b15621-12 (1)

8/20/2019 b15621-12 (1)

http://slidepdf.com/reader/full/b15621-12-1 1/46

315

11.1 Introduction

Landslides are requently responsible or considerable losses o both money and lives. Te severity o

the landslide problem intensifies with increased urban development and change in land use. In view

o this consideration, it is not surprising that landslides are rapidly becoming the ocus o major scien-

tific research, engineering study and practice, and land-use policy throughout the world. International

cooperation among various individuals concerned with the fields o geology, geomorphology, and soil

and rock mechanics has recently contributed to improvement o our understanding o landslides in

recent years.Landslides and related slope instability phenomena plague many parts o the world. Japan leads

other nations in landslide severity with projected combined direct and indirect losses o $4 billion

11Landslide RiskAssessment and

Mitigation

Mihail E. PopescuIllinois Institute

of Technology

Aurelian C.

TrandafirFugro GeoConsulting Inc.

11.1 Introduction ...................................................................................... 31511.2 Landslide Hazard Assessment and Landslide Risk Management ...................................................................................... 316

Landslide Hazard Assessment • Landslide Risk Management

Process • Landslide Inventory Maps

11.3 Landslide ypes and Causal Factors ..............................................320Landslide Classification • Landslide Causal Factors

11.4 Slope Stability Analyses and Selection o Design SoilParameters .........................................................................................325Introductory Remarks • Methods o Slope Stability Analysis • Shear

Strength Parameters or Slope Stability Analysis • Backward Analysiso Slope Stability • Seismic Slope Stability Analysis

11.5 Landslide Risk Mitigation ...............................................................336Landslide Risk reatment Options • Landslide Remedial

Measures • Levels o Effectiveness and Acceptability Tat May Be

Applied in the Use o Remedial Measures

11.6 Landslide Monitoring and Warning Systems .............................. 351Landslide Monitoring • Landslide Warning Systems • Forecasting

the ime o Landslides

11.7 Concluding Remarks ........................................................................354Reerences ......................................................................................................355Further Reading ............................................................................................358

Relevant Websites .........................................................................................359

8/20/2019 b15621-12 (1)


316 Bridge Engineering Handbook, Second Edition: Substructure Design

annually (Schuster, 1996). Te United States, Italy, and India ollow Japan, with an estimated annual

cost ranging between $1 billion and $2 billion. Landslide disasters are also common in develop-

ing countries, and monetary losses sometimes equal or exceed the gross national product o these

countries.

Te paramount importance o landslide hazard management or transportation acilities, includingbridges, is by and large recognized. Repairs and maintenance afer landslides on U.S. highways cost an

estimated $106 million annually.

As an integral part o transportation systems, bridges are designed to move people, goods, and ser-

vices efficiently, economically, and saely. Landslides can disrupt or damage these systems at a variety o

spatial and temporal scales, dramatically reducing network serviceability, increase costs, and decrease

saety. Recurrence intervals or landslide events span rom daily to centuries, whereas the associated

consequences range rom inconvenient to catastrophic.

Landslides can significantly impact bridges. Tey can knock out bridge abutments or significantly

weaken the soil supporting them, making bridge structure unusable or hazardous or use. In addition

to the damage or reduced serviceability o the structure, in some instances, landslides can crush or bury

vehicles and result in death. Some landslides occur unexpectedly, whereas others arrive with significantwarning, but all are amenable to some level o prediction and mitigation.

Herein lies the guiding principle o the current chapter, that is, to describe landslide hazards and

methods to mitigate the associated risks in an appropriate and effective manner.

Tis chapter provides the basic principles and inormation needed by the bridge engineer to plan and

design sae and cost-effective structures in areas prone to or already affected by landslides.

11.2 Landslide Hazard Assessment and

Landslide Risk Management11.2.1 Landslide Hazard Assessment

Landslide hazard identification requires an understanding o the slope processes and the relationship

o those processes to geomorphology, geology, hydrogeology, climate, and vegetation. From this under-

standing, it wil l be possible to

• Classiy the types o potential landsliding. Te classification system proposed by Varnes

(1978) as modified by Cruden and Varnes (1996) constitutes a suitable system. It should be

recognized that a site may be affected by more than one type o landslide hazard. For example,

deep-seated landslides occur at the site, whereas rockall and debris flows will initiate rom

above the site.• Assess the physical extent o each potential landslide being considered, including the location,

areal extent, and volume involved.

• Assess the likely causal actor(s), the physical characteristics o the materials involved, and the

slide mechanics.

• Estimate the resulting anticipated travel distance and velocity o movement.

• Address the possibility o ast-acting processes, such as flows and alls, rom which it is more di-

ficult to escape.

Methods commonly used to identiy hazards include geomorphological mapping, gathering o his-

toric inormation on landslides in similar topography, geology, and climate (e.g., rom maintenance

records, aerial photographs, newspapers, review o analysis o stability). Some orms o geologic andgeomorphic mapping are considered to be an integrated component o the fieldwork stage when assess-

ing natural landslides, which requires understanding the site while inspecting it.

8/20/2019 b15621-12 (1)


317Landslide Risk Assessment and Mitigation

Beore embarking on a regional landslide hazard assessment, the ollowing preparatory steps are to

be taken (Hutchinson, 2001):

1. Identiy the user and purpose o the proposed assessment. Involve the user in all phases o the

program.

2. Define the area to be mapped and decide the appropriate scale o mapping. Tis may range rom1:100,000 or smaller to 1:5,000 or larger.

3. Obtain, or prepare, a good topographic base map o the area, preerably contoured.

4. Construct a detailed database o the geology (solid and superficial), geomorphology, hydrogeol-

ogy, pedology, meteorology, mining and other human intererence, history, and all other relevant

actors within the area, and o all known mass movements including all published work, newspa-

per articles, and the results o interviewing the local population.

5. Obtain all available air photo cover, satellite imagery, and ground photography o the area.

Photography o various dates can be particularly valuable, both because o what can be revealed

by differing lighting and vegetation conditions and to delineate changes in the man-made and

natural conditions, including slide development.

11.2.2 Landslide Risk Management Process

Te risk management process comprises two components: risk assessment and risk treatment. Landslide

and slope engineering have always involved some orm o risk management, although it was seldom

ormally recognized as such. Tis inormal type o risk management was essentially the exercise o

engineering judgment by experienced engineers and geologists.

Figure 11.1 shows the process o landslide risk management in a flowchart orm (Australian

Geomechanics Society—AGS, 2000). In simple orm, the process involves answering the ollowing

questions:

1. What might happen?

2. How likely is it?

3. What damage or injury may result?

4. How important is it?

5. What can be done about it?

Tere is a clear distinction between hazard, risk, and probability. Hazard is usually defined as a condi-

tion with the potential or causing an undesirable consequence (Fell, 1994b). Te description o landslide

hazard should include the location, volume (or area), classification, and velocity o potential landslides

and any resulting detached material, and the probability o their occurrence within a given period o time.

Risk is a measure o the probability and severity o an adverse effect to health, property, or the environ-ment. Risk is ofen estimated by the product o probability and consequences. However, a more general

interpretation o risk involves a comparison o the probability and consequences in a nonproduct orm.

Probability is the likelihood o a specific outcome, measured by the ratio o specific outcomes to

the total number o possible outcomes. Probability is expressed as a number between 0 and 1, with 0

indicating an impossible outcome and 1 indicating that an outcome is certain.

Te intent o a landslide hazard assessment is to identiy a region’s susceptibility to landslides and

their consequences based on several key or significant physical attributes comprising the previous

landslide activities, bedrock eatures, slope geometry, and hydrologic characteristics. In a development

program (planning process) concerning a landslide-prone area, one needs to determine the acceptable

risk. It is indispensable to recognize the vulnerability and degrees o risk involved and to instigate a

systematic approach in avoiding, controlling, or mitigating existing and uture landslide hazards inthe planning process. Accordingly, either a planner should avoid the landslide-susceptible areas i it is

deemed appropriate or else he or she needs to implement strategies to reduce risk.

8/20/2019 b15621-12 (1)



Risk analysis

Risk assessment

Risk managementRisk treatment

(or risk control)Treatment optionsAccept riskAvoid riskReduce likelihoodReduce consequencesTransfer risk

Monitor and review Risk changesMore informationFurther studies

Treatment planDetail selected options

Implement planPolicy and planning

Feedback

Reconsider

Risk evaluationCompare to levels of tolerable of acceptable riskAssess priorities and optionsClient/owner/regulator to decide to accept or treatTechnical specialist to advise

Risk estimation

Scope definitionEstablish brief, proposed methodology

Classification of landslide e.g., slide, debris flow, rockfallExtent of landslide e.g., location, area, volumeTravel distance of landslideRate of movement e.g., creep, slow, fast

Hazard identification

Consequence analysis

Elements at riskProperty Roads/communicationsServicesPeopleTravel distance

Temporal probability e.g., Vehicles, persons

Vulnerability Relative damageProbability of injury/loss of life

Risk calculation

Risk = (likelihood of slide) × (probability of spatial impact) × (temporal probability) × (vulnerability) × (elements at risk) considered for all hazards

Estimate frequency QualitativeQuantitative

Historic performance

Relate to initiating eventsRainfall

Construction activity EarthquakeServices failure/malfunction

Frequency analysis

FIGURE 11.1 Process o landslide risk management. (From Australian Geomechanics Society (AGS), Sub-

Committee on Landslide Risk Management, Landslide Risk Management Concepts and Guidelines, 49–92, 2000.

With permission.)

8/20/2019 b15621-12 (1)



11.2.3 Landslide Inventory Maps

International Union o Geological Sciences (IUGS) Working Group on Landslides (WG/L), or-

merly International Geotechnical Societies’ UNESCO Working Party on World Landslide Inventory

(WP/WLI), which was established in the ramework o the United Nations International Decade or

Natural Disaster Reduction (1990–2000), defined the landslide hazard as “the probability o occur-rence within a specified period o time and within a given area o a potentially damaging phenom-

enon” (Cruden, 1997). Furthermore, IUGS described the landslide risk assessment as “the expected

degree o loss due to a landslide (specific risk) and the expected number o lives lost, people injured,

damage to property and disruption o economic activity (total risk).” As shown in Figure 11.2, the

integrated assessment o landslide hazard and risk requires a broad-based knowledge rom a wide

spectrum o disciplines including geosciences, geomorphology, meteorology, hydrogeology, and geo-

technical engineering (Chowdhury et al., 2001).

Landslide hazards are commonly delineated on inventory maps, which display distributions o haz-

ard classes and identiy areas that potential landslides may be generated. Inventory maps show the

location and, where applicable, the date o occurrence and historical records o landslides in a region.Tese maps are prepared by different techniques, and, ideally, provide inormation concerning the

spatial and temporal probability, type, magnitude, velocity, runout distance, and retrogression limit

o the mass movements predicted in a designated area. Details o inventory maps depend on available

R e s e a r c

h , o

b s e r v a t i o n , a n

d c o m m u n i c a t i o n : i t e r a t i o

n

Elements atRisk

Risk assessment

Geology Existing

Landslides

PotentialLandslides

Analyses of historical dataElement

Vulnerability

Hazard assessment

Landslide probability

Consequences of failure

Sociopolitical andeconomic

acceptable risk

Risk prioritization andtreatment

FIGURE 11.2 Methodology or landslide risk assessment. (From Chowdhury, R. et al., A Focus on Hilly Areas

Subject to the Occurrence and Effects o Landslides, Global Blueprint for Change, 1st Edition—Prepared in conjunction with the International Workshop on Disaster Reduction, August 19–22, 2001. With permission.)

8/20/2019 b15621-12 (1)



resources and are based on the scope, extent o study area, scales o base maps, aerial photographs, and

uture land use.

Evidently, the extent o inormation required concerning the landslide hazard analysis wi ll depend on

the level and nature o proposed development or a region. Te negligence o incorporating the impact o

potential landslide activity on a project or the prospects o new development on landslide potential maylead to increased risk. Einstein (1988 and 1997) has presented a comprehensive mapping procedure or

landslide management. Following are key eatures o mapping procedures proposed by Einstein (1988

and 1997):

State-of-nature maps: Tey are used to characterize a site, present data without interpretation,

such as geologic and topographic maps, precipitation data, and results o site investigation.

Danger maps: Tey are used to identiy the ailure modes involving debris flows, rockalls,

and so on.

Hazard maps: Tey are used to exhibit the probability o ailure related to the possible modes

o ailure on danger maps. Alternatively, the results can be expressed qualitatively as high,

medium, or low. Management maps: Tey are incorporated to entail summaries o management decisions.

Furthermore, ollowing scales o analyses or landslide hazard zonations have been outlined by the

International Association o Engineering Geology (Soeters and van Westen, 1996):

• National Scale (<1:1,000,000): Tis is a low level detail map intended to provide a general inven-

tory o nationwide hazard. It is used to notiy national policy makers and general public.

• Regional Scale (1:100,000 to 1:500,000): Because landslide hazards are considered to be undesir-

able actors as ar as the planners are concerned, the regional mapping scale is used in evaluating

possible constraints due to instability related to the development o large engineering projects

and regional development plans. In general, these types o maps are constructed in early phases

o regional development projects with low level details and cover large study areas, on the ordero 1000 km2 or more. Tey are used to identiy areas where landsliding could be a constraint con-

cerning the development o rural or urban transportation projects. “errain units with an areal

extent o several tens o hectares are outlined and classified according to their susceptibility to

occurrence o mass movements,” as stipulated by Soeters and van Westen (1996).

• Medium Scale (1:25,000 to 1:50,000): Tis range is considered to be a suitable scale range or land-

slide hazard maps. As such, they are used to identiy the hazard zones in developed areas with

large structures, roads, and urbanization. Considerably greater levels o detail are required to pre-

pare the maps at this scale, and the details should encompass slopes in adjacent sites in the same

lithology with the possibility o having different hazard scores depending on their characteristics.

Furthermore, distinction should be made between various slope segments, located within thesame terrain unit, such as rating o a concave slope as opposed to a convex slope.

• Large Scale (1:5,000 to 1:15,000): Maps o this scale are generally prepared or limited areas

based on both interpretation o aerial photographs and extensive field investigations that use

various techniques applied in routine geotechnical engineering, engineering geology, and

geomorphology.

11.3 Landslide Types and Causal Factors

11.3.1 Landslide Classification

Te UNESCO Working Party’s definition o a landslide is “the movement o a mass o rock, earth ordebris down a slope” (Cruden, 1991, 1997) and recognizes that the phenomena described as landslides

are not limited either to the land or to sliding; the word has a much more extensive meaning than its

8/20/2019 b15621-12 (1)



component parts suggest. An idealized diagram depicting the nomenclature o various eatures o a

complex earth slide–earth flow is shown in Figure 11.3 (Cruden and Varnes, 1996).

As there is a wide spectrum o landslide types, the potential or already occurred landslides should

be classified as clear as easible. Te criteria used by the UNESCO Working Party on World LandslideInventory in classification o landslides ollow Varnes (1978) in emphasizing the type o movement and

type o material. Te divisions o materials are rock, debris, and earth. Rock is characterized as an intact

hard or firm mass in its natural place prior to the initial movement, whereas soil is reerred to as an

aggregate o unconsolidated solid particles, either transported or derived in place via the weathering

processes. Te latter is urther divided into earth, in which 80% or more o the particles are smaller than

2 mm in size, and debris, whereby 20%–80% o the solid particles are larger than 2 mm.

Movements are divided into five types (Figure 11.4): alls, flows, slides, spreads, and topples. In real-

ity, there is a continuum o mass movements rom alls through sl ides to flows. In many instances, it is

difficult to determine whether masses o material have allen or slid, and similarly there are a number

o instances in which material has both slid and flowed. Very large alls can result in various types o

flow involving fluidization with either water or air. Te Department o Environment (DOE, 1994) rec-ognized the existence o complex landslides in which ground displacement is achieved by more than

one type o mass movement and emphasized that this should not be conused with landslide complex,

that is, an area o instability within which many different types o mass movement occur. Cruden and

Varnes (1996) suggested that landslide complexity can be indicated by combining the five basic types

o movement and the three divisions o materials. I the type o movement changes with the progress

o movement, then the material should be described at the beginning o each successive movement.

For example, a rockall that has been ollowed by flow o the displaced material can be described as a

rockall–debris flow.

Te landslide designation can become more elaborate as more inormation about the movement

becomes available. Adjectives can be added in ront o the noun string defining the type o landslide tobuild up the description o the movement. Te adjective magnitude reers to the volume o displaced

material involved in a landslide hazard, whereas the intensity renders a collection o physical parameters

Backward rotated tree

Main scrap

Minor scrap

Original ground surface

Slope reversed

Ponded

waterswamp

Transverse cracks

TransversecracksRadial cracks

Toe

Surface of rupture(shear plane)

Transverse ridges

FIGURE 11.3 Nomenclature o various eatures o a landslide. (From Cruden, D.M., and Varnes, D.J.,

Landslide ypes and Processes, In Landslides Investigation and Mitigation, urner, A.K., and Schuster, R.L. (eds.),

ransportation Research Board Special Report 247, National Research Council, Washington, DC, 1996. With

permission.)

8/20/2019 b15621-12 (1)



that describe the destruction or destructive potential o a landslide hazard. Te qualitative expression

o the ormer is small, medium, or large, whereas the latter is expressed qualitatively as slow, moderate,

or ast as in downslope velocity o a debris flow. A landslide is known to be active when it is presentlymoving. An inactive landslide is one that last moved more than one annual cycle o seasons ago. Inactive

landslides are urther categorized into dormant i the causes o movement are apparent and abandoned

i the triggering action is no longer present (Popescu, 1984).

(a)

20 m

0 m

20 m

0 m

20 m

0 m

3 m0

500 m0

(b)

1 22

3 4

(c)

(d)

(e)

FIGURE 11.4 ypes o movement: (a) all, (b) topple, (c) slide, (d) spread, and (e) flow. (From Cruden, D.M., and

Varnes, D.J., Landslide ypes and Processes, In Landslides Investigation and Mitigation, urner, A.K., and Schuster,

R.L. (eds.), ransportation Research Board Special Report 247, National Research Council, Washington, DC, 1996.

With permission.)

8/20/2019 b15621-12 (1)



11.3.2 Landslide Causal Factors

Te processes involved in slope movements comprise a continuous series o events rom cause to effect

(Varnes, 1978). It is o primary importance to recognize the conditions that caused the slope to become

unstable and the processes that triggered the movement, when assessing landslide hazard or a particu-

lar site. Only an accurate diagnosis makes it possible to properly understand the landslide mechanismsand thence to propose effective treatment measures.

In every slope, there are orces that tend to promote downslope movement and opposing orces that

tend to resist movement. A general definition o the actor o saety o a slope results rom comparing the

downslope shear stress with the shear strength o the soil, along an assumed or known rupture surace.

Starting rom this general definition, erzaghi (1950) divided landslide causes into external causes that

result in an increase o the shearing stress (e.g., geometrical changes, unloading the slope toe, loading

the slope crest, shocks and vibrations, drawdown, changes in water regime) and internal causes that

result in a decrease o the shearing resistance (e.g., progressive ailure, weathering, seepage erosion).

However, Varnes (1978) pointed out that there are a number o external or internal causes that may be

operating either to reduce the shearing resistance or to increase the shearing stress. Tere are also causesthat simultaneously affect both terms o the actor-o-saety ratio.

Te great variety o slope movements reflects the diversity o conditions that cause the slope instability

and the processes that trigger the movement. It is more appropriate to discuss causal actors (including

both “conditions” and “processes”) than “causes” per se alone. Tus, ground conditions (weak strength,

sensitive abric, degree o weathering and racturing) are influential criteria but are not causes (Popescu,

1996). Tey are part o the conditions necessary or an unstable slope to develop, to which must be added

the environmental criteria o stress, pore water pressure, and temperature. It does not matter i the

ground is weak as such—ailure will only occur as a result i there is an effective causal process that acts

as well. Such causal processes may be natural or anthropogenic but effectively change the static ground

conditions sufficiently to cause the slope system to ail, that is, to adversely change the state o stability.

Seldom, i ever, can a landslide be attributed to a single causal actor. Te process leading to thedevelopment o a slide has its beginning with the ormation o the rock itsel, when its basic properties

are determined and includes all the subsequent events o crustal movement, erosion, and weathering.

Te computed value o the actor o saety is a clear and simple distinction between stable and unsta-

ble slopes. However, rom the physical point o view, it is better to visualize slopes existing in one o the

ollowing three stages: stable, marginally stable, and actively unstable (Crozier, 1986). Stable slopes are

those where the margin o stability is sufficiently high to withstand all destabilizing orces. Marginally

stable slopes are those that will ail at some time in response to the destabilizing orces having attained a

certain level o activity. Finally, actively unstable slopes are those in which destabilizing orces produce

continuous or intermittent movement.

Te three stability stages must be seen to be part o a continuum, with the probability o ailure beingminute at the stable end o the spectrum, but increasing through the marginally stable range to reach cer-

tainty in the actively unstable stage. Te three stability stages provide a useul ramework or understand-

ing the causal actors o landslides and classiying them into two groups on the basis o their unction:

1. Preparatory causal actors that make the slope susceptible to movement without actually initiat-

ing it, and thereby tending to place the slope in a marginally stable state.

2. riggering causal actors that initiate movement. Te causal actors shif the slope rom a margin-

ally stable to an actively unstable state.

A particular causal actor may inflict either or both unctions, depending on its degree o activity and

the margin o stability. Although it may be possible to identiy a single triggering process, an explanationo ultimate causes o a landslide invariably involves a number o preparatory conditions and processes.

Based on their temporal variability, the destabilizing processes may be grouped into slow-changing

(e.g., weathering, erosion) and ast-changing processes (e.g., earthquakes, reservoir drawdown). In the

8/20/2019 b15621-12 (1)



search or landslide causes, attention is ofen ocused on those processes within the slope system that

provoke the greatest rate o change. Although slow changes act over a long period o time to reduce the

resistance/shear stress ratio, ofen a ast change can be identified as having triggered movement.

Because the assessment o landslide causes is complex and landslides are not always investigated in

great detail, UNESCO Working Party on World Landslide Inventory (1994) adopted a simple classifica-tion system o landslide causal actors as shown in able 11.1. Te operational approach to classification

o landslide causal actors, proposed by this system, is intended to cover the majority o landslides.

TABLE 11.1 A Brie List o Landslide Causal Factors

1. Ground Conditions

(1) Plastic weak material

(2) Sensitive material

(3) Collapsible material

(4) Weathered material

(5) Sheared material

(6) Jointed or fissured material (7) Adversely oriented mass discontinuities (including bedding, schistosity,

cleavage)

(8) Adversely oriented structural discontinuities (including aults, unconormities,

flexural shears, sedimentary contacts)

(9) Contrast in permeability and its effects on groundwater

(10) Contrast in stiffness (stiff, dense material over plastic material)

2. Geomorphological Processes

(1) ectonic uplif

(2) Volcanic uplif

(3) Glacial rebound

(4) Fluvial erosion o the slope toe

(5) Wave erosion o the slope toe (6) Glacial erosion o the slope toe

(7) Erosion o the lateral margins

(8) Subterranean erosion (solution, piping)

(9) Deposition loading o the slope or its crest

(10) Vegetation removal (by erosion, orest fire, drought)

3. Physical Processes

(1) Intense, short period rainall

(2) Rapid melt o deep snow

(3) Prolonged high precipitation

(4) Rapid drawdown ollowing floods, high tides, or breaching o natural dams

(5) Earthquake

(6) Volcanic eruption (7) Breaching o crater lakes

(8) Tawing o permarost

(9) Freeze and thaw weathering

(10) Shrink and swell weathering o expansive soils

4. Man-Made Processes

(1) Excavation o the slope or its toe

(2) Loading o the slope or its crest

(3) Drawdown (o reservoirs)

(4) Irrigation

(5) Deective maintenance o drainage systems

(6) Water leakage rom services (water supplies, sewers, stormwater drains)

(7) Vegetation removal (deorestation) (8) Mining and quarrying (open pits or underground galleries)

(9) Creation o dumps o very loose waste

(10) Artificial vibration (including traffic, pile driving, heavy machinery)

8/20/2019 b15621-12 (1)



It involves consideration o the available data rom simple site investigation and inormation urnished

by other site observations. Landslide causal actors are grouped according to their effect (preparatory

or triggering) and their origin (ground conditions and geomorphological, physical, or man-made pro-

cesses). Ground conditions may not have a triggering unction, while any ground condition or process

may have a preparatory unction.Ground conditions or the material and mass characteristics o the ground can be mapped on the sur-

ace o the landslide and the surrounding ground and explored in the subsurace by drilling, trenching,

and adits. Mechanical characteristics can be determined by testing. Geomorphic processes, or changes

in the morphology o the ground, can be documented by preexisting maps, aerial photographs, surveys

o the landslide, or careul observation over time by the local population. Physical processes concern

the environment and can be documented at the site by instrumentation, such as rainall gauges, seismo-

graphs, or piezometers. Careul local observations o water wells or damage rom earthquakes may be

acceptable substitutes. Variations in mechanical properties with distance rom the surace may, in some

circumstances, indicate changes o these properties with time. Man-made processes can be documented

by site observations and rom construction or excavation records at the site. Separate identification o

man-made and natural landslides is useul or both administrative and theoretical reasons.

11.4 Slope Stability Analyses and Selectionof Design Soil Parameters

11.4.1 Introductory Remarks

Tere are two major approaches in the analysis o slope stability. Te first one is the “orward” approach

in the analysis o slope stability that requires data on shear strength properties and pore pressure condi-

tions. Te ormer are derived rom a range o field and laboratory techniques, whereas the latter demand

improved techniques capable o instrumenting rapid groundwater and soil suction responses to rainal lwithout damping the transient peak conditions. Probable worst-case parameter values are assumed, and

a conservative value o the actor o saety is derived.

Te second approach is the “backward” approach in the analysis o slope instability that requires

detailed inormation on ailure surace geometry and pore water pressure distribution. Accurate deter-

mination o the position and shape o the slip surace using surace and subsurace monitoring data is

essential to a reliable backward analysis o a given slope ailure. Considering that the actor o saety is

one, the backward analysis o the ailed slope can give a measure o the shear strength mobilized along

the slip surace. In many cases, when there are considerable difficulties in obtaining undisturbed sam-

ples, backward analysis is an effective tool, and sometimes the only tool, or investigating the strength

eatures o a soil/rock deposit. Both “orward” and “backward” approaches are generally carried out

using limit equilibrium methods o slope stability analysis.

11.4.2 Methods of Slope Stability Analysis

Slope stability assessments are generally preormed using methods o limit equilibrium. Such meth-

ods make use o the shear strength parameters along the sliding surace, but they do not require any

inormation on the stress–strain properties o the slope materials. Limit equilibrium methods assume

the slide mass as a rigid body in equilibrium and use static orce and moment equilibrium equations to

derive the slope saety actor. Consequently, limit equilibrium methods cannot provide any inormation

related to slope deormations. For a specified sliding surace, some o the limit equilibrium methods

such as the Ordinary (or Fellenius) method, Bishop’s simplified method, or Janbu’s simplified methodcan be used even with manual calculations to determine the slope saety actor.

As shown in Figure 11.5, limit equilibrium methods involve discretization o the slide mass into a

number o slices to take into account the inherent irregularities associated with the geometry o the

8/20/2019 b15621-12 (1)



sliding surace and ground surace as well as variation in strength properties o various geologic layers

intercepted by the sliding surace. In selecting the optimum number o slices or the portion o sliding

surace passing through a specific layer, the basic requirement is to approximate as closely as possible

the shape o actual nonlinear sliding surace within each slice by a planar surace o a given inclination

(α) relative to the horizontal (Figure 11.6). Tis allows or a rigorous determination o the orientation

o normal (N ) and shear (S) orces acting at the sliding surace within each slice, inormation which

is critical or the limit equilibrium equations (Figure 11.6). Additionally, vertical slice boundaries are

introduced at the points o change in geometry o the ground surace (Figure 11.5).Figure 11.6 shows the orces acting on an individual slice within the slide mass. Te variables rom

Figure 11.6 are defined in the ollowing (Krahn, 2004):

W = total weight o a slice o width b and height h.

N = total normal orce on the base o the slice.

Sm = shear orce mobilized at the base o each slice.

E = horizontal interslice normal orces. Subscripts L and R designate the lef and right sides o the

slice, respectively.

X = vertical interslice shear orces. Subscripts L and R define the lef and right sides o the slice,

respectively.

D = an external point load.kW = horizontal seismic load applied through the centroid o each slice.

R = radius o a circular slip surace or the moment arm associated with the mobilized shear orce,

Sm, or any shape o slip surace.

N

D

Water A L

E L

E R

S m

X L

X R

Tensioncrack

zone

Bedrock

W

A R

kW

f

x

Ra L

h

e

b

a R

d

FIGURE 11.6 Forces acting on a slice o the slide mass. (Afer Krahn, J., Stability Modeling with Slope/W. An Engineering Methodology , Geo-Slope/W International Ltd., 396, 2004.)

Ground surface

Slice within slide mass

Sliding surface

FIGURE 11.5 Slice discretization o the slide mass.

8/20/2019 b15621-12 (1)



f = perpendicular offset o the normal orce rom the center o rotation or rom the center o

moments. It is assumed that f distances on the right side o the center o rotation o a negative

slope (i.e., a right-acing slope) are negative and those on the lef side o the center o rotation

are positive. For positive slopes, the sign convention is reversed.

x = horizontal distance rom the centerline o each slice to the center o rotation or to the centero moments.

e = vertical distance rom the centroid o each slice to the center o rotation or to the center o

moments.

d = perpendicular distance rom a point load to the center o rotation or to the center o moments.

h = vertical distance rom the center o the base o each slice to the uppermost line in the geom-

etry (i.e., generally ground surace).

a = perpendicular distance rom the resultant external water orce to the center o rotation or to

the center o moments. Te L and R subscripts designate the lef and right sides o the slope,

respectively.

A = resultant o external water orces. Te L and R subscripts designate the lef and right sides o

the slope, respectively.ω = angle o the point load rom the horizontal. Tis angle is measured counterclockwise rom

the positive x -axis.

α = angle between the tangent to the center o the base o each slice and the horizontal. Te sign

convention is as ollows. When the angle slopes in the same direction as the overall slope o the

geometry, α is positive and vice versa.

Te slope saety actor or a specific sliding surace is given by the ollowing equation:

F R

S=s

m

(11.1)

where R is the available resistant orce along the sliding surace and Sm is the mobilized shear orce along

the sliding surace. Te assumption in limit equilibrium analysis is that Sm is mobilized in the same

proportion (relative to the available resistant orce) or each slice o the slide mass thus F s is the same or

each slice. Te saety actor can also be regarded as a ratio between capacity and demand. Te capacity

(R) is provided by the Mohr–Coulomb ailure criterion (i.e., R = N tanφ + cβ) and thus depends on the

material shear strength parameters (i.e., cohesion intercept, c, and internal riction angle, φ). On the

other hand, the demand (Sm) is obtained rom the equations o statics applied to each slice and thereore

represents the required shear orce to maintain the slide mass in limit equilibrium.

ypical limit equilibrium equations are as ollows:

• Equations o orce equilibrium in the vertical and horizontal directions or each slice can be usedto obtain the normal orce at the base o the slice (N) and the interslice normal orce (E).

• Te equation o equilibrium o orces in the horizontal direction or all slices can be used to derive

the orce equilibrium saety actor, F sF.

• Te equation o moment equilibrium about a common point or all slices can be used to derive the

moment equilibrium saety actor, F sM.

A detailed mathematical ormulation o various limit equilibrium equations can be ound in Krahn

(2004). Because the number o unknowns in these equations is larger than the number o equations

available, additional assumptions need to be introduced. Such assumptions are typically made in respect

to the magnitude and orientation o the interslice orces. Major differences among various limit equi-

librium methods are associated with the specific equilibrium equations used and the assumptions madewith respect to the interslice orces. able 11.2 outlines the equations used by various limit equilibrium

methods o slope stability analysis together with the assumptions related to interslice orces associated

with each method.

8/20/2019 b15621-12 (1)



Te Ordinary method o slices provides lower saety actors compared to more rigorous methods

(e.g., Morgenstern–Price method). Te shape o the sliding surace has a major effect on the calcu-

lated saety actors using Bishop and Janbu simplified methods. Computational results rom slope

stability analyses conducted or various geometries o the sliding surace (Krahn, 2003) indicate that

or circular sliding suraces Bishop’s simplified method provides accurate saety actors similar to

those obtained using Morgenstern–Price method, whereas Janbu’s simplified method underestimates

the slope saety actor. Conversely, Janbu’s simplified method agrees very well with Morgenstern–

Price method yielding accurate saety actors or planar sliding suraces, whereas Bishop’s sim-

plified method underestimates the slope saety actor or planar sliding surace geometries. TeMorgenstern–Price saety actor along composite sliding suraces involving combinations o planar

and circular segments is typically bounded by the saety actors associated with Bishop and Janbu

simplified methods.

Stability charts are useul in preliminary stages o a project or sensitivity analysis since they

enable to perorm a quick and simple slope stability assessment. hey are developed using dimen-

sionless relationships that can be established between the saety actor and other parameters

characterizing the slope geometry, soil shear strength, and pore water pressure. Most charts are

developed or homogeneous slopes with simple geometries. In case o nonhomogeneity o the soils

layers, average parameters should be evaluated. Several published stability charts developed by

various investigators can be ound in the literature (Fel lenius, 1936; aylor, 1937; Janbu, 1954; Hoek

and Bray, 1974).Commercial sofware is currently available to conduct limit equilibrium slope stability analyses using

personal computers. Tis tool has become critical or slope stability studies as it allows or the investi-

gation o complex problems involving multiple stratigraphic layers, various pore pressure conditions,

and reinorcement. It also provides the advantage o conducting quick sensitivity analyses addressing

the influence o various input parameters on slope stability. As outlined by Duncan and Wright (2005),

the computer programs or slope stability can be divided into analysis programs and design programs.

Te output provided by analysis programs is the slope saety actor corresponding to a prescribed set

o input parameters (e.g., slope geometry, material properties, pore pressures, external loads, rein-

orcement). On the other hand, design programs aim at determining the appropriate slope conditions

required to provide one or more design actors o saety specified by the user. Many o the computerprograms used to analyze reinorced slopes all in this latter category. able 11.3 (Duncan and Wright,

2005) presents perormance ratings o various computer programs or slope stability analysis used in

geotechnical engineering practice.

TABLE 11.2 Statics Satisfied and Interslice Forces in Various Methods

Method

Moment

Equilibrium

Horizontal Force

Equilibrium

Interslice

Normal (E)

Interslice

Shear (X) Inclination o X/E Resultant

Ordinary or Fellenius Yes No No No No orce

Bishop’s simplified Yes No Yes No HorizontalJanbu’s simplified No Yes Yes No Horizontal

Spencer Yes Yes Yes Yes Constant

Morgenstern—Price Yes Yes Yes Yes Variable

Corps o Engineers—1 No Yes Yes Yes Inclination o a line rom

crest to toe

Corps o Engineers—2 No Yes Yes Yes Slice top ground surace

inclination

Lowe–Karafiath No Yes Yes Yes Average o ground surace

slope and slice base

inclination

Source: Afer Krahn, J., Canadian Geotechnical Journal , 40, 643–660, 2003.

8/20/2019 b15621-12 (1)



11.4.3 Shear Strength Parameters for Slope Stability Analysis

Te limit equilibrium methods orming the ramework o slope stability/instability analysis gener-

ally accept the Mohr–Coulomb ailure criterion, which can be expressed in terms o effective or total

stresses. Te Mohr–Coulomb shear strength in terms o effective stresses is as ollows:

τ = ′ + ′σ ′ϕtan c (11.2a)

where τ and σ′ are the shear stress and effective normal stress, respectively, on the ailure surace, and

c′ and φ′ are the shear strength parameters (i.e., cohesion intercept and internal riction angle, respec-

tively) in terms o effective stresses. For a slope stability analysis in terms o total stresses, the shearstrength is given by the ollowing equation:

cτ = + σ ϕtan u u (11.2b)

TABLE 11.3 Characteristics o Various Computer Programs or Slope Stability Analysis

UEXAS4 SLOPE/W SLIDE XSABL WINSABL RSS SNAIL GoldNail

Accuracy 5 4.5 4.5 4 3.5 4 2 4

Program

Computationime

5 5 5 5 4 (1a) 4 2 4

ime or

Learning Curve

3 5 5 4 3.5 3.5 3 3.5

ime to Enter

Data and

Complete

Analysis

3 5 5 4.5 4 3.5 3 2.5

Ease o

Reinorced

Slope Design

1.5 2.5 2.5 5—Initial

only—no final

design

capabilities

3 5—Horizontal

reinorcement

only

4 5

Ease oUnreinorced

Slope Data

Entry

3.5 5 5 4 3.5 3 3 3.5

Ease o Soil Nail

Data Entry

2.5 3.5 3.5 No provision or

reinorcement

3.5 5—Horizontal

reinorcement

only

5 4.5

Ease o ieback

Data Entry

2.5 5 5 No provision or

reinorcement

4 5—Horizontal

reinorcement

only

3 3.5

Ease o Geogrid

Data Entry

2.5 3.5 3.5 No provision or

reinorcement

4.5 5 3 3.5

ime Requiredto Make

Output Report

Ready

4 5 5 3 2 3 3 1

Quality o

Graphical

Output

4 5 5 3 2 3 3 1

Note: 1 = Poor; 2 = Fair; 3 = Average; 4 = Good; 5 = Excellent.a In WINSABL, Spencer’s method has a computation time o up to several minutes.

8/20/2019 b15621-12 (1)



where σ represents the total normal stress on the ailure surace, and cu and φu are the shear strength

parameters (i.e., cohesion intercept and internal riction angle, respectively) in terms o total stresses. σʹ

and σ are related through the effective stress principle, that is,

′σ = σ − u (11.3)

where u represents the pore water pressure at the sliding surace. An illustration o the strength enve-

lopes in terms o total and effective stresses is provided in Figure 11.7.

A detailed discussion on the types o field and laboratory experiments required or the evaluation o

shear strength parameters in total and effective stresses can be ound elsewhere (Duncan, 1996; Duncan

and Wright, 2005).

Since soil shear strength is directly related to the effective normal stress along the sliding surace,

a slope stability analysis in terms o effective stresses based on Equation 11.2a is always desirable.

However, selection o Equation 11.2a or b in slope stability assessments depends largely on the degree to

which the pore water pressure along the sliding surace is known. I a slope is likely to ail under drainedconditions (i.e., no excess pore pressures develop until the onset o slope ailure), then the pore water

pressure can be estimated rom field measurements and seepage analyses. Under such circumstances,

the effective stress along the sliding surace can be determined rom Equation 11.3 and thereore an

effective stress slope stability analysis based on Equation 11.2a can be undertaken. However, a slope

may also experience ailure in undrained conditions when subjected to an increase in driving orces

(e.g., due to earthquake loading, excavation o material rom toe, placement o material at crest) at a

rate quick enough to not allow or dissipation o excess pore pressures during the loading process due to

the relatively low hydraulic conductivity o the slope materials. In such situations, a total stress analysis

based on Equation 11.2b is typically used to analyze the slope instability in undrained conditions since

the excess pore pressures generated along the sliding surace due to undrained loading/unloading are

not easy to predict and thereore the magnitude o effective stresses at ailure cannot be determined.

Detailed recommendations on how to select appropriate undrained shear strength parameters or total

stress slope stability analyses are provided by Duncan (1996) and Duncan and Wright (2005). Duncan

(1996) also provides a rational approach o assessing whether undrained conditions are likely to occur

under a specific loading scenario using a time actor–based methodology similar to one-dimensional

primary consolidation analysis.

Failure envelope for

effective stresses

Failure envelope for

total stresses

cu

uf

c

u

3f

1f

1f

, 3f

FIGURE 11.7 Failure envelopes or total and effective stresses.

8/20/2019 b15621-12 (1)



It is also worth noting that the stress–strain behavior o overconsolidated soils is characterized by

a peak strength reached at relatively small strains ollowed by a gradual reduction in strength with

progressive increase in strains, culminating in a residual strength value smaller than the peak strength.

Under such circumstances, the selection o appropriate strength parameters or slope stability analysis

must be made based on the amount o shear deormation experienced on the slope. Te stability o aslope likely to experience ailure or the first time can be analyzed using peak shear strength parameters,

whereas the stability o reactivated landslides (i.e., landslides exhibiting recurrent movements over cer-

tain periods o time) must be analyzed using residual shear strength parameters.

11.4.4 Backward Analysis of Slope Stability

Backward analysis is an effective approach to derive the design shear strength parameters or slope

stabilization. A slope ailure can reasonably be considered as a ull-scale shear test capable to give a

measure o the strength mobilized at ailure along the slip surace. Te back-calculated shear strength

parameters, which are intended to be closely matched with the observed real-lie perormance o the

slope, can then be used in urther limit equilibrium analyses to design remedial works. Shear strengthparameters obtained by back analysis ensure more reliability than those obtained by laboratory or in-

situ testing when used to design remedial measures.

Procedures to determine the magnitude o both shear strength parameters or the relationship between

them by considering the position o the actual slip surace within a slope are discussed by Popescu and

Yamagami (1994). Te two unknowns—that is, the shear strength parameters c′ and φ′—can be simul-

taneously determined rom the ollowing two requirements:

1. F s = 1 or the given ailure surace. Tat means the back-calculated strength parameters have to

satisy the c′ − tan φ′ limit equilibrium relationship.

2. F s = minimum or the given ailure surace and the slope under consideration. Tat means the

actors o saety or slip suraces slightly inside and slightly outside the actual slip surace shouldbe greater than one (Figure 11.8a).

Based on the abovementioned requirements, Saito (1980) developed a semigraphical procedure using

trial and error to determine unique values o c′ and tan φ′ by back analysis (Figure 11.8b). An envelope

o the limit equilibrium lines c′ − tan φ′, corresponding to different trial sliding suraces, is drawn, and

the unique values c′ and tan φ′ are ound as the coordinates o the contact point held in common by the

envelope and the limit equilibrium line corresponding to the actual ailure surace. A more systematic

procedure to find the very narrow range o back-calculated shear strength parameters based on the same

requirements is illustrated in Figure 11.8c.

Te undamental problem involved is always one o data quality, and consequently the back analysis

approach must be applied with care and the results interpreted with caution. Back analysis is o use onlyi the soil conditions at ailure are unaffected by the ailure. For example, back-calculated parameters

or a first-time slide in stiff overconsolidated clays could not be used to predict subsequent stability o

the sliding mass, since the shear strength parameters will have been reduced to their residual values by

the ailure. In such cases, an assumption o c′ = 0 and the use o a residual riction angleφ′r is warranted

(Bromhead, 1992). I the three-dimensional geometrical effects are important or the ailed slope under

consideration and a two-dimensional back analysis is perormed, the back-calculated shear strength

will be too high and thus unsae.

Additionally, one has to be aware o the many pitalls o the back analysis approach that involves a

number o basic assumptions regarding soil homogeneity, slope and slip surace geometry, and pore

pressure conditions along the ailure surace (e.g., Leroueil and avenas, 1981). A position o totalconfidence in all these assumptions is rarely i ever achieved. While the topographical profile can gen-

erally be determined with enough accuracy, the slip surace is almost always known in only ew points

and interpolations with a considerable degree o subjectivity are necessary. Errors in the position o

8/20/2019 b15621-12 (1)



the slip surace result in errors in back-calculated shear strength parameters. I the slip surace used

in back analysis is deeper than the actual one, c′ is overestimated and φ′ is underestimated and vice versa. Te data concerning the pore pressure on the slip surace are generally ew and imprecise.

More exactly, the pore pressure at ailure is almost always unknown. I the assumed pore pressures

are higher than the actual ones, the shear strength is overestimated. As a consequence, a conservative

assessment o the shear strength is obtainable only by underestimating the pore pressures.

o avoid the questionable problem o the representativeness o the back-calculated unique set o shear

strength parameters, a method or designing remedial works based on the limit equilibrium relation-

ship c′ − φ′ rather than a unique set o shear strength parameters can be used (Popescu, 1991).

Te method principle is shown in Figure 11.9. It is considered that a slope ailure provides a single

piece o inormation, which results in a linear limit equilibrium relationship between shear strength

parameters. Tat piece o inormation is that the actor o saety is equal to unity (F s = 1), or the horizon-tal orce at the slope toe is equal to zero (E = 0) or the conditions prevailing at ailure. Each o the two

conditions (F s = 1 or E = 0) results in the same relationship c′ − tan φ′, which or any practical purpose

might be considered linear.

Trial slidingsurfaces, T88

Envelope

tan'

Region II

Contact points

Failuresurfaces, F8

(a)

(b)

(c)

01

1

Region II

1.0

F

T S S 1 ’ T S S

1

1 0

Range of shear strengthparameters

F8

1c

c

tan'

FIGURE 11.8 Shear strength back analysis methods. (Afer Popescu, M.E., Schaeer V.R., In Proceedings of the

10th International Symposium on Landslides and Engineered Slopes, Xi ’an, China, pp. 1787–1793, 2008.)

8/20/2019 b15621-12 (1)



tan'

c'

c'

LoadUnload

WT 1WT 2

No pile

Pile interval ( B/ D)1

Pile interval ( B/ D)2<( B/ D)1

d d B D

PileRow

Initial geometry

Unloading active parts

Loading passive parts

Before drainage (WT 1)

After drainage (WT 2)

tan ' 0

tan'

tan'

tan'

tan' nec

c' nec

c' 0

c'

c'

Limit equilibrium relationshipfor the failed slope: F s=1 or E n=0

Limit equilibrium relationshipfor the stabilized slope

FIGURE 11.9 Limit equilibrium relationship and design o slope remedial measures. (Afer Popescu, M.E. and

Schaeer V.R. Proc. 10th Intern. Symp. on Landslides and Engineered Slopes, Xi’an, China, p. 1787–1793, 2008.)

8/20/2019 b15621-12 (1)



Te linear relationship c′ − tan φ′ can be obtained using standard computer sofware or slope stabil-

ity limit equilibrium analysis by manipulations o trial values o c′ and tan φ′ and calculating the cor-

responding actor o saety value. It is simple to show that in an analysis using arbitrary φ′ alone (c′ = 0)

to yield a nonunity actor o saety, F φ*, the intercept o the c′ − tan φ′ line (corresponding to F s = 1) on

the tanφ′ axis results as

′ϕ = ′ϕ

ϕ

tantan

0 *F (11.4)

Similarly, the intercept o the c′ − tan φ′ line (corresponding to F s = 1) on the c′ axis can be ound

assuming φ′ = 0 and an arbitrary c′ value, which yield to a nonunity actor o saety, F c*:

′ = ′

0

c*

cc

F (11.5)

Using the concept o limit equilibrium linear relationship c′ − tanφ′, the effect o any remedial measure(drainage, modification o slope geometry, restraining structures) can easily be evaluated by considering

the intercepts o the c′ − tan φ′ lines or the ailed slope c ϕ′ ′( , tan )0 0 and or the same slope afer installing

some remedial works (c′nec, tan φ′nec), respectively (Figure 11.9). Te saety actor o the stabilized slope is

= = ′′

= ′ϕ

′ϕ

ϕmin ,tan

tanc* 0

nec

* 0

nec

F F c

cF (11.6)

Errors included in back calculation o a given slope ailure will be offset by applying the same results,

in the orm o c′ − tan φ′ relationship, to the design o remedial measures.

Te above outlined procedure was used to design piles to stabilize landslides (Popescu, 1991) taking intoaccount both driving and resisting orce. Te principle o the proposed approach is illustrated in Figure 11.9,

which gives the driving and resisting orce acting on each pile in a row as a unction o the nondimensional

pile interval ratio B/D. Te driving orce, F D, is the total horizontal orce exerted by the sliding mass corre-

sponding to a prescribed increase in the saety actor along the given ailure surace. Te resisting orce, F R ,

is the lateral orce corresponding to soil yield, adjacent to piles, in the hatched area as shown in Figure 11.10.

F D increases with the pile interval, whereas F R decreases with the same interval. Te intersection point o

the two curves, which represent the two orces, gives the pile interval ratio satisying the equality between

driving and resisting orce. Te accurate estimation o the lateral orce on pile is an important param-

eter or the stability analysis because its effects on both the pile and slope stability are conflicting. Tat

is, sae assumptions or the stability o slope are unsae assumptions or the pile stability, and vice versa.

Consequently, to obtain an economic and sae design, it is necessary to avoid excessive saety actors.

11.4.5 Seismic Slope Stability Analysis

Slopes in earthquake-prone areas may experience ailure during a seismic event due to inertia orces

imparted by the earthquake to the slide mass and/or loss o shear strength in slope materials during

earthquake shaking. Depending on the ailure mechanism, Kramer (1996) divided seismic slope insta-

bilities into the ollowing two major categories: inertial instabilities and weakening instabilities. Slopes

composed o materials susceptible to shear strength loss during an earthquake (e.g., liquefiable soils)

all in the latter category. Slopes undergoing inertial instability during an earthquake are characterized

by relatively constant shear strength, and incremental downward deormations occur when earthquakeaccelerations exceed the yield acceleration o the slide mass. Since such slopes typically remain stable

at the end o the earthquake, the seismic analysis ocuses on the determination o earthquake-induced

permanent slope displacements. Slopes susceptible to shear strength reduction during earthquake may

8/20/2019 b15621-12 (1)



experience instability also afer the earthquake i the seismically induced shear strength loss along the

sliding surace is large enough to bring the available shear strength below the static driving shear stress

acting at the sliding surace. Te seismic analysis in such circumstances will ocus on determining the

reduction in shear strength during the earthquake, and a static postearthquake slope stability analysis

using conventional limit equilibrium methods and employing the reduced shear strengths determined

rom the previous seismic analysis needs to be undertaken. A step-by-step procedure to be ollowed in a

seismic slope stability analysis is provided by Duncan and Wright (2005).Seismic slope stability can be assessed using the pseudostatic approach. Tis approach uses tradi-

tional limit equilibrium techniques and involves the application o an additional static orce to replicate

the earthquake loading. Te additional orce is computed as the weight o the slide mass multiplied

by a seismic coefficient and may be regarded as the equivalent o a seismic inertia orce acting on the

slope. Pseudostatic slope stability analyses are typically conducted during the preliminary stages o a

seismic landslide hazard assessment to evaluate the susceptibility to earthquake-induced slope ailure

and decide whether more advanced investigations (e.g., extensive laboratory testing and seismic slope

stability assessments using more sophisticated methods) should be undertaken to better characterize

the seismic slope response. A detailed discussion on the pseudostatic approach including the selection

o seismic coefficients and allowable saety actors or seismic slope stability analysis can be ound else-where (e.g., Kramer, 1996; Abramson et al., 2001; Duncan and Wright, 2005).

Te dynamic displacement analysis o slopes during earthquake is typically carried out using the

Newmark sliding block methodology (Newmark, 1965). Tis method appears to provide a compromise

Pilerow

d B d D

Downhill

Uphill

F D F R

B/ D

F D , F R

Narrowly spaced piles

Largely spaced piles

F D – Driving force

F R – Resisting force

FIGURE 11.10 Driving versus resisting orce or stabilizing piles. (Afer Popescu, M.E. and Schaeer V.R. Proc.

10th Intern. Symp. on Landslides and Engineered Slopes, Xi'an, China, p. 1787–1793, 2008.)

8/20/2019 b15621-12 (1)



between the simple pseudostatic approach, which gives a actor o saety as the only indicator o seismic

slope stability, and the more sophisticated finite element method, which produces detailed results o

seismic perormance but requires quite complex constitutive models or simulating the relevant aspects

o soil behavior. Te Newmark model is basically a one-block translational or rotational mechanism

along a rigid plastic-sliding surace, activated when the ground-shaking acceleration exceeds a criticallevel. Tereore, this rigid block approach lacks the ability o modeling the seismic compliance o a soil

slope. However, despite this deficiency, the Newmark sliding block concept is still widely used in engi-

neering practice mainly due to the act that it requires only undamental design inormation (e.g., geom-

etry o the problem), a minimum number o material properties (i.e., unit weight and shear strength

parameters), and involves a robust computational process. Details o the conventional Newmark sliding

block method and the solution procedure can be easily ound in the literature (e.g., Newmark 1965;

Kramer 1996; Abramson et al., 2001). A Newmark sliding block methodology accounting or the degra-

dation o yield strength along the sliding surace with progressive landslide deormation was developed

and applied to seismic stability evaluations o slopes susceptible to earthquake-induced catastrophic

ailure in liquefiable soils (randafir and Sassa, 2004, 2005).

11.5 Landslide Risk Mitigation

11.5.1 Landslide Risk Treatment Options

Risk treatment is the final stage o the risk management process and provides the methodology or con-

trolling the risk. At the end o the evaluation procedure, it is up to the client or to policy makers to decide

whether to accept the risk or not, or to decide that more detailed study is required. Te landslide risk analyst

can provide background data or normally acceptable limits as guidance to the decision maker but should

not be making the decision. Part o the specialist’s advice may be to identiy the options and methods or

treating the risk. ypical options would include (Australian Geomechanics Society, 2000) the ollowing:

• Accept the risk: Tis will usually require the risk to be considered to be within the acceptable ortolerable range.

• Avoid the risk: Tis will entail avoiding the project, thus seeking an alternative site or orm o

development so that the revised risk becomes acceptable or tolerable.

• Reduce the likelihood: Tis requires stabilization measures to control the initiating circumstances,

such as reprofiling the surace geometry or installing groundwater drainage, anchors, stabilizing

structures, protective structures.

• Reduce the consequences: Tis requires provision o deensive stabilization measures, ameliora-

tion o the behavior o the hazard, or relocation o the development to a more avorable location

to achieve an acceptable or tolerable risk.

• Monitoring and warning systems: In some situations, monitoring (such as by regular site visits orby surveys) and establishment o warning systems may be used to manage the risk on an interim

or permanent basis. Monitoring and warning systems may be regarded as another means o

reducing the consequences.

• ransfer the risk: Tis requires that either another authority to accept the risk or to compensate

or the risk such as by insurance.

• Postpone the decision: I there is sufficient uncertainty, it may not be appropriate to make a deci-

sion on the data available. Further investigation or monitoring will be required to provide data

or better evaluation o the risk.

Te relative costs and benefits o various options need to be considered so that the most cost- effective

solutions, consistent with the overall needs o the client, owner, and regulator, can be identified.Combinations o options or alternatives may be appropriate, particularly where relatively large reduc-

tions in risk can be achieved or relatively small expenditures. Prioritization o alternative options is

likely to assist with selection.

8/20/2019 b15621-12 (1)



11.5.2 Landslide Remedial Measures

Correction o an existing landslide or the prevention o a pending landslide is a unction o reduction o

the driving orces or increase in the available resisting orces. Any remedial measure used must involve

one or both o the above parameters. Many general reviews o the methods o landslide remediation have

been made. Te interested reader is particularly directed to Cornorth (2005), Duncan and Wright (2005),Popescu and Seve (2001). ransportation Research Board (1996), Bromhead (1992), Zaruba and Mencl

(1982), and Hutchinson (1977).

IUGS WG/L (Popescu, 2001) has prepared a short checklist o landslide remedial measures arranged

in our practical groups, namely, modification o slope geometry, drainage, retaining structures, and

internal slope reinorcement, as shown in able 11.4.

A flow diagram (Figure 11.11) exhibits the sequence o various phases involved in the planning,

design, construction, and monitoring o remedial works (Kelly and Martin, 1986). Te ollowing gives a

short description o the most commonly used remedial measures.

TABLE 11.4 A Brie List o Landslide Remedial Measures

1. Modification o Slope Geometry

(1) Removing material rom the area driving the landslide (with possible substitution by lightweight fill)

(2) Adding material to the area maintaining stability (counterweight berm or fill)

(3) Reducing general slope angle

2. Drainage

(1) Surace drains to divert water rom flowing onto the slide area (collecting ditches and pipes)

(2) Shallow or deep trench drains filled with ree-draining geomaterials (coarse granular fills and geosynthetics)

(3) Buttress counterorts o coarse-grained materials (hydrological effect)

(4) Vertical (small diameter) boreholes with pumping or sel-draining

(5) Vertical (large diameter) wells with gravity draining

(6) Subhorizontal or subvertical boreholes

(7) Drainage tunnels, galleries, or adits

(8) Vacuum dewatering

(9) Drainage by siphoning

(10) Electroosmotic dewatering

(11) Vegetation planting (hydrological effect)

3. Retaining Structures

(1) Gravity retaining walls

(2) Crib-block walls

(3) Gabion walls

(4) Passive piles, piers, and caissons

(5) Cast-in-situ-reinorced concrete walls

(6) Reinorced earth-retaining structures with strip/sheet–polymer/metallic reinorcement elements (7) Buttress counterorts o coarse-grained material (mechanical effect)

(8) Retention nets or rock slope aces

(9) Rockall attenuation or stopping systems (rocktrap ditches, benches, ences, and walls)

(10) Protective rock/concrete blocks against erosion

4. Internal Slope Reinorcement

(1) Rock bolts

(2) Micropiles

(3) Soil nailing

(4) Anchors (prestressed or not)

(5) Grouting

(6) Stone or lime/cement columns

(7) Heat treatment (8) Freezing

(9) Electroosmotic anchors

(10) Vegetation planting (root strength mechanical effect)

8/20/2019 b15621-12 (1)



Planning phase

Preliminary design of

general project optionsDesk study

Finish

Monitor

YES

YES

NO NO

Geotechnical and geomorphological

study to locate extent, position,possible depth, and mechanism of

landslide

Conceptual design to improve

stability of landslide

Is there sufficient space to avoid

landslide and its stabilizing works?

Design and implement proving site

investigation

Continue periodic monitoring

Review preliminary design and

select optimum method of stabilizing

landslide

Maintenance phase

Construction phase

Is project

performing

satisfactorily?

Revise

design NO

YES

Design phase

Are the overall consequences of

avoiding it in terms of cost,

safety, etc., acceptable?

Select preferred site for project taking

account of cost, economics, safety,

and so on and likely stabilizing works

Install monitoring system

Construct

Design and implement detailed site

investigation on preferred site

Detailed design of project and the

required stabilizing works, i.e.,

earthworks, drainage, structural

works, etc.

Detailed design of short and

long-term monitoring systemSpecify and special measures

specific to construction

through landslide zone

FIGURE 11.11 Various phases involved in planning, design, and construction o landslide remedial works. (From

Kelly, J.M.H., and Martin, P.L., Construction Works on or Near Landslides, In Proceedings o the Symposium oLandslides in South Wales Coalfield, Polytechnic o Wales, 85–103, 1986. With permission.)

8/20/2019 b15621-12 (1)



11.5.2.1 Drainage Measures

Hutchinson (1977) has indicated that drainage is the principal measure used in the mitigation o

landslides, with modification o slope geometry the second most used method. Tese are also generally

the least costly o the our major categories, which is obviously why they are the most used. Te experi-

ence shows that while one remedial measure may be dominant, most landslide repairs involve use o acombination o two or more o the major categories. For example, while restraint may be the principal

measure used to correct a particular landslide, drainage and modification o slope geometry, to some

degree and by necessity, are also used.

Drainage is ofen a crucial remedial measure due to the important role played by pore water pressure

in reducing shear strength. Because o its high stabilization efficiency in relation to cost, drainage o

surace water and groundwater is the most widely used and generally the most successul stabilization

method. As a long-term solution, however, it suffers greatly because the drains must be maintained i

they are to continue to unction.

Drainage may be used to prevent surace or subsurace water reaching the slide area or to remove

it rom the slide area. Surace water is diverted rom unstable slopes by ditches and pipes. Drainage o

shallow groundwater is usually achieved by networks o trench drains. Drainage o the ailure suraces,

on the other hand, is achieved by counterort or deep drains, which are trenches sunk into the ground to

intersect the shear surace and extending below it. In the case o deep landslides, ofen the most effective

way o lowering groundwater is to drive drainage adits into the intact material beneath the landslide.

From this position, a series o upward-directed drainage holes can be drilled through the roo o the

tunnel to drain the sole o the landslide. Alternatively, the adits can connect a series o vertical wells

sunk down rom the ground surace. In instances where the groundwater is too deep to be reached by

ordinary trench drains and where the landslide is too small to justiy an expensive drainage adit or

gallery, bored subhorizontal drains can be used. Another approach is to use a combination o vertical

drainage wells linked to a system o subhorizontal borehole drains.

Figure 11.12 shows a selection o drainage measures applied to a landslide. Tis figure was compiledby Bromhead (1992) rom a number o case records, and all the drainage measures adopted in the figure

have been used successul ly, either singly or in combination, to stabilize landslides.

Subhorizontal drains may be ineffective in clays and other fine-grained soils. Tereore, the pos-

sibility o poor perormance should be considered when assessing the relative merits o subhorizontal

drains to other remedial measures. Figure 11.13 illustrates some o the more common situations where

subhorizontal drains can be used or slope stabilization (Cornorth, 2005). A case study o a large land-

slide on the southern Oregon coast, stabilized by a vertical shaf and horizontal drains, is discussed by

Cornorth (2005) and illustrated in Figure 11.14.

Modification o slope geometry as illustrated in Figure 11.15 is a most efficient method. Balancing the

volume o cut and fill makes it unnecessary to dispose o excavated material off-site or to import soil

or fill area. However, the success o corrective slope regrading (fill or cut) is determined not merely by

size or shape o the alteration but also by position on the slope. Hutchinson (1977) provided details o

the “neutral line” method to assist in finding the best location to place a stabilizing fill or cut. Tere are

some situations where this approach is not simple to adopt. Tese include long translational landslides

where there is no apparent toe or crest (Figure 11.16). Also, situations where the geometry is determined

by engineering constraints and where the unstable area is complex and thus a change in topography,

which improves the stability o one area, may reduce the stability o another area.

Schuster (1995) discussed recent advances in the commonly used drainage systems while briefly men-

tioning less commonly used, but innovative means o drainage, such as electroosmotic dewatering and

vacuum and siphon drains were also presented. In addition, buttress counterorts o coarse-grained

materials placed at the toes o unstable slopes ofen are successul as remedial measures (Figure 11.17).Tese methods are listed in able 11.4 under both “Drainage,” when used mainly or their hydrological

effect, and “Retaining Structures,” when used mainly or their mechanical effect.

8/20/2019 b15621-12 (1)



11.5.2.2 Structural Measures

Retaining structures include a variety o structural solutions starting with traditional concrete or

masonry gravity retaining walls and concrete cantilever retaining walls (Figure 11.18) as well as crib,

bin and gabion retaining walls (Figure 11.19). Mechanically stabilized earth walls (Figure 11.20) or

landslide applications can be built to support shallow slides. Tey can be also incorporated in buttress

remediation where available land is restricted and a steeper buttress slope is required or to help road-

widening projects.

Heavily reinorced concrete piles are also used as retaining structures to stabilize landslides (Figure

11.21). Spaced and staggered piles are more requently used than tangent or secant piles. Such stabilizing

piles are easy to construct and may be buried within the slide mass, making the remediation less intru-

sive than other techniques. Design principles or stabilizing a slope with piles are shown schematically

in Figure 11.22.

Piles, piers, buttress, or walls are passive stabilization systems; that is, urther movements o the slope

increase pressure on them, and the system reaction orces put into the slide mass lead to stabilization.

On the other hand, prestressed anchors are active stabilization systems; that is, they use preloading to

put the stabilizing orces into the landslide mass rom the beginning. Anchor loads are spread into the

slide mass by pads so that the bearing capacity ailures o the ground are avoided (Figure 11.23). Pads

with a small number o anchors are preerred. Figure 11.24 shows anchors used to stabilize a landslide

above ablachaca Dam in Peru (Duncan and Wright, 2005).

Micropiles that are essentially an outgrowth o the technology used in the construction o groundanchors are passive systems. Applications o micropiles or slope stabilization are schematically illus-

trated in Figure 11.25: Case 1 micropiles are directly loaded and resist the loads applied by the slide

Outfall details need careand manholes at breaksin slope if pipeworkis installed in trenchdrains

Sand drains connect to drainage blanket toeliminate construction pore pressures underfill. ey can, but do not need to be fully penetrating

Vertical sand drains operate by draining pore pressures intounderlying permeable strata

Bored drains lead into shafts or tunnelsor could be bored from ground levelat the toe of the slide

Perched water tableintercepted by tunnel

Trench drains highly effectivein shallow slides or slopewash

Overland flows intercepted by cut-off drain or extensions to main downslopetrench drains

Run-off captured by chevron (top) orherringbone pattern (lower) secondary drains

FIGURE 11.12 Various drainage measures applied to landslide stabilization. (Afer Bromhead, E.N., Slope

Stability. 2nd Edition, Blackie Academic & Proessional, London, 411, 1992.)

8/20/2019 b15621-12 (1)



mass, whereas Case 2 micropiles are an interlocking, three-dimensional network o reticulated piles and

are not as heavily reinorced as Case 1 micropiles.

An example o the use o both passive and active systems in the same landslide is shown in Figure 11.26.

Various methods o retaining rock slopes are illustrated in Figures 11.27 and 11.28. All the categories

o stabilization treatment or soil slopes have their analogies in rock slopes, but they have different sets

o priorities. Te most effective techniques or rock slope stabilization are those which increase thestrength o discontinuities in the rock mass—anchoring, bolting, and grouting. A comprehensive trea-

tise on the rock slope subject is given by Hoek and Bray (1974) and Fell (1994a).

Horizontal drain

Horizontal drain

Horizontal drains

Loweredgroundwater level

Originalgroundwater level

Groundwater level

Groundwater level

Mixed soils

Broken

basalt

Weakly cementedgravel

“Ponded” waterat head of slide

Claystone

Colluvium

Permeablesedimentary

strata

Claystone

Tuff beds

Slip surface

(a)

(b)

(c)

Slip surface

Ancient landslide debris(intermixed clay and gravel)

FIGURE 11.13 Examples o geological conditions in which subhorizontal drains may be an appropriate option.

(Afer Cornorth, D., Landslides in Practice, Investigation, Analysis, and Remedial/Preventative Options in Soils,

Wiley, 2005.)

8/20/2019 b15621-12 (1)



RealignedU.S. Highway 101

Dischargepipe

Dischargepipe

P a c i fi c

O

c e a n

Rock headland

Limits of landslide

Scale in feet

Note: All horizontal drain locations

higher up the slope are not shown

(a)

(b)

Vertical shaft(18-foot diameter)

Horizontal drains arrayedwithin the drainage shaft

(three levels)Discharge drainfrom shaft

U.S.Highway 101

Conventional horizontal draininstallations in upper slope

Slip surface

5000

Scale in feet

5000

Cut slope Mainheadscrap

Graben

NORTH

FIGURE 11.14 (a) Plane view and (b) cross section o a vertical shaf and horizontal drains applied to stabilize

the Arizona Inn Slide, near Brookings, OR. (Afer Cornorth, D., Landslides in Practice, Investigation, Analysis, and

Remedial/Preventative Options in Soils, Wiley, 2005.)

Fill area

Diverted

road

Diverted

river

Waste material

Finished profile

Former waste fill profile Cut areas

Bedrock

Water table

S u p e r fi c i a

l d e p o s i t s

0 50

meters

100

FIGURE 11.15 Slope stabilization by cut and fill. (From Duncan, J.M., and Wright, S.G., Soil Strength and Slope

Stability , Wiley, 2005. With permission.)

8/20/2019 b15621-12 (1)



11.5.2.3 Nonstructural MeasuresDuring the early part o the post–World War II period, landslides were generally seen to be “engineering

problems” requiring “engineering solutions” involving correction by the use o structural techniques.

Tis structural approach initially ocused on retaining walls but has subsequently been diversified to

include a wide range o more sophisticated techniques including passive piles and piers, cast-in-situ-

reinorced concrete walls, and reinorced earth-retaining structures. When properly designed and

constructed, these structural solutions can be extremely valuable, especially in areas with high loss

potential or in restricted sites. However, fixation with structural solutions has in some cases resulted

in the adoption o overly expensive measures that have proven to be less appropriate than alternative

approaches involving slope geometry modification or drainage.

Over the last several decades, there has been a notable shif toward “sof engineering,” nonstruc-tural solutions, including classical methods such as drainage and modification o slope geometry, but

also some novel methods such as lime/cement stabilization, grouting, or soil nailing. Te cost o non-

structural remedial measures is considerably lower than the cost o structural solutions. In addition,

Buttress at base of slope is ineffective

Existinghighway

Flood plainsediments

Active shear Ancient slipsurface

Bedrock

Active ancientslip surface

Landslidedebris

FIGURE 11.16 Ineffective location o a buttress where the slip surace passes deep below the slope base.


Wiley, 2005.)

590

New grade

Original slope

Random fillButtress fill

Drain pipes Weathered basalt

Rupture surfacein landslide

Natural soil: clayey and sandy silt

580

570

E l e

v a t i o n

( f t )

560

550

FIGURE 11.17 Buttress or slope stabilization in Portland, OR. (From Duncan, J.M., and Wright, S.G., Soil

Strength and Slope Stability , Wiley, 2005. With permission.)

8/20/2019 b15621-12 (1)



structural solutions, such as retaining walls, involve exposing the slope during construction and ofen

require steep temporary excavations. Both o these operations increase the risk o ailure during con-

struction or oversteepening or increased infiltration rom rainall. In contrast, the use o soil nailingas a nonstructural solution to strengthen the slope avoids the need to open or alter the slope rom its

current condition (Figure 11.29).

Environmental considerations have increasingly become an important actor in the choice o suit-

able remedial measures, particularly issues such as visual intrusion in scenic areas or the impact on

nature or geological conservation interests. An example o a “sof engineering” solution, more com-

patible with the environment, is the stabilization o slopes by the combined use o vegetation and

man-made structural elements working together in an integrated manner known as biotechnical slope

stabilization (Schuster, 1995). Te basic concepts o vegetative stabilization are not new—vegetation

has a beneficial effect on slope stability by the processes o interception o rainall, and transpiration

o groundwater, thus maintaining drier soils and enabling some reduction in potential peak ground-water pressures. In addition to these hydrological effects, vegetation roots reinorce the soil, increasing

soil shear strength, while tree roots may anchor into firm strata, providing support to the upslope soil

mantle and buttressing and arching. A small increase in soil cohesion induced by the roots has a major

Granular backfillwith toe drain

Masonry orconcrete wall

Landslide debris

Slip surface

Slip surface

Firm base

(a)

Drain

Shear key (optional)

(b)

H

H

0.5 to 0.7 H

0.4 to 0.67 H

8 to 12"

Temporary construction slope(typical 1:1)

Landslide debris

Counterfort (toprovide structural

stiffness on higherwalls)

Minimum2.5 feetor tofrost level

AB

FIGURE 11.18 (a) Concrete or masonry gravity retaining wall and (b) concrete cantilever retaining wall.


Wiley, 2005.)

8/20/2019 b15621-12 (1)



effect on shallow landslides. Te mechanical effect o vegetation is not significant or deeper-seated

landslides, whereas the hydrological effect is beneficial or both shallow and deep landslides. However,

vegetat ion may not always assist slope stabi lity. Destabilizing orces may be generated by the weight

o the vegetation acting as a surcharge and by wind orces acting on the exposed vegetation, although

both o these are very minor effects. Roots o vegetation may also act adversely by penetrating and

dilating the joints o widely jointed rocks. For detailed inormation on research into the engineer-

ing role o vegetation or slope stabilization, reer to Greenway (1987). In addition, the “Geotechnical

Manual or Slopes” (Geotechnical Control Office, 1981) includes an excellent table noting the hydro-

logical and mechanical effects o vegetation.Te concept o biotechnical slope stabilization is generally cost-effective as compared to the use o

structural elements alone; it increases environmental compatibility and allows the use o local natural

materials. Interstices o the retaining structure are planted with vegetation whose roots bind together the

Temporary cut slope

(a)

(b)

(c)

Granular soil backfill

Prefabricatedconcrete or

timber elements

Firm foundationrequired

Slump block

Slip surface

Slip surface

Slip surface

Geotextile separator

Temporary cut slope


Prefabricatedconcrete ortimber elements



Gabion baskets

Select stone


Landslide debris

Slump block

FIGURE 11.19 (a) Crib, (b) bin, and (c) gabion retaining walls. (Afer Cornorth, D., Landslides in Practice,

Investigation, Analysis, and Remedial/Preventative Options in Soils, Wiley, 2005.)

8/20/2019 b15621-12 (1)



soil within and behind the structure. Te stability o all types o retaining structures with open gridwork

or tiered acings benefits rom such vegetation. Figure 11.30 shows an example where plants are installed

on the level tiers. iers offer improvements in slope appearance as compared to linear slopes and are help-

ul to construction o mechanically stabilized earth walls, soil nail walls, and anchor block walls.

11.5.3 Levels of Effectiveness and Acceptability That MayBe Applied in the Use of Remedial Measures

erzaghi (1950) stated that, “i a slope has started to move, the means or stopping movement must be

adapted to the processes which started the slide.” For example, i erosion is a causal process o the sl ide,

an efficient remediation technique would involve armoring the slope against erosion or removing the

source o erosion. An erosive spring can be made nonerosive by either blanketing with filter materials or

drying up the spring with horizontal drains, and so on.

Te greatest benefit in understanding landslide-producing processes and mechanisms lies in the use

o the above understanding to anticipate and devise measures to minimize and prevent major landslides.

S

(a) (b)

(c) (d) D

FIGURE 11.21 Reinorced concrete pile arrangements: (a) tangent, (b) secant, (c) staggered, (d) spaced pile walls.


Wiley, 2005.)

Finshed grade

Reinforcedbackfill

Facing

Leveling pad

Foundation

Geosynthetic reinforcementtypically 1 to 3 feet verticalseparation

Excavation limit

Original groundsurface

Retained

backfill

FIGURE 11.20 Mechanically stabilized earth wall. (Afer Cornorth, D., Landslides in Practice, Investigation,

Analysis, and Remedial/Preventative Options in Soils, Wiley, 2005.)

8/20/2019 b15621-12 (1)



Te term major should be underscored here because it is neither possible nor easible, nor even desirable,

to prevent all landslides. Tere are many examples o landslides that can be handled more effectively and

at less cost afer they occur. Landslide avoidance through selective locationing is obviously desired—

even required—in many cases, but the dwindling number o sae and desirable construction sites may

orce more and more the use o landslide-susceptible terrain.

Selection o an appropriate remedial measure depends on (1) engineering easibility, (2) economic

easibility, (3) legal/regulatory conormity, (4) social acceptability, and (5) environmental acceptability.

A brie description o each method is presented herein:

1. Engineering easibility involves analysis o geologic and hydrologic conditions at the site to ensurethe physical effectiveness o the remedial measure. An ofen-overlooked aspect is being certain

that the design will not merely divert the problem elsewhere.

2. Economic easibility takes into account the cost o the remedial action as composed to the benefits

it provides. Tese benefits include deerred maintenance, avoidance o damage (including loss o

lie), and other tangible and intangible benefits.

3. Legal–regulatory conormity provides or the remedial measure meeting local building codes,

avoiding liability to other property owners, and related actors.

4. Social acceptability is the degree to which the remedial measure is acceptable to the community

and neighbors. Some measures or a property owner may prevent urther damage but be an unat-

tractive eyesore to neighbors.

5. Environmental acceptability addresses the need or the remedial measure to not adversely affect

the environment. Dewatering a slope to the extent that it no longer supports a unique plant com-

munity may not be an environmentally acceptable solution.

Design principles

Critical slip

surface (with P )

e moving soil imposes a force P on the portionof each pile above the slip surface, at a distanceY above the slip surface. e distance Y isdetermined by the distribution of unit resistance.

p=unit resistance

P = resultante distribution of pis determined by theestimated relativemovement and p- yresistance

Portion of pile below slip surface is subjectedto shear load P and moment load M = PY

Portion of pile below slip surface

P

(a) (b)

(c)

(d)

Y

Soil above slip surface force P on pile,at distance Y above slip surface

Portion of pile aboveslip surface P

P

Y

M = PY

M P

FIGURE 11.22 (a) Design principles or stabilizing a slope with piles; (b) unit resistance p and resultant P pile;

(c) portion o pile above slip surace; (d) portion o pile below slip surace. (Afer Duncan, J.M., and Wright, S.G.,

Soil Strength and Slope Stability , Wiley, 2005.)

8/20/2019 b15621-12 (1)



Spray-on insulation

Spray-oninsulation

Shotcrete

3 -ick shotcrete, wire mesh

reinforced, facing colored to matchexposed rock and rough-screenfinished

6 -ick treatedtimber lagging

Final backfill tooriginal slope

Separationgeotextile

Drain rockblanket toslope edge

Lowerbackfill

Buried anchor block

(a)

(b)

(c)

6 -icktimber lagging

Timber lagging

Flowable fill behindbearing block

Anchor block

Anchor block

Anchor

Anchor block

Anchor hole

Anchor

Geocompositedrain board

13.12

13.12

6.56

6.56

11.48 6.56

6.56

11.48

8.2

8.2

16

18

FIGURE 11.23 Richardson Highway Slide remediation. Lower anchor block wall: (a) section, (b) elevation,

and (c) plan showing outer protection. (Afer Cornorth, D., Landslides in Practice, Investigation, Analysis, and

Remedial/Preventative Options in Soils, Wiley, 2005.)

2950

2900

2850

2800

E l e v a t i o n (

m )

2750

2700

2650

2600150 200 250 300 350 400

Distance (m)

Anchors

Estimatedbase of

landslide

Ancient

slidingsurfaces

Tunnel S-250

Tunnel S-200 E

Buttress

Active slidingsurfaces (Typ)

Ground surface

EI 2695 m

Sediments

450 500 550 600 650

FIGURE 11.24 Landslide repair o ablachaca Dam. (From Duncan, J.M., and Wright, S.G., Soil Strength and

Slope Stability , Wiley, 2005. With permission.)

8/20/2019 b15621-12 (1)



Just as there are a number o available remedial measures, so are there a number o levels o effec-

tiveness and levels o acceptability that may be applied in the use o these measures. We may have a

landslide, or example, that we choose to live with. Although this type o landslide poses no significant

hazard to the public, it will require periodic maintenance through removal due to occasional encroach-

ment onto the shoulder o a roadway. Te permanent closure o the Manchester–Sheffield road at Mam

or in 1979 (Skempton et al., 1989) and the decision not to reopen the railway link to Killin ollowing the

Glen Ogle rockslide in the United Kingdom (Smith, 1984) are well-known examples o abandonment

due to the effects o landslides in which repair was considered uneconomical.Most landslides, however, usually must be dealt with sooner or later. How they are handled depends

on the processes that prepared and precipitated the movement, the landslide type, the kinds o materi-

als involved, the size and location o the landslide, the place or components affected by or the situation

Concrete cap

Soil

Bedrock Micropiles Micropiles

Slip surface Final grade

Reinforcedsoil mass

Wall facing

Original grade

Concrete cap Surcharge loading

Case 1 micropiles

(a)

Case 2 micropiles

(b)

Road

FIGURE 11.25 Slope stabilization with micropiles: (a) Case 1—slip surace reinorcement; (b) Case 2—reticulatedpile soil mass reinorcement. (Afer Cornorth, D., Landslides in Practice, Investigation, Analysis, and Remedial/

Preventative Options in Soils, Wiley, 2005.)

Surface crack

Original ground(reinforced earth wall)

38° natural slope

Tied-back shear pile wall(4-foot dia. reinforced concretepiles at 7-foot 6-inch spacings)

Overburden

Approximatebedrock surface

G r o u

n d a n c

h o r

( b o n d e d l e

n g t h

2 5 f e e t )

U n b o n

d e d l e n

g t h

4 5 f e e t

Estimated positionof slip surface

Embedmentinto bedrock

14 feet atpile center

15° on

alternateanchors

25°

Slip surface measuredby inclinometer

0 10

Scale in feet

20 30

FIGURE 11.26 ied-back anchor shear piles (Goat Lick Slide, Essex, Montana). (Afer Cornorth, D., Landslides

in Practice, Investigation, Analysis, and Remedial/Preventative Options in Soils , Wiley, 2005.)

8/20/2019 b15621-12 (1)



created as a result o the landslide, available resources, and so on. Te technical solution must be in

harmony with the natural system, otherwise the remedial work will be either short-lived or excessively

expensive. In act, landslides are so varied in type and size, and in most instances, so dependent upon

special local circumstances that or a given landslide problem, there is more than one method o pre-

vention or correction that can be successully applied. Te success o each measure depends, to a large

extent, on the degree to which the specific soil and groundwater conditions are prudently recognized in

an investigation and incorporated in design.

As many o the geological eatures, such as sheared discontinuities are not known in advance, it ismore advantageous to plan and install remedial measures on a “design-as-you-go basis.” Tat is, the

design has to be flexible enough to accommodate changes during or subsequent to the construction o

remedial works.

Overburden slope flattened

Sealing of smallloose material

Dowel

Rock anchor to prevent slidingalong bedding or clay seam

Sealing of loose blocks

‘Dental concrete’ or masonry

Weephole

Warning sign

Bolt

Catch fence or wall

Rocktrap ditch with part gravel infill

Structural facing dowelled to base

Graded filter or sandbags

Argillaceous stratumor shear zone

FIGURE 11.27 Rock slope stabilization methods (Afer Bromhead, E.N., Slope Stability. 2nd Edition, Blackie

Academic & Proessional, London, 411, 1992.)

8/20/2019 b15621-12 (1)



11.6 Landslide Monitoring and Warning Systems

11.6.1 Landslide Monitoring

Monitoring o landslides plays an increasingly important role in the context o living and coping with

these natural hazards. Te classical methods o land surveys, inclinometers, extensometers, and piezom-

eters are still the most appropriate monitoring measures. In the uture, the emerging techniques based

on remote sensing and remote access techniques will undoubtedly be o main interest.

Te Department o Environment (1994) has identified the ollowing categories o monitoring,

designed or slightly differing purposes but generally involving similar techniques:

1. Preliminary monitoring involves provision o data on preexisting landslides so that the dangerscan be assessed and remedial measures can be properly designed or the site be abandoned.

2. Precautionary monitoring is carried out during construction to ensure saety and to acilitate

redesign, i necessary.

Hanging nets or chains toslow blocks tumbling from above

Free-hanging meshnet suspended from above

Bench as arock fall collector

Fence

Mesh secured by

bolts and gunitedto protectfriable formation

Rocktrap ditch and control fence

Stay

Stay

FIGURE 11.28 Rockall stabilization methods. (Afer Bromhead, E.N., Slope Stability. 2nd Edition, Blackie

Academic & Proessional, London, 411, 1992.)

Failure surface

Failure surface

Failure surface

(c)(b)(a)

FIGURE 11.29 Potential ailure suraces that need to be studied in soil nail design: (a) external ailure, (b) internal

ailure, and (c) mixed ailure. (Afer Cornorth, D., Landslides in Practice, Investigation, Analysis, and Remedial/

Preventative Options in Soils, Wiley, 2005.)

8/20/2019 b15621-12 (1)



3. Postconstruction monitoring is considered to check on the perormance o stabilization measures

and to ocus attention on problems that require remedial measures.

Observational methods based on careul monitoring—beore, during, and afer construction—are

essential in achieving reliable and cost-effective remedial measures.

11.6.2 Landslide Warning Systems

When dealing with a slope o precarious stability and/or presenting a risk that is considered too high, a pos-

sible option is to do nothing in regard to mitigation, but to install a warning system to insure or improve the

saety o people. It is worth noting that warning systems do not modiy the hazard but contribute to reduc-ing the consequences o the landslide and thus the risk, in particular the risk associated to the loss o lie.

Various types o warning systems have been proposed, and the selection o an appropriate one should

take into account the stage o landslide activity:

1. At preailure stage, the warning system can be applied either to revealing actors or to triggering

or aggravating actors. Revealing actors can be, or example, the opening o fissures or the move-

ment o given points on the slope; in such cases, the warning criterion will be the magnitude or

rate o movement. When the warning system is associated with triggering or aggravating actors,

there is a need to first define the relation between the magnitude o actors controlling the stabil-

ity condition or the rate o movement o the slope. Te warning criterion can be a given hourly

rainal l or the cumulative rainal l during a certain period o time, increased pore water pressure,a given stage o erosion, a minimum negative pore pressure in a loess deposit, and so on.

2. At ailure stage, the warning system can only be linked to revealing actors, generally a sudden

acceleration o movements or the disappearance o a target.

FIGURE 11.30 iered retaining structure o mechanically stabil ized earth with landscaped benches. (Afer

Cornorth, D., Landslides in Practice, Investigation, Analysis, and Remedial/Preventative Options in Soils,

Wiley, 2005.)

8/20/2019 b15621-12 (1)



3. At postailure stage, the warning system has to be associated to the expected consequences o the

movement. It is generally associated with the rate o movement and runout distance.

Leroueil (1996) defined the ollowing our possible different stages o landslide activity:

1. Preailure stage when the soil mass is still continuous. Tis stage is mostly controlled by progres-sive ailure and creep.

2. Onset o ailure characterized by the ormation o a continuous shear surace through the entire

soil or rock mass.

3. Postailure stage, which includes movement o the soil or rock mass involved in the landslide rom

just afer ailure until it essentially stops.

4. Reactivation stage when the soil or rock mass slides along one or several preexisting shear sur-

aces. Tis reactivation can be occasional or continuous with seasonal variations o the rate o

movement.

Te majority o remedial measures, outlined above, can be cost-prohibitive and may be social ly and

politically unpopular. As a result, there may be a temptation to adopt and rely instead upon the instal-lation o apparently cheaper and much less disruptive monitoring and warning systems to “save” the

population rom uture catastrophes. However, or such an approach to be successul, it is necessary to

ulfill satisactorily each o the ollowing steps (Hutchinson, 2001):

1. Te monitoring system shall be designed to record the relevant parameters, to be in the right

places, and to be sound in principle and effective in operation.

2. Te monitoring results need to be assessed continuously by suitable experts.

3. A viable decision shall be made, with a minimum o delay, that the danger point has been reached.

4. Te decision should be passed promptly to the relevant authorities, with a sufficient degree o

confidence and accuracy regarding the orecast place and time o ailure or those authorities to

be able to act without ear o raising a alse alarm. 5. Once the authorities decide to accept the technical advice, they must pass the warning onto the

public in a way that will not cause panic and possibly exacerbate the situation.

6. Te public needs to be well-inormed and prepared in advance to respond in an orderly and prear-

ranged manner.

In view o the preceding discussion, it is not surprising that, although there have been a ew successes

with monitoring and warning systems, particularly in relatively simple, site-specific situations, there

have been many cases where these have ailed, because one or more o requirements (1) through (6) above

have been violated, ofen with tragic and extensive loss o lie. It is concluded, thereore, that sustained

good management o an area, as outlined above, should be our primary response to the threat o land-

slide hazards and risks, with monitoring and warning systems being in a secondary, supporting role.

11.6.3 Forecasting the Time of Landslides

Landslides are very complex phenomena and are difficult to predict. Tey involve materials ranging over

many orders o magnitudes in size, rom fine-grained particles to masses o earth/rock o several cubic

kilometers. Te velocity o mass movements also varies over a wide range, rom creeping movements

o millimeter per year to extremely rapid avalanches that travel at several hundred kilometers per hour

(Cruden and Varnes, 1996). Moreover, they span the geologic–hydrologic interace rom completely dry

materials to viscous fluid type flows. As a result, orecasting the time o landslides remains a crucial and

still an unresolved problem.Landslide prediction can be classified as long term, intermediate term, or short term (Hamilton, 1997).

Long-term prediction o landslides is typically attained via landslide hazard maps, which are actually

susceptibility maps, or large areas. As mentioned previously, these maps contribute to assessments o

8/20/2019 b15621-12 (1)



long-term characteristics and warning o landslide hazards; hence, they provide a ramework or iden-

tiying the need or additional data, and effective mitigation techniques, along with zoning or land-use

planning (United Nations, 1996).

Landslide monitoring is considered to provide the necessary data that can be used or intermediate-

term prediction. Appearance o cracks, fluctuation o moisture in soils, and acceleration o surace orsubsurace movements provide precursory evidence o landslide movement. Specifically, the accelera-

tion o surace or subsurace movements enables the most direct detection o impending landsliding

(Voight and Kennedy, 1979).

Monitoring, described above, entails compilation o meteorological, hydrological, topographical, and

geophysical data. Te advent o automatic sampling, recording, and transmitting devices has enabled

practical prediction o landslide movements (Hamilton, 1997). Although prediction o landslide move-

ment, based on interaction between climate and slope movement, is a daunting task at this time, it may

become more viable in the uture due to ongoing research and monitoring o regional weather patterns.

Among approaches to the mitigation o landslide risk, the prediction o the time o occurrence or

a first-time landslide deserves special consideration (Saito, 1965). Te task is ar rom being simple

because the undamental physics controlling the nature and shape o the creep curve o geomaterialshas not been ully elucidated yet. Moreover, all the relevant parameters and boundary conditions are

not clearly defined, and it is impossible to orecast the triggering actors originating outside the sliding

mass (e.g., heavy rainall). An important key to the prediction o landslide ailure time should be the

stress–strain–time relations, but the heterogeneity o the geological conditions, groundwater seepage

conditions, associated pore water pressures on the potential sliding surace, and scale effects make the

laboratory evaluation o the geomechanical parameters barely adequate or the simulation o the tem-

poral evolution o a potential slide using numerical models.

Several methods have been proposed or the prediction concerning the time o occurrence o land-

slides. In engineering practice, such methods, that iner the time to ailure by means o monitored

surace displacements, are preerred or a prediction, given that they remove all uncertainties involvedin these problems. One o the first, most spectacular and well-documented predictions o slope ailure,

based upon displacement monitoring, was carried out at the Chuquicamata mine in Chile (Kennedy

and Niermeyer, 1970): the date o ailure was exactly predicted by means o a rough extrapolation o

displacement data. Hoek and Bray (1977) pointed out that the circumstance is not o great importance;

in act, rom the point o view o an engineer, even a prediction with an error o ew weeks is reasonable

and helps in making decisions. As a consequence, one may state that the key to the prediction is the

correct choice and a good monitoring o the relevant physical actors, rather than the principle selected

or inerring the time to ailure.

Regardless o the technique used or extrapolating the time to ailure, the quality o the prediction

depends on the quality o the data, so that a clear identification o the critical points or variables selected

or monitoring is strongly required to get a consistent prediction. Tis entails the need or developing oan understanding o preailure deormations and other precursory signs o different landslides mecha-

nisms. Accordingly, the help offered by slope monitoring methods, particularly global positioning sys-

tem and time domain reflectometry, can be noticeable. For some methods, the requency o observation

seems to condition the effectiveness o the prediction, as well as the extent o the time span o data col-

lection (i.e., the monitoring system should be installed as soon as possible). Te observation needs also to

be extended to other parameters, different than displacements, such as pore pressure or crack aperture.

11.7 Concluding Remarks

Assessing the landslide hazard is the most important step in landslide risk management. Once thathas been done, it is easible to assess the number, size, and vulnerability o the fixed elements at risk

(structures, roads, railways, pipelines, etc.), and thence the damage they will suffer. Te various risks

have to be combined to arrive at a total risk in financial terms. Comparison o this with, or instance,

8/20/2019 b15621-12 (1)



cost–benefit studies o the cost o relocation o acilities, or mitigation o the hazard by countermea-

sures, provides a useul tool or management and decision making.

Sites where there is undue risk rom landslides to communities and inrastructure should be identi-

fied and ranked using well-established methods o landslide hazard and landslide risk analysis and then

to mitigate these risks appropriately and effectively. Te necessary actions should be taken as soon aspossible, while there is yet time.

It should be emphasized that these include not only various direct measures, such as relocation o

inrastructure or slide stabilization, but also “good housekeeping” o the region as a whole, as or exam-

ple, sustained, ecologically sensitive management o land use, sound planning, obtaining inormation,

making emergency arrangements, and so on. In Hong Kong, such approaches have had dramatic suc-

cess, reducing the average rate o landslide atalities per year per person to 5 × 10−7, a tenth o what it was

beore the introduction o a slope–saety regime (through what is now the Geotechnical Engineering

Office) in late 1972 (Powel, 1992).

A pragmatic approach o living with landslides and reducing the impact o landslide problems in

urban areas is well illustrated by the strategy adopted to cope with landslide problems at Ventnor, Isle

o Wight, United Kingdom (Lee et al., 1991). Ventnor is an unusual situation in that the entire town lieswithin an ancient landslide complex. Te spatial extent and scale o the problems at Ventnor has indi-

cated that total avoidance or abandonment o the site are out o question, and large-scale conspicuous

engineering structures would be unacceptable in a town dependent on tourism. Instead, coordinated

measures have been adopted to limit the impacts o human activity that promote ground instability by

planning control, control o construction activity, preventing water leakage, and improving building

standards. In addition, good maintenance practice by individual homeowners proved to be a significant

help, because neglect could have resulted in localized instability problems.

Much progress has been made in developing techniques to minimize the impact o landslides,

although new, more efficient, quicker, and cheaper methods could well emerge in the uture. Tere are a

number o levels o effectiveness and levels o acceptability that may be applied in the use o these mea-sures, or, while one slide may require an immediate and absolute long-term correction, another may

only require minimal control or a short period.

Whatever the measure chosen, and whatever the level o effectiveness required, the geotechnical engi-

neer and engineering geologist have to combine their talents and energies to solve the problem. Solving

landslide-related problems is changing rom what has been predominantly an art to what may be termed

an art-science. Te continual collaboration and sharing o experience by engineers and geologists will

no doubt move the field as a whole closer toward the science end o the art–science spectrum than it is

at present.

References

Abramson, L.W., Lee, .S., Sharma, S., and Boyce, G.M. 2001. Slope Stability and Stabilization Methods.

2nd Edition, John Wiley and Sons, New York, NY.

AGS. 2000. Landslide Risk Management Concepts and Guidelines, Sub-Committee on Landslide Risk

Management, Australian Geomechanics Society, St Ives, Australia, pp. 49–92.

Bromhead, E.N. 1992. Slope Stability. 2nd Edition, Blackie Academic & Proessional, London, 411 pp.

Chowdhury, R., Flentje, P., and Ko Ko, C. 2001. “A Focus on Hilly Areas Subject to the Occurrence and

Effects o Landslides,” Global Blueprint for Change, 1st Edition—Prepared in conjunction with the

International Workshop on Disaster Reduction, August 19–22, 2001.

Cornorth, D. 2005. Landslides in Practice, Investigation, Analysis, and Remedial/Preventative Options in

Soils, Wiley, Hoboken, New Jersey, NJ.Crozier, M.J. 1986. Landslides—Causes, Consequences and Environment , Croom Helm, London.

Cruden, D.M. 1991. “A Simple Definition o a Landslide,” Bulletin International Association of Engineering

Geology , 43, 27–29.

8/20/2019 b15621-12 (1)



Cruden, D.M. 1997. “Estimating the Risks rom Landslides Using Historical Data,” In Landslide

Risk Assessment , Cruden, D.M., and Fell, R. (eds.), Balkema, Rotterdam, Te Netherlands, pp.

277–284.

Cruden, D.M., and Varnes, D.J. 1996. “Landslide ypes and Processes,” In Landslides Investigation and

Mitigation, urner, A.K., and Schuster, R.L. (eds.), ransportation Research Board Special Report247, National Research Council, Washington, DC.

Department o Environment. 1994. Landsliding in Great Britain, Jones, D.K.C., and Lee, E.M. (eds).,

HMSO, London.

Duncan, J.M. 1996. “Soil Slope Stability Analysis,” Chapter 13 In Landslides—Investigation and Mitigation,

urner, A.K., and Schuster, R.L., (eds.), ransportation Research Board Special Report 247, National

Research Council, National Academy Press, Washington, DC, pp. 337–371.

Duncan, J.M., and Wright, S.G. 2005. Soil Strength and Slope Stability , Wiley, Hoboken, New Jersey, NJ.

Einstein, H.H. 1988. “Landslide Risk Assessment Procedure,” In Proceedings o the 5th International

Symposium on Landslides, Lausanne, Switzerland, Balkema, Rotterdam, Te Netherlands,

2, 1075–1090.

Einstein, H.H. 1997. “Landslide Risk—Systematic Approaches to Assessment and Management,” In LandslideRisk Assessment , Cruden, D.M., and Fell, R. (eds.), Balkema, Rotterdam, Te Netherlands, pp. 25–50.

Fell, R. 1994a. “Stabilization o Soil and Rock Slopes,” In Proceedings o East Asia Symposium and Field

Workshop on Landslides and Debris Flows, Seoul, 1, 7–74.

Fell, R. 1994b. “Landslide Risk Assessment and Acceptable Risk,” Canadian Geotechnical Journal ,

31(2), 261–272.

Fellenius, W. 1936. “Calculation o the Stability o Earth Dams,” ransactions o the 2nd Congress on Large

Dams, International Commission on Large Dams o the World Power Conerence, Washington,

D.C, 4, 445–462.

Geotechnical Control Office. 1981. Geotechnical Manual for Slopes, Public Works Department, Hong Kong.

Greenway, D. R. 1987. Vegetation and slope stability, In Slope Stability Geotechnical Engineering andGeomorphology, Anderson, M.G., and Richards, K.S., (eds.), John Wiley & Sons, Chichester, UK,

pp. 187–230.

Hamilton, R. 1997. Report on Early Warning Capabilities for Geological Hazards, International Decade

or Natural Disaster Reduction (IDNDR) Secretariat, Geneva, http://www.unisdr.org/unisdr/docs/

early/geo/geo.htm, accessed on October, 8, 2003.

Hoek, E., and Bray, J.M. 1974. Rock Slope Engineering , Institute o Mining and Metallurgy, London.

Hoek, E., and Bray, J.M. 1977. Rock Slope Engineering , Institute o Mining and Metallurgy, London.

Hutchinson, J.N. 1977. “Te Assessment o the Effectiveness o Corrective Measures in Relation to

Geological Conditions and ypes o Slope Movement.” Bulletin of International Association of

Engineering Geology , 16, 131–155.

Hutchinson, J.N. 2001. “Landslide Risk—o Know, to Foresee, to Prevent,” Journal of echnical andEnvironmental Geology , 3, 3–22.

Janbu, N. 1954. Stability Analysis of Slopes with Dimensionless Parameters, Harvard Soil Mechanics Series

46, Harvard University Press, Cambridge, MA.

Kelly, J.M.H., and Martin, P.L. 1986. “Construction Works on or Near Landslides,” In Proceedings o the

Symposium o Landslides in South Wales Coalfield, Polytechnic o Wales, pp. 85–103.

Kennedy B.A., and Niermeyer, K.E. 1970. “Slope Monitoring System Used in the Prediction o a Major

Slope Failure at the Chuquicamata Mine, Chile,” In Proceedings o the Symposium on Planning

Open Pit Mines, Johannesburg, pp. 215–225.

Krahn, J. 2003. “Te 2001 R.M. Hardy Lecture: Te Limits o Limit Equilibrium Analyses,” Canadian

Geotechnical Journal , 40(3), 643–660.Krahn, J. 2004. Stability Modeling with Slope/W. An Engineering Methodology , Geo-Slope/W International

Ltd., Calgary, Alberta, Canada, 396 pp.

Kramer, S.L. 1996. Geotechnical Earthquake Engineering , Prentice-Hall, Upper Saddle River, NJ.

http://www.unisdr.org/




8/20/2019 b15621-12 (1)



Lee, E.M., Moore, R., Burt, N., and Brunsden, D. 1991. “Strategies or Managing the Landslide Complex

at Ventnor, Isle o Wight,” In Proceedings o the International Conerence on Slope Stability

Engineering, Isle o Wight, Tomas elord, London, pp. 219–225.

Leroueil, E. 1996. “Landslide Hazard—Risk Maps at Different Scales: Objectives, ools and Developments,”

In Proceedings o the International Symposium on Landslides, rondheim, June 17–21, (Senneset,K., [Ed.]), pp. 35–52.

Leroueil, S., and avenas, F. 1981. “Pitalls o back-analyses,” In Proceedings o the 10th International

Conerence on Soil Mechanics and Foundation Engineering, 1, 185–190.

Newmark, N.M. 1965. “Effects o Earthquakes on Dams and Embankments,” Géotechnique,

15(2), 139–159.

Popescu, M.E. 1984. “Landslides in Overconsolidated Clays as Encountered in Eastern Europe,” State-o-

the-Art Report,” In Proceedings o the 4th International Symposium on Landslides, oronto, ON,

pp. 83–106.

Popescu, M.E. 1991. “Landslide control by means o a row o piles. Keynote paper.” In Proceedings o the

International Conerence on Slope Stability Engineering, Isle o Wight, Tomas elord, pp. 389–394.

Popescu, M.E. 1996. “From Landslide Causes o Landslide Remediation, Special Lecture,” In Proceedingso the 7th International Symposium on Landslides, rondheim, 1, 75–96.

Popescu, M.E. 2001. “A Suggested Method or Reporting Landslide Remedial Measures,” International

Association of Engineering Geology Bulletin, 60(1), 69–74.

Popescu, M.E., Schaeer V.R. 2008. “Landslide stabilizing piles: a design based on the results o slope ailure

back analysis,” In Proceedings o the 10th International Symposium on Landslides and Engineered

Slopes, Xi’an, China, pp. 1787–1793

Popescu, M.E., and Seve, G. 2001. “Landslide Remediation Options Afer the International Decade or

Natural Disaster Reduction (1990–2000),” Keynote Lecture, In Proceedings o the Conerence

ransition rom Slide to Flow—Mechanisms and Remedial Measures, ISSMGE C-11, rabzon,

pp. 73–102.Popescu, M.E., and Yamagami, . 1994. “Back analysis o slope ailures—A possibility or a challenge?”,

In Proceedings o the 7th Congress International Association o Engineering Geology, Lisbon, 6,

4737–4744.

Powel, G.E. 1992. “Recent Changes in the Approach to Landslip Preventive Works in Hong Kong,” In

Proceedings o the 6th International Symposium on Landslides, Christchurch, 3, 1789–1795.

Saito, M. 1965. “Forecasting the time o occurrence o slope ailure,” In Proceedings o 6th International

Congress o Soil Mechanics and Foundation Engineering, Montreal, Canada, 2, 537–541.

Saito, M. 1980. “Reverse calculation method to obtain c and φ on a slip surace,” In Proceedings o the

International Symposium on Landslides, New Delhi, 1, 281–284.

Schuster, R.L. 1995. “Recent Advances in Slope Stabilization,” Keynote paper. In Proceedings o the 6th

International Symposium on Landslides, Christchurch, 3, 1715–1746.Schuster, R.L. 1996. “Socioeconomic Significance o Landslides,” Chapter 2 In Landslides—Investigation

and Mitigation, urner, A.K., and Schuster, R.L., (eds.), ransportation Research Board Special

Report 247, National Research Council, National Academy Press, Washington, DC, pp. 12–31.

Skempton, A.W., Leadbeater, A.D., and Chandler, R.J. 1989. “Te Mam or Landslide, North Derbyshire.”

Philosphical ransactions of the Royal Society , London, 329(1607), 503–547.

Smith, D.I. 1984. “Te Landslips o the Scottish Highlands in Relation to Major Engineering Projects.”

British Geological Survey Project 09/LS. Unpublished report or the Department o Environment.

Soeters, R., and van Westen, C.J. 1996. “Slope Instability Recognition, Analysis, and Zonation,” In

Landslides: Investigations and Mitigation, urner, A.K., Schuster, R.L. (eds.), ransportation Research

Board Special Report 247, National Research Council, Washington, DC, pp. 129–177.aylor, D.W. 1937. “Stability o Earth Slopes,” Journal of the Boston Society of Civil Engineers, 24(3).

Reprinted in Contributions to Soil Mechanics, 1925–1940, Boston Society o Civil Engineers,

Boston, MA, 1940, pp. 337–386.

8/20/2019 b15621-12 (1)



erzaghi, K. 1950. Mechanisms o landslides, In Application of Geology to Engineering Practice, Berkley

Volume, Savage, J.L. et al., (eds.) Geological Society o America, Boulder, Colorado, pp. 83–123.

randafir, A.C., and Sassa, K. 2004. “Newmark Deormation Analysis o Earthquake-Induced Catastrophic

Landslides in Liquefiable Soils,” In Proceedings o the 9th International Symposium on Landslides,

Rio de Janeiro, Volume 1, Balkema, Rotterdam, Te Netherlands, pp. 723–728.randafir, A.C., and Sassa, K. 2005. “Seismic triggering o catastrophic ailures on shear suraces in satu-

rated cohesionless soils,” Canadian Geotechnical Journal , 42(1), 229–251.

ransportation Research Board. 1996. “Landslides: Investigation and Mitigation,” ransportation Research

Board Special Report 247, 1996. (Various contributing authors)

United Nations. 1996. “Mudflows—Experience and Lessons Learned rom the Management o Major

Disasters,” United Nations Department of Humanitarian Affairs, New York, 139 pp.

Varnes, D.J. 1978. “Slope Movements and ypes and Processes,” In Landslides Analysis and Control ,

ransportation Research Board Special Report 176, pp. 11–33.

Voight, B., and Kennedy, B.A. 1979. “Slope ailure o 1967–1969, Chuquicamata Mine, Chile,” In Rockslides

and Avalanches, Elsevier, Amsterdam, 2, 595–632.

Working Party on World Landslide Inventory (WP/WLI): International Geotechnical Societies’UNESCO Working Party on World Landslide Inventory, Cruden, D.M., Chairman. 1991. “A

Suggested Method or a Landslide Summary,” International Association of Engineering Geology

Bulletin, 43, 101–110.

Working Party on World Landslide Inventory (WP/WLI): International Geotechnical Societies’ UNESCO

Working Party on World Landslide Inventory. Working Group on Landslide Causes, Popescu, M.E.,

Chairman. 1994. “A Suggested Method or Reporting Landslide Causes,” International Association of

Engineering Geology Bulletin, 50, 71–74.

Zaruba, Q., and Mencl, V. 1982. Landslides and Teir Control , Elsevier, Amsterdam, 324 pp.

Further ReadingBrandl, H. 1995. “Observational Method in Slope Engineering,” In Proceedings o, International

Symposium on 70 Years o Soil Mechanics, Istanbul, pp. 1–12.

Bromhead, E.N. 1992. Te Stability of Slopes, Blackie Academic & Proessional, London.

Carrara, A., and F. Guzzetti (eds). 1995. Geographical Information Systems in Assessing Natural Hazards,

Kluwer Academic Publishers, Dordrecht, Netherlands, 353 pp.

Central Laboratory o Bridges and Roads. 1994. Monitoring o Unstable Slopes, Guide echnique,

Paris.

Fell, R., and Hartord, D. 1997. “Landslide Risk Management,” In Landslide Risk Assessment , Cruden,

D.M., and Fell, R. (eds.), Balkema, Rotterdam, Te Netherlands, pp. 51–110.

Fell, R., Hungr, O., Leroueil, S., and Riemer, W. 2000. “Geotechnical Engineering o the Stability o NaturalSlopes, and Cuts and Fills in Soil,” Keynote Lecture, In Proceedings o GeoEng 2000, Melbourne,

pp. 21–120.

Fukuzono, . 1985. “A New Method or Predicting the Failure ime o a Slope.” In Proceedings o the 4th

International Conerence and Field Workshop in Landslides, okyo, pp. 145–150.

Hartlen, J., and Viberg, L. 1988. “General Report: Evaluation o Landslide Hazard,” In Proceedings o

5th International Symposium on Landslides, Lausanne, Balkema, Rotterdam, Te Netherlands, 2,

1037–1057.

LCPC: Central Laboratory o Bridges and Roads. 1994. Monitoring of Unstable Slopes, Guide echnique,

Paris.

“Primer on Natural Hazard Management in Integrated Regional Development Planning. 1991.”Department o Regional Development and Environment Executive Secretariat or Economic and

Social Affairs Organization o American States, Washington, DC.

8/20/2019 b15621-12 (1)



Varnes, D.J., and Te International Association o Engineering Geology Commission on Landslides and

other Mass Movements. 1984. “Landslide Hazard Zonation: A Review o Principles and Practice,”

Natural Hazards, Volume 3, Paris, France, UNESCO, 63 pp.

Relevant Websites 1. http://en.wikipedia.org/wiki/Landslide

2. http://landslides.usgs.gov/

3. http://geology.utah.gov/utahgeo/hazards/landslide/index.htm

4. http://en.wikipedia.org/wiki/Slope_stability

5. http://www.bt.cdc.gov/disasters/landslides.asp

6. http://daveslandslideblog.blogspot.com/

7. http://www.ga.gov.au/hazards/landslide/landslide-basics/where.html

8. http://www.geotechnicalino.com/slope_stability_publications.html

http://en.wikipedia.org/


http://landslides.usgs.gov/


http://geology.utah.gov/




http://www.bt.cdc.gov/


http://daveslandslideblog.blogspot.com/


http://www.ga.gov.au/


http://www.geotechnicalinfo.com/










8/20/2019 b15621-12 (1)


b15621-12 (1)

Documents