Loading...
6.0 Evaluation of Potential Supplemental Stabiliaztion Measures 6.0 EVALUATION OF POTENTIAL SUPPLEMENTAL STABILIZATION MEASURES • As presented in the previous sections, the proposed grading, dewatering, and surface drainage improvements have the potential to at least temporarily arrest movement in the eastern portion of the landslide. However, even in the best case, the proposed POC will only be capable of improving the stability marginally and the landslide may still creep intermittently and be susceptible to reactivation. Conditions which contribute to reactivation of the landslide include shoreline erosion, successive years of above average rainfall, lapses in the de-watering or surface drainage maintenance programs, and continued movement of the Seaward and/or West-Central subslides. A preliminary evaluation of the most promising supplemental stabilization measures has been conducted as part of this investigation. These measures include: (1) PB clay strength enhancement; (2) the construction of a revetment along the shore line; and (3) a more extensive dewatering program. The effects of each of these measures are discussed in this section. 6.1 PB Clay Strength Enhancement • As discussed in Section 4.5.3, the PB clay samples which were analyzed as part of this investigation are a type of bentonite composed predominantly of sodium montmorillonite. It is well established that sodium ions within this type of clay can be replaced by calcium through a cation exchange process. This process has been used in the age-old practice of improving the engineering behavior of clay soils through the addition of lime. Extensive testing by Mesri (1969) and others has established that, under identical normal stresses, the shear strength of calcium montmorillonite is typically 100% to 300% higher than that of sodium montmorillonite (Figure 4.20). Exchanging calcium for sodium within the bentonite has the potential for increasing the shear strength along the rupture surface and substantially improving the stability of the landslide. In order to evaluate the feasibility of this approach under controlled conditions, two separate ring shear tests were added to our study. In the first test, 6% calcium hydroxide [Ca(OH)2] by weight was mixed into the bentonite sample prior to testing. The results of this test are summarized along with S 77 those for a control sample [without Ca(OH)2] in the following table and graphically shown • in Figure 6-1. Table 6.1.1. Effect of Ca(OH)2 on Shear Strength of PB Clay Residual Shear Strength (psf) Normal Stress Relative Shear (lsf) Control Sample Sample with 6% Strength of Treated Ca(OIO2 Sample 250 169 188 111% 500 215 245 114% 1,000 315 375 119% 2,500 487 668 137% 2,500 475 880 185% 5,000 700 1,150 164% 9,000 990 1,950 195% • Note that the residual shear strength of the PB clay sample treated with calcium hydroxide was substantially higher at each of the normal stress levels tested. At the normal stress level that most closely approximates the average overburden above the landslide rupture surface (9,000 psf), the residual shear strength of the Ca(OH)2-treated sample was nearly twice that of the control sample. The addition of lime to soil causes microfabric changes by forming compounds such as calcium silicate and calcium aliiminate hydrate. The microfabric changes and the presence of these compounds account for the increased strength and decreased dependence on normal loads. Since the slide surface of PBL is on average about 100 feet below the existing ground surface and the overlying material is relatively stiff, the conventional "lime column" approach is probably not economically feasible. The lime column process has been used worldwide in the past several decades primarily for soft clays at shallow depths. For stiff clays at shallow depths (10 to 15 feet), pressure injected lime solutions have been used ID (e.g., Lundy and Greenfield, 1968; Thompson&Robinet, 1976) for treatment of swelling 78 soils. This method involves combining lime slurry with surfactant in a mixing tank and 11 injecting it under high pressure as a relatively rapid method to treat foundation soils under light structures. The treated soil after lime injection also relies on the cation exchange phenomenon to decrease soil expansion potential. BYA considered the use of CaC12, which has a much higher solubility than Ca(OH)2, and investigated the effect of cation exchange by diffusion in lieu of physically mixing it with the sliding material. In the second strength enhancement test, an untreated PB clay sample was sheared for a distance of 5.4 inches under a normal stress of 5,000 psf in the ring shear testing system. At that time, the test was stopped and the water in the sample bath was replaced with distilled water to which calcium chloride (CaC12) had been added to yield a concentration of about 1,000 ppm by weight. This is intended to facilitate diffusion of the calcium chloride solution into the sodium montmorillonite and the subsequent exchange of sodium by calcium within the clay. After a period of 24 hours, shearing was re-initiated and continued up to a total displacement of 8.5 inches. The results of this test . are presented in Figure 6.2. As can be seen in this figure, a well defined secondary peak occurred when shearing was reinitiated following the addition of the CaC12 solution. This peak strength (1,360 psf) was approximately 70% higher (810 psf) than that which was measured prior to the addition of CaC12. With further displacement, the shear strength of the sample stabilized at approximately 745 psf. This is approximately 15% higher than the residual strength of the untreated control sample. It is noted that 24 hours is probably not long enough to fully develop the potential cation exchange. The rupture surface of the lab sample was examined about 48 hours after the introduction of the CaC12 solution using Energy Dispersive X-Ray Spectroscopy (EDS). The results of this test are presented in Figure 6.3. Comparison of Figure 6.3 with Figure 4.13 (PB clay slickensided surface without exposure to CaC12), shows that there is a stronger presence of calcium within the treated sample than the untreated sample. Even though EDS is only a qualitative analysis, it is encouraging to note that the calcium cations within • 79 the solution had reacted with sodium and penetrated to the rupture surface over the • relatively short duration of the test. From an engineering perspective, there are two primary issues related to the practical application of lime treatment of the PB clay: (1) the rate of cation exchange once the solution is present, i.e., how fast can the reaction be expected to occur relative to the strength increase with time; and (2) how much strength can be expected for a given time. In addition to BYA's preliminary testing results shown in Figure 6.2, recent research by Rao and Rajasekarau (1996) involving the injection of lime slurry into a soft marine clay in India has shown promise regarding both issues. The plasticity of the marine clay used in Rao's tests was almost identical with PB clay (LL=85 & PI=53). For example, the lime content (measured as a percent of CaO) required to fully satisfy the requirements of cation exchange for the Indian marine clay is about 1.6% to 1.7%. Within days, the lime concentration reached the desired value when measured radially from the lime column and the strength of the clay had increased 5 to 8 times (measured by vane shear test). • To assess the potential impact of strength enhancement to the factor of safety in the Eastern PBL, cross section D-D' was re-analyzed assuming the strength of a 400-foot wide strip of sliding surface is enhanced by cation exchange. The location of this strip is conceptually shown in Figure 2.9 as a "hatched" section along the basal rupture surface. This strip is chosen because the basal rupture surface is steeply dipped and relatively shallow. The results of a two-dimensional stability analysis are summarized below. Table 6.1.2. Evaluation of Strength Enhancement of a 400-Foot Wide Strip on the Factor of Safety for Cross Section D-D' Strength Enhancement Factor of Safety for D-D' 0% 1.12* 25% 1.17 50% 1.21 Note: See Table 4.5.5.2. The current factor of safety is more likely to be 1.0 when 3-dimensional effect and rate of movement is considered. 80 Given that the reported strength increase achievable by cation exchange can be potentially • much-higher, (see Figure 4.20 and Rao et. al, 1996, which shows 200% to 800%) than the 25% or 50% assumed for calculation of factor of safety increase in Table 6.1.1, the potential of strength enhancement as a supplemental or alternative method to stabilize PBL warrants further investigation. In summary, the results of BYA's ring shear tests suggest that the shear strength of PB clay can be increased alongthe rupture surface through a cation exchange process resulting in greater stability of the landslide. BYA's tests show promise, but additional laboratory tests are recommended to confirm and supplement these results. It is further recommended that the additional tests should be run using a lower rate of ring shear displacement such as 1 inch per day. Theoretically, the results should be comparable because the test data were normalized, but confirmation is needed. More importantly, a pilot field testing program should be undertaken. The pilot program would identify the most practical and effective method of introducing the highly soluble calcium chloride within the rupture surface and to monitor its effect in arresting the movement. A brief discussion on the pilot field program is described in Section 6.4. There are a number of ways to introduce CaC12 to the PB clay once information such as the rate and magnitude of strength increase is better known after the pilot field program. For example, it may be possible to use existing wells (or abandoned, sheared-off casings) be used for CaC12 introduction into the slide during dry seasons. It may also be feasible to introduce chemicals at the graben area in the Seaward subslide area whose seaward movements represent a continued withdrawal of lateral support to the East-Central subslide. 6.2 Revetment to Buttress Seaward Subslide One mitigative option which has received some discussion in the past involves the • construction of a revetment along the shoreline at the toe of the landslide. Such a 81 revetment could improve the stability of the landslide by providing resistance to movement • of the Seaward subslide as well as reducing shoreline erosion. A preliminary stability analysis has been completed as part of this investigation to evaluate the potential benefits associated with the construction of such a revetment. A preliminary configuration of the revetment which was considered in this analysis is shown in Figure 6.4. Protected by rip- rap, approximately 200,000 cubic yards of material would be needed to construct this revetment from Inspiration Point on the west to the landslide margin on the east. For the purposes of this analysis, the revetment material is assumed to have a saturated unit weight of 140 pcf and an effective friction angle and cohesion of 38° and 0, respectively. The results of a preliminary stability analysis are summarized below: Table 6.2. Potential Impact of Shoreline Revetment on Eastern Half of PBL Estimated Combined 'Condition Cross Factor of Weight Relative factor of Section Safety (Tons per Foot Mass Safety • of Width) With Grading Currently Proposed @ the C-C' 0.93 13,800 49.3% 1.07 Head of the Landslide D-D' 1.21 14,200 50.7% C-C' 0.99 13,950 49.3% With Proposed Grading + Revetment 1.14 D-D' 1.3 14,350 50.7% For Seaward subslide with head located 900 feet from shore, see Table 5.2.3. As shown, it is estimated that the presence of this revetment would increase the factor of safety of the landslide as a whole by approximately 7%. The results of a similar analysis for the most critical Seaward subslide are presented below: I 82 Table 6.3. Effect of Shoreline Revetment on Seaward Subslide Estimated Combined Condition Cross Factor of Weight Relative factor of Section Safety (Tons per Foot Mass Safety of Width) With Grading C-C' 0.95 4,200 45.2% Currently Proposed @ 0.97 the Head of the D-D' 0.98 5,100 54.8% Landslide With Proposed C-C' 1.1 4,350 45.3% Grading + 1.12 Revetment D-D' 1.13 5,250 54.7% ' See Figure 5.2. As shown, it is estimated that the presence of this revetment would increase the factor of safety of the Seaward subslide by approximately 15%. 6.3 Dewatering • Additional dewatering is included in the current control plan in the form of the subdrain system and the four to six dewatering wells which have been proposed. Although these improvements are believed to be a necessary and important element of this plan, it is unlikely that they will result in extensive reductions in groundwater levels within the remaining portion of the landslide. Previous dewatering efforts have established that removal of groundwater from the landslide is difficult due to the low permeability of the landslide debris and underlying bedrock. High groundwater levels, particularly in the eastern portion of the failure, contribute substantially to the ongoing movement. Therefore, a more intensive dewatering effort may be needed in conjunction with the proposed surface drainage improvements. Additional stability analyses have been conducted as part of this investigation in order to evaluate the potential benefits of a more rigorous dewatering program. Uniform reductions in the groundwater level within the - landslide ranging from ten to fifty feet were evaluated as part of this analysis. The results are summarized in Table 6.4. IP 83 Table 6.4. Potential Impact of Dewatering on Landslide • Estimated Combined Cross Factor of Weight Relative Condition Section Safety (Tons per Foot Mass factor of of Width) Safety With Proposed C-C' 0.93 13,800 49.3% Grading1.07 D-D' 1.21 14,200 50.7% With Proposed C-C' 0.98 13,650 49.2% Grading& Groundwater 1.13 Lowered 10 D-D' 1.28 14,100 50.8% Feet With Proposed C-C' 1.02 13,550 49.2% Grading& Groundwater 1.19 Lowered 20 D-D' 1.35 14,000 50.8% Feet With Proposed C-C' 1.05 13,400 49.1% Grading& Groundwater 1.23 • Lowered 30II D-D' 1.41 13,900 50.9% Feet With Proposed C-C' 1.08 13,250 49.0% Grading& Groundwater 1.28 Lowered 40 D-D' 1.48 13,800 51.0% Feet With Proposed C-C' 1.1 13,150 49.0% Grading& Groundwater 1.33 Lowered 50 D-D' 1.55 13,700 51.0% Feet As shown, the benefits of lowering the groundwater elevation are theoretically significant- particularly in the eastern portion of the landslide. However, to lower the water table an average of more than 20 feet may not be feasible because of the high cost associated with lowering groundwater within the low permeability material. Further analysis of this alternative, including an estimation of the number and configuration of wells that would be required to achieve specific local drawdown, or to intercept the inflow, may be needed _• for an economic analysis and comparison with other options. At this time, we believe that 84 one cannot practically expect to lower the water table an additional 20 feet below the • October 1996 level across the PBL as a whole. 6.4 Pilot Field Testing Program Comparison of Table 6.1.1 with Tables 5.2.1, 5.2.5 and 6.3, show that the effect of a conservatively assumed strength enhancement example hypothetically implemented in a 400-foot wide strip along the basal rupture surface can increase the factor of safety to 1.21. A factor of safety of 1.21 is equivalent to the net effect of regrading plus some permanent reduction of groundwater surface. The advantage of using the proven phenomenon of cation exchange to enhance the strength is that the process and its effect will increase with time. Theoretically, infiltration of surface water through cracks and fissures which normally carry water downward and downgradient to the basal sliding plane can further transport and spread the calcium ions to the sodium rich PB clay beyond the source where the calcium chloride is introduced. However, even though lime slurry injection has been used in the past for treating shallow foundations embedded in stiff clays, no known applications have been used for stabilizing landslides with a deep-seated sliding surface. Hence, the feasibility of field implementation and cost effectiveness need to be established and a Pilot Field Testing Program is recommended. The objectives of the Pilot Field Testing Program are as follows: 1. Determine sliding base bentonite strength under in-situ conditions: This will establish the baseline strength data at the test site. Previous strength measurements on PB clay were made on laboratory test specimens several square inches in size. (The ring shear tests conducted in this study has a shearing surface about 22 square inches.) The in-situ test will involve more than 100 square feet of relatively undisturbed sliding material. • 85 2. Evaluate the rate of diffusion of the calcium ions and the rate of strength . increase as a result of the cation exchange at the test site; i.e., to test the effectiveness to increase the factor of safety. 3. Evaluate feasible method(s) to introduce the calcium ions into the sliding plane and the probable range of cost. A test site (Figure 6.5) has been selected in the northwestern part of the Eastern PBL which meets a number of criteria. The sliding surface at the test site is relatively well defined based on previous studies as consisting of an approximately 3 inches thick layer of PB clay dipping approximately 15 degrees (based on two nearby borings). The sliding base is shallow (located about 10 feet below existing ground surface) and the ground surface is relatively flat. Initial calculations indicate that the site will have a factor of safety close to 1.0, a condition ideal to achieve controlled sliding movements prior to or after the introduction of calcium ions. i Sampling and testing to further document the characteristics of the PB clay at the test site will be necessary prior to beginning the pilot testing. While it is not the intention in this report to detail the testing protocol, the overall approach entails isolating a relatively large block for in-situ testing and monitoring. An automatically controlled loading and resisting force system will be installed in the direction of the sliding planes true dip (15° SW). In- situ sensors (such as ion-specific probes) or their equivalent will be installed in monitoring holes for diffusivity measurements. The monitoring holes may also be used to insert neutron probes to monitor bulk moisture migration from the calcium ion source. Provisions to restrain the testing block from sloughing, displacement measurements and a data acquisition system need to be utilized in order to evaluate injection, other options and limitations. 86 C 3000 a w U Z w N LU z 2000- PB Clay mixed with CA(OH)2- c PB Clay UJ 1000- ______ � 0 1500 3000 4500 6000 7500 9000 NORMAL PRESSURE,psf LEGEND • PB Clay A PB Clay mixed with 6%CA(OH)2 by Weight NEE BING YEN & ASSOCIATES, INC LOA Nom Geotechnical& Environmental Consultants RESIDUAL SHEAR STRENGTH OF LIME TREATED AND UNTREATED PB CLAY TESTED IN RST PROJECT NAME:PORTUGUESE BEND LANDSLIDE PROJECT NO,G94-0989 DATE:MARCH 1997 FIGURE 6.1 8 . •� : : EXPLANATION Jul 6- . • ME UJE5cc cc 4- PROJECT NAME: H96-0989 Y C, �— 2- : Sample IPB-2 • : : : :• • Highly Plastic Tan Bentonite 6 • :• • : Sart of Test: 08-09-1996 o : : : : : : : End of Test: 08-14-1996 cc a •4 Test Duration: 3.75 Days �' .2- • `� .. -� • ....:,�`"`~�-�-_. _. Total Disp.: 8.5 Inches • • o- Data File:PB-13.DAT • Diep. Comment • 0.00 Sample PS-2 0.00 Pre-Consolidate 110 ket 0.00 Rate• 15 In/Der 0.00 Normal Stress•5 KSF • LI— 2000 fl7 . 3.057 Decrease Rete to I In/Day d G 5.361 Add CaCl to Sseple Seth • 6.599 End Test cn . . u7 . . • W 1— . . un •. = 1000 . • 0 i 2 3 4 5 6 7 8 9 DISPLACEMENT (INCHES) NNE BING YEN & ASSOCIATES, INC NOR Geotechnical& Environmental Consultants PEAK AND RESIDUAL SHEARING STRENGTH PROJECT NAME:PORTUGUESE BEND LANDSLIDE OF PB CLAY SUBMERGED IN CACI 2 SOLUTION PROJECT NO.G94-0989 DATE:MARCH 1997 FIGURE 6.2 I/ 14-Aug-1996 14 : 47 : 16 PB2M Preset = 100 secs Vert = 10061 counts Disp = 1 Elapsed = 100 secs Si 0 • 1 20 KV , Al S lMg • I CI Ca I r K Ca Ti Fe ri 11 12 13 14 15 16 17 8 9 I 4- 0 . 000 Range = 10 . 230 keV 10 . 110 —4 Integral, 0 = 506228 0■r BING YEN & ASSOCIATES, INC 1O. Geotechnical& Environmental Consultants ENERGY DISPERSIVE SPECTROSCOPY OF PB CLAY SUBMERGED IN CaCl2 SOLUTION PROJECT NAME:PORTUGUESE BEND LANDSLIDE FOR 2 DAYS DURING RST PROJECT NO.G94-0989 DATE:MARCH 1997. FIGURE 6.3 80 — 40 - E Drainage Blanket z Seaward Subside ; ,� 2.5 Ohu Q Riprap w "' 0 — Hydraugers� • Ocean . ., •� Approximate Basal Rupture Surface 40i I I I I I I I I 0 40 80 120 160 200 240 280 320 DISTANCE(FEED norBING YEN & ASSOCIATES, INC hidid, Geotechnical& Environmental Consultants CONCEPTUAL REVETMENT DESIGN USED PROJECT NAME:PORTUGUESE BEND LANDSLIDE FOR STABILITY ANALYSIS PROJECT NO.G94-0989 DATE:MARCH 1997 FIGURE 6.4 ``` �� \ _\ / ---- ''—ii .„ ... -- .4Pg - i \N.--'''--*-11411.1....... 181".4441.1111"114.\ . . \''' ,.- -_________ "\\\ * • . '9 1 2 4 -`-------..<_,,, / '' TEST SITE 6P - \,N, 9`""-- 8 3r__ „N...... .. , ,• 171 . 1 6 j94,4 N '% -.'''--,.„.,,,,,,,''-'"----! _....__._.s_____,s,,,___..._.„\------______,:, 4% 4\ I.... . •„, va f `\ . WI .f 88 - 3\ \ ' \\ , .... ,...„... ....%' , 140,..„ !pp' . , 111P 2 • ,, \•.__, o \ ' \ 0 . • Ai:4%W 44' dObak.N. irn BING YEN & ASSOCIATES, INC EH. Geotechnical& Environmental Consultants PILOT FIELD TESTING SITE 0 '0 100 Ft. PROJECT NAME:PORTUGUESE BEND LANDSLIDE (N4,019,880; E4,178,430) Scale: 1"=50' IN PROJECT NO.:G94-0989 DATE:MARCH 1997 FIGURE 6.5