STATE OF CALIFORNIA EARL WARREN. Governor DEPARTMENT OF NATURAL RESOURCES WARREN T. HANNUM, Director DIVISION OF MINES FERRY BUILDING. SAN FRANCISCO 11 OLAF P. JENKINS. Chief SAN FRANCISCO SPECIAL REPORT 2 OCTOBER 1950 GEOLOGY OF PART OF THE DELTA-MENDOTA CANAL NEAR TRACY, CALIFORNIA By PARRY REICHE U. S. Department of the Interior, Bureau of Reclamation Digitized by the Internet Archive in 2012 with funding from University of California, Davis Libraries http://archive.org/details/geologyofpartofd02reic GEOLOGY OF PART OF THE DELTA-MENDOTA CANAL NEAR TRACY, CALIFORNIA By Parry Reiche* OUTLINE OF REPORT Page 3 3 5 ABSTRACT INTRODUCTION GEOLOGIC SETTING CANAL GEOLOGY E5 Stratigraphy 5 Neroly formation 5 Post-Neroly unconformity (i Tulare formation 6 Quaternary sediments and land forms i) Structure 10 Engineering geology 11 REFERENCES CITED 12 Illustrations Figure 1. Index map of central and northern California. The district described is west of the town of Tracy 3 2. Geologic sketch map, showing project features noted in text 4 3. View northwest from station 257. Outlet of Moun- tain House Road siphon 6 4. Phi-quartile parameters of representative Tulare formation sediments of intermediate grade 7 5. Generalized geology of canal cut, stations 191 to 211 10 ABSTRACT The Delta-Mendota Canal excavation exposed, for about 6 miles, Miocene and Plio-Pleistocene continental beds in a district of few natural outcrops. The cut has since been lined with concrete, and record is here made of the materials and of a hitherto unnoted uncon- formity. The 450-500 feet of Plio-Pleistocene strata are referred to the Tulare formation of which they are an unfamiliar facies. An east- ward shift of about 15 miles of the axis of the Great Valley of Cali- fornia during Quaternary times is suggested. Engineering applica- tions of the geology are briefly discussed. INTRODUCTION Canal-line studies were so important in the early development of geology as to be of perennial interest to all followers of William Smith. This paper summarizes the geology of an 8-mile stretch of a canal now under con- struction by the United States Bureau of Reclamation, and illustrates a few engineering applications of geologic study. The Delta-Mendota Canal will extend 117 miles south-southeastward along the -western margin of the lower San Joaquin Valley of central California. Much of the water it will carry falls in northernmost California and is controlled by Shasta Dam. After release the flow will be via the Sacramento River and a by-pass. This near-sea- level by-pass consists of a short canal and system of natu- ral channels, across the polder and slough country (the "Delta" to Calif ornians) in which the Sacramento and San Joaquin Rivers unite before turning west to San Francisco Bay and the Pacific. From this cross channel, 4,600 cubic feet of water per second will be drawn off and • Geologist, U. S. Bureau of Reclamation, Sacramento. Early phases of the investigation were carried on by Aubrey Walker. The observations and conclusions of this paper are, however, those of the writer, except where specifically noted otherwise. Manuscript sub- mitted for publication December 1949. directed southward for irrigation. The first 2\ miles of the Delta-Mendota is a sea-level channel, which increases in depth to about 70 feet. Pumps, operated by power devel- oped at Shasta Dam, 217 miles to the north, will then lift the water about 200 feet through a mile of closed conduit to the 106^-mile gravity section. 1 This section, the main canal, is to be lined for most of its length with 4 inches of concrete, below a berm 18 feet above the canal's flat 48- foot-wide base. The sides slope at 1^ to 1 (top width 102 feet) ; a gradient of 0.00005, or about \\\ inches to the mile is provided. The northern end of the canal, herein considered, crosses the area shown on the U. S. Geological Survey Bethany and Midway quadrangles (scale 1 : 31,680). Geo- logic studies have been made of the pump-plant founda- tion (fig. 2) and the first h\ miles to the gravity section, where the cut is in bedrock. For many miles southward the foundation is the flat-lying unconsolidated alluvial complex of the Great Valley. 1 Boden, O. G., Excavating for Central Valley's heartbeat : Exca- vating Engineer, vol. 42, no. 12, 1948. Green, R. J., Nine million yards closer to Mendota : Excavating Engineer, vol. 43, no. 5, 1949. SCALE OF MILES 25 50 75 100 Approximate boundary of th« Greot Volley DELTA MENDOTA CANAL PT. ARGUE LLO Figure 1. Index map of central and northern California showing location of Delta-Mendota Canal. The region described is west of Tracy. (3) Special Report 2 LEGEND Note equations in Conol stationing as follows : 219 1 69 back = 218 ahead 350 + 80 back =351 ahead Figure 2. Geologic sketch map showing project features noted in text. Delta-Mendota Canal GEOLOGIC SETTING The canal extends along the western margin of the Great Valley, which is a north-northwestward trending 40-mile wide tectonic trough lying between the Sierra Nevada and the Coast Ranges. The flattish upper surface of the Sierran block passes westward beneath the alluvium and Tertiary-Cretaceous sediments in the valley ; to the west is the Diablo Range, of folded and faulted Jurassic to Quaternary sediments. A broad transverse swell, not reflected in the present valley surface, affected Tertiary deposition locally. A thousand feet or more of marine Eocene and Miocene sediments found to the north and south 2 are absent in both the subsurface and the foot- hills to either side ; instead, several thousand feet of continental upper Miocene and Pliocene sediments rest on the marine Upper Cretaceous Panoche formation, at least at, and west of the valley axis. The deepest part of the trough is apparently well west of the present axis of the valley. The western parts of the Sierran block are relatively stable. The Diablo Range has several known active faults and is subject to periodic quakes. It was much folded and thrusted during its elevation in Tertiary and Pleistocene times. CANAL GEOLOGY Stratigraphy Bedrock exposed by the canal cut comprises con- tinental upper Miocene and Plio-Pleistocene strata (Neroly and Tulare formations, respectively) in an east- northeasterly dipping homocline which flattens valley- ward from 15 to less than 2 degrees. These strata have already been covered by the concrete lining of the canal. One of the few exposures afforded by the outer foot- hills is the section along the Western Pacific Railroad cut several miles to the south. This is interrupted and, according to Huey, 3 contains a hidden fault contact between Tulare and Neroly beds. 4 The facies of the Tulare found in the canal has not been recorded heretofore. Neroly Formation The Neroly formation (upper Miocene) forms the uppermost part of the San Pablo group. It was character- ized initially in terms of a distinctive echinoid fauna, 5 found north of Mt. Diablo. The name has since been extended by general consent rather than by demonstrated equivalence or coextension to a series of continental beds on the eastern side of the Diablo Range, and is so used herein. In the adjoining Tesla quadrangle the Neroly formation has recently been divided into a lower member composed in large part of bluish, opal-coated andesitic sands and cobble to pebble conglomerates, and an upper member consisting of silts and soft clavstones. 6 The two 2 Beckwith, H. T., Tracy gas field: California Div. Mines Bull. 118, p. 587, 1943. 3 Huey, A. S., Geology of the Tesla quadrangle : California Div. Mines Bull. 140, 75 pp., 1948. * Anderson, Ralph, and Pack, R. W., Geology and oil resources of the west border of the San Joaquin Valley north of Coalinga, Cali- fornia : U. S. Geol. Survey Bull. 603, 220 pp., 1915. 5 Clark, B. L,., and Woodford, A. O., The geology and paleon- tology of the type section of the Meganos formation of California : California Univ., Dept. Geol. Sci. Bull., vol 17, p. 69, 1927. Huey, A. S., Geology of the Tesla quadrangle, California : Cali- fornia Div. Mines Bull. 140, p. 43, 1948. parts are related gradationally and attain, respectively, thicknesses of 700 and 2,000 feet, Vertebrate finds re- ported here indicate that the latter figure included sev- eral hundred feet of the Tulare formation. The Neroly appears to be lapped out west of the pump-plant site (fig. 2) by the Tulare formation. West of canal station 187 a total of 270 ± feet, stratigraphically, is present. Canal exposures of the Neroly are in the transition zone noted by Huey. These occurred near the head of the gravity section from station 187 to 193+40, and again from station 206+35 to 210+20 (fig. 5). Light-gray, yellowish-gray, and olive-gray to pale yellowish-brown colors ' predominate, with a little pale blue, where opaline grain coating is especially marked. The sediments are lenticular, and vaguely bedded. Firm-friable to fairly well cemented (calcareous) silty, fine- and medium-sized angular sand grains, with rare lustrous hexagonal biotite flakes and angular quartz bipyramids, are common. Spo- radic small ocherous clayey pockets, rarely with small pebbles, have thick selenite veins. Massive, friable, in some places weakly cemented, clayey soft siltstones with ash shards and much selenite veining, grade locally into mas- sive, very silty soft claystones. More conspicuous, though volumetrically less important, are lenses of hard, calcare- ous massive to slabby, pebbly and granuliferous, ill-sorted sandstones, many with bluish opaline grain-coatings. One channel-fill, a few tens of feet wide, of hard, calcareous volcanic-pebble conglomerate was exposed. All of these materials are similar to those described by Huey in the lower Neroly of the quadrangle to the south. The coarser fractions are largely of dark volcanic rock, and the finer are of ash, as shown by quartz and biotite euhedrons and by relic shard textures in the less altered members. There are also two or three fairly persistent 30- to 60-foot units of pale-green and greenish-gray soft clay- stones and siltstones, with minor amounts of fine-grained sorted sandstones interbedded with the materials already described; one of these was exposed in the cut, at inter- vals between stations 191 and 211 (fig. 5) and another between stations 233 and 237. These materials merit the following more detailed description. North of station 212, about 35 feet of pale-green to grayish-green, massive, variably silty soft claystone occurs with a few poorly marked beds of dusky-green, massive, unctuous, internally slickensided clay. Near the top of this partial section several 2- to 4-inch beds of light-gray fine-grained limestone are traceable for several tens of feet, beyond which a few nodules of the same material mark their horizons. No fossil material was found. The cut for the Mountain House Road siphon, exposed a similar green clay overlain by ill-defimited, pale-green beds of massive, slightly clayey, compact but not cemented, siltstone, and sorted, clean, very fine-grained soft sand- stone. A preconstruction drill hole near station 235+50 cored more than 47 feet of this material. Near station 233 attitudes were taken on two thin limestones, like those noted above. In the lowest beds here exposed, flecks of carbonized wood and unidentifiable fragments of thin- shelled pelecypods and gastropods were found. That they were fresh-water forms appears highly probable. 7 Color descriptions throughout are in agreement with the Rock Color Chart (1948) distributed by the National Research Council, based on the Munsell system of color identification. Special Report 2 Figure 3. View northwest from station 257, showing outlet of Mountain House Road siphon. As it was suspected that bentonitic minerals might be present in significant amounts, four samples of the Neroly clays were sent to the Denver laboratories of the U. S. Bureau of Reclamation for petrographic and X-ray diffraction studies. 8 The samples consisted of about 50 percent montmoril- lonite-type clay, 5 to 30 percent (average 18.5 percent) illite-type clay, and minor amounts of quartz and feldspar grains. One of the four samples had 5 percent and a second had 15 percent unidentified clay mineral, probably kao- linitic. Calcite, zircon, and carbonaceous matter were recorded in two of the four, glaucophane in three, and serpentine or chlorite in one. None of these last named constituents amounted to more than a few percent. The glaucophane and serpentinous matter suggest partial derivation from the Franciscan formation. It is note- worthy that neither tourmaline nor staurolite was recorded. Both of these appear to characterize marine sedi- ments formed during the Upper Cretaceous and Tertiary periods in this part of the Coast Ranges. 9 Post- Neroly Unconformity Probably most of Pliocene time is represented by the unconformity which flatly bevels the Neroly beds. For the most part, angular discordance is very slight (fig. 2), and even in the canal cannot be measured, due to lack of trustworthy attitudes near the contact. Near the head of the gravity section, for example, dips between 12 and 15 degrees were taken on Neroly beds, and one of 11 degrees on the basal Tulare, although the latter visibly "Julius Schlocker of U. S. Bureau of Reclamation made the studies "Klein, I. E., I'etrographer for U. S. Bureau of Reclamation, Sacramento, California. laps out a 30-foot clay bed, near station 191, as shown in figure 5. An exceptional condition was exposed by the cut for the siphon, at and near station 233. There thin limestone interbeds dip 30° northeastward ; the over- lying Tulare sands only 16°. This minor and local flexing may have influenced the position of a hill or interstream rise 100 to 150 feet high which was later buried by the Tulare beds. This pre-Tulare erosional feature seems to extend, approximately, from station 230 to station 270. Lithologic character beneath the unconformity like- wise varies. At the siphon cut the lower beds preserve their characteristic appearance quite to the contact, and are overlain either by yellowish-brown ill-sorted soft sand- stones with a 0.1 foot basal pebbly and cobbly gravel (station 233+00, left) or by 3 feet of reworked material of their own substance, which in turn is overlain by loose cross-laminated sands (station 233+00, at base of right wall cut). Northward from station 212, however, the beds below the contact for a distance of 1 foot to several feet are conspicuously whitened by an ancient caliche (carbonate) accumulation both in cracks and fissures, and inter-granular. This affects both the Neroly clays and silts and, where they are lacking, the Neroly sandstones and ashy silts. Bedding is obscure or destroyed. Most of the carbonate is earthy, but occasional lenses and narrow knobby masses of pale-pink, dense, hard limestone, have resulted from its local solution and reprecipitation. This type of pre-Tulare surface alteration agrees with pub- lished paleobotanic evidence as to subarid Pliocene climate of the region. Tulare Formation The beds above the Neroly scarcely appear in natural outcrop. A scanty vertebrate fauna, representing finds Delta-Mendota Canal near both top and bottom of the section, is adjudged Plio-Pleistocene by Savage. 10 This and the rather catch- all character of the term Tulare seems to justify their inclusion in that formation. As shown in figure 2, the canal trenches these beds extensively. They also appear in small buried hills, centered at stations 449 and 471. The total thickness represented is between 450 and 500 feet, of which the upper 30 to 35 feet constitutes a dis- tinctive bed having a continuity measurable in miles ; it will be discussed separately. The remainder is a confused, varying assemblage of clastic sediments, probably depos- ited as an alluvial apron by many small, ephemeral streams from the Diablo Range. Several tens of feet at the base of the Tulare, in the headward part of the gravity section, have an ashy admix- ture due to reworking of Neroly beds. With this exception the Tulare is not, hereabouts, noticeably ashy. At the Mountain House Road siphon the lowermost 10 to 12 feet consists of light-gray to pale-yellowish-gray, loose, clean sand, ranging from fine-to coarse-textured, in lensing units, and minor amounts of clean coarse silts. These are cross-laminated at low angles, and at several horizons have calcareous concretions up to 18 inches in length. This association of concretions with loose cross-laminated sands was seen elsewhere along the canal only in the cut for a box culvert for the Mountain House Creek drainage at station 302+50. The top of the Tulare is eroded and over- lain by essentially horizontal sandy and gravelly Pleisto- cene and Recent alluvia. The Tulare strata are chiefly light greenish and yel- lowish grays, pale to dusky yellows, and moderate yellow- ish brown ; the more clayey beds are pale green to pale olive. In the excavation for the pump plant the upper beds are light yellowish browns and pinkish browns, which give way in the course of 5 or 10 feet to light greenish gray, pale olive and grayish green, although a coarsely mottled zone, dominated by the colors of the lower part of the exposures, extends vaguely for several tens of feet. The change takes place, oddly, not at the water table, present at depths of 8 to 13 feet, but approximately at sea level, 60 to 65 feet below the orignal ground surface. There is no evidence, or probability, of recent or past incursion of salt water. Throughout the Tulare exposed during construction, excepting the upper 30-foot member, separately consid- ered, the dominant lithology is a massive, firm-friable, silty and clayey, very fine- to medium-grained soft sand- stone. This grades in short distances and erratically to soft sandstone, and silty and very finely sandy soft clay- stone. Throughout the lower two-thirds or more of the formation, ocherous channel fills, 3-5 feet thick and 10-20 feet wide, of ferruginous coarsely sandy, unsorted, pebble conglomerate are a characteristic minor constituent. The pebbles are of many different kinds, but "rotten" basic or intermediate volcanic rocks, presumably from the lower Neroly conglomerates are noticeable in most places, especially in the lower third of the formation. Well- cemented fine- and medium-grained dark-gray graywacke, vein quartz, gray and red chert, largely of Franciscan (Jurassic) derivation, and calcareous fine-grained sand- stones and finely crystalline gray limestone, similar to exposed Upper Cretaceous rocks, are common. Many of 10 Savage, Don, personal communication. .the channels have sporadic flat pebbles of white siliceous Kreyenhagen (?) shale, and less commonly, of a white weakly schistose, sugary quartzite. Between stations 273 and 282, and again between stations 310 and 318, a 24- to 30-foot bed of massive, greenish-gray clay is underlain by a comparable thickness of light-gray to pale-yellow, low- angle cross-laminated, loose, clean, fine and medium sands, which include a very few pebbles at the base. A characteristic feature of the Tulare formation is its small-scale variation ; in very short distances, rapid lens- ing and channeling and less readily noted rapid gradua- tion in the proportion of the various grade fractions are the rule. One can seldom match a given bed or lens across the canal cut, a distance of only 48 to 102 feet. Trust- worthy attitudes are rare. Where wet, as in the pump plant excavation, bedding is indistinct. Secondary fea- tures include erratic limonite staining in bands, patches, and wavy lines ; and discrete crystals and veinlets of sele- nite, especially in the more clayey materials. Mechanical analyses of 27 Tulare sediments were made by personnel of the Sacramento laboratory of the U. S. Bureau of Reclamation, in the course of routine — l o m m m c 9 • M m • phi quortll < »o 1 o • 1 t • • • O • 1 » O 2 o Itviotion 9 °C 1 1 o 1 • i quortll* ( 1 « 1 • D. • < < I > 1 .01 .02 OS O.I 0.2 0.5 1.0 mm. Median diameter ( circlet indicate doubtful valuet, bated an extrapolation!.) Figure 4. Phi-quartile parameters of representative Tulare forma- tion sediments of intermediate grade. 8 Special Report 2 investigations. These are not of the highest accuracy, and in no case were hydrometer readings made beyond the time necessary to establish percentages of fractions smaller than 0.005 millimeter ; thus, in five cases the median diameter had to be extrapolated, and in three others, extrapolations for the lower quartile were not warranted. Recorded median diameters are plotted against phi quartile deviations and skewness n in figure 4. For the eight cases mentioned above as not thus repre- sentable, the median diameters, following reasonable extrapolations, or read directly from the cumulative curves, range from 0.003 to 0.019 millimeter, with six of the eight less than 0.005 millimeter. There is little cor- relation of sorting with size and no natural grouping into distinct classes. No analyses of the coarse gravelly sand- stones and conglomeratic lenses are available ; but though conspicuous they are so minor a constituent of the forma- tion that this does not vitiate the general picture offered. Seven representative samples of the sandy materials and one of the clay have been the subject of heavy min- eral analysis. 12 The results are given in table 1. In addi- tion, summaries of the heavy mineral content of a repre- sentative sample of the diatomaceous siltstone at the top of the Tulare, and of a sand from a Recent fan of Corral Hollow Creek, 9 miles to the south, are tabulated. Klein concludes from field and laboratory studies that abundant glaucophane, common staurolite-tourmaline, and domi- nant hornblende (more than 50 percent of the heavies) are diagnostic, respectively, of derivation from terranes of Franciscan (Upper Jurassic) sediments and green- stones, Cretaceous-Tertiary sediments, and Sierran crys- tallines. 13 Judging by these criteria the local Tulare, " Krumbein, W. C, and Pettijohn, F. J., Manual of sedimentary petrography, New York, D. Appleton-Century Co., 549 pp., 1938. 12 Klein, I. E., U. S. Bureau of Reclamation. 13 Daviess, S. N., Mineralogy of late Upper Cretaceous, Paleo- cene, and Eocene sandstones of the Los Banos district, west border of San Joaquin Valley, California: Am. Assoc. Petroleum Geologists Bull., vol. 30, pp. 63-82, 1946. except for its uppermost bed, discussed below, was derived from Coast Range sources. Two Tulare clays were studied by Schlocker. 14 In each, the associated silt grains (10 and 20 percent) com- prise quartz and subordinate feldspar, and very minor amounts of heavies. One specimen, from station 280+90, yielded chalcedony, biotite, zircon, glaucophane, sphene, and about 5 percent of calcite ; the other, from station 315-)-00, had chlorite or serpentine, zircon, glaucophane, and sphene. In both samples, the clay minerals are reported as predominantly of the montmorillonite type, with lesser amounts of the illite type, and up to 15 per- cent of an unidentified clay mineral, probably of the kaolinite group. The uppermost 30 to 35 feet of the local Tulare is a single nearly homogeneous bed of marked continuity. Beginning at the top of the cut near station 326, in the Wi sec. 17, T. 2 S., R. 4 E., M.D., it is intermittently present along the canal as far southeast as station 473, in sec. 28, and in two long stretches, between stations 337 to 351 and stations 426+50 to 445+50, it forms the entire cut. No natural outcrops are known, but the Western Pacific Railroad cut in sec. 2, T. 3 S., R. 4 E., M.D., exposes a similar and probably identical bed. The mate- rial is pale yellow to yellowish-white, massive to obscurely laminated, friable, low density, very porous, siltstone, grading locally to a sorted, very fine-grained silty sand- stone. Small circular diatoms, rare in some places and very common in others, can be seen with a hand lens. A few bleached biotite flakes are present. In many places minute vugs, bounded by sharply intersecting concave surfaces, suggest collapsed air or gas bubbles. Four samples run by the regional laboratory of the U. S. Bureau of Reclamation averaged 4 percent fine sand (Wentworth's size classifica- tion of sediments). A mineral grain study made by Klein of one of these showed predominant quartz, with lesser 11 Schlocker, Julius, U. S. Bureau of Reclamation. Table 1. Heavy minerals in nine Tulare formation specimens from the canal cut, and in a specimen from a Recent fan of Corral Hollow Creek. (All data from studies by I. E. Klein) Sta. 242 + 41 Sta. 280+00 Sta. 250+80 Sta. 275+30 Sta. 256+95 Sta. 317+34 Pump plant exc. #10 Pump plant exc. #11 Sta. 329 diatoma- ceous silt Corral Hollow Creek (Recent) Index Significance (derivation) r D „\ A a r a a a a s s (a) Glaucophane r c c (a) r c c (a) r c c c r c c c s " { [Franciscan formation. c (a) c /Cretaceous and Tertiary Staurolite ---] \marine sediments. Sierran crystalline c a D c (a) s s s c c terranes (Jurassic intrusives). (a) A A (a) c r (a) (a) a (a) A c s (a) (a) (a) c (a) a a A a c s c (a) A (a) c a A A (a) s (a) A A (a) s (a) a c s s (a) s a a s s Probably from r r r Franciscan serpentine. Probably Sierran (a) (a) metamorphic . Explanation of symbols: D, dominant (50-85% of heavies) ; A, very abundant (25-50%); a, abundant (10-25%); (3-10% ) ; c, common (1-3% ); s, scarce (<1% ) ; and r, very rare (trace). (a), fairly abundant Delta-Mendota Canal amounts of feldspars, including orthoclase, microcline, and plagioclase; bleached biotite flakes are conspicuous. The much less abundant heavies are enumerated in table 1. Much of the material consists of the broken to comminuted opaline tests of diatoms; neither volcanic glass shards nor calcareous fossil fragments were noted. For the same four samples the average clay content, by hydrometer analysis, was 44 percent and the median diameter was 0.006 millimeter. This figure, however, is of questionable value, since minute opaline fragments of the diatoms, by virtue of their relatively low density, are probably slower settling than their dimensions would indicate. Klein estimated, that not much more than 10 percent of the sam- ple he studied under the mircroscope, was of truly argil- laceous material. Near the Highway 50 bridge (station 347) the dia- tomaceous silt bed is interrupted about 21 feet above its base by a 0.5- to 1.5-foot zone containing sporadic grains and narrow lenses of clean, loose, sorted quartzose, medium- to fine-grained sand. This is traceable for several hundred feet. A 0.2-foot zone, about 3 feet higher, but of less continuity, has discontinuous lenses of loose, slightly iron-stained subrounded small-pebble to granule gravel. Its sorting and grade suggest that the diatomaceous siltstone was deposited either in a river floodway or in a shallow lake subject to flood overflow, something like the Tulare Lake of the southern San Joaquin Valley, as it was 70 and more years ago. Water was there retained except in flood season by the flat outer margin of an alluvial cone built by Kings River nearly across the Valley. During flood period much of the water together with its suspended clay could escape to the San Joaquin River and the sea. In this way, lake beds of relatively low clay content might have accumulated. 15 Either environ- ment would have occupied an axial position in the Great Valley, as it then was. Subsequent Pleistocene uplift of the adjacent Diablo Range has shifted that position 15 miles or so to the east. In either case debris from the Sierra Nevada should have formed a notable fraction of the lake sediments. This conclusion is confirmed by Klein's tabulated observations (data for station 329), and conclusions. The following vertebrate remains are from beds here classed as Tulare. 16 For all of them he estimates a "late Cenozoic, probably Pleistocene," age. 17 Equus fragments : Canal station location right Ma 265 left upper cheek tooth 277 Mylontidae, cf. Paramylodon, distal portion of tibia and fibula 364 + 50 Camelidae, cf. Camelopsis, fragments of lower cheek tooth 265 Tapirus, cf. T. haysii Leidy, left maxillary with complete molar series 241 + 80 Quaternary Sediments and Land Forms During the Pleistocene the Diablo Range was further folded and uplifted ; and the Tulare beds in the foothill belt were tilted slightly to the east. Sandy alluvium has yielded Pleistocene vertebrate remains from the inter- 16 Tulare Lake is an historical playa-type sump ; the Tulare for- mation is a Plio-Pleistocene continental accumulation. 10 Savage, Don, personal communication, February 28, 1949. 17 Many of the fossils noted here and also those noted from the Quaternary terraces and alluvium, were collected by Inspector George E. Sharp of the staff of the U. S. Bureau of Reclamation at Tracy, California. stream stretch of canal between stations 336 and 363 (above berm height, and so for the most part still avail- able for study). This material occurs as a narrow strip, 20 to perhaps 35 feet thick, and was probably deposited as part of a former cone of Mountain House Creek, which at that time appears to have had a course, long since lost by piracy, almost eastward from the N£ cor. sec. 19. Clean coarse sands and pebble gravels, cross-bedded, extend downward as steepbanked channel fills well below berm elevation in several places. The present lower course of Mountain House Creek is bordered on both sides by terrace remnants which have yielded Pleistocene vertebrate fragments, almost certainly in place. The terrace north of the creek was excavated for the canal from station 282+50 to 295, or thereabouts ; and had a surface elevation between 190 and 200 feet. South of the creek the riser of a high terrace, tread elevation about 220, was trenched and formed the entire foundation of the canal from station 304 to near station 310, arid most of the left bank of the canal thence most of the way to station 318. North and east of the pump plant site Pleistocene beds 10-15 feet thick extended an unascertained distance along the canal. Extensions of the same blanket were exposed by the siphon cut to the south, and along the canal southward from station 446. The Pleistocene materials differ from the Tulare beds in being coarser, looser, and generally cleaner. Worn fragments of Ostrea bourgeoisie (from the Cierbo forma- tion which is exposed several miles to the southwest) are almost characteristic. Colors are chiefly grayish yellow to grayish orange. The upper 10 to 12 feet exposed south- ward from station 450 is a vagely bedded, firm, finely sandy and slightly clayey silt. Except for this, the Pleisto- cene sediments are loose to readily friable, channeled and cross-laminated (low angle), clean angular-grained sands with minor pebbly streaks and lenses. The coarser sands are ill-sorted and angular ; medium and fine sands are mostly well-sorted, and contain occasional clayey laminae and sporadic pebbles. The following vertebrate remains from Quaternary alluvium and terrace deposits exposed by the canal cut are reported by Savage. 18 His estimate of the age of these unsatisfactory fragments is also given. Canal station location Late Cenozoic, probably Pleistocene : Equus fragments of right cheek tooth and a lower left jaw with Ms 354 (above berm) upper molar 370 4 worn upper cheek teeth 447 Proboscidae, probably Mammuthus : fragments of vertebral centrum — 306 fragments of tusk 235 Pleistocene : Mammuthus, fragment of cheek tooth 296 Later Pleistocene : Bison, distal end of radius 295 + 80 fragments of a proboscidean tusk and bones (not catalogued) 133 — 65 The foothill belt, for a width of nearly 2 miles, appears to have been developed by incision and partial 18 Savage, Don, personal communication, February 28, 1949. 10 Special Report 2 destruction of a sloping erosional surface, the remnants of which rise from about an elevation of 275 to about 375 feet in the district north of Mountain House Creek. This belt is cut almost entirely in ill-consolidated sedimentary- rocks ; it exhibits, at slightly differing elevations, flatfish spur remnants which are not gravel-veneered. The con- tact of the higher inner margin of this belt, with the steep hill front is sinuous both in plan and profile. Stream lateral corrasion rather than faulting or differential ero- sion probably caused this physiographic disconformity. The sloping surface appears to decline northward because of the relation of its remnants to the present topography. South of Mountain House Creek, where its outer margin was not snubbed off by side-cutting streams during a Pleistocene humid episode, interstream surface remnants descend smoothly to the Great Valley. The foothill sur- face, like much of the sloping plain to the east, which now merges with the valley flat, is probably a slightly upwarped complex of rock fans and apron. Mountain House Creek has alternately cut and filled during its incision of the foothill belt ; with that exception, allu- vium is probably thin to absent over much of the flatly sloping valley margin. Pleistocene vertebrate remains have been taken from terraces of Mountain House Creek and from alluvium only slightly above the general level of the Great Valley at several places. The sloping foothill surface, which bevels Plio-Pleistocene and earlier strata, is probably of Pleistocene age. Structure The general structure of the Tulare beds is a homo- cline, whose northeasterly dip decreases valleyward from 15° to about 2°. Near Mountain House Road there appears to be a very flatly open north-plunging anticlinal nose. This structure is predicated on attitudes taken within 30 feet or so of station 232+40, and coincides with a local sharp eastward steepening of the dip of the Neroly formation beneath the unconformity. It is possible that a fault, hidden by recent alluvium, parallels the foothill front about a mile northeastward from the edge of the foothills. The epicenters of a few minor earthquakes each year appear to plot along a narrow belt extending northwestward from Bethany. 19 Further- more it can be shown that 10 miles to the southeast the valley has subsided independently of the foothills in Recent or late Pleistocene times, and by an amount meas- urable in hundreds of feet. A possibly active fault trends northwest into the Cretaceous terrane about li miles west-southwest of the " Byerly, P., personal communication. SCALE FOR PLAN 100 200 300 FEET Tulare J formati on [ SCALE FOR SECTION O 25 SO 100 200 FEET J I I LEGENO t'/*// Tuiore formation : pale yellow, soft sandstones, clayey siltstones, and silty clays. Sporadic ochreous pebble conglomerate lenses. Neroly formation: massive grayish -green bentonitic clay; few thin limestone beds. Neroly formation: gray, weakly cemented sandstone, tuffaceous silt, and minor andesitic pebble conglomerate. FiciiKE 5 Generalized geology of the canal cut, stations 191 to 211. Delta-Mendota Canal 11 canal. This minor break, the Midway fault, 20 is shown on Huey 's map of the Tesla quadrangle, which adjourns the Bethany quadrangle on the south. According to Huey it has an apparent relative eastern upthrow of about 100 feet near the north boundary of his area, and is marked southward, by "furrows, sag ponds and springs". It is clearly responsible for the low west-facing battered fault scarp which bounds Midway Valley on the east. Drainage pattern and land forms indicate that the foothill belt east of the Midway fault has experienced actual elevation, the district to the west has remained essentially unmoved. This, corresponding to the limited length of the fault, is a local, not a general, condition. Several relief features for which pre-Recent fault- ing would be an attractive explanation appear on the air photos and on the ground, westward from the canal and north of Highway 50, within the area shown by figure 2. Outcrops are very sparse and proof of faulting was not obtained. Engineering Geology As outcrops were essentially lacking, preconstruction investigations were largely limited to drilling operations. Initial choice of the location and the evaluation of exca- vation and foundation characteristics of the material down to canal grade were based on a core-drilling program car- ried on under geologic supervision. Altogether 61 exploratory holes were drilled north- ward from Mountain House Creek between January 1944 and January 1947, before construction started. These were cored throughout, and had a total footage of 5,007 feet. They furnished data for final choice of the pump plant site and gravity section alinement of the canal. Later the foundation of the pump plant was intensively drilled. Finally, during canal excavation, periodic inspec- tions were made, and a map and sections were prepared and submitted with specific recommendations. The geologic observations were directed toward answering questions regarding (1) excavation character- istics, (2) slope stabilities, and (3) foundation properties. A study of cores indicated that the entire cut for the canal would classify as ' ' common excavation ' ' ; that is, blasting would not be necessary. This was confirmed. The gravity section of the canal is being excavated by walking draglines (Moneghans) ; in most of the stretch here dis- cussed a 13-yard bucket and 200-foot boom were used. Slope stabilities were of special concern only in the tide-level stretch and the pump plant excavation, where the canal is to be left permanently unlined. Geologists' field reports and the laboratory study of cores were utilized by the designers. Special conditions arising from local aquifers exposed by the cut were reported and con- sidered, as was a possibility of heaving by plastic flow over a 30-foot clay bed found below bottom grade in one stretch. Questions regarding foundation characteristics have differed in the pump plant site and the gravity section. Because of the likelihood of earthquakes and the impera- tive necessity of continuing alinement of the pumps, the designers elected a pile foundation. Possible settling was not of controlling importance in this decision, as cores indicated suitably compacted materials. The design called 20 Vickery, F. P., Structural dynamics of the Livermore Basin: Jour. Geology, vol. 33, pp. 608-628, 1925. for 40-foot piles to be driven from the bottom of the excavation, about 80 feet below the original ground sur- face. During the final preconstruction phase Drilling Inspector Manuel Bonilla was directed to develop data to facilitate qualitative comparison of the difficulty of pile driving at the depths indicated, as against that of pile driving in the uppermost 40 feet. Bonilla 's solution of this problem was to slip 5-inch lengths of 3-inch diame- ter core, immediately on recovery, into a short length of casing, taking care that no distortion of the core occurred, yet that the seating was reasonably snug, to provide lateral support. A standard penetration resistance meter, with 0.1-inch needle, was then used on each such specimen. Sporadic sand and gravel lenses were not amendable to this treatment, and the number of tests run was not sufficient for detailed statistical analysis. 21 However, in spite of wide ranges in penetration resistance, it was pos- sible to estimate, and report, that the materials below a depth of 50 feet would offer approximately three times as great resistance to pile driving as those above that depth. Twelve test piles were then driven under the super- vision of Project Engineer George Imrie. It was found necessary to shoe all wooden piling and impossible to secure full specified depth (40') in eight cases where loading tests were made. Confirmation of the crude tests performed during core drilling was forthcoming when the foundation piling was driven. It was found impossible to drive these piles, and permission had to be granted the contractor to pre- drill all holes for piling at a diameter 2 inches less than that of the pile, to within 1 foot to 5 feet of the total depth. Even so, considerable difficulty was experienced. Sporadic lenses of well-cemented pebble conglomerate were encountered in the test holes and beneath the founda- tion cut, and were not included in the analysis, having been separately reported and their importance evaluated. Foundation problems in the gravity section of the canal centered on the recognition of possible plastic or swelling clays which might settle under bridge footings or heave sections of the concrete lining. As the strata dip appreciably in the stretch between stations 187 and 320, the possibility existed that abrupt changes in foundation character might subject the lining to differential strain. The excavation was therefore mapped, beds suspected of heaving potentialities located, sampled for laboratory study, and their attitudes determined. The most serious situation was that of Neroly clays exposed in several stretches between stations 187 and 212. Free swell values up to 135 percent were reported by the Denver laboratory technicians, and the material was definitely montmoril- lonitic, as had been suspected in the field. A few hours before the concrete lining in this stretch was poured a moderately heavy rain fell. As a result, the foundation swelled, during and after the period of set, sufficiently to bow up a bottom section of the lining for about 150 feet to an elevation 0.1- to 0.2-foot above normal grade. After the set was complete, the lining was cracked and lifted in several small areas. Beyond station 320 the northeasterly dip of the strata decreases rapidly, and beyond station 474 the canal -i This work was done in January 1947. Since then the Bureau of Reclamation has adopted and specifies the in situ determination of penetration resistances, making use of a Bx "spoon" and 140-pound hammer, in all explorations of ill-consolidated foundations. 12 Special Report 2 enters the fiat-lying alluvial sediments of the Great Valley. Under such circumstances, any changes in founda- tion character may be expected to be gradual, not abrupt as in the case of tilted beds. REFERENCES CITED Anderson, Ralph, and Pack, R. W., Geology and oil resources of (he west border of the San Joaquin Valley north of Coalinga : V. S. Geol. Survey Bull. 603, 220 pp., 1915. Beckwith, H. T., Tracy gas field: California Div. Mines Bull. 118, p. 587, 1943. Boden, (). G., Excavating for Central Valley's heartbeat: Excavating Engineer, vol. 42, no. 12, 1948. Clark, B. L., and Woodford, A. O., The geology and paleon- tology of the type section of the Meganos formation of California : California Univ., Dept. Geol. Sci., Bull., vol. 17, 1927. Daviess, S. N., Mineralogy of late Upper Cretaceous, Paleo- cene, and Eocene sandstones of the Los Banos District, west border of San Joaquin Valley, California : Am. Assoc. Petroleum Geologists Bull., vol. 30, pp. 63-82, 1946. Green, R. J., Nine million yards closer to Mendota : Excavating Engineer, vol. 43, no. 5, 1949. Huey, A. S., Geology of the Tesla quadrangle : California Div. Mines Bull. 140, 75 pp., 1948. Krumbein, W. C, and Pettijohn, F. J., Manual of sedimentary petrography, New York, D. Appleton-Century Co., 549 pp., 1938. Vickery, F. P., Structural dynamics of the Livermore Basin : Jour. Geology, vol. 33, pp. 608-628, 1925. 30663 7-50 2M