International Journal of

Environmental Research and Public Health Article

Back-Analyses of Landfill Instability Induced by High Water Level: Case Study of Shenzhen Landfill Ren Peng 1, *, Yujing Hou 2 , Liangtong Zhan 3 and Yangping Yao 1 Received: 8 September 2015; Accepted: 7 January 2016; Published: 12 January 2016 Academic Editor: Miklas Scholz 1 2 3

*

Department of Civil Engineering, Beihang University, No.37 Xue-Yuan Road, Beijing 100191, China; [email protected] Institute of Geotechnical Engineering, China Institute of Water Resources and Hydropower Research, No.20 Che-Gongzhuang West Road, Beijing 100048, China; [email protected] Institute of Geotechnical Engineering, MOE Key Laboratory of Soft Soils and Geo-environmental Engineering, Zhejiang University, No.866 Yu-Hangtang Road, Hangzhou 310058, China; [email protected] Correspondence: [email protected]; Tel.: +86-010-6878-6900

Abstract: In June 2008, the Shenzhen landfill slope failed. This case is used as an example to study the deformation characteristics and failure mode of a slope induced by high water levels. An integrated monitoring system, including water level gauges, electronic total stations, and inclinometers, was used to monitor the slope failure process. The field measurements suggest that the landfill landslide was caused by a deep slip along the weak interface of the composite liner system at the base of the landfill. The high water level is considered to be the main factor that caused this failure. To calculate the relative interface shear displacements in the geosynthetic multilayer liner system, a series of numerical direct shear tests were carried out. Based on the numerical results, the composite lining system simplified and the centrifuge modeling technique was used to quantitatively evaluate the effect of water levels on landfill instability. Keywords: landfill instability; high water level; back analyses

1. Introduction Since millions of tons of municipal solid waste (MSW) are being dumped onto “garbage mountains” every day, disposal of this waste is an increasing concern throughout the world [1]. Further, as MSW consists of large amount of organic and vegetable wastes, the leachates from these wastes cause leachate levels to rise, which affects both the stability of the slope and the drainage of polluted water. These are new engineering geology and hydrogeology problems. For example, Hendron et al. [2] and Caicedo et al. [3] independently concluded that high pore water pressure in an excessively wet waste body were mainly responsible for the instability and failure of the Dona Juana landfill in Bogota, Colombia. Koerner and Soong [4] found that for ten large solid waste landfill failures, the triggering mechanisms were excessive liquid in the waste and buildup of pore water pressure. Four of them were due to the buildup of leachate in the waste resulting in the failure surfaces occurring above the low permeable soil or geomembrane liner at the base of the landfills. In China, a large proportion of MSW is kitchen waste, which produces a significant amount of leachate. If the discharge mechanism is clogged, or the long-term drainage is impeded, this leachate accumulates thereby causing the water level in the landfill to rise. A survey shows that in China, there are many existing landfills have high water level problems, which may seriously challenge the stability of the landfills [5]. Therefore, there is an urgent need to study the deformation characteristics and failure

Int. J. Environ. Res. Public Health 2016, 13, 126; doi:10.3390/ijerph13010126

www.mdpi.com/journal/ijerph

Int. J. Environ. Res. Public Health 2016, 13, 126

2 of 17

mode landfills This study has identified the factors that are important Int.of J. Environ. Res.with Publichigh Healthwater 2016, 13,levels. 126 2 of in 17 the prediction and evaluation of landfill stability, and for the prevention of landfill instability. factors that aremechanism important inand the prediction evaluation of landfillhave stability, and for the prevention The failure the modeand of landfill instability been studied through field of landfill instability. investigations [6–9], physical model tests [10,11] and numerical simulations [12–14]. However, none of The failure mechanism and the mode of landfill instability have been studied through field these studies used an integrated system to monitor the landfill. Such a system can monitor the surface investigations [6–9], physical model tests [10,11] and numerical simulations [12–14]. However, none horizontal displacement, the deep lateral displacement, and the water level. of these studies used an integrated system to monitor the landfill. Such a system can monitor the In this paper, the Shenzhen landfill (Shenzhen Xiaping landfill) landslide case study is presented. surface horizontal displacement, the deep lateral displacement, and the water level. This landfill haspaper, undergone severe deformation and failure. integrated monitoring In this the Shenzhen landfill (Shenzhen Xiaping An landfill) landslide case studysystem, is including water level gauges, electronic total stations, and inclinometers, was used monitor the presented. This landfill has undergone severe deformation and failure. An integrated to monitoring failure process. A case water reportlevel documenting these fieldtotal monitoring been published previously system, including gauges, electronic stations, data and has inclinometers, was used to in Chinese present study a continuation of that earlier work,monitoring aiming to data back-analyse monitor[1]. theThe failure process. A iscase report documenting these field has been the published previously in Chinese [1]. The present study a continuation of that technique earlier work, aiming landfill instability induced by the high water level. Theiscentrifuge modeling was used to to back-analyse the landfill instability high water level. The centrifuge modeling quantitatively evaluate the effect of waterinduced level onby thethe landfill instability. technique was used to quantitatively evaluate the effect of water level on the landfill instability.

2. Geological Setting 2. Geological Setting

As shown in Figure 1, the Shenzhen landfill is located in the village of Caopu, Shenzhen City, As shown in Figure 1, the located in the The village of Caopu, Shenzhen City, Guangdong Province, China. It isShenzhen a typicallandfill valleyistype landfill. topography at the base of the Guangdong Province, China. It is a typical valley type landfill. The topography at the base of the landfill is in a north to south direction As shown in Figure 2, a retaining wall was built at the narrow landfill is in a north to south direction As shown in Figure 2, a retaining wall was built at the narrow valley end to hold the waste body. Behind the retaining wall in the north-west direction, the slope valley end to hold the waste body. Behind the retaining wall in the north-west direction, the slope of of the waste body was between 1:3.5 and 1:4. The maximum height of the waste body was about the waste body was between 1:3.5 and 1:4. The maximum height of the waste body was about 40 m. 40 m.AtAt this landsite, topography greatly. theside, west side, a ridge that under extended this landsite, the the topography variesvaries greatly. On theOn west there is athere ridgeisthat extended under the waste body, making the terrain very steep. The slope of the terrain along the base of the the waste body, making the terrain very steep. The slope of the terrain along the base of the waste ˝ waste body was steep 24 The . The bedrock mainly consists sandstone siltstone quartz body was as as steep asas 24°. bedrock mainly consists of of sandstone (i.e.,(i.e., siltstone andand quartz sandstone). There is noisweak intercalated layer or structural plane. The ground is therefore considered sandstone). There no weak intercalated layer or structural plane. The ground is therefore a solid foundation. considered a solid foundation.

N

Location of study area

0

50

100km

Figure 1. Location of the Shenzhen landfill.

Figure 1. Location of the Shenzhen landfill.

Int. J. Environ. Res. Public Health 2016, 13, 126 Int. J. Environ. Res. Public Health 2016, 13, 126

3 of 17 3 of 17

Figure 2. Site plan of monitoring points and pumping wells. Figure 2. Site plan of monitoring points and pumping wells.

The Shenzhen landfill was one of the earliest sanitary landfills equipped with a composite liner TheasShenzhen one ofon the earliestresults sanitary landfills equipped with in a composite liner system, shown inlandfill Figurewas 3. Based research and practical experience, the composite system, as shown in Figure 3. Based on research results and practical experience, in the composite liner, the shear strength is low at the high-density polyethylene (HDPE) geomembrane/geotextiles liner,the thecompacted shear strength is low at the high-density and clay/geotextiles interface [15].polyethylene Hence, these(HDPE) are thegeomembrane/geotextiles weak interfaces for the and the compacted clay/geotextiles interface [15]. Hence, these are the weak interfaces for the stability stability of the waste body. of the Inwaste 2000, body. the waste started to accumulate behind the retaining wall. The main components were In 2000, thewaste wasteand started to accumulate behind the retainingwaste. wall. The components were municipal solid a small amount ( DR DR >DR DR< >1010 mm/d

Monitoring

Monitoring frequency frequency

Once a month

Once a month

Once a week

Once a week

Once a day

Once a day

Twice a day

Twice a day

Considering the in in water level, there were increases in the Considering the rainfall rainfall condition conditionininJune Juneand andthe therise rise water level, there were increases in average daily displacement rates for all three gauges. In particular, the large displacement rates in the average daily displacement rates for all three gauges. In particular, the large displacement rates 6–7 June, and 13 June correspond to the two slides. As shown in Figure 6, the gauge W7 shows that the in 6–7 June, and 13 June correspond to the two slides. As shown in Figure 6, the gauge W7 shows average surface displacement rates at the centre of the waste were body 135 mm/d that the daily average dailyhorizontal surface horizontal displacement rates at the centre of body the waste were and 185 mm/d during the two slides. The average displacement rate in the centre of the waste is 135 mm/d and 185 mm/d during the two slides. The average displacement rate in the centrebody of the generally greater than that for the rest of the body. The displacement rate of the posterior region in the waste body is generally greater than that for the rest of the body. The displacement rate of the waste body is alsoingenerally that smaller. the gauges W6 and W7 slip posterior region the wastesmaller. body isThis alsoshows generally This shows thatwere the within gaugesthe W6same and W7 plane, with the centre part of the waste body moving outward. were within the same slip plane, with the centre part of the waste body moving outward.

Int. J. Environ. Res. Public Health 2016, 13, 126

3.3. Monitoring Displacement Int. J. Environ. of Res.Deep PublicLateral Health 2016, 13, 126

7 of 17

7 of 17

At the monitoring point, elevation of the bottom of the landfill and the thickness of the waste can 3.3. Monitoring of Deep Lateral Displacement be determined based on geological data. The boreholes could be drilled to the required depth, where Atofthe monitoring the bottom of the landfill thickness of the waste the bottom the boreholespoint, kept aelevation certain of distance (ranged from 3 to and 5 m)the from the liner. Inclinometer can be determined based on geological data. The boreholes could be drilled to the required depth, casings were then lowered into the borehole, and subsequently these boreholes were backfilled with where the bottom of the boreholes kept a certain distance (ranged from 3 to 5 m) from the liner. gravel (3 to 6 mm in grain size). The bottom of the inclinometer casing was not fixed, so both the Inclinometer casings were then lowered into the borehole, and subsequently these boreholes were surface horizontal monitoring and the lateral monitoring ought backfilled withdisplacement gravel (3 to 6 mm in grain size). Thedeep bottom of thedisplacement inclinometer casing was not fixed,to be carried out simultaneously. In this way, by combining the monitored horizontal displacement so both the surface horizontal displacement monitoring and the deep lateral displacement monitoringat the top ofought the inclinometer with monitored lateral displacement at different of thehorizontal inclinometer, to be carried outthe simultaneously. In this way, by combining thedepths monitored the actual displacement at of different depths of thethe waste body lateral can then be derived. displacement at the top the inclinometer with monitored displacement at different depths of the inclinometer, actual displacement different depths of themonitor waste body can then be derived. Figure 2 shows thethe locations of the deepat lateral displacement points. Using the monitoring Figure 2 shows the locations of the deep lateral displacement monitor points. Using thethree results of gauges CX1, CX3 and CX5, Figure 7 shows a set of typical monitoring results. These monitoring results of gauges CX1, CX3 and CX5, Figure 7 shows a set of typical monitoring results. gauges are in a line parallel to the x–y line. The gauge CX3 was located in the middle of the waste body. These three gauges are in a line parallel to the x–y line. The gauge CX3 was located in the middle of The gauge CX1 was located at the bulging area at the lower part of the waste body. The gauge CX5 was the waste body. The gauge CX1 was located at the bulging area at the lower part of the waste body. located near the top of the waste body. Figure 7 shows that there was very little displacement at the The gauge CX5 was located near the top of the waste body. Figure 7 shows that there was very little lowerdisplacement part of the waste body,part while there wasbody, largewhile displacement in thedisplacement middle part. recordings at the lower of the waste there was large in The the middle of gauge CX5 show that the displacements are consistent throughout the entire depth of gauge, part. The recordings of gauge CX5 show that the displacements are consistent throughout the the entire wheredepth the displacement was as as 1.5 mwas on 8 At1.5 this of thethe sliding of the gauge, where thelarge displacement asJune. large as m location, on 8 June.the At depth this location, wastedepth bodyofwas m. Abody comparison of12 the of the three gauges show the about sliding12 waste was about m.displacement A comparisonresults of the displacement results of the that three gaugeswent showthrough that the the slip middle-lower surface went through of the the slip surface part ofthe themiddle-lower CX1 inclinedpart tube, butCX1 not inclined through the tube, but not through the CX5 tube. This result suggests that the deformation occurred at the deep CX5 tube. This result suggests that the deformation occurred at the deep composite liner system. composite liner system. The translational slide is the most likely failure mode. The translational slide is the most likely failure mode.

Figure 7. Monitored deep lateral displacements at various time. Figure 7. Monitored deep lateral displacements at various time.

3.4. Emergency Management 3.4. Emergency Management the landslide, a variety of emergency measures have been out, carried out, which AfterAfter the landslide, a variety of emergency measures have been carried which subsequently subsequently proved to be effective. After the 6–7 June heavy rainfall event, 15 pumping wells were proved to be effective. After the 6–7 June heavy rainfall event, 15 pumping wells were drilled into the drilled into the waste body, as shown in Figure 2. It can be seen from Figure 5 that after the start of the waste body, as shown in Figure 2. It can be seen from Figure 5 that after the start of the pumping, the pumping, the water level in the wells fell. At the same time, a plastic film was laid on top of the water level in the wells fell. At the same time, a plastic film was laid on top of the waste body so as to waste body so as to reduce infiltration to the body. Further, the drains were unclogged so that the reduce infiltration to the body. Further, the drains were unclogged so that the surface water in concave surface water in concave area could flow out. At the point where the leachate overflowed, area could flow out. At the point where the leachate overflowed, the drainage pipe was connected to a gravel ditch so that the leachate could flow out. Finally, sandbags were used to exert a back pressure to

Int. J. Environ. Res. Public Health 2016, 13, 126 Int. J. Environ. Res. Public Health 2016, 13, 126

8 of 17 8 of 17

the drainage pipe was connected to a gravel ditch so that the leachate could flow out. Finally, sandbags were tofoot exert back pressure the rescue bulging area at the footbeofseen the from slope.Figure After 6these the bulging areaused at the ofathe slope. After to these measures, it can that rescue measures, it canof bethe seen from Figure 6 that the displacement rates the waste heavy body slowed the displacement rates waste body slowed down significantly. In theof subsequent rainfall down significantly. In the slip. subsequent heavy rainfall event, therea was no in obvious slip. further These event, there was no obvious These emergency measures played key role preventing emergency measures played a key role in preventing further landslide during the 2008 rainy season. landslide during the 2008 rainy season. 4. 4. Analysis Analysisofofthe theLandslide LandslideMechanisms Mechanisms The useofof liner systems for landfill design and construction brings someConstruction challenges. The use liner systems for landfill design and construction brings some challenges. Construction of a landfill includes the cover placement of soil cover and wasteheights layersover up these to various of a landfill includes the placement of soil and waste layers up to various liners heights over these liners can result in application of substantial down-slope shear stresses on the can result in application of substantial down-slope shear stresses on the liners leading to development liners leadingliner to development of significant liner tension.ofIn interface strengths of significant tension. In addition, interface strengths theaddition, composite liner system may of notthe be composite liner system may not be enough to resistwhich the increased gravity forces along gravity, enough to resist the increased gravity forces gravity, resulted in deep sliding thewhich weak resulted slidingofalong the weak interface at the bottom of landfill. interfacein atdeep the bottom landfill. Jones Jones and and Dixon Dixon [16] [16] stated stated that that the design of landfills must consider the stability stability both within within and elements of of aalining liningsystem. system.However, However,when whena liner a liner system is anchored side, and between elements system is anchored on on oneone side, the the displacement of the geosynthetics on that side is then limited. As such, the distribution of forces displacement of the geosynthetics on that side is then limited. As such, the distribution of forces within within each component is complex as it is due mainly duedeformability to the deformability and frictional interaction each component is complex as it is mainly to the and frictional interaction between between components. Therefore, to use an analytical method to determine the tensile stress in the components. Therefore, to use an analytical method to determine the tensile stress in the geomembrane, geomembrane, and the displacement between eachmore interface becomes difficult. Hence, and the displacement between each interface becomes difficult. Hence, more numerical simulation is numerical often used to liner simulate the composite liner and interface characteristics and often usedsimulation to simulateisthe composite interface characteristics the tensile properties of the the tensile properties of the composite liner [17–21].modelling The explicit difference modelling code, FLAC (Fast composite liner [17–21]. The explicit difference code, FLAC (Fast Lagrangian Analysis of Lagrangian of Continua 2-Dimensions) (Itasca International Inc., Minnesota, MN, commonly USA) [22], Continua inAnalysis 2-Dimensions) (ItascainInternational Inc., Minnesota, MN, USA) [22], is the most is the numerical most commonly used numerical analysis technique for simulating used analysis technique for simulating landfill lining systems.landfill lining systems. Based on the field monitoring data, it can be deduced that the Based on the field monitoring data, it can be deduced that the slip slip is is most most likely likely to to occur occur at at the the side slope. Therefore, Therefore,FLAC FLACwas was mainly used to study the characteristics the composite liner side slope. mainly used to study the characteristics of theof composite liner system system at the sideofslope of Shenzhen In the current a three-layer geosynthetic lining at the side slope Shenzhen landfill.landfill. In the current study, astudy, three-layer geosynthetic lining system system has been analyzed. Beam elements to sub-grids on both with have interfaces has been analyzed. Beam elements attachedattached to sub-grids on both side with side interfaces been have used been used to the represent the geosynthetic (Itasca 2007) [22]. A direct shear tests were to represent geosynthetic elements elements (Itasca 2007) [22]. A series of series directof shear tests were used to used to simulate the behavior of the critical interface in the geosynthetic multilayer liner system. simulate the behavior of the critical interface in the geosynthetic multilayer liner system. In In the the simulation simulation model, model, the the length length has has been been set set at at 22 m. In In the the model, model, the the left left side side of of the the liner liner system been fixed fixedsosoasastoto simulate operation condition in field the field where the fastening system has been simulate thethe operation condition in the where the fastening posts posts were anchored the the slope, as shown in Figure 8. were anchored at the at top oftop the of slope, as shown in Figure 8. Load-plate(sand) Geotextile(Beam-3)

interface4

Geomembrane(Beam-2)

interface3

Geotextile(Beam-1)

interface2

Subgrade(native soil)

interface1

fastening the post

Right

Left 0

0

0.4

0.8

1.2

1.6

2 (m)

Figure 8. 8. Composite Composite liner liner system system in in direct direct shear shear test. test. Figure

The beams were modelled using a linear elastic law. The tensile strength data for the textured The beams were modelled using a linear elastic law. The tensile strength data for the textured HDPE geomembranes (GM) and the geotextiles (GT) were acquired from the laboratory tests, as shown HDPE geomembranes (GM) and the geotextiles (GT) were acquired from the laboratory tests, as

Int. J. Environ. Res. Public Health 2016, 13, 126 Int. J. Environ. Res. Public Health 2016, 13, 126

9 of 17 9 of 17

shown in Table 2. The interface behavior was assured to be elastic-perfectly with a Mohr-Coulomb in Table 2. The interface behavior was assured to be elastic-perfectly with a Mohr-Coulomb failure failure criterion. The interface parameters resulting from the shear box characterization tests are criterion. The interface parameters resulting from the shear box characterization tests are shown in shown in Table 3. Villard et al. [18] describes the method on how to determine the values of these Table 3. Villard et al. [18] describes the method on how to determine the values of these parameters. parameters. The system used, and the results obtained for the GT/GM interface, are shown in The system used, and the results obtained for the GT/GM interface, are shown in Figure 9. An elastic Figure 9. An elastic element with a bulk modulus of 2e5kPa, and a shear modulus of 1e5kPa was element with a bulk modulus of 2e5kPa, and a shear modulus of 1e5kPa was used to simulate the used to simulate the load-plate (sand) and subgrade (native soil). load-plate (sand) and subgrade (native soil). Table 2. Mechanical properties of geosynthetics (modified from Lin, 2009 [15]). Table 2. Mechanical properties of geosynthetics (modified from Lin, 2009 [15]).

Geosynthetics Thickness (mm) Tensile Stiffness (kN/m) (mm) Tensile Stiffness TexturedGeosynthetics HDPE geomembrane Thickness1.5 380(kN/m) Textured HDPE geomembrane 1.5 4 380 33.3 Geotextile(400 g/m2) Geotextile(400 g/m2 )

4

33.3

Table 3. Friction parameters of the various interfaces (modified from Lin, 2009 [15]). Table 3. Friction parameters of the various interfaces (modified from Lin, 2009 [15]).

Interface

Cohesion (kPa)

Interface

Cohesion (kPa)

Granular soil/GT Granular soil/GT GT/GM GT/GM GT/Native GT/Nativesoil soil

Friction Angle (°) Friction Angle (˝ )

0

24

0 33 00

(MPa/m)

(MPa/m)

70

24 19.8 19.8 2525

70 60 60 40 40

100

7 6 4

7

6 4

Normal stress-25kPa Normal stress-54kPa Normal stress-99kPa Normal stress-213kPa Normal stress-237kPa

80

Shear stress /kPa

Normal Stiffness Shear Stiffness Normal(MPa/m) Stiffness Shear (MPa/m) Stiffness

60 40 20 0 0

10

20

30

40

50

60

70

Shear displacement /mm Figure 9. 9. Geotextile-geomembrane friction tests (reproduced from Lin, 2009 [15]). Figure Geotextile-geomembrane friction tests (reproduced from Lin, 2009 [15]).

AA series of interface tests tests conducted on a large apparatus (made by Geotest series of interface conducted on adirect largeshear direct shear apparatus (made Company, by Geotest Shaoxing, China), and the standard used in the test is ASTM (American Society for Testing Society Material) Company, Shaoxing, China), and the standard used in the test is ASTM (American for D5321. Interfaces are given the normal and shear stiffness values, and the interface shear strength. Testing Material) D5321. Interfaces are given the normal and shear stiffness values, and the In the shear simulated direct shear test, a constant shearing speed of 1 mm/min was used together interface strength. with the normal stresses of 25, shear 50, 100 anda 200 kPa. The tension developed at the beam anchorages In the simulated direct test, constant shearing speed of 1 mm/min was used together inwith all ofthe thenormal numerical analysis is shown in Figure 10. It can be seen that the pulling force of the stresses of 25, 50, 100 and 200 kPa. The tension developed at the beam anchorages geosynthetic is dependent the position. The10. relative displacements at the force interfaces in all of thematerials numerical analysis ison shown in Figure It canshear be seen that the pulling of the are shown in Figure 11. As the peak strength is minimum at the GT/GM interface, the largest relative geosynthetic materials is dependent on the position. The relative shear displacements at the shear displacements occurred at interface-3, the displacements measured between synthetic interfaces are shown in Figure 11. As the and peak strength is minimum at the GT/GMthe interface, the waste and the geotextile (interface-4) do not reach the post peak values. While the characteristics of largest relative shear displacements occurred at interface-3, and the displacements measured interface-2 are similar, but since the GM tensile stiffness is relatively it therefore between and the interface-3 synthetic waste and the geotextile (interface-4) do not reach thehigher, post peak values. undertakes a larger shear force coming from the upper part of the composite liner. This limits the While the characteristics of interface-2 and interface-3 are similar, but since the GM tensile stiffness

is relatively higher, it therefore undertakes a larger shear force coming from the upper part of the

Int. J. Environ. Res. Public Health 2016, 13, 126 Int.J.J.Environ. Environ.Res. Res.Public PublicHealth Health2016, 2016,13, 13,126 126 Int.

10 of 17 10ofof17 17 10

composite liner.This Thislimits limitsthe therelative relative displacement transferring tointerface-2 interface-2 and interface-1. So, composite liner. displacement transferring interface-1. So, relative displacement transferring to interface-2 and interface-1. So, to there is only aand small tensile stress there is only a small tensile stress in the geotextile under the GM, as shown in Figure 10. there only a small tensile stress the geotextile in theisgeotextile under the GM, asin shown in Figureunder 10. the GM, as shown in Figure 10.

Tensile Tensileforce force/(kN/m) /(kN/m)

25 25 20 20

Beam-1 Beam-1 Beam-2 Beam-2

15 15

Beam-3 Beam-3

10 10 55 00

00

50 50

100 150 100 150 Appliedload load/kPa /kPa Applied

200 200

250 250

Figure10. 10.Tension Tensionininthe thegeosynthetic, geosynthetic,atatanchorage anchoragefrom fromFLAC FLACmodels. models. Figure Figure 10. Tension

Relative Relativeshear sheardisplacement/m displacement/m

0.2 0.2 0.16 0.16 0.12 0.12

interface-1 interface-1 interface-2 interface-2

0.08 0.08

interface-3 interface-3 interface-4 interface-4

0.04 0.04 00

00

Left Left

0.5 1.5 0.5 11 1.5 Location of monitoring point /m Location of monitoring point /m

22

Right Right

Figure 11.Relative Relativeinterface interfaceshear sheardisplacements displacementsbetween betweendifferent differentinterfaces interfaces(Normal (Normalstress stress===100 100kPa). kPa). Figure 11. displacements between different interfaces (Normal stress Figure 11. Relative interface shear

According to the results of the models, the trends of the tensile stresses in the According to to the the results results of of the the numerical numerical models, models, the the trends trends of of the the tensile tensile stresses stresses in in the the According numerical geosynthetics and the relative displacements at the interface give the quantitative data. The sliding geosynthetics and the relative displacements at the interface give the quantitative data. The sliding geosynthetics and the relative displacements at the interface give the quantitative data. The sliding path the side slope ofof Shenzhen landfill is most likely to occur between the pathwithin withincomposite compositeliner lineratatat the side slope ofShenzhen Shenzhen landfill most likely occur between path within composite liner the side slope landfill isismost likely totooccur between upper GT/GM interfaces (interface 3). theupper upperGT/GM GT/GMinterfaces interfaces(interface (interface3). 3). the Therefore, when considering the stability which consists ofofof a acomposite liner layer, it Therefore, when considering the stability landfill which consists acomposite composite liner layer, Therefore, when considering the stabilityof ofofaaalandfill landfill which consists liner layer, is important to know how much stress is transferred between the interfaces. The weakest interface as important toto know know how how much much stress stress isis transferred transferred between between the the interfaces. interfaces. The The weakest weakest itit isis important well as the most likely slip surface must be identified, so that a reasonable conclusion can be drawn. interface as well as the most likely slip surface must be identified, so that a reasonable conclusion interface as well as the most likely slip surface must be identified, so that a reasonable conclusion canbe bedrawn. drawn. can 5. Back-Analysis of the Waste Landslide

Back-Analysis ofthe theWaste Waste Landslide To further understand theLandslide Shenzhen landfill landslide, an experiment using a physical model 5.5.Back-Analysis of was carried out atunderstand an enhanced-g level. In order to generate thean enhanced-g, geotechnical centrifuge Tofurther further theShenzhen Shenzhen landfill landslide, experimentausing usingaaphysical physical model To understand the landfill landslide, an experiment model atwas China Institute of Water Resources and Hydropower Research (IWHR) was used. The IWHR carried out out atat an an enhanced-g enhanced-g level. level. In In order order toto generate generate the the enhanced-g, enhanced-g, aa geotechnical geotechnical was carried geotechnical centrifuge apparatus has a design capacity ofHydropower 450g–ton, a maximum model acceleration centrifuge at China Institute of Water Resources and Research (IWHR) was used. used. centrifuge at China Institute of Water Resources and Hydropower Research (IWHR) was of 300 g and a maximum payload mass of 1500 kg, as shown in Figure 12. The centrifuge modeling TheIWHR IWHRgeotechnical geotechnicalcentrifuge centrifugeapparatus apparatushas hasaadesign designcapacity capacityofof450g–ton, 450g–ton,aamaximum maximummodel model The technique is commonly used toaexamine deformations and instability ofkg, geological slopes [23]. Earlier acceleration of 300 g and maximum payload mass of 1500 as shown in Figure 12. acceleration of 300 g and a maximum payload mass of 1500 kg, as shown in Figure 12. studies on groundwater fluctuations induced slope instability in a centrifuge include Timpong et al. [24], The centrifuge modeling technique is commonly used to examine deformations and instability The centrifuge modeling technique is commonly used to examine deformations and instability ofof Zheng and Tang [25], Knappett and Yang et al. [27]. Previous studiesinduced by the authors were basedin on geological slopes [23]. Earlier[26] studies ongroundwater groundwater fluctuations slopeinstability instability geological slopes [23]. Earlier studies on fluctuations induced slope in aa centrifugeinclude includeTimpong Timpongetetal. al.[24], [24],Zheng Zhengand andTang Tang[25], [25],Knappett Knappett[26] [26]and andYang Yangetetal. al.[27]. [27]. centrifuge

Int. J. Environ. Res. Public Health 2016, 13, 126 Int. J. Environ. Res. Public Health 2016, 13, 126

11 of 17 11 of 17

Previous studies by the authors were based on simple hypothetical steep slope models of inclinations of 60°–90° (i.e.,slope initially unstable slopes) [28]. main objective of unstable this experiment to ˝ (i.e., simple hypothetical steep models of inclinations of The 60˝ –90 initially slopes) is [28]. gain a quantitative understanding between the variations in pore water pressure and their effect on The main objective of this experiment is to gain a quantitative understanding between the variations the slope landfill stability. In addition, none of the earlier studies made reference to a full-scale in pore water pressure and their effect on the slope landfill stability. In addition, none of the earlier landfill landslide. studies made reference to a full-scale landfill landslide.

Figure Figure 12. 12. The The centrifuge centrifuge facility facility at at IWHR. IWHR.

5.1. Preparation of Model Slope Fine sand, kaolin clay, and peat are the main materials used to model the MSW. The stress-strain curves, density, void ratio, shear strength, loading and unloading curves, strength, permeability permeability coefficient, loading unloading characteristics parameters parameters could could fit fit well well with with some some of of the the features features of of the the MSW MSW (Zhu, (Zhu, et et al. al. [11]; [11]; characteristics In this study, the model waste was assumed isotropic, and the anisotropy effects Peng, et al. [29]). In (especially permeability anisotropy) were not considered. The moisture content of model waste was 45%. To construct construct the model model slope, the model model waste was compacted, compacted, such that the unit weight and void ratio should be equal to 99 kN/m and1.6 1.6 respectively. respectively. Table Table 4 shows the main properties of the kN/m33 and waste. model waste. Table 4. Physical properties of model waste. Table 4. Physical properties of model waste.

Model Fill Moisture Model Moisture Fill Age Waste Content Waste Age Content Middle 3.6–6.6 3.6–6.6 Middle 45% 45% aged waste year year aged waste

Void Void Ratio Ratio

Unit Unit Weight Weight

1.6 1.6

9 kN/m kN/m33

Coefficient Permeability Coefficient of of Permeability Compressibility Compressibility Coefficient Coefficient ´1 −1 2.62 MPa 2.62 MPa

Shear Strength Cohesion = 15 kPa, Cohesion = 15 kPa, −4 cm/s 6.6 6.6 ˆ 10 ×´4 10cm/s ˝ ϕ =ϕ28 = 28° Shear Strength

The model was constructed constructed in in aa rigid rigid aluminum aluminumbox boxwith withinner innerdimensions dimensionsofof135 135ˆ× 40 40ˆ × 90 cm × width (length ˆ width ׈height). height).Based Basedon onthe thesection sectionmap mapalong along the the x–y x–y profile profile of the Shenzhen landfill (Figure 2), the gravity acceleration in the model was designed to be 90 g. Figure 13 shows the model slope, which corresponds to the prototype slope slope of of approximately approximately 108 108 m m long. long. If an anacceleration accelerationofof N times the Earth’s gravity (g) is applied to the centrifuge model, N times the Earth’s gravity (g) is applied to the centrifuge model, according according to the laws, stiffness the tension must be N intimes smaller the model as to the scaling laws,scaling the tension muststiffness be N times smaller the model as in compared to the compared to the prototype, if the interface frictional behavior is to be the same as the prototype [11]. prototype, if interface frictional behavior is to be the same as the prototype [11]. It is impractical It is impractical to recreate completely a prototype composite system structure in test. the to recreate completely a prototype composite liner system structureliner in the centrifugal model centrifugal test. Aliner simplified linermade system therefore made for the model. A simplifiedmodel composite system composite was therefore for was the model. Based on the previous analysis, the weakest interface of the composite liner layer is the upper GT/GM GM, GT/GMinterface. interface.Because Becausethe thetensile tensile stiffness stiffness of of GT GT is is reduced reduced by one order of magnitude than GM, the deformation deformationof ofthe theGM GMisismuch much smaller than that of GT. in this experiment, the which GM which smaller than that of GT. So, So, in this experiment, the GM used used the was fieldadhered was adhered theofbase the model. A thinner GTa with a thickness of 0.5 mm, a in thein field to the to base the of model. A thinner GT with thickness of 0.5 mm, a tensile

Int. J. Environ. Res. Public Health 2016, 13, 126

12 of 17

strength smaller than that in the actual site, but has similar interface frictional angles (Figure 9) was anchored to the GM on the right side.

Figure 13. Experimental setup of model slope (Note: all dimensions are in mm).

To measure the surface settlements of the slope, three laser displacement transducers (LS) were placed on the surface of the slope. To measure the pore water pressure within the slope, four pore pressure transducers (PPT) were installed at the base of the model waste body. Figure 13 shows the positions of LS and PPT. One additional PPT (P0 in Figure 13) was placed at the bottom of seepage tank. 5.2. Test Procedures and Results The centrifuge was subjected to a stepwise increase to the design acceleration of 90 g. Figures 14 and 15 respectively show the monitored time histories of the surface settlements and pore water pressures during the centrifuge flight. Upon reaching the design test acceleration, the water level in the model waste body was raised by introducing water into the water supply chamber at the upslope end at time t2 . Figure 15 clearly shows at time t2 , the positive pore water pressure started to buildup. Between time t2 and t3 , the water supply to the chamber was kept at a constant rate of 800 mL/min and the groundwater table moved upward. This period between time t2 and t3 was approximately 7 min. When the rate of water supply to the chamber increased to 2000 mL/min at time t4 , there was a significant surface settlement as shown by the monitored displacement of L1 transducer. This phenomenon is useful to predict the occurrence of slope failure in the test model (because L1 laid above the side slope in the model, if it slips at this area, an additional settlement occurs). Back calculations were performed using the measured pore water pressure to establish the variations of groundwater table in the model slope, as shown in Figure 16. When the rate of water supply to the chamber was at 2000 mL/min, the back calculations show that the depth between the ground surface and the groundwater table was about 2.5 cm (corresponding to 2.25 m in the prototype). This result is very similar to the monitored data in the field on 6~7 June. During the model testing between time t3 and t4 , flooding was observed near the toe of the slope. In the back calculations, the groundwater table appeared to be higher than the ground surface. This may be attributed to the water flowing out through the tension cracks, which resulted in flooding at the toe of the slope.

Int. J. Environ. Res. Public Health 2016, 13, 126 Int. J. Environ. Res. Public Health 2016, 13, 126 Int. Int. J.J. Environ. Environ. Res. Res. Public Public Health Health 2016, 2016, 13, 13, 126 126

13 of 17 13 of 17 13 13 of of 17 17

Displacement/mm Displacement/mm

Time/sec

0 00 -5 -5 -5 -10 -10 -10 -15 -15 -15

0 00

Time/sec Time/sec 1000 1500 1000 1500 1000 1500

500 500 500

2000

2000 2000

2500

2500 2500

L1 L1 L1 L2 L2L2

-20 -20 -20

L3 L3L3

-25 -25 -25 -30 -30 -30 -35 -35 -35

tt11

tt3t33

ttt222

tt44t4

-40 -40 -40 Figure Variations settlement with time model slope. Figure 14. Variations of slope. Figure14. 14.Variations Variationsof ofsurface surface settlement settlement with time inin model slope. Figure 14. of surface settlementwith withtime timein inmodel model slope. 350 350 350

P0 P0 P0 P1 P1 P1 P2 P2 P2 P3 P3 P3 P4 P4 P4

Pore water pressure/kPa Pore water pressure/kPa Pore water pressure/kPa

300 300

300

250 250 250 200 200 200

tt11 t

tt22 t

1

tt33 t

2

tt44 t

3

BB

AA

CC

C

4

B

A

150 150 150 100 100 100 50 50 50 00 0 00 0

500 500 500

1000 1000

1500 1500 1000 1500 Time/sec Time/sec

2000 2000

2500 2500

2000

2500

Time/sec Figure water pressurewith withtime timein modelslope. slope. Figure 15. Variations of Figure15. 15.Variations Variationsof of pore pore water water pressure pressure with time ininmodel model slope. Figure 15. Variations of pore water pressure with time in model slope.

100 100 Time Time (Figure.15) (Figure.15) Time AA (Figure.15) BB ACC

100 80 80

y/cmy/cm

80 60 60

Seepage SeepageRate Rate (ml/min) (ml/min) Seepage 800 800 Rate (ml/min) 1400 1400 800 2000 2000

B

1400

C

2000

Measured MeasuredTotal Total Head(cm) Head(cm)

Measured Total Head(cm)

Slope Slopeat at90g 90g

60 40 40

Original Originalslope slope

Slope at 90g Original slope

40 20 20 20 00 0

00

0

25 25

25

50 50

75 75

50

x/cm x/cm 75

100 100

100

125 125

125

Figure 16. Variations water Figure ground waterlevel levelin inmodel modelslope. slope. Figure16. 16.Variations Variations of of ground ground x/cm water level in model slope.

Figures 17 and show the photographs of failure. slope did not experience Figures and1818 18show showthe the photographs of the the model model slope failure. The not experience Figure 16.photographs Variations of of ground waterslope level modelThe slope. Figures 1717 and the model slopein failure. Theslope slopedid did not experience aa large large slide slide but but tension tension cracks cracks were were developed developed in in the the landfill landfill indicating indicating the the onset onset of of sliding. sliding. a large slide but tension cracks were developed in the landfill indicating the onset of sliding.

Figures 17 and 18 show the photographs of the model slope failure. The slope did not experience a large slide but tension cracks were developed in the landfill indicating the onset of sliding.

Int. J. Environ. Res. Public Health 2016, 13, 126 Int. J. Environ. Res. Public Health 2016, 13, 126

14 of 17 14 of 17

Deep slip surface along the interface of geotextile and geomembrane in the liner system at the side slope was The the slipinterface distanceofatgeotextile this location of the model was upliner to about cm, Deep slipobserved. surface along and geomembrane in the system1 at thebut at the theobserved. slope, theThe displacement the GT/GM interfaces sidebase slopeofwas slip distancebetween at this location of the model was was up tosmaller, about 1 as cm,shown but at in Figure 17. It thatthe thedisplacement shear stress on the liner lessinterfaces than the shear strength. The results the base ofmeans the slope, between the was GT/GM was smaller, as shown in Figure It meansfirst thatoccurred the shearalong stress the on the wasof less the shear Thecontinued results show that17.slippage sideliner slopes thethan landfill, and strength. the failure thatalong slippage along slopes of the This landfill, and mode the failure continued with to to show develop thefirst baseoccurred towards the the toe side of the landfill. failure is consistent develop the base towardsanalyses, the toe of the landfill. failure mode is consistent withhas Filzbeen et al.’s [20] to Filz et al.’s along [20] finite difference where the This progressive failure mechanism used finite this difference analyses, where the progressive failure mechanism has been used to explain explain phenomenon. this phenomenon.

Figure mode(side (sideview). view). Figure17. 17.Photographs Photographs of of failure failure mode

Figure 18. Photograph of failure mode (top view). Figure 18. Photograph of failure mode (top view).

In addition, as can be seen in Figure 14, the model waste has experienced a larger In addition, as can be seen in Figure 14, the model waste has experienced a larger auto-settlement auto-settlement before the failure of the landfill slope. So, for this failure mode, it is a challenge to before the failure of the landfill slope. So, for this failure mode, it is a challenge to select parameters select parameters for the slope stability calculation. It is because the MSW is a strain hardening for the slope stability calculation. It is because the MSW is a strain hardening material [11], but material [11], but geosynthetics interface possess mainly strain softening properties. Based on the geosynthetics interface possess mainly strain softening properties. Based on the numerical analysis in numerical analysis in Section 4, it is necessary to develop a property constitutive model which can

Int. J. Environ. Res. Public Health 2016, 13, 126

15 of 17

Section 4, it is necessary to develop a property constitutive model which can reflect the characteristics of the MSW and the geosynthetics interface. Then, both the stability and lining system integrity could be considered. It is important to note that the section in Figure 7 is much closer to the bulge area, where a ridge extended to the bottom of landfill site. Due to the steep terrain, a deep sliding deformation might appear at this location. But when considering the overall stability of the slope, x–y section that used in this model test (as shown in Figure 2) is a more representative area that is closer to the middle portion of the sliding mass. Hence, there are slightly different deformation and failure characteristics between the centrifugal model test and the monitoring results. On the other hand, the data collected under controlled laboratory conditions are useful for verifying the results of the analytical and numerical modelling methods. 6. Conclusions Based on the monitored field data at Shenzhen landfill and the back-analyses, the following conclusions are reached: (1)

(2)

(3)

(4)

Due to the failure of the leachate drainage system, the water level was relatively high. With increasing infiltration, this led to a further rise in the water level, which increased the weight of the MSW. As such, the waste body experienced a large deformation, which triggered the landslide within the liner system under the drag load. The rescue measures implemented after the landslide proved to be effective. These measures included pumping and draining water in the waste body so as to lower the water level; laying plastic film on top of the waste body so as to reduce infiltration to the body; unclogging the drains so that the surface water can flow out; and using sandbags to exert a back pressure to the bulging area at the foot of the slope. Based on interface test results of the geosynthetics, the shearing behavior of the landfill composite liner system was investigated using FLAC numerical simulation to conduct the integrated shearing test. The output data include the axial strain in geosynthetics, relative shear displacements and deformations. It provides a basis for decision-making, which used a simple geosynthetic liner system in centrifugal model test. For a typical geological profile of the landslide, a centrifuge model was used to quantitatively evaluate the effect of water level on the landfill instability. The failure process shows that progressive failure can occur along geosynthetic interfaces in lined waste landfills when peak strengths are greater than residual strengths.

This research provides an improved understanding of the physical behavior and failure mode of MSW landfill subjected to high water level. It has also shown that the deformation characteristics of waste (strain hardening), and the strain softening properties of geosynthetic interfaces should be considered in future stability analysis. Acknowledgments: This work was supported by the National Basic Research Program (973) of China (2012CB719803, 2014CB047001). And deep appreciations give to ZuyuCHEN of IWHR and Yunmin CHEN of Zhejiang University, for their patient instructions to this paper. Author Contributions: Ren Peng performed the analysis and drafted the manuscript. Yujing Hou and Yangping Yao designed the research. Liangtong Zhan collected the field data. All authors discussed the results and implications, and commented on the manuscript at all stages, contributed extensively to the work presented in this paper. Conflicts of Interest: The authors declare no conflict of interest.

References 1.

Zhan, L.T.; Guan, Q.R.; Chen, Y.M.; Liu, Z. Monitoring and back analyses of slope failure process at a landfill. Chin. J. Rock Mech. Eng. 2010, 8, 1697–1705.

Int. J. Environ. Res. Public Health 2016, 13, 126

2.

3.

4. 5. 6. 7. 8. 9.

10. 11.

12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25.

16 of 17

Hendron, D.M.; Fernandez, G.; Prommer, P.J.; Giroud, J.P.; Orozco, L.F. Investigation of the cause of the 27 September 1997 slope failure at the Dona Juana landfill. In Proceedings of the Seventh International Waste Management and Landfill Symposium, Cagliari, Italy, 4–8 October 1999; pp. 545–554. Caicedo, B.; Giraldo, E.; Yamin, L. The landslide of Dona Juana landfill in Bogota. A case study. In Environmental Geotechnics, 4th ICEG; de Mello, L.G., Almeida, M., Eds.; Balkema: Lisse, The Netherlands, 2002; Volume 1, pp. 171–175. Koerner, R.M.; Soong, T.Y. Stability assessment of ten large landfill failures. Geotech. Spec. Publ. 2000, 103, 1–38. Lan, J.W. Mechanism of Leachate Generation, Transport and Mound in MSW Landfills and Control of Leachate Level. Ph.D. Thesis, Zhejiang University, Hangzhou, China, 9 January 2012. Merry, S.M.; Kavazanjian, J.E.; Fritz, W.U. Reconnaissance of the10 July 2000, Payatas landfill failure. J. Perform. Constr. Facil. 2005, 2, 100–107. [CrossRef] Eid, H.T.; Stark, T.D.; Evans, W.D.; Sherry, P.E. Municipal solid waste slope failure. I: Waste and foundation soil properties. J. Geotech. Geoenviron. 2000, 5, 397–407. [CrossRef] Stark, T.D.; Eid, H.T.; Evans, W.D.; Sherry, P.E. Municipal solid waste slope failure. II: Stability analyses. J. Geotech. Geoenviron. 2000, 5, 408–419. [CrossRef] Gunn, D.A.; Chambers, J.E.; Hobbs, P.R.N.; Ford, J.R.; Wilkinson, P.B.; Jenkins, G.O.; Merritt, A. Rapid observations to guide the design of systems for long-term monitoring of a complex landslide in the upper lias clays of North Yorkshire, UK. J. Eng. Geol. Hydrog. 2013, 3, 323–336. [CrossRef] Jessberger, H.L.; Kockel, R. Determination and assessment of the mechanical properties of waste materials. Proc. Sardinia 1993, 93, 1383–1392. Zhu, B.; Yang, C.B.; Wang, L.; Chen, Y.M. Compounding model MSW and centrifugal model tests of landfill. In Proceeding of the 1st China National Symposium on Coupled Phenomena in Geomaterials and Environmental Geotechnics, Hangzhou, China, 24–26 November 2011; pp. 425–432, in Chinese. Thusyanthan, N.I.; Madabhushi, S.P.; Singh, S. Centrifuge modeling of solid waste landfill systems. Part 1: Development of a model municipal solid waste. ASTM Geotech. Testing. J. 2006, 3, 217–222. Seed, R.B.; Mitchell, J.K.; Seed, H.B. Kettleman hills waste landfill slope failure. II: Stability analyses. J. Geotech. Eng. 1990, 4, 669–690. [CrossRef] Huvaj-Sarihan, N.; Stark, T.D. Back analyses of landfill slope failures. In Proceedings of the 6th International Case Histories Conference, Arlington, AV, USA, 11–16 August 2008; pp. 11–16. Lin, W.A. Shear Stress Transfer, Strength Characteristics and Safety Control of Composite Liner Systems. Ph.D. Thesis, Zhejiang University, Hangzhou, China, 6 June 2009. Jones, D.R.V.; Dixon, N. Landfill lining stability and integrity: The role of waste settlement. Geotext Geomembr. 2005, 23, 27–53. [CrossRef] Long, J.H.; Gilbert, R.B.; Daly, J.J. Effect of waste settlement on sloped lining systems. In Proceedings of the Geosynthetics ’95, Nashville, TN, USA, 21–23 February 1995; Volume 2, pp. 729–744. Villard, P. Modelling of interface problems by the finite element method with considerable displacements. Comput. Geotech. 1996, 19, 23–45. [CrossRef] Meiβner, H.; Abel, K. Numerical investigations of inclined landfill liner systems. In Proceedings of the GeoEng 2000, Melbourne, Australia, 19–24 November 2000, published on CD. Filz, G.M.; Esterhuizen, J.J.B.; Duncan, J.M. Progressive failure of lined waste impoundments. J. Geotech. Geoenviron. ASCE 2001, 127, 841–848. [CrossRef] Fowmes, G.J. Analysis of Steep Sided Landfill Lining Systems. Ph.D. Thesis, Loughborough University, Leicestershire, UK, November 2007. Itasca Consulting Group, Inc. Flac 7.0 Slope Stability Analysis. Available online: http://www.itascacg.com/ software/flac/videos/flac-70-slope-stability-analysis (accessed on 8 September 2015). Taylor, R.N. Geotechnical Centrifuge Technology; CRC Press: Glasgow, UK, 2003. Timpong, S.; Itoh, K.; Toyosawa, Y. Geotechnical Centrifuge Modelling of Slope Failure Induced by Ground Water Table Change // Landslides and Climate Change; Taylor and Francis Group: London, UK, 2007; pp. 107–112. Zheng, Y.; Tang, X. Stability analysis of slopes under drawdown condition of reservoirs. Chin J. Geotech Eng. 2007, 8, 1115–1121.

Int. J. Environ. Res. Public Health 2016, 13, 126

26.

27. 28. 29.

17 of 17

Knappett, J. Investigation of Slope Stability Influenced by Change of Reservoir Water Level in Three Gorges of China. Available online: http://www.seiofbluemountain.com/upload/product/201002/ 12657115158jkxyle5.pdf (accessed on 8 September 2015). Yang, C.B.; Zhu, B.; Kong, L.G.; Han, L.B.; Chen, Y.M. Centrifugal model tests on failure of silty slopes induced by change of water level. Chin. J. Geotech. Eng. 2013, 7, 1261–1271. Ling, H.; Ling, H.I. Centrifuge model simulations of rainfall-induced slope instability. J. Geotech. Geoenviron. ASCE 2012, 9, 1151–1157. [CrossRef] Peng, R.; Hou, Y.J.; Chen, Z.Y.; Zhang, X.D.; Sun, Q.L.; Yao, Y.P. Centrifuge tests on the instability of valley-type municipal solid waste landfill. J. Civ. Architect Environ. Eng. 2014, 6, 67–72. © 2016 by the authors; licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons by Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).

Back-Analyses of Landfill Instability Induced by High Water Level: Case Study of Shenzhen Landfill.

In June 2008, the Shenzhen landfill slope failed. This case is used as an example to study the deformation characteristics and failure mode of a slope...
7MB Sizes 0 Downloads 19 Views