet works control house. The gound-water table over the whole area downstream from the dam, including the valley sides, continued to rise during the rising of the reservoir; and wells within a mile or more of the dam site were affected. On the left abutment, high water pressures were evidenced around the end of the dam. Because of sealing by impervious layers, the water table was apparently forced to the surface through the abutment mantle and along the semipervious zone of the embankment. Because of these observations, special drainage construction was instituted along the toe of the embankment. Ground water analyses in the area showed 244 to 396 parts per million dissolved solids, 160 to 210 parts per million calcium carbonate, 144 to 279 parts per million bicarbonate, and 10.1 to 70.1 parts per million sulphates. These quantities indicate a comparatively low water-soluble solids content. One dye test indicated a rate of travel in sand zone D of 0.54 feet per minute.

In the valley proper, additional exploration outlined an erosional spur trending in a northeasterly direction from the right abutment, and an ancient stream channel adjacent to the left abutment encircling this spur. The thalweg of the stream channel is approximately at elevation 3000, about 10 feet lower than the top of the Ogallala formation, as originally indicated. This channel is contiguous with sand zones D and E in the left abutment. The ancient stream channel is filled with river sands and fine gravels to about elevation 3030. The stream sands and gravels are overlain by 3 to 12 feet of silty fine sands, silts, organic silts, and swamp muck. From well permeameter tests, the computed coefficient of permeability varied from 900 to 255, 000 feet per year. The altered concepts of foundation conditions as disclosed by the geologic anomalies made modifications to the original foundation designs necessary. From the original and supplemental data the waste areas, abutment cutoffs, and channel blanket were designed.

18. Materials Explorations. - Explorations for earth materials were carried out in three phases--preliminary, preconstruction, and construction. Investigations were initiated in October 1944. A reconnaissance exploration was made concurrently with the preliminary foundation studies in 1945. Prospective borrow areas A-1, A-2, B-1, B-2, B-3, and B-4, all for impervious material, were explored by 12 test pits and 26 power auger holes. Reconnaissance exploration indicated sufficient impervious material was available in the area. No positive source of pervious material was located; however, the alluvial sands in the river channel were suggested as a possible source. Pervious borrow areas G-1 and G-2 were located subsequently. Rock quarries at Lincoln, Kans.; Golden, Colo.; and Guernsey, Wyo., were suggested as possible sources of riprap. In 1946, quarries near Franklin, Nebr., were selected as possible riprap sources after investigation of 13 other possible sources.

It was recognized when the specifications were written that locating and obtaining suitable embankment materials for Enders Dam would present difficult problems and that because of limited exploration extensive preconstruction investigations would be required. A preconstruction investigation program for borrow materials was initiated in October 1946. Investigations were made principally with a Buda earth auger capable of drilling a maximum depth of 24 feet. A churn drill was used to drill deeper than 24 feet. A hand auger was used for preliminary drilling. Forty-two holes were drilled in

the vicinity of the B-1 borrow area, and 20 holes were drilled in the G-2 area. Neither the G-1 nor G-2 areas, sources of pervious material, were entirely satisfactory because of high ground-water surface and pockets of organic material. Both were located in the valley bottom close to the toes of the dam, one upstream and one downstream. Preconstruction exploration located a deposit of Ogallala formation sand on the left abutment downstream from the dam site. Since this area was in the same general vicinity as G-2, except higher on the left abutment, the same designation was applied to the new location. Preconstruction zoning studies indicated that sufficient quantities of material were available in the B-1 and G-2 areas and from required excavation, to construct the dam and to supply pervious materials for the dike.

During construction, a lower yield from area B-1 and the wasting of larger quantities of required excavation than had been anticipated made the location of additional sources of both pervious and impervious materials necessary. Hence, areas B-1-E, D, and A-1 were developed. Area B-1-E, containing impervious material, was an upstream extension of B-1. Area D, located high on the left abutment and downstream from the extended dam axis in a previously unexplored area, provided additional impervious material. Area A-1, on the right abutment upstream from the spillway area, provided pervious material. These areas were explored by hand auger and rotary drill. The location of all developed borrow areas is shown on figure 5. A loess deposit of fine sandy silt on the right abutment, area T, was explored by hand auger and Buda earth drill and was selected to provide impervious material for the abutment cutoffs required under extra work order No. 5. Borrow areas for the dike embankment were not explored prior to the date specifications were issued. Materials were to be obtained from "excavations required for other parts of the work or from borrow pits in the vicinity of the work. Therefore areas No. 1, 2, 3, and 4 were explored prior to and during construction by hand auger holes on 200-foot centers.

Explorations for earth materials indicated that plenty of impervious material suitable for zone 1 was available close to the site. However, required yardages of suitable pervious materials, rockfill, and riprap would be difficult to obtain because adequate deposits were not available locally, because suitable materials were overlain with loess, and because stratification in the areas selected prevented proper mixing and selection of materials above the water table.

19. Borrowed Impervious Materials. impervious, zone 1 material from borrow areas. obtained from borrow areas B-1, B-1-E (part 1), from selected parts of the required excavation. tained from borrow areas No. 1, 2, 3, and 4.

It was necessary to obtain much of the
Zone 1 materials for the dam were
B-1-E (part 2), D, G-2, and T and
Zone 1 materials for the dike were ob-

Because all of the impervious materials were considered to be essentially similar, originally only samples of soil from area B-1 were tested in the earth materials laboratory. Information from 10 test holes indicated that borrow area B-1 was 5 to 15 feet of interlensed, predominantly fine to coarse sand with silt-clay binder, SM-5C, or sandy silty-clay mixture, ML-CL. This material, generally known as loess or reworked loess, covers a variable thickness of sands and gravels which in turn overlie the Ogallala formation. The water table varied from 5 to 9 feet below the ground surface. For four representative samples, in-place dry unit weight ranged from 84.0 to 93.9 pounds per cubic foot, and moisture content ranged from 5.0 to 12.8 percent. Average properties of composite samples from four test pits for full depth of material in borrow area B-1 are shown in figure 6. Testing in borrow area B-1-E, parts 1 and 2, indicated the average properties for the remolded material as shown in figure 6. The results of laboratory testing of materials from the G-1 and G-2 areas indicated that the upper 10 feet of soil would be impervious silty sand and sandy silt. Since the soil properties of these areas are similar to the soil properties of the B-1 and B-1-E areas, these data are omitted. Dike impervious borrow materials varied from USC (Unified Soil Classification) group symbol SM to ML with ML material apparently dominant. Laboratory test results for four representative samples from all the dike borrow areas are also shown in figure 6.