Loading...
4.0 Geologic And Geotechnical Characteristics of Landslide 4.0 GEOLOGIC AND GEOTECHNICAL CHARACTERISTICS OF LANDSLIDE IP 4.1 Geologic Setting and Properties of Bedrock Bedrock involved in the Portuguese Bend landslide occurs within the middle Miocene Altamira Shale member of the Monterey Formation. This unit was deposited in a marine environment when volcanism was active in the area.during the time interval between approximately 16 and 14.5 million years before present. Rocks younger than 14.5 million years old occur along upper Crenshaw Extension and across the top of the peninsula but they only play a passive role in the deeper seated landslides. During the time when the subject part of the Altamira Shale was being deposited, periodic volcanic eruptions produced ash which consolidated to form a type of rock called "tuff". After burial, ocean water in combination with volcanically derived heat caused much of the volcanic ash to alter into a montmorillonite clay. A tuff in which the ash (tiny glass • fragments) has been altered to clay is called "bentonite". Pure bentonite is one of the weakest geologic materials in existence. The PBL and other landslides in the area usually occur along bentonite beds within the Altamira Shale. For this investigation, the Portuguese Bend bentonite clay, which constitutes the slide plane material, is denoted as PB clay in this report. Rocks younger than 14.5 million years old within the Palos Verdes Hills contain some thin tuff beds but, generally speaking, these tuff beds have not been converted to bentonite. The strength of bentonite varies with its specific chemical composition and with its purity - both of which depend upon its depositional environment. Volcanic ash was generally deposited by one of two possible mechanisms: (1) by atmospheric eruptions; or (2) by undersea eruptions in the shallow near-shore water. Atmospheric eruptions blew the ash into the air from which it fell (airfall ash) onto the ocean surface and, hence, through the water to the ocean floor. Ash of this origin generally forms thin layers consisting of tiny flakes of volcanic glass that lack admixed materials of other origins. When this glass is 19 11 converted to clay following burial, it produces the pure bentonite which is exceptionally weak. Volcanic eruptions that occur in shallow water mix volcanic ash, pumice and rock fragments with water. The dense fluid mixture flows down the submerged flank of the volcano as a turbidity current and onto the adjacent basin floor where the volcanic debris is deposited. A turbidity current generally picks up material as it moves across the sea floor and mixes it with the volcanic debris. As the turbidity current slows down, the coarsest debris settles out first and the finest settles last. The coarsest part of the deposit tends to be gritty and have a comparatively higher frictional resistance, even after volcanic glass has been converted to clay. Volcanic ash is the main constituent of the upper part of the deposit but the ash is likely to contain admixed diatoms and other fine-grained material picked up from the sea floor. When this ash alters to montmorillonite clay, the diatoms alter to the silica minerals cristobalite, tridymite and chalcedony. These minerals • form abrasive agents which increase the frictional resistance (angle of internal friction) of the bentonite. Thus, bentonite formed from turbidite deposits tends to be stronger than bentonite formed from airfall ash. The non-explosive part of the volcanic eruptions formed basalt lava flows and intrusions. When hot basalt lava flows onto the sea floor, its outer surface quickly solidifies in contact with ocean water. The mass of solidified lava has a tuberous or pillow shape and is called pillow basalt. A mass of pillow basalt has about the same properties as a pile of interlocking concrete blocks. Basalt within each pillow has a high angle of internal friction but the pile of pillows comes apart when subjected to tensile stress. The most common occurrence of basalt is in sills that were intruded along bedding planes in sedimentary strata. Basaltic magma is more dense than most sedimentary strata; consequently, it can intrude along weak bedding planes and float the overlying sedimentary rock above it. Although magma along the contact with sediments is quickly cooled and solidifies, magma may continue to be emplaced within the interior of the sill while floating • its roof above it. Sills are often thick in the vicinity of the magma source but decrease in 20 • thickness away from that source. This creates topography on the sea floor and causes sediments to be deposited in topographic hollows between sills. During subsequent compaction, the sedimentary materials will be highly compressible relative to the basalt. This creates an undulatory geologic structure of compactional origin that is superimposed upon the geologic structure of deformational origin. In addition, sills tend to be very strong and are unlikely to have slide surfaces pass through them. As a result, slide planes tend to pass above or below sills. The undulatory feature described above exists in the northern and central portions of PBL. Rocks of volcanic origin comprise less than half of the rocks present in the Altamira Shale. The majority of the Altamira shale unit is derived from Catalina Schist debris and biogenic materials, particularly diatoms. Most of this material has a high frictional resistance to sliding. In the PBL part of the Palos Verdes peninsula, the Catalina Schist debris came from an island located northeast of the landslide area. This debris consists mainly of sand • and silt composed of quartz, sodium feldspar and other granular minerals. These are abrasive and generally do not alter to weaker minerals, such as clay. As a result, strata derived from this material do not host slide planes. Sandstone derived from Catalina Schist was originally cemented by dolomite. This cement has been removed by solution within the weathering zone. This has left fine-grained unconsolidated sand that is abundant above the base of the uphill part of the PBL. The sandy material can be an aquifer of moderate to low permeability. Biogenic material in the shale comes from organisms that lived in the ocean. The most abundant constituent was diatoms, unicellular algae that had delicate porous skeletons composed of amorphous silicon dioxide (opal). Following deposition, the combination of volcanic heat and the weight of overlying strata produced successive changes in the diatom skeletons. First the amorphous opal changed to the ultra finely crystallized silica minerals, cristobalite and tridymite. Eventually much of it changed to fine-grained quartz (variety chalcedony). The process involves progressive hardening and compaction of the strata. • Strata composed mainly of cristobalite and tridymite appear similar to unglazed porcelain 21 4110 and are called porcellanite. Porcellanite is abundant within the sequence of strata that are involved in the PBL. It tends to be abrasive and does not host slide planes but it does hold large amounts of water. Unweathered porcellanite contains an ultra fine-grained network of pore spaces containing organic matter. The organic matter readily oxidizes to carbon dioxide and water. This leaves tiny interconnected pores that hold large amounts of water but are too small to allow water to flow out as free groundwater. (For this reason, dried porcellanite is used as kitty litter.) Tree roots can remove water from porcellanite through capillary action and vapor diffusion but wells can not, which is another reason why some of the wells do not produce much water even though the surrounding porcellanite appears to be saturated. Other materials of biogenic origin include fish scales, chert and phosphate nodules. Fish scales remain well preserved in the strata as long as oxygen is absent. During weathering, the fish scales oxidize leaving pore spaces that can hold water. Chert is a glassy nonporous rock formed when pure diatomite (a rock composed of diatom skeletons) is converted to ultra fine-grained chalcedony. Chert is hard but brittle and has a high frictional resistance to sliding. The phosphate nodules are generally lens shaped, less than an inch long and occur in siltstone and mudstone. They consist of soft cream-colored fluorapatite and were formed where calcium phosphate was concentrated by the reprocessing of organic matter by bottom dwellers and anaerobic organisms. However, the presence of these nodules has had no perceptible effect on the development and propagation of the landslides in the area. Impure dolostone is an important rock type within Altamira Shale found in the surrounding areas but is a minor rock type within the Portuguese Bend area. The mineral dolomite [CaMg(CO3)2] is the dominant constituent of dolostone. Dolostone was formed by precipitation of dolomite in what were initially permeable beds composed of sand-sized grains. The dolostone occurs as beds and lenses that are generally in the range between 0.2 to 5 feet thick. Dolostone is hard like concrete and tends to break into blocks up to • several feet long. 22 • In summary, bentonite is the only truly weak material within the bedrock in the Portuguese Bend area. Major landslide surfaces are generally restricted to bentonite beds, especially those derived from airfall ash. In addition, bentonite has a tremendous capacity to absorb and store water but has a very low permeability. Its ability to hold water is determined by its chemistry and the amount of compressive stress applied to it. The montmorillonite clay which comprises bentonite contains exchangeable sodium, calcium and other cations. Clay rich in sodium holds more water and is weaker than clay rich in calcium. Calcium within the bentonite was typically introduced during prolonged weathering. Consequently, calcium-rich clay is more abundant in deeply weathered uphill areas and sodium-rich clay is more abundant in unweathered bedrock and in downhill areas where weathering is less extensive and where groundwater tends to leach calcium from the clay. The bentonite found along the PBL sliding surface contains primarily sodium with lesser amounts of calcium. As will be discussed in the geotechnical engineering and strength section, the types of cations present within the bentonite, and the capacity of the clay to exchange one • type of cation for another, have a significant influence on the shear strength of that clay. 4.2 Geologic Structure and Stratigraphy Prior to the current study, there were lingering uncertainties about the geologic structure (the three dimensional geometric shape) of the base of the PBL primarily due to inconsistencies in the locations of early borings caused by landslide movement between the time the referenced topographic maps were made and the time of the subsurface investigation. Considerable uncertainties also existed regarding the geometric arrangement of strata in the bedrock beneath the slide. The bedrock structure beneath the slide must be known in order to evaluate the effect of any POC which involves regrading. For example, if the underlying bedrock does not have the same configuration as the base of the slide or consists of weaker material than that found at the existing sliding plane, it is possible for a new and deeper slide base to develop beneath the regraded surface. 1 23 110 Though limited in scope, the drilling program(see Section 3.2 and Appendix B)completed as part of this investigation has located the base of PBL in all borings with an estimated vertical accuracy of about one foot (based on the 1995 topography maps adjusted to movement since the map was made). Direct measurements of the base of the slide were made in bucket auger borings while for each of the four core holes the slide plane was determined from the cored samples. In holes drilled by air rotary methods, the slide base was marked by a change from yellowish brown oxidized slide debris containing sheared bentonite to gray unoxidized bedrock fragments. There was also a corresponding change from easy drilling within the relatively soft and broken slide debris to more difficult drilling within the harder and more competent underlying bedrock. Two inch diameter PVC casing was installed within the air rotary borings to confirm the position of the rupture surface. A tape measure was then used to measure the depth at which movement subsequently bent and closed this casing. The elevations at which the rupture surface was encountered in each of the 1996 borings are summarized in the following table: • • 24 1 Table 4.2. Geologic Contact Elevations from 1996 Borings Rupture Surface(RS) Portuguese Tuff(PT) Surface Boring Elevation Elevation Depth Top RS to PT Base Thickness (feet) (feet) (feet) Elevation (feet) Elevation (feet) PBS-1 179.6 4 175.6 -29 33 -91 62 PBS-2 156.1 60 96.1 32.1 27.9 -30 62 PBS-3 215.6 96 119.6 69.6 26.4 11 59 PBS-4 249.3 152 97.3 127.3 29.7 74 53 PBS-5 287.8 ' 201 86.8 173.8 27.2 116 58 PBS-6 310.2 222 88.2 179 43 121 58 PBS-7 298.6 269 29.6 221.1 48 161 60 PBS-8 281.2 209 72.2 180.2 29 116 64 PBS-9 185.1 98 87.1 78.1 20 12 66 PBS-10 170.4 44 126.4 14.4 30 -39 53 PBS-11 279.5 269 10.5 234.5 35 170 65 PBS-12 340.9 313 27.9 269.9 43 212 58 96W-1 276.2 247 29.2 * -- - -- fa96W-2 279.2 247 32.2 * - -- - B96-1 190 150 40 * - - -- B96-2 276 256 20 * - - - B96-3 477.4 399.4 78 * -- -- - B96-4 464.7 378.7 86 * - - -- B96-5 437.5 375.5 62 * - - - B96-6 433.5 353.5 80 * - - -- B96-7 434.8 361.8 73 * - -- -- B96-8 436.7 374.7 62 * - -- - B96-9 454 401 53 * - - -- B96-10 185.1 158.1 27 * - - -- B96-11 186.7 162.7 24 * -- - -- B96-12 293.5 283 10.5 * -- -- -- B96-13 289.3 279 10.3 * -- -- -- B96-14 284 247 37 * - -- - B96-15 313.8 237.8 26 * - - -- B96-16 283.8 242.8 41 * -- -- -- B96-17 282.3 241.3 41 * - -- -- • B96-18 291.8 227.6 64.2 * -- - Note: Borings not deep enough to reach PT bed. 25 Global Positioning System (GPS) measurements were used to locate the surface coordinates and the vertical elevation of each of the 1996 borings with an accuracy of better than one foot. It is our opinion that there is good control on the location of the slide base in the area north of Palos Verdes Drive South and east of Portuguese Canyon as a result of this study. Since no borings were made in the developed area west of Portuguese Canyon, the data available for this area are considered sparse. However, only minor grading to restore proper surface drainage is contemplated for this area and no analytical model for stability analysis is in the scope of the present study. Contours of the elevation of the PBL basal rupture surface are presented in Figure 4.1 along with the data points (boring locations) upon which they are based. In the areas where discrepancies exist, data from the new borings has been relied upon. As shown in this figure, the slide base has relatively steep seaward dips of 15 to 25 degrees along its uphill(northern)edge. The rupture surface flattens to a seaward dip of less than 6 degrees • in an anticlinal undulation that extends westward along the uphill edge of the central Subslides. Seaward of this anticlinal undulation, the seaward dip of the rupture surface steepens, especially along the east side of the slide where this surface forms a basin which extends a few feet below sea level. The ramp between the anticlinal flat and the basin has seaward dips as steep as 17 degrees. On the west side of this basin just east of Portuguese Canyon, the rupture surface is 55 feet above sea level on the north side of Palos Verdes Drive South and the maximum seaward dip between here and the anticlinal undulation is about 9 degrees. The structure contours do not differ significantly from those on earlier maps except for the placement of a north-south striking fault that crosses Palos Verdes Drive South along the general alignment of Portuguese Canyon. The rate of movement on the east side of this fault has historically been roughly twice as fast as that on the west side. Accordingly, the trace of this fault is visible along the ground surface. The fault apparently dips steeply to the base of the slide where it is marked by a vertical discontinuity in the elevation of the • rupture surface. This fault serves as the natural boundary of the East and West Central 26 PBL Subslides, as was described earlier in this report(Figure 1.2) and in the geologic map (Plate 1) which is in the pocket of this report. The new borings were especially helpful in determining the relationship between the base of the slide and the underlying sequence of strata. Uncertainty had existed because masses of tuff occur within the slide from its uphill edge all the way to the beach. The borings indicated that the Portuguese Tuff is at depth beneath the slide base throughout the northern portion of PBL. Thus, tuff within the slide comes from the area locateduphill from the head of the slide. Ancient sliding must have uprooted the seaward dipping tuff on the slopes bounding Crenshaw Extension to the east of Portuguese Canyon. Then, seconcy sliding moved the tuff onto the PBL. This "conveyer belt" system reloaded the head of the slide and caused recurrent movement in the same way that the 1995 secondary slide reactivated the ancient landslide movement. This understanding helps to develop a geotechnical model, such as the slide plane shown in Figures 2.7 and 2.8 as will be • discussed later. In all but the most northern portion of the slide, the rupture surface occurs along a sheared bentonite bed located about 30 to 40 feet above the top of the Portuguese Tuff, as indicated in Table 4.2, above. The bentonite bed at the base of the slide is the same bed that caused recent sliding onto Yacht Harbor Drive immediately east of Klondike Canyon, see geologic cross section D-D' (in pocket of this report). Here, the bentonite is about three inches thick, intensely sheared, and is nearly free of gritty inclusions. This bentonite most likely formed from the airfall ash, as described earlier. The same bentonite bed appears to form the base of the secondary landslide that occurred at the uphill edge of the Portuguese Bend landslide in 1995 (Figure 4.2), except that the tuff is thicker and less sheared at this location. The roughly 30 feet of strata between the slide base and the Portuguese Tuff consists • mainly of strong strata derived from Catalina Schist debris and siliceous biogenic material. 27 A few bentonitic layers are present but they do not appear to be especially weak because of impurities as described in Section 4.1. The Portuguese Tuff is typically about 60 feet thick beneath the PBL (Table 4.2). It is significantly thicker than any other tuff units within the Palos Verdes Peninsula and therefore forms an easily recognized reference horizon. It was probably formed by a tremendous underwater eruption to the southwest of this area. Ash, pumice and volcanic rock fragments were mixed with water and then transported into this area by turbidity currents. The lower two-thirds of the Portuguese Tuff is coarse-grained and contains extensive gritty material that give a fairly high frictional resistance to sliding. The upper third was mainly derived from ash. The uppermost part of the tuff appears to have been redeposited by erosion of ash from slopes surrounding the basin. It contains thin interbeds rich in biogenic material that create uncertainty regarding the precise top of the Portuguese Tuff. The Portuguese Tuff is commonly referred to as bentonite because nearly all ash • and pumice in it has been altered to clay. The Portuguese Tuff is underlain by strata derived from Catalina Schist and biogenic material with tuff beds scattered throughout the unit. The sequence contains intrusions and extrusions of basalt at varying depths below the Portuguese Tuff. For example, at the beach next to Portuguese Canyon, basalt was found to occur almost immediately below the tuff in a core hole funded by the City of Ranch Palos Verdes in 1996. However, the first basalt found uphill from the head of the slide is pillow basalt which occurs roughly 200 feet below the tuff. The stratigraphic details of this sequence are important in the area of seaward dipping beds located uphill from the Portuguese Bend landslide but are not as significant beneath the landslide because the beds are confined by overlying strata and are therefore not as susceptible to sliding. The undulating shape of the landslide base is controlled by the structure of the underlying bedrock. The east edge of the slide is defined by a sharp upturn in the bedrock which also 4111 forms the ridge immediately east of PBL. This upturn may mark the west edge of the 28 basalt which is exposed within the abandoned quarry area and along the invert of Klondike Canyon to the east. The conduit that brought basalt to the surface probably did not extend further west than Klondike Canyon. Because basalt undergoes little consolidation relative to the marine sediments, the structural embayment which extends westward from Klondike Canyon was probably formed during late middle Miocene by differential compaction of these sediments. The western limit of the structural embayment that contains the ancient PBL complex also appears to be controlled by basalt in a similar manner. Three anticlinal (arched upward) undulations extend westward beneath PBL. One is located immediately south of Palos Verdes Drive South. The other passes beneath the uphill part of the slide. The axis of the northern anticline crosses immediately seaward of the intersection of Sweetbay Road with Peppertree Drive. A third anticlinal arch extends in an east-west direction to the north of the slide beneath the downhill side of Peacock Hill. The seaward flank of this arch forms the ramp at the uphill margin of PBL. • The anticlinal arches are asymmetric folds with short uphill limbs and long downhill limbs. Their "wave length , the distance between the undulations, is roughly two thousand feet. These folds probably formed during the early stages of development of the regional Palos Verdes anticline. The southward (seaward) tilting of this flank of the Palos Verdes Peninsula causes the anticlinal arches to have relatively flat north(uphill)limbs and seaward dipping ramps on their south (downhill) limbs. As discussed above, these folds are superimposed upon structures formed by differential compaction of the sedimentary rocks relative to basalt. The combination of these structures produces the undulations which occur along the base of the PBL, see geologic cross sections (Plate 3) in the pocket. 4.3 Subslides: A General Geotechnical Engineering Assessment The buttressing effect of Inspiration Point (Figure 1.3) divides the PBL into two parts. The smaller part of about 65 acres, which is not dealt with here, moves into the cove west of Inspiration Point. The larger part of about 200 acres moves into Portuguese Bend to • the east of Inspiration Point. This eastern part has an average thickness of about 100 feet, 29 an average length of about 3,200 feet and an average width of about 2,700 feet. As discussed previously, PBL is a relatively thin and flexible blanket with a length and width to thickness ratio of roughly 30:1. Undulations in the rupture surface and the fault dictated by geologic structure divide the eastern part of PBL into the four Subslides previously described as (1) the Landward subslide, (2) the East-Central subslide, (3) the West-Central subslide and (4)the Coastal or Seaward subslide. These undulations and the associated boundaries between each of the four Subslides maintain relatively fixed positions as slide material moves seaward. The relative position and lateral extent of these Subslides have been shown in Figures 1.2 and 4.1. The Landward subslide extends in a crescent shape around the west and north margins of the west-central subslide and the north (uphill) side of the east-central subslide. This subslide has an area of about 50 acres. The rupture surface beneath the uphill part of this subslide dips relatively steeply seaward at inclinations of 15 to 25 degrees (see Figure 41) 2.7). As a result, the slide mass generates much more driving force than resisting force in this region. On the downhill side of the Landward subslide, the ramp flattens to dips of less than 5 degrees across the previously described anticlinal undulation in the underlying bedrock. Here, the slide mass generates a net resisting force as long as the water table is low. The East-Central subslide covers an area of about 60 acres. Its uphill boundary is marked by a west trending graben that continually regenerates at the same location as slide material moves from the flat anticlinal undulation in the slide base on the north into an area where the rupture surface dips as much as 17 degrees seaward on the south. Slide material moving down this ramp generates the majority of the driving force which has propagated movement of this subslide. The ramp is steepest on its east side where the rupture surface extends into a spoon-shaped trough. This trough has a long axis oriented north-south and a gently sloping bottom that extends slightly below sea level landward from Palos Verdes Drive South and rises to a few feet above sea level seaward of Palos Verdes Drive South. 30 1111, The seaward dipping ramp on the west side of the East-Central subslide is considerably flatter with an average inclination of only 7 degrees as opposed to 15+ degrees on the east side of the subslide. As indicated previously, the western edge of this subslide is defined by a near vertical fault which extends in a north-south direction along the general alignment of Portuguese Canyon. The canyon probably developed by following a rift along the fault. The fault is controlled by a discontinuity in the underlying bedrock structure. The eastern side of the East-Central subslide terminates against upturned bedrock along the adjacent ridge. This bedrock strikes in a north-south direction and dips steeply to the west forming a dip-slope along much of the west face of this ridge. The West-Central subslide includes about 40 acres west of Portuguese Canyon. The rupture surface dips more gently beneath this subslide than beneath the East-Central subslide. At the downhill margin of the West-Central subslide, the rupture surface is marked by a gentle synclinal-anticlinal pair of west trending undulations that tend to • buttress the subslide. These features have contributed to the historically lower rate of movement for the western portion of the landslide. The Seaward subslide includes about 45 acres of land that is mainly seaward of Palos Verdes Drive South. It consists of rotational slide blocks that remain active because of continuous wave erosion along its seaward edge. Mitigating this wave erosion will be a necessary component of any long term plan to stabilize this portion of the landslide. Continued movement of the Seaward subslide has a tendency to remove support from the upslope East- and West-Central Subslides. From a geotechnical engineering viewpoint, the above discussion of Subslides substantiates that the first step in attempting to "stabilize" PBL consists of decreasing the driving force in the Landward slide; arresting the fast-moving East-Central subslide; preventing the withdrawal of lateral support by the Seaward subslide together with other stability • enhancing measures such as lowering groundwater and strength enhancement of the PB 31 110 clay. Accordingly, the geotechnical analyses are focused in this manner. In the following sections, groundwater and soil strength issues are reviewed and evaluated. 4.4 Groundwater Soil moisture is present throughout the slide material from the ground surface to the base of the slide. Under saturated conditions, the soil moisture capacity of slide material is generally on the order of 25% to 40% by weight or between 3 and 6 gallons per cubic foot of soil. The amount of soil moisture remains constant below the groundwater surface. However, moisture levels can vary significantly within the vadose zone above the groundwater surface. The soil moisture content in the vadose zone generally reaches a maximum value following periods of high rainfall in winter and early spring. Conversely, soil moisture reaches a minimum following dry periods, particularly at the end of the dry season in late fall. During the dry season, unirrigated surface soil loses much of its • moisture due to transpiration by plants and surface evaporation. Deeper soils may also have a portion of their absorbed water removed by transpiration -provided deep rooted plants are present. Groundwater is stored in openings large enough to prevent ionic and molecular forces from holding the water as soil moisture. The PBL debris and underlying bedrock typically contain a high percentage of fine-grained particles which have a relatively high capacity to absorb and retain water, yet only a small portion of this water will drain from the soil under the force of gravity. On the average, only about 20% of this water is free liquid, i.e., will drain by gravity from the soil if the groundwater table is lowered. At present the groundwater table is above the basal rupture surface throughout nearly all of the PBL. Figure 4.2 (and Plate 2) shows contours representing the approximate elevation of the groundwater table based on measurements made during the summer of • 1996. This figure also shows the locations of wells upon which these contours are based. Groundwater is perched above the landslide base because bentonite along the base of the 32 • slide is nearly impermeable and the underlying bedrock generally lacks groundwater storage capacity and permeability. The water table has its highest elevation at about 335 feet near the uphill edge of the landslide and slopes seaward at an average gradient of about 10%. The general direction of groundwater flow is southerly and down gradient. However, groundwater flow follows a circuitous route because secondary slide surfaces form barriers of low permeability and zones of extension have above average permeability and both are oriented at a high angle to the groundwater gradient. The water table is known to be higher today than it was when the slide began moving in 1956. However, only sparse data are available for the period prior to the installation of wells in 1984, see Figure 2.2. During the year after sliding began in 1956, twenty-two holes were drilled to the slide base in the area uphill from Palos Verdes Drive South, see Appendix A. Of these, groundwater was reported above the slide base in only six holes. Three were in the developed area west of Portuguese Canyon and three were east of the • canyon. Of those to the west, two were bucket auger borings (DW-2 and DW-5)that were downhole logged by Ehlig. In boring DW-2, located near the northwest corner of the slide, groundwater was found at one foot above the slide base at an elevation of 326 feet above sea level. At present, monitoring well B88-4, located 180 feet south of DW-2, has the water table at an elevation of 331 feet. Another hole, EE, located 220 feet southwest of B88-4, encountered the water table at an elevation of 315 feet. Further south near Cherry Hill Lane, DW-5 encountered the water table at an elevation of 137 feet. This is about 40 feet above the slide base and appears to be close to the present groundwater table projected from nearby wells. However, another boring, DW-3, located 450 northwest of DW-5, penetrated the slide base at an elevation of 147 feet without encountering groundwater. Most holes drilled east of Portuguese Canyon during the year after sliding began did not report groundwater above the slide base. One of these, DW-6, located near the east • central edge of the slide, was a bucket auger boring that was downhole logged by Ehlig. This hole was covered and kept open for several weeks to see the offset on the slide base. 33 No water was present in this hole at any time. Of the three holes that contained groundwater above the slide base, KK and Z were immediately north of Palos Verdes Drive. Their logs report the water table at elevation of 67 and 30 feet, respectively. At present, the water table is about 70 feet higher in this area, see Appendix B. The third hole (V), where the water table was reported above the slide base, was located in the uphill part of the slide near PBS-6. Its water table was at elevation 255 feet, whereas PBS-6 has the present water table at 269 feet. Other 1956-57 borings probably penetrated the water table above the slide base but passed through slide material of low permeability. However, it was our understanding that these holes may have not stood open long enough for the water level in the hole to reach equilibrium. Following activation of the PBL in 1956, surface drainage was quickly disrupted by the formation of fissures and grabens. This facilitated the accumulation of groundwater. By 1968, the groundwater table was near its present level based on scattered data from • borings. The high rainfall during the period from 1977-78 through 1982-83 caused a significant rise in the water table that accelerated the rate of slide movement. In 1984, wells were drilled to monitor and remove groundwater and in 1986 and 1988, Phase I and Phase II grading restored surface drainage in much of the slide area. This resulted in a lowering of the water table. However, in recent years, high rainfall in combination with drainage problems has resulted in significant groundwater recharge suggesting that relying on the present system for controlling groundwater is inadequate. During recent years (Appendix B), there have been fourteen producing wells in parts of the landslide. Of these, four (W6A2, Figueroa, Peppertree and Nancy) intercept groundwater entering the slide from the northwest. Collectively, they produced an average of 16,042 gallons per day during the first six months of 1996. Three wells (Tangerine, Cherryhill West and Fischer) are in the developed area west of Portuguese Canyon. Collectively, they produced an average of 2,131 gallons per day during the first six months of 1996. Of seven wells east of Portuguese Canyon, W6B and three sump • wells are in the uphill part of the slide. Collectively, W6B, West Sump and Central sump 34 411 produced an average of 1,054 gallons per day during the first six months of 1996. East sump was out of order much of this time but it has recently been producing about 250 gallons per day. Of the four wells in the southeastern part of the slide, only three have operational meters (Ish. EC, Ish. SE and Ish. W). Collectively, the three have produced an average of 689 gallons per day during the first six months of 1996. Ish. NE lacks an operational flow meter but is known to be a poor producer. The current average production from all fourteen wells is about 20,000 gallons per day. This compares with an average production of about 220,000 gallons per day from eighteen wells in the adjacent Abalone Cove Landslide Abatement District. Of these wells, six individually produce nearly as much or more water than the collective production from the fourteen wells. The disparity in groundwater production is the result of bentonite that clogs slide debris below the water table in the PBL. The low production is important in that there appears to be a limit below which wells in the PBL are ineffective at lowering • the water table. Underground drainage galleries were suggested, but not investigated because of concern over cost and installation of galleries in a moving mass. Therefore, a special effort is needed to limit groundwater recharge to as small an amount as possible. In summary, groundwaterelevations based upon data from borings and standpipes installed during the late 1950s following the reactivation of the landslide are listed in Appendix A and the locations of these borings and standpipes are shown in Figure 2.2. Groundwater elevations, based on borings and observation wells, are listed in Appendices A and B, and their locations are shown in Figure 2.1. Comparison of the groundwater table between 1955 and 1996 indicated that the groundwater levels in the northern and western portion of the landslide are generally similar to those of 1996. However, the 1955 groundwater levels in the eastern portion of the landslide appear to have been much lower than those which exist today. Hence, the rise in groundwater levels in the eastern portion of the landslide which has occurred since 1955 is a significant factor contributing to the • continued movement of the PBL and needs to be evaluated. 35 111 Contours representing the estimated current depth of submergence of the basal rupture surface beneath the surface of the groundwater table are presented in Figure 4.3. These contours are estimated from the summer of 1996 groundwater surface readings and the basal failure surface elevation contours presented in Figure 4.1. As shown, all but the extreme Landward margin of the rupture surface is presently submerged. It is important to note that the areas with the highest hydrostatic pressures correspond with depressions in the rupture surface and occur in the east-central and southwestern portions of the landslide. The maximum level of submergence in these areas is on the order of 50 to 100 feet. The average piezometric head acting on the rupture surface within the limits of the landslide is approximately 40 feet. The mean depth of the landslide debris is on the order of 100 feet. Accordingly, on the average, the lower 40 feet of this slide debris is saturated. Note that the contour maps related to depth of submergence are based on the limited amount of available data and are, therefore, approximate. • Groundwater elevations have periodically been measured in four wells located within the northern portion of the active landslide since the mid to late 1980s. The locations of these wells are shown in Figure 2.1 and the recorded groundwater elevations for each well are given in Appendix B and summarized in Figures 4.4 through 4.7. Monthly and annual precipitation data for the area are also presented in these figures for comparison. The groundwater levels recorded for these wells exhibit a wide variation in their response to precipitation patterns. All wells have perceptible, and in some cases substantial, decreases in groundwater levels which appear to be associated with the drought conditions and/or surface drainage improvements of the late 1980s, as described earlier. The groundwater levels in two of these wells (W-2D and W-4E) currently remain four and twenty-five feet, respectively, below their 1984 levels. The groundwater levels in these two wells reached their minimum levels in 1993 or 1994 and have increased slightly since then. The groundwater elevations in the remaining wells are presently one to five feet above the levels recorded at the time of their installation in 1988. • 36 tip The two most easterly monitoring wells (B88-5 and B88-9) show significant short-term fluctuations in response to periods of high precipitation. Well B88-5 near the northeast margin of the slide has had the most pronounced fluctuations with the water level in the well casing increasing approximately 15 to 25 feet following one or more months of high rainfall. The response of these two wells appears to be indicative of relatively rapid rates of infiltration in the northern portion of the landslide and suggest that this is an area in which mitigation of surface and subsurface drainage warrants attention in the POC. As indicated previously, the available data suggest that the permeability of the landslide debris is generally low. The average effective hydraulic conductivity of the saturated deposits within the landslide typically appears to be on the order of 0.2 foot per day (7 x 10-5 cm/sec) based upon calculations using production rates from dewatering wells operated during the mid-1980s. Field observations and laboratory data suggest that a significant portion of this apparent permeability is due to flow along cracks and fissures • within the landslide debris. The results of laboratory permeability tests on small samples (i.e., approximately 2.5-inches in diameter) typically indicate hydraulic conductivities one to two orders of magnitude lower than values calculated based on well yields. The hydraulic conductivities back-calculated from the production rates of these wells (except Cherry Hill) are summarized in the following table: 37 ,4. Table 4.4. Estimated Hydraulic Conductivity From Well Production Typical Average Thickness of Range of Back- Well Production Rate Saturated Slide Calculated Hydraulic (gpm) Debris Conductivity W-6A2 7 75 0.85 W-6B 0.3 50 0.06 W-6I 0.5 45 0.14 • W-6J 0.3 25 0.17 W-6K 1.7 70 0.26 I-SE 1 100 0.08 Nancy 1.4 80 0.17 Average: 0.25 ft/day Typical groundwater flow velocities towards the ocean through the voids of the landslide material are difficult to estimate and one cannot generalize hydraulic conductivities • calculated above to the entire PBL. Groundwater flows much faster through cracks and fissures. The cracks and fissures, however, do not constitute a continuous network throughout the slide mass to the ocean. Recharge of the groundwater within the landslide occurs through the following four primary mechanisms: 1. Direct precipitation and infiltration through the surface of the landslide. 2. The drainage of offsite water onto the surface of the landslide and the subsequent infiltration of this water; 3. The infiltration of water from private sewage disposal systems installed _ . within the limits of the landslide; • 4. The flow of groundwater into the landslide region. 38 r10 During periods of heavy rainfall, large quantities of runoff flow onto the landslide from the tributary canyons, especially Paintbrush Canyon as discussed previously. Although the water from these canyons is conveyed across the landslide through a combination of natural and improved drainage courses, it appears that significant quantities of this run-off still infiltrate into the ground within and around the periphery of the failure. The evaluation of the regional groundwater basin needed to determine the groundwater budget of PBL is not within the scope of this study nor do we believe it is warranted at this point in time. Some general comments can be made, however, to the extent which is pertinent to the goal of this study. For example, approximately 56 million gallons of precipitation falls on the surface of the 200 acre landslide during an average year of 10.4-inches of rainfall. Even though a varying amount of this water is lost through evapotranspiration, infiltration of rainfall into the landslide is significant due to fissuring and generally poor drainage conditions within and around the margins of PBL. On a relative basis, approximately two dozen residences are present within the western portion of the landslide. Assuming an average 250 gpd per residence recharge to the groundwater, this amounts to about 2.2 million gallons per year. Although substantial, this quantity is relatively small compared to the infiltration from rainfall, plus the inflow from drainage courses north of the slide. Significant inflow of groundwater comes from the extensional zone of the northwestern area of the slide and the northeastern area which was described earlier from the well data. For the groundwater, which is already within PBL boundary, the gradient and inflow or outflow across the slide plane is important from a geotechnical engineering viewpoint. In this regard, three multi-stage pneumatic piezometers were installed as part of the current investigation. Schematics illustrating the configuration of each installation and the recorded piezometric levels are provided in Figure 4.8. As shown, each installation consisted of three separate pneumatic piezometer tips grouted in-place within a single bore hole. One of the tips was positioned 10 to 20 feet beneath the basal rupture surface while • the other two tips were placed at various intervals above that surface. The data from two 39 ,.. northerly installations indicates the piezometric level above the rupture surface in that area is approximately one foot higher than the piezometric level beneath that surface. These data indicate that the downward flow across the rupture surface and out of the landslide in this area is negligible. The most southerly installation (BYA-1) shows the opposite tendency. BYA-1 was installed in an area where the rupture surface dips steeply downward. Data collected from this location indicates approximately 10 feet of hydraulic head beneath the rupture surface and little or no head above the surface. The data indicate potential uplift forces and a tendency for water to flow upwards across the rupture surface and into the landslide in this area. Considering the general lack of data on hydraulic head measured across the slide surface, the data obtained from BYA-1 is significant. However, the extent of this upward trend is unknown. For these reasons, it is recommended to include BYA-1, a multi-stage pneumatic piezometers, be included in the City's regular monitoring program. • 4.5 Shear Strength and Engineering Properties 4.5.1 Overview Identification of the shear strength characteristics and behavior of the bentonitic clay within which movement is taking place represents a crucial component of this investigation. The conventional simplified manner of expressing soil shear strength (ti) is by a linear function of the effective overburden pressure (a'), and cohesion c and friction angle (4) components: i = c + a' tang (1) In the case where large shear displacements have occurred, such as in landslides, residual strength parameters are usually used: 110 T = Cr + a' tan 4r. (2) 40 The subscript "r" is used to denote the residual strength, i.e., a residual strength after the soil has experienced large shear displacement such as the PB clay after the 1956 landslide. Another assumption used in equations (1) and (2) in describing soil strength is that, the soil behaves in a rigid-plastic manner. This is to say that the soil, when under stress, either behaves rigidly and without deformation when the stress is less than a certain level, or deforms plastically without limit when this level is exceeded. In reality, few materials behave in this manner. Soil deforms under stress. When the stress is small, it deforms imperceptibly, but it is not rigid. As stress increases, soil deformation accelerates. Depending upon the type of soil, it may deform continuously without any apparent increase in stress if the stress applied is large enough. This is commonly referred to as soil creep. The conventional simplified equations (1) and(2)are adopted for convenience and are not necessarily • factual or realistic. While there are certain types of soil or rock whose behavior can be reasonably approximated by equations (1) and (2), these simplified strength models are not applicable to PB clay. PBL moves and creeps nearly continuously, even though the rate of movement responds to the rainfall and remedial grading episodes. Therefore, it is clear that a conventional simplified rigid-plastic type of equation is insufficient for analyses evaluating the feasibility of "stabilizing" a site like PBL. In the present investigation, BYA has utilized three separate approaches to identify the shear strength characteristics of the PB clay which occurs along the rupture surface. These include: S 41 1. Research and evaluation of previous testing results generated by other investigators for samples of bentonitic clay obtained from both within, and outside of, the PBL area; 2. The completion of a series of laboratory tests on samples of bentonite clay collected from the basal rupture surface. These procedures included Atterberg limits, gradation, specific gravity, moisture/density, cation exchange potential, x-ray diffraction, energy dispersive X-ray spectroscopy (EDS)and large displacement ring shear tests. 3. The back-calculation of anticipated landslide factors of safety and displacement rates based upon the adopted shear strength parameters followed by the comparison of these results to the • observed behavior of the landslide at various points in time. Each of these approaches will be discussed separately in the following sections. 4.5.2 Previous Testing Results Since the reactivation of PBL in 1956, the sliding prone material - generally referred to as Palos Verdes bentonite has been tested by a number of organizations. The majority of these tests have been performed in conjunction with land development projects on samples collected outside the limits of PBL. The residual shear strength of Palos Verdes bentonite was measured in conjunction with at least 38 separate investigations between 1957 to 1992 (Watry and Ehlig, 1992; Vander Linden, 1972). While these results are typically not site specific to PBL, the material tested shares a similar geologic genesis. In order to provide a frame of reference, BYA performed a statistical analyses of the available data. The results • of this analysis are summarized in Table 4.5.2. 42 Table 4.5.2. Statistical Evaluation of Palos Verdes Bentonite Samples from 1957 - 1992 Soil Parameter/Statistical Plasticity Index Residual Friction Residual Parameter (PO Angle(degrees) Cohesion �r (Psi) Mean 54 10.5 147 Standard Deviation 1.3 1.5 2.2 10% probability of 35 5.4 41 being less than 10% probability of 82 20.8 526 being greater than Data Population 53 39 32 Log-normal distribution of the data provided best fit, as shown in Figures 4.9 and 4.10 for PI and 4)r. •• The following observations can be made with reference to this table and these figures: • The results of the plasticity index of Palos Verdes (PB) bentonite appears to be readily useful from an engineering application perspective. For example, a plasticity index(P.I.) = 54 could be considered a representative sample for Palos Verdes bentonite, which put the soil classification most likely in a high plastic clay range (CH) even though with a 90% probability that the soil could have a PI value range from 35 to 82. • The same cannot be said of residual friction(4)r)based on these test results from an engineering analysis perspective. For example, Table 4.5.2 shows that there is a 90% probability that the residual friction angle (4)r) of PB bentonite varies from 5.4° to 20.8°. This variation(5.4° to 20.8°) is much too wide to be useful. A wrongful choice between 5.4° to 8° could theoretically change frictional resistance thereby affecting the calculated the 43 factor of safety from 1.0 to 1.5. Hence, for this project, the strength of the slide material has to be tested and better defined. A considerable amount of literature and testing data has been generated regarding physio-chemical and engineering characteristics of montmorillonitic clays in the past four decades. Published results on the residual shear strength of pure montmorillonitic clays represent another source of information which was considered as part of this investigation. Several selected studies were identified as part of the current investigation and are cited in the reference section of this report. The results of these studies provide valuable insight into the general behavior and chemical composition of these clays. One of the most significant characteristics documented in this body of work is the strong influence that the chemical composition of the clay-water system can have on the shear strength of • those clays. This dependancy required that the chemical composition of PB clay be better defined before any useful correlation with published results could be attempted. The chemical composition of the PB clay and the resulting shear strength correlations will be discussed in subsequent sections of this report. 4.5.3 Laboratory Testing Results For the reasons outlined above, it was concluded that laboratory testing of bentonite samples collected from the basal rupture surface of PBL would be necessary to fulfill the objectives of this investigation. The testing program which was subsequently developed involved the following tasks: 1. The collection of bentonite from the rupture surface at 16 locations • either from borings or surface samples within the limits of the active failure. 44 2. The completion of laboratory index tests (moisture content, Atterburg limits, gradation, etc.). The closest match with the statistical mean (plasticity index = 53) is the sample taken at the southwestern part of PBL (see Figure 4.1, identified as PB-2). 3. The selection of a representative sample based upon steps 1 and 2, above, followed by detailed testing of that sample to determine its chemical composition and shear strength characteristics. These tests included the following: a. X-ray diffraction to identify the crystalline structure of the sample and confirm the specific type of clay which was present; • b. Energy dispersive X-ray spectroscopy (EDS)to identify the specific elements which are present within the clay; c. Cation exchange potential to measure the potential mobility of the primary cations which are present within the clay; d. Scanning electron microscopy (SEM) to identify and evaluate the microscopic surface texture and particle orientation along the rupture surface; e. Large displacement ring shear tests to directly measure the residual shear strength of the clay under different simulated burial depths and displacement rates. These work items have been completed and the results are discussed in this section and Section 5.2, Geotechnical Analysis of Proposed Improvements. 45 The variation of moisture, density and Atterburg limits of the bentonite clay samples tested is indicated below: Moisture Content: 20% to 60% Dry Density: 57 to 92 pcf Liquid Limit = 110% +10% Plastic Index = 56 ±6 Following the completion of these index tests, a representative sample of the bentonitic clay from the rupture surface was selected for more detailed testing. Samples (PB-2, Figure 4.1) of bentonite collected from a surface exposure of the rupture surface in the southwest portion of the landslide were selected for this testing based upon(1) the representative appearance and index properties for this sample; (2) the large quantity of sample which had been obtained from this • location; and (3) the accessibility of this location for the collection of additional samples in the future, if necessary. Clay from this location was subsequently subjected to a series of laboratory tests to measure various geotechnical parameters. As outlined above, these tests included scanning electron micrograph, x-ray diffraction, cation exchange potential, energy dispersive X-ray spectroscopy (EDS). The results of these tests are summarized in the following table: Table 4.5.3.1. Summary of Geotechnical Parameters of a PB Clay Sample (PB-2) Used For Analysis Parameter Results Reference Physical Description Tan, highly plastic clay with a waxy appearance Sodium Montmorillonite XRD(x-ray diffraction) [Na0.3(A1,Mg)2Si4O10(OH)2 Figure 4.11 - x 1120] Exchangeable Sodium 13,200 mg/kg (EPA 9081) 46 r Parameter Results Reference • SEM (Scanning Aligned clay platlets along Figures 4.12; 4.15 Electron Microscopy) slickensided surface EDS (Energy Dispersive cation/anion content Figure 4.13 Spectroscopy) Dry Density 58 pcf Moisture Content 60%+ Specific Gravity 2.65 • Liquid Limit 110% Plastic Limit 54% Plastic Index 56 Ring shear testing was selected as the method of choice for measuring the residual shear strength parameters of the bentonitic clay. This test method involves torsionally shearing a ring of soil in a circular direction as shown schematically in Figure 4.14. The ring shear apparatus used for this testing was designed, • fabricated, and calibrated by BYA. It should be acknowledged that the concept of the Ring Shear Test(RST) was first developed in 1937 by Professor Juul Hvorslev and its principle and testing protocol for the RST were advanced through joint research between the Imperial College, UK, and the Norwegian Geotechnical Institute (NGI) in the early 1970s. The RST has many advantages over the conventional direct shear or triaxial shear machines. A key advantage is that the RST is capable of measuring "soil shearing resistance and the slow plastic flow before failure and the temporary or permanent decrease of the shearing resistance after failure" (A. Bishop, 1971) at very large displacements. This capability makes RST uniquely suitable to test the soil residual strength in landslide investigations. Bentonite from sample location PB-2 was tested in BYA's ring shear apparatus under a variety of normal stresses and displacement rates. The results for these tests are summarized in the Table 4.5.3.2. 47 • Table 4.5.3.2. Summary of Ring Shear Results for 15 in./day Displacement Rate Normal Stress Consolidation Peak Shear Residual Shear Test No. (psf) Pressure(psi) Stress (psf) Stress(psi) 250 8 10,000 - 169 500 8 10,000 - 215 1,000 8 10,000 - 315 1 10,000 1,350 487 6 10,000 - 475 2,500 7 10,000 - 510 11 10,000 - 465 15 10,000 1,600 326 2,800 5 10,000 2,079 705 1 10,000 - 700 4 20,000 3,000 913 • 5,000 13 10,000 2,608 741 15 10,000 - 696 7,500 15 10,000 - 900 1 10,000 - 870 10,000 - - — 15 10,000 - 1,033 It is interesting to compare the naturally slickensided surface (Figure 4.12) with that after ring shear testing (Figure 4.15). Note the polished surface of parallel clay platelets produced by shear displacement as well as the clay particles below the sheared surface. The natural slickensided sample and the lab sample are comparable under similar magnifying power and angle of view, suggesting the pertinence of using ring shear tests. Each ring shear sample was remolded at 10% above the in-situ moisture content and then over-consolidated in the ring shear testing apparatus at a normal stress of 10 to 20 ksf. After consolidation, the normal stress was reduced to the specified level and the sample was allowed to . II equilibrate. Each sample was then sheared at a displacement rate of 15 inches per 48 • day until both a peak and subsequently a relatively constant residual resistance level were attained. At that point in the testing process, either the normal stress or the rate of displacement was typically changed and shearing was continued until a new constant residual stress level was attained. For most tests, this process was repeated for several different normal stresses or displacement rates using the same sample. During the testing process, the normal stress, shear stress and sample displacement were electronically monitored and recorded using a computerized data acquisition system. An example plot illustrating shear stress and normal stress versus displacement is presented in Figure 4.16. As shown, the ratio between the shear stress and normal stress imposed on the sample (i.e., ti/v) is also presented in this figure. Note that each of the overconsolidated samples produced a relatively well defined peak shear strength after approximately 0.1 inches of displacement. After this peak, shear strength decreased significantly with further displacement as the failure surface developed and became more defined and polished. A nearly • steady state residual stress level was generally achieved after three to four inches of total displacement. However, the shear strength of the samples often continued to decrease slightly with continued displacement beyond this range. Shearing was continued for one sample until a total displacement of 460 inches (38 feet) was reached. For this sample, no further decrease in shear strength was perceptible beyond a displacement of approximately 20 feet. Cyclical variations in shear strength were typically noted as the samples were rotationally sheared in the ring shear testing apparatus, with each cycle corresponding to one full rotation of the sample. This pattern is visible in the shear stress versus displacement plots. Based upon examination of the samples after the completion of each test, the cyclic pattern is believed to be attributable to small variations in the surface profile and minor alignment of the actuator and other hardware, which in our opinion, needs to be refined for additional testing. However, while refinements are necessary, they are not expected to affect the general conclusion of the testing results which are presented below. 49 111 The ring shear test data and the associated peak and residual failure envelopes for the PB clay samples are presented graphically in Figures 4.17 and 4.18, respectively. The shear strengths presented in these figures represent those associated with a displacement rate of 15:inches per day. As shown and as expected in Figure 4.17, the peak shear strength for this material is dependant upon both the preconsolidation pressure and the normal stress under which shearing actually occurs (i.e., dependency on the overconsolidation ratio, OCR). The residual shear strengths of the samples tested did not show any perceptible dependence upon the pre-consolidation pressure, as anticipated. As can be seen in Figure 4.18, the residual shear strength is nonlinear with respect to the normal stress level which is applied during shearing. The slope of the failure envelope or the effective residual friction angle at low normal stress levels (i.e., <500 psf) for this material is on the order of 11.5°. At higher stress levels (5,000 to 10,000 psf), this angle decreases to approximately 6.5°. • As will be presented in Section 6 of this report, in-situ lime injection into the sodium montmorillonite rich PB clay is being considered as a potential strength enhancement alternative to stabilize PBL. Figure 4.17 included peak strength test results when the same PB clay samples were mixed with 6% lime. Figure 4.18 confirms that the simplified conventional linear relationship between overburden and shear strength (Equation 2) does not provide an accurate representation of the shear strength envelope of PB clay. Accordingly, a best fit polynomial equation shown below was utilized to represent the non-linear relationship between overburden and residual shear strength: i = c + (kl • (7) + (k2 • CP) (Egn. 3a) Where, • 50 ti = residual shear strength a = effective overburden or normal stress c = soil cohesion k1, k2, and k3 = constants "Best fit" values for the four constants in this equation(i.e., C, k1, k2, and k3)have been calculated based upon the test results which are presented in Figure 4.18. These constants and the resulting equation are as follows: till = 125 psf + (0.67 • a) - (0.304 • a107) (Eqn. 3b) In the above equation, t15 represents the residual shear strength at a displacement rate of 15 inches per day. The units associated with this expression are pounds per square foot (psf). • It was anticipated that the residual shear strength of the bentonite would exhibit some dependency on the rate of shearing due to "viscous" behavior which is rate dependent. To evaluate the degree of this dependency, the residual shear strength of several samples was measured at multiple displacement rates. The results of these measurements are summarized in the following table: Table 4.5.3.3. Summary of Shear Tests to Define Rate Dependency Displacement Rate Relative Shear (In./Day) Test No. Strength(%) 1 0.917 4 0.945 1 5 0.951 10 0.943 • 11 0.952 15 0.938 51 Displacement Rate Test No. Relative Shear (In./Day) Strength(%) 10 1 0.985 15 All 1.000 45 11 1.019 65 10 1.019 80 11 1.038 275 1 1.089 10 1.349 2,000 11 1.400 3,500 6 1.525 5,000 10 1.600 11 1.547 The shear strengths listed in Table 4.5.3.3 were measured sequentially for • relatively small displacements. In the rightmost column of Table 4.5.3.3, the shear strength of each sample has been normalized to the strength measured at a displacement rate of 15 inches per day. This accounts for the variation in strength between separate samples and allows a more direct comparison of rate effects. The normalized shear strengths presented have been plotted against the corresponding displacement rate in Figure 4.19. As shown, the residual shear strength increases non-linearly with higher displacement rates. As discussed in Section 4.5.1, the conventional rigid-plastic assumption on soil shear strength(Equations 1 and 2) is not capable of accounting for sample viscosity or the effect of displacement rate on shear strength. In the current investigation, this deficiency has been overcome by introducing a shear displacement rate or velocity term into the general residual shear strength equation, as follows: _ Trate = f(Rate) • [125 psf + (0.67 • a) - (0.304 • a1.07)] (Eqn. 4a) 52 1111 In this expression f(Rate) represents a variable which is a function of the displacement velocity (Rate). For this investigation, f(Rate)was defined to express the residual shear strength at any displacement rate relative to the residual shear strength at a displacement rate of 15 inches per day. For the bentonite samples tested, parameters have been calculated for a polynomial equation which provide the "best fit" for this relationship. This equation is as follows: f(Rate) = 0.93 + 0.024 • (Rate)°38 (Eqn. 4b) Where "Rate" represents the displacement rate in inches per day. The f(Rate) represents an adjustment function applied to the soil strength which is determined at the standard testing rate of fifteen inches per day. Equation 4b provides a means to account for residual shear strength reduction at low rates of displacement based upon values measured at a higher laboratory testing rate. As can be seen from this • equation, as the displacement rate approaches "zero", the effective shear strength is projected to be about 93% of that measured at the tested fifteen inches per day displacement rate. It should be noted that the choice of using 15-inch/day displacement rate is somewhat arbitrary. The circumference of the RST sample is about 30 inches (see Figure 4.14). We chose one half (1 ) revolution of shear per day as a starting point of the testing program to save time (i.e., 15-inch per day). For confirmatory testing in the future, it is recommended to use a rate of 1 to 1.5 inches per day or slower to test the validity and variability of Equation (4b). 4.5.4 Comparison of Strength Data of PB Clay with Montmorillonites For a period of about 15 years between 1960 to 1975, research on clays was conducted at the University of Illinois and the Massachusetts Institute of Technology under the primary sponsorship of the National Science Foundation and 53 U.S. Army Corps of Engineers Waterways Experiment Station (WES), among others. Portions of these research projects were conducted on montmorillonite with emphasis on the montmorillonite-water system and strength [e.g., Mesri and Olson(1970), Ladd, et. al. (1977)]. Considering the rich sodium-montmorillonite content in the PB clay (Table 4.5.3.1) and the limited scope of BYA's laboratory tests, it is considered essential to validate the curve-linear strength relationship of the PB clay, such as Figure 4.18, with published results of both sodium- montmorillonite and calcium-montmorillonite. This comparison is shown on Figure 4.20, which illustrates that the strength envelope measured by BYA for PB clay is consistent in its curve-linear configuration and lies approximately' in between the reported strength envelopes for pure calcium and pure sodium- montmorillonites. With respect to BYA's findings, the behavior of PB clay shows that under reduced • overburden pressure, the strength changes. Figure 4.21 illustrates that when a sample of PB clay is sheared under a reduced overburden pressure, a peak is first observed and the difference between the peak and the residual strength increases as the effective normal stress decreases. When the overburden pressure is increased, no such peak occurs. This phenomenon is significant in that the shearing resistance to be overcome when a landslide is reactivated on a pre-existing slide plane may exceed the residual value if the movement is associated with a decrease in effective overburden pressure (e.g., regrading by cut as proposed in the POC). Similar findings were noted in brown London clay (Bishop, et. al., 1971). For example, Figure 4.22 shows ring shear test results on brown London clay conducted at the Imperial College. • 1 Note that theoretically, no direct comparison can be made because Mesri's data were obtained in triaxial shear tests and plotted on p-q coordinate. BYA's results were tested in RST and plotted on a-T coordinate. No closed-form formula can be related to these two coordinates systems for a non-linear strength relationship. For general discussion on p-q and CY-T, see Lambie and Whitman(1969), for example. 54 .• The practical significance of this fmding lies in planning for the regrading operation. With a reduction of overburden pressure by cut or removal of soil above a slide surface, movement along which was arrested before the cut, the slide material will have a higher resistance to slide reactivation. In other words, by decreasing the normal force above the sliding surface, the PB clay will have a higher resistance to slide reactivation. The magnitude of the apparent increase in shearing resistance has to be tested to a level more than what is shown here in Figure 4.21. 4.5.5 Back-Calculation From Landslide Behavior In order to evaluate the reasonableness of the available data and associated interpretation, a two dimensional slope stability analysis program has been used to back-calculate factors of safety for the East-Central Subslide at various points in • time. This program uses Spencer's Method modified by BYA to incorporate the curve-linear residual strength relationship (Equation 3b) and to account for interslice forces within the failure mass. Equation (4b) from the previous section was also incorporated into BYA's computer software in order to model the displacement rate of the bentonite along the failure surface under a variety of past, and potential future, conditions. Cross sections C-C' and D-D' (Figure 2.8), which transect the eastern portion of PBL have been utilized in our analysis. Analysis of these cross sections have been completed for three separate episodes since PBL reactivation in 1956. The approximate dates of these episodes and the significance of each, are summarized below: • 55 • Table 4.5.5.1. Time Frame of the Three Episodes Used for Analysis Date Rate of Groundwater External Source of Data Movement Level Factors 1956 High Low Original Boring and/or caisson (2+ inches/ topography at excavations by Los Angeles day) time of County, Palos Verdes reactivation. Properties, the Donald R. Warren Company, and (Figure 2.2 and (Figure 2.2, Mackintosh&Mackintosh (Figure 3.2) (Appendix B) Figure 4.1) (Figure 2.2). 1984 to Moderate Moderate to High High rainfall Survey data, borings and 1988 (=1 inches/ during the early monitoring wells by Fhlig day) 1980's. Prior during the mid to late to dewatering 1980's and grading (Figure 2.4 and activities by (Figure 3.3) Appendix B) Ehlig. (Figures 2.4, 2.6). 1995 to Slow Moderate After the Survey data, borings, 1996 (=0.2 inches/ dewatering and monitoring wells, and day) regrading piezometers by activities of the Ehlig and BYA (Figures 3.3, (Figures 4.2 and late 1980's. (Figure 2.1 and • 3.4) Appendix B) Appendix B) These three episodes were selected for analysis based upon the availability of data for each time frame as well as characteristic differences in the prevailing conditions at the site (e.g., variations in groundwater levels, displacement rates, topography, etc.) which allow for an approximate evaluation of the general validity of the model. The results of this portion of the stability analysis are summarized in the following table: • 56 Table 4.5.5.2. Calculated Factors of Safety with Rates of Movement Incorporated Approx. Displacement Shear Weight Combined Strength Cross Factor of Relative Date Rate Rate Factor Section Safety °ns per Mass factor of (Inches/Day) (��� Foot of Safety Width) C-C' 0.87 15,500 49.4% 1956 2 + 0.97 1.00 D-D' 1.12 15,900 50.6% C-C' 0.88 17,400 48.7% 1984 1 - 0.95 0.99 D-D' 1.09 18,350 51.3% C-C' 0.87 15,450 49.8% 1995 to 0.2- 94.2% 1.00 1996 D-D' 1.12 16,200 50.2% The displacement rates shown in the second column of Table 4.5.5.1 represent the average displacement rate of the landslide at the time of the associated analysis • based upon the available survey data. Using equation(4) from Section 4.5.3, these rates were used to calculate a shear strength rate variable to account for the observed viscous behavior of PB clay along slide plane. The factor of safety which was calculated for each cross section is presented in the fifth column. An overall factor of safety for the Seaward subslide (last column) was calculated based upon the proportional mass of the two cross sections and is presented in the last column. Preliminary observations based upon this portion of the stability analysis can be summarized as follows: • Generally speaking, the response predicted by the computer model is consistent with the observed behavior of the eastern portion of PBL where cross sections C-C' and D-D' were analyzed. The model appears to be able to quantify the effects of grading or topographic changes, variations in the • groundwater elevation, and variations in the rate of displacement with a reasonable degree of accuracy. As with any "back-calculation" the 57 potential for offsetting errors exists in the analysis and refinement of the model is needed as additional data is collected. • The calculated factors of safety for cross section C-C', which is through the western portion of the subslide, are consistently on the order of 20% lower than those for the eastern portion of the failure (cross section D-D'). This appears to be attributable to a more uniform inclination of the failure surface along this section. • Between 1956 and 1984, the landslide moved seaward more than 500 feet and the head of the failure decreased in elevation by approximately 60 feet (Figure 3.5). This significantly decreased the amount of force which was driving the failure. However, the available data indicate that groundwater levels rose substantially during this same period-in some areas by as much • as 100 feet or more. The down-dropping of the Landward portion of the landslide created a depression to increase water infiltration as well as contributed to the submergence of the slide plane under the water table. The stability analysis indicates that the lower driving forces were largely offset by reduced shear resistance resulting from the higher piezometric levels along the rupture service. The net effect of these changes was the continued movement of the failure at a lower rate of displacement. • The two phases of re-grading (see Figure 3.5) which occurred in late 1986 and early 1988, respectively, are responsible for the approximately 3% increase in the calculated factor of safety for cross section D-D' between 1984 and 1995, i.e., from 1.09 to 1.12. Except for short term variations following periods of high rainfall, available data indicate that fluctuations in the groundwater elevation within the landslide during this period were generally small and had no significant impact on the factors of safety. Although, the reduction of the factor of safety (0.01) for cross section C-C' 58 • between 1984 and 1995 may be considered too small to be certain, available data and our analysis suggested that it is attributable to the reduction of shear strength as a result of the decreased rate of movement (i.e., 0.2 versus 1 inch/day). • Based upon the behavior of the bentonite samples which were tested as part of this investigation and the current average rate of movement of the failure (i.e., approximately 0.2-inches/day), it is estimated that the effective shear strength of the bentonite along the rupture surface may be reduced by an additional 1.4% due to viscosity effects if the displacement rate of the landslide were to approach "zero". As is always the case, the quantity and quality of data which is available to support the analysis of this landslide is less than ideal. Additional laboratory work is • necessary to refine and extend this analysis to the entire PBL. However, the field observations, laboratory data, and analytical evaluation are encouragingly consistent. Incorporating residual strength as a function of overburden and the rate of displacement represents a pragmatic and logical step towards evaluating means to reduce the rate of movement to zero. Data regarding groundwater levels during the 1950s, 60s and 70s is particularly sparse (see Figures 2.2 to 2.4 ) and/or inconsistent because of apparent variations in drilling and measurement procedures. Nonetheless, the previous data and that generated by this investigation appear to provide a reasonable basis for development of a preliminary analytical model for the current level of analysis. Accordingly, the model is adopted to analyze the proposed POC and other stabilization alternatives for PBL. These analyses are presented in Section 5. • 59 $ '8o------/ ' S� ,. \\� �� • ) LLL Fiysyq` w , slir 4, •B96-3 / 4001?) .� 4?/ •B9. 9 (4 40 p 00:70 ` 896-4 •, • 374 X28/ • •p A� WPT •B8 3 1 S •` • U •B3 _._ i k'EE�.4 I. 67 696- • 7 4 • B96-7 3 511111 WGAair26 <317 • —N4.020,000 231® • N4,020,4•• ytP • '" * °PBS- ••8996-12 3 8963 .• O 24/ 69 B96-13 10 �WN BYA- if <5' 233 • 2968 2/318 B96-14•• 696- - 0,0,-,..4•243 •� G't • PBS-6 896- • •222 •242 y230 W2C `�32 3_1 •239/ •96- 96W- 2 BYA-2 244 c\ • 96W-1• • \\ \, J8,4 W4E 247 •+ 340 * 2 219 280— 6PBS-11 • •6• ............ � CC 2 896-1 + h PBS-5 •217 2 2 . h �W2F - I- __� ' ®201PBS-:, 21 (*) 20• "-Z�_e% ,j 02) IV �` ��PF _____,7- Portuguese Can • •205 FQO ��F X280 /y a Y 18/ W 178 Kaioil 78 118 . %F.._ .. ---160---- 896-11 a - - • 62 �' 16 DW-3 �- :. PBS-4 / YA-0 896- . 152 _� I58 - '� 14/ 696- .-2.. 1 ---?� -- / 50 C---/// ' 24O --All - J o 12 8v. 7___--- �' 2H r\ .....,14--- 104.....- _ ��--\ r ( 85 :B 99 <100 ... , 57 ?__ DW �0 a — 'i ) 24‘-- -_.--.?"--- 4 ------ ------ ----8 ,D !. 0 Q 220__ v v 2� �� ivb (,, 2•—� ChZc - , APB o is)-r •W6K -P,py ---60. �._ ..... b 44 :NoLL !N'• 1 WCH PBS-2 KK izit 1 :, 60 •57 r t r r n• i i r i r r • '���PALOS VERDE r r ?4o1 8,000 r N4,018,9•• 5• • SDR/VE SOUTH rt PBS-1 74 DD i , 4 AV- W21. ---- i 56 - 0 1 ao-� � i.�_ ► r7 eW2M '�•7....,= � t GG __ .,---- ,2 A r 'it e lhialii C.„" o> • -444 - C_ • C-11 E-3, 9 29 �, 2A&2 Ado. ', .•A r� -27 DW -, ••22 10�C-2 7 -:.---' G -27 0• C 1 no I1 B A,1,2 . 3 (.7------- (3 B 1 NI"r� X41 111 ,01- qv3 \.. 4 -2' ��- . DW 1 . 4 , \• - ii y-9,QV a•W�Ij , , \C�O ------;_____ ______±._,_ • DW 8• / is , . sit. PACIFIC OCEAN 20 +Pp LEGEND ____80____ BYA-1 PIEZOMETER INSTALLED BY BYA 1996 1111 B-96-18• 1996 SOIL BORING PB-1 • BULK SOIL SAMPLE H •'�-1 , - PBS-2• BORING DRILLED IN 1996 IP W4E• WELLS MONITORED FROM 1984-96 ELEVATION CONTOUR FOR BASAL WCH O WELLS MONITORED FROM 1984-86 0 300 600 FEET (IN__20-- RUFTURE SURFACE IN FEET FROM 96W-10 1996 BORING I (Base Map, 1995) MEAN SEA LEVEL BB• 1956-57 BORING 0■V BING YEN & ASSOCIATES, INC. F40 1956-57 CAISSON bH4 GeoreehiIcel&Environmental Consultants,Esta&lshf979 APPROXIMATE ELEVATION CONTOURS FOR PROJECT NAME: PORTUGUESE BEND LANDSLIDE BASAL RUPTURE SURFACE PROJECT NO:G-940989 I DATE: MARCH 1997 FIGURE 4.1 --------------------480------Y )) - --- \, -/i) c I. V \ \IT, AIIIIIIIINjlIlo4"*p SNS s6o� P �4 '0 w11111k 4. Aripe S'e ' l'O 4 tp,,, -kiow fiiQ2, :96-3 ` <359 % 8 `42o 11* SECONDARY1111111111k • <3 ;� SLIDE 3/1995 c 6-4 •<36 SECO DAR Y,\.'IL � • 12 jos SLIDE /1995 di WPT ®B88-4 7% MU Al$y,FE� iiik ® 330 ,� � , • � AyR� 696-�� <31 I ? A •, . N4,020,OOQ N4,020,/ , W6A' q B, 300- :�.�� B96- � � \ �� I -33 B96-12 49 j LANDWARD SUBSLIDE , x'2967 286 ' 4 B96-13 `i 3;i t W6B • <27: 688-5 • N • EiN-L, _/-----267'i WSE.• 258 3 � �,Q� WSC • • B96 B96-16 2 GD � ' 1 F' 2 ,10B962- � 26 Ga • ..... PBS-6 F •--'•. I • 11, - • fp 258 Q�� •896-17 .� �`,. •• ♦ :YA-2 5 xI :961 .7 9• -2 /,,✓ 8 WSW ,� �i= ♦ '•• • '.5W-1 B96-2 �,'• •W4E� .-" , ••. -257\ W2D I �' ... 253-.7� I,,'IPyg,}L-? .' • 41 B96-1: • 1 PBS-5 >22 'Do I •234 ailti ' 1t /PBS-: (3 B88-9 • di \ \ R� A093. !� `. J 300 //� ) I 1 i_Travinio k117•.--i0 Q? 'lJIP,' / • 7. •-----••T. ZQ 1 , •nrfl y ' o Cony 11111111r o t 696-11 • • 1,6.1 • `� -28• ®PBS-4 •B96.10 174 G 50 56 2.• WLT , /*I J is 69a- ► / 150 IC -�.� 40 ...--__ r / \ 0 ,,---,„ ..... 4irr-\ .\\ ._...._ A ( , _WI1\IIP PBS-9 EAST—CE TRAL SUB - LIDE 142' c:\V it 240 / 0 .15.3 22. OPBS-10 WI, / 4 �,-----W6K%.. • r WF 114 • l r-180--\ 1‘ 18• • W12 N$ 2B. WEST—CENTRAL SUBSLIDE (iiiii° t 80 4111.111 �`__ 10 J ....... �, W 13 p.....___I_..7_. N4,018.SIS I —__.._ �`���-� PALO VERD; , ' NA •':S-1 ) N4,018,000 -- 3PSOUTH 102 / / • • illiPmilm'r -41111114......,,w ,-----\ 41„ 4. (____ Is .S1.0 • V% *ib.,_ i111111140, . ii 126 1.s Iq, 120 s l •.,.,, SEAWARD SUBSLIDE i , / -- . .....-4397„ . 4 60 v I Irfillit '— PB-2)00„__ 111.114111b• S'O/ PACIFIC OCEAN • �.p O Ar LEGEND ' ---80— TOPOGRAPHIC CONTOURS BVA 1 O PIEZOMETER INSTALLED BY BYA 1996 �` IN FEET(MSL) B_96-18• 1996 SOIL BORING D' ` .•''••......•• APPROXIMATE SLIDE BOUNDARY PB-1• BULK SOIL SAMPLE , �,"' TOE OF SLIDE PBS-2• BORING DRILLED IN 1996 \ DIRT ROAD W4E• WELLS MONITORED FROM 1984-96 ROAD 0 300 600 FEET N. I IIIIII BUILDING —50— GROUNDWATER CONTOUR 0■� BING YEN & ASSOCIATES, INC. IN FEET(MSL):DOTTED WHERE :�: Geotechnlcel&Environmental Consultants,Established 1979 APPROXIMATE OCTOBER 1996 ESTIMATED GROUNDWATER SURFACE PROJECT NAME: PORTUGUESE BEND LANDSLIDE ELEVATION CONTOURS (Base Map, 1995) PROJECT NO: G-940989 I DATE: MARCH 1997, FIGURE 4.2 • y i -----------------480---...----/ • d 5 �� y8) ` \660 PVP-II ��A_ �) I \ 4,0 .`i------1,460 'Sym s6o� i-,7 ---A. soo��0 `�flo 42. Lt siIr •B96-3 <359 • \420 ?f D • OB96-r <390 -, 00 illibilio, 16. •B96-, 28/ •P <36 :S-12 v 4401 the arsk ®6:- • y •B -5 C -BAY,„ EE 330 f064 <358 •B96-7 •96-8 •?...-__ 26i • <317 <367 W W•i42 WF o • 96020,OOQ —N4,020,000 PBS-70 •696-12 y -)...",_ 333 / ('5cZP 296 286 3:. p gal/ W66 696-13�, • 2g33� YA_3 ?p 0., ,� • WSC 6,5 +•258 cpm _ ^14.) \89rIA B9CHANN 4 •-256 6 �.� . PBS-6 t •�696-1 �`• 258 2 W2C >240 `00 360 �, .. __ BYA-2 0 •'V .• ___.' o \96W-2 250 � • 696-1 `. •WSW ♦• 246 ?, •`s6W-1 • C d• \\\X\_, �• ,896-2\ • v V •,r >, � W4 ��♦♦�• 1 3 •9 •...• J&4 254_-- i .i •.� PBS-11-.- 216: G� 25 - 4.0.00....280--- •696:11': �i a 0 �� `.>225' s �� p� .7"---..g. `0'.#02B20 5 265 ••W2F • 1' PBS-8 <20 , ��9 : : lI % :'ice' - _p_ 30�WT t1/4 0 2 1/4 *s • . c 20s 141 (010 ,pQ "�' •44. O Y�,a, 18/ W 2:e I / • "'� - - 161.5 �W-3 s• j 160 i �.B96-10 1 • ePBS-4 ,,� g 150" �.. -�.(0) .ilia -----4 )?. 174 \\�/ a 140 - -; --- -� -- . - 14- ---.� • 6-1 • .. - 1 50 _ _ - = 224 222- : . . : . . . . . . . . . : . . O 220- • ; i i • w218 � .� . . . , • • 216 . . . •. .• •. .• •. . . . . •. .• •. •. 214 . . . . . . . . . . . . . . . . . . . . . . . . . . . • 22- •• • • • • • • • • • • • • • • • • . .• .• .• .• .• .• .• . . . .• .• .• .• . 20 .• 16- : : : : : : : : : : : : . . . • . . . . . . . . •. . . . . . . . . •. •. •. •. •. •. •. . . •. •. •. •. • . 16- . G, a, . . •. . - L . . . . . . . . • . . . . 14- : • • : • • • • . • • • • • • • v 12 • • a a 10- . Z Z H ~ 8 Q Q . • cc cc • • • 2(� 1 • t l I n '� '(�'�,_Th _ �� Fill rn n : • -.n _ n 1 1, . 1/171980 1981 1982 1983 1984 1985 1986 1987 1988 1989 � 1991 1992 ;993 1994'L 185 1996 DATE MOD BING YEN & ASSOCIATES, INC mom Geotechnical& Environmental Consultants FLUCTUATION OF GROUNDWATER ELEVATION IN MONITORING WELL 2D PROJECT NAME:PORTUGUESE BEND LANDSLIDE Note:Rain Gage Station LAFCD 44 PROJECT NO.G94-0989 DATE:MARCH 1997 FIGURE 4.4 290 • • • • . . . . . •. .• •. . . •. . •. . •. . • . 280— . . •. . . . . . . . . • • : : : . • 0 270— : -`• ; i : 260— . • 250— • • . . . . . . . . . . . . . . . . 240 . . . . . . . . . . . . . . . . . . . . . . . . . . . 22- • • • • • • • • • • • • • • • • . . . . . . . . . . . . . . . . L • a) 20 . m 18— • . • . . . . . . . . . . . . . . . 16- • • • • •• a, . •. . . •. . . • . . • aa 10- . . . . . ... . . . .. .. . . . . ... . . . . zt=ti .cc 4— . l - 1 nil 1 II ' n _ 1' ' , I ni TL _ Hi' :lin n ' _ n: n ri• - 7t, f 1 o r Y t - t r „,-:fill r ""1 r---� r t - ' � 1/111980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 DATE Nor BING YEN & ASSOCIATES, INC 1om Geotechnical& Environmental Consultants FLUCTUATION OF GROUNDWATER ELEVATION IN MONITORING WELL W-4E PROJECT NAME:PORTUGUESE BEND LANDSLIDE Note:Rain Gage Station LAFCD 44 PROJECT NO.G94-0989 DATE:MARCH 1997 FIGURE 4.5 335 • • • • • • • • • • • • • • • • +, . •. • . 62330 • ; Z • • • . . . . . •• o H •a 325- u, . .• •. .• . J . . LL., • w 320- : a • . •. .• • • •• . . . . . . . . . • • . •. •. . . . . . . . . . 315 • . . . • . . 22- •• • • • • • • • • • • • • • • • • • • • •• • • • • • • • • • • • • • ezi m - 18- : •: •. : •: •: •: : •: •: : •: •: •• •: •• . . . . . . • • •• •• •• •• •• •• •• • • Tri �, 16- a L 12 •- a Li_a- 10- •• • " •• * * * * *•* . . . . ZZ a a _ : :• : - : : :: : : : • 1 • ~ z 4- • • }ii0 ( 1.F- 'L I H I� 1 IT : n_ . IH , ill TL • I r� n I Tl, 1• 1/1/1980 1981 1982 1983 1984 1985 1988 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 DATE NEE BING YEN & ASSOCIATES, INC NEN Geotechnical& Environmental Consultants FLUCTUATION OF GROUNDWATER ELEVATION IN MONITORING WELL B88-4 PROJECT NAME:PORTUGUESE BEND LANDSLIDE Note:Rain Gage Station LAFCD 44 PROJECT NO.G94-0989 DATE:MARCH 1997 FIGURE 4.6 290 . 4-, 285 LT 280 . . . . . . . : . : . . . . • 0 275 a 270 ' 265 • _ W • . . . . . . . w 260 • • 3 255 • • • • • .• . . . . . . . . . . .• . . . . 250 . . 22 • • • • • • • • • • • • • • • • . . . . . . . . . . . .• . . . . . . . . . . . . . . . . . . . . CD 73 c_ c20 • . . . . . • . • • m d 18 : . : : . : : : : : : : . : . 16 • . •• • • �, 1213 w 5 c c 14 . • • • • : . J J 12 • • • • • • • • • : • • • • . • • • : • aa 10 . . . . . ... . . . • ' • •• ' . . . ... . . . . Z Z QQ O cr cc_ _ - • Z 4 . CD a . _ . • • QIP _t 7 �y' n fh i IL TL II� Fill_ �r{� ., ' r 1/1/1990 1981 1982 1983 1984 �' 1986 1987 1988 1989 1990 1991 1992 1993 i99 1995 1996 DATE Nor BING YEN & ASSOCIATES, INC mom Geotechnical& Environmental Consultants FLUCTUATION OF GROUNDWATER ELEVATION IN MONITORING WELL B88-5 PROJECT NAME:PORTUGUESE BEND LANDSLIDE Note:Rain Gage Station LAFCD 44 PROJECT NO.G94-0989 DATE:MARCH 1997 FIGURE 4.7 1/—Ground Surface x' , r 10-Inch Diameter Borehole r Piezometer C > Piezometer with Sand Sock Y1 4-Inch Diameter Schedule 40 PVC Casing Sealed at Bottom • y jf < Cement Grout With b%Bentonite Piezometer B •,;•:,/(-1 • ' Failure Surface 4 Itir % 5 • 4 .:,.. .,,, Total Depth Approximately 75 Feet 4 Piezometer A r Piezometer(Depth) APPROXIMATE Well I.D. DEPTH TO A B C SUDING SURFACE BYA-1 48 33 25 40 BYA-2 98 75 50 90 BYA-3 47 32 25 40 NNE BING YEN & ASSOCIATES, INC •❑N Geotechnical& Environmental Consultants INSTALLATION SCHEMATIC FOR PROJECT NAME:PORTUGUESE BEND LANDSLIDE PNEUMATIC PIEZOMETER PROJECT NO.G94-0989 I DATE:MARCH 1997 FIGURE 4.8 l 99 1 1 I 1 1 I III 1 1 1 1 1 1 1 1 98 — 95 - 90 — 1::....-:0 80 v , ✓• 70 — C 7 60 E. O 50 — 40 — a• 30 — E 20 U 10 1----- . 5 = ,11 . 2 — 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 10 100 1000 Plasticity Index •Data base from S.M.Watry and P.L.Ehllg(1992) 0Er BING YEN & ASSOCIATES, INC 10ilGeotechnical& Environmental Consultants STATISTICAL ANALYSIS OF PALOS VERDES PENINSULA BENTONITES PROJECT NAME:PORTUGUESE BEND LANDSLIDE LOG-NORMAL DISTRIBUTION OF PI PROJECT NO.G94-0989 DATE:MARCH 1997 FIGURE 4.9 1 99 1 , 1 I I I III I 1 II 1 I I 1 98 - - 95 - - 90 - - °.• 80 ' ` - ,, - 0 c70 - 60 - - it 50 - - > 40 - ,'� - 330 - - - E _ V -20 ,' = - - 10 , - , 5 = ' - , ' 2 i 1 i 1 1 1 1 1 i i II 1 11111 1 10 Lal_ Friction Angle (Degrees) Data base from S.M.Watry and P.L.Ehlig(1992);Except one set by Eben Vey(1961)where friction angle Is equl to 3.5 degrees.Slx sets of unpublished BYA(1986)data from samples taken at flying triangle area were also added. nor BING YEN & ASSOCIATES, INC mom Geotechnical& Environmental Consultants STATISTICAL ANALYSIS OF PALOS VERDES PENINSULA BENTONITES LOG-NORMAL DISTRIBUTION PROJECT NAME:PORTUGUESE BEND LANDSLIDE OF RESIDUAL FRICTION ANGLE PROJECT NO.G94-0989 DATE:MARCH 1997 FIGURE 4.10 • 3000 . 1 . , . , 1 , , , , 1 . , , 1 , , , 1 . , • 1 , , . , 1 . . , , 1 , . , Cu Ka rod. - 35KV/l5mA 2500 — 1 _ 2000 —tii _ • T I — (/) . C 1500 — — a) .-N C \ - - 1000 - _ \I\ • 500 — _ 5 10 15 20 25 30 35 40 45 . • Diffraction Angle 20 GilaXRD B-2-96 .29-1499 Nees(A1.Mp251aO1(OH)2aH2O a A 41 611 d A M I Idd 21.5 Iial OM Sodium Magnesium Aluminum Silicate M,mlm.mllunite-21A 111.6 IM 1102 Hydroxide Ilydrate 4 45 55 11 02 Rad. C'u6 o 1. 1.5415 Neer Ni bq 3.15 5 CM an 1.1. IhHrnmt trrtcr VI , 2.56 335 I..20 Ref. Hnndlcy,G..Penn Stair University.University Park Peans51. 1.69 M 151.150 vania.t'SA,!CPIS c;mnain Aid Report.11977) 1.493 25 IM.An 1.325 a 25.11 Sys. Monoclinic S.C. 5.2 b 9.0 c 21.0 A C I5 90.0 7 Z iiBrown.Si..X-Ru.Idenrifoatu.n and l'rysml Sm,nurn.(('la.. (1wdl ISpsCin:.•o I.....V.s.MuiRMined•. .VSA. ed over wale..Lim s days.%', n•,ln...lt.pi..baldy ctislobable.11MI'5 rclalise s hun,idm.'.•n.,tiro Croup.diatahcdral subgroups C I). ss Ce1Ca:..t 1619,6 9Mil.c=5.2110.a111-2.3333. .\ 1 Y 1 1 0'6=1157,a.11.u.III,..1.12,2.191 `� • -._.--- "ti.^ ,x,0 4-.+..� I 1 I I 1 I I. 10 20 30 4;; SO m 75 ni❑.0 BING YEN & ASSOCIATES, INC mom Geotechnical& Environmental Consultants X-RAY DEFRACTORGRAM RESULTS FOR SAMPLE TAKEN AT PB-2 PROJECT NAME:PORTUGUESE BEND LANDSLIDE WITH PATTERN MATCHING PDF#29-1499 PROJECT NO.G94-0989 DATE:MARCH 1997 FIGURE 4.11 ms,,. Rrilik e ._-lia .41111.'111,2 °/,p1 Alg ven wei-- . , if -, : ,, , , .... .... .. . ,, 4 .41,, . _ .'.`K"=. k ,_ f ,tib. ., .WI) J , s:} a r w _ dir - ,‘, - . ., . ri„...,..: ,_" ‘ • ., ' C., ....le D - *41 -Mit . ar,.. i- '-' .... li " , 4 ....• _. ob.- ... . , _ * ,..., , , ..., b . . - -, - . 1 ir ,,- -..t,.. -: 4.110g", ''':.Atiklik, ' 7 • - `.'s‘ a jri 1fijt. ` Sr . atit ► ''i.4 0 20KV 0 . 50KX 2 0 . ON 0009 PhotoMet A. Oblique view showing clay platelets beneath slide surface , --.., AF A ORTLIGUES gi. , _ L Y-76 4E -n '+s .* A; . c M r Visior .4%dtc. 20KV 02 . 5KX 4 . 001' 0003 PhotoMet B. Slide Surface ?IS' BING YEN & ASSOCIATES, INC NEN Geotechnical& Environmental Consultants SEM OF SLICKENSIDED PB CLAY TAKEN AT PB-1 PROJECT NAME: PORTUGUESE BEND LANDSLIDE PROJECT NO.G94-0989 DATE:MARCH 1997 FIGURE 4.12 • 22-Jul -1996 13 : 48 : 24 PORTUGUESE BEND RPV, SLICKEN SIDE ! Preset = 100 secs Vert = 2711 counts Disp = 1 Elapsed = 58 secs Si 0 20 KV • Al lMg � 1 S K Ca Fe \- -- -- -- - Y - 1 12 T3 T 4 T5 16 17 � 8 9 4- 0 . 000 Range = 10 . 230 keV 10 . 110 -4 Integral 0 = 113637 MI00 BING YEN & ASSOCIATES, INC mom Geotechnical& Environmental.Consultants ENERGY DISPERSION SPECTROSCOPY OF A PB CLAY SLIDING MATERIAL PROJECT NAME:PORTUGUESE BEND LANDSLIDE TAKEN AT PB-1 PROJECT NO.G94-0989 DATE:MARCH 1997 FIGURE 4.13 10" 411111111141111110 Upper Porous Stone 5 .5 (Rotates) y° Soil A Sample 09" NN I , ',. Soil Sample r : ,;::: ' - - - ""' Sample Used I 0.08"X0.11"X 0.50" I 411910._______i,,,,, Lower Porous Stone(Fixed) Outer Confining Ring Inner Confining Ring Isometric View -------_ L vs., Sample Slice ox, Used for Photography and EDS Sample Configuration Pneumatic Cylinder _ a r9. IIIIIIIIIIIP / Reaction Frame Load Cell 'ti• Rotary Bearing Actuator Upper Porous Ston- ,Jli Inner Confining Ring (Rotates) Soil Sample i Outer Confining Ring il IA . ......_ arr i . IiiiiM Base -�- Lower Porous Stone(Fixed) pr Reaction Frame NOT TO SCALE Cross Section on❑ BING YEN & ASSOCIATES, INC EtaGeotechnical& Environmental Consultants ❑■ SCHEMATIC OF THE PYA RING SHEAR PROJECT NAME: PORTUGUESE BEND LANDSLIDE TESTING DEVICE PROJECT NO.G94-0989 DATE: MARCH 1997 FIGURE 4.14 - , 4 y-"1'` 4_ _ ., 411. - . <` ' a 1' �t...-•,rte.. . •c` ;, . >>.•.N4 W �14" 4, 0 41 . t s E illitille X1_1 f:: I.:l A5ci::.:; 2001-' A0018 Phc toMet A. Oblique view showing clay platelets beneath sheared surface PB2N \ )111111L . _,.....ismilio ----:".e4,,.....7.11.)., . gm: - , �. '"� 20KV 02 . 5KX 4 . 00F 0014 PhotoMet B. Shear Surface 0•P BING YEN & ASSOCIATES, INC KAIGeotechnical& Environmental Consultants SEM OF PB CLAY AFTER RING SHEAR TEST PROJECT NAME:PORTUGUESE BEND LANDSLIDE PROJECT NO.G94-0989 DATE:MARCH 1997 FIGURE 4.15 i? i • EXPLANATION 17 2C w PROJECT NAME: H96-0989 z c~rI Y 4 . . . . . . .. . . . . . . . . . . . . . . Sample IPB-2 Highly Plastic Tan Bentonite 0 . s . . . . . . . . . .:. . . . . . . . . . . Start of Test: 07-02-1996 Cl)i o vEnd of Test: 07-10-1996 wQ .4 Test Duration: 7.62 Days U' f= .2 Total Disp.: 83. Inches 0 • Data File: PB-i.DAT Map. Comment 0.00 Pre-Consolidate 110 kef 0.00 Initis] Hormel Stress-2.5 kef 1200 • 0.00 Initial Rate - 15 in./day 9.749 Decrease Disp. Rete to 0.01 In/Day LL � 9. 51 Return to 16 In/Day� 10.. Rate to In/Day Reduce Ra Li_ to S In/Dey 10.57 Increeee Normal Stress to 5 KSF U3 in 11.93 Increeee Hormel Stress to 10 KSF LLJ 8 O O I- 14.41 Reduce Rete to 10 In/Day. 1--- 11.71 Increeee Rete To 15 In\Day. . 14.88 600 p Ce(H Added to Sample Bath. Q - - - - 25.35 Increeee Rete to 275 In./Oey Q LIJ - 84.10 Replace Beth Neter. 6 pH w 1F Acid. 2 U7 88.22 pH-7.3 . . . Lower to 5 w HF Acid 400 - 83.90 test stopped 0 1 -r r r 1 r r r r 0 10 20 30 40 50 60 70 80 90 DISPLACEMENT (INCHES) N�0 BING YEN & ASSOCIATES, INC mo. Geotechnical& Environmental Consultants TYPICAL RING SHEAR TESTING RESULTS FOR PB CLAY SAMPLE (PB-2) PROJECT NAME:PORTUGUESE BEND LANDSLIDE PROJECT NO.G94-0989 DATE:MARCH 1997 FIGURE 4.16 4000 w N U 3000- -.4' :`. 20 ksf Preconsolidatio LT, N A Cl) W yr a CD z ♦ 4 S' J, w 2000- '° 10 ksf Preconsolidation Zk '' N w .: n a 0 !il Q • g 1000- 0 i I I I I 1 0 1000 2000 3000 4000 5000 6000 7000 NORMAL PRESSURE psf LEGEND 0 SAMPLE PRECONSOLIDATED TO 10 ksf O SAMPLE PRECONSOLIDATED TO 20 ksf A SAMPLE MIXED WITH 6%LIME AND PRE- NKr BING YEN & ASSOCIATES, INC CONSOLIDATED TO 10 ksf KEIU REMOLDED AND PRECONSOLIDATED PB CLAY °k LIME AND PRE ��■ Geotechnical& Environmental Consultants ♦ SAMPLE MIXED WITH 6 CONSOLIDATED TO 20 ksf PEAK STRENGTH SHEARED IN RST AT PROJECT NAME:PORTUGUESE BEND LANDSLIDE 15 INCHES PER DAY DISPLACEMENT RATE • PROJECT NO.G94-0989 DATE:MARCH 1997 FIGURE 4.17 2000- I I I I I I I I I I I I I II I I I I F Q _ I _ J _ L _ I _ J _ L _ I _ J _ L _ I _ J _ L _ I _ J _ L _ I _ J _ L I _ N I I I I I I I I I I I I I I f t= 125+0.67 6 -0.304&07 4(5 1000 -- - I - -I - I- - I - -1 - I- - I - -1 - L -+- 7 - 17 - I - 71 - - NI I I I I + I I I I I I I 1 I I r I I I I I I I I I 0 f l l l l l l l l l l 0 1000 2000 3000 4000 5000 6000 7000 8000 9000 10000 Normal Stress (psf) o.r BING YEN & ASSOCIATES, INC LOA Geotechnical& Environmental Consultants RESIDUAL SHEARING STRENGTH MEASURED IN RING SHEAR TEST AT 15-INCHES PER DAY PROJECT NAME:PORTUGUESE BEND LANDSLIDE DISPLACEMENT RATE PROJECT NO.G94-0989 DATE:MARCH1997 FIGURE 4.18 1.600 — + + 1.500 — 1.400 — + L 0) 1.300 — C N cn — f(d)= 0.93 + 0,024 (d)0,38 N 1.200 — 0) 0 1.100 — + + + +. 1.000 — I _ I 0.900 1 1 1 1 1 1 1 I I 11 1 1 1 1 1 1 t l 1 1 1 11111 I I1 1 11111 1 1 I I I 157d 0.1 1.0 10.0 100.0 1000.0 Displacement Rate (In./Day), d Door BING YEN & ASSOCIATES, INC m0. Geotechnical& Environmental Consultants EFFECT OF DISPLACEMENT RATE ON SHEARING STRENGTH OF PB CLAY PROJECT NAME:PORTUGUESE BEND LANDSLIDE PROJECT NO.G94-0989 DATE:MARCH 1997 FIGURE 4.19 30 1 1 1 1 II 1IIIII � � � I I ] � iii ] Calcium Montmorillonite(ICU,OCR=1) • Calcium Montmorillonite(ICU,OCR>1) A Calcium Montmorillonite(ICD,OCR=1) A * Sodium Montmorillonite(ICU,OCR=1) _ BYA PB Clay Residual Strength Envelope 20 — Al N _ _ a n - N-Calcium Montmorillonite - V' • A N N 10 — — Az-PB Clay • \- - Sodium Montmorillonite- 1 0 I l I � � I X11111111It 0 10 20 30 40 50 60 70 80 90 100 P=(SIG1+SIG3)/2(psi) ❑o❑ BING YEN & ASSOCIATES, INC CILIA Geotechnical& Environmental Consultants COMPARISON OF PB CLAY RESIDUAL STRENGTH ENVELOPE WITH THAT OF SODIUM AND CALCIUM PROJECT NAME:PORTUGUESE BEND LANDSLIDE MONTMORILLONITES BY MESRI, 1969 PROJECT NO.G94-0989 DATE:MARCH 1997 FIGURE 4.20 3 • J cn 2 • - rn LLI CC CC CI) Li Y • •0 • cn o W ~ • cc FQ cn a .5- . . . • 0 600 - 7 cn CL Cr) cn Q 400 - I CI) CC w 0 200 - 0 0 .5 1 DISPLACEMENT (INCHES) nor BING YEN & ASSOCIATES, INC •DA LII• Geotechnical& Environmental Consultants EFFECT OF REDUCTION OF OVERBURDEN PRESSURE ON SHEARING STRENGTH PROJECT NAME:PORTUGUESE BEND LANDSLIDE OF PB CLAY PROJECT NO.G94-0989 DATE:MARCH 1997 FIGURE 4.21 Total elapsed time 110 days • 0.30 __I _ � —_18' -16' 025 — — x+-x