Document Actions

ijege-13_bs-dykes-et-alii.pdf

background image
Italian Journal of Engineering Geology and Environment - Book Series (6) www.ijege.uniroma1.it © 2013 Sapienza Università
Editrice
133
DOI: 10.4408/IJEGE.2013-06.B-10
A GEOMORPHOLOGICAL RECONNAISSANCE OF
STRUCTURALLY-CONTROLLED LANDSLIDES IN THE DOLOMITES
A
lAn
P. DYKES
(*)
, E
dwArd
n. BROMHEAD
(*)
, S. M
Ahdi
HOSSEYNI
(**)
& M
AiA
IBSEN
(*)
(*)
School of Civil Engineering and Construction, Kingston University - London, United Kingdom
(**)
Dept. of Civil Engineering, I.A.U. Azadshahr Branch, Shahrood Road, Azadshahr, Golestan, 49617, Iran
Within about 100 km (by road) from Longarone
and Vaiont, many landslides of many different types
and characteristics can be readily identified and ac-
cessed. In general, the different types can be clearly
associated with their respective topographic and geo-
logical contexts. Those associated with ‘soft rocks’ are
found at the southern edge of the Dolomites region and
within the northern half of the region, where the respec-
tive lithologies generally underlie more gentle slopes
below prominent limestone outcrops. Small shallow
slides and extensive landslides, including the Tambre
and Tessina earthflows, occur in the Eocene flysch
in the Alpago Basin at the eastern end of the Belluno
Trough. The mid-slopes of the Alpago Basin are also
characterised by the forms of deep-seated rotational
and/or compound slides in the flysch. Further northeast,
the soft mudrocks such as those of the San Cassiano
Formation have given rise to many mass movements
including the Corvara earthflow and the shallow land-
slides in and around Cortina d’Ampezzo, as well as
shallow rotational slides and deep-seated rotational or
compound slides involving failure within the mudrocks
resulting in major displacements of the overlying lime-
stones. The upper valley slopes immediately west of the
Gardena Pass are characterised by very large tension
cracks and subsided limestone blocks that can be cat-
egorised as ‘cambering’ failures, again due to failure of
the weaker underlying San Cassiano Formation.
Landslides founded entirely in the limestones/do-
lostones have occurred as far south as the ancient Piave
INTRODUCTION
The mountainous region of northeast Italy is
known as ‘The Dolomites’, the name arising from the
dominance of dolomitic limestones among the geo-
logical formations that characterise the region. The
mountains are the product of a complex tectonic his-
tory which, in the region that includes Longarone and
Cortina d’Ampezzo, is thought to have included uplift
of perhaps 3-5 km (d
oglioni
, 1987). Present uplift
rates are high, possibly as much as 6-7 mm y
–1
(Dr F
Podda, University of Trieste, pers. comm. 2012). The
present mountain slopes, that provide the potential
for mass movements, were largely shaped by glacial
and subsequent fluvial erosion processes. The persist-
ence of the (north-facing) Marmolada glacier down to
~2850 m altitude and the terminal moraines from the
Piave Glacier south of the Venetian Pre-Alps near Vit-
torio Veneto are indicative of significant Quaternary
glaciation extending into this region. The summit of
the Marmolada mountain, at 3342 m, is the highest
point in the region. Maximum summit elevations de-
crease eastwards from here, across the region of in-
terest to this paper, being around 2500 m adjacent to
the Piave Valley and <2000 m around the Tagliamento
Valley east of Ampezzo. Local (valley-scale) relief is
typically in the range 1200-1600 m. Mass movements
are unsurprisingly common in such a tectonically ac-
tive mountainous region, their types and characteris-
tics being strongly related to the geology, particularly
with respect to structural controls.
background image
A.P. DYKES, E.N. BROMHEAD, S. MAHDI HOSSEYNI & M. IBSEN
134
International Conference on Vajont - 1963-2013 - Thoughts and analyses after 50 years since the catastrophic landslide Padua, Italy - 8-10 October 2013
of landslide is more common in hard rocks where the
structural weakness provided by the bedding planes
contrasts strongly with the intact rock mass strength.
Bedding-controlled landslides are commonly found
in argillaceous rocks, particularly stiff plastic over-
consolidated clays (e.g. B
Arton
, 1984; C
ooPEr
et alii,
1998). The basal part of the deep-seated rotational or
compound slip surface in most failures of these materi-
als is planar, being controlled by a ‘weak layer’ arising
from a (brief) interruption to an otherwise continuous
and generally uniform sedimentary accumulation, i.e.
a bedding plane (e.g. B
roMhEAd
et alii, 2002; d
ixon
&
B
roMhEAd
, 2002). The low strength of the ‘weak layer’
gives rise to the instability and then controls the depth
of the failure, but does not dominate the morphology or
internal geometry of the landslide.
CONTROLS ON DIP-SLOPE FAILURES
The occurrence and consequences (geomorpho-
logical and possibly socio-economic) of dip-slope
failures can be strongly related to the geometric
relationship(s) between the dip and strike of the bed-
ding planes and the pre-failure topographic surface.
The configuration of the bedding planes and any other
structural features results from the tectonic history of
a location, having developed over geological times-
cales. The pre-failure topographic surface, on the oth-
er hand, results from the combined effects of various
geomorphological processes acting on the slope and/
or adjoining landform units (such as the river channel
at the foot of the slope) over typically much shorter
timescales, of the order of perhaps 10
3
-10
5
years rather
than e.g. 10
6
-10
8
years.
C
rudEn
(2000) classified these geometric relation-
ships based on observations of rockslides in the Cana-
dian Rockies (Tab. 1). As is usually the case, reality
often exhibits more complex arrangements than any
simple classification scheme can represent. Hence, if
Valley at Fadalto and Nove, where very large rockfalls
or rockslides formed landslide dams that have been ex-
ploited for hydro-power electricity generation (C
oPPo
-
lA
& B
roMhEAd
, 2008). The Fadalto landslide involved
failure across or through the bedding of the limestone,
and the Nove landslide can be assumed to be simi-
lar in this respect although it is so old that the origi-
nal source area cannot be reliably identified. Further
north, a moderately large and very distinct deep-seated
rotational landslide on the western side of the Pelmo
massif (summit elevation 3168 m) has also developed
across the bedding, which further highlights the role of
joints and fractures through the rock mass associated
with the regional tectonics in promoting large-scale
instability. The other landslides that are not causally
associated with the bedding of the limestones are the
debris flows. Although small debris flows are common
throughout the Dolomites, two distinctive concentra-
tions occur along the eastern slope of the Boite Val-
ley southeast of Cortina and below the 2000-2200 m
high ridge immediately north of the same town. Most
of these are appear to originate from deeply dissected
fracture or bedding zones where structural weaknesses
have allowed water penetration and thus promoted lo-
calised preferential freeze-thaw weathering.
The primary focus of this paper is a group of land-
form-defining rockslides, some of which may be better
described as ‘block slides’, that appear to have been
primarily controlled by the bedding within, or perhaps
adjacent to, the major dolomitic lithological units. Most
of the examples involve units of the Dolomia Princi-
pale of Upper Triassic age. We make a distinction at this
point between ‘bedding-plane failures’ (also known as
‘dip-slope failures’) and ‘bedding-controlled failures’.
In a dip-slope failure, the landslide occurs due to failure
along one or more parallel planar or near-planar bed-
ding planes and the failed bedding surface dominates
the post-failure morphology of the slope. This type
Tab. 1 - Classification of slope–bedding relationships (af-
ter C
ruden
, 2000)
Tab. 2 - Characteristics of the study landslides
background image
A GEOMORPHOLOGICAL RECONNAISSANCE OF STRUCTURALLY-CONTROLLED LANDSLIDES IN THE DOLOMITES
Italian Journal of Engineering Geology and Environment - Book Series (6) www.ijege.uniroma1.it © 2013 Sapienza Università
Editrice
135
ALLEGHE LANDSLIDE
The major rockslide at Alleghe occurred in Janu-
ary 1771 and was followed later that year by anoth-
er, smaller one from the same slope (E
iSBAChEr
&
C
lAguE
, 1984; d
i
B
iASio
et alii, 2000; C
oPPolA
&
B
roMhEAd
, 2008). Figure 1 shows the planar form of
the (combined) landslide scar, the failure surface be-
ing defined by a small number of closely spaced bed-
ding planes in the limestone (Fig. 3). The displaced
mass comprised a thick wedge of rock that slipped
from a valley side slope that was previously convex
in profile and plan. The northern edge of the failure
surface marks the line where the critical bedding
plane(s) daylighted, whereas the southern margin
comprises a high lateral scar defined by high angle
joints (Fig. 2).
The original failure formed a landslide dam that
was subsequently engineered to become effectively
permanent. Before the failure, the river was 30-40 m
below the present lake level. If the landslide scar is
projected downslope at the same gradient, it appears
to coincide with the approximate position of the river
at that elevation. Since the post-failure scar is planar,
the river at the foot of a dip slope incises rapidly to cre-
ate a narrow gorge, then the lower part of the slope be-
comes an overdip slope and the likelihood of instability
is significantly increased. Furthermore, although this
framework is designed for the intepretation of slopes
underlain by planar bedding, it does provide a basis for
assessing the stability status of slopes formed on anti-
clinal or synclinal structures. For example, the Vaiont
landslide occurred on a dominantly dip (or underdip)
slope but instability was promoted by the overdip slope
at the toe, where the extremely steep sides of the river
gorge allowed the critical bedding plane(s) to daylight.
DIP-SLOPE LANDSLIDES IN THE DOLO-
MITES
We have examined the main features of four dis-
tinct examples of large dip-slope landslides and what
we think may be a very large dip-slope or bedding-
controlled landslide, in the southeastern Dolomites
region. These landslides and the coordinates of their
head scars are as follows:
Alleghe
046°24’19”N, 011°59’44”E
Borta
046°22’04”N, 012°48’17”E
Pineda
046°16’54”N, 012°21’04”E
Vaiont (East) 046°15’15”N, 012°21’04”E
Cinque Torri
046°30’15”N, 012°02’00”E
Summary characteristics of these landslides are
shown in Table 2. The lengths and heights (vertical
differences in elevation between the head and the toe)
are derived from various topographic map sources,
with the consequence that the derived slope gradients
may be ±1–2° of the stated value. This level of ap-
proximation does not affect the interpretations or the
key message of this paper.
Fig. 1 - The Alleghe landslide in 2011, seen from above
Alleghe village
Fig. 2 - The upper part of the scar showing the bedding
control and the lateral joint-controlled margin
Fig. 3 - Part of the failure surface (seen from below the
landslide) showing it to be controlled by very few
separate bedding planes
background image
A.P. DYKES, E.N. BROMHEAD, S. MAHDI HOSSEYNI & M. IBSEN
136
International Conference on Vajont - 1963-2013 - Thoughts and analyses after 50 years since the catastrophic landslide Padua, Italy - 8-10 October 2013
lope) and the formation of a landslide dam. Finally,
the geomorphological context suggests a similar
geometric predisposition to instability, although the
evidence is less clear. It seems likely that the incis-
ing tributary streams on the main valley slope cre-
ated overdip slopes along the upslope sides of the
streams. As such, they also defined the downslope
extent of the failure and thus limited the scale and
persistence (only a few weeks) of the dam.
PINEDA LANDSLIDE
The age of the Pineda rockslide is uncertain, but
its scale is readily apparent (Tab. 2, Fig. 6). The ridge
between the present Vaiont Lake and the Mesazzo
Stream has been identified as a deposit of the Pineda
slide (e.g. P
Aronuzzi
& B
ollA
, 2012), which strongly
suggests that a large landslide dam existed for some
period of time following the landslide. Unlike the
other landslides, the upper scar is less steep than the
overall slope but nevertheless it appears that failure
of bedding planes in the Soverzene Formation defined
the event (Fig. 6b), resulting in a triangular sub-planar
scar in the landscape (Fig. 6) similar to the landslides
described above.
The Pineda slide is characterised by a lower mean
gradient of the bedding plane failure surface com-
pared with the other landslides, although the form of
this upper slope unit is slightly convex (Fig. 6b) and
a dip of 30° is shown for the lower part of this unit
by P
Aronuzzi
& B
ollA
(2012). Clearly the geological
control over this slope failure is more complex than at
the landslides discussed previously, but it seems rea-
sonable to suggest that, as a dip slope, the upper slope
unit was susceptible to the development of instability,
and that this instability was ultimately brought about
it follows that the original valley slope adjacent to
the river was probably steeper than the bedding, i.e.
that instability was promoted by the lower slope be-
ing an overdip slope. However, this geomorphologi-
cal context of a somewhat overdeepened river valley
also reduced the impacts of the landslide. The dam
was formed because the narrow lower valley pre-
vented runout of the debris, and the permeable nature
of the debris enabled it to persist without catastroph-
ic failure and flood (C
oPPolA
& B
roMhEAd
, 2008).
The village was, however, hit by a devastating wave
from the second slide (E
iSBAChEr
& C
lAguE
, 1984).
BORTA LANDSLIDE
The rockslide that destroyed the village of Bor-
ta, on the Tagliamento River near Caprizi, occurred
in August 1692. It is similar to the Alleghe slide
in many respects, but although the affected slope
was significantly longer and higher than at Alleghe
(Tab. 2), the failed mass was much smaller with an
estimated volume of around 30,000 m
3
(C
oPPolA
&
B
roMhEAd
, 2008). Inspection of aerial imagery (e.g.
GoogleEarth) suggests that the failure involved the
upper part of a main valley slope that had been dis-
sected by first-order tributary streams. The extent of
the steeper upper slope scar can be clearly seen in
Fig. 4, where the bedding planes in the dolostone
that define the failure surface are clearly visible (Fig.
5). Figures 4 and 5 both show the lateral scar on the
eastern side of the failure, controlled by joints.
Several point of similarity with the Alleghe
landslide can be identified. These include the bed-
ding plane shear surface with the same gradient, the
joint-controlled lateral scar on one side, the triangu-
lar form of the main scar (width increasing downs-
Fig. 4 - The Borta landslide in 2011, seen from upstream
Fig. 5 - Part of the failure surface showing the dolostone
bedding planes
background image
A GEOMORPHOLOGICAL RECONNAISSANCE OF STRUCTURALLY-CONTROLLED LANDSLIDES IN THE DOLOMITES
Italian Journal of Engineering Geology and Environment - Book Series (6) www.ijege.uniroma1.it © 2013 Sapienza Università
Editrice
137
above. The scar is much wider than those of the other
landslides but (i) its pre-failure profile characteristics
are similar (Table 2), (ii) it formed a landslide dam (Fig.
7), (iii) it has a joint-controlled lateral scar at its (east-
ern) lateral margin, and (iv) the visible failure surface
has a roughly triangular (actually more of an arch) plan
form. The latter feature is easily missed on casual in-
spection of aerial imagery because of a large forested
unit at the extreme southeast corner of the landslide
that was displaced but remained on the failure surface.
This displaced block clearly resulted from retrogressive
development of the landslide. Such retrogression often
accounts for the upward-narrowing plan forms of land-
slide scars, as observed in all of the cases in this paper.
The view of the landslide in Fig. 7 is somewhat
misleading in that it does not show how the Vaiont
Gorge turned slightly northwards, away from Mt. Toc,
where the landslide dam now begins. In any case, cross-
sections indicate that the East side failure surface was
probably mostly planar. Furthermore, the sides of the
gorge were much steeper than the typically 30-40° dip
that defines the failure surface and thus the toeslope of
the entire landslide was an overdip slope.
CINQUE TORRI LANDSLIDE
The small rocky outcrop near Cortina d’Ampezzo
known as the Cinque Torri (‘Five Towers’) (Fig. 8)
has been demonstrated to be the product of mass
movements involving failures of the weak argilla-
ceous rocks under the load exerted by the Dolomia
Principale and suggested to be a small part of a much
larger-scale landslide system involving bedding-
controlled failure(s) (V
iEro
et alii, 2010). The argil-
laceous rocks (the Travenanzes and Heiligkreuz For-
mations) are 150-180 m thick and underlain by the
Dolomia Cassiana, with the entire sequence having
a moderate dip of 30° towards N10°E. V
iEro
et alii
by glacial oversteepening of the lower part of the val-
ley slope. Figure 6a appears to show the landslide scar
extending some distance below the upper slope unit.
This landslide may therefore have begun with failure
of the underlying ‘soft’ rocks (the marly Scaglia Rossa
(C
AnCElli
et alii, 1991) and the Marne di Erto which
is transitional from the Scaglia Rossa to the Flysch),
and retrogressed upslope.
The lithology of the upper bedding failure surface
of the Pineda landslide and its general structure, i.e.
the folded front of a laterally extensive over-thrust,
are the same as for the cliffs above Casso village that
overlook the Vaiont dam. There have been several
potentially damaging rockfalls/rockslides from these
cliffs in recent years, requiring significant expendi-
ture on mitigation and stabilisation measures. It seems
ironic that a landslide hazard assessment of the valley
before 1963 would probably highlight the risks associ-
ated with these outcrops of the Soverzene Formation,
based on the evidence of Pineda and the Casso depos-
its, but entirely fail to recognise any realistic hazard
and risk from the northern slopes of Mt Toc or indeed
from the reservoir, some 200 m lower than the town.
VAIONT LANDSLIDE
This research led us to re-examine some features of
the Vaiont landslide in the context of other major rock-
slides in the region. For this reason, attention is focused
here on the East side of the Vaiont slide, which has of-
ten been largely overlooked in the past because of the
closer proximity of the West side to the dam.
If considered as a separate landslide, Vaiont (East)
is similar in many respects to the landslides described
Fig. 6 - The Pineda landslide: (a) seen from the Vaiont
landslide in 2012; (b) geology according to B
esio
& s
emenza
(r
iva
et alii, 1990)
Fig. 7 - The Vaiont landslide in 2011, seen from the east
background image
A.P. DYKES, E.N. BROMHEAD, S. MAHDI HOSSEYNI & M. IBSEN
138
International Conference on Vajont - 1963-2013 - Thoughts and analyses after 50 years since the catastrophic landslide Padua, Italy - 8-10 October 2013
(2010) show everything to the right of our Fig. 8 as a
‘reference stable area’.
In our view, the geomorphology of the site indi-
cates a landslide system possibly extending over sev-
eral square kilometres. The influence of the 30° dip of
the bedding can be seen in many parts of the landscape
in Fig. 8. The left side of Averau appears to be the
head of a large translational dip-slope landslide (Fig.
9) that probably extends northeastwards for up to 4
km down to the Falzarego River with the upper sur-
face of the Dolomia Cassiana as the failure surface.
This is consistent with V
iEro
et alii’s (2010) sugges-
tion of such a landslide although we think it extends
another kilometre further upslope. However, it is un-
clear what structural forms or failure surface geometry
in the lower half of the slope give rise the low overall
gradient of the failed slope (Tab. 2).
DISCUSSION
We have examined five major landslides in the
southeast Dolomites region, four of which are dip-slope
failures and the fifth may also be of the same type al-
though details of the lower part of its slope are not
known.The dip angles of the bedding plane failure sur-
faces (‘Upper slope’ in Tab. 2) are very similar for all of
the landslides and the dip-slope failures are characterised
by almost identical mean pre-failure slope gradients.
Detailed investigations of the Vaiont landslide
over many years have confirmed the presence of thin
clay layers between limestone beds within the Fonzaso
Formation, which is the stratigraphic unit that forms
the visible failure surface. The Pineda landslide in-
volved a lower dip angle than the other cases, which is
lower than the normal range of friction angles between
unweathered limestone surfaces (S
ElBy
, 1993), and is
thus consistent with a failure strongly controlled by the
strength of clay layers. However, in deformed strata
like this (Fig. 6), flexural slip between beds is likely to
have reduced the available shear strength - in this case,
of the clay - to its residual value, with the cohesion re-
quired for stability probably having been provided by
rock-bridges (S
turzEnEggEr
& S
tEAd
, 2012).
At the Borta landslide, only the upper part of the
slope failed, but both here and at Alleghe the bedding
appears to be almost entirely planar. The dip angles lie
just inside the range of friction angles for unweathered
limestone surfaces. This can be interpreted as indicat-
ing that failure in these cases was controlled either by
clay layers and rock bridges or simply by the bedding
contacts without any clay but possibly weathered due
to rainwater penetration directly into the bedding at
the top of the slopes.
The Cinque Torri landslide is perhaps the easiest
to explain. In this case, the controlling shear strength
would have been that of the basal layer of the domi-
nantly argillaceous Heiligkreuz Formation. The ob-
served ~30° dip angle in the upper part of the slope is
steep enough to account for the occurrence of the land-
slide, but its apparently limited development may be
associated with the much lower overall slope gradient.
The four dip-slope landslides are typical of others
described from around the world. The largest known
terrestrial landslide on Earth, Saidmerrah in the Za-
gros Mountains of southwest Iran, is a dip-slope land-
slide involving failure of planar bedding surfaces in
Cretaceous limestone, although the dip angle is rather
steeper than in many other cases. The role of rock
bridges as contributors to the available shear strength
in folded limestone sequences was demonstrated re-
cently using the Palliser Rockslide, one of many such
dip-slope landslides in the limestones of the Canadian
Rockies (S
turzEnEggEr
& S
tEAd
, 2012). Interestingly,
Fig. 8 - The Cinque Torri and Averau, with Pelmo (3168 m)
in the distance.The white line indicates the hypoth-
esised failure surface (broken line = concealed at
depth). Snow highlights the topographic details
Fig. 9 - View towards Averau from the Cinque Torri. The
planar slope on the left is the bedding-plane fail-
ure surface highlighted in Fig. 8
background image
A GEOMORPHOLOGICAL RECONNAISSANCE OF STRUCTURALLY-CONTROLLED LANDSLIDES IN THE DOLOMITES
Italian Journal of Engineering Geology and Environment - Book Series (6) www.ijege.uniroma1.it © 2013 Sapienza Università
Editrice
139
provided the major source of debris that the flood wave
from the Vaiont landslide was able to entrain and which
consequently further increased the intensity of the de-
struction at Longarone (B
roMhEAd
et alii, 1996).
CONCLUSIONS
Two key messages arise from this brief survey of
the large structurally controlled landslides of the south-
eastern Dolomites. Firstly, failure along the bedding is
as likely in this region as it is in any other mountain-
ous region dominated by strong sedimentary rocks,
often carbonates, and subjected to significant tectonic
deformation followed by vigorous glacial and/or flu-
vial erosion. Secondly, we have shown that the eastern
half of the Vaiont landslide appears to be entirely typi-
cal of the other dip-slope landslides in the study area.
We make no assessment of the influence (or not) of
the western half on the occurrence of the failure of the
eastern half, but we do emphasise the fact that there
should be nothing inherently surprising about the east
side failure. The detailed mechanisms of the overall
landslide may still be very uncertain, but the underly-
ing general context and characteristics are clear. The
critical implication is that any assessment of the risks
of major landslides that may arise from any future in-
frastructure developments in mountain regions, in the
Dolomites or indeed elsewhere in the world, should
include the Vaiont example as a typical case and not
exclude it as being ‘unexplained’.
ACKNOWLEDGEMENTS
We are grateful to Stefano Devoto, Fulvio Podda
and Emanuele Forte for assistance with fieldwork at
Vaiont in 2012 and for further information and ongo-
ing discussions relating to this and the other landslides
in the region.
the Frank Rockslide in Alberta was for many years
thought to have been primarily bedding-controlled,
although it has now been shown that the structural
control was dominated by the joint sets through the
massive limestone (e.g. P
EdrAzzini
et alii, 2012).
Two further geomorphological issues became ap-
parent as a result of our study of the landslides in the
Dolomites. Firstly, the formation of a landslide dam
depends not so much on the absolute volume of the
failed mass and the duration of the movement (i.e. rate
of accumulation of the deposit), but more importantly
on how these two characteristics relate to the scale of
the valley and the discharge in the river at the time of
failure. Three of the landslides are of moderate scale
but all involved rapid accumulation of their debris
within narrow and steep-sided valleys. At Vaiont, the
glacial-scale valley trough was effectively dammed as
the fluvial gorge was entirely infilled.
Secondly, although formation of the landslide dams
will have supplied some sediment to the downstream
river channels, in the Dolomites it appears that the more
significant sediment impact has been for the dams to
interrupt the downstream transfer of very high (glacio-)
fluvial sediment loads. Very rapid accumulation of sedi-
ments upstream of landslide dams is best demonstrated
by reference to infilling of the lake formed by the Al-
leghe landslide, which had extended as far as Caprile (3
km upstream of the present residual lake) within weeks
of the event. Such rapid sedimentation is also readily
apparent in the residual lake upstream of the Vaiont
landslide and in the river above the Borta slide. Finally,
although not relating to a dip-slope failure, mention
must be made of the infilling of the Piave Valley fol-
lowing the landslide and dam formation at Fadalto. The
net accumulation of sediment arising from this event
can be traced as far upstream as Castelmezzano and
REFERENCES
B
Arton
M.E. (1984) - The preferred path of landslide shear surfaces in overconsolidated clays and soft rocks. Proc. 4
th
Int.
Symp. Landslides, 3: 75-79.
B
roMhEAd
E.n., C
oPPolA
l. & r
EndEll
h.M. (1996) - Valley blocking landslide dams in the eastern Italian Alps. 7
th
Int. Symp.
Landslides, 655-660. Trondheim, Balkema.
B
roMhEAd
E.n., i
BSEn
M.-l., P
APAnAStASSiou
x. & z
EMiChAEl
A.A. (2002) - Three-dimensional stability analysis of a coastal
landslide at Hanover Point, Isle of Wight. Q. J. Eng. Geol. Hydrogeol., 35: 79-88.
C
AnCElli
A., C
roStA
g. & r
oMAni
P. (1991) - Discussion. In: C
hAndlEr
R.J. (E
d
.). Slope Stability: Engineering Developments
and Applications, 82-83. ICE Publishing, London.
C
ooPEr
M.r., B
roMhEAd
E.n., P
EtlEy
d.J. & g
rAnt
d.i. (1998) - The Selbourne cutting stability experiment. Géotechnique,
48: 83-101.
background image
A.P. DYKES, E.N. BROMHEAD, S. MAHDI HOSSEYNI & M. IBSEN
140
International Conference on Vajont - 1963-2013 - Thoughts and analyses after 50 years since the catastrophic landslide Padua, Italy - 8-10 October 2013
C
oPPolA
l. & B
roMhEAd
E.n. (2008) - Fossil landslide dams and their exploitation for hydropower in the Italian Dolomites.
Boll. Soc. Geol. It., 127: 163-171.
C
rudEn
d.M. (2000) - Some forms of mountain peaks in the Canadian Rockies controlled by their rock structure. Quat. Int., 68:
59-65.
d
iBiASio
E., t
urrini
M.C. & B
uSin
w. (2000) - A historical landslide in the Province of Belluno Eastern Dolomites, Italy): study
of past events and provision for the future. In: B
roMhEAd
E.n., d
ixon
n. & i
BSEn
M.-L. (E
dS
.). Landslides in Research,
Theory & Practice (Vol. 1). Proc. 8
th
Int. Symp. Landslides, 415-420. Thomas Telford, London.
D
ixon
N. & B
roMhEAd
E.N. (2002) - Landsliding in London Clay coastal cliffs. Q.J. Eng. Geol. Hydrogeol., 35: 327-343.
d
oglioni
C. (1987) - Tectonics of the Dolomites (Southern Alps, Northern Italy). J. Struct. Geol., 9: 181-193.
E
iSBAChEr
g.h. & C
lAguE
J.J. (1984) - Destructive mass movement in high mountains, hazard and management. Geological
Survey of Canada Paper, 84-116.
P
Aronuzzi
P. & B
ollA
A. (2012) - The prehistoric Vajont rockslide: an updated geological model. Geomorphology, 169-170:
165-191.
P
EdrAzzini
A., F
roESE
C.r., J
ABoyEdoFF
M., h
ungr
o. & h
uMAir
F. (2012) - Combining digital elevation model analysis and
run-out modeling to characterize hazard posed by a potentially unstable rock slope at Turtle Mountain, Alberta, Canada.
Eng. Geol., 128: 76-94.
r
iVA
M., B
ESio
M., M
ASEtti
d., r
oCCAti
F., S
APigni
M. & S
EMEnzA
E. (1990) - Geologia delle valli Vaiont e Gallina (Dolomiti
orientali). Annali dell' Università di Ferrara, Sezione: Scienzedella Terra, 2 (4): 55-76.
S
ElBy
M.J. (1993) - Hillslope materials and processes. 2
nd
Ed. Oxford University Press, Oxford.
S
turzEnEggEr
M. & S
tEAd
d. (2012) - The Palliser Rockslide, Canadian Rocky Mountains: characterization and modeling of a
stepped failure surface. Geomorphology, 138: 145-161.
V
iEro
A., t
EzA
t., M
ASSironi
M., J
ABoyEdoFF
M. & g
AlgAro
A. (2010) - Laser scanning-based recognition of rotational
movements on a deep seated gravitational instability: the Cinque Torri case (North-Eastern Italian Alps). Geomorphology,
122: 191-204.
Statistics