Bluff Slumping Workshop proceedings, Romulus, Michigan, February, 1982

[From the U.S. Government Printing Office, www.gpo.gov]

AM
............ .....

0
fr

GB
450.2 ,oceeidings of the 19a'r%F2% Worksbab
.B65
B!"ff-

1982 Sponsored by
U.S. Geologkal Survey and Micbigan Sea Grant

PROCEEDINGS

BLUFF SLUMPING WORKSHOP

Romulus, Michigan
Februaryl 1982

Sponsored by

Michigan Sea Grant
The Earth Sciences Assistance Office of the
United States Geological Survey

US DepartSment of Coln erce
AA Coastal ervices Center LibrarY
NO@ 2234 South 1-7107hscn A--TC-27.10
Charleston, SC

MICHU-SG-82-901

Michigan Sea Grant
Publications Office
2200 Bonisteel Boulevard
Ann Arbor, Michigan 48109

This publication is a result of work sponsored by the Michigan Se2 Grant Program,
Project R/CE-4, with a grant NA-80AA-D-00072, from the Office of Sea Grant, Nationdl
Oceanic and Atmospheric Administration (NOAA), U.S. Department of Commerce and
funds from the State of-Michigan.;

TABLE OF CONTENTS

1. Introduction

11. State-of-the Art Papers

A. Causes & Mechanics of Coastal Bluff Recession
in the Great Lakes T. Edil .................................................................. I
B. Methods of Preventing Bluff Recession - E. Brater .................................. 51

III. Selected Papers

A. Assessment of the Effectiveness of Corrective Measures in
Relation to Geological Conditions and Types of Slope Movement
J. N. Hutchinson ...................................................................................... 61
B. Regulations to Reduce Coastal Erosion Losses - D. A. Yanggen ............... 87

IV. Summaries of Workshop Discussion Groups

A. Problem Identification & Assessment ........................................................ 109
B. Controlling Factors ..................................... ............ ill
C. Mitigation Measures ........... o.................... o......................................... o ........ 112

V. Research Needs ......o ..... o...................................................... o ........................... 117
VI. Public Dissemination .... oo ............o..........o................................................ o ......... 121

VII. List of Workshop Attendees ... o......................................... o ............................... 125

VIII. Additional Sources of Information ................................................................... 129

Problems of Great Lakes Bluff Erosion

Coastal erosion in the- Great Lakes has resulted in public and private costs
through loss of property and damage to structures, construction of protective devices,
and public diaster relief and recovery expenditures. An estimated two-thirds of Great
Lakes shoreline property owners have experienced some type of erosion damage. A
key factor in shoreline erosion is bluff slumping. The U.S. Army Corps of Engineers
estimated that 150,716 M3 of Great Lakes bluffs were eroded between 1972 and 1976.
A total of 41,196 residences lie within 60 M of a bluff edge. In Michigan alone, 80
houses were destroyed between 1974 and 1978, and an estimated 800 more were in
imminent danger.

Although the effects of blu 'ff slumping are often substantial, there had been no
workshop for researchers which focused on the role of groundwater seepage and other
geologic, hydrologic, and engineering factors in Great Lakes bluff stability until Michigan
Sea Grant, in conjunction with The Earth Sciences Assistance Office of the U.S.
Geological Survey, held such a workshop on bluff slumping in mid-February. The goal
of this workshop was to improve management of coastal hazards by increasing
professional and public knowledge of the role of groundwater and other elements which
affect ground stability. The Bluff Slumping Workshop brought together engineers,
geologists, hydrologists, and other scientists who specialize in slope stability. The
objective was to assess the present scientific and technical knowledge dealing with
bluff slumping and its mitigation and to determine further research and data collection
needs.

The workshop was developed around two state-of-the-art papers dealing with
causes and mechanics and mitigation and prevention. Participants dealt with questions
of problem identification, assessment, controlling factors, mitigation methods, and
research needs.

Results of the workshop will be published by the end of June. Copies can be
obtained by requesting the Proceedings of the Bluff Slumping Workshop from:
Publications, Michigan Sea Grant Program, 2200 Bonisteel Blvd., Ann Arbor, MI 48109.

I
I
I
I
I
I
I
STATE-OF-THE-ART PAPERS
I
I
I
I
I
I
I
I
I
I
I
I

CAUSES AND MECHANICS OF COASTAL BLUFF
RECESSION IN THE GREAT LAKES

prepared by

Tuncer S. Edit

Department of Civil and Environmental Engineering
University of Wisconsin-Madison

CAUSES AND MECHANICS OF COASTAL BLUFF

RECESSION IN THE GREAT LAKES

The State-of-the-Art-Paper
by Tuncer B. Edil
Department of Civil and Environmental Engineering
University of Wisconsin-Madison

Proceedings of the Workshop on Bluff Slumping in the Great
Lakes

February 18-19,1982
Detroit, Michigan

INTRODUCTION

This paper presents a review of the literature available
on the causes and mechanics of coastal recession in the Great
Lakes with the purpose of delineating the most likely role
and significance of various factors such as wind and wave
action, composition of bluff materials, ground water, and
vegetal cover. Theoretical information and the reports of
investigators working on coastal slope stability problems in
other regions were also used as needed in order to clarify
the issues.
The physical characteristics, use, and ownership of the
Great Lakes shoreline and the scope and economic impact of
the coastal erosion, flooding, and bluff slumping problems
have been documented in some detail (International Joint
Commission, 1973; U.S. Army, Corps of Engineers, 1971;
Environment Canada, 1975; Great Lakes Basin Commission,
@1981). Table 1 gives a summary of these characteristics
(International Joint Commission,'1973). Nearly 65 percent
(10,444 km) of the.16,047 km-long Great Lakes shoreline is
designated as having significant erosion with about 5.4
percent (860 km) of it being critical. The total damage to

the United States shoreline of the Great Lakes due to wave
action during the high lake level period May 1951 through
April 1952 is placed at about 50 million dollars (1952 price
level) in the same report by the International Joint
Commission.
Nearly 32 percent of the U.S. shoreline, not including
the islands, consists of erodible bluffs (Table 2). Sand
dunes are encountered in 8 per cent and erodible low plains
in 17 per cent of the U.S. shoreline (Great Lakes Basin
Framework Study: Appendix 12, 1975). Extensiveness of the
shoreline formed in erodible bluffs and dunes and often
complex response of this type of shoreline to wave erosion
make slope processes an important part of the shore recession
problem. The bluff processes encountered in the Great Lakes
are described in the next sections. A discussion and
comparison of the role and significance of the causes and
processes follow it. Finally, the conclusions and the
recommendations for further research are presented.

SLOPE PROCESSES

Slope movement is an expression of force overcoming
resistance. Coastal bluffs are examples of systems in which
force@and resistance are continually opposed. This
equilibrium is altered by changing environmental conditions
that can initiate downslope movement. When the forces
within a moving mass become less than the resistances to
movement, the material will slow down and eventually stop.
Force requires energy and all energy in geomorphic
systems is ultimately derived from either gravity or climate
(Carson and Kirkby, 1972). The forc e provided by gravity is
simply that of the weight of bluff forming materials
including soil water and the external loads such as
buildings, snow, piled materials, etc.-placed on the bluff.
To these forces can be added vibrations resulting from
earthquakes, blasting and machines. The intensity of shear
force is a function of gravity and it varies from point to

point in a bluff; however, the factors which control it,
along a potential failure surface, include slope inclination,
height and unit weight of slope forming materials and
included water.
Climate, through its control on water, air, and
temperature, provides energy for the most important forces on
bluffs. Climate related forces include wind and wave action,
surface and ground-water flow, rain impact, moisture and
temperature related ground expansion, and ice action.
Resistance to these forces that tend to move.materials
down and/or away from the bluffs is provided primarily by the
shear strength of these materials. Vegetation and man-made
structural systems such as shore protection and bluff
stabilization structures may provide additional resistance.
Shear strength is not constant for a given material. @It may
change in time due to weathering and ground water pressure
changes which are in turn controlled by certain climatic
factors. Furthermore, use of certain technologies can
improve shear strength of soils or minimize the detrimental
effects of the climatic factors.
The interaction of driving force and soil resistance
results in a number of processes leading to debris production
and removal. Gray (1977) summarized debris production and
removal processes and the involved variables in a diagram.
Basically all processes are alike in that material begins to
move only when the forces involved become greater than the
resistances. Once movement has begun, the mode of
interaction of force and resistance may differ greatly from
one process to another and these differences are commonly
used to classify geomorphic processes (Sharpe, 1938). The
commonly encountered processes in the Great Lakes coastal
bluffs include wave/current erosion, wind erosion, ice
erosion, rain fall erosion (rain impact and sheet/rill
erosion), ground water sapping, sliding/slumping,
solifluction, debris flow, and creep.
It is possible to separate these processes into two
broad groups, namely mass movement and particle movement. In
the former group debris begins to move as a coherent unit.

It the movement of the mass is along a well-defined surface
without internal shear (rigid body movement) it is termed
slide (or slump). On the other hand, if the shear is
distributed throughout the mass (viscous flow) without a
sharply defined failure surface it is termed flow. In flows
all the movement occurs as differential movement within the
body of the flowing mass. Movements in which particles move
as individuals, with little or no relation to their neighbors
are particle movements. Distinction between these categories
is often difficult. Nevertheless, some processes seem to be
mainly particle movements, especially the erosional processes
caused by waves, currents, rain, ground water, and winds
(Carson and Kirkbyl 1972).
The concepts presented above are summarized in Table 3.
There are other geomorphic processes and there may be other
ways of classifying these processes; but we are concerned
with the processes most commonly encountered in the Great
Lakes region here. In the next sections the nature of the
various slope processes is described.

WAVE EROSION

Probably the most significant geomorphic process along
the Great Lakes shoreline is the erosion and removal of
shorelines materials by waves. Wave action is important both
in itself and in initiating and perpetuating other geomorphic
processes in those segments of the shoreline where bluffs are
encountered. The nature and magnitude of wave erosion
depends broadly on wave energy and erodibility of materials
at a specific location along the shoreline. The wave energy
available at a given point on the shoreline for erosion of
beach and bluff materials depends on the following factors:

1. wave climate (wind velocity, duration and fetch)
2. water level relative to the beach and bluff toe
3. nearshore and offshore bathymetry
4. shore configuration (orientation)

These factors are not constant but instead change
significantly with time. Most notable is the water level
fluctuations in the Great Lakes.

Lake Levels

Records going back to 1860 indicate that the lake levels
have varied considerably. Superimposed on the very long-term
trends resulting from crustal tilting (about 0.52 m per
century) and the seasonal fluctuations (as high as 1 m) are
some extremely short periods of changes, of varying
magnitudes (Tovell, 1977). The most temporary of these are
caused by winds that blow along the long axis of a lake and
drive the waters to one end. In Lake Erie wind set- ups have
caused differences in water levels of more than 4 m between
Buffalo, New York and Toledo, Ohio. A second cause of
temporary changes is seiches, which are changes in lake
levels due to differences in atmospheric pressure at
different ends of a lake.
The Great Lakes also exhibit long-term water level
changes without regular periods. The intervals vary from 10
to 30 years. The magnitudes of these long-term variations
are three-and-a-half to six times greater than the average
seasonal variations. These changes are associated with the
changes in the volume of water in the lakes. Fig. 1 shows
the regular seasonal cycle, with lowest levels in the late
fall and winter months and highest levels in late spring and
summer, along with the long-term fluctuation for the period
1952 to 1973. The volume of water is dependent upon the
amount of precipitation over the lake, the amount of water
delivered by rivers and streams flowing into the lake, the
inflow from the lake above, the flow of groundwater into the
lake, and any artificial diversion into the lake from outside
the basin. The water is lost by evaporation, outflow
(natural or artificial diversion), and withdrawal for
municipal and industrial use.
The principal factor that determines the water budget in
the lakes is climate. The periodic fluctuations in Great

Lakes water levels are largely due to the confined outlet
conditions of the lakes combined with variations in climate,
i.e. precipitation and evaporation. There is also a certain
amount of regulation of the lake levels between the upper
lakes and the lower lakes by controlling the water discharge
through the dams at the lake outlets.
It has been generally accepted that the lake level
fluctuation is the primary factor affecting the rate of shore
erosion and the consequent bluff recession (Brater, 1975;
Berg and Collinson, 1976; Carter, 1976; Seibel 1972; Davis et
al., 1973; Quigley and Di Nardo, 1978; Zeman,1978). There is
considerable local variation in erosion rates due to the
presence of the controlling factors other than the lake
level. Along shoreline segments with bluffs, slope processes
also contribute to bluff recession usually with a complicated
time lag relationship with respect to toe erosion. For these
reasons it is not easy to correlate lake levels with erosion
rates especially if the time period considered is short.
Therefore, today, in spite of a wide perception that there is
a direct influence of lake level on accelerated erosion,
there is not a comprehensive study demonstrating this fact
clearly and directly. Johnson and Hiipakka, (1976) even
present a statistical study indicating the lack of
correlation between the rate of erosion and lake levels.
Nevertheless, available evidence, observations,'and
theoretical considerations (see International Joint
Commission, 1973) all indicate that high lake levels play a
passive role, as pointed out by Davis et al. (1973) in that
they allow erosion to take place at a rapid rate.

Erodibility of Coastal Materials

The amount of wave erosion depends on the erodibility of
coastal materials in addition to the wave energy delivered.
Erodibility is largely a function of composition. The.Great
Lakes are surrounded by a wide variety of coastal types and
deposits, although the coasts are generally dominated by
Pleistocene glacial deposits. Each of the lakes has a

portion of its coast which is comprised of bedrock. In
general, bedrock portions of the shoreline are most resistant
to wave action. However, erosion in solid rock may present a
potential hazard if unrecognized as reported by Phillips
(1978) for the north and east shores of Lake Superior.
Most of the recent critical erosion is, however,
confined to the coasts of unconsolidated sediments. These
may be in the form of glacial till or outwash, lake sediments
or dunes which are reworked from glacial sediments. In
general cohesionless materials such as sand and silt are the
most erodible materials with cohesive soils (clay, silty
clay, etc.) having somewhat more resistance to wave erosion
due to cohesion between the particles.
Presently there are analytical procedures available to
evaluate wave climate at'a specific site using numerical
hindcasting procedures. There are a number of such
procedures (Pierson et al., 1955; Sverdrup and Munk, 1957;
Gelci et al., 1957; Bretschneider, 1958) and they produce the
wave energy density spectrum at a deep water hindcasting site
based on the wind data for the site. A representative wave
height from the wave energy density spectrum may then be
calculated. A wave entering the nearshore zone will be
slowed, shortened, and steepened as it moves into
progressively shallower water. This process is known as
shoaling. Furthermore, a wave arriving at an angle to the
shoreline will be bent toward alignment with the underwater
depth contours, since that portion of the wave front bending
is called refraction and shoaling (Dobson, 1967). The
refraction analysis requires, together with the deep water
wave heights and directions of propagation, the hydrographic
information (bathymetry). These studies are useful for the
design of shore protection structures and for estimating long
shore currents, rip currents and sediment transport by
littoral drift. Wave climate has been evaluated at different
sites in the Great Lakes (Brater and Ponce Campos, 1978; Edil
et al., 1979; Gelinas and Quigley, 1973; Resio and Vincent,
1976; Keillor and De Groot, 1978).

Rate and Amount of Wave Erosion

In summary, it can be stated that there are quantitative
procedures developed to evaluate wave energy on a
site-specific basis; however, the physical information used
in the analysis involves a number of variables which are
usually difficult to measure precisely and have a
probabilistic nature. Therefore, extensive field calibration
of the analytical procedure is required for a realistic
evaluation. Furthermore, the relationship of wave energy to
the rate and amount of shore and bluff-toe erosion has not
been satisfactorily established. Quigley and Gelinas (1976)
reported an approximately linear relationship between the
150-year erosion rate and break wave energy. However, a
closer examination of their data shows poor linearity and
wide scatter. Nevertheless, a qualitative general trend for
increasing erosion rate with increasing wave energy is
apparent. Berg and Collinson (1976), based on a detailed
study of bluff recession at more than 21 sites on the
Illinois shore of Lake Michigan, suggested that serious bluff
erosion from wave attack becomes significant beyond a
threshold lake level. They also noted the time lag between
lake level drop or rise and recession rate.
In another study Fisher et al. (1975) tried to correlate
wave energy (expressed as wave height) with shoreline change
for a segment of the Atlantic coast. He found promising
qualitative but poor quantitative correlation on a regional
scale. The correlations improved somewhat on the local
scale. This shows that local conditions are significant in
controlling recession rates and wave energy but it is only
one of the factors involved. This latter study involved a
shoretype with sand dunes. When high bluffs formed in
cohesive soils are present the recession is further
complicated by landslide activity that makes such
correlations even more difficult especially when a relatively
short time period is considered.
Wave action, as a geomorphic process, is the primary
factor responsible for the changes in slope geometry on

coastal bluffs. The effect of wave action on the bluff takes
place in an active and/or passive manner. In the former,
waves directly attack and erode the intact native toe
material resulting in steepening of slopes, initiation of
slides, loss of vegetation, and other surficial slope
processes. In the passive case, wave action causes the
removal of material which may collect at the toe and on the
beach. This material may be contributed from the bluff face
due to free degradation or slides in the upper parts of the
bluff (termed colluvium). In the passive case, the effect of
wave action is to perpetuate instability by preventing the
flattening and stabilization of the bluff by natural
processes. However, after sufficient recession and
stabilization a bluff would not be affected by this type of
wave activity barring an increase in wave activity from the
statistical average such may be induced by, for example,
increasing lake level.

SLIDING/SLUMPING

The processes of downslope movement what
geomorphologists referto as mass wasting or mass movements,
includes many types of movement. An earlier and widely
accepted engineering classification referred to these
processes as landslides (Varnes, 1958). A recent update of
the same classification adopts the term "slope movements"
(Varnes, 1978). The chief criteria used in this
classification are type of movement primarily and type of
material secondarily (Table 4). Types of movement are
divided into five main groups: falls, topples, slides,
spreads, and flows. A sixth group, complex slope movements,
includes combinations of the other five types. Materials are
divided into rock and soil and soil is further divided into
debris (coarse-grained) and earth (fine-grained). Any of
these slope movements may occur along the Great Lakes
shoreline. However, slides (both rotational and
translational) and flows (including solifluction) are two

types of movement that are encountered most commonly and will
be addressed herein. Falls in oversteepened bluffs have also
been observed and Quigley and Gelinas (1976) report a
toppling failure. These processes do not lend themselves to
quantitative analysis and furthermore, are not encountered on
a wide-spread basis.

Shear Strength Behavior of Soils

It is appropriate to discuss the shear strength behavior
of soils briefly since it is one of the important factors
which control the occurence and mode of sliding. The
Mohr-Coulomb criterion is most widely used to define shear
failure in soils. According to this criterion the shear
strength can be expressed consistently in terms of effective
normal stress (a') acting on the failure surface as

S=C.-+ (cy-u) tan@' = c6 + a' tan@' W

where c' and @' are the effective (or drained) strength
parameters: cohesion intercept and angle of internal
friction, respectively. There are two parts, in general, to
shear strength: a constant cohesional part and a variable
frictional part. The effective strength parameters and the
unit weights of the materials encountered in the bluffs of
the Great Lakes shorelines are compiled in Table 5 from the
published reports of various investigators.
Effective stress (ai) is defined as

0, = a - u (2).

where a is total normal stress and u is pore-water pressure.
Total stress is generated by gravity. Therefore, it
increases generally in a linear fashion with depth in a soil
mass and it is a function of the unit weight of the soil (Y).
Unit weight of soil depends on its density (void ratio),
specific gravity of the solids, and the degree of saturation.

The actual stress at any point within the soil will
depend on the distribution of stresses between the liquid and
solid phases for a saturated soil. The water pressure within
the soil pores is termed pore-water pressure. In general,
pore pressure consists of two parts: one is related to static
or flowing ground water and the other is the pore pressure
change (excess pore pressure or pore pressure deficiency)
resulting from changing stresses. Placing a load on the soil
surface, making an excavation in soil, removal of part of the
slope materials by erosion and slips all cause changes in the
state of stress (both normal and shear stresses) at different
points in the soil mass. As soil tends to contract or dilate
in response to these stress changes, a change in pore
pressure occurs. Therefore, as the pore pressure changes in
response to ground-water levels and/or in response to stress
changes in the soil, the effective stress and consequently
the shear strength changes at a given point in the soil mass.
Water flows in and out of a soil element on a potential
failure surface as the pore pressure changes at the point.
In the case of granular soils (sands and gravels) this flow
and adjustment of pore pressure takes place rapidly (as
rapidly as it is induced) due to high permeability of such
materials. In soils containing significant quantities of
fine-grained material (clay and silt), however, there is a
delay in response to pore pressure changes due to low
permeability.

Effective Stress Method (Long-Term Stability)

In the case of natural slopes stability is usually
considered as a long-term problem. The long-term stability
analysis is performed using the effective stress computed by
subtracting the pore pressure related to ground water from
the total stress and assuming that the pore pressure change
due to stress changes will be dissipated in the long-term and
will be equal to zero. Coastal bluffs are in constant
evolution due to the combined effects of toe erosion, slides,
and face degradation. These processes, in.general, cause a

decrease in total lateral normal stress and increase shear
stresses along potential failure surfaces accompanied by a
decrease in pore pressure. Subsequently the soil swells as
water flows into this zone and shear strength decreases with
time. This situation becomes more critical as the
overconsolidation of clays increases.
Therefore, the long-term stability should be considered
in terms of effective stresses and assuming drained
conditions for the coastal bluffs formed in stiff clays of
the Great Lakes region. The effective stress method is also
used in well-drained granular soils in which the short and
long-term stabilities coincide due to the immediate
dissipation of pore pressure changes. Edil and Vallejo
(1977) described bluff stability at two sites on the shore of
Lake Michigan. Where unexpected stability occurred, it could
be explained in a rational manner by this process of delayed
failure. Analyses of short-term stability tended to indicate
stability while long-term stability analyses indicated
instability or low stability. Quigley and Gelinas (1976)
analyzed the stability of a typical bluff of clay till having
a height of 40 m and an increasing amount of toe-material
being removed. An analysis of their results assuming
short-term and undrained conditions indicates that nearly 52
m of toe erosion is required to initiate an immediate
failure. This is unlikely. On the other hand, about 20 m of
erosion (a reasonable possibility), whether it occurs all at
once or gradually, creates instability in the long-term
analysis.

Rotational Slides (Slumps)

In slides, the movement consists of shear strain and
displacement along one or several surfaces that are visible
or may reasonably be inferred, or within a relatively narrow
zone.(Varnes,.1978). Slides are subdivided into rotational
slides (slumps) and translational slides. The former is a
slide along a surface of rupture that is curved concavely
upward. In many slumps this surface is spoon-shaped and the

movement is more or less rotational about an axis that is
parallel to the slope. The classic purely rotational slump
on a surface of smooth curvature is relatively common in
fairly homogeneous materials such as constructed embankments
and fills. Slides in natural slopes tend to be complex or
are at least controlled in their mode of movement by internal'
inhomogeneities and discontinuities such as the presence of a
very weak layer or fractures. Barring the presence of such
gross inhomogeneities, rotational slides involving
approximately circular rupture surfaces have been observed
and analyzed in the Great Lakes bluffs formed in cohesive
soils (Quigley and Tutt, 1968; Edil and Vallejo, 1977; Edil
and Haas, 1980).
Deep-seated rotational slips occur in clayey soils and
do not occur in sands. One method of analysis of rotational
slides that is accurate for most purposes is that advanced by
Bishop (1955). The forces acting on a typical slice of the
slumping mass is shown in Fig. 2. From a consideration of
the moment equilibrium about the center of rotation, the
equation for safety factor (F) is obtained:

Z1[c1b + (W-ub) tano') (1/m
EW sina (3)

where

m cosa[l + (tana tanol)/F (4)

Safety factor is defined as the ratio of the available
shearing resistance along a given slip surface to the
calculated shearing resistance required for equilibrium. In
other words, it is that factor by which the shear strength
parameters may be reduced in order to bring the slope into a
state of limiting equilibrium along a given slip surface. A
value of F equal to. unity indicates failure with values
greater than unity indicating increasing degrees of safety.
Normally a number of circles are examined to locate the most
critical circle with the lowest factor of safety. The

failure arc predicted by the Bishop method has been found to
compare well with actual failure surfaces in bluffs on the
Great Lakes (Edil and Vallejo, 1977) and other places or
(Sevaldson, 1956).
Vallejo and Edil (1979) developed stability charts for
rapid evaluation of the state of stability of actively
evolving Great Lakes coastal slopes using the effective
stress approach and the Bishop method. These charts indicate
the stability status as well as the type of potential
failure, whether deep or shallow, to which the bluffs may be
subjected. The geometric changes that the bluffs will be
subjected to before becoming stable can be discerned from the
charts. An example of these charts is given in Fig. 3. The
family of curves represents height-inclination combinations
for the limiting long-term safety factor of unity with r'
respect to a deep-slip type of failure. Skempton and
Hutchinson (1969) defined the deep slip failures as the ones
having values of the ratio between the maximum thickness of
the slide, D, and the maximum length of the slide, L, ranging
from'O.15 to 0.33 (Fig. 4). The slopes with
height-inclination combinations falling within Zone A are
unstable slopes with potential deep slips in the long-term.
The stable angle below which no more rapid mass movement will
take place (except perhaps creep) is given by the ultimate
angle of stability (@u) (Skempton and DeLory, 1953). This
angle defines the upper limit of Zone C of stable slopes. A
slope which is in Zone B (between Zones A and C) will
experience shallow failures. The shallow failures are
defined by Skempton (1953) as those having values of D/L
between the limits of 0.03 and 0.05. The shallow slips could
be planar slides (slab slides), small rotational slips, and
flows. For this reason, the slopes in Zone B can be
classified as stable slopes with local instability or
quasi-stable slopes.

Influence of Slope Parameters on Stability of Bluffs

Edil and Vallejo (1980) made a theoretical study of the

influence of slope parameters on the long-term stability of
the coastal bluffs. Based on this analysis, various slope
parameters are involved in the following ways:
a) the cohesive component of the shear strength of soil
is the dominant factor providing resistance to
failure for slopes with@ heights less than about 25m.
b) the friction component is the dominant factor for
slopes greater in height than about 25m.
c) an increase in ground-water level produces an overall
decrease in the safe slope angle at any height.
d) a high slope (greater than 25m) will reach unstable
conditions faster than a low slope if the two slopes
are steepened equally.
e) unit weight influences slope stability differently
depending upon the position of the ground-water
table. For low ground-water levels (less than one
quarter of the slope height measured from the slope
base) slopes of lower unit-weight materials are more
stable; for high water levels (more than three
quarters of the height measured from the slope base)
higher unit weight materials yield more stability.
For intermediate water levels the effect of unit
weight depends on the slope height.
For a given slope most of these parameters, such as
height and materials (c',@ ', and Y) are fixed. The ones which
are likely to vary with changing environmental conditions are
inclination and ground-water level. Fig. 3 shows the effect
of slope inclination on the stability in terms of height and
cohesion intercept for fixed angle of friction, unit weight
and ground-water level. The nature of the relationship
varies only quantitatively for different values of the latter
parameters. Fig. 5 shows the height-inclination
relationships at limit equilibrium conditions for different
relative ground-water levels, Hw , and different values of V
for constant c' and y. An increase in H from 1/4 to 3/4 of
w
the slope height has a significant effect on reducing
critical slope inclination at a given slope height. This
trend demonstrates the need for accurate assessment of the

ground-water conditions when considering the stability of
bluffs as well as the effectiveness of slope drainage in
improving the stability against deep-seated slumps.

Translational Slides

in translational slides the mass progresses down and out
along a more or less planar or gently undulating surface and
has little of the rotational movement or backward tilting
characteristic of slump. A translational slide in which the
moving mass consists of a single unit that is not greatly
deformed or a few closely related units may be called a block
slide. An example of such a failure involving a block of
fractured till in the upper part of a coastal bluff in
Milwaukee County, Wisconsin was,reported by Sterrett and Edil
(1982). The forces acting on such a block is shown in Fig.
6. The safety factor is given as
CIB/cos@) + (W cos@) - U - V sin @) )tanol (5)
F p P
W sin@ P + V cos@P

The movement of translational slides is commonly-controlled
structurally by surfaces of weakness, such as faults, joints,
bedding planes, and variations in shear strength between
layers of bedded deposits, or by the contact between firm
bedrock and over lying detritus. It is evident that the
proportions of block slides are controlled largely by the
spacing of the discontinuities which bound the block, and D/L
ratios thus vary widely.
Translational slips can also occur in a homogeneous soil
mass. In particular, granular materials such as sand and
gravel fail in surface ravelling and shallow slides with the
failure surface parallel to the slope surface. Similar
failures occur in a mantle of weathered or colluvial
(granulated) material on clayey slopes and are referred 'as
slab slides. An infinite slope analysis is often
representative of such failures. In this analysis, the slip
surface is assumed to be a plane-parallel to the ground

surface and the end effects can be neglected. With small
ratios of D/L this type of analysis is often appropriate.
The forces acting on a slice of an infinite slope is shown in
Fig. 7. There is no internal distortion and end effects are
neglected. From the consideration of force equilibrium, the
safety factor (F) is obtained explicitly as (Morgenstern and
Sangrey, 1978)

F-"--(C'/Yd)seca coseca + (tan@'/tana)[l-(y'h/yd)sec 2a] (6)
w

The ultimate angle of stability, @u, is obtained by
setting effective cohesion intercept (co) equal to zero
keeping other parametric values the same in Fig. 3. The zero
value for the effective cohesion intercept reflects the
long-term effects of weathering, unloading, and previous mass
failures on the cohesive component of the shear strength on
the face of a natural slope (Skempton, 1964). The straight
line indicates that slope stability for c'=O depends only on
the ground-water pressure and the value of V and is
independent of the slope height. For H w=0 and Hw =H the
values of @u can be computed from Eqn. 6. The values of @ u
for the intermediate values of H are obtained from shallow
w
rotational slips. Influence of ground-water level on @ur
i.e. the shallow slide stability, is shown in Fig. 8. The
values of @u for different @' and H w are given in Table 6.
These values can be used in regrading a slope to a uniform
stable angle. A segment of a bluff at Madigan Beach, Ashland
County, Wisconsin was indeed regraded to a stable inclination
of 2211 to 25'0 based on these values in a demonstration
project (Edil et al. 1979).
Sterrett (1980) reported slab slides with a depth of
about 0.6m from Milwaukee County, Wisconsin. This depth
coincided closely with the depth of desiccation cracking and
soil structure change from fine prismatic peds to massive
intact blocks. The latter was attributed to repeated
freeze-thaw cycles. Sterrett also observed that frozen slabs
of soil measured 0.6 m by 10 m by 13 m failed in early
spring. This failure was attributed to differential melting

of the bluff face. Certain parts of the bluffs melt faster
because of difference in orientation. The upper part of the
bluff face generally melts before the middle and lower parts.
Sterrett believed that melt water seeped into the ground and
exerted hydrostatic pressure behind the frozen slabs of the
Jower parts.

SOLIFLUCTION, FLOW, AND CREEP

Distributed movements within debris are recognized as
flows. Slip surfaces within the moving mass are usually not
visible and the boundary between moving mass and material in
place may be a sharp surface of differential movement. Flows
commonly result from unusually heavy precipitation or from
thaw of snow or frozen soil. The flows observed in the Great
Lakes bluffs take-place mostly in spring and result from
primarily ground thawing and snow/ice melting. Therefore,
they can be classified largely as solifluction. Solifluction
(literally means soil flow) occurs in areas of perennially or
permanently frozen ground and takes many forms involving a
variety of mechanisms. Soliflucion is the downslope movement
of water saturated materials which follows thawing in
previously frozen slopes. Seasonally frozen subsurface
layers of soil prevent percolation of water from upper thawed
layer (termed "active layer"). The active layer becomes
saturated with water from melting of ice lenses within it as
well as from melting snow and rainfall. The effect of excess
water is to fluidize the active layer reducing its strength
and cause it to move downslope by gravity.
The extent of solifluction depends on the grain size of
the slope material, availability of water, depth of frost
penetration, number and duration of freeze-thaw cycles,
inclination of the ground surface, and competence of the
vegetative cover (Embleton and King, 1975). Vegetation
appears to be th e most restraining factor for solifluction.
The size of the flows along the Western Lake Michigan
shoreline varies from 0.3 to 0.6 m wide to 15 to 20 m wide

and 21 m long.
A number of approaches for the analysis of solifluction
failures have been suggested. Chandler (1970) used effective
stress analysis and the mechanism called "ice-blocked
drainage". McRoberts and Morgenstern (1974) also used the
effective stress method along with the-theory of
thaw-consolidation. In both approaches, the residual
strength parameters were used. Hutchinson (1974), on the
other hand, developed a method of using the total stress
method along with the undrained strength of the soil. All
three approaches were based on infinite slope analysis. In
other words, it was assumed that the moving mass at least
started as a rigid and continuous mass sliding on a surface
parallel to the ground surface. The least slope angle at
which the mudflows are mobilized in the field is usually
smaller compared to the least slope angle obtained from the
stability analyses based on the infinite slope approach for
residual strength conditions. Vallejo (1979, 1980a, 1980b,
1981) introduced a new approach to the analysis of
solifluction that reflects the particulate structure of the
flowing mass. The structure of the frozen soils forming
natural slopes in cold regions consists of a reticulate ice
vein network subdividing the frozen soil into irregular
blocks (McRoberts and Morgenstern, 1974). Upon thawing, the
structure will then consist of a mixture of hard pieces of
soil and water. This water can change to a liquid-like soil
slurry after the failure. Vallejo analyzed the stability of
this system of large soil pieces and water, as shown in Fig.
9, using the finding of Bagnold (1954) regarding the movement
of concentrated grains in dispersion (grain flow). The
safety factor with respect to flow is given as

Lcl + (Y Yf) d cos@ tan@l C
F r s r
fyf + (ys yf) C1 d sina
where the terms are as defined in Fig. 9. Vallejo and Edil
(1981) applied this analysis to a coastal bluff in Kewaunee,
Wisconsin with successful field verification. The critical

depth of thaw (normal to the slope face) at which failure
occured was measured to be 0.25 m.
Flows other than solifluction and creep (deformation
under constant stress ) are also evident on the Great Lakes
bluffs. However, these processes appear relatively minor in
comparison to the predominant solifluction in spring and the
associated mass wasting.

RAIN IMPACT AND RILL/SHEET EROSION

Raindrop impact is the dominant factor in the detachment
of soil particles. Sheetwash is the unconfined flow of water
over the ground surface after a rainfall. Depths of flow are
generally only a few millimeters. The majority of rainfall
detachment studies relate total soil loss from a*storm to the
total energy of the storm. At the present, there is no
unified theory which will relate soil loss to raindrop
impact. Nevertheless, grain size, soil structure, and
permeability are the important soil properties controlling
detachability.
Once the particles are detached they transport
downslope. The transport capacity of interrill flow is
primarily a function of runoff rate, slope steepness,
roughness of the surface, transportability of the detached
soil particles, and the effect of raindrop impact (Foster and
Meyer, 1975).
Rills are the concentrated (channelized) flow of water
on a hillslope. Rill erosion takes place in terms of
detachment by flow and transportation by flow. Rill
formation tends to follow the zones of weaknesses on the
slope face. Rill flow, unlike sheet flow, can attain high
enough shear stresses to detach and transport soil particles.
There are only a few reports of sheet/rill erosion on
the Great Lakes bluffs even though it is commonly,observed
along the bluffs. Sterrett (1980) investigated this process
along with slumping and solifluction on a systematic basis at
several sites along Lake Michigan and Lake Superior

shorelines'. Based on field observations, Sterrett concluded
that most of the material removed from the slopes during
summer is via sheetwash and rill erosion. However, it was
not possible to relate soil loss simply to precipitation. He
found the Universal Soil Loss Equation useful in predicting
soil-loss from steep slopes when modified as suggested by
Foster and Wischmeier (1974). The modified equation gives
average soil loss per unit area per unit time, A as
n S. X. 1+M -SA i+m
E 1 1 3 j-1
A = RKCP j=1 A e(72.6)m (8)

where R=rainfall factor, K=soil erodibility factor,
C=cropping management factor, P=erosion-control practice
factor, S =slope factor of the jth segment, X j=slope length
of the jtA segment, X e=entire slope length, m=coefficient of
Variation (about 0.3).

SAPPING

Sapping is the removal of soil particles by seeping
water. This proces is most effective with cohesionless
materials such as silts and sands. Coastal bluffs are often
ground-water discharge areas and seep points and seepage
faces are commonly encountered on the bluffs. Erosion of
cohesionless deposits at the bluff face may lead to the
collapse of overlying cohesive soils when support is removed.
In many areas along the Lake Michigan bluffs in
southeastern Wisconsin relatively impermeable clayey tills or
lake sediments are overlain by more permeable sand or sandy
silt. Following periods of abundant rainfall, perched water
tables may form in these more permeable beds. Lines of
springs will develop where these beds are exposed along the
bluffs. In saturated sand and silt associated with the

springs along the bluffs numerous small failures occur
(Hadley, 1974). There is no known report regarding the
relative contribution of this process to the overall mass
wasting in the Great Lakes bluffs.

WIND EROSION

Wind, in addition to generating waves, may also directly
attack the beach and produce some beach erosion. Beach
sediment is carried landward and accumulates in the form of
dunes. This type of beach erosion takes place when there is IF
a well-developed beach and results in a net gain of sediment
to the coastal zone (Davis et al., 1973). Wind transport of
sand-and-silt-sized material picked from dunes themselves is
increased by the removal of vegetation that. both slows the
wind speed and binds the soil. Wind erosion becomes dominant
in dry areas and on bluffs consisting of cohesionless
materials. Dunes are dominant along the south and eastern
shoreline of Lake Michigan in the Great Lakes and wind
erosion may reach significant levels there. Along bluffs
formed in cohesive soils it does not appear to be very
significant in comparison to other processes. There are not
many reports of significant wind erosion of the Great Lakes
bluffs. Marsh et al. (1973) report wind erosion of 4 in.
sand per winter from A@270 ft high bluff near Grand Marais.

ICE EROSION

During a normal winter season, substantial ice develops
along the shores of the Great Lakes. The long, narrow,
continuous ridges of grounded ice, separated by broad areas
of low-relief ice that parallel much of the shoreline is
called an icefoot (Marsh et al., 1973). The icefoot is of
geomorphic importance, as it protects the shoreline from the
high-energy waves of winter and spring, thereby reducing
rates of erosion which otherwise might be expected. Another

geomorphic influence of lake ice ridges is the scouring of
the lake bottom, especially along sandy shorelines. During
the final phases of icefoot decay, masses of free ice, driven
by strong winds and storm waves can also cause extensive
damage to shore protection structures along the lake shore.
With the opening of-the Great Lakes to winter navigation
other effects such as the liquefaction and flow of nearshore
sand on slopes by wave action of water confined under the ice
may be expected at some places (Wuebben et al., 1978).

DISCUSSION

In the previous section the processes of downslope
movement and the factors that control them were discussed.
In this section an attempt will be made to delineate the
relative importance of each process in bluff retreat and
approaches of past and future research on coastal bluff
processes on Great Lakes shorelines will be discussed.

Objectives of Rese arch

Most research that has been done or will be done has
practical objectives. Although a number of theoretical
concepts have developed in this work, the end product and
therefore research design, has been aimed at solving specific
problems. Two main types of research design are typical
(Wisconsin Coastal Management Program, 1979).
The first category, site-specific studies, have been
undertaken at numerous locations. These are often associated
with structural solutions to shore recession problems. In
these studies an attempt is usually made to identify and
understand slope-stability problems at a single site over a
relatively short period of time. I will later consider this
aspect in more detail because it is apparent from the
literature and from personal observation that there are
definite problems and misconceptions in this area.
The second approach to shore erosion problems is

generally associated with non-structural solutions over a
longer stretch of shoreline. In order to develop zoning
regulations for building setbacks, relocation policy, etc.
units of shoreline the size of a state or at least a county
are usually considered. These studies are usually aimed at
minimizing future losses. In this case the need for
understanding bluff processes is more acute because
predictions of future recession over a long period of time
with changing water-level and climatic conditions is

necessary.
This type of study requires an initial survey of the
shoreline to establish rate of recession (usually with aerial
photographs) and identify problem areas. This should be
followed by a field survey where geologic units, position of
ground water seeps, type of failure, bluff height and bluff
angles are established. This was done in Wisconsin
(Mickelson et al. 1976) as part of a project supported by the
Coastal Zone Management Program. One conclusion of this
project was the realization that predictions of future
recession could not be made without an understanding of the
time scales involved in bluff degradation and slope
adjustment to external change. Another conclusion was that
there was a need to understand the relative importance of the
different processes in removing material from the bluff and
the relative importance of factors leading to downslope
movement. Finally, the need for models of bluff evolution
was recognized. Progress in each of these need areas is
discussed in the following sections.

Time Scale

Different parts of geomorphic systems respond at
different rates to changes in external parameters. In
addition, some bluffs appear to pass through an evolutionary
sequence which is only interrupted when external variables
surpass a threshold level.' In other words, it appears that
once toe erosion initiates an unstable slope, a predictable
progression of failures take place on the bluff irrespective

of water level or the amount of wave erosion (Edil and
Vallejo, 1977; Peters, 1982). The problem with prediction is
understanding response times to environmental changes and to
understand the time necessary for bluffs to pass through an
evolutionary sequence.
Evidence from other areas with evolving slopes such as
river banks and marine coasts (McGreal, 1979; Cambers, 1976;
Hutchinson, 1973) as well as from the Great Lakes (Peters,
1982) suggest that there are possibly three time scales over
which the natural cycles of evolution take place. These
scales are 2-3 years, 50 to 100 years, and thousands of
years.
It is unfortunate that more information in this area is
not available. Without a clearer understanding of the time
scales of change on coastal bluffs predictions of future
change are likely to be misleading or, in fact, incorrect.
In Wisconsin, we are attempting to -understand time
scales by looking at segments of shoreline with uniform
characteristics through time. In some cases these shore
segments may be as short as 100 m or as long as 10 km. If
internal characteristics can be made uniform, we should be
able to examine the effect of changing environmental factors
on bluff evolution.

Relative Importance of Environmental Factors

As discussed in a previous section, water level (because
of its control on toe erosion) is a primary long-term cause
of bluff instability. Given a low water level for long
periods of time (100 years) most slopes along the shoreline
would become nearly stable.
A study by Brandon and Rideout (1980) indicated that the
majority (more than 90%) of the property owners in three
regions along the eastern Lake Michigan shoreline perceived
wave action (and lake level ) as the cause of erosion damage.
Ground-water seepage, ice erosion, and spring than received
30 to 50% of the respondents secondary causes. These answers
are obviously biased by the conditions in that region. The

eastern shore of Lake Michigan has extensive shoreline
consisting of cohesionless dunes which immediately respond'to
changes of water level and wave energy.
In the technical literature practically every report on
the subject cites wave action as a controlling factor on
bluff erosion and there are quite a number of studies that
attempt to document it (Quigley and Gelinas, 1976; Quigley et
al., 1977; Quigley and Tutt, 1968; Gelinas and Quigley, 1973;
Davis et al., 1973; Brater and Ponce-Campos, 1978; Brater et
al., 1974; Edil et al., 1979; Zeman, 1978).
It should be pointed out, however, that along bluffs of
cohesive materials wave erosion is often only the "trigger"
that initiates slope failure. Decrease in water level or toe
protection may control wave erosion but mass wasting on the
bluff normally continues to be a problem for a long period of
time. Relatively large rotational slides have been observed
to take place in areas free of significant wave erosion, for
instance during times of low lake levels (Quigley et al.,
1977). A large slide occured in an area south of Bender
Park, Milwaukee County, Wisconsin in 1979. The toe has been
protected from wave erosion for at least 20 years by the
beach that built up north of the power plant.
There are a significant number of reports that refer to
the role of ground water in coastal bluff slumping (Sterrett
and Edil, 1982; De Young and Brown, 1979; Lee, 1975; Palmer,
1973; Berg and Collinson, 1976; Bird and Armstrong, 1970;
Marsh, et al., 1973; Pincus, 1964). In some of these studies
ground water was singled out as the most critical factor,
acting quite independently from contributing factors. Ground
water is certainly an important factor in determining
stable-slope angles and can cause long term stability
problems. Sterrett (1980) monitored a series of wells near
the bluff edge south of Milwaukee. In this fractured till
water level rose quickly after a precipitation event.
Photographs and profile measurements seem to demonstrate that
movement on slump blocks is directly tied to this rise in
water level. Similar observations were made in Racine County
Coast Watch Program (1981). Slump activity was observed to

be related to substantial precipitation events; storm wave
events alone did not seem to have a direct effect on bluff
recession rather an indirect one.
Mickelson (personal communication, 1982) reports several
localities where slopes are near vertical and stable
where ground water is not present. In identical materials
only 100 m away, where ground-water is present, slopes are at
angles of less than 300 and are actively moving.
Drexhage and Calkin (1981) studied historic bluff
recession along the Lake Ontario shoreline of New York as a
function of nine parameters. They found a strong
relationship between erosion rate and bluff height when the
bluff is higher than 6 M. The rate also correlated with
bluff composition (greater for clayey/silty till than sand
and sandy till) and with bluff slope (higher for inclinations
greater than 45"). It appears that bluff slumping is a
prominent mode of slope retreat in this region.
Vegetation and its role on slope processes and
stabilization is another topic that should be mentioned here.
Much of the information pertinent to this subject is
contained in the Proceeding of the Great Lakes Vegetation
Workshop and in the excellent summary provided by Gray (1977)
in the same volume. It is known that vegetation is
particularly effective in controlling particle movement, e.i.
the sheet/rill erosion and solifluction. Fowle et al (1978)
reported from their study of Scarborough bluffs on Lake
Ontario that vegetation can become established on the bluffs
and progress to forested slope. However, this can only
happen in well-drained areas protected from toe erosion and
where ground-water seepage is controlled. Recession will
continue until these erosive forces and deep slumps are
reduced or eliminated. When this has been done, plants may
well serve an important role in erosion control and check the
recession which now continues apace.

Relative Importance of Slope Processes

There are a few attempts to study a number of factors or

processes irrespective of the causes on a comparative basis
Sterrett (1980) provided a comprehensive and quantitative
view of the bluff processes resulting in slope retreat at
selected sites along the western Lake Michigan shoreline
where high bluffs in glacial till.and related materials are
abundant. The total soil loss per year for two of the sites
monitored for three years and the relative importance of each
process are given in Table 7. Face degradational processes
cause a considerable amount of mass-wasting on a nearly
continuous basis on active slopes lacking vegetation.
A number of investigators have studied slumping because
it is a major and well-defined event (Edil and Vallejo, 1977;
Quigley and Di Nardo, 1978; Berg and Collinson, 1976).
Slumping, unlike face degradational processes, is a discreet
event and involves large volumes of material. McGreal (1979)
designates slumping as the most significant mass wasting
process in coastal cliff recession in Ireland.

Potential of Developing Slope Evolution Models

From the review of the literature it is apparent that
bluffs or dunes composed of granular materials respond to
wave action and other erosive processes directly and tend to
have parallel retreat of the bluff face without deep slips
and time delays. Therefore, modelling their behavior
qualitatively and even quantitatively is an achievable task
(Peters, 1982) unlike the clay bluffs where possible deep
slips, delayed response and other factors make modelling
rather difficult. Hutchinson (1973) described a cyclic
response model of coastal slopes in England to toe erosion.
Quigley et al. (1977) and Edil and Vallejo (1977) described
qualitative models for bluffs studied on Lake Erie and Lake
Michigan, respectively. Peters (1982) building on research
and data base of Vallejo (1977) and Sterrett (1980) proposed
evolutionary models for bluffs at five carefully selected
sites along western Lake Michigan shoreline and provided
predictions for their future for different lake levels.
Quantitative models, or even generalizations of existing

qualitative models of coastal bluff recession are not
available at this time.

CONCLUSIONS

Based on the review of the literature available to the
author and his own experiences the following conclusions are
advanced:

A. Structural (Stabilization) Approach

1. The tools for the analysis, design and construction of
structural solutions on a site-specific basis are
currently available. The problems associated with the
execution of this category of solutions seem to be of
two types: (a) many attempts are not engineered and
fail to cope with the problems and (b) those
engineered solutions quite often neglect to consider
all aspects of the problem.

2. A stabilization program should include the follwing
steps:
a) Protection against wave action: this may include
shore protection structures such as groins,
seawall, breakwaters, etc. and/or beach building
(nourishment) or a natural drop in lake levels.
b) Stabilization against deep slips: a bluff may have
a stable or unstable profile at the time of
stabilization. This should be verified by a
geotechnical analysis. If not safe against a deep
slip in the long-term bluff should be stabilized.
Methods include regrading to a stable angle,
toe-loading with a berm, lowering of ground water
and attendant pore pressures, controlling and
intercepting surface and ground water. If the
bluff is safe against a deep slip proceed to step
(c).

c) Stabilization against face degradation: this may be
achieved by flattening the slope to the ultimate
angle of stability plus vegetation or allowing a
steeper angle than this but less than the angle to
initiate a deep slip plus occasional maintenance
after shallow slips. Vegetation is also needed.

3. Additional research is needed primarily in the
technology and development of economic ways of
stabilization in all aspects of slope failure.

B. Nonstructural Approach (Planning and Management)

Our technical understanding of coastal recession over a
long period of time, say 30 to 50 years, appears quite
limited for quantitative predictions. Tools for such an
analysis even on a site-specific basis, are not
well-established. However, this does not mean that there has
not been any progress. Research conducted primarily during
the last decade or two has identified the operating processes
and their possible magnitudes. We have at least a
qualitative appreciation of the factors and the processes and
some conceptual models of the interrelationship of these.

FURTHER RESEARCH NEEDS

Models of long-term erosion/recession processes need to
be established. This will require introduction of
probabilistic modelling and long-term monitoring at least at
selected sites. Schultz (1980) applied the probabilistic
assessment of slumping potential to southwestern shoreline of
Lake Superior on a reach by reach basis. This sort of
approach should be expanded to include the potential for the
other significant mass wasting processes.
Systematic monitoring is the single most important
recommendation for further research. Monitoring of sites for
a few years will not provide the data basis needed for the

formulation of long-term trends. Such a program should
obtain quantitative data on the causative factors discussed
and the response of the shoreline in terms of geomorphic
changes.

ACKNOWLEDGEMENTS

The author wishes to acknowledge the contributions of
his colleague Dr. David M. Mickelson, to his understanding of
the coastal bluff slumping problem over the years. He also
reviewed this manuscript. Many former students at the
University of Wisconsin-Madison contributed to the shore
erosion research. Among them Drs. L. E. Vallejo and R.J.
Sterrett and Messrs. B.J. Haas, M.E. Schultz, A. Bagchi, R.
De Groot, and C. Peters are acknowledged. The financial
support provided by the National Oceanic and Atmospheric
Administration, U.S., Department of Commerce through the
Wisconsin Sea Grant Program and the Wisconsin Coastal Zone
Management Program over the years supported the research.
Finally, the author wishes to acknowledge the sponsorship of
this paper by the Michigan Sea Grant Program as part of a
project supported by the U.S. Geological Survey. Miss Nermin
Demircioglu's effort and patience in rapid typing and
processing of the manuscript is appreciated.

REFERENCES

Bagnold, R.A. (1954), "Experiments on a Gravity-Free
Dispersion of Large Solid Spheres in a Newtonian Fluid Under
Shearl" Proc. Royal Soc., Series A 225, pp. 49-63.

Berg, R.C. and Collinson, C. (1976), "Bluff Erosion,
Recession Rates, and Volumetric Losses on the Lake Michigan
Shore in Illinois," Illinois State Geological Survey.,,
Environmental Geology Notes, No. 76.

Bird, S.J.G. and Armstrong, J.L. (1970), "Scarborough
Bluffs-A Recessional Study," Proc. 13th Conf. Great Lakes
Res., pp. 187-197.

Bishop, A.W. (1955), "The Use of the Slip Circle in the
Stability Analysis of Slopes," Geotechnique, Vol. 5, No. 1,
pp. 7-17.

Braden, P.L. and Rideout, S.R. (1980), "Consumer Investment
in Shore Protection," Michigan Sea Grant Program,
MICHU-SG-80-200.

Brater, E.F. (1975), "Beach Erosion in Michigan: An
Historical Review," University of Michigan.

Brater, E.F., Armstrong, J.M. and McGill, M. (1979),
"Michigan's Demonstration Erosion Control Program, Evaluation
Report," Michigan Department of Natural Resources.

Brater E.F. and Ponce-Campos, C.D. (1978), "Coastal
Engineering and Erosion Protection: Report for the Year
1976-77," Michigan Sea Grant Program, MICHU-SG-78-201.

Bretschneider, C.L. (1958), "Revisions in Wave Forecasting:
Deep and Shallow Water," Proc. 6th Conf. Coastal Engineering,
Engineering Foundation.

Cambers, C. (1976), "Temporal Scales in Coastal Erosion
Systems," Trans. Inst. British Geogr. New Ser. Vol. 1, No.
2, pp. 246-256.

Carson, M.A. and Kirkby, M.J. (1972), Hillslop@@ Form and
Process, Cambridge university Press, London.

Carter, C.H. (1976), "Lake Erie Shore Erosion, Lake County,
Ohio: Setting, Processes, and Recession Rates from 1876 to
1973,11 Ohio Geol. Survey Rept. Investigations No. 99.

Chandler, R.J. (1970), "Solifluction on Low-Angled Slopes in
Northamptonshire," Q.J. of Eng. Geol., Vol. 3, pp. 65-69.

Davis, Jr., R.A. Seibel, E. and Fox, W.T. (1973), "Coastal
Erosion in Eastern Lake Michigan-Causes and Effects," Proc.
16th Conf. Great Lakes Res., pp. 404-412.

DeYoung, B. and Brown, L. (1979), "Controlling Bluff
Groundwater Along the Great Lakes," New York Sea Grant
Program, Page 102.00.

Drexhage, T. and Calkin P.E. (1981), "Historic Bluff
Recession Along the Lake Ontario Coast, New York," New York
Sea Grant Institute Albany, New York.

Dobson,R.S. (1967), "Some Applications of a Digital Computer
to Hydraulic Engineering Problems," Stanford Univ., Dept.
Civil Eng., Tech, Rep., No. 80, 172 p.

Edil, T.B. and Haas, B.J. (1980), "Proposed Criteria for
Interpreting-Stability of Lakeshore Bluffs," Engineering
Geology, No. 16, pp. 97-110.

Edil, T.B.-, Monkmeyer, P.L., Becker, N.M., Shands, J.A and
Wolf, P.R. (1979), "Evaluation of Shore Protection
Demonstrations at Madigan Beach and Madeline Island,
Wisconsin," In: The Red Clay Project Final Report.
Environmental Protection Agency.

Edil, T.B. and Vallejo, L.E. (1977), "Shoreline Erosion and
Landslides in the Great Lakes," Proc. 9th Int. Conf. Soil
Mech. and Foundation Eng., Tokyo, Vol. 2, pp. 51-57.

Edil, T.B., and Vallejo L.E. (1979), "Mechanics of Coastal
L'andslides and the Influence of Slope Parameters,"
Engineering Geology, No. 126 pp. 83-96.

Embleton, C. and King, C.A.M. (1975), Glacial and Periglacial
Geomorphology, Edward Arnold, London.

Environment Canada (1975), Canada-Ontario Great Lakes Shore
Damage Survey Technical Report.

Fisher J.S., Gulbrandsen L.F., and Goldsmith V. (1975),
"Analytical Prediction of Shoreline Erosion," Proc. Ocean
Engineering III ASCE Specialty Conference Newark, Delaware.

Foster, G.R. and Meyer, L.D. (1975), "Mathematical Simulation
of Upland Erosion by Fundamental Erosion Mechanics," In:
Present and Prospective Technology for Predicting Sediment
Yields and Sources, U.S. Department of Agriculture,
Agricultural Research Service, ARS-S-40, pp.-190-206.

Foster, G.R. and Wischmeier, W.H. (1974), "Evaluating
Irregular Slopes for Soil Loss Prediction," Transactions,
American Society of Agricultural Engineers, Vol. 17, pp.
305-309.

Fowle, C.D., Collishaw, G. and Lewis, M.F. (1978),
"Vegetation and Erosion on Scarborough Bluffs, Lake Ontario,"
Proc. 2nd Workshop on Great Lakes Coastal Erosion and
Sedimentation, Burlington, Ontario, pp. 33-36.

Gelcij R., Cazale, H. and Vassal, J. (1957), "The
Spectroangular Method: Forecasting Ocean Waves," Min. Trav.
Publics Transports, Meteorol. Natl. (Paris), Notice Inform.
Tech., Vol. 22, no. 3.

Gelinas, P.j. and Quigley, R.W. (1973), " The Influence of
Geology on Erosion Rates Along the North Shore of Lake Erie,"
Prod. 16th Conf. Great Lakes Res., pp. 421-430.

Gray, D.H. (1977) "The Influence of. Vegetation on Slope
Processes in the Great Lakes Region," Prod. Workshop on the
Role of Vegetation in Stabilization of the Great Lakes
Shoreline, Great Lakes Basin Commission, Ann Arbor, Michigan.

Great Lakes Basin Commission (1981), Coastal Hazards Plan and
.Environmental Impact Statement, Ann Arbor,Michigan.

Great Lakes Basin Framework Study (1975), Appendix 12.

Hadley, D.W. (1976), "Shoreline Erosion in Southeastern
Wisconsin," Geol. and Natural History Survey Report,
University of Wisconsin-Extension, Madison, Wisconsin.

Hutchinson, J.N. (1973), "The Response of London Clays to
Differing Rates of Toe Erosion," Prod. Geologica Applicata E
Idrogeologica, Vol. 8, Part. 1, Naples, Italy.

Hutchinson, J.N. (1974). "Periglacial Solifluction: an
Approximate Mechanism for Clayey Soils. Geotechnique, Vol.
24, pp. 438-443.

International Joint Commission, (1973), Regulation of Great
Lakes Water Levels, Appendix C: Shore Property.

Johnson, C.N. and Hiipakka, L.W. (1976), "Sand-Bypass and
Shore Erosion, Bridgeman, Michigan," Prod. 15th Coastal
Engineering Conference, Honolulu, Hawaii . 856-858.
11 pp

Keillor, J.P. (1978), "Recent Recession Of Lake Michigan
Shorelines in Racine County," Wisconsin Sea Grant Program
Advisory Services, Special Report No. 507.

Lee " K.K. (1975), "Bluff Recession in Lake Michigan Shoreline
Erosion," J. Soil and Water Conservation, pp. 138-139. -

Marsh, W.M., Marsh, BD. and Dozier, J. (1973), "Formation,
Structure, and Geomorphic Influence of Lake Superior
lcefootg," American J. Science, Vol. 273,, pp. 48-64..

McRoberts, E.C. and Morgenstern, N.R. (1974), "The Stability
of Thawing Slopes," Can. Geotech. J., Vol. 11, No. 4, pp.
447-469.

McGreal, W.S. (1979) "Cliffline Recession Near Kilkeel N.
Ireland: An Example of a Dynamic Coastal System," Geogr. Ann.
Vol. 61A, No. 3-4, pp. 211-219.

Mickelson, D.M. (1982), Personal communication.
Mickelson, D.M., Acomb, L.J'. and Edil, T.B. (1979), "The
Origin of Preconsolidated and Normally Consolidated Tills in
Eastern Wisconsin, U.S.A.," In: Moraines and Varves, Balkema,
Rotterdam, pp. 54-63.

Morgenstern N.R. and Sangrey D.A. (1978), "Methods of
Stability Analysis," In: Landslides, Ed. R.L. Schuster and
R.J. Krizek, Special Report 176, Transportation Research
Board, Washington, D.C. pp. 155-171.

Palmer, H.D. (1973), "Shoreline Erosion in Upper Chesapeake
Bay: the Role of Grounwater," Shore and Beach, pp. 19-22.

Peters, C. (1982),"'Models of Short and Long-Term Geomorphic
Evolution of Bluffs in the Great Lakes," M.S. Thesis, Dept.
Geology and Geophysics, Univ. of wisconsin, Madison.

Philips, B.A.M. (1978), "Coastal Erosion, North and East
Shores of Lake Superior," Proc. 2nd Workshop on Great Lakes
Coastal Erosion and Sedimentation, Burlington, Ontario,
pp.37-40.

Pierson, Jr., W.J., Neumann, G. and James, R.W. (1955),
"Practical Methods for Observing and Forecasting Ocean Waves
by means of Wave Spectra and Statistics," U.S. Naval
Hydrograph, Office Publ., 603.

Pincus, H.J. (1964), "Retreat of Lakeshore Bluffs," J.
Waterways and Harbors Division,, ASCE, Vol. 90, No. WW1, pp.
115-134.

Quigley, R.M. and Di Nardo, L.R. (1978), "Cyclic Instability
Modes of Eroding Clay Bluffs Lake Erie Northshore Bluffs at
Port Bruce, Ontario, Canada," Proc. AAG-IGU Coastal Symp. New
Orleans, Louisiana.

Quigley, R.M. and Gelinas, P.J. (1976), "Soil Mechanics
Aspects of Shoreline Erosion," Geoscience, Canada, Vol. 3,
No. 3, pp. 169-173.

Oulgley, R.M., Gelinas, P.J., Bou, W.T. and Parker, R.W.
(1977), "Cyclic Erosion-Instability Relationship, Lake Erie
North Shore Bluffs," Can. Geotech. J., Vol. 14, pp.310-323.

Quigley, R.M. and Tutt, D.B. (1968), "Stability-Lake Erie
North Shore Bluffs," Proc. llth Conf. Great Lakes Res., pp.
230-238.

Racine County Coast Watch Program (1981), Final Report,
Racine County Planning and zoning Department, Wisconsin.

Resio, D.T. and Vincent, C.L. (1976), "Estimation of Winds
Over the Great Lakes," Misc. Paper H-76-12, U.S. Army
Engineers Waterways Experiment Station, Vicksburg,
Missisippi.

Schultz, M.N. (1980), "Probabilistic Stability Analysis of
Slopes Along the Southwestern Lake Superior S@oreline,"-M.S.
Independent Study Report, Dept. Civil and Environmental
Engineering, Univ. of Wisconsin, Madison, Wisconsin.

Seibel, E. (1972), "Shore Erosion at Selected Sites Along
Lakes Michigan and Huron," Ph.D. Thesis, Dept. Geology, Univ.
of Michigan, Ann Arbor, Michigan.

Sevaldson, R.A. (1956), "The Slide in Lodalen, Oct. 6. 1954,"
Geotechnique, Vol. 6, pp. 167-182.

Sharpe, S.F.S. (1938), Landslides and Related Phenomena,
Colombia Univ. Press, New York.

Skempton, A.W. (1953), Discussion in "Soil Stability Problems
in Road Engineering," Proc. Inst. Civ. Eng. Vol. 2, No. 2,
pp. 265-268.

Skempton, A.W. (1964), "Long-Term Stability of Clay Slopes,"
Geotechnique, No. 14, pp. 77-101.

Skempton, A.W. and DeLory, F.A. (1957), "Stability of Natural
Slopes in London Clay," Proc. 4th Int. Conf. Soil Mech. and
Foundation Eng., London, Vol. 2, Pp. 378-81.

Skempton, A.W. and Hutchinson, J.N. (1969), "Stability of
Natural Slopes and Embankment Foundations," State-of-the-Art
Report,'Proc. 7th Int. Conf. Soil Mech. and Foundation Eng.,
Mexico, pp. 291-334.

Sterrett R.J. (1980), "Factors and Mechanics of Bluff Erosion
on Wisconsin's Great Lakes Shorelines," Ph.D. Thesis, Dept.
Geology and Geophysics, Univ. of Wisconsin, Madison,
Wisconsin.

Sterrett, R.J. and Edil, T.B. (1982), "Ground-Water Flow
Systems and Stability of a Slope," Ground Water, Vol. 20, No.
1, pp. 5-11.

Sverdrup, H.U. and Munk, W.H. (1947), "Wind, Sea, and Swell:
Theory of Relations for Forecasting," U.S. Naval Hydrograph,
Office Publ. 601.

Tovell, W.M. (1977), "Not Ours to Control," Rotunda, Vol. 10,
No. 3.

U.S. Army, Corps of Engineers (1971), Great Lakes Region
Inventory Report National Shoreline Study.

Vallejo L.E. (1977), "Mechanics of the Stability and
Development of the Great Lakes Coastal Bluffs," Ph.D. Thesis,
Dept. Civil and Environmental Eng. Univ of Wisconsin,
Madison, Wisconsin.

Vallejo, L.E. (1979), " An Explanation for Mudflows,"
Geotechnique, Vol. 29, No. 3, pp. 351-354.

Vallejo, L.E. (1980a), "Mechanics of Mudflow Mobilization on
Low Angled Clay Slopes," Engineering Geology, No. 16, pp.
63-70.

Vallejo, L.E. (1980b), "A New Approach to the Stability
Analysis of Thawing Slopes," Can. Geotech. J., Vol. 17, pp.
607-612.

Vallejo, L.E. (1981), "Stability Analysis of Mud Flows on
Natural Slopes," Proc. loth Int. Conf. Soil Mech. and
Foundation Eng., Vol. 3, Stockholm, pp. 541-544.

Vallejo, L.E. and Edil T.B. (1979), "Design Charts for
Development and Stability of Evolving Slopes," J. Civil Eng.
Design, Vol. 1, No. 3, pp. 231-252.

Vallejo, L.E. and Edil T.B. (1981), "Stability of Thawing
Slopes: Field and Theoretical Investigations," Proc. loth
Int. Conf. Soil Mech. and Foundation Eng., Vol. 3, Stockholm,
pp. 545-548.

Varnes, D.J. (1958), "Landslide Types and Processes," In:
Landslides and Engineering Practice, Ed. E.B. Eckel,-H.R.B.
Special Report 29, Washington D.C. pp. 20-47.

Varnes, D.J. (1978), "Slope Movement Types and Processes,"
In: Landslides, Ed. R.L. Schuster and R.J. Krizek, Special
Report 176, Transportation Research Board, Washington D.C.

Wisconsin Coastal Management Program (1979), Wisconsin@s
Shore Erosion Plan: an Appraisal of Options and Strategies by
R. Springman and S.M. Born, Madison, Wisconsin..

Wuebben, J.L., Alger, G.R. and Hodek, R.J. (1978), "Ice and
Navigation Related Sedimentation," Proc. IAHR, Symp. on Ice
Problems, Lulea, Sweden, pp. 393-403.

Zeman, A.J., (1978), nNatural and Man-made Erosion Problems
Along the Port Burwell to Long Point Shoreline," Proc. 2nd
Workshop on Great Lakes Coastal Erosion and Sedimentation,
Burlington, Ontario, pp. 49-52.

TABLE 1

THE GREAT LAKES SHORELINE, DESCRIPTION OWNERSHIP AND USE, 1970

United States (a) Canada (b
Great Lakes Shoreline Total Miles Total Miles

1. PHYSICAL CHARACTERISTICS

With a beach zone 23@107 5,306
Without a beach zone 1,572 981

Total' 3,679 6,287

2. USE

Residential 1,216 1,261
Commercial and Industrial 189 329
Agricultural and Undeveloped 633 695
Forest 1,159 3,396
Recreation 365 357
Public Building and Related Lands 60 99
Fish and Wildlife Wetlands 57 148

Total 3,679 6,286

3. OWNERSHIP

Federal 133 374
Non-Federal Public 517 2,378
Private 3,029 3,535
Total 3,679 6,287

4. PROBLEM-IDENTIFICATION
Non-'Eroding 1,704 839
Significant Erosion
Critical 214 320
Non-Critical 1,046 4,907
Subject to Flooding 335 72
Protected 380 149

Total 3,679 6,288

5. TOTAL SHORELINE MILEAGE 3,679 6,288

(a) Source: Department of the Army, Corps of Engineers, North Central
Division, Great Lakes Regional Inventory Report National
Shoreline Study, August 1971, does not include islands and
connecting rivers.
(b) Source: 1966 Field Surveys, Department of Public Works, Canada,
includes Canadian national reach to Trois Kivieres.

1 mile = 1.61 km 39

Table 2. SHORETYPES (USA)

ERODIBLE BLUFFS 1188 MILES 32/0
ERODIBLE LOW PLAINES 623 17%
SAND DUNES 325 9%
NONERODIBLE 904 'v 24%
OTHER (FILLS, WETLAUDS, 675 18%
ETC.)
TOTAL 3715 MILES 100%

Table 3- FORCES, RESISTANCES AND SLOPE PROCESSES

GRAVITY MASS MOVEMENT
SLIDING
ROTATIONAL: "SLUMPS
TRANSLATIONAL: -BLOGK"SLIDE
VIBRATIONS SLAB SLIDE
FLOW C/)
C/)
rn
SOLIFLUCTION cn
CLIMATE DEBRIS FLOW

CREEP
v) SHEAR STRENGTH
PARTICLE MOMENT

WAVE EROSION
2 VEGETATION
01 WIND EROSION
ICE EROSION
STRUCTURAL SYSTEMS RILL/SHEET EROSION
SAPPING

Table 4. Abbreviated Classification of Slope Movements (from Varnes, 1978)

TYPE OF MATERIAL
TYPE OF MOVEMENT ENGINEERING SOILS
BEDROCK Predominantaly Predominantly
Course rine

FALLS FTock fall Debris fall Earth fall
TOPPLES Rock topple Debris topple Earth Topple

ROTATIONAL FEW Rock slump Debris slumD Earth slump
UNITS -
SLIDES Rock block slide Debris block slide Earth block slide

TRANSLATIONAL MANY Rock slide Debris slide Earth slide
I UNITS I i Earth spread
LATERAL SPREADS I R ck spread Debris wread
FLOWS Rock flow Debris flow I Earth Flow
I(deep creep) (sail creep)
COMPLEX Cn i on of t principal types of movement

Table 5. Summary of Great Lakes Coastal Soil Properties

Effective
Unit Strength
Weight Parameters
Location Wl/m 2 c'(kN/m2 0'(Deg.) Source

Lake Erie
Geography Field Stn. 22.4 35.0 26.0 Quigley & Tutt (1968)
21.6 24.0 28.0
21.5 17.5 28.0
Patrick Point 20.9 10.0 26.0 Quigley et al (1977)
21.8 23.0 25.0

Iona 20.4 6.0 26.0
20.9 0 30.0
21.8 23.0 25.0
Pumping Station 20.4 15.0 28.0
21.3 9.0 28.0
21.8 14.0 34.0

Lake Michigan
Till 1A 19.6 0 34.6 Mickelson et al (1979)
Till 2A 18.6�0.6 0 31.1A.1 10
28 18.5�0.6 0 31.4-
2U 18.7 a 30.5
Till Ozaukee 17.9�0.4 0 31.4�0.8
Haven 18.6�0.9 23.8�5.6 31.2�0-5
Vaiders 17.7�1.2 28.3�6.9 29.3�0.6
Glacio-Lacustrine Clays 4 to 60 26 to 29 Edil & Haas (1980)

Lake Superior
Till Douglas Creek 18.2�0.5 50.0 26.4�3.0 Schultz (1980)
Hanson Creek 19.0�0.2 50.0 28.0
Jardine Creek 19.6 0 40.0
Madigan Beach Till-I 21.4 76.6 19.0 Edil et al (1979)
Madigan Beach Till-2 19.0 0 21.0
Silty Sands, Sands 18.8-21.2 0 31-38 Schultz (1980)

Table Ultimate Angle of Stability
(in degrees)

0 11"al
(0) 0 1/4 1/2 3/4 1

25 25 21.5 16.5 10.5 13
30 30 28.5 18.5 16 16
35 35 34 24 21 19

Table 7. ESTIMATES OF AMOUNT OF MATERIALS
REMOVED FROM BLUFFS BY DIFFERENT
PROCESSES (AFTER STERRETT, 1980)

TOTAL
ANNUAL RILL
SOIL EROSION/
LOSS SLUMP- SOIL- SHEET-
3
&-a PROFILE --EI ING FLUCTION- WASH
BENDER 1 452 3% 93% 40/70
PARK 001 97
2 470 11@ 91% /a

C7
3 1167 6 'u, % 35% 5yo

PORT 1 774 6 8 23% 9%
WASH. r1m
2 430 3% 90/0 1%
3 381 0% 70% 21%

Figure 1
Hydrographs of Great Lakes Water Levels.

DATUM IGLD 1955
62 71 F19 -2 11973
184 1952! 1953'119541.19551 195611957 _195811959,196 6TI19' 1 196311964 11965FI`966TI967TI@681 1969! 197o! 19
602

601
600
183 - r- -
CHART DrL#,4 600fr, 1829in M
599
w LAKE SUPERI F. :
R M

581
560 Z
177 0

579 2
578 -11
Cr 176
IU 2 M
-t- 577 --J M
CHART (TATUM 5760 FY 175 Bin M
576
A
515 > co
-'HURON
0
5 6 Z
UJ
> 575
LLJ cn M
-j w 574
175 Z
4b. cc < 573 M
W w
572
114 r- M
w CHART DATUM 5;N7FT 174 2-, 571 >
z cl 570 71
M
Uj LAKE ST CLAIR CA M
<
51`3 0
U) z -> r
4
> 572 c >
CO < 571
7
< Z
cl
570
(D
cf) 174 'Ie 4 - x
Ln
w 569 LIT M
cc L
33
173 568
w CHART L)ATU1.4 568 6F T 1733-:
LAKE ERIE
1 9
z 246 z

L 247
z A- 4 -4
0
- 75 - 246
I'- M
C
> I A- 245 ca
w M
-j - 244 0
w
74 -
243
CHARfoAlUM 2A28FI 740.
242
LAKE ONTARIO 241
1952 1953! 1954 0 1 1961 1962119( 16
I Li= "i7-12!n I i 1972 1973 L
t@52 1953,,19541

J\-V

(from Environment Canada, 1975)

C., -@s

/hr

Ion 0

Figure 2. Forces Involved in Effective Stress
Slip Circle Analysis

90
80- o'-250
Cm hw=1/4H
70- 'Y - 21 KN/M3
ZONE A
60-
LLJ 50- C'(KN/ M2)
1 \41
Q 40 50
z ZONE 6 30
< 30
LLJ 10
0- 20- -
0
-j ZONE C
10-
01 1 1 1
10 20 30 40 50 60
SLOPE HEIGHT H (m)

Figure 3. Stability and Slope Develpoment Chart
C
- Y-Vz,
r

LA Aj

L
D

SECTION A -A

L MAXIMUM LENGTH OF SLIDE: UP SLOPE
D MAXIMUM THICKNESS OF SLIDE
B MAXIMUM BREADTH OF SLIDE

Figure 4. Landslide Proportions (after Skempton
and Hutchinson,1969)

FC7 25 k1l, 21 kN/ 2
90

80-
35- H.-I H
3 4
70- 25,

66 -

50-

40-

2n
10 20 30 40 50 60
80

7
0 35- H
3 2
0
60- 23*

cz 50-
Uj
-j 40-
z
41 30-

Uj
CL 20-
0
U) 10
0 10 20 30 40 50 60
70 1 1 1
60-0'- 35' -
30'
50- 25*

40-

30-

20-

V 10

0
0 10 20 30 40 50 60

SLOPE HEIGHT H (m)

Fig.5. Influence of effective friction angle on stability for different water levels.

W= Weight of block
Z U= Water force on bottom of block
V= Water force on side of block

Z= Depth of jointing
W v Zw= Depth of water in joint

P
Figure 6. Failure Block Analysis

E+dE
x
W X+dX
d17 h
E

-::z S

Figure 7. Forces Acting on an idealized Slice of
Infinite Slope
U

c 0
t 40@- 0, -30*
y, - 21 WM3

CL
0
20-
2

16.
4

0 10 20 30 40 50 60

SLOPE HEIGHT H (rn)
Fig.8. Influence of water level on Ultimate angle of stability for a uniform slope.

fluid: water or mud UNIT WEIGHT OF FLUID
Y
f
Soil lump Y UNIT WEIGHT OF SOIL LUMPS
S
MP S
C VOLUME RATIO OF LUI

RESIDUAL STRENGTH PARAMETERS
r

Figure 9. Solifluction

METFK:)DS OF PREVENTING BLUFF RECESSION

prepared by

Ernest F. Brater

Department of Civil Engineering
University of Michigan
Ann Arbor, Michigan

Methods of Preventing Bluff Recession
Introduction E. F. Brater

The principle objective of - this paper is to outline and describe methods of
preventing bluff recession. However, some preliminary discussion of shore processes is
necessary to provide a better understanding of the function of the various protective
measures. Of the more or less synonymous terms, shore erosion, beach erosion and
bluff recession, the latter provides the most graphic description of the damaging phase
of shore processes. Recession is started by the action of storm waves on the toe of
the bluff. The material loosened by the waves is kept in suspension by the violent
turbulence and carried away by the littoral currents created by the waves and the
wind. The action of the waves leaves the lower face of the bluff in a potentially
unstable condition. The bluff will eventually slump to a more stable slope. The
slumping may occur while the storm is still in progress or the bluff may retain its
steep slope for many months. The energy which causes the erosion is provided primarily
by the wind.

The relative importance of the various factors are illustrated in Fig. I by the
size of the blocks. A very important factor is the water level. During high levels
the waves can more easily reach the bluff whereas during lower levels it may require
a major storm for the waves to reach the bluff. On the lower Great Lakes, long term
lake levels vary over a range of about 6.5 feet due to changes in long term precipitation.
Shorter term changes in level are produced by wind, and to a lesser extent by local
changes in barometric pressure. The effect of long term water surface elevations on
erosion is illustrated by measurements at many locations along a 3-mile reach of Lake
Michigan during three decades from 1938 through 1970. (Brater and Siebel, 1973).The
average recession rates varied from less than 2 feet per year during periods of low
levels to 8 feet per year during high levels.

In assessing the effectiveness of protective measures an important factor is the
irregular occurrence of major storms. Thi6 is illustrated in Fig. 2 which shows. all
storms having waves of 5 feet or more in height on the east coast of Michigan's lower
peninsula over a 46-year period. During I or 6 year periods no major storms occurred
and while in other periods of the same length, many storms occurred. The wave heights
shown in Fig. 2 are the significant wave heights which are the average of the largest
one-third of a group of waves. Fig. 2 also shows the annual lake levels during the 46
years. The effectiveness and durability of a protective measure obviously cannot be
judged until its life has extended into a period which includes major storms or above-
average lake levels.

Preventing Bluff Recession

The intensive settlement and use of the coastal areas of the world without
consideration of shore processes has produced vast losses of property and of the natural
beauty of the shore line. Efforts to prevent damage to structures in vulnerable bluff
areas has produced an unbelievably large variety of devices. Most of these are ugly
and often neither durable nor effective. Some do more harm than good. The forces
produced by waves in vulnerable areas are so great that the cost of complete protection

may exceed the value of the property. The use of lower cost procedures which will
protect against the more frequent moderate storms and reduce the damage due to
major storms may be economically feasible if protection is sufficiently well designed
and well constructed to resist destruction by major storms. The following outline lists
procedures for preventing shore damage according to their basic function.

Preventative Methods,

1. Vacating vulnerable shore areas.

2. Creating a beach.
a) Artificial nourishment.
b) Artificial nourishment with groins.
c) Grain systems.

3. Protecting the toe of the bluff from waves.
a) Sea walls.
b) Revetments.
c) Breakwaters.

Vacat6 Vulnerable Shore Areas

Were it not for this first item in the outline of "Preventative Methods" the
heading could have been "Protective Methods". Shore erosion and accretion are natural
processes which cannot be easily modified. Vacating vulnerable property takes this
fact into consideration. The method has the great advantage of retaining the natural
beauty of the shore line while making it available for non-damaging recreational uses.
In many locations where buildings, roads, or other structures have already been built
in vulnerable shore areas the most economical solution may be to move the structures
away from the edge of the bluff. In locations that are still natural, building can often
be prevented by zoning or by making potential users aware of the rate of bluff recession.

Creating A Beach

A beach is one of the most efficient wave energy absorbers. Much of the energy
is lost in the breaker zone and most of the remainder is lost when waves run up on the
flat beach. Very little energy is reflected from the beach. Therefore, one of the
most basic methods of protecting the toe of a bluff is to cause the waves to break on
a beach well away from the bluff. The value of a protective beach is clearly
demonstrated by the decrease in bluff recession rate during low water periods on the
Great Lakes and by the rapid recession rates in areas where the beach is starved of
its normal littoral drift by the presence of a jetty or inlet. This method also has the
advantage of creating improved recreational conditions.

Artificial Nourishment - From an aesthetic point of view this is by far the best
method of protection. However, the lack of sufficient beach material at the erosion
site indicates that there is not sufficient natural littoral drift to maintain the beach.
Therefore, the artificial nourishment will be gradually, or perhaps rapidly, moved away
and then will have to be replaced.

Artificial Nourishment Plus Groin s - Groins are low walls extending seaward from
the bluff. Groins form pockets which tend to retain the sand. They are very effective
even in areas of high wave energy. An advantage of groin systems over structures
built parallel to the shore is that they interfere very little with the recreational use of
the beach.

Groin Systems - Groin systems, without artificial fill, are very effective in areas
having considerable littoral drift. They will-often fill with sand during the first major
storm. After they are filled the littoral drift will pass by as it did before they were
built. However, during the time they are filling the downdrift shore area will not
receive its normal littoral drift. If this appears to be a serious problem, it is necessary
to fill the groins artificially.

Protecting the Toe of the Bluff

There are many locations where the structures cannot be moved back or where it
is not feasible or even desirable to create a beach. In such locations bluff recession
can be stopped only by preventing waves from attacking the toe of the bluff. When
the toe is protected, the top of the bluff will continue to recede only until the face of
the bluff has attained a stable slope.

Sea Walls - Sea walls are structures built parallel to shore usually of a solid
material such as wood, steel, concrete or asphalt. More massive constructions are
sand-filled tubes or bags, rock-filled wire cages or rubble mounds. Sea walls are built
more often than any other method of shore protection. This is unfortunate because
most sea walls tend to increase the erosion rate at the toe of the wall and, unless they
are well designed and securely constructed, they fail more easily than other procedures.
The increased erosion is due to the violent turbulence and strong littoral currents
created as t1he waves strike and reflect from the smooth wall. The most common cause
of failure is tipping or sliding lakeward due to back pressure from saturated soil.
Vertical walls are the worst, but many of the same problems occur to some extent
with sloping impermeable walls. Sea walls should only be used when no other method
is feasible.

Revetments - Revetments are layers of rubble placed at the toe of the bluff
after the bluff has been graded to a reasonably stable slope. A well designed and well
built revetment is one of the very best methods of toe protection. One advantage of
a revetment is that it tends to absorb rather than reflect the wave energy and therefore
it does not accelerate erosion at the toe. Another advantage is that a revetment
inhibits wave run-up and therefore does not have to be built as high as a smooth. wall.
Our experience has shown that it is possible to underdesign a revetment, to keep costs
low, without as much danger of destruction during major storms as is true of other
forms of protection. Finally, a well built rock revetment fits into the natural beach
environment much better than a wall.

A revetment consists of a cover or armor layer of large stones placed on a
foundation of one or more layers of smaller stones. The armor stones are designed for
the particular wave height that can be expected at that location. The actual size
depends not only on the wave height but also on the bluff slope and the shape and
specific gravity of the stones. The purpose of the foundation is to prevent undermining
of the armor stones by the penetration of jets from the impact of the waves. It is
therefore very important that the foundation stones have a mixture of sizes. Revetments
must be entrenched at the toe to prevent undermining and sliding.
53

Off-share Breakwaters One method of preventing wave energy from reaching
the bluff toe is to use a breakwater to protect the area. However, the cost of off-
shore breakwaters is so great compared with structures built near shore that this method
is only used where there is a need to create a harbor or protect a high value structure.
Breakwaters also interfere with navigation and littoral drift.

Some success has been observed during moderate storms in the use of a small
permeable near-@shore wall at a depth of only about one foot. These are usually
accompanied by impermeable groins. The permeability of the wall (about 40 per cent)
helps to mitigate the reflections and turbulence caused by solid walls.

Slope Stabilization

A bluff which has its toe protected from wave action is still subject to the less
violent natural forces caused by surface runoff, wind and groundwater seepage. More
serious erosion problems may also be caused by people, especially if they are allowed
to use vehicles on the bluff. The less violent natural erosion can be reduced by reducing
the steepness of the slope and encouraging vegetative cover. In the writer's experience'
a slope of 1-1/2 horizontal to 1 vertical is borderline and slopes of 2 to I or flatter
are much more secure. In most areas, natural vegetation will cover such slopes but
the process i's often speeded up with artificial planting. At some locations, it is
necessary to provide drains for the storm runoff from adjacent urbanized areas. If
there are layers of clay or rock in the bluff, perched water tables may provide seepage
over the exposed edges of the impermeable layers. If this is sufficient to cause a
problem, it can be reduced by placing drains or wells landward of the face of the bluff.

Shore Protection Demonstration Project

An important contribution to the knowledge of low-cost protection procedures
was made by the Michigan Shore Protection Demonstration Project which was funded by
a grant from the Michigan Department of Natural Resources to the University of
Michigan. The Michigan Sea Grant Program provided additional funds Jor observing
the installations and for the published reports. The various reports and papers resulting
from this project are listed in "Selected References." A revised copy Brater et al. 1977
is provided as an appendix to this paper. This particular report was selected for the
appendix because it provides descriptions of the various installations as well as an
outline of the laboratory research.

Loosens +
ranspol 5 f
Waves beach aterial
direction Twbulen"
heirt Rapid erosion
'a Loo
pa are soft material
ad alterr long
refraction
Wind dif Ifactloa
let I#C Direct action
a i an bluf f
Littoral
currents Transport
t 1 Suspendedlood Flooding
direction Bed load Wave damage
ma
Currents
Water level Oval topping andwar
barriers Y f barriers.
Ice movement Scow __j
Precip Local runat f Surface runoff Lac
Itation, River Ground water. unsfab@
dischargg Traospart 61 p

,,n
,,n
Fig.

10 a
Alan"
5rao

L.J,
78
C 6 Ur- -0 44
a 00.60 40
39 40 4C -so @
60 76 V76

Fig. 2 - Storm Wave Heights and Lake Elevations

Selected References

1. Armstrong, J.M., (1976) "Low-Cost Share Protection for the Great Lakes, A
Demonstration/Research Project", Proc. 15th Coastal Engineering Conference,
Amer. Sac. of Civil Engineers.

2. Brater E.F., (1979) "Observations on Low-Cost Shore Protection", Journ. of the
Waterway, Port, Coastal and Ocean Division; Amer. Sac. of Civil Engineers, Vol.
105.

3. Brater, E.F., J. M. Armstrong and M. McGill, (1979) "The Michigan Demonstration
Erosion Control Program in 197611, Michigan Sea Grant Report No. 55. Published
with the assistance of the Michigan Department of Natural Resources and the
Michigan Coastal Zone Laboratory.

4. Brater, E.F., N. Billings, and D.W. Granger, (1975) "Low-C-ost Shore Protection for
the Great Lakes", republished by the Michigan Department of Natural Resources.

5. Brater, E.F. and M. Cortright, (1976) "Beach Erosion in Michigan, A Historical
Review." Published by the Michigan Department of Natural Resources.

6. Brater, E.F., and H.W King, (1976) Handbook of hydraulics, 6th Ed., McGraw-Hill
Book Co., Inc., New York.

7. Brater, E.F., and C.D. Ponce-Campos, (1981) "The Michigan shore protection project
final report", Michigan Sea Grant Report No. MICHU-SG-81-203.

9. Brater, E.F., and C.D. Ponce-Campos, (.1978) "Coastal Engineering and Erosion
Protection, Report for the year 1976-77", Michigan Sea Grant Technical Report
No. 59.

9. Brater, E.F. and D.C. Ponce-Campos, (1976) "Laboratory Investigation of Shore
Erosion Processes", Proc. 15th Coastal Engineering Conference, Amer. Sac. of
Civil Engineers, Vol. IV, 1976.

10. Brater, E.F., and E. Siebel, (1973) "An Engineering Study of Great Lakes Erosion
in the Lower Peninsula of Michigan", Published by Michigan Water Resources
Commission, Dept. of Natural Resources, State of Michigan.

11. U.S. Army Corps of Engineers, Coastal Engineering Research Center, (1977) "Shore
Protection Manual.11

I ,
I
I
I
I
I
I SELECTED PAPERS
I
I
I
I
I
IV
I
I
I
1.
1,
1.

ASSESSMENT OF THE EFFECTIVENESS
OF CORRECTIVE MEASURES
.IN RELATION TO
GEOLOGICAL CONDITIONS AND TYPES OF SLOPE MOVEMENT

J. N. Hutchinson

Reprinted from:
Norwegian Geotech, Institute
Publication 124
Oslo 1978

NORWEGIAN GEOTECH INST
PUBLICATON NO. 12.4

OSLO 1978

_S IN RELA71ON TO GEOLO-
ASSESSMENT OF THE X&F OF CORRECTIVE MEASURE
GICAL CONDMONS AND TYPES OF SLOPE MOVEMENT

EVALUATION DE L'EFFICACTTE DES MESURES DE STABILISATTON PAR RAPPORT AUX CON-
DITIONS GEOLOGIQUES ET AUX TYPES DES MOUVEMENTS DU TERRAIN
HUTMNsw j.N., imperial College of Science and Technology, London, United Xingdom
Summary: p 1, the various types of correctivemeasuresare briefly reviewed. Attention
The General Report consists of three P . art
is then concentrated on the two, @Po rt= measures, modification of the slope profile by excavation and nj ling, and
drainage. An analysis is made of the optimum positioning of corrective cutsor fills, making use of the influence line concept i
borrowed from structural engineering. In this wa
y a neutral point, neutral line and neutral zone are defined for circular and
nou-circular landslides and for various values of B with respect to the applied change in total stress. Drainage is then discussed
in more detail, with particular attention being Sim to horizontal drains and to trench (and counterfort) drains. Performance
data for tretich drains in the UJL an then revieved anod analysed. From this a tentative basis for design is drnloped. Thecloggingof
draicage system by siltation or by geochemical effects, is also discussed.

In Part 2 the papers contributed to Theme 3 of the Symposium are reviewed. Finally, in Part 3. some suggestions are made
as to the desirable directions of future research. An crtensive list of references is provided.

Rdsunid:

Le Rapport Gindral se compose do trois parties. Dans la premitre paftie les diffirentes mesures de stabilisation sont
brRriernent examinies. L'attention est ensuite port6e sur les deux mesures les plus courantes: la modification du profil do la
pente par excavation at remblayage, et par drainage. Une analyst est faite de I!emplacement optimal des tranch6es ou des
remblais de stabilisation qui utilise le concept de la ligne d'influence emprunti au ginie structural. De cette mani6re un point
neutre, une ligne neutre et une zone neutre sont difinis pour les glissements de terrain circulaires et non-circulaires ainsi que
pour des valeurs diverses de If en tenant compte du changement apportd i la tension totale. Le drainage est ensuite discuti
do faqon plus approfondic en insistant plus spdcialement sur les drains horizontaux et les drains tranches (et contreforts) on
Partimilier. Les r6sultats d'essai sur lutilisation do drains tranchds dans le Royaume Uni sont prdsenties et analysdes. A partir
de ces donnees une tentative de mise an point de drains tranchhs est developpie. Le colmatage des syst@mes de drainage par d6p6t
do limon et par efTets g6ochimiques est 6galement discut6.
Dans Ja deuxi@mc partie les contributions 6crites du Suitt 3 sont discut6es. Enfin, dam la troisiime partie sont prkesentbes des
Suggestions sur lorientation i donner aux futurs projets de recherches. Une liste d6taill6e de r6rerences est dortake.

Tbis paper vu the Ge Report of Theme 3 at the Symposium on Landslides and odwr Mass Movements, Prague, September 1977, and
was published in the Bulletin of the International Association of Engineering Geology, No. 16. [Tn. p. 131 - 155.

wad Rich cracks against surface inflow Usually fall, any Aal
will tend to be broken by the slightest further movement.
in vww of its con=4 this General Report concentuttes; on the It is better. therefore, to make artangements to drain cracki.
stabilmtjon of dopes consisting of soils and rocks in their natural
%at*, in either natural dapes or cutongs, and e3tcludes ones that c/ Blanket- slope with free-drainJng material, with filters as ne.
comprise chiefly MIL onsary: This combines measure I /d/ with drainap /Root,
The theme chosen by the Organising Committee is a good one. It 19581 and is particularly effective in the case Of Slopes expand
directs our attention Vecirmily towards the assessment of the to rapid drawdown /Skempton. 1946, Finzi & Niccolai. 196 1.,
efficacy of sb&ilization niessum which a reading of the literature C4derpenj 96 71 Kleagel at al. 1974/.
diom to have bean a matter Imply neglected hitherto. This situation
has doubtless arisen PaWY from a natural desire to clan &a ilk on dj Trench drains: then are generally nwow* and aligned directly
ajob and partly from the reluctance of the client of owner to accept downslope /Early &Skemptan.1 9721, thus Imply avoiding the
continuing expenditure on long-term motatauns. risk of reactivating the landslide being treated. They are
sometimes supplemented by shallower drains laid in a chwmn
In accordance with the wishes of the Organning Committee, the or herring-bone pattern /Duvivier,1940/. An etarlier version of
General Report cons-ts of thme parts. The first reviews the main the trench drain is the counterfort Azaim In " the invert is
aspects of the theme in the light of the p - 1 - - stase of knowledge, located in firm pound beneath the slip surfax so that@ in 01
the second evaluates; the contribution of the submitted papers addition to reducing ground-water pressures, the drains also
while in the third, suggestions am truide as to the nature of the provide some mechanical support /Gregary, IS", Callin.1 $461.
stM unsolved. pmblenu and the dimcdon of future msearch. Open or Onvel-f1illed drain trenches; running ctos"lope are
sometimes built above the crest of a dip or dope. when they
I - REVMV are usually termed interceptor or cut-off drains /Toms &
Bardett,1%24 Smith,19641. Shallow ones merely intercept
Gamma surface run-aff -, deeper ones an intended to intercept ground-
water flowing towards the slope. A deep cut-off trench, ex-
Many goad reviews of methods of dope stabilization have ban leaded downwards by drain holes into a drainage gallery, was F
made. Some of the mom meant at by Root /1953, 19581, Bakar& constructed at the head of the colluvial, slope being stabilized
Marshall 119581, Brawner /1959/, Mehra &Natmajan /1966/, ZAM- at Weirton /D'Appolonb at aL,1967/. Cue must be taken to
ba & Mand /1%9/. Duncan /197i, 19761, Schwab= &Wright avoid siting cut-off drum so that they could act as a tension
/19741, Smith /19741 and Broms /19751. A useful review of we- czack in any future landslide.
thods particular to rock dopes is provided. by Peckover & X=
/19761. al Horizontal drains. usually ddled into a dope an a slightly
doing padient and provided with perforated at porous liom
The main methods of stabilization used we santmarised brielly /Smith & Staffwd,1957; Root,19584 Rob;nsm,1%7; Menke.
below. They may be employed singly or in embintion. 1968q Nonveffla.197(h Ent. 19741 Tong & Maher, 197S.,
Bmd,1976/. The maximum practice length of such draims
1. E%dendon &Mng Is generally around 100 in, though one 231 m long is reported
by Zisuba AMencl 11976/. Lengths of up to about 60 in am
a/ Excente at toe. until stability is attained : a crude method man common. In slides of large scale, horizontal drains can
which miles on stimulating retrogression of the dide until be used to advantage in conjunction with vertical drainap
Its average dope is sufficiently gentle to be readily maintainv& obafts /Nat. Conf, Landslide Control, 19721. with trench
Large quantities of excavation sic generally involved. The clas- Armin /La Rochelle at &L, 1976/, or with galleries /we 2f
sic application of the method is in the Culebra nub of the bolow/. In cold climates it may be necessary to prevent the
Gaillard Cut on the Panams. Costal Utton &Banks.1970/. outim of horizontal drains from freezing /Golder, 19711.

b/ Remove and replace dipped masend: either wholly by ftos- f/ Galleries: expensive, but am be appropriate to use in the
draining material /Symons.19701 or. man economically, by treatment of very large slides Mwer-Brady.19S51 Kezdi,1%9-,
rocompacted do debris provided with drains /Newman.189(h National Conf. Landslide Cantral.19724 Rico at &L.19761.
Duncan.19711. The method is obviously applicable only to Supplementary drainage boruip can be made through the
dips of modest size. A variant of this method is to destroy sides. Mior or roof of the planes as required /Tantpchi &
pm4xisting sherz sarflaces at shallow depths by dissing oat. Watad.1%5; Rodriguez at aL.1%71 Ziruba &Mencl.1969t
remoulding and woompacting the excovated material JWaeki6 Rock &Bray,1974, Nilsen &Lien.19761. For galleries running
1970/. parallel to the dope is=, Shur /1970/ has made a study of the
optimum locations for various ratios of horizontal to vertical
cl Excavate to unload dope: either by a general flattestim with purneability, asio$ a variable resistance analogue.
or without berms /Baker &Marjbal1.I9Sgj Broms,1969/. or
locally at the had of a dide /Peck &Imiand.19S31 Lotton & S1 Vatical drains: these may dischup by gravity through horie
Banks,1970/. As discussed law, it is impot t that such ex- zontal drains or adits /well-drainst /Sexton, 1938; Palmer et
a" tion an correctly posidoned. al. 1950; Shermil, 1971-, Rat, 19761, by siphonft within the
al limitation of depth IR*ot.1958/. or by automaticARY
dj Filling to load slope: generally by means of I possibly activated pump /National ConE Landslide Control,1972.
combined with other gravity structures, such n a gabion or Hock & Bray. 1974/. Alternatively, the water may be blown
mWatcad earth wells, at its too lViner-Brady. 1955, Brows out of the wells at intervals by compressed air. Under favour-
1%9, Early & Sitempton.19774 Ziruba &Mcnd.19761. Again able hydrogeological conditions it is sometimes feasible to
the correct positioning of stabdistag 111h is of great im;mr- discharge downwards into an underlying squiller at louer
tance, a Is their proper drainage. The disturbing effects of piazometric pressure /Parrott.19551 Witson,1961/. In some
embankments that hot, unavoidably. to be placed in do- coaM however, such measures have led to fresh stabilky
stabilizing positions a= be reduced by the use Of light-weight problems associated with the under-draining stratum JZirubi
fill%, such as fly-ash. & Mpnd,1%9; Lefebvre A Lafleur.1976/. Vertical driins
may also be used as relief wells, discharging upwmds@ to lessen
2. Drainage /Ced*rpvnI%7,I975# Rat.19761 artesian groundwater pressures at depth. The use of und
drain in this way, to stabilize a slope of quick clay. is descrr
a/ Load away sudice water. this should generally be done im- bed by Haim 11 %1/.
MedlittelY /Urubs, &MencU%9/.
b/ Piev t the buildup of water in tansion cracks: this sham Wider Armin- in which a bulldozer can operate, we used in the
be attended to t away /East. 19741. A U to U.S. /Root,19591.
62

h/ Electv-oamoshc used in the drainage of low permeability stabilisation of shallow instability in seek facas /Lang,1961,
$091% even INUO ClaYL Water migrates ftom anodes towards Price & KmM.1%71. together with other measures /Fookes
cathodes, whence it a rasnoved, with at without ing & Sweemy,1976; Pechaves & Xaff.1976/. Information on
/Cangrande at aL.I%Ij Bjenwn at aL.1%7% MitebelL,19701 the long-term behaviour at pre-aftessed anchor%, especially
Wade,19761. with rapid to am= losses through creep and corstision. is
W311346
j/ Vegetation: acts chiefly through reduction of pore pressures
by evapo4razispiration. Ziruba & Mand /1%9/ point out *at 4. Miscellaneous melhods;*
in this, and other ipqppcts@ deciduous - we superior tocmd@
ism 'There is also some contribution at shallow depth front the a/ GsoutizV a classical use of this is to seduce the permeability of
strength and binding action of the roots /Toms,1948; vim der the pound in order to taiduce the mpass of groundowater to
Burgt &van Bendegom.1948, Mehra &Naft*=,1%61 Zimba a landslide /Mitclaall,1970i Matsubayadii,19724 Cement
& Menci.1%9; Gray,1970, 1974, Brown &Sheu,197Sj Coin pmtin& n-rly aessited, has also been shown to be effective
at aL,1976/. In addition, a ve
. Ptationcmaaffectstlaeamount when hkiected into the slip surface of dides in cutting dopes
or-EWraliati AFMijOig Rod4a at &L,19761. The rob of of stiff clay /Aym,1961/ andin other materials /Duman.1971..
vaigatation in controlling susface arosion is mentioned in 5b/ Zdruba & Menci,1%9/. Hdone without care it can, of course,
below. tsigger a dide.

Raftaining b/ Chemical staitillization: by less ctm-- and other p
In the present comsecdcat it is deep sail stabilization rather
The uss of rigid resimaining structures is gesaimilly leas appropriate than surface treatment t1sat is relevant. Handy & Williams
than that of methods involving d-in- or msbmp* of the dope. /1967/ claim to have slabilked a slide by the introduction of
Numerous cam of failu of such struc- are reported by Root quick Hme into the sliding zone through Ihne wv" partly
119SSI and Baker & Marshall 119S81. When pzope* gi I by pozzalanic effects and partty through drying. Mouni at al.
howam, they can have a usaflal roh4 particularly where is /1968/ quwdon the efficacy of Ume wells but suggest that the
restricted. diffusion of vatious salts through wells may be a practicable
stalpftation method. For the quick clay that they investipted,
a/ Reaming wails, founded beneath the unstable powid:, potassium chloride had the best overall effect. Art application
indication of 1he great variety of designs nod is pren by Root of pousaium chloride diftsion in practice is described by
119581, Bak- & Marshall 119S81, Zdruba Mend /1%9/ and FolearaA & Sam /19761. Sam thearisticst background is Zinn
Costa, Nunes /1969/. Bujak at al. 11967/ -111 P the stabilize- by Mitchell /1976/.
tion of a rock all& by means of a concrete retaining block.
held down by prestressed rock suchoss and supporting a too c/ Suppression of. natural elect, ods: a method of slope
OIL In thk cm the efrmcy of the F, an a P N e measures was stlibilization described in wmal papers /Lg. 1968, 1973/ by
cbecked by monitoring; Veder. Under certain conditions it is claimed that ground-watei
pramaim can be reduced by the imartian of short circuit
bi Pilm- a wall of continuous at closely spaced driven cantilever elactiodes". which suppress an inherent, naturally occurring
piles can be effective in stabilizing shailow dides /Toms & slactro-camosis in the ground. Several landslides in Austria and
Bartleu.1%24 Zimba & Mencl,1%9/. Mon deep-mad slides Italy an stated to have been stabilized in this war. the method
ham been successfully stabilned by, for Instance. anchored dogs not appear to have been evaluated yet in the UX or the
sheet or bond pfle wails /D'Appotonis at &L.19671 Bandl, U.S.A. Veder /Pels. Com:04 points out thO the Principle is
1976, Zdruba & Mericl 1976/ or by large diameter cylinder widently inapplicable to cans what the ground-water p
pie nuisung was, generally of cantilever " /Andrews & as controlled by infliftafton under panty ibraiugh very
Klasell,19641 De Bmr,1%9; WUm,1970, Do Boa & Wallay% pame" layers.
1970/. A general discussion of this type of structure is Om
by North-Lewis A Lyons /1975/. A particularly met i P design, d/ Elactso-camotic anchorr. a technique wanted under the
employing anchored foundation plan 30 to 35 in deep and 13 m name elactzo-osmouc do-backs" by Casagraside & McIver
to diameter at 24 in contras to stabilize a rock Stop* in Italy, 11971/ for improving the stability of tailings d&ML It MEY 11150
is described by Badovin & Fattore /1974/. In dopes of soft be of mom ganard application in fine-grained materials.00im-
clay, there is a dan that the displacements and excess Pan- plessiv incres are reported in the pull-out resistance of thin
water pressues induced by pile-driving will trigger a Landslide steel rods pushed into the pound and then used a electrodes
/11jerrons & Johannessart,1964h Broms & Dennenowk,19681. in a direct current system for I to 2 weeks. In the soils tested
both electrodes showed to incresse in pun-out resistance,
c/ Sail and rock anchors, generally Pre-stsessO& than Us GEOP, though this was naturally more marked at the =odes.
loyed either in conjunction with re+sii" sauctures, as in 3a/
&b/, or alone to ratio= the driving foam of a lazuldide and to a/ Fmcing: expensive and rarely use& The best known appli-
the normal effective stresses on its sliP surface. For a cation is in temporarily stabilizing a now of silt during con-
planar slide, it can readily be shown that, for a given, anchor stsuction of the Grand Con Dam /Gordon.19371. Use of the
f'osoa P, the maximum improvement of the fictor of s1fatY method in stabilising a slide in a tunnel roof is described by
of ft AW is achieved when P is inclined tr 'a slope at Gunlaikgrud /1%8/. A general review of the technique is given
an MOO with the dip Puface equal to taZ17- by Sanger /1%9/.
This result is independent of the magnitude of c , and IT9.
pound-water pressures obtaining V Heating. in the course of railway construction in 19th century
England, it was a fairly common practice to stabilize the
Amhozs have been perhaps most commonly used On tMzWAti*- slopes of clay cuttings by bursting". Wide counterfort drains
nal rock dides /Zlruba & Mend,l%% Hook & Bmy,1974; were dog and filled with alternate layers of clay spoil and coal.
Raidovin & Fattore.19741 Lang.1976/. They can also be used The coal was then ignited, thus baking the intermixed and the
to sUbilin SOJL slides /Cmbefort,1%61 Costa Num 1%6/- adjacent clay /we, for example, Copperthwaitel 880/. Although
Allowance must then be made for short-term CZ0035 Pore- particularly advantageous when applied to clays and shales of
wam pressures and subsequent Consolidation Under the high carbonsecous content, which were virtuallY self-1111119.
anchor pads with concomitant losses in anchor stem Anchors &a technique was also used in non-carboitaccous days such as
ale also coming into increasing use in open pit mines. Pard- the London Clay. Note recently, unstable dopes of loess and
cularly in Amer;M for example to permit samiparnianerit
dopes to be cut mate steeply /Bamm at aL.1971& Littlejohn Some of these my still be regarded as being In an experimental
at aL.1977/. Another application is in the securing Of 1111P stage.
boulders. for instance in dopes of residual son /Costa Nunes
a venaso,m3l. Cables or chains may also be used for this -.rbc sachnique may be of un in the construction of rainfoxced
purpose /Bjerrum & J@rstzd,I%8/- Rock bolts are nod for earth wails with clayey bwkML
63

of day have been stabdind by PassiM hot Psm through & Modiracation of siope psofille by excavation and filling
system of tunnels or borcholes /HM,1934i Bela JL StanculescV6
1958, Litrimay et aL,I%lt Zimba & Menc:191%9; Mitchell, Gonaw remarks
1970/, Litrinov et &L also mention a tech*ue P a bining
beating with chemical treatment, which they am thermo- The respective merits of removing the head of an actual or potential
ohernicaL stabillsation. dide, of flattening the slope uniformly or benching it, or of building
a be--m at its toe, am discussed by Baker & Marshall /19581. Root
g/ Blasting: a controversial and unreliable technique, intended /1958/ and Mehra A Natarajan /1 %61. In general, such cuts and fUls
to disrupt a slip surface or to imp.ov drainage /Root,195gs would appear to be most effective when applied to deep-seated
Baker&Marshall, 1958;Zdruba&Mencl. 1%9;Mitchell, 1970/. landslides, in which the slip surface tends to fall steeply at the
head and rise appreciably in the moon of the toe. Ocarly. however.
Ill Bridging: a technique occasionally used to carry a road over them is also a scale effect, so that the influence of a given cut or
an active or potential landslide. It is'more often employed fill on the overall factor of safety will diminish with the size of the
on stem slopes affected by translational slides ot moderate landslide being treated. For ixample, in the can of the large deep.
or mug width /Root, 1958. Baker & Marshall, 1958; Zimba =led coastal landslides at Folkestone Warren the toe-weightine
A mencl 1969-.Costa Nunes. 1%9/. constructed as part of the stabilization works [Vim-Brady,19551,
although wall positioned, Jimpsove the overall loritterm factor of
S. Erodon Canbol, safety by only about 3.5 to W.

The close link between man mowen ents and erosion is a ba* It I& of course, very important to ensure that neither cuts nor
demerit in the geological cycle of demidation. It follows, therefore, fills UiW the existing or potential dide that they are designed
that the control of crodion, In both the general and the specific to stzbffiw, nor generate fresh slides local to themselves. It should
sanso, is ftindarriental to the prevention of landslides. 71his. point also be borne in mind that a IM is a rather specific measure and
has been emphasised by Bjerruns et AL /1969/ and by Hutchinson while it may deal satisfactorily with the particular slide that it was
1197j-r- Furthermore, once a landslide has bum stabilited it is designed to stabilin /abc an Fig. I /. it way be quite ineffective
impo,rtant to prevent any fiLrdier worsening of its condition through against an almost equally serious over-rider- dide /on abdl. The
emision. The potential croding agent is usually water, though ero- danger is greatest in the case of translational landslides, as Ulustra.
sion by wind and other agencies can occasionally be eguXuant ted in Fig. 1: the geometry of rotational slides makes them less
/P-v- Colenuit- 19284 prone to, though not necessarily safe from. thk type of faure.

A/ Control of too erosion: In cam whom the tac of a landslide
Is situated in the sca, a lalm, a reservoir or a river, it is of
prime importance to prevent erosion at dds most critical a
point. Measures commonly used Include concrete or crib walls,
4*rap and other revetments, sroynes and spur dikes /Viner-
Brandy, 1955, National Conf. LmddWe ControL*1972/.
3
b/ Control of sadwe on slopes: generally achieved d 2
through proper attention to the design of drains for surface
water /Mehra & Natarajm,1%6/ and the encouragement of t'
suitable vegetation cover /Buztz,I%9/. The latter was done C
originally by topsoillng and seeding sometimes in combination
with jute netting to prevent sail erosion while the vegetation Fig. I Translational slide stsbilized by a too fill. showing the
was becoming established /Mehra et aL,1%7/. More recently danger of a potential over-rider slide.
various techniques of hydraulic seeding, in which a mulch (1) Slip surface. (2) Too fUl. (3) Over-rider slide.
containing the plant seeds is sprayed onto the slope, ham been
developed /Schiechti, 19651. In then the necessary interim A point frequently neglected in the literature on slope stabiTmation
erosion protection b often provided by asphalt emulsion, or is that cuts and fills. depending an the ground condition and the
other chemical or plastic admixtures. A layer of sofi-cement speed of their constructim cocriprise mom or lea andmined un-
has sometimes been used n a sentilmmiment control of loadings and loadings respectively. The matter is cleariy presented
surface wasion /Costa Mum AVel1os%1963/. by Bishop and Djerrum /1960/. Thus in the can of a fill. th value
cl Control of -; -g crodow this type of erosion occurs wheri of the factor of safety, F, will--cenerally be lea in the short-tenn
the seepap drag of groundwater disdiarging at a free face is than in the lont-term. The ooposite will usually apply the cue
Large enough-to dislodge and remove individual soil particles. of cuts. In both cam it is good practice to check both the short-
The remiltant back-sapping tends to undezridne the and long-term values of F. An important advantage of a comctive
cumbent stmta and mntWY to cam their collapse. This M is diat, once succesdully placed. its stabilizing effect will tend
form of fall= is quite common in both natural and cut to Increase until the ground ' , th it is fully consolidated and
slopes, and can be very damaginst Soils in the cown silit to thereafter, unless by some accident the fill is removed, its coritrp-
fine and mop am particularly prone to it /Terzaghi,1950i bution to stability will be a permanent one*? As discussed later.
Hutchinson, 1%8/. I'm flee face may also, under certain this reliabilLy in the long-term is generally less assured in the case
conditions, migrate into the slope In a localized manner. of drainage measures.
leading to piping by subsurface erosion /Tmaghi & Peck,1967 I
Sherord et &L,1972/. The chief method of controlling scepage DW "neatni Ime" t
erosion is by placing inverted filters over the area of discharge The efficacy of & corrective cut or fill is controlled by its location,
./Wud.1948, Teizaghi & Peck,1%71, or by intercepting the
seepage at some distance beck from the face with wells or weight and shape and by the chmacteristics of the actual or po-
sand drains /Anon,1965/. tential landslide that it Is intended to stabilize. These factors me
all known, or can be determined, at the design stage. It follows.
From this brief review of types of corrective measure, two points therefore, that while it is advisable to check their performance by
emerge. The first is that, while most of the papers suggest that the appropriate monitoring during and after construction, the efficac)@
measures applied were effective. at least in the short term, It is of cuts and fills should first be 4etermbied by analysis befai the
rather ram for the efficacy of the corrective works to be properly am carried out.
assessed by appropriate monitoring. The second is that drai 11 CAlculated for sections W6 and W8 of Hutchinson A %91.
appears to be the most wrierally used stabilkwtion measure, with
Me modification of slope proffie by cuttint and fillins also fre- **Provided that proper drainage of the fill has been arriinged, for
quentlY employed, Tbus. in the remainder of this paper, attention
d

is concentrated an then two coneetive methods. dimple by an underlying drainage blanket and by appropriate
surface dnims.

In sawtund engineering design a f-Diat device is to consider The - F P Of the nestud points in plan is termed the NEU.
that an jnfluence load- of some convenient magnitude passes rnt@ LINE /Fig. 2c/ and this forms the boundary between the
ova a structure, for inistance a bridge, and to determine the m. am of the on which a fill /or cut/ would impgov its
swung influence lines", typically for banding moments and shear stability and the am for which the reverse would hold.
forces at any point@ = a function of the position of this load on the
struMm The same idn may be applied to a landslide, to determine Determissation of the location Of this neutral point
inflinum'lines for changes in its factor of safety produced by an
influence load moving across it. Let the sUp suface /Fig. 2a/ be of general sham with effective
shear strength parameters c' md 0'. Let the influ re load, in the
0) Section Y-Y &* instance, represent a uniformly distributed fill, of intensity w
&W per unit of horizontal area. acting on a dice, i, of horizontal width
b. Thus A W - wb. Now in the gencM cut. application of Phis
loW will at up undrained me-pressures; an the subadjaCentL Slip
Slim i surfisce, of mqnitude A u - f. a * * 11 fog
Aa a I w If owing Skempton
A Hutchinson /1%9/, we apply a Con"ntlonal Method to this
tion,circular slip =d6ce, and let
Fi- 2: tradable misting form
2; daving form Z D 0
the new fector of xd&W F I uill then be
F A Nj' ton 0' + 2 R,,
b) Influence line for Y-Y I & a sO) AW sin a,+ ZDO
where a, is defined In FW 2& and A W Is the change in effiective
normal form produced an the ban of @icc i by the infl,woe load
AW, givess by

46 Z DJAW (cos a, - IF sec a, tanO: R- 1
C I Plan
X 1.0 Us V3 I X F0 ZR 0(AW sin CtI + X DO)
Now, at the neutW point,
OD
Y Y F,
I and c@ - an ( FW 2a)
F
0
2 Z The position of the umtmk point Is thus given by
tan%_(, _g WC2 %) tana,

fig. 2 a) Ciaiwiectioa of a landslide, with an Influence load A W Alternatively, the Bishop Simplified Method /Bishop,19S41 may be
acting. used as the basis for the assalysis. Then, with ft symbols used in
b) Influence line (diagrammatic) for the effect of A W *on that paper.
the overall factor of safety of the landslide.
c) Plan of the landslide, showing diagrammatically the
positions of the neutral lines for different D values. Fo jjc'b + tan V (W - ubj
2: V/ sin a + tan tang
(1) Slip surface. (1) Influence fill. (3) Position of neutral
point (drawn for B @- 0). (4) Positions of neutral line. for 0
different 19 values. Hc=c6 putting
Consider, for simplicity, a slope that can be mpresersted two-dimen- jec a
sionally by die cross-section in F* 2A. If a co..v nienit, arbibiry in- [cb + ton 4V (W - ub)] a J, and I + L&E @Ian a " M 0
fluence load, A W, is now assumed to act, in tum at successive po- F
sitions between the head and too of the landslide we can derive an 0
influence line for the nesulting effect an the Overall Safety factor. (jMO)
as indicated in Fig. 2b. 11m ratio Fos IV an a
F I(= F0 + AM. the overall F with AW acting Then the now factor of safety, withA W acting an slic:c 4 is
F0, the original value of F Fla I - [AW ton OP (I-b) M + Z UM
AW sin a + JW sin a
is taken as a convenient measure Of dLis effect when I
An iniluence load representing ML. with A W positive and acting sec a.
downwards, will of cow= tend to deciasse the existing factor of get: a and M
Plan

Z 2

safety, F0. when it sets in the vicinity of the head of the slide and CC !!netana-
to incmde this when it acts neat the toe. Of particular interest is I + tan 0 FXan a F
the point where A F - 0. or FI/F0 a 1.0, termed the NEUTRAL
POINT. which indicates where an tOuem load will hwe no effect
65

Hama
F, * [AW tan OP (I - 2) M14 + Z(Jml)] XW da a
F. W sin 9@ + 2; W sin a) 2; UM.)

F
To flad the position of the neutral point, put -@ - I and a, an as
before. 0 O.S-
Then sin a. - (I - A) M tan 0
in FO
but at the neutral point MI; h( Olk , therefore

U - 9) sac a. 0
sin% a . . tan -9 02-0 4.0 0
I + taDA' tan Fo Who 0
ta.- (Nn
FO

which reduces to Fig. 3 : Approximate variation of ado'mob with The refs.
tionship is not perfectly linear, but the divergencies are or
tan % WC2 an) tan JV , as before /eqA (1)/. insignificant within the normal range of ast4obyalues'
FO
the neutral point for R - 0 is displaced towards the dope by a lio-
This is a quadratic in tan The negairre root gives high negative rizootal dista equal to the radius of the appropriate friction Ir
vahm of 411, which doubL Chiefly reflect the known anomalies circle, Le.
of both th Conventional Bishop Simplified Methods in
cam of =iy rising 34 s'urfam's"lWhitman &Baflay,1967, Turn- P. /Fig. 4a/
bull & Hvorslev 1 % 7/. Then negative values of a (0' b - 90 and 0
steeper) in not of practical importance and are e6folo hq*te&

Equation (1) is a general solution which applies for any value Of
J to positive or negative innuence loads, Le. to M14 if ammed. to
be of zero strength, and to cuts if the complications of changes in
slide geometry and the breakdown ofpossible negative pore-pressures
an neglected. It also holds for sUp surfaces, of circular or general
shape and, in principle, rejardless of whether these are pre-existing
at potential. It would obviously be questionable, however..to apply
the idea to a potential slip stirface which had no physical reality.
It will be noted that the position of the neutral point is indepen-
dent of A Wand c'.
r 10
Two cases can be distinguished:
(D for 11 - 1.0, when
tan an - (I - 3OC2 wan dl (2)
an) r-
0
which is satisfied by a,, a 0 (3) U

and CU) fat 0. when tan a. tan'#.,)b (4)
a O;ob b) .06
12

In the generd case of I.o > A > 0. the neutral point wM occupy
positions intermediate between those of cams (I) and (H) above, .001.00.00
in accordance with equation (1). 7he solution of this equation is
shown graphically in Fig. 3.
While B is an undrained parameter, the cam of 0 may also
conveniently be regarded as equivalent to the fully drained con&
tion with respect to a corrective earthwork of any initial pore-water 6.0
Pressure responscO. In principle this idea can be applied to any 0 Sol
value.
For circular dip surfaces, the result for can (I) confirm the salf-
evident fact that for I - 1.0 the position of the neutral point 09, -6 -6
13 situated vertically below the centre of the slip circle /Fig. 4&t. 2D 30 'a So 70
In comparison, the solution for cage (h) &bows that the position of Fig. 4 a) and b)
ro
Lq-) L-@

71c lOng4erm OfTect of the carthworks on the migirW steady
state pore-water pressures in the Ianddidg as & whole is ne*Md.

P*03 In addition it Is clear that for Son-citcular dip surfaces which
inctude a Planar SOCUOn, the 1201021 IMe will widen to form, a
NEUTRAL ZONE should the inclination of this section to the
horktontal coincide with the value of(In for the particular conditions
obtaining
. ......... @.*%%.
Discussices

Partly a a check on the show results for the positions of the
neutral point and partly in order to define the Chosen influ
lines Jm their entirety, a number of computer analyses have ben run
10 20 30 1.0 W 60 70M for a constant slice width b. The effects of various influence loads
acting on a typxW circular slip surface have been detentuned
Fig. 4 Influence lines for F, /F for a typical 1-ular suds: - Bishop's Simplified Method /1954/ and similar investigations for a
a) Craia-section of lai@Wida. typical noncircular slip surface have been made by the Morgenstern
b) Influence lines for an influence "cut" and an influence &Pzics Method /1%5/. In all these analyses an arbitrary influence
M for 'D - 1.0 and I a 0 NW for Vat us original piazo- load of 20.39 tones f /200 kN1 an a I Matte width /Measorad
Metric lines. homontally in a valleywartl directiont has been used in the circular
c) Influence lines for an influence, drainago. for various cam and the sonse load acting on a 2 main width in the non.
original pissometric lifts. circular ones. In all cam the section of landslide analysed is I to
(1) Ground surface and piescenstric line u (iu = 0.49). dikk in a am slope direction.
(2) Plazometric line v (iu a 0.29). (3) Slip surface and
piezossetric line w (iu a 0). (4) Influence fEL (5) Influence The complete kauence lines for each case considered we given
"cut". (6) Influence drainage. (7)-Friction circle with In Figs. 4a, b, S, 6, and 7a. b. As will be secri from the plot& of
radius - It sin 01 - (8) Influence fill with I - 0. FI/Fo q*= a. in Figs. 5 and 6. the computations confirm the
(9) Influence fin w4l a LOT!) Influence -cut- with t1knitical predictions for the poutiou of the neutral points.
0. (11) Influence "cut" with 1.0.
NR. Curves (9) and (11) we independent of the original
piazonsetric line.
1-06

F1
FO G) Q (1@00 nob
U 0.491 O;mu a 22 -18'

I
V 0-294 O@n, - 16 - Ize
1-04- W 0.000 46,w a 11 - 28'
4,

1-03

102 -

1100,

1101
1-0 /011

0-99- U V W

0-98-

0-91Y
'-70 -60 -650 .40 *30 #20 010 0 -10 -20 a -30 -1-0
CL
V

FiS.S: Influarics lines for F,/,,, for stypka circular slide (min 4), platted avain"a-
(1) Original piesometric lins. (2) Average pore pressure ratio. (3) Mobilisad affect*" angle of shearing ame, (4) Influems lines for
an inFuence, fill with S a 0 and various original piezometric levew (s) Insuence line for an influence fM with 1.0. which Is
independent of original piezometric level. 67

1.08
Fi
FO - ru Vm.b
106- U 0-49 O@Mu = 20-420
V 0.00 46V = 10 - 600
104- 4 .0, -

1-02- 4@% 10000
100000

1.00
00000@
wol@

0.98-

O-g6

0-94 - MIN
+50 +40 +30 +20 +10 0 -10 -20 -30

a
Fit. 6 zInfluence lines for F, IF* for a typical son-circuiar slide (as in Fig. 7). platted against a.
(1) to (5) are as defined for F* S.

A check. for the circular case with 0, on the sensitivity of the Suing analysis refer strictly to the undrained change in pore-wat:r
results to the magnitude of a positive influence load, indicates that pressure associated with the applied cut or ra and the conclusions
dw values of F IF - I are proportional to this load up to a value reached are thus independent of the ..background" drainage condi- a
of at least twice. le maximum total vertical stress acting an the tions in the slip as a whole, except insofar as these determine F
dip surface conskiered, that is. well beyond any prac*41 limit Mius, for example. the neutral line for an influence loading "'A
This proportionally does break down. however, if the influence load a - i.o will be dermed by % - 00 regardless of whether the slip
is hmmsed gmtly, for instance by a factor of 100. itself is in the short-tam, intermediate or long-term condition. For
the run two of these conditions, however, the continuing change
A pre-requisite for application of these ideas * of couM deter- in pore-water pressures in the slip itself. as these move towards
mi-tion of the location of existing or likely potential ft surfaces. equilibrium, must also be considcred, in relation to its effect on both
As this information is needed in my case. however, to Oermit the overall long-term factor of safety and the long-term position
piezometers to be installed in their proper positions, this is not of the neutral point.
seen as a disadvantage. The examples considered all comprise singic
SUPL In practice, complex assemblages of slips am ftequently en- in the past, vectors of surface movement in an active landslide
countered and then a set of neutral lines /as in Fig. 2c/ may need to have sometimes been used as a guide to the proper location of
be deflned separately for several, or all, of the component single corrective earthworks, for instance by limiting the extent of a
dipL counterweight berm to the area which has shown a tendency to
rise MIL Early, pers. comm./. On the assumption that the ground
It follows from equation (1) that the positions of the neutral lines, surface movements closely reflect the dWc of the underlying
for any conditions other than that represented by the special case ft surface, we now see that this method is accurate for a fully-
of 8 - 1.0, will tend to shift valleywards as the initial factor of undrained MI but on the conservative side for a fully drained one.
safety. FO. is increased towards F by the stabilization works.
The pracdcal 6gniflcance of this ;di depend partly on the shape Some engineers hold the view that. in effect, the neutral line is
of the slip surface and partly on the designed improvement in F 01 located vatically beneath the centre of gravity of the slip mass.
In principle itsep, s better to work as far aspossible with the finil From the foregoing. however, it is clear that, in general, this is
value of F. not correct. In the landslides examined so far, two of which are
It should perhaps be emphasised that the values of A in the fore, shown in Fig. S. the vertical through the centre of gravity of the

2 iMns 32")
LL_OS L
a CI-0
(cs.. 06 1

alto- lop
%%% so
0.96 - (QX 12s)
1k.0
21-0
01" lafts 00 1

Mg. Examples of the relationship between the extrem
0-22a 10 20 30 40 so 60 70 position of the neutral point and the contra of gravity of a
slide.
a) Circular slide at Select (after Skernplon & Brown 3961)
c) b) Mon-eircutar did* at Bury HM (aftw Hutchinson. at aL
1973)
V) Position of the ceatre of ruvity. (2) Neutral point for
102 U a 1.0. (3) Neutral point for B - 0. (4XSUp surface.
slip man falls between the 9 a 1.0 emd the 0 - 0 positions of the
neutral point. This would scern to be generally the case, but, may
1-00 riot invariably be so.

It is anticipated that the ddef value of the work presented above
will be in the initial stages of &'design, when a knowledite-Rf the
10 20 30 40 SO 60 70 positions of the neutral line fair various values of B Le.z. F - 2c/,
a @ should be of help in choos - for instancVhe timum route
Fi& 7 Influance Ham for F, for a typical son-circular stfide: - for-& -road to crosS an exist!t landslide or e NZ location for
corrective earthworits. The influence lines may also be and to
a) Cr@ion of landslide. make preliminary quantitative estimates of the Fl/F ratio pro-
b) Influence Lines for an influenc* -cut' and an influence duced by a given carthwork. A final check an this s6t& however.
fill for '11 - 1.0 and 11 a 0 and for various original piazo- always be roade by nutning orthodox stability anslyseL
metric lines.
C) Influence lines for in influence drainage for various
original piesometric lima. Drainage
(1) Ground surface and piezoinstric live u 0.49). General ks
(2) Slip surface and piexomet. Una v (;u - 0).
(3) Influence fUL (4) laft" 'a "Cut". (5) Influence The neutral fine carwept is clearly inapplicable in this -ft. as
drainage. drainage of any pan of a landslide i-@'Wways beneficial. Influence
(6) Influence fM with S 0. (7) Influence fM with 1.0
lines for the effect of an @iafluence drainage groducmig a r -
, I - iduc
(a) Influence "cut" with - 0. (9) Influence "cut" withl tion in piezometric pressure of 10.2 tonn'e fl 00 kN/m / an
-1.0 successive I in wide slices for three different piezometric lines in
NL Curves (1) and (9) am indepandem of original piezo- a typical circular 4, in given in Fig. 4c. Similar influence lines
metric level. for a piezometric reduction of 10.2 tonne flm' an successive 2 m
wide slices for two different piezometric lines in a typical non-cir-
cular slip are given in Fig. 7c. From these examples we we that the
effect of a given drainage is rather constant throughout, except
for a dot rise at the toe. ne rapid reduction in effect at both
head and toe of the slides results from the neglect of negative pore-
water pressures once the reduced piezometric line falls below the
level of the associated put of the slip surface-

Some indication has been given earlier of the great variety of drai- the drainage of homogeneous 3:1 dopes by horizontal drain., and
nap measures that have been used for slope stabdization. Of these, produce - design charts which am believed to apply with reasonaole
two will be tinted in more detail hem accuracy to haingencous slopes with inclinations of bet-avan 2.S
and 3-5:1. Two depths to an impervious lower boundary are con-
Horizontal d. sideredL As the authors state@ the design charts ban yet to be
calibrated against field experience.
This tam refers to small diameter pipe drains that are insulled
within a slope, usually by helical auger or rotary drill*, an a rising As indicated above, monitoring of horizontal drain systems has 0
gradient of typically 2% to 20% so as to discharge by gravity. While garierally beeii confined to measurement of drain discharge. In
the origin of this type of drain is obscure, much of the early develop- some cam the effect of drainage on the movements of the slide
ment work was cwricd out in California, where horizontal drains being treated have also been reported /Teixeira & Kanji, 1970/.
have sinci: been widely used /Smith &Stafford.1957, Root,1958/. The effects of horizontal drainage installations an ground-water
Mon recently. they have been employed to stabilize slopes 'in many pressures have, as yet, be= measured rather rarely. A brief report
other countries, including Japan /Taniguchi &Watari.1%S/, Britain is given by Brawner /1971! of reductions of between 9 and 12 in in
/Aym.1%1; Robinson.1%7/, Germany. /Henke.1969/, Czechodo- deft-water pressures in a 137 in high rock slope, occurring within
vakia jUiiiba A Mend, 1%9/,'YugosIavia /Nonieiller, 19701, New 30 days of installing four trial horizontal dminL A fuller description
Zealand /Eas4 1974/, Hong Kong /Tong &Maher, 197S/, C4nada of the stabilization by horizontal drains of slides in a 15 in hW
Aa Rochelle at aL,1976/ and FranceX=befart,l%6j Rat,1976/. 3:1 cut dope of slits and sands in Now Zealand is prcrvided by
Horizontal drains ate usually between about 5 and 20 cm in dia- Eag. /1974/. Flfty4lve drains of 30 in nominal length and 3 in
2CM& in three tows. were installed in the space of tan days
meter. In most cam they an spaced between 3 and b in apart during the winter of 1972. Within a further 5 days, pietainetric
and, as mentioned &Um, have lengths of about 30 to 100 m. They lavels had dropped by between 1.0 and 2-5 m /1.7 in on average/.
may often with advantage, be installed from several davations. A check in the winter of 1974 showed that the average depression
Discharges from a single drain live varied from about 175,000 of piezametric levels was then 2.0 in. No further movements of the
litres/day /13rawner,1971/ to zem Early installations generally dides was recorded in the two years following instcallation of the
consisted of perforated steel pipes without filteri and were prone drains. It is now clear that this Installation was over-designed. It' J
to both corrosion and siltation. Smith A Stafford /1957/ recom- was. however, an emergency operation to save a high-tension cle:c-
tuended that the 6 in length of drain nearest to the outlet should be tricity pylon threatened by the dides and as such was both expe-
galvanised and nouperforated, to dow down corrosion and hinder Alton and successful. Various divergencies from the idealized
choking of the pipe by roots. More recently plartic pipes have been slopes explored by Kenney at &L 11976/ prevent this case record
used, with filters formed of porous concrete /Robinson,l%7/. from mv g a a check on their design charts.
resin banded sand /Nonveiller,1970/ or synthetic filter fabrics. In
jointed rock and residual soll masses, Choi /1974/ recommendi; There is a gint need for can tecards of well instrumented and
the use of drains with Impermeable inveru monitored horizontal drain in-all-tions in various sails and rocks.

The use of horizontal drains is most appropriate In do case of Tioiiich isad cosmaisfart &a=
slopes where the $round-water Ho too deep to be imbed by
tranck dndnL Such conditions. no fkaquently associated with Gonad remarks
relatively song slopes and deep-seated slip sufaces /NonveEel,
19701. Most of the slopes treated live been between about 15 As indicated above, the tenn counterfort is used heir to describe
and 4 So in inclination, though horizontal drains have beim insWled drains which penctizie into solid ground beneath a slip surface,
in Hong Kong in slopes of weathered ipeous rocks of up to 70 and therefore also provide wine machartical buttressing to the
Inclination /Tong &Maher,1975/. From a practical point of view dope, while the tes. trench drains is reserved for those which do
it is important that the rock at sell bivalved is reafflY drillable and not thus penetrate, and so voutribute to stability only dirougli
does not cme, and it is an advantage if the necessary drain length their drainage action.
does not exceed about 90 in /Smith &Stafford,1957/. As with my
form of drainage, it is helpful If the mass to be treated contains The counterfort dim seems to have been the first of dim tylies
previous layers or zones, but this is not asundaL. to be used n a principal. stabilintion measure. While the origins
have not been explored, it is clear that drains of this type were
It can be diMcult to install horizontal drain in a slide that is being widely used in France and England during the first half of
still moving, as the drill strinp may jam Such an operation was the 19th century, in both embankments and cuttings /Gregary,
successfidly carried on t, howmer, on an active slide an the Slo Paulo 19"; Collin.19461. Collin. in particular, developed his designs to
to Santos highway /Temaira &Kanji,1970/. The dide had an an of a considerable degreti of sophistication, arranging, far exampic.
about 20 hectares and was moving at between 2 and S m/month. the width of the counterforts to increase stepwisc down-dope so
Because of this high movement rate, the boazontid dram wait as to improve the stability of the day mass lying between tbarn.
Installed In two stages. In the first apreliminary stabilization was To the same end, in remedying a landslide, the tog between the
attained by butallfrig drains about 40 in Long. ft wu than possible counterforts was often excov ted successively and then, if suffi.
to burtall the go r - d stage drains in lengths up to 120 in. After 5 ciently dry, replaced and con. wad in thin layers /CoUln.1846
yew the slWe had shown no significant further movement. P.71/. In contemporary Enelsh practice /R. Stephemn reported
by Dockray, 1844, Gregory. 18441 UW operation was omitted, as
Until recently, horizontal drains have ban insulled entirely on an indeed it is nowadays.
empirical basis. with the quantity of water discharged as the main
criterion of success. Latterly both Nonveiller /1970/ and Kenney at At that time it seerns to have been scrienny believed that drain$
al. /19761 have criticized this approach and re-em hasised that the that did not penetrate ban th the mat of diding wen of little or
primasy aim of such drainage is to reduce pose-water pressures no we /r-g- WhWcy, I &RO/. The basis for such a view was removed
which, in day slopes, may be achieved with a very small yield of by Tevaglit's enunciation of the principle of effective sum in
water. Both Teixeira & Kanji /1970/ and, to a man detailed extent, 1925, but it took about a further three decades before trench
Noweitler /1970/ make use of flow nou to estimate the reduction drains took their proper place as a stabilization measure in the
in pare-pressure that a gtivit horizontal drain butallation will UYL
achiem A very thorough theoretical study of the stabilizing cf-
facts of hotiz tal drains has man 1 tly been made by Choi Approximate dwory
/19741.
Up to now, the design of trench drams for slope stabilization
Kenney at al. /1976/ make a throodimensional model study of in the U.K. has proceeded an a sami-empirwal basis. Typwgdly,
the average lowering of piczameter level an the dip surface m
quired to produce the dad d increase in factor of safety has been
A case where horizontal drains were installed by hydraulic jacking calc-ulated: the trench drams needed to effect this avenge Imueting
is dewribed by La Rochelle at al. /19761. hirre then generally been dimensioned on the basis of experieva.
70

Them is an extensive Ifterstu're an Sro'un&wstef now to drains. centrated on the plane EH Ift 9/ at drain invest level. Placometric
especially in the fields of soil physics and agricultural engineering surfaces, with P P vp P P - to this plane, are platted for the two depth
/reviewed, for example, by van SchAfgaarde 1970, 1974, van Horn ratios and various vahm of SID In Fig. 12, an the assumption that
19741. This is largely concerned. however, with the prevention of
water4ogging of crop toots and thus with the determination of the
phreatiq surface between horizontal draw of relatively shallow depth
/often I to 1.5 'ml. Drainage for dope stabilization, on. ft other
hand, requires a imawledge of the piezometric levels at depths Of 7=74
typically up to 8 or 10 in. 7be problem has therefore been approa-
86
ched through the use of flow nets, as was so d by Henkel
/1957/.

Even in the long-term, the actual threedimensional trench drain
problem, with intermittent racharge from inktration, a variable
inflow of ground-water from updope, and non-hemopneous and
notropic permeability. varying with effective sum Involves non-
steady saturated and partially saturated flow, and is highly camp-
lox*. Furthermom the ' P reloan occasioned by the excavation
of the drains generally produces a signific t short-term modifica-
don of the original pon-water pressures, which is followed by an
to stage of consolidation and possilily swellimg, before
the long-term condition is reached. The long-am condition is of
ilization, however, and
chid in die context of slope stabi
the approximate treatment of this, described below, has theWore
ban developed.

The initial assumptions are that both, dw pound sarikee and the
. . piezometric surface we horizontal, that the permeability
of the ground is hornoymous and isotiWic and that the tranich
drains are of rectangular cmwaction, and parallel to each other.
The effects of artisotropic permeability we explored subsequently.
This arrangement and defirtitions of the various symbols und are
shown in the key cross-section of F* 9. The drains are also assumed
to be of past length L upslope compared with their spacing S. so
that a two-dimensional approach will have some validity. A photo.
paph of such drains being excrvated /for which B 0.6 in, D - 5
to 6 a and A I I to 12 in/ is shown in Fig. 10.
i @ Rg. to.: Photograph of tramh drain being dug In a till slope at
A Low Worsall. Yorkshire, April - May 1977 (with acknow-
S Istig*ments to the Northumbrian Water Authority).
7@07- ro it
P the Penveability ratio Rk is 1.0. For any given S/D ratio,

die same plots, of course, also yield the piezometric surfaces for
0 varying values of Rk through scale transformation.
he
It is now assumed djs4 in effect@ a horizontal ponded surface
h r*4 exists at some depth p - D - h below ground level. This approxims-
not ble for clay SODS in temperate
tion, which is probably unremnlk
X th,, 0' ... climates, particulady as k often decreases rapidly with depth,
----j en-b! the approach to be made man general by substituting ho
A/2 for D in Fig. 12. The approximation is. of count, also implicit iii
tha theoretical cones in FISL 13. 20 and 21.

Fig. 9 Key diagram: crose-section of t"ical is each drah2& From Fig. 9. the following deflniti and relationships can be
(1) Ground surface. (2) Original piezom*tric level 00 Plan@ drawn with respec to plane EFGH after long-term drainage:
EI-L (3) Piezometric levels on piano FG after drainsm
(4) Mean piezometfiC level on plane FG after "nags. N The avenge pigtometric had, fi, on the plant FG between
(5) Moan piezomatric [we[ on drainage inverts after A.,.in. can be expressed as - f h . where f is a factor re-
drainage. (6) Trench or counterfort drain. (7) Clay s*&L aecting the shape of the rele;Aininjitzomeil surikee /3 in
(a) Impermeable boundUT at depth- Fig. q/ and hin is the piezometric head mid-way between drains-
Flow nets, derived from rugibs element analyses by E,N, Bromhead, The a"rage pi=OMetriC lit2d, h ale an the whole p1m EH,
have been drawn for an SID ratio Of 2. for two depths to an imp", is Oven by hav - hda + R S
meable layer /defined by depth ratios. n, of 1.0 and 4.5'0? and for
assumed ponding of water at pound level, Le. It = D /Figs. 9 and
11/. From thew, the piezometric levch on ;L@ at any depth can whed hd Is the piezometlic head at the drain inverts. In
readily be determined. For simplicity, however, attention is con- generai, in the rest of the paper, h d is assumed to be zero.
In addition, the solutions am generally a function of the parameter when
,he ratio of the inflitration and the permeability. which is hav' FS
q/k, A
not readily determined very Closely.
The expressions under /i/ and 1W above apply to horizontal ground.
-Increase of n abWa this value RTOM to haw little effect On the - For pound of inclinsition 0. with trench drains of high US ratio,
flow net within the drain depth. 71

000

%b
OOP .000,

doo .00
Ole de

011
00

C2@r

11 Typical flow nets for Rk 1. drawn from finite element analyses;
a) for My penetrating drain (n - 1)
b) for partially penetrating drain with n - 4.s.
(1) Trench draix (2) Impermeable lower boundary. (3) Ponded upper boundary.

constructed down the line of steepest dope, the modified T" while the overall efilcieW of the drains, on any gN= cmm-section,
of qkzoMetric head will bf Siv= approxims ly by ho' a ho cas is
h cos-'A 1@v - hav Cos A ate- NW- Wo - Way . ho - hay
lie data an Fig. 12 have been used to show how the ratios hm/h wo Ito
and Who vary with Slho, for n values of 1.0 and 4.5 and Rk
Plots of 11 at 17, against S/ho can readily be made from the cuzves
hm in Fjp 13, as Sown later for q. It will be noted that both I and
As n so and 11M, 0 qav an independent of
V S7
this plot is "M for 0.
Although, for convenience, the ensuing discussion is carriedinaWy
The efficiency of tmch drains on a given crosmection is defined, in terms of 17 it should be bome in mind that the stabdWng effect
with respect to the intervening mass of soil, as of the actual tmnch drain installation will be a function of the
!2Lh' . ho - h true overall efficiency of the drains, which will depend on the
variation of i1a, over the site.
Wo 'h*

1.0

S/0 26

0.8- -0.8

2
0-6- 0-6

0.4-
-0-4

N..
0-2- %V Q-2

0-2 0-4 1 0-6 0-8 1-0
x / S
a) n=1 b) n =4-5

FIS.'12 Curves showing the variation of piesometric level (at drain invert level and assuming hd = 0) between trench drains, for Rk
I and various values of SID-
a) For fully penetrating drains (n - 1)
b) For partially penetrating drains (n - 4.5)
Not& The ratios SID and hJD Am dso t&kGn to be Approximately equivalent to S/h, and h/h, respectively. The effect of Rk;ft 1.0
can sim be obtained from these curves by the appropriate scale transformation.
t 0 Fig. 13 Curves showing the relationship between hmfh, and S/ho.
ho and between filho and S/ho. for fully penetrating drains
?z (with n = 1.0) and partially penetrating drains, with n
F; a 4.S. Rk is taken as 1.
he NS. The note appended to the caption (6t Fig. 12 also
h! applies here.

X.
M. 0. 1. JI le n /L-S Can records
a2-
Details of the performance of a number of trench and counterfort
drain installations are given in Table 1. Only natural slopes have

411M MIR wn im aw

wilt.. VOID H W@vg@ 42
Ora va vo OX 1. 1
a
log 2--X a
A 3'
Valli V
0 i
its is
0,
43

!S Nis Oil
1.9 j 4.@O g IVA..
o - .2 E 10,
a 181.0" A = A I -<1V 9 1 .. I ra JV
21 Jq a 0 %o
IL7 W
Va. %al
>.
CA
Alt!] P m
4-1 -9 1- Id's
41 1 .1 .
0 IN W, ?k
:A. 1.9-61 :2 1 V
%D
A E6 0
V
V Is
0 r
9 9
14 .9 -1 H 11 . I
I

%0
9A."'.9- I

Site Approx. Principal Drain details Drain performance
dope material -
Ref.No. Name 0 D 3 110 S/ho Ilm h" hm/ho 11 q4V T years
I Bredon Hill /Ur. Slip/ 140 Ur. Lies .8 3.0 11.3 9.1.6 4.35 .90 .720 .670 .33 .720 .740 1.2 Skem
11958
2 Sevenoaks /Lobo D1 7* Lr. Greensand .9 4.6 17.4 3.1 3.61 1.60 .95 .90 .32 .69 .71 2.2 Weeks
0 over Wealden
3 Guildford /P40241 7 London Clay .8 5.0 15.3 .3.97 3.8S 2.36 1.96 1.77 .59 .53 -.5S 10.3 Larsel
4 Hodson /EI-2/ so Gault clay .6 3.7 8.3 3.53 2.39 1;1-2 rg.1.62 1.51 .62 .34 37 1.3 Skem

5 Hodson /E34/ 5 Gault clay 6 3.7 8.3 3.53 2.39 r,.I.5 g.1.12 1.05 .42 ..68 .70 1.3
6 Burderop Wood 60 Gault clay .6 3.7 11.6 3.32 3.49 2.28 1.36 lA8 .69 .53 .55 1.6 Skent
IJ6-91 0
7 Burdetop Wood 6 Gault clay .6 3.7 25.0 2.90 8.86 c-.2.35 1.93 1.89 .81 .33 JS 1.6
11(9-141
a Boulby /N2-3/ 14 Till .5 3.2 11.0 i.86 2.26 2.52 2.02- 1.93- .52 .58* .60 5.8 Clevel

9 Boulby /T21 is Tal .5 S.5 6.0 2.9 2.07 1.0 .80* .740 .34 .720 .74 3.9 and A
10 Barnadale /DI-9/ lie Ur. Liao .7 3.3 11.3 2.77 4.08 1.35 .88 .83 .49 .68 .70 0.7 Chan
V

I I Barnsdale /D22-241 79 Ur. LIAS .7 3.0 11.3 3.43 3.29 3.0 2AI 2.27 .87 .30 .34 0.4
Notes. Ali dimensions in mattes. IP - estimated values. ojsbq an ra value of O.S.
TABLE 1. Details or the performance or some trench and counterfort drains In natural dopes.

200 -
M On ythe Beds

ISO X
Atherfi Silt
Atherfield Clay
W"d Clay --IDJ X

IDO 200 300 4OOM

b) X
Au

2-9

3-SA
0-5 A

Piezmeter
Nos. ft CM C14

Fig. 14: Comterflort drain inswitatioa at Seveafths. Kent (*ftWW*eJC1@l%9.1970. & SYMOAS & Booth,1971):
dinal esaian.
=U in piezoosetric JrM, A w. at -L9 as depth oe contretine bee wee drain as a Mult of The drainage (Measured c. I jeonth
b
aftw installation).
(1) Upper soliftction lobe. (2) Lower solifluction shaft. (3) Controllne of proposed mad. later reroute& (4) Extent of counterforts
CLA 14.5A). 9.1 se to each sido of the piezonmer Use.

In cwtrw to Sevenoaks, dds site Is an dw flank of = isclued based upon date kindly provided by Ckveland Potash Ltd. Again
knoll, with probably a relatively snall, Inflow of ground-water from dw period for which the pore-pressuin won monitored before
"dope, installation of the drains is shorter than desirable, but the very rapid
initial drawdown. diminishing as the depth of pinorinter tips below
A Ions-torm for ter P404 at Guildford, kindly pro- trotich invert level increases, is well shown for all three piezometers.
vided by Prof. N.E. Simons /19771, is compared with the pre- Of put interest is the somewhat disturbing fact that during the
dminage pmomewc loveh briefly recorded by a nearby poexamorter flM winter after installation of the drains, the piezometric leivels
/PI03/ in Fig. 18. Unfortsinaftly the original piezometer was son strongly, almost to their Pre-drainage values. This is particularly
destroyed during drain construction and not replaced for am* time, striking at pinameter N3*, where the pre-drainage artesian piezo-
so dm is no record a( the kmediate t1rawdown produced at ammic level of 1.68 m above original ground level was brought down
P1 03. not of dw behaviour In *0 succeaft 3 yum TIM IongAerm to at ban 1.3 m below this ground level in the September following
road, from March 1969 onwards to the present M nmmd*y
saady bowe"L A I"or of laminated soils and clays was noted in the borehole for
N3, in the vicinity of the piezometer tip. Dir P.R. Vauglian suggests
A shorter, but umm continuous Is provided bY a dadnese that an cuning the dram trench through this layer the silt layers
X

O@
- @X

Wsaiktion in a dop of dL with some sand and larnmied day would wash out, allowing the clay laminae to told at slump down,
layw% at Soulby. Yodmhize- This la overt In FI& 19, which is thus effectknir scaling off the more permeable laminae.

E IL W r

18-3 m

.4-57 -4-57 4-57 1 4-57
a. 9r 0.1 9 [-
Piezo. No& 215 209 211

4-6m

Fill. 13 Crosso4ection z - x of counterfort drain installation at Sevenuaks, Kew, shpwing the drea of drainage (after Weeksl%9, 1970).
Note.in this paper all such emss-sections; me drawn looking downslope. (I)Approximate piezometric level before drainage.
(2) Plezometric levels after drainage.

W E
O-Gm 14-6m O-6M

2-44 2." 2-" 2-" 2-"

Me= Nos. A R J2 J9

IS AA.70
2
---2 Sept-M
3-7M
---I Dec. 70

Doc- 71
Lj

Sg. 16: Cros*section of trench drain Installation at Burderop Wood, Wiltshire. showing the effects of drainage at various times after
Installation (after Skempton.1972).
(1) Approximate original position of the piezometric surface. (2) Positions of the piezometric surface following drainage (the drains do
Wei constructed between I and 3 July 1970). (3) Approximate position of the slip surface.

w construction of the drains in July 1971, only to return to an artesian
I pressm of + 1.37 above the same datum in the following winter
IMM VIL of 1971-72 /U. 90% of the September draw-down was temporarily
--4 losti. By the next winter of 1972-73, however, the drains am to
have taken hold, and this situation has been satisfactorily main-
tained up to the present
4
Although data are sparse, there an indications from other sites
that this stronil: rim in piezomettic levels during the first winter
after drain instillation may be a fairly general phenomenon. Some
I.-I law, but dpiricant, see nal rise may even occur in the second
Lj winter after Onstmil-tion, but after that the pment evidence suggests
that seasonal rise is strongly damped by the presence of the drains
and that around the yen reduction in piezometric levei is achieved.
Whether this can be regarded as permanent is of course quWion-
Fig. 17 Cross-section of trench drain Installation at GuHdfbr4 able, as discussed subsequently. An important lesson to be dra%,..n
Surrey, showing the long-term effect of drainage (largely fwni this delayed effect of drains is that some additional corrective
after Simons, 1977). (1) Approximate piezometric level meamse, such as a too M may be required to ensure stability.
before drainage. based on piezometer P103. (2) Position
of the piezometric surface about 10.5 years after installa- not only during the period of initial dtawdown and consoLidation
tion of the drains. but also during the first one, or possibly two, winters.

m aio

PLAN

N C0.08M

63-
'Tr

PIM PM

62-

Go-

S9-

S-R1
1265 1966 1957 1968 1969 1970 1971 1972 1973 1974 97S 1976 1977

Fig. IS: Long-term comparison of pic2ometric levels before and after drainage at Guildford, Surrey (largely after Simotks@1977@ (I)Treach
drain. (2) Sm&U lowering of ground Wei during construction of drain&. (3) Dowaslope.

Henkel /1957/ n=atks that the rate of drop in ponwan passures 74S for a typical drain installation with S/h a 3, for n values of 1.0
following trench at counurfaft drain construction is much mate and 4.S and R values of 1 and S. As can A men, the Various theo.
rapid than would be expamd from a calculabon band upon the jeficaL curves a not differ greatly.
cadricient of consolidation obtained from oedomew tesm More
recently Chandler /in praiV has moniWad this process for a treach Against this background in platted the available field obWrVRldOffL
dram installation in die Upper Lizz, and concludes that the observed All these am from natural slopes, except those from the cutting at
rate is at least 4 times as fast as that occurring in one-dimensional Cumnor Hill. to general then is an appreciable scatter, with the
consolidation in the field. predominantly through relief field observations falling on at below the theoretical curves. The
effects. theoretical values of fr for the curves shown, range from 0.76 to
0.90. It is suggested that in practical cam where only the mid-2aint
Dr. M. Hamra. of Imperial College, is currently examining the piezometric height, h in, is known, an estimate of the value of h can
immediate, undrained pore-pressm changes that are caused by the be --,fp by taitins da as a f h . with f s- 0.2. This approximation
stress rehmse consequent upon ouzvating drain trenches. Using Is probably slightly
t
non-linear finite element p, @ gi , he finds, for trenclics ex-
irted in saturated, overconsolidated clay. with S - 15 in. D a 5 in, in Fig. 21a, theoretical curves for the relationship between hm/h
X - 1.0, B 1.0 and h a 4.5 in. that the aver immediate and Sib we shown for n values of 1.0 and 4.5 and Rk Values bf
Quction in h for the central thre"uarters of the man between and S. Vith two exceptions /Bredon Hill and Seve&aks/ these
the eluong is -aWout 1.0 in for a = 0, and 1.5 in for a - -0,120 /a is bucket the available field data. Finally, in Fig. 21b. theoretical
as defined by Henkel 1%0/. This sum raican effect is clearly an curves for the variation of long-terni drain efficiency ff are Sivev,
important component of the rapid drawdowns observed soon after for similar values of a and Rk' A amparison of the available field
drain installation /eg. Fig. 19h Sivw lamurable boundary cmidi- data with thew mum leads to the following conclusions:
itiong, & further component will come from the following consolida- ji/ With the .exception of point I I for Bamsdale, which is rather
tion stage. short-terrn and whose efficiency is probably still increasing,
Compadion of theory with field behaviour none of the field points lies below the curve for a - 4.5 and
R k = I /curve Gi.
From the approximate theory developed earlier, the families of
curves in Figs. 20 and 21 are readily derived. These can be compared 1W Up to an S/h ratio of 5, the bulk of the field data lies between
with the field data stimmarised in Table I. cam C andothe curve for n a 1.0 Rk * 5*0/curve H/. If site
&wrap in considered, all the data, with the exception of
The aim of Fig. 20 is to determine values of the shape factor fe for point I for Bredon Hill, fail between curves G and H and lie
against
the piezometric imi& between drains. Thus h/h in is platted generally in the upper half of that sp
are available, but 5 has
Equiralent to & value of the poreTressurg parameter A of + 0.16. *ONo relevant measurements of the ratio Rk
been chosen as a 10tely upper limit for the materials involved.
blitchell /1956/ provided wine guidance for this choice.
77

It is suggested that, untfl further observations became available,
the curves H and G in Fig. 21b can be ngarded as tentative upper
Cr and lower bounds for the design of trench or counterfort drains in
W CID ft.0 overconsolidated clay slopes under temperate climatic condition&,
un un Q !D Ln up to an S/h ratio of 5 or 6. Higher ratios thanthis are unlikely to
be under @@on. Cum G is likely to be a conservative lomer
bound, unless in a given cam the is and Rk values or the hydrology
are particularly unfavourable.

All the drainage installations in natural dopes that are listed in
Table I appear to have been successful as a stabilizing measure. The
efficacy of such drains in cut slopes is less assured. and several
z examples of faures subsequent to the installation of counterfort
-7
drains in railway ;uttings in stiff fissured clays are given by Cassel
/1948/ and by Ayres /1%11. This is probably due in @W to d,
Peneraw slopes and shorter lengths of dziun mychled.

It b hoped to make a more detailed examination of tremch and
countediort drams, including an examination of the necessary ct@-
teris for preventing the occurrence of slips between such drains and
the effects of underdrainar, at of artesian pressures at depth, in a
fordicasning papert with Mr F-N. Bromhead.

aami at duins

Drainage, as mentioned earlier, is one of the most effective and most
and of stabilizing measura for slopes. At the sanse am. there is
generaW associated with such drainage installations some degree of
doubt cossanning their long-term performance.

Clogging of both pipe and permeable aggregate drains probably
occurs most cornmonly by the ingress and lodging of silts and fine
sands. Much attention has been devoted to this problem, particularly
to the design of filtats, and there is a correspondingly large literatiar
an this subject, briefly summarised, for example. by Spalding /1970,,
and CAdcrg= /1975J.

The matter is not Yet fully resolved, however, and as in practice
both specifications and workmanship, frequently fan below a good
standard. cum of dogging by siltation are probably quite common,
as indicated by Keene /1951/. Well detailed caw records of this
z type Of fail= me less so, but there are indications that permeable
aggregate dritinage systems without proW filter protection my
tend to silt up after a working life of about 10 to 20 years.

Ln In the case of some bated hortcontal drains in Derbyshire, the
following information has been kindly provided by Mr C.V. Under-
w , the County Surveyor /pers. comml. Fifteen perforated PVC
drains, of internad diameter 40 mm and lengft between 3 2 and 40 m.
were insuLled without filters in 1971 at an angle of 5 above the
horizontal through laminated silty dAyL By 1975 all the drains
had became inoperatim largely through siltation. The earlier ho-
nizontal drain installation at Otley in Yorkshire, however, ca&
structed in 1964-67 using galvanised iron drains provided generally
with pre-cast porous concrete filters, still appears to be working
satWactatily /GJL Forrestet, pers. catzintl.
CID Some Other ways which drams become ciogpd an listed by Keene
/19SI1. NeVtfth9le21, after siltation it is probable that po. and bio-
chemical factors pow the greatest threat to the satisfactory operation
Of drainage systems. yet this subject has been laWly neglected by
-4 civil engincers. There is a fairly extensive literature an the subject,
0 however, in the firlds of agricultural engineering and water supp!y-
3 One Of the most* common geochernical effects is the precipitation
of hydrated ferric oxide /iron ochre/. Useful studies of iron */and
ochre in near-surface agricultural drains have been mad:
by Alcock 119731 and by Thorbum &Trafford /19761. The lana
-#-OW z authors distinguish two man groups of ochre fanning soils; peats
z and slightly OrPnic marine sediments and pyrite-bearing rocks.
A detailed discussion of measures adopted to prevent ad= depL*
sition Occurring in a deeper, civil engineering drainage system in
Antwerp is Sinn by Brand /1969/. In the earth dam context.
problems ansing from the geochemical and biochemical precipitation
of iron compounds we discussed in a pioneering paper by Wanti &
/1974/. They present a photograph showing a sand filter ce-

Fig. 19 a, b) 78

Ml too
on, Ow M, Am

m
2 N3 reinstated
Piezorneters
io

.0
31 Hkg5j: N
P. T V.I., Nil I
Igi. 3* 1 a - 0
ILI. 0 a
4--n
;P 12
n
H-ato -woo
X Z 9 "r-
0 =I- X w broken
CLIX I > R-- t f 0
2 -
g-j 5 . . t -9 910
PC;, .19- - : @ 0. ;- 10,
=tmj 1.3
CL
'; ;
z A
t- ! -- N2
CL v S,
Z broken
CL
C1
;;aj F reinstated

NI unblocked

blocked
it". 5:oC F-
H w
0 :C 8-
COL
N3

ion IFeb.lMar 'Apr, IMayl Jun!Jul. IAuq!S9p!OcV Nov.'Decl
1971 1972 `1973 1974 19

1.0

S
=3, n =I-O.Rk=S

0.8-'
h

0-6- A

S 3, n 0-0. P* 1
ho

s
0-2- 'FO =3. n=4-5, NX1

0 0.1 0-2 0 -3 0.4 0.5

X/S

Key: m Savenocks. Lobe 0 a Barnsdale . 07-9
A Burderop Wood. J 2 & J6-9 + Barnsdaie. 022-24
Hodson. avg. E 1-A. x Cunmor Hill 115-17
F[S. 20 Computed fWatJoMh* bw"M h/hIn wd XIS tot 4 1 Awd S Old M` I and 4.S. COMPGrOd wfth the awadaMe flaid date.

1.0

0.8 -

04
ho 0-5- 03
02
0-4- 154 leo, n =1-0
Rkz 5
09 01 n =4-5

0-2- n =1-0
X@n=4-5 RI,
0

Fig. 21 a)

b) 1-o

0-8-
91 2
1% .01o n 1-0
8@ Rk=5
0-6- - "14-. '*@ . n=4-51
4 6 03

0-4-

,H
0-2- Rk= 1 n=1-O
n = 4 - 5

0
0 2 4 10
S/ ho
Fig. 21 Comparison of approximate theory for trench drains %%ith the field data summarised in Table 1.
a) Computed.Magiombp between hmlll and Stho for Rk - I and 3 and is - I and 4.5. compared with the field data.
(For key to the numbering of the point, (in Figs. 2 1 ) and b)). refer to Table 1. Solid circles indicate the more reflable data).
b) Computed relationship between ;I and S/ho for Rk - I and 5 and n = I and 4.3, compared with the field data.
(in both Flip 2 la) and b). points from thip same site are connected by a broken fine.)

mected up by such ptomsses, whicit are probably akin to thosc The paper by Bazyfiski A Frankowsid describes the investigation
which form iton pans in naturAl ground. There is also some experien- and stability analysis of a 'tacit' de affecting a railway cutting at
cc of well screen dogging through bacterial and other wtivity Sadowie. The slide originated in 1934 and involves loess overlying
!M.S. Esunton, pem comm./. Miocene dayL In the investigation, Dutch penetrometer tesu were
much used, but it is not den to what extent them were useful. In
the stability andysts of this slide on pre-existing slip surfaces it
Cam of dogging through the precipitation of other materials is disturbing to find no mention of residual strength, and that the
seem to be less common but bkwJmp of drains, flowing partly full. strength testing and stability analyses am carried out in terms of
by encrustations of calcium carbonate have been reported from total stream
limestone areas.
Ark interesting, though tantalisingly brief, description of landslide
Drains made from artificial fabrics are currently being developed problems an about 600 km of mountain roads in N. Bengal and
for use in - civil engineering /eg. Healy & Long.1972, Long &Healy, Sikidm, is giveti by Chopra. The complexity of mass movements
19771. Then sum promising and do enable the manufacture of in such a region, and the importance of good road location. surface
the filter element. to be closely controlled under factory. rather drainage and the control of erosion, am well brought out.
thin ate, conditions. Preliminary tests howmr, iridicate that
dogOng by Air is still a problem /Hoogendoom &van der Mculen, In his pa; PY, Fujita reviews the occurrence of slides around reservoir
1977/. Further development and continuing testing of these mater- margins in Japan, with particular reference to the Influence an
Ws is clearly neonsary. The scope of the testing should also be these of variations in reanwir wales level. The paper gives a number
widened to include pochemical and biochemical furors. of can records of Aides occurring both on flooding as well as
on dmwdown. Much of the work done at Vitiont /eg. Kenney. 1967/
2 - SUMMED PAPERS is relevant to this topic and one would also have expected to find
The 10 papers on Them 3 cover the following main topics: site reference to the concept of critical pool level.
lirvestiption, regional landslide studies, various slope stabilization Mend, Papoulek & Paseks report an a cam where an unstable
works and dynamic effects. slope of marly Neogene days NW of Brno was successfully built
upon. Despite complex geological ancVhydrogeological conditions,
The paper by Yarnapichi describes the use of repeated electrical useful backcalculations of the stabill' Oya of some of the landslides
resistivity surveying an a recent landslide in order to guide in tl)e am made. As often found elsewhere, at comparable stim ranges,
loc;ition of dminage wells and to check their efficacy. Wells are the laboratory value of residual strength lies 2 or 3 degrees below
sited in zones of low apparerit electrical resistivity, considejed to the values obUined by back analysis. The slopes were stabilised
correlate with an abundance of ground-water, or greatest slide mainly by a combination of an anchored pillar wall and some
duger. The rust such well gave a strong discharge and a subsequent horizontal drains and by minimising the size of cuts for roads and
'Veat of the resistivity survey showed that the associated zones of sewers. No monitoring of these meiisures is reported.
low apparent resistivity had disappeared. The method seems a very
indirect way evaluating ground-water pmaure distributions but is The paper by Flimmal deals with horizontal drain and pile %%-all
clearly much cheaper and quicker than the conventional one using installations in a more general way. Useful practical experience,
Piezometers. It could add to the value of a future trial of this pined on numerous landslide stabilisation schemes in Czechoslo-
method if the pre- and post-drainage situations were to be checked vakia. is given for both the above types of corrective measure. Some
with piezometers. guidance is also given on the drain discharges and lowerings of

ground-water level achievc& No mention is made of filters being Concerning the considerable variety of other corrective techniques,
provided on the horizontal drains. only three research needs will be touched upon. In the car@e of p6r-
manent soil or rock anchors, efforts should be made to explore
A brief account of the stabilization of an 80 m high rock slope in more fully their lorWterm behaviour, particularly with regard to
N. Bohemia, mainly by the use of pre-stressed, horizontal cable stress loss and corrosion. For other corrective structures, and
anchors, is provided by Zajfc. An outline is given of the comprehen- particularly for heavy cantilevezed or anchored pile restraining
sin joint survey in&& and of the anchor design. It is to be hoped structures there is @ a need for more measurements of deflections and
that long-term monitoring of this installation, which is currently earth pressures, and study of the relation between these. There is
being carried out, will be undertaken. also room for further exploration of the extent to which geophysical
investigation techniques can be of help in stabilization work.
The final three papers all deal with dynamic aspects of slope stabili-
ty, with particular ref@rence to blasting. Musselyan, Kochurov &Lj- Finally, in view of the dose link between landslides and erosion,
Yrusevitch describe field experiments in which the response of it would seem advisable to direct moire attention towards methods
to= slopes to blasting was examined. The blasting was not cairried of controlling the latter, on both the local and the regional Scale.
out until the settlement of the loess, following preliminary wetting
had been completed. The seismicity factor measured, for which the
slope movements are known, is compared with those associated
with natural earthquakes. Dvaidic discusses the technique of re-
moving parts of a landslide by blastin& for Instance as part of the The author is grateful to Mr. E.N. Brombead, of Kingston Poly.,
stabilization of a rock mass. He gives an example of a rock slide technic, for running the stability analyses for the various influence
unintentionally initiated by blasting and suggests an approach lines, for carrying out the finite element seepage analyses, and for.
whereby the appropriate size of blasting charge, that will not helpful dIscussionL He also wishes to thank Dr.E.G. Youngs, of the
endanger the ovetall stability, can be detaminedL Finally, in a A.R.C. Unit of Soil Physics, for a useful discussion. At Imperial
rather general. paper, Shahunants A Fedorenko discuss principles College he wishes to thank Prof A.W. Skempton for his generous
of slope stabilization and management in folded mountain regionL provision and discussion of unpublished data and Professor A.W.
Then include, in addition to normal slope stabilization measures, Bishop, Dr. R.J. Chandler, Dr. M. Hamza. Dr. SX Sarma and
the provision of dykes to deflect or retain landslide and mudflow Dr. P.R. Vaughan for various discussions. He is also indebted to the
debt* the provision of overflow channels around dams formed b librarians of the Civil Engineering Department, and to many fellow
y
large natural landslides and the use of conventional and nuclear and engineering geologists elsewhere, for help in covering
explosion to shake down threatening masses. In the latter connec- the relevant literature.
tion it is pointed out that reliable methods of predicting the run-out
of slides am not yet avagable. For the Guildford data, the author acknowledges with thmks the
contributions of Prof. N.E. Simons, of the University of Surrey,
3 - RESEARCH NEEDS Prof R.E. Gibson, of King's College, London, and Mr. J.H. Arm-
strong and Mr. A. Aarons of the Building Design Partnership. For
Development of out knowledge of the efficacy of slope stabilization the Hodson and Burderop Wood data he is grateful, not only to
measures depends primarily on comprehensive and long-tam Professor Skempton, but also to Sir Alexander Gibb and Partners
monitqring of the performance of the various methods in use. and the South-Eastern Road Construction Unit, while for the
In general, the essence of such monitoring is the cairrying out of Boulby data he is indebted to Mr. A. Brockbank and Mr.C.G. Patti.
We operations, such as the measurement of movements, stresses. son, of Imperial Chemical Industries Ltd Agricultural Division,
pore-water pressures and drain discharges, in a well planned and and to the Cleveland Potash Ltd. For the information on Bredon
thorough manner. In other words, although there is obviously Hill and on Cumnor Hill he is watel'44 respectively, to Professor
room for its further development@ stifEcient technology already A.W. Skempton A Dr. D.J. Henkel and Coventry Corporation, and
exists; it is interest and determination that no crucial if a valuable to A.G. Weeks and Partners and Mr. J. Peverel-Cooper, County
body of detailed cast records is to be built up.. Surveyor and Engineer. Oxfordshire County Council.

In the case of corrective cuts and fUls, the main pre-requisite of References
success is a proper investigation of the landslide itself. The monito-
dug of movements, it the wound surface and preferably also with ALCOCX M. 11973/: A survey of ochrous land drainage sites in
depth, should be an invariable practice and the values of 9 dugua& England and WaICL Min. of Ag., Fisheries &Food. Field DrAi-
construction should be measured by suitable piezometers. The main nage Experimental Unit, Tech. Bull. 73.2, 1-7.
research needs heire concern the difficulties of s the degree ANDREWS G.H. - KLASSELL J.A. /19641: Cylinder pile retaining
of stability of the existing slopes or landslides and of deciding what wall. Highway Research Record. No. 56, 83-97.
is an appropriate degree of improvement. In this connection. more ANON /1 %51: Cliff stabilization scheme. East Cliff , Bournemouth,
knowledge of stress-strain conditions throughout the whole landsLip being protected with sand draiftL The Surveyor and Municipal
mass especially in the approach to failure is required. This advance Engineer, 125, /2 January/, 25.
beyond the constraints of the limit equilibrium approach will also AYRES D.J. /196 1: The treatment of unstable slopes and railway
need to be accompanied by a re-examination of the usual ideas r_1 __ alck formations. J. A Traim Soc. of EnM-52, I I I- I lit __' 11
ofJactor of safety-. BAUR R. trot and correction
- In: Landslides and engineering practice. /ed. E.B. Eckel/.
With respect to drainage measures; then is a great need for 150-188. Highway Research Iloard-Special Report -29.---- --- ---
monitored performance records of all the methods. The initial %4cr -A. 119 741: Example of slope st2bili-
wound-water conditions should first be established for at least sation in marly-saridstone Flysch. Proc. 3rd Cong. Int. Soc.
one full mason, and preferably longer. In addition, measurements Rock Mech., Denver, 11-8, 759-764.
of the variation of permeability with depth, and the permeabdity BARATA F.E. /1%9/: Landslides in the tropical region of Rio de
ratio kh/lcv are highly relevant. As drainage can take a considerabi .e Janeiro. Proc. 7th Int. Conf. Soil Mech. & Foundn Engrg.
time 16 bicome effective, long-term monitoring of Its prowess is Mexico, 2, 507-S 16.
essential. This needs to cover not merely the consolidation phase BARRON X. -COATES D.F. - GVENGE M. 11971/: Support for
but also the succeeding steady state phase. in order to discover any pit slopes. Canadian Institute of Mining & Metallurgy Bull.
deterioration through clogging or other causes. 64, /March/, 113-120.
BAZY@ISKI J.-FRANKOWSK1 Z. 11977/: Site investigations
In most drainage systems, the danger of dogging is usually present. and calculations of the stability of slopes in the landslide arc:1
Much effort is required hue, not only In a research context but also at Sadowic near Cracow. Symp. Int. Assoc. Engrg Geol..
to devise mum of protection against ciog8ft which will be satis- Prague. Paper on Theme 3.
factory under site conditions. It is also important for this research BELES A.A. - STANCULESCU Ll. /1958/: Thermal treatment
to cover clogging through geo- or bio-chemical processes as well as a means of improving the stability of earth masses. Gcutcchni-
that which can arise by siltation. qua, 8, 158-165.

BISHOP A.W. /19541: The use of the dip circle in the stability COLLIN A. /19"1: Recherches expirimentales suf les glissements
analysis of slopes. Proc. European Conf. an Stability of Earth spontanis des terrains argileux. Paric Carilian -Goeury
Slopes /StockhoW, 1. 1-13, and Geotechnique, 5, 7-17. et V. Delmont.
91SI10P A.W. - BJERRUM L /1960/: The relevance of the triaxial COPPERTHWArM H. 11880A Earthwork dips on the Leeds &
too to the solution of stability problems. Proc. Research Wetherby Bran6 Railway. Min. Proc. Insta Civil Engr3.,
Conf. Shear Strength of Cohesive Soils, Boulder, Colorado, 62, 285-287.
437-501. COSTA NUNES AJ. /1966A Slope stabilization; improvements
BJERRUM L - JOHANNESSEN I.J. /1%0/: Pore pressures resul- In the techniques of prestressed anchorages in rocks and
ting from driving piles in soft clay. Conf. Pare pressure & soils. Pne. Ist Cong. Int. Soc. Rock Mech., Lisbon, 2,
suction in soils, 14-17. London: Butterworths. 141-146.
BJERRUM L - JORSTAD IF. /1968/: Stability of rock slopes in COSrA NUNES AJ. 11%9t* Landslides in soils of decomposed
Norway. Norwegian Geatechnical Institute, Pub. No. 79, rock due- to intense rainstorms. Proc. 7th Int. Conf. Soil
1-11. Mach. &Foundn Engr&, Mexico, 2, 547-S54.
BJERRUM L-L(3KEN T.-HEIBERG &-FOSTER R. /1969/: COSrA NUNES AJ.-VELLOSO D.A. /1963A Estabiliza;1o de
A field study of factors re3ponsible for quick day dides. taludes em capas residuan de ongern gramtolpamuca.
Proc. Ith Int. Cont Soil Mach. &Foundn. Engra Mexico, Proc. 2nd Panam. Conf. Soil Mach. &Foundn Engrg., Sla
2,531-540. Paula, 2,383-394.
BJERRUM L - MOUM J. - EME 0. /1967/: Application of electro- D-APPOLONIA E. - ALPERSTEIN R. - VAPPOLOMA D.J. /1%7/:
osmosis to a foundation problem in a Norwegian Quick Clay. Behaviour of a calluvid slope. J. Soil Mechs AFoundns
Geotechnique, 17, 214-235. Div., ASCE., 93, /SB441,447-473.
BRAND E.W. /1%8/: Solution of an unusual subsurface drainage DE BEER E.E. /1%9/: Experimental data concerning clay dopes.
problem. J. Irrign &Drainage Div. A.&C-L. 94, IM2/.199-221. Proc. 7th Int. Conf. Soil Mech. &Foundn Engr& Mexico,
BRANDL H. /1976/: Die Sbcherung van hohen Ansctmittan in 2,517-525.
DE BEER E.E. -WALLAYS M. /1970/: Stabilization of a dope in
sutschgef[hrrietan Verwitterunpb6dwL Proc. 6th European by mum of bated piles reinforced with steel bUML
Conf. Sod Mach. & Foundn EnSM Vienna, 1.1, 19-28. achi
BRAWNER C.O. /1959/: The landdide problem in British Columbia Proc. 2nd Consr. Int. Soc. Rock Mech. Belgrada, 3, 36 1-
highway construction. Proc. 13th Canadian Soil Mech. 369.
Conf., 45-60. DOCKRAY J.& /1844/: Discussion on Gregory /1844/. Min. Proc.
BRAWNERC.O. 11971A Cam studiesof stability onmining projects. Instri Civil Engrs, 3,148-150.
Chap. 12 in Stability in Open Pit Minmg. /Eds. Brawner DUNCAN J.M. /1971A Prevention and *correctlon of landdideL
C.O. & Millipn V.1 Proc. I st Int. Conf. Sotbility in Open 61h Ann. Nevada Street &Highway Conf. Section 11. 1-42.
Pit Mining, Vancouver, 205-226. 'DUNCAN J.M. /19761: Notes an slope stabilization. University
BROMS B.& /1%9/: Stability of natural dopes and embankment of California, Balteley. Unpublished.
foundations: Discussiom Proc. 7th Int. Cont Soil Mech. DUVIVIER J. /19401: Cliff-stabilization works in London Clay.
&Foundn Engrg, Mexico, 3, 30-394. J. hum. Civil Engrs, 14, 412-426; Discussion 427-446.
BROMS B.B. /I 975A Landslid*L Chap. I I in Foundation Enginee- DV00AK A. /19771: Landslides caused by blasting. Symp. Int.
ring Handbook /Eds. Winterkom, H.F. &Fang H-Y/, 373- Aesoc. Engrg Geol., Prague. Paper on Theme 3.
401. New York: Van Nostrand Reinhold Company. EARLY JLIL-SKEMPTON A.W. /1972/. Investigation of the
BROMS B.B. - BENNERMARK H. / 19681: Shear strength of soft landslide at Walton's Wood, Staffordshim Q.J. Engrg Geoi.,
day, Discussion. Proc. Gcouch. Conf.. Odo, 2. 118-120. 5,19-41.
BROWN CIL, SHEU ILS, /1975/: Effects of do-forestation on do- EAST G.R.W. /1%741: Remedial mesiures and cam histories:
pes. J. Geotachnical Div. ASCE, 101, /GT21, 147-165. Inclined plane dope ftilures in the Auckland Waitemata
BUJAK M. - GLAB W. - MORACZEWSKI K. - WOLSKJ W. / 1967/: soils. Three can with different remedial measuraL Proc.
Preventive messures against the rock *dide at Tresna Dam Symp. Stability of Slopes in NaturW Ground, Nelson, /New
site. Proc. 9th Conp. Larp Dams, Istanbul, 1, 1027-1036. EGG Z=Iaqd Goomechanics Societyi, 5.17-5-U.
CAMBEFORT H. /1%6/: Les ouvrages ancris au sol. Travaux, ESTAD L - SEM H. /1976A Stability of excavations impro-
Mai 1%6, 834-947. ved by salt diffusion from deep WdIL P=. 6th European
Conf. Soil Mech.&Foundn. Engrg, Vienna, 1.1, 211-216.
CASAGRANDE L - LONGHNEY R.W. - MATICH M.A.J. /1%1/: FINZI D. - MCCOLAI C. / 1% 11: Slope consolidation of the bmb
Electro-ounatic stabilization of a high dope in loose 53- of the Monguelfo Reservoir, Italy. Proc.'Sth Int. Conf.
turated sit Proc. 5 th Int. Conf. Soil Mech. & Foundn. Sod Mech. AFoundn Engr& Paris, 2,591-S94.
Engrg Pads,2,555-S61. FLU01MEL L /1977A Horizontal drainage borinp and cast-in-situ
CASAGRANDE L - McIVER B.N. / 197 1/: Design and construction pHs wells as stabilization treatment of landslides in sedi.
of tailings dams. Chap. 11 in Stability in Open Pit Mining mentary rocks. Symp. Int. Assoc Engrg Geol., Prague.
/Eds. Brawner C.O. & Milligan. VJ Proc. Ist Int. Conf. Paper on Theme 3.
Stability in Open Pit Mining. Vancouver, 181-203. FOOKES P.G. -SWEENEY M. /1976/: Stabilization and control
CASSEL F.L. /1948/: Slips in fissured clay. Proc. 2nd Int. Conf. of local rockfalls and degrading rock slopes. Q.J. Engrg
Soil Mech. &Foundn Engr& Rotterdam, 2,46-50. GeaL, 9, 37-55.
CEDERGREN H*R, /1967': Seepage, drainage and flow ntL New FUJrrA H. 11977/: Muence of water level fluotuations in a re"
York: John Wiley &Sam Inc. servoir an slope stability. Symp. Int. Assoc. Engrg Goal.,
CEDERGREN H.R. /197S/: Dminage and dewatering. Chap. 6 in Prague. Paper on Themi 3.
Foundation Engineering Handbook./Eds. Winterkom H.R. GOLDER H.Q. /1971/: The stabilization of dopes in open-pit
& Fang H-W, 221-243. New York: van Nostrand Reinhold mining. Chap. 10.in Stability in Open Pit Mining. fEds.
Company. Brawner, C.O. &Milligan VJ Proc. Ist Int. Conf. on Sta-
CHANDLER R.J. /in press/: Back analysis techniques for slope bility in Open pit Mining, Vancouver, 169-180.
stabilization works: a can record. Geotechnique. GORDON G. /1937/: Freezing arch across toe of East Forebay
CHOI Y.L. /1974/; Design of horizontal drains. J. Engrg Soc. of Slide, Grand Coulee Dam. The Reclamation Era, 27,12-16.
Hong Kong, Dec. 1974, 3 7.49* -AY D.H. /1970A Effects of forest deu-cutting on the stability
CHOPRA B.R. /19771: Landslides and other maw moven ents of natural slopes. Bull. Assoc. Engrg Geologists, 7, 45-66.
along roads in Sikkim and North Bengal- Symp. Int. Assoc. @GRAY DAL /19741. Reinfomement and stabilization of soil by
Engrg Geol, Prague. Paper on Theme 3. vegetation. J. GeoteclL Engrg Div., ASCE, 100 tGT6/.
COLAS G. - MOGIER 1C. - URSAT P. /1976/: Stabilisation des sots 69S-699.
stiriles par enherbernent. Bull. do Haison des laboratoires GREGORY CH. /1844/: On railway cuttings and embankryients
des pants et chaussies, Stabititi des talus". Numiro spe- with in account of some slips in London Clay, on the
cial HL 161-168. line of the London and Croydon Railway. Min. Proc.
COLENUTr G.W. 11928/: The cliff-founder and Landslide at Gore Instn. Civil Engrs, 3.135-145.
Cliff, Isle of Wight. Proc. Isle of Wight Nat. Hist. & Ar- GUNLEMMUD T. /1%8/: Frysestabiliserinit i forbindelse med
cheol. Soc, 1, 56 I-S70. 83

tunnelms. Nordiska Geoteknam Matet. Vig-och vatten LONG R. -HEALY .X. /1977A Fabric filters on pre-rabricated
bygpmn, 8, 30-33. underdrains. Int. Conf, an Use of Fabrics in Geotechnics,
HANDY R. L -WILLIAMS W.W. /I % 7/: Chemical stabilization of Paris, 2, 237-24 1.
an active landslide. Civil Engineering, 37,62-65. LUITON R.J. -BANKS D.C. /1970/: Study of clay dopes alone
HEALY X.A.-LONG R.P. /1972A Prefabricated fliter-fin for the Panama Canal. Rept 1. East Culebra and West Culebra
subsurface drains. J. Irrign & Drainage Div. ASCE. 98, 'Odes and die Model Slope. U.S. Army Engineer Water.
/IR4/, 543-552. ways Experiment Station, Vicksburg, Mississippi.
HENKE X.F. I 1 %9/: B83chunpsicherung dutch borizontale Drains- MATSUBAYASHI M. /1972/: On the groundwater CUt-Ofr Work
sebohrungen. Slrat n u. Tieft, 22, 74-80. with chemical groutings and its effectiveness in landslide
HENKEL D.J. /1957/: Investigations of two long-term failures in arm Landslide, 9, 21-32.
London Clay slopes at Wood Green and Northolt. Proc. MEHRA S.R. - NATARAJAN T.K. /1%6/: Handbook on landsi;de
4th Int. Conf. Sail Mech. & Foundn Engrg. London. 2. analysis and correction. New Delhi: Central Road Res.-rach
315-320. Institute.
HENKEL D.J. /19601: The shear we gth of saturated resnouWed MEHRA S.R. - NATARA;AN TX . GUPTA S.C. /1967/: Practical
clays. Proc. Reseuch Conf. Shear Strength of Cohesive lessons in landslide correction. Road Research Papers.,
Soi)s, Boulder, Colorado, 533-554. Delhi, No. 89. 1-3 1.
HILL R.A. /19341: Clay suatum dried out to Prevent landslipL MENCL V. - PAPOUSEX 7- - PASEKA A. / 1977/: Building on a
Cirtil Engineering, 4, 403-407. luddide am situated on the boundary of the Carpathian
HILLS R.C. /1971/: The influence of land management and soil foredeep. Symp. Int. Assoc. Engrg Geol., Prague. Paper on
dwacteristics on kfiltration and the occurrence of Theme 3.
land flow. J. Hydrology. 13, 163-191. MITCHELL J1C. /1956/: The fabric of natural clay and its relation
HOEK I- -BRAY J.W. /1974/: Rock dope engineezin& London: to engineering propertieL Proc. Highw. Res. Bd., 35,
Instn of Mining &Metallurgy. 693-713.
HOLM O.S. 11%11: Stabilization of a quick clay dope by vertical MITCHELL J.K. /1970/: hi-place triestment of foundation soils.
sand drains. Proc. 5th IAt. Conf. Soil Mach &Foundn Engrg. J. Soil Mech & Foundns Div., ASCE. %, /SM l/. 73-110.
Pub, 2,625-627. MITCHELL JX 11976/: Fundamentals of sod behaviour. New York:
HOOGENDOORN A.D. -VAN DER MEULEN T. /1977/: Prdimi- John Wiley &Sons. Inc.
nay investigations on clopft of fabricL Int- Conf. on Use MORGENSTERN N.R. -PRICE V.E. 11%51- The analysis of the
of Fabrics in Geotechnics, Paris, 2, 177-182. stability of general ft surfaces. Geotechnique, 15, 79-93.
HUTCHINSON J.N. /I%8j:Mass movement. In The Encyclopedia MOUM J.-SOPP O.L-LOKEN T. /1%8/: Stabilization of un-
of Geomorphology /Ed. R.W. Fairbridp/, 689495. - disturbed quick day by salt WCJIL Norwegian Institute
HUTCWNSON J.N. /1%9/: A reconsideration of the coastal land- Pubn. No. 91, 1-7.
slides at Folkestone Warren. KenL Geotedusique, 19,6-38. MUSAELYAN A.A. - KOCHUROV Y.G. - LANRUSEVITCH LV.
HUTCHINSON JA 119731: The response of London Clay cliffs to /1977A Some research results of dynamic stability of
differing rates of toe erosion. Geologia Applicata e Idro- 4opes formed of collapsible sod& Symp. Int. Assoc. Engrg
geologia, 8, 221-239. Geot., Prague. Paper on Theme 3.
HUTCHINSON J.N. - SOMMERVELLE S.H. - PETLEY D.J. /1973A NATIONAL CONFERENCE OF LANDSLIDE CONTROL /1972/:
A landslide in perigiacially disturbed Etruria Mari at Bury Landslides in Japan. The Japan Society of Lindslide.
Hill, Staffordshire. Q.J. EnV& Geol.. 6, 3 77-404. NEWMAN J. /1890/: Earthwork slips and subsidences upon
INFANTI N.Jr. - KANJI M.A. /1974/: Preliminary considerations Works. London: E. &F.N. Spon.
on geochemical factors affecting the safety of earth dams. NILSEN K.Y. - LIEN R. /1976/: Description of a drainage gallery
2nd Int. Congr. Int. Ass= Engrg Geology, SEo Paulo, 4, for stablilising the left abutment of Muravatn Dam, South-
33.1-33.11. ern Norway. TranL 12th Int. Congr. on L4rV Dams, Mexi-
KEENE P. 11951t: Present practice in subsurface drainage for co City, 2, 4 t5-427.
highways and sirporm Ifthway Research Bull. No. 45, NONVEILLER E. /1970/: Stabilization of landslides by means of
1-20. horizontal borings. European Civil Engineering, 5, 221-
KENNEY T.C. 119671: Stability of the Vajont Valley slope. Fels- 228.
mechanik u. Ingenieurgeologie, 5, 10-16. NORTH-LEWIS J.P. -LYONS G.H.A. /1975t: Contiguous bored
KENNEY T.C. - PAZIN M. -CHOI W.S. /1976/: Horizontal drains piles. Proc. Conf. an Diaphragm Walls & Anchorages.
in homogeneous slopeL 291h Canadian Geotech. Cant, lustn Civil Engrs, 189-194.
Vancouver, VLI-VLI5. PALMER LA. - THOMPSON J.D. - YEOMANS CM. / 1950/: The
KEZDI A. /1 %9/: Landslide in loen along the bank of the Danube. control of a landslide by subsurface drainage. @roc. Highway
Proc. 7th InL Conf. Soil Mech & Foundn Engrg. -Mexico. Research Board, 30,503-509.
2,627-626. PARROTT W.T. 119SSI: Control of a slide by vertical sand drAinL
KLENGEL K.J. - HAMMERSCHMIDT IL - MOLLER E. /19741: Highway Research Board Bull. 115,51-52.
Bbschungssichwung und -sanicrung bei Autobahnneubau. PECK X& -IRELAND H.O. /1953/: Investigation of stability
Z. pol. Wiss. Berlin, 2, 371-380. problefnL Am. Railroad Engrs Assoc. Bull., 507, 112-133.
LANG H.J. led./ / 1976A Nstiondstrasse N8 in Obwalden. Bauab- PECKOVER F.L. -KERR J.W.G. 11976/: Treatment of rock slopes
schnitt AIpvachstad4Cantwqp= Nidwalden. Mitt. Inst. on transportation rOUteL 29th Canadian Geotech. Conf..
hir Grundban a. Bodenmechanflc ETH Zilrich, Nr. 106. Vancouver, L144.40.
LANG T.A. /1%11: Theory and practice of rock bolting. Trans. PRICE D.G.-KNILL J.L /1%7/: The engineering
Amer. Instn Min. EnSM 220,333-348. Edinburgh Castle Rock. Gootechnique. 17, 411-432.
LA ROCHELLE P. - LEFEBVRE G. - BILODEAU P.M. /1976A RAT M. /1976/: Drainages. BuIL de liaison des laboratorires des
The stabilization of a dWe in Saint-Terome, Lac Saint- Pont& et chaussies, Atabiliti des talus', Numiro
Jean. 29th Canadian Geoftch. Conf., Vancouver, XI.37- III, ISI-160. special
XL53. RICO A. - SPRINGALL G. - MENDOZA A. /1976/: Investigations
LEFEBVRE 0. - LAFLEUR J. /19761* EValuation of vertical drai- of instability and remedial works on the rijuana-Ensenada
nage as a stabilizing agent in a clay dope at Hall, Quebec. highway, Mexico. Geotechnique, 26,577-590.
29th Carumilan Geotech. Conf, Vancouver, VI.16-VI.30. ROBINSON B, /1967/: Landslip stabilization by horizontaily
LITTLEJOHN QS.-NORTON ?J. i-TURNER MJ. /1977/: bored drainL Highways &Public Works, 35, /June 1967.
A study of rock dope reinforcement at Westfield open pit -No. 1690/, 32-33, 35 &37.
and the effect of blasting an pre-stressed anchors. Proc. RODDA J.C - DONVNING R.A. - LAW F.M. /1976/: Systematic
Conf. on Rock Engineerfiq6 University of Newcastle upon Hydrology. London, Boston: Nownes-ButterworthL
Tyne, 293.3 10. RODRIGUEZ AX - MENCHACA L.M.A. - PECERO G.M. /1967!-
11TVINOV I.M..RZHANITZIN B.A.-BEZRUX V.M. /1961/: Estabilizacion de un deslizarniento en la carretera Tijcan2-
Stabilization of soil far constructional purpose. Proc. Ensenada. Proc. 3rd Panam. Conf. Soil Mech & Foundn
Sth InL Conf. Soil Mech & Foundn EW&. Par* 2, 775- Engrg, Caracas, 2. 51-64.
780.
84

Selections from

REGULATIONS TO REDUCE COASTAL EROSION LOSSES

Prepared by

D. A. Yanggen

University of W isconsin-Ex tension
for the
Wisconsin Coastal Management Program
January 1981

DETERMINING THE EROSIO N HAZARD

There are two basic approaches to determining erosion hazard -- the

site specific method and the reach method. The site specific method requires
a geotechnical engineering analysis at each site at the time development is
proposed. This method requires a report analyzing among other things: (1)

wave-induced erosion based upon recession rates and wave energy calculations;
(2) geologic conditions including the soils at the site and their properties
and stability; and (3)*groundwater and surface later conditions.. While the

site specific approach may be technically Accurate, it is too costly and

time consuming for all but the most expensive development.

The reach method uses generalized formulas to estimate the two components

of the erosion hazard, i.e., the recessi.on rate and stable slope angle. Much

of the information needed is available from studies made through the Wiscons in

Coastal Management Program. There is a Shore Erosion Technical Report with

technical appendices. Erosion Hazard Area Maps at a scale of 1 inch equals

2,000 feet-delineate areas with erosion potential. These maps also show short-

term recession rates (1966-1975) and long-term recessi on rates at selected

intervals. For a further description of this information see Appendix A.

Estimating A Stable Slope Angle

Assuming for a moment that no further wave-induced erosion takes place,

it is possible to estimate the stable angle of repose of a bluff. The

ultimate angle of repose of a stable slope reflects the angle of internal

friction of the materials comprising the bluff. The angle of internal fric-

tion of various materials has been documented by engineering analysis. Even

though actual blu@f failure at a particular site depends upon local varia-

tions in the soil profile, groundwater conditions, vegetative cover, surface
drainage and other factors, the stable angle of repose of.various classes
of materials can provide a reasonable rule of thumb to estimate slope sta-
bility. Thus knowing the height of a bluff, its slope angle, and the pre-

dominant material of which it is comprised takes into account some key site-

specific factors.

It is possible to generalize further and establish an average stable
slope angle for a range of erodible materials. A stable slope angle of
21.8 degrees (2-1/2 feet horizontal distance to I foot of vertical distance)

appears to be a reasonable general figure based upon studies of relative

slope stability of bluffs along Lake Michigan which took into account strati-

graphy, parent materials, bluff height and slope angle (Shore Erosion Study
Technical Report, February, 1977). This report shows an average drained

angle of internal friction based upon.laboratory tests, of 31.40 - 29.80

with one unit showing a 22.3* angle. However, the report also states that

below the groundwater table, water pressure reduces the stable slope angle

to about half the drained internal angle of friction. It further states

that localized conditions such as artesian pressures and excess hydrostatic

pressures due to seepage effects tend to reduce this stable slope angle even

more. More importantly, the study did not take into account frost actions,

surface wash, mudflow, and similar forms of mass wasting. A recent study
suggests that these processes may be'responsible for up to 50 percent of

bluff retreat in some cases. Investigators on the Shore Erosion Study have

indicated that on Lake Michigan slopes of approximately 21.8 degrees
(2 1/2: 1), natural vegetation occurs and that vegetation can effectively

control many mass wasting proce sses. The predominantly clayey soils on

Lake Superior tend to be less stable. A generalized stable slope angle of

three feet horizontal distance to one foot vertical distance (18.40) has

been suggested for regulatory purposes in Douglas County as a result of

studies done by the Red Clay Project.

Estimating Recession Rates

Wave-induced erosion can be expressed in terms of an average annual

recession rate. The Wisconsin Coastal Management program has measurements

of recession rates in the form of a reconnaissance survey. Two types of
recession rate data are available: (1) Long-term (approximately 100 years)

which integrates periods of high and low erosion and thus reflects fluctua-

tion of lake levels; (2) Short-term (1966-75) recession rates. The measure-

ment points along the shoreline are generally closer together for short-

term rates. Short-term rates are usually considerably higher than long-term

rates because the short-term rates were measured during high lake levels

when erosion is accelerated. In some instances short-term rates are lower.

This may reflect the episodic nature of slope failure (a bluff which failed

and is now temporarily stable) or the effects of structural protection.

In general it is preferable to use the long term rate as a measure of
recession. In speaking of the variation over time in average-ktreat rates,

a technical paper of the Corps of Engineers notes:

"Engineers are sometimes criticized for placing too much reliability in
average retrea-t rates derived from a limited number of measurements widely
spaced along the-shore.. However, the practicing engineer is interested in
overall conditions affecting a large section of shore, and in long-term
results affecting the lifetime of a project or structure (e.g., 30 years.),.
It is worth pointing out that as the temporal scale increases some of the
problems that originally contaminated da 'ta tend to cancel one another
rather-than accumulate as the time between observations is extended."

"A problem frequently faced by engineers is to choose a sampling inter-
val adequate to determine a mean recess in rate for a given beach....It is
well known that for a fixed level of long shore variability, the precision r,
of the estimated regional mean can be improved by increasing the-number of
survey stations.. Less well recognized is that inherent variability usua-Ily
does not increase greatly with time. Thus, the probable error or mean
rates and the gercent error in mean recession tend to decrease with time.
TWe-variance-of these estiFates would also tend to decrease (thus, the pre-
cisions increase).in direct proportion between the number of years between
surveys."

Source: Hands, Edward B. "Changes in Rates of Shore Retreat, Lake
Michigan, 1967-76. Technical Paper No. 79-4, U. S. Army
Corps of Engineers, p. 27-30

Determining theRecession Rate Setback

A recession rate setback distance can be established by multiplying the

average annual recession rate by the assigned design life of the structure to
be protected (e.g., 30 years, 50 years or 100 years for a residence). The

selection of the appropriate regulatory time span during which buildings are

to be protected from recession is a decision to be made by local policy makers

in Wisconsin. The State of Michigan requires permanent structures to be set

back the distance of the 30 year recession rate, but recommends that a greater

setback is desirable. The Province of Ontario measures the 100 year recession

rate and the stable slope angle.
"The 100-year erosion limit was established by extending i,n.3@nd from the
edge of the bluff the average annual recession rate multiplied by 100 years,
with an additional distance added on for a stable slope. To determine stable
slopes, soil characteristics, stratigraphy, bluff height and observed stable
bluff profiles were analyzed. As a result of this analysis, slopes of 2:1
and 3:1 were most frequently used." (A Guide For The Use Of Canada/Ontario
Great Lakes Flood and Erosion Prone Area Mapping, Ministry of Natural Resources,
Ontario, March, 1978, p 16)

A 50 year rate appears to be a reasonable minimum figure, since it
approximates the useful life of a typical residence. To illustrate, assum-
ing a 50 year design life and a long term recession rate of 2 feet per year;
regulated structures would have to be set back 100 feet from the ordinary
high watermark. The recession rates shown in the Technical Report Appendices

and Erosion Hazard Maps should be considered as a general guide for deter-

mining the recession rate in a given area. In areas with highly variable

recession rates or where structures have accelerated erosion, it may be

necessary to make additional studies or to determine the recession rate at

the particular site when development is proposed.

Determining the Stable Slope Setback

Structures, such as residences, that would be damaged by slope failure

can be,protected by requiring them to be located outside of unstable slope

areas. This determination can be made by applying general rules to a specific

site. Here is an example of the way it would work. Assume a bluff is 50 feet
high. An angle of 21.81 (2 1/2 feet horizontal distance to I foot vertical
distance) is measured from the ordinary high watermark. The point at which

this angle intersects the bluff is the edge of the stable slope. This means

that the stable slope setback would be 2.5 feet (stable slope angle) x 50
feet (bluff height) or 125 feet from the ordinary high watermark.

Establishing The Erosion Hazard Setback

These computations of recession rate and stable slope angle can be used

to establish an erosion hazard setback in a zoning ordinance. Within this

setback line high value structures which would be severely damaged by erosion

or activities which would accelerate erosion can be regulated. Using our

previous examples, in an ordinance that required a 50 year period for pro-

tection against recession the erosion hazard setback would be 100 feet from

the ordinary high watermark for a beach area with a 2 foot per year reces-

sion rate. Assume there is another area with the same recession rate but

which also has a 50 foot high bluff. Here the erosion hazard setback would
be the stable slope setback (50 ft. x 2.5 ft.) = 125 feet plus the recession

rate setback of 100 feet or a 225 foot erosion hazard setback line.

The erosion hazard setback can be modified if the landowner provides

technical data proving that a different recession rate is warranted, slope

conditions are more stable than assumed, or that the-erosion hazard, although

correctly estimated, can be mitigated by structural protection.

REDUCING THE EROSION HAZARD

The basic causes of shoreline erosion, i.e., wave erosion, unstable

slopes and surface erosion, can be reduced in some instances by protecting

the shoreline from waves, stabilizing slopes, and controlling surface erosion.

Protecting The Shoreline From Waves

There are two primary methods for structural protection against waves.

The first method directly armors the shoreline against wave attack through

the use of revetments, bulkheads or si.milar structures. Revetments are sloping

rock or concrete structures placed parallel to the shoreline to protect against

wave action. Bulkheads are vertical walls with their base well below the lake

level, whose primary purpose is to prevent the sliding of earth or slope failure
with a secondary purpose of protecting against wave action. The second method

of protecting against waves is to build a.protective beach by promoting
beach accretion and retarding beach erosion. This method involves the use of

nearshore breakwaters or groins.

For structures placed above the ordinary high water mark the regulatory
jurisdiction lies primarily with local government. In the case of structures

extending below the ordinary high water mark, local government can exercise

concurrent jurisdiction with the Wisconsin Department of Natural Resources

and the U. S. Army Corps of Engineers.

Further information about these methods of structural shore position is

contained in the publications Great Lakes Shore Erosion Protection: A

General Review with Case Studies;.Great Lakes Shore Protection: Structural

Design Examples, both available from the Wisconsin Coastal Management Program;
and Help Yourself -- a pamphlet discussing erosion problems and alternative

methods of shore protection,and the three volume Shore Protection Manual, an

engineering handbook -- both available from the North Central Division of

the U. S. Army Corps of Engineers. A few general comments about shore protec-

tion measures are warranted, however. Improperly designed, installed or main-

tained protective works are a waste of money and may have adverse off-site

effects. Most classes of protective works are expensive. The effective life

of a structure is generally reflected in its construction cost. "Careful

site analysis and design must precede the placement of all structural devicesq

and even then the 'success' is measured in terms of a few decades. Without

proper engineering and maintenance, structural failure can be expected at an

even earlier point. Virtually all emergency structures and many low cost

50 VeaP Recession + Stable Slope Proposed Erosion Hazard
Rate Allowance Setback

Stable Slope IN.

Ordinary High Watermark
21.8
----------- &----

structures (those under $100 per lineal foot) do not last beyond ten years."
[Wisconsin Shore Erosion Plan p. 25] The chances for an effective structural
approach are enhanced.if a group of. lot owners join together to build pro-

tective devices that are compatible and complement one another. For

example, revetments and bulkheads constructed along a reach of shoreline

which is exposed to wave attack may be subject to erosion (flanking) at

either end of the structure. It may be difficult to secure both ends of the

structure against flanking where the property involved constitutes only part

of the reach subject to wave approach from a given direction.

Bluff Stabilization and Surface Erosion Control

It is virtually impossible to stabilize a bluff unless the base of the

bluff has been protected against wave attack. However, once this has been

done several methods of bluff stabilization are available. These methods

include reshaping the bluff to a stable angle, mechanically terracing the

bluff face through retaining walls, increasing the strength of the soi-l by

removing excess groundwater, and controlling runoff over the top of the bluff.

They are usually employed in combination. The most common method is regrading

the bluff to a stable slope and constructing a rip-rap revetment at the toe,

but different procedures may be required depending upon the particular situa-

tion. The publication Harmony With The Lake: Guide To Bluff Stabilization

suggests the following stabilization measures be taken as necessary in the

following order of priority: (1) When necessary, and if possible, reshape
the bluff face to a stable angl e of slope; (2)'Control any excessive surface

water runoff; (3) Control any excessive groundwater seepage; (4) Revegetate

the bluff face as necessary.

Vegetation is important in'surface erosion control because it protects

the soil from the impact of rain, slows runoff, acts as a filter to catch

sediment, and helps hold soil particles in place. Grasses and low-growing

shrubs are preferred when establishing a vegetative cover. They provide

protection soon after growth begins, form a denser root mat, and do not tend

to loosen soil around the roots as would occur with tree roots during wind

storms, The presence of vegetation, especially trees and shrubs, may be a
general indicator that the bluff is stable at the present time, i.e., the

toe is not undercut and the slope is at a stable angle of repose. It does

not mean that continued erosion will not occur in the future.

INCREASING THE EROSION HAZARD

As previously indicated, structural attempts to control shore erosion may

increase erosion of nearby properties. Improperly managed storm water may

also increase erosion. During periods of heavy rainfall, surface water flows

over the top of the bluff and can erode the entire face of the bluff. Devel-

opment which increases runoff by creating impervious surfaces, concentrating

runoff, or destroying vegetation, accelerates this erosion. A bluff may also

fail because of the added weight of buildings, swimming pools, and other heavy

structures placed too close to the bluff top. Septic tank sewerage systems
add weight, and the liquid effluent can reduce friction between soil particles
causing unstable bluff material to slump and slide toward the beach.

ADJUSTING LAND USE TO THE EROSION HAZARD

Types of Land Uses and Development Patterns

The damages that will result from shoreline, erosion depend upon both the

severity of the erosion hazard and the type of land use that will be affected.

As the shoreline continues to erode, the land will eventually be lost but the

major portion of the damage comes from destruction of structures on the land.

Open space land uses such as agriculture, forestry and parks may be the most

appropriate land use in many erosion hazard areas, other things being equal.

However, some facilities such as marinas, water intakes, sewage treatment

plants, ports, and certain industries may require a location in the immediate

shoreline area. For these shoreline dependent uses careful siting to avoid

high hazard erosion areas and well designed erosion mitigation measures are

important to avo id unnecessary damage. In the main, these uses are ones for

which it may be economically feasible to provide effective structural protec-

tion. An investigator of shoreline erosion in Southeastern Wisconsin,

commenting on structural erosion protective measures, notes "For the most

part, the successful structures observed were built either by units of

government or, to a lesser extent, by industry. These structures are massive,

well engineered and constructed, and probably much too expensive to be justi-

fied for even the most valuable residential properties." (Shoreline Erosion

In Southeastern Wisconsin, David W. Hadley, Wisconsin Geological and Natural

History Survey, 1976, Special Report Number 5, p. 27.)

It is up to the comprehensive land use planning and zoning process to

allocate lands to their most appropriate use. In this discussion we assume

that the proposed.use is appropriate.. The focus here is on ensuring that the

use is developed in a manner consistent with theerosion hazard through the
use of land jise controls. Special erosion hazard provisions for a zoning.,

ordi,.na,nce:and.-a subdivisi-on regulation ordinance are presented in Part II

to illustrate-how%.thts may be@accomplished.

Since-regulations-generally apply only to new development.- the effective-

ness of regulations depends upon the existing land use development and owner-

ship patterns. These patterns vary widely but may be characterized.by the

followfng general categories: (1) Rural areas where the land consists of

large tracts of open space use in single ownership, e.g., farms and forests;
(2) Rural areas where the land has been divided into smaller tracts through

subdivision plats or-sale of individual lots but is not yet developed or only
partially developed; (3) Suburban areas where.the land has been substantially

developed along the immedi-ate shoreline and development consists of infilling,
i.e., construction on-the undeveloped shoreline lots; (4) Developed areas

where the first@tier of lots has been largely built upon and development is

occurring within the second tier of lots within an area still subject to ero-

sion; and (5) Urban areas where almost the entire shoreline is developed in

depth. (In general, regulations have their best potential in relatively
undeveloped areas.)
Develoeed Areas
Lots already occupied.by buildings are largely beyond the scope of regu-
lations- The-only appropriate regulatory provisions-are those designed to

control activities which would accelerate erosion or which control the expan-
sion of structures subject to damage. The owner of an existing structure
subject to substantial erosion damage has two basic options: (1) Attempt to

mitigate the erosion hazard by protecting against wave erosion and stabiliz-
ing the slope or (2) relocate the structure. Permanent relocation outside

the erosion hazard area could mean moving the structure to the rear of the

same parcel if the lot is of sufficient depth, or moving it to a different

lot not subject to erosion. "Relocation is an alternative that cannot be

overemphasized. Erosion is a natural geologic process that is extremely

difficult to stop. The alternatives to build shore protection or to relocate

must be weighed against the consequence of failure. Depending upon the type

of structure you might consider, it may cost the same to relocate as it would

to build shore protection. Should a protective structure fail, then your

investment in the structure is lost and your home or cottage still in danger."
[Help Yourself, U. S. Amy Corps of Engineers, North Central Division. p. 14]

A number of factors affect the cost of relocation. "They include lot

depth, the availability of new building sites, ease of site access, building

configuration and size, amont of subfloor access, number of'public utility

disconnections, and the availability of experienced movers. Because reloca-

tion is typically only considered during emergency periods, the amount of

land lakeward of a building is a critical factor. Between 15 to 20 feet of

clearance is normally required.for safe operation of equipment. Moving costs

of a small cabin or cottage, medium size ranch style house, and large mansion

can be expected to range between $3,000-$4,000, $7,000-$9,000, and $30,000-
$40,000, respectively. These costs do not include site preparation costs
at the new location." (Note: 1979 cost estimates) [Wisconsin's Shore
Erosion Plan. p.87]

In cases of individual hardship where lots are too shallow to permit

construction meeting the erosion hazard setback, it may sometimes be reason-

able to permit a moveable structure such as a mobile home or residence

designed so that it can be readily relocated. Allowing such structures with-

in the setback line should be done only on a case by case basis after a care-

ful investigation of the particular situation. Appropriate conditions should
be attached to development permission in-these instances.

Undeveloped Areas

On lots of adequate depth the most satisfactory approach is to properly

locate the structure in the first place. This means that structures should

be safely set back from erosion hazard areas. The setback should be based

upon a consideration of the recession rate, in all cases, and a stable slope

angle in the case of erodible bluffs. The setback approach is preferable

largely because of the limitations of structural attempts from both a private
and public point of view. Among these limitations are: (1) Attempts to

adequately protect against waves may not be feasible from an engineering point

of view, e.g., effective protection may require stabilization of a coastal
reach which is longer than the site in question; (2) Structural measures are

usually too costly in relation to the value of the land proposed to be pro-

100

tected. 1977 figures for approximate cost ranges perlineal foot of pro-
tected shoreline are as follows: temporary devices (less than 5 years
expected life) $504100, intermediate life devices (5-25 years)..$100-$200,
and "Permanent" devices (25-50 years).above $200. The actual life of a

structure depends upon proper design, construction and maintenance [Great

Lakes Shore Erosion Protection: Structural Design Examples, p. 5];
(3) Structural measures may have adverse off-site effects. Groins may cause

accelerated erosion by starving down drift beaches. Shore armorment may
deflect waves which erode adjoining property; (4) The form of shore protec-
tion most commonly used by individual property owners is loose dumping of

stone or concrete rubble. This practice affords only short term protection.

Besides destroying the natural beauty of the shoreline, this material often

ends up on the bed of th e lake, impairing the public rights in navigable

waters.

REGULATIONS TO ADJUST LAND USE TO EROSION HAZARD

Zoning and Subdivision Regulations

Zoning ordinances and subdivision regulations are important*tools that

local government can use to require that new land uses take erosion hazard

into account. Subdivision regulations and zoning complement each other.

Zoning focuses primarily on the uses of land, the dimension of lots, and

the location of structures on the lot. Lot dimensions are important to

ensure the lot is deep enough to permit structures to be safely located

behind the required erosion hazard setback line. Zoning carAalso control

grading, filling, vegetative removal, installation of protective devices

101

and other activities that may accelerate erosion. Thus activities can be

made conditional uses to require that they be undertaken in a manner that

avoids adverse effects.

Subdivision regulations focus on the process of dividing larger tracts

of land into lots for purposes of sale or building. For undeveloped areas

which have not been divided into lots, subdivision regulations have particu-

lar promise. The larger size of the parcel involved makes it more likely

that economically feasible engineering solutions can be found to storm water

management, grading and filling and erosion protection measures. Subdividers

can be required to designate erosion hazard areas on the plat, and restrict

this area to park or open space for the use of the residents of the subdivi-

sion.

Status of Zoning and Subdivision Regulations Along the Coast

All Wisconsin coastal counties have adopted shoreland regulations which

include zoning ordinances and subdiVision regulations which apply to the

unincorporated portions of the Great Lake shorelands. [Milwaukee County does

not contain any unincorporated areas.] County shoreland regulations were

designed primarily for inland lakes and most do not take into account the
special erosion hazards of the Great Lakes. [Exceptions are Douglas, Ozaukee

and Racine Counties.] All of Wisconsin's 33 coastal cities and all but two

of its villages have zoning ordinances. Twenty-five of the coastal munici-
palities have also adopted subdivision regulations. Most municipAl regula-
tions do not contain special provisions for coastal erosion hazard areas.

102

Sample erosion hazard provisions for zoning ordinances and subdivision
regulations are contained in Part II. The general approach suggested in
these provisions is to: identify erosion hazard areas; restrict or pro-
hibit uses which are vulnerable to erosion damage or which may impair public

rights in navigable waters; require special review of erosion protection

devices to ensure that they are properly designed, installed and maintained;

and regulate land disturbance, storm drainage and other activities which may

increase erosion. Erosion hazard regulations which.restrict the use of

private property must meet certain basic constitutional tests or they may be

found invalid by a court. These legal tests provide guidelines for drafting

and administering erosion hazard regulations.

CONSTITUTIONAL CONSIDERATIONS IN SELECTING REGULATORY POLICIES

General Tests of Validity

Regulations which restrict the right to use private property will

usually be found constitutionally valid if they meet the following basic

conditions:

(1) The regulations serve valid public objectives which promote

public health, safety and general welfare.
(2) The regulatory provisions are a reasonable'means to achieve

these objectives.

(3) There is a reasonable basis for the classification of uses and

lands subject to the regulations.

103

(4) The property owner is left with some reasonable use (usually
framed in economic terms) of the property, i.e., there is no taking of

property. without compensation.

(5) The ordinance provides sufficient standards to prevent the arbi-

trary exercise of power by administering.agencies in reviewing conditional

uses and in other discretionary activities.

Zoning regulations are presumed to be valid but this presumption may be

rebutted by evidence provided by persons contesting the validity of the

regulations. The importance of the particular facts in each case has been
emphasized.. "However, each case in which the validity of such regulations

is challenged, must be determined on the facts that are directly applicable

to the property of the parties complaining." Kmiec v. Town of Spider Lake

(Wis 1974) 211 NW2d 471, 477. As a consequence, in most instances the court

determines the validity of the'regulations only as they apply to the particu-

lar property in question. Landowners typically challenge the validity of

zoning where their land is zoned differently from nearby lands and the zoning

prohibits a use which would permit a higher economic return than the use to

which it is restricted. In Kmiec the Wisconsin court invalidated the applica-

tion of agricultural zoning to the particular lakeshore property in question
on two grounds: -(I) The parcel was improperly mapped, i.e., the classifica-

tion was without a reasonable basis; and (2) the classification resulted in

a substantial negative value of the land.
Erosion hazard regulations-which severely restrict the use of private
property on the basis of generally delineated erosion hazard areas may be

104

particularly subject to challenge on the grounds of "taking,, i.e., the

restrictions permit no reasonable use of the property; and improper classi-
fication, i.e., the.area is not subject to severe erosion or the estimated
erosion hazard@has been incorrectly kmapped".

The possibility of challenge to erosion hazard regulations makes it

important to carefully review legal precedents which validate reasonable

erosion hazard regulations.

105

I
I
I
-I
I
I
I
I SUMMARIES OF WORKSHOP DISCUSSION GROUPS
I
I
I
I ..
1,
I
I
I
I
I
I

PROBLEM IDENTIFICATION & ASSESSMENT

1. The group concurred with the state-of-the art paper on the causes and
mechanics of bluff slumping. They made the following specific findings
and recommendations in this regard, viz.:

a) The information in the state-of-the art papers should be converted
to user-oriented guidelines and released in the form of pamphlets
or leaflets.

b) The public should be made aware of the futility of low-cost solutions
without adequate problem identification and assessment.

c) Coastal slope failures frequently involve both terrestrial and marine
processes of degradation often in complex interaction with one
another.

d) As a result of (c) above successful systems to protect coastal
bluffs must combine a shoreline defense against wave action along
with protection against slope failures in the backshore slope area.

2. Some important distinctions were noted between two major approaches
to identification and assessment of slope stability problems in the
coastal zone, viz.:

a) Reach specific vs. site specific studies

b) Long term vs. short term analysis

c) Management (zoning) vs. engineering (structural) solutions.

3. Different constituencies are interested in different approaches listed
in #2 above. Government agencies and planning bodies tend to be
interested in the reach (long-term management) approach while property
owners prefer site specific (short-term structural) solutions.

4. The group perceived a need to develop improved coastline hazard or
zonation maps for slope degradation processes. This zonation mapping
involves the following key steps:

a) Collection of data and preparation of basic data maps

b) Development and implementation of decision models.

c) Preparation of zonation or use suitability maps.

These steps are schematically illustrated in Figure 1. Item (a) ab 'ove could be done
by technicians and trained volunteers, (b) by scientists and engineers, and (c) by
multidisciplinary teams from the public sector (e.g., state resource agencies, U.S. Corps
of Engineers, U.S. Geological Survey).

109

5. These are technical, economic, and political problems with a major
project of coastline zonation or hazard mapping. These include:

a) Public acceptance. Can the negative connotations of "hazard" or
"risk" maps be avoided by making slope evolution maps or "remedial
action" maps which focus on positive action to reduce damages
along particular reaches of shoreline?

b) Reliability of decision models. Long term data to calibrate various
models may be lacking.

c) Cost. The public must be convinced of the benefit of such a
mapping program. The loss mitigation made possible by such
analysis (either by avoidance or by pinpointing effective correction
action) must be demonstrated. Put another way, the cost of such
a program must be shown to be a small fraction of the cost of
failing to implement it.

Basic Data Maps

Lithology Topography H drology Vegetation Climate
composition height GWT type ppt.
structure slope angle pare press. % cover wind

DECISION MODELS:
mass-soil movement (slumping) Share Profiling
surficial erosion (face degradation) with time
toe erosion (wave energy delivery)
sediment transport

USE SUITABILITY OR
HAZARD (zonation)
MAPS

FIG. 1; Schematic illustration of steps for preparation of zonation or hazard maps in
the coastal zone.

6. The group concurred that many bluff stability problems particularly those
involving slumping in high bluffs in cohesive sediments can only be understood
and described properly in long-term time frames. This requires that funding
for such studies be long term and sustained.

110

CONTROLLING FACTORS

1. Bluff Lithology Many (or most) bluffs in the Great Lakes are comprised
of two or more stratigraphic units which often include cohesive and
noncohesive units in a single profile. This carries significant
implications for bluff modelling research.

2. Geomorphic Monitoring There is a need to differentiate between bluff
erosion and retreat and between retreat based on crest recression and
toe recession. This also relates to low angle and high angle failure
and the net loss of bluff material in cohesive and noncohesive lithologies;
further, the optimum frequencies of aerial overflights may be different
for different bluff types.

3. Beach Sediment Budget Most littoral sediment in the Great Lakes
originates from bluff erosion and the body of beach sediment (in transit
material) is an important control on wave erosion of bluffs; bluff retreat
in many cases can be related to beach sediment budgets over different
time f rames.

4. Bluff Recession Models: Slope profiles of bluff retreat appear to vary
with different lithologies; sandy bluffs appear to show high seasonal
variation in morphology whereas clayey bluffs appear to show greatest
morphological variabilities over longer time frames. In general, slopes
are gentler during low lake level periods and steeper during high lake
level periods. Shallow failures are more common during high lake
levels. Deeper seated failures are more common during low lake level
periods.

5. Offshore Erosion: Bluff retreat over the long term may not only be
related to recent forces and processes but to offshore (remnant) features
such as residual boulder fields; such fields have been tied to "hard
points" along retreating coast lines.

6. Climatic Effects: The influence of snowbelts on coastal hydrology and
bluff retreat is largely unknown. On cohesive bluffs, there seems to
be little activity in winter. In spring, the primary loss is by mud flows
and other bluff face activities. In fall, most loss is due to sheet loss
and rill erosion. Toe erosion is greatest in spring and in fall during
periods of high wave energy. Wind erosion is the most common erosion
process in winter.

7. Wave Forces: Wave growth on the Great Lakes appear to be much
larger than ocean-based forecasting models would indicate. This is
apparently related to the high intensity of atmospheric instability over
the lakes in the fall and winter. Sediment movement initiated by
storm wave events appears to last much longer than the wave. event
itself.

MITIGATION MEASURES

1. Non-structural mitigation measures

a) Non-structural methods only work in undeveloped areas where you
can control land use (excluding reloc ation of houses).

b) Michigan, Wisconsin, and Illinois approach is to encourage setback
requirements. ' (Wisconsin's program is voluntary allowing for use
of recession rates, slope stability, and/or combination of these).

c) Various interest groups have different views regarding disclosure
of erosion/bluff slumping hazards. For example, interest has been
shown by property assessors, banks, and realtors representing
prospective buyers, vs. realtors representing sellers). Property
owners, realtors, and banks should be made aware of erosion bluff
hazards.

d) Property owners should be made aware of federal insurance
assistance policy. Increased public awareness is needed that under
existing federal insurance laws losses from erosion are not covered.
Any changes in insurance laws should discourage property owners
from building in high risk zones.

e) There are different approaches for calculating bluff hazard
setbacks. Some states use historic recession rates, others include
slope stability, and site-by-site review. The traditional method
uses recession rates, but more emphasis should be placed on the
stable slope angle. Manitowoc and Ozaukee, counties in Wisconsin
have recently incorporated both.

f) Most programs allow for reduced setbacks with either toe protection
or movable structures (homes).

g) While there is disagreement, the scientific basis for establishing
erosion hazards is generally agreed upon.

2. Bottom and beach formation

a) Shore protection methods that are of moderate cost, replaceable,
small-scale, and work with natural forces should be tested further.
The public should be made aware of these types of devices.

b) Artificial nourishment of groin system provides immediate benefit
for shore protection.

112

3. Toe Protection

a) Armoring the shoreline is not always the answer because these are
site specific problems.

b) There is no such thing as low-cost, as in cheap, bluff protection.
The public should be made aware of this.

c) Property owners should be made aware that any shore protection
requires periodic inspection and maintenance.

d) Within the framework of general site criteria a list of certain
devices can be suggested. However, property owners should be
made aware of the site-specific nature of their effectiveness.

e) Need better scientific documentation of many devices now
marketed and developed. Many have short history, weathered few
storms (need monitoring and bathymetry).

f) Cost of structure must be weighed against durability.

4. Drainage

a) Subsurface Drainage

-Dewatering of bluffs with a deep dewatering problem is only
viable on highly valuable land due to the high cost and limited
zone of effectiveness of each well.
-Problem with dewatering bluffs is inconsistency of bluff materials
between bluffs and within bluffs.
-Drainage techniques include chimney drains, curtain drains, and
eductor wells.
-Chimney drains must be done in conjunction with toe protection
and may also require slope grading to the stable angle of repose.
-The difficulty in dewatering bluffs is due mainly to depth of
water, source of water, (e.g. aquifers, septic systems, etc.).
-Almost impossible to drain pure clay slope.
drains only handle shallow dewatering problems. Deep
draining is done with eductor well.

b) Surface Drainage

-Problems associated with surface drainage include freezing and
bursting of pipes that transport water.
-Curtain drains are also a device for surface drainage.
-Surface drainage is the easiest problem to control.
-Vegetation is very effective on surface runoff and small seeps.

113

5. Modification of Slope Profile

a) Group did not discuss methods of grading slope, terracing, or fills.

b) Most property owners do not want their property cut to a 2:1
angle and it can also be very expensive.

6. Revegetation

a) It was felt that vegetation was only effective in controlling surface
drainage and small seeps.

b) Revegetation should be conducted after other mitigation has
stabilized slope.

7. Combined Approaches

a) Public awareness of effects of on-site septic systems on both
subsurface and surface drainage is needed.

b) Need to look at all problems - controlling wave action and
groundwater before initiating mitigation methods.

c) Community action to reduce bluff slumping is desirable. Property
owners should be encouraged to work with their neighbors on shore
protection, thus benefiting from economies of scale and increased
effectiveness and durability.

114

I
I
I
I
I
I
I RESEARCH NEEDS
i
I
I
I
I
a
I
I
I
I
I
I

A. General

Each discussion group formulated a list of research needs. These needs
were discussed in a plenary session. There was considerable overlap between
the lists of each group; accordingly they have been consolidated and are
presented in a single section.

No attempt is made herein to assign priorities to these research needs. It
was generally recognized, however, that research funding in the foreseeable
future would be quite limited. Maximum emphasis should be placed,
therefore, on data collection and monitoring using local resources and trained
volunteers. The Coastwatch Program in Racine, Wisconsin was cited as an
example of this approach.

Problems amenable to individual solution have already been handled.
Challenges remaining in research are multidisciplinary requiring a team
approach.

B. Research Items

1. Slope evolution models. Slope evolution models for coastal bluffs should
be developed and refined. One or two general models of bluff erosion
.and retreat are not meaningful for the Great Lakes as a whole. Instead,
models should be developed based on specific analysis of geomorphic
processes and temporal trends on a reach basis.

2. Causal factors. Improved understanding of "lake" variables (wave
events, lake or water level changes, longshore transport, sediment
transport, sediment sources, foot ice, off-shore topography, and off-
shore topography) individually and in various combinations is necessary
to better understand the primary causes of bluff failure and erosion.
This information would be helpful in developing slope evolution models
and in forecasting bluff retreat at different sites.

3. Bathymetry Studies. We currently lack extensive or detailed data an
the nearshore zone. Most of the data gathering and monitoring has
focused on the beach, backshore slopes, and offshore deep zones.
Bathymetry studies could provide much of the needed information cited
under "lake" variables in item #2.

4. Relative importance and significance of coastal slope processes. Major
types of slope processes include deep slumping, shallow translational
slides or flows, and surficial erosion (rilling/sheet wash). We need to
get a better idea of the relative importance of these different processes
in terms of:

a) volumes of degraded materials

b) rates of bluff recession attributable to each

C damages to property and threat to lives.

117

5. Episodic nature of coastal slope erosion and retreat. The magnitude
and frequency of episodes of bluff erosion and retreat are incompletely
understood. Moreover, the magnitude and frequency of these events
may vary with different bluff types, based on their lithology,
morphology, ground water, location@ etc.... The issue is further
complicated by:

a) combinations of causal factors such as high intensity and frequent
storms coupled with weak foot ice along the shore.

b) delayed response of bluff failure (particularly high clay bluffs) to
certain driving forces such as storms and high lake levels.

6. Nature of bluff slumping in the coastal zone. The following aspects
of bluff slumping in high, clay till bluffs are poorly or incompletely
understood:

a) Where and when do large slumps occur? These are usually one of
two types; both are likely to be destructive. The first type
degenerates into a fast moving flow or slide which generates
offshore debris and damages offshore protective structures. The
second is a slow moving, deeper seated type which destroys property
atop the bluff to a considerable distance behind the bluff face. OF
b) What is the role and significance of progessive failure and strain
softening that initiates at the toe of bluffs composed of lacustrine
clays and water laid tills?

c) Can potential failure surfaces (e.g., weathered zones, highly sheared
bands) be identified and located in clay till bluffs?

d) How extensive and significant are fractured tills in coastal bluffs?

7. Mitigation. Approximately 70% of the eroding shoreline along the
Great Lakes is in private ownership. Accordingly, materials should be
developed to help property owners analyze their bluff stability problem
and provide information on integrated solutions to their problems (i.e.,
solutions that encompass both terrestrial and marine processes). in
particular, guidelines should be developed which outline:

a) Measures which property owners can implement themselves (e.g.,
traffic, vegetation, and water management) vs. measures which
require advice and services of trained professionals (e.g., subsurface
drainage installations and protective structures.)

b) Combined approaches to bluff protection including simultaneous
use of vegetation, structures, grading, beach fills, and drainage
systems.

c) Setback requirements along specific reaches of shoreline.

Continued monitoring and evaluation are also required on the long term
effectiveness and performance of low-cost shore protection systems which
have been installed at various sites as demonstration projects.
11B

I -
I
I
I
I
I
I PUBLIC DISSEMINATION
I .
I
I
I
I
I
I-
I
I
I
I
I

A. General - The workshop participants were asked to identify priority areas
for information dissemination to the public on bluff slumping problems and
to suggest possible vehicles for such dissemination. The resulting list of
items was culled from both small group discussion and a general discussion
in plenary session.

A number of difficulties were noted at the outset of such an information-
transf er program. These included but are not limited to the following
observations:

1. -Many people are not particularly concerned about the problem until it
affects them directly and acutely. By then it may be too late to do
anything, or available options may be severely limited.

2. The consciousness threshhold tends to fluctuate with lake levels. Periods
of low lake level may be the best time to initiate remedial action,
but people are less worried then and less inclined to take action. This
underscores a need for continuous education and publicity about the
problem

3. The information needs of different constituencies vary in their technical
content and format. The level and type of information desired by
property owners most likely will differ from that required by a
consultant engineer, contractor, or public agency official. This suggests
a need for different levels of publications which address different needs
and concerns.

4. Making these various constituencies aware of sources and availability
of information can be a problem. In many cases shore property owners
live elsewhere (e.g., Chicago and Detroit) and are only part-time
residents.

B. Format for Dissemination - A number of different methods for dissemination
were discussed including pamphlets, fold out brochures, slide shows, fact
sheets, handbooks, reports, annotated bibliographies, and short courses.
There was some concensus that the best format for a public information
document was a brochure or small handbook.

A revised and updated version of the Illinois fold out brochure (Illinois
CZM, 1979) was cited as a possibility. The general feeling was that a
revised brochure or handbook should present an overview of the causes of
bluff slumping; outline procedures for correct identification and assessment
of the problem at a specific site, provide general guidelines for solving the
problem, and cite sources of additional information and/or assistance.

121

C. Content of Public Information Document. - In addition to the above suggestions
for organization and focus of the publication, a number of specific
recommendations about content were also made. These include the following:

1. A section should be included with a flow chart or checklist that takes
property owners through a step-by-step analysis of their particular bluff
problem. One suggestion was made that 'such a checklist be modelled
on the one found at the beginning of the coastal zone vegetation
guidebook. Another suggested that the home owner identify his
particular bluff type based on the Michigan Lake Shore Classification
System (MSU Resource Development, 1958) which in turn would describe
the slope process most likely to affect that particular bluff type and
the range of remedial measures that might be considered.

2. The publication should tell property owners when they need to call in
experts and who these experts are. In addition the publication should
inform the owner what measures he can implement himself at relatively
little coast. These measures tend to fall in the general category of
"management" action, viz.:

a) Traffic Management - routing and channeling- of vehicular and
pedestrian traffic from vulnerable slopes by use of sign posts,
barriers, stairs, etc.

b) Water Management - minimizing infiltration and runoff into bluff
area from lawn watering, drain fields, roof gutters, road ditches,
etc.

c) Vegetation Management - maintaining woody shrubs and trees on
critical slopes. Planting dune grass and suitable herbaceous cover
on sandy slopes.

Lakeshore Classification Map and Shoretype Bulletin, #1-28, (1958). Michigan State
University, Department of Resource Development, Agric. Exp. Stn., East Lansing, Mich.
(1958).

3. The publication should describe warning signs or tell-tale indicators of
an impending slope failure or continuing stability problem in coastal
bluffs. These could include:

a) vegetative indicators - bare slopes, incongruent vegetation,
phreatophyte plants growing on the slope, displaced vegetation.

b) hydrologic indicators - springs or seeps in the face of the bluffs,
presence of rills or gullies, perched ground water tables.

c) morphological indicators - presence of scarps, cracks, depressions;
uneven or broken slopes; alluvial fans at the base of slopes.

d) lithological indicators - weak or erodible material exposed at the
toe of bluff, lack of protective beach.

122

I
I
I
I
I
I
I
WORKSHOP ATTENDEES
I
I
I
I
I
I
I
'I
I
I
I
I

WORKSHOP ATTENDEES

Gary L. Boyd John Lyon
Dept. of Fisheries & Oceans Dept. of Civil Engineering
867 Lakeshore Dr. 470 Hitchcock Hall
P.O. Box 5050 Ohio State University
Burlington, Ontario, Canada L7R 4A6 Columbus, OH 43210

Robert M. Quigley William M. Marsh
Faculty of Engineering Science Dept. of Resource & Comm. Science
Univ. of Western Ontario University of Michigan - Flint
London, Ontario, Canada N6A 5B9 Flint, MI 48503

A. J. Zemen H. W. Belcher, P.E.
Shore Processes Sec., Hydraulics Div. U.S.D.A. Soil Can. Services
NWRI Can. Cen. for Inland Waters 1405 S. Harrison Rd.
P.O. Box 5050 East Lansing, MI 48823
Burlington, Ontario, Canada L7R 4A6 Luis Vallejo
John G. Housley Dept. of Civil Engineering
HQ USACE (DAEN-CWP-F) 147 Engineering Build.
U.S. Army Corp. of Engineers Michigan State University
Washington, D.C. 20314 East Lansing, MI 48824

Norm Wingard Thomas A. Coleman
ESAO-USGS Soils & Materials Section
720 National Center Mich. Dept. of Transportation
Reston, VA 22092 P.O. Box 3005
'arkle, WESHH Lansing, MI 48909
Dennis M,
U.S. Army Engineer Martin R. Jannereth
Waterway Experiment Station Land Resource Programs
P.O. Box 631 P.O. Box 30028
Vicksburg, MS 31980 Lansing, MI 48909

Gerald Mitchell, WESGE Susan Prinz
U.S. Army Engin. Waterway Exper. Sta. 335 Morris Ave., SE
P.O. Box 631 Grand Rapids, MI 49503
Vicksburg, MS 31980
Frank Neilson, WESHI John McKinney
Mich. Sea Grant Ext. Agent
U.S. Army Engin. Waterway Exper. Sta. Governmental Center
P.O. Box 631 400 Boardman Avenue
Vicksburg, MS 31980 Traverse City, MI 49684
Lawson Smith, WESGR Dr. Tuncer'Edil-
U.S. Army Engin. Waterway Exper. Sta. Dept. of Engineering
P.O. Box 631 1415 Johnson Drive
Vicksburg, MS 31980 Madison, WI 53706

125

Charles H. Carter Guy Meadows
Div. of Geological Survey Dept. of Atmos. & Oceanic Sciences
P.O. Box 650 2218 Space Res.
305 E. Shoreline Drive University of Michigan
Sandusky, OH 44870 Ann Arbor, MI 48109

Constantine N. Papadakis David M. Mickelson
Wolverine Tower, Suite 1014 Dept. of Geology - Weeks Hall
3001 S. State.St. University of Wisconsin
Ann Arbor, MI 48104 Madison, W1 53706

Charles L. Kureth, Jr. Douglas Yanggen
The Traverse Group, Inc. Land Use Specialist
2480 Gale Road 1815 University Avenue
Ann Arbor, MI 48105 Madison, W1 53706

Ernest Brater Edith KcKee
Dept. of. Civil Engineering 416 Maple Street
304 West Engineering Winnetka, ILL 60093
University of Michigan
Ann Arbor, MI 48109 Robert W. Fleming
Engin. Geological Br. M5903
Donald F. Eschman U.S.G.S., Denver Fed. Center
Dept. of Geological Sciences Denver, CO 80225
4010 CC Little Bldg.
University of Michigan J. Phillip Keillor
Ann Arbor, MI 48109 Sea Grant Institute
University of Wisconsin
Donald Gray 1800 University Avenue
Dept. of Civil Engineering Madison, WI 53706
312 West Engineering
University of Michigan Suzanne Tainter
Ann Arbor, MI 48109 Communications Coordinator
Sandra Gregerman, Info. Officer Michigan Sea Grant
2200 Bonisteel Blvd.
Michigan Sea Grant Ann Arbor, MI 48109
2200 Bonisteel Blvd.
Ann Arbor, MI 48109 Ellen Fisher
1815 University Avenue
John H. Judd, Assist. Director University of Wisconsin
Michigan Sea Grant Madison, WI 53706
2200 Bonisteel Blvd.
Ann Arbor, MI 48109

126

I .
I
I
I
I
I ADDITIONAL SOURCES OF INFORMATION
I
I
I I
I
I
I
I
I
a
.1,
I
I
I