Document Actions

ijege-13_bs-genevois-tecca.pdf

background image
Italian Journal of Engineering Geology and Environment - Book Series (6) www.ijege.uniroma1.it © 2013 Sapienza Università
Editrice
15
DOI: 10.4408/IJEGE.2013-06.B-02
THE VAJONT LANDSLIDE: STATE-OF-THE-ART
R
inaldo
GENEVOIS
(*)
& P
ia
R
osella
TECCA
(**)
(*)
University of Padova - Department of Geosciences - Padova, Italy
(**)
CNR-IRPI - Padova, Italy
Rupture on the slope and road, ….., the enlargement
of the main landslide scarp,…., make me think the
worst. Water level is lowering, ….. It is my intention to
reach the 695 level in order to have a safety margin in
case of waves. … May God help us!
On October 9, 1963 at 10:39 p.m. local time,
about 270÷280 million m
3
of rocks slid into the Vajont
reservoir creating a giant water wave that overflowed
the dam, destroyed the many villages, killing about
2000 people.
On July 1965, a parliament Board of Inquiry stated
that could not have been foreseen. On February 1968
three engineers of SADE, the firm charged of the dam
construction, and of the department of public works
were convicted; one of them was acquitted in appeal.
The mechanics of the final collapse, still poorly un-
INTRODUCTION
The next 9
th
of October 2013 is the 50
th
anniversary
of the Vajont landslide. Many questions, legal, econom-
ic, social and scientific have accompanied the history
of the dam and the management of the emergency con-
cerning the Vajont reservoir slopes instability.
The dam was built by SADE-Società Adriatica
di Elettricità (Adriatic Energy Corporation), the elec-
tricity supply and distribution monopolist in North-
Eastern Italy at that time. The dam and the basin were
intended to be at the center of a complex system of
water management in which water would have been
channeled from nearby valleys and artificial basins
located at higher levels.
The long lasting story of Vajont dam begins at the
dawn of the 20th century and ends abruptly on 9
th
Oc-
tober 1963 at 10:39 p.m. local time, in a 4-5 minutes
of apocalypses. It is not only a story, as the journalist
Tina Merlin wrote, of the “arrogance of strong power
and subjection of humanity and science to economic
reasons”, but has to be regarded as a clear sign of the
engineering geological knowledge existing at that
time. At that time the dam was the tallest (265.5 m)
concrete double arched dam in the world. After a three
years long period of slow but on the whole accelerat-
ing movements on the morning of 9
th
October the situ-
ation was really critical, so that the engineer director
of the construction site wrote to his head a dramatic
letter: “In these past few days translational speeds
within the landslide have clearly increased.
…….
Fig. 1 - The village of Longarone after the catastrophic
water wave
background image
R. GENEVOIS & P.R. TECCA
16
International Conference on Vajont - 1963-2013 - Thoughts and analyses after 50 years since the catastrophic landslide Padua, Italy - 8-10 October 2013
summary of the worldwide researches on this
event, embracing different disciplines from geolo-
gy to engineering geology, hydrogeology, geomor-
phology, geomechanics and to hydraulics. It is not
a review of all the papers written on this subject,
but the main papers are reported.
4) The current studies: they refer only to the researches
carried out in the range of the direct collaboration
of the Geosciences Department of the University
of Padova with the Simon Fraser (Canada), Mi-
lano La Biococca and Pavia Universities.
ἄριστον μὲν ὕδωρ
(Pindar - Greatest however [is] water)
THE VAJONT DAM HISTORY
THE DESIGN OF THE DAM: 1925 - JANUARY 1957
Preliminary studies for the Vajont dam construc-
tion began as early as in 1920 and in 1928 two pos-
sible cross sections were preliminarily individuated in
derstood, has constituted and still constitutes a challenge
to physicists, geologists and engineers over the last 50
years. The reasons why some catastrophic landslides
move so fast and so far, cannot be fully explained yet,
and the 1963 Vajont landslide is a tragic example of it.
Owing to its intrinsic geological and geomechanical
complexity, the Vajont rockslide has been considered as
the starting point for the development of modern rock
mechanics and rock engineering. Besides, the disaster
exerted a great influence in the field of both engineer-
ing and geology, marking a turning point in the attention
given in hydro-projects to the reservoir slopes, as was
not the case before. Nowadays the impact of unstable
slope must be evaluated and an implicit, if not explicit
comparison is always made with the Vajont slopes.
For that, reports and scientific literature on the
Vajont landslide are very numerous: no other slide in
the world has been the subject of so many studies and
researches.
The scientific community has never stopped the
thorough study of the combined causes and the final
mechanisms of this tragic event. But, despite the great
scientific interest related to this catastrophic episode,
most of the questions concerning the failure mecha-
nism, the high velocity and the general characteristics
of the slide should be considered not yet satisfactorily
explained. Many geological, hydrogeological, and
geotechnical aspects still remain unexplained, par-
ticularly as regards whether the 1963 slide was a new
landslide or the reactivation of a prehistoric slide.
On the other hand, the event was unusual in that
the flood was not caused by the dam collapse, but a
catastrophic landslide from the slope of Monte Toc
that pushed over the dam about 30 million m
3
of the
water in the newly created reservoir. Nevertheless the
dam resisted!
This keynote will be a short review of the events
that preceded the Vajont disaster and of the studies and
researches carried out both before and after the tragic
landslide. It is structured in four main parts:
1) The chronicle of a disaster: it represents a tem-
poral list of the events and the data gradually ac-
quired up to the catastrophic landslide.
2) The studies before the failure: it represents the
state-of-the-art of geological and engineering
geological knowledge as they increased from the
beginning to the ending failure.
3) The studies and researches after the failure: it is a
Fig. 2 - The Vajont gorge at Colomber bridge
background image
THE VAJONT LANDSLIDE: STATE-OF-THE-ART
Italian Journal of Engineering Geology and Environment - Book Series (6) www.ijege.uniroma1.it © 2013 Sapienza Università
Editrice
17
1
st
April 1958 - The High Council of Public Works ap-
points the "Commissione di Collaudo", a board
of experts with the aim to control works execu-
tion and procedures: first inspection on 19
th
July
1958: Prof. Penta and Prof. Mueller agree on the
consolidation of the down-valley slopes with nails
and rock anchors.
September 1959 - The construction of the dam is ac-
complished: 261.6 m high, 190.2 m long, 22.1
m thick at base and 3.4 m at crest, 725.5 m a.s.l.
at crest.
22
th
October 1959 – Second field survey of the "Com-
missione di collaudo".
26
th
November 1960 - "Commissione di collaudo"
field survey.
10
th
April 1961 - Fourth field survey of the members
of the "Commissione di collaudo" Prof. Penta and
Eng. Sensidoni.
17
th
October 1961 - Fifth field survey of the "Commi-
sione di collaudo": the filling of the reservoir may
continue, but further coincidental movements
cannot be escluded.
16
th
November 1961 - The filling of the reservoir may
continue up to 640 m a.s.l. elevation and later up
to 655 m a.s.l. (December 1961) and to 675 m
a.s.l. (6
th
February 1962).
GEOLOGICAL AND GEOMECHANICAL STU-
DIES
After the design phase, a great number of geologi-
cal studies and reports accompanied the construction
and the operation tests of the dam as a consequence of
both the variations requested by SADE and the con-
tinuously raising problems related to the geological
and stability problems of the area.
31
st
January 1957 - Prof. G. Dal Piaz report, integrat-
ing his 25
th
March 1948 report, is favorable to
the increase of the dam height to 266 m, but with
some precautions regarding the dam shoulders
and the needed waterproofing works.
6
th
August 1957 - A report of dr. L. Mueller underlines
the poor characteristics of the rock mass on the
left slope of the valley.
September 1957 - Prof. G. Dal Piaz report (raft dates
9
th
June 1957), attached to the Eng. C. Semenza
detailed design (2
nd
April 1957) confirms the va-
lidity of all the studies until then carried out.
29
th
October 1958 - The left Vajont slope is studied by
the Vajont valley by SADE.
30
th
January 1929 - SADE asks for the permit to
use the Vajont torrent waters (Eng. C. Semenza
project).
4
th
August 1928 - Prof. Giorgio Dal Piaz, the geolo-
gist charged by SADE, suggests the dam location
downstream of the Colomber bridge, where actu-
ally the dam will be built, rejecting the 1925 hy-
pothesis of the Swiss geologist Prof. J. Hug.
7
th
-9
th
August 1937 - Prof. G. Dal Piaz confirms the
chosen location for a 180 m high dam.
22
th
June 1940 - SADE designs the hydraulic Boite-
Piave-Vajont system. A further geological report
of Prof. G. Dal Piaz corroborates the choice al-
ready made and underlines once more the suitable
conditions of the Vajont valley. The project is ap-
proved by the Public Works High Council on 15
th
October 1943
15
th
May1948 - SADE submits the detailed design of
a 202 m high dam with a new geological report of
Prof. G. Dal Piaz (25
th
March 1948) which repeats
and confirms his previous favourable conclusions.
11
th
October 1948 - Eng. C. Semenza asks to Prof. G.
Dal Piaz the possibility to increase the height of
the dam to an elevation of about 730 m a.s.l.
21
th
December 1948 - In another geological report,
further improved in 18
th
November 1953, Prof.
G. Dal Piaz pays attention to the stability of the
La Pineda landslide and, mainly, to the stability
condition of the village of Erto on the right slope
of the valley.
THE CONSTRUCTION OF THE DAM: JA-
NUARY 1957 - SEPTEMBER 1959
January 1957 - Construction works begin.
31
th
January 1957 - SADE presents the "Great Vajont
Project": impounded waters will derive from the
torrents Vajont, Maè and Boite. SADE requests
the maximum level of the reservoir to be raised
to 722.5 m a.s.l., accompanying the demand with
an integration of Prof. G. Da Piaz to his geologi-
cal report dated 25
th
March 1948, that confirms his
previous conclusions.
2
nd
April 1957 - SADE presents the detailed design
for a 266 m high dam (150x10
6
m
3
capacity), later
approved (15
th
June 1957) by the High Council of
Public Works that, however, asks for further geo-
logical surveys.
background image
R. GENEVOIS & P.R. TECCA
18
International Conference on Vajont - 1963-2013 - Thoughts and analyses after 50 years since the catastrophic landslide Padua, Italy - 8-10 October 2013
degradation of the rock mass characteristics due
to the deformation processes.
2
nd
March 1961: - Prof. Caloi, in a letter to the Eng. Toni-
ni writes that the bedrock was sufficiently sound.
5
th
May 1961 - Dr. Pacher, assistant of dr. Mueller,
analyses the results of a 100 m long exploratory
tunnel dug on to the left of the Massalezza creek
at an elevation of about 900 m a.s.l. (spring 1961),
and concludes with the existence of a number of
although limited slip surfaces, a fact that contrasts
with the hypothesis of a unique deep shear surface.
31
st
October 1961. Prof. Penta, on the basis of 10
th
April
and 17
th
October 1961 field surveys, confirms pre-
vious conclusions (15
th
April 1961 letter) on the
existence of a shallow, possibly dormant landslide.
3
rd
May 1962 - No new landslides are observed by
Eng. Beghelli (Belluno Public Works Bureau
("Genio Civile") during the field survey.
THE CHRONICLE OF A CATASTROPHE
A mess of almost impenetrable and contrasting
events, technical aspects and human behavior im-
printed the long inexorable approach to the 9
th
Octo-
ber disaster. The beginning might be set on 22
th
March
1959 when a 3x10
6
m
3
landslide occurred in the near
valley of the torrent Maè, falling into the reservoir of
Pontisei and generating a water wave about 20 m high.
The consulting geologist is Prof. Penta and a few days
after Prof. P. Caloi write that the landslide had been
exactly forecasted in one of his report. Considering
only the lapse after the end of the dam construction
(September 1959), a concise list of the main facts is
the following.
19-21
th
July 1959 – First field survey of the "Commis-
sione di collaudo", the board of experts charged to
control works execution and procedures: no prob-
lems are reported.
2
th
February 1960 - The first filling of the reservoir (up
to 595 m a.s.l. elevation) begins.
Prof. G. Dal Piaz for the construction of a road:
only small rock falls are forecasted.
10
th
October 1959 - Doubts are expressed dr. L. Muel-
ler on the stability of the left slope which requires
new field surveys.
4
th
February 1960 - Prof. P. Caloi, resting on data ob-
tained with a seismic station installed on Decem-
ber 1959 near the dam, states that the left slope is
composed by sound rocks showing very high elas-
tic moduli, with a shallow (10-12 m) detritic cover.
June 1960 - On the basis of field surveys carried out
(1959-spring 1960) the presence of an old wide
and deep landslide on the left slope is supposed by
geologists dr. E. Semenza and dr. Giudici.
9
th
July 1960 - Prof. G. Dal Piaz confirms his previous
conclusions and only small and slow phenomena
might be possible. However, he doesn't consider
the Semenza-Giudici hypothesis.
October 1960 - On M.t Toc slope a 2 km long M-
shaped fissure opens up at 1200-1400 m a.s.l.,
defining an area about 1700 m long and 1000 m
large.
15-16
th
November 1960 - A sketch in the report of
Prof. Mueller shows that he had actually consid-
ered the instability of a huge rock mass on a deep
slip surface.
26
th
November 1960 - Prof. Penta supposes that the
M-shaped fissure, opened in October 1960, could
be the proof either of a very deep landslide or of
a more superficial and slow one. More accurate
and deep geological surveys are suggested (memo
1
st
December 1960) and were accepted and pre-
scribed by SADE.
3
rd
February 1961 - Resting on all surveys and studies
carried out by Prof. P. Caloi, by the geologists E.
Semenza, Giudici and Penta and by Dr. Broili, dr.
Mueller concludes: i) the volume of the moving
mass is about 2x10
8
m
3
; ii) the moving mass is
split in different independent parts; iii) if so, the
three parts of the landslide will occur in different
times; iv) the landslide could be partially control-
led by artificial landsliding.
10
th
February 1961 - Prof. Caloi, integrating his 23
rd
February 1961 report, concludes: i) seismic wave
velocities are less than those previously meas-
ured (1959); ii) the velocity decrease might be
attributed to both unavoidable differences in the
execution of the seismic surveys and an intense
Fig. 3 - The M-shaped perimeter fissure (photo E. Se-
menza)
background image
THE VAJONT LANDSLIDE: STATE-OF-THE-ART
Italian Journal of Engineering Geology and Environment - Book Series (6) www.ijege.uniroma1.it © 2013 Sapienza Università
Editrice
19
21
st
February 1961 - The risk of the failure of the Vajont
left slope is reported in an article of the local daily.
3
rd
February 1961 - Dr. L. Muller supposes a 2x108 m3
landslide and such a landslide cannot be stabilized.
10
th
April 1961 - Prof. Penta and Eng. Sensidoni, mem-
bers of the "Commissione di collaudo" state: i)
surface displacements are no more present; ii) no
indication about a deep landsliding phenomenon
is evident; iii) no immediate danger is discernible.
15
th
April 1961 - Prof. Penta after a new field survey
affirms that no hazard can be detected with the
reservoir water level at 600 m a.s.l..
20
th
April 1961 - After the drawdown no more dis-
placements are detected, but landslides phenome-
na are suspected (letter of Eng. C. Semenza), even
if Prof. G. Dal Piaz and Prof. Penta are optimist.
10
th
May 1961 - The by-pass gallery is accomplished.
August-September 1961 - Four standpipe piezometers
are installed on the left slope at depths between
167 m and 221 m. One of them is immediately
out-of-order.
31
st
October 1961- the filling of the reservoir can be
continued as the by-pass gallery is finished (Prof.
Penta report on 10
th
April and 17
th
October 1961
field surveys).
10
th
April 1962 - On the basis of data obtained by the
seismograph installed near the dam during 1959,
Prof. P. Caloi concludes: i) the microseismic ac-
tivity observed in 1960, during the first reservoir
water level raise to 630 m a.s.l., has not been
observed during the second raise to 650 m a.s.l.
(November-December 1961 to January 1962); ii)
corresponding displacements velocities are very
low with respect to those monitored in the 1960
fall; iii) no other surface facts are observed.
15
th
June 1962 - A rather intense seismic activity is
registered in the period 27
th
April-13
th
May 1962:
it is attributed to natural phenomena and not de-
pendent on the reservoir filling, as shakes are
external to the landsliding area (Report of Prof.
Caloi and his assistant dr. Spadea).
3
rd
July 1962 – The 700 m a.s.l. reservoir water level
is considered precautionary: for that elevation,
results of physical modelling at "Centro Modelli
Idraulici di Nove di Fadalto" indicate that, in the
worst case of possible landslides, the volume of
the displaced water would be about 20.000 m
3
with
an instant discharge of 2.000 m
3
for 20 seconds.
March 1960 - A landslide occurs on the left slope just
upstream of the Massalezza creek confluence, at
an elevation of about 850 m a.s.l., that is at the
border of Pian della Pozza.
June 1960 - The geologists E. Semenza and Giudici
expose the hypothesis of the existence of a huge
old landslide on the left slope.
11
th
June 1960 - A reservoir water level to 660 m a.s.l.
is allowed.
October 1960 - A long M-shaped fissure, 1800 m long
and 1100 m wide as a mean, opens on the left
slope, defining an area of about 2 km
2.
4
th
November 1960 - A landslide of 7-8x10
5
m
3
occurs
on the left slope.
15
th
÷16
th
November 1960 - Following this event, the
Eng. C. Semenza, the dr. L. Mueller, the geologist
E. Semenza and alii decide the drawdown of the
reservoir and the construction of a by-pass gallery
with the aim to control the water level in the east-
ern part of the reservoir in case of the occurrence
of the deep and huge landslide that dr. L. Muller
hypothesizes.
November 1960÷January 1961 - A slow drawdown of
the reservoir down to 600 m a.s.l. is carried out
and the level maintained until October 1961. The
lapse is necessary for the construction of the by-
pass gallery, that begins on january 1961.
31
st
January 1961 - SADE charges the "Centro Model-
li Idraulici di Nove di Fadalto", a SADE research
centre entrusted to Prof. A. Ghetti (Institute of
Hydraulics, University of Padova), to model the
dam and the reservoir and to evaluate the effect of
a landslide into the reservoir. The assumed model
hypotheses are: two different and consecutive 20-
40x10
6
m
3
landslides; reservoir water levels at
680 m and 720 m a.s.l.; the rock mass modelled
by rounded gravels.
Fig. 4 - The 4
th
November 1960 landslide (Photo Semenza)
background image
R. GENEVOIS & P.R. TECCA
20
International Conference on Vajont - 1963-2013 - Thoughts and analyses after 50 years since the catastrophic landslide Padua, Italy - 8-10 October 2013
8
th
July 1962 - A displacements acceleration is ob-
served during the reservoir filling from 660 to 685
m a.s.l. New fissures open parallel to the valley
axis; displacements at monitored points show dif-
ferent settlement values; water levels in installed
piezometers are as in the reservoir (field observa-
tions of the Government’s person in charge). But,
further landslidings is deemed to occur, in case,
slowly and involving a mass limited uphill by the
new fissures. A new geological survey is consid-
ered, however, necessary, and SADE requests the
possibility of local inhabitants evacuation.
17
th
November 1962 - Water level in the reservoir
reaches 700 m a.s.l.
2
nd
December 1962 - Drawdown up to 647.5 m a.s.l.
achieved on 10
th
April 1963.
10
th
January 1963 - With respect to July 1962 an in-
crease of the displacements velocities is observed:
it begins with the December 1962 drawdown from
700 to 650 m a.s.l., but at a reservoir elevation of
650 m a.s.l. velocities are very slow. Piezometers
show water levels always as in the reservoir.
20
th
March 1963 - SADE requests the filling of the
reservoir up to 715 m a.s.l.
22
nd
July 1963 - The presence of "cloudy water of
reservoir, continuous rumbles and tremors in the
area" is exposed (The Major of Erto).
4
th
September 1963 - The water level in the reservoir
reaches 710 m a.s.l.
2
nd
September 1963 - A new seismic shake. The dis-
placements of the monitored points increase con-
tinuously.
15
th
September 1963 - A new fissure opens in the slope
and the perimeter M-shaped fissure widens.
27
th
September 1963 - The drawdown of the reservoir
begins.
2
nd
October 1963 - New fissures open and further dis-
placements are observed.
5
th
October 1963 –A settlement area in Pian della Poz-
za is observed (Prof Caloi and dr. Spadea report).
7
th
October 1963 - New fissures about 10 m long open.
In the night the evacuation of inhabitants of Mt.
Toc area is ordered.
9
th
October 1963, morning - The discharge channel of
the dam is clogged up and the displacement veloc-
ity further increase.
9
th
October 1963, 13:00 local time - A new fissure 5 m
long and 0.5 m wide opens and develops.
9
th
October 1963, 20:00 local time - The road on the
left side is no more passable.
9
th
October 1963, 22:39 local time: a single 2.7x10
8
m
3
landslide occurs. The resulting water wave flows
in the Vajont razing a number of villages to the Est
(Frasein, San Martino, Col di Spesse, Patata) and,
after overtopping the dam, to the West (Longar-
one, Codissago, Castellavazzo, Villanova, Pirago,
Faè, Rivalta). About 2.000 people dead, but the
dam is intact!
11
th
October 1963 - The Public Works Minister ap-
point a bord of enwuiry on the disaster.
15
th
February 1997 - After about 34 years all the penal
and civil suits are definitely concluded.
THE STUDIES BEFORE THE FAILURE
Before the catastrophe, many geological studies
were carried on the Vajont valley. Initial geological
studies interpreted the Vajont Valley as a typical exam-
ple of a syncline (the Erto Syncline), located along the
axis of the valley (B
oyeR
, 1913). Specific geological
studies, related to the dam construction project, date
back to 1901 A.D., even if for a dam only 8 m high.
The first geological studies for an important dam
date at 1925 A.D., when the Swiss geologist Prof. J.
Hug recommends the neighbourhood of the Casso
Bridge for the dam location, where he considered the
Fig. 5 - The hollow of Pian della Pozza
Fig. 6 - The dam reservoir after the landslide
background image
THE VAJONT LANDSLIDE: STATE-OF-THE-ART
Italian Journal of Engineering Geology and Environment - Book Series (6) www.ijege.uniroma1.it © 2013 Sapienza Università
Editrice
21
and sound. The rock masses at the Colomber Bridge
section are uniform and compact and permeability
problems would be easily passed considering the tech-
niques available at that time. Instead, Giurassic and
Cretaceous marly limestones outcropping at the valley
bottom near the Casso Bridge area are characterized
by many fractures both parallel and normal to the val-
ley axis as a consequence of the overturning of the
whole stratigraphic series.
In the following Prof. G. Dal Piaz studies in
more depth the geological features of the valley, but
considering only the rock mass of the abutments and
the right side of the valley (5
th
June 1940, 25
th
March
1948, 21
st
December 1948, integrated on 18
th
Novem-
ber 1953, 31
st
January 1957 and 9
th
June 1957 geologi-
cal reports). The Vajont valley has been cut in Jurassic
to Eocenic rocks: in its narrower lower part, the strata
(Dogger) dip to East and their thickness progressively
decrease towards the valley bottom . The overturning
of the stratigraphic series on the right side of Vajont
valley is confirmed: a great overturned anticline (Dog-
ger and Lias limestones) recumbent to South and with
an axis inclined to East, lying unconformably on the
Eocenic "Scaglia Rossa", is described. This geologi-
cal structure is complicated by the presence of two
synclines folds having the Eocenic "Scaglia Rossa"
formation at the core one of which is found near the
outcropping rocks ("Fonzaso" and "Scaglia Rossa"
Formations) more suitable in comparison with those
outcropping near the Colomber Bridge, supposed to
be too permeable.
Except these very preliminary studies, the geolog-
ical and geomorphological setting of the valley for a
significant high dam was studied by Prof. G. Dal Piaz,
the head of the Geology Institute of the University of
Padova, from 1928 A.D. until on 20
th
April 1962.
The Vajont valley, between M.t Toc to the South
and M-t Salta to the North, is a very narrow gorge
stretching from the confluence with the Piave river to
the West up to the site of Casso village. Proceeding
towards East the valley widens, remaining however
very narrow in its lower part. The steep slopes are
characterized by many wide morphological steps and
one of these, 700-800 m long, is present on the left
valley side at an elevation of about 850 m a.s.l.
According to his initial geological report (4
th
Au-
gust 1928 and 7
th
-9
th
August 1937) the Vajont valley is
eroded along the axis of an East-West trending asym-
metrical syncline plunging upstream to the East (Erto
syncline). It is set in dolomitic limestones, no tectonic
accidents are evident and rock masses are uniform
Fig. 7 - The village of Longarone before (left) and after
(right) the catastrophe
Fig. 8 - Longitudinal sketch of the ajont valley with the
two considered location of the dam. Approximate
length scale 1:100.000
Fig. 9 - The Erto syncline (left valley slope) and the over-
turned anticline (right valley slope) (from S
emenza
& G
hirotti
, 1998)
Fig. 10 - Map of the pre-1963 Vajont landslide area (from
G
enevoiS
& G
hirotti
, 2005)
background image
R. GENEVOIS & P.R. TECCA
22
International Conference on Vajont - 1963-2013 - Thoughts and analyses after 50 years since the catastrophic landslide Padua, Italy - 8-10 October 2013
at depths from a few meters to 18 m under a detrital
cover considered stable.
The left side of the Vajont valley is studied and de-
scribed by G. Dal Piaz only on 1960 A.D. (9
th
July 1960
report), also in the wake of the landslide occurred (4
th
November 1960) on the left slope just upstream of the
Massalezza creek confluence (December 1960 annex).
Some but shallow landslides, probably involving the
morainic deposits, are described between the "Pineda"
and "Pian della Pozza" areas. The "Pian della Pozza" is
interpreted as a glacial morphological terrace charac-
terized by a shallow elliptical doline where limestones,
under the detrital cover, present very intense karst fea-
tures (karren) and long and deep fractures parallel to the
valley. Many other dolines are observed near the border
of the cliff were occasional rock falls and shallow but
substantially slow slides might occur.
Landslide deposits are described between "Pian
della Pozza" and the dam site, but the rock mass is
depicted as generally sound. In the "Pian della Pozza"
and "Pian del Toc" areas long fractures parallel to the
valley are reported and attributed to movements of
the rocks underlying the morainic deposits. Vertical
fractures normal to the valley are also indicated. As
observed outcropping limestones are deeply fractured
on the higher parts of both the left and right abutments
but, downwards, the quality of the rock mass progres-
sively improves. The concave feature of the left slope
between "Pian della Pozza" and the Vajont river bed is
attributed to a previous landslide.
At the beginning, the good quality of the rock mass
is confirmed by the results of a seismic survey carried
out on the left valley side (4
th
February 1960, Prof. P.
Caloi report). The bedrock, with high elastic moduli,
results covered by 10 m to 18-20 m thick landslide
deposits. However, further seismic surveys (Decem-
ber 1960) showed the presence of 20-40 m thick loose
materials over a rock mass deeply fractured with elas-
village of Casso. The presence of Flysch (alternances
of marls and sandstones) and some small shallow
landslides are observed in the area of the village of
Erto. Two joint systems, parallel and normal to the
valley, are described on the right side and considered
responsible for the development of existing rockfalls.
From a geomorphological point of view, the pres-
ence of morainic and landslide deposits is described on
the uphill part of the Vajont right side. There, a huge
rock mass is hypothesized to have slid in the past cov-
ering also great part of the left valley side where, new
but not relevant landsliding phenomena could occur
as a consequence of the filling-drawdown operations.
The area of the village of Erto on the right valley side
was investigated by a number of boreholes and small
exploratory galleries dug at 730 m a.s.l. elevation. The
bedrock turned out to be constituted by sound marls
(“Scaglia Rossa” and “Scaglia Cinerea” formation)
under a 20-40 m thick detrital cover made by calcare-
ous and marly blocks and fragments in a sandy-clayey
matrix (Technical report on boreholes carried out dur-
ing 1954 in the Erto area by geologist E. Milli).
The good quality of the rock mass of the abut-
ments is corroborated by the results of some boreholes,
carried out for the preliminary project and by the new
road gallery excavated on the right side at about the
same elevation of the planned crest dam (725 m a.s.l.).
Unconfined compression tests, carried out on Jurassic
limestones, shown on the average a strength decrease
from 209 MPa, at the bottom of the valley, to 119-134
MPa at 725 m a.s.l. elevation (crown of the dam). In
conclusion, geological and morphological conditions
at the Colomber Bridge area are deemed suitable as a
location for the planned dam.
In order to complete the field surveys in the area
of Erto, new boreholes were carried out in 1960
finding the bedrock (Flysch and “Scaglia Rossa”)
Fig. 11 - The “Pian della Pozza” depression
Fig. 12 - Semenza’s hypothesis on slip surfaces of the an-
cient landslide (from S
emenza
& G
hirotti
, 2000)
background image
THE VAJONT LANDSLIDE: STATE-OF-THE-ART
Italian Journal of Engineering Geology and Environment - Book Series (6) www.ijege.uniroma1.it © 2013 Sapienza Università
Editrice
23
The geological structure of this sector of the Va-
jont valley is reconstructed and its role in the landslide
mobilization is stressed. The Vajont Valley coincides
with the axis of an East-West trending, asymmetrical
syncline eastward plunging. The northern slope of
Monte Toc has a “chair-like” structure, clearly vis-
ible from the Piave Valley in front of Longarone. The
landslide involved mainly limestones and marls of the
Socchér Formation (Jurassic and Cretaceous), that is
supposed to slid down over one or more thin clay lay-
ers contained in the Fonzaso Fm (Malm). The failure
surface was supposed to be bounded to the North by
the level of cataclasites outcropping at 600 m a.s.l., to
the South by the elongated depression of "Pian della
Pozza", and to the East by a vertical fault. In conclu-
sions the landslide mass could be mobilized.
The possibility of a very deep and huge landslide
was hypothesized also by Prof. Penta based on the 28
th
November 1960 field survey. He concluded, however,
that the data so far collected were not sufficient to
specify the nature and the characteristics of the ob-
served phenomenon and recommended a very slow
drawdown of the reservoir. After the opening of the
long M-shaped fissure on October 1960, the geological
studies of G. Dal Piaz and E.Semenza-F. Giudici were
considered substantial by L.Mueller and his collabo-
rators dr. Fally and dr. Broili. They integrated them
with stratigraphic and displacement data collected at
the time and evaluated the volume of the landslide in
about 2x10
8
m
3
with a shear surface located at a depth
variable from 250 m, in the western part, to 200 m
in the eastern one. Besides, they recognized two sec-
tors in the landslide mass separated by the Massalezza
creek and characterized by different sliding mecha-
nisms. The western sector with displacement vectors
dipping N20° as a mean with inclinations decreasing
tic moduli 10 times smaller than before, while more
rigid material was individuated only at depths of about
100-150 m (10
th
and 23
th
February 1961, Prof. P. Caloi
report). The strong drop observed in both longitudinal
velocities (from 5-6 to 2.5-3.0 km/s) and frequencies
(from 200 to 50 Hz) is attributed to the rocks crush-
ing due to the failure of rock masses located at higher
elevations consequent to the seismic shakes occurred
all along the 1960.
On the same year, geologists E. Semenza and F.
Giudici recognized the presence of an enormous and
ancient landslide deposit on the left side of the Vajont
valley, in the area of “Pian del Toc” and “Pian della
Pozza”. The landslide deposit was deeply cut by the
Vajont stream forming the present gorge and leaving a
portion on the right side of the valley ("Colle Isolato).
This hypothesis was confirmed later on during the ex-
cavation of the by-pass tunnel (1961) when at the base
of the "Colle Isolato" alluvial gravels and a thin layer
of cataclasites were found.
Fig. 13 - The 1959 Semenza's sketch of the Vajont valley
before (I) and after the ancient landslide (II). The
new river channel cut in the landslide body (III)
and the aaluvial deposits found under the remnant
of the ancient landslide (“Colle Isolato”)
Fig. 14 - Part of the perimetral M-shaped fissure opened on
October 1960
background image
R. GENEVOIS & P.R. TECCA
24
International Conference on Vajont - 1963-2013 - Thoughts and analyses after 50 years since the catastrophic landslide Padua, Italy - 8-10 October 2013
from 27°-34° to almost zero proceeding downwards.
In the flat part of the "chair" there are antithetic frac-
tures attributed to rotational movements.
Muller hypothesized a pulsating glacier-like move-
ment of the lower half of the landslide mass initiated
since many centuries or more before and facilitated
by the presence in the lower part of the landslide mass
of a number of folds with axes almost horizontal. The
seismic surveys of P. Caloi, indicating the absence of
an effective slip surface, supported on one hand this
hypothesis, but on the other hand were opposed, indi-
cating the presence of sound rocks in the lower part
of the landslide mass. The eastern sector was supposed
to behave as the higher part of the western one, that is
sliding on a more or less inclined and planar surface.
Many triggering factors were indicated (low values
of the shear and tensile strength; strata, fractures and
fold axes orientation; high and different permeability
of rock masses; high water pressure and cyclic stresses
due to earthquakes). The water level in the reservoir is
indicated to influence directly the displacement rate, as
can be inferred by the 4
th
November 1961 landslide,
occurred with the reservoir level being increased to 646
m a.s.l.). Fast movements are forecasted to occur with
the drawdown operations. In conclusions the landslide
cannot be stopped, but only controlled. Besides, con-
sidering that the 4
th
November 1960 landslide caused
in the reservoir a 2 m high wave, a greater landslide
is supposed to determine a water wave 40 m high pos-
sibly causing the instability of the dam abutments that
have to be, thus, consolidated (3
rd
February 1961 re-
port). A small test gallery was excavated during 1961
in the Massalezza area with the aim to get the hypoth-
esized failure surface: at 28
th
February 1961 it had a
length of 37 m, but no data are available.
In a following report (20
th
September 1962) the
displacements, measured by the instruments installed
on the dam abutments in the period 1959-1961, were
considered by L. Mueller a consequence of the first
filling and drawdown operations and confirmed the
results of the tests carried out by ISMES on a physi-
cal model. However, up to this time no alarming
deformation was observed on the abutments: small
deformations were observed as a consequence of the
reservoir level variations.
At the beginning of 1961 SADE proposes to
the”Centro Modelli Idraulici” (Hydraulic Modeling
Center) of the University of Padova to carry out a
research on a hydraulic model to examine the effects
of a landslide on the Vajont reservoir, as regards
mainly the consequent flood on the reservoir slopes
and over the dam.
The 1:200 scale model (29 m long, 12 m wide and
1.80 m high) was considered suitable to represent the
wave motion. The natural landslide is reproduced with
round gravels either placed on a metallic plane surface
or placed on a curved surface representing the geologi-
cally forecasted failure surface. The mass is, then, ar-
tificially drag in such a way to obtain both independ-
ent movements of the two hypothesized sectors and
different velocities, so that the landslide effects could
be examined in different cases. Tests were carried out
considering the water level at 722.5 m, 700.0 m and
670.0 m elevation. Considering the maximum elevation
of the water in the reservoir, the effects of the landslide,
occurring in 3 or 1.5 minutes, are most relevant for
the downstream sector: discharges over the dam range
from 12.000 m
3
/s to 30.000 m
3
/s and wave heights over
the dam from 11.5 m to 22.0 m. Hypothesizing two dif-
ferent sectors, landslide time results between 1 and 3.5
minutes with a time interval of 2.5 and 10.5 minutes:
discharge results of 20.000 m
3
/s with a wave height
over the dam of 16.0 m. Lower values of both discharge
and wave height were obtained considering the occur-
rence of the only downstream sector of the landslide or
Fig. 15 The 4
th
November 1960 landslide on the left reser-
voir slope
background image
THE VAJONT LANDSLIDE: STATE-OF-THE-ART
Italian Journal of Engineering Geology and Environment - Book Series (6) www.ijege.uniroma1.it © 2013 Sapienza Università
Editrice
25
tragic 8
th
October 1963 should be checked considering
also the diagram showing the trend of monitored data in
conjunction with both the filling-drawdown operations
and specific events.
FIRST FILLING AND DRAW-DOWN OF THE
RESERVOIR
It initiates on February 1960. A small detachment
occurred by March 1960 when the level of the reser-
voir reached about 610 m a.s.l. In October 1960, with
the reservoir level at 650 m a.s.l., the displacements
rate rapidly increases to about 3.5 cm/day and a long
fissure opens up, suggesting that a quite large land-
slide could have been mobilized. On 4
th
November a
700,000 m
3
landslide occurred. The reservoir level is
gently dropped back to 615 m a.s.l. and the displace-
ments rate reduces drastically.
The designers were conscious that was not possible
to arrest artificially the landsliding mass or to acceler-
ate the sliding process and that the danger arising by
an uncontrollable reservoir level would have been too
great. They assumed that the huge landslide mass could
have been controlled carefully varying the reservoir
level. They, besides, calculated that, in case of the oc-
currence of a sudden failure, over-topping of the dam
would be avoided, so long as the landslide would have
not filled the reservoir in less ten minutes. The follow-
ing draw-down to 615 m a.s.l. leads to a decrease of
the movement rates to less than 1 mm/day. By this time
the landslide mass had moved an average of about 1 m.
SECOND FILLING AND DRAW-DOWN OF
THE RESERVOIR
It begins on October 1961 and through to early Feb-
ruary 1962 the reservoir level is raised to 665 m a.s.l. In
November 1962 the level reached 715 m a.s.l.. By the
end of the filling (March 1963) velocities, substantially
constant, increased to 1.2 cm/day. The draw-down to
665 m a.s.l. in four months leads the displacements rate
to zero and the designers were convinced that the con-
trol of the landslide was effectively possible.
THIRD FILLING AND DRAW-DOWN OF THE
RESERVOIR
The reservoir level was increased to 711 m a.s.l.
in two months (April-May 1963), the velocities never
exceeding 0.3 cm/day. and The displacements rate
reached 0.4 cm/day for a reservoir level at 717 m a.s.l.
a water level in the reservoir maintained below 700 m
a.s.l. (Prof. A. Ghetti 3
rd
July 1962 report).
As it may be noted, until approximately the Spring
1959 the concern of people involved on the dam con-
struction was mainly, if not only in the stability and char-
acteristics of rocks of the dam abutments, while the sta-
bility of the whole reservoir slopes was not considered
with exception of the Erto area. Studies on the whole
reservoir area began when the construction of the dam
body was already accomplished (1959), driven also by
the 4
th
November 1960 landslide occurred with about
700.000 m
3
on the left side of the valley. The water level
in the reservoir was at 650 m a.s.l. following the first fill-
ing operations initiated on February 1960.
Piezometers and topographic marks were planned
and installed on the left side slope to monitor groundwa-
ter level and displacements. Besides, a seismometer was
installed on the dam and operated from May 1960. The
main purpose, behind the limited instrumentation at that
time available, was to possibly relate the reservoir level
with the vertical and horizontal displacements. In gener-
al, data on horizontal displacements seem to suggest that
the slide was moving as a rigid body. It should be noted
that piezometers didn't reach the sliding surface and they
were perforated pipes providing, thus, only information
on the average water pressures crossed by the tube.
Geological knowledge of all the people involved
in the dam planning and construction from 1960 to the
Fig. 16 - Relationships between precipitation, horizontal
rate of movement and levels of piezometers and
reservoir. (after h
endron
& P
atton
, 1985)
background image
R. GENEVOIS & P.R. TECCA
26
International Conference on Vajont - 1963-2013 - Thoughts and analyses after 50 years since the catastrophic landslide Padua, Italy - 8-10 October 2013
(June1963) and 0.5 cm/day when the level reached 720
m a.s.l. (mid-July 1963). During the following month
the level was maintained (mid-August), but the veloci-
ties increased to 0.8 cm/day, further increasing in early
September to as much as 3.5 cm/day, when water level
reached 725 m a.s.l.. In order to control creep velocities,
the reservoir level was slowly lowered reaching 715 m
a.s.l. by 9
th
October 1963 , but the velocities continuous-
ly increase, and rates of up to 20 cm/day were registered.
In short, some general considerations might be
drawn from the observation of data gathered by the
installed instrumentation.
During the period 1960-1963, 43 extensional and
compressional events were registered, ascribed to a re-
gional stress field characterized by N–S compression
(C
aloi
& s
Padea
, 1966) . The seismic activity regis-
tered on the Vajont valley on 23
rd
, 29
th
April, 2
nd
May
1962 and 2
nd
September 1963 is attributed to natural
phenomena: seismic shakes have a too much deep hy-
pocenter to be due to the presence of the reservoir (P.
C
aloi
& s
Padea
, 15
th
June 1962).
In general, the water levels in the piezometers fol-
low rather closely the levels of the reservoir with the
only exception of Piezometer n. 2, during the initial
part of the recording period when it showed water
pressure corresponding to a water column of 90 m
above the reservoir level. It is also possible to note
how increasing the water level the sliding velocity
increased, but this relationship is not linear tending
towards an asymptotic limit, indicating the failure.
The observation that the second reservoir filling leads
to a different asymptotic value for the reservoir wa-
ter level, may indicate that the movements along the
shear surface are not governed by the effective normal
stresses acting on it. This consideration, perhaps, in-
duced to the third filling operation when rate of move-
ments continuously increases and the lowering of the
reservoir level in the attempt to reduce the velocity of
the slide did not work.
From the above observations the Vajont event is
to be considered a slide reactivation with the sliding
surface located in discontinuous layer of plastic clays:
thus, the effective friction angle should not be larger
than 10-14º. Furthermore, no direct information of the
position of the failure surface and on the characteristics
of this material is available.
In order to estimate the evolution of the landslide
a sort of conceptual method has been applied based on
displacements rates and reservoir levels: a reduction of
the reservoir level has to follow too much high displace-
ments rates, accepting that a unique slide was possible
but generating a water wave whose height was mod-
eled by carried out experiments. However, reservoir
levels do not necessarily provide true values of existing
pore pressures on the failure surface and the conceptual
model was not based on a mechanical analysis.
However, considering the difficulties that continue
to be present still today to handle large landslides, we
must reconsider what Carlo Semenza, the designer of
the dam, wrote in a letter dated April 1961 : "... things
are probably bigger than us and there are no adequate
practical measures ... I am facing something that, due
to its dimensions, seems to escape from our hands
…" .
At 22:38 GMT on 9
th
October 1963 the catastroph-
ic landslide occurred.
Albert Einstein
Learn from yesterday,
live for today, hope for tomorrow.
The important thing is not to stop questioning.
STUDIES AND RESEARCHES AFTER
THE FAILURE
After the tragedy, many studies and researches
were carried out on different aspects of the phenome-
non, in order to explain both the kinematics and dynam-
ics of the landslide and its effects. Information reported
in related papers greatly increased the understanding of
such phenomena, so strongly reducing the risk condi-
tions for the populations in mountainous areas.
In order to re-evaluate the currently existing in-
formation on the slide, an electronic bibliographic da-
tabase has been developed, giving to researchers the
availability of a great part of the technical reports and
scientific articles, published or not (s
uPeRChi
et alii,
2010). For simplicity, only a part of these papers is con-
sidered here and main conclusions are shortly reported.
Fig. 17 - A panoramic view of the Vajont reservoir before
(a) and immediately after (b) the landslide failure
background image
THE VAJONT LANDSLIDE: STATE-OF-THE-ART
Italian Journal of Engineering Geology and Environment - Book Series (6) www.ijege.uniroma1.it © 2013 Sapienza Università
Editrice
27
the landslide is evaluated in about 100 s and the maxi-
mum velocity in about 16 m/s. The causes of the land-
slide are indicated in the stratigraphical (overlapping
of deformable strata on more rigid ones and presence
of clayey thin strata) and tectonic (strata dip and fold-
ing of deformable strata) setting, besides the morphol-
ogy of the valley consequent to the glacial and fluvial
erosion. Triggering causes are indicated mainly in the
effects of groundwater level variations, due to rainfall
and reservoir filling-drawdown cycles. However, the
sudden change in the velocities of the landslide cannot
be explained.
The dynamics of the water wave induced by the
landslide has been firstly analyzed in the “Studio
sull’onda di piena seguita alla frana del Vajont” (Study
on the water wave induced by the Vajont landslide) by
Prof. M. Viparelli and Prof. G. Merla, charged by the
Government Board of Inquiry (1964). The water wave,
with a maximum height of 200 m, is due to the 48 Mm
3
of rocks entered in the reservoir, while the volume of
the water overtopping the dam is estimated in 25 Mm
3
.
The water wave, after the dam overtopping, took 4 min-
utes to reach the Piave river at a distance of 1.400 m
and the discharge at the dam site is estimated in 50-
100.000 m
3
/s: lower mean values are obtained for the
whole event (30.000 m
3
/s) and much larger values for
the instantaneous discharge (1 Mm
3
/s). The water wave
at the entrance in the Piave river valley is calculated
100 m high; this wave reaches at 8:00 o’clock of the
day after a site 84 km distant, as high as 2.33 m.
The geological and mechanical factors and the hy-
draulic effects of the landslide have been investigated
by a Commission composed by lawyer M. Frattini and
Professors F. Arredi, A. Boni, C. Fassò and F. Scarsella,
charged by the Board of Inquiry of the National Organi-
zation for the Electrical Energy (ENEL) on 1964.
Fundamentally, the members of this Commission
conclude that the landslide, exceptional for volume
and velocity, is the consequence of the nature of the in-
volved rocks, their “chair” structure and previous ero-
sional processes but, above all, of the effective stresses
decrease due to the saturation caused by the reservoir
infilling. The observed displacements are due to creep
phenomena and a rather good correlation is found be-
tween horizontal displacements and reservoir levels
while correlations with rainfall and seismic activity
are questionable. Maximum displacements, North di-
rected, are initially located between the dam and the
Papers, written just after the catastrophic landslide,
are mainly represented by the technical reports written
by experts charged by public organizations to investi-
gate factors and causes of the landslide.
Immediately after the disaster, Prof. A. Moretti and
Prof. A. Valdinucci, members of the Italian Geological
Service, were charged by the government to investigate
on: the possibility of future extension of the landslide;
the risk conditions of the villages of Erto, Casso and
Cimolais; the future behavior of the dam stressed by the
landslide-induced water wave (“Relazione sulla frana del
Vajont - Nota preliminare” - Preliminary report on the
landslide of the Vajont basin, 27
th
November 196 -). After
a summary of the geological and geomorphological set-
ting of the area, the authors conclude on the unfavorable
geological, tectonic, morphological and hydrogeological
setting of the area, underlining the wrong interpretation
and forecast of the landslide evolution. Considering the
stresses acting on the dam by the landslide body, the hy-
pothesis of the demolition of the higher part of the dam
is also contemplated.
The stability of Casso and Erto villages is nega-
tively considered and their transfer recommended: for
Casso, in consequence of the possible re-activation of
an ancient rock fall whose equilibrium was perturbed
by the 1963 landslide, and for Erto, due to the possibil-
ity of a new huge landslide.
The geology of the Vajont valley is the object of
the “Sutdio geologico della frana del Vajont” (Geo-
logical Study of the Vajont Vallet) by Prof. R. Selli and
Prof. L. Trevisan for the Government Board of Inquiry
(1964) . The description of the area and of the events
is quite accurate. Oucropping formations, consisting of
cherty limestones (Malm) and marly limestones with
clay interbeds (Cretaceous), are folded and faulted
by the tectonic sliding over the Dogger limestones of
the bedrock. The landslide volume is estimated in 200
Mm3 and the total displacement in 450-500 m. Two
different rock masses are individuated separated by the
Massalezza creek, the western one being characterized
by glaciers-like movements. The slip surface is set in
the Malm strata, while in the eastern part it is set in
the stratigraphical higher Cretaceous strata. The whole
mass moved in two different steps: initially it slipped on
a deep surface until it arrived on the opposite slope; af-
terwards, it slipped on surfaces progressively less deep,
but always characterized by the presence of "plastic"
terms in the stratigraphical series. The running time of
background image
R. GENEVOIS & P.R. TECCA
28
International Conference on Vajont - 1963-2013 - Thoughts and analyses after 50 years since the catastrophic landslide Padua, Italy - 8-10 October 2013
Massalezza creek but, after June1963, they are found
just above the dam. The displacement could be inter-
preted also as simply preparatory to a series of small
landslides and not to an instability involving the whole
slope. The measured phreatic groundwater level var-
ies with rainfall until the reservoir level reaches the
groundwater level, from here on it strictly follows
the reservoir level. The landslide, with an estimated
volume of 250 Mm
3
, developed contemporary in the
eastern and western parts in a total time of 10’30”, run-
ning up on the opposite valley side up to 125 m with
a rather small rotation. Triggering factors are indicated
in: i) the low friction resistance due to the presence of
clayey interbeds, to previous movements (presence of
cataclasites and milonites) or to the effective stresses
decrease; ii) the seismic activity in the neighborhood.
The water wave induced by the landslide had a volume
of 48 Mm
3
with a maximum elevation of 200 m on the
previous reservoir level; 30 Mm
3
is the water volume
that overpassed the dam. The results of the experiments
carried out on the 1:200 scale model by the University
of Padova are considered hydraulically reliable and the
differences with what really happened is to be attrib-
uted to the great differences in involved volume and
movement rate. Finally, the phenomenon, considered
exceptional for volume and velocity, is the reactivation
of the ancient landslide.
A first summary of both the characteristics of the
area and the events associated with the landslide, to-
gether with some interpretations of the occurred land-
slide mainly derived from the technical reports of the
different boards of inquiry, is edited by A. De Nardi on
an Italian journal (1964). The first scientific paper pub-
lished on an international journal, describing the 1963
Vajont landslide and related events, is that of Müller
(1964). After a detailed and thorough description of the
studies carried out and of the phenomena observed dur-
ing the different phases of the Vajont reservoir history,
he observes: “… The peak velocities increased progres-
sively during the early days of October. According to
the report of the “Commissione di Inchiesta” the veloc-
ity had reached 20 cm per day by October 9. Compared
with the final velocity of the sliding mass (about 25 m/
sec), all movement, even the last phase, must be con-
sidered a creeping movement up to the very instant of
the slide itself...
”. He concludes that the transition from
the initial long creeping stage of the mass to a true rock
slide was caused by, “the slight excess of driving forc-
es, due to the joint water thrust or to the decrease in re-
sisting forces, resulting from the buoyancy and soften-
ing of clayey substances during higher water level [...]
with a progressive rupture mechanism at the base of the
moved mass”. Besides, Müller attributes to the land-
slide a velocity of 25-30 m/s, consequence of a “spon-
taneous decrease in the interior resistance” and favored
the hypothesis of a new first-time landslide contrasting
his initially agreement with G
iudiCi
& s
emenza
(1960)
on the existence of a prehistoric landslide. In conclu-
sion, he strongly believes in the substantial unpredict-
ability of many aspects of the landslide.
K
ieRsh
s
(1964, 1965) considers the existence of a
prehistoric landslide and the presence of a weak zone
of highly fractured rocks due to the effects of the last
glacial period. He concludes that the collapse was trig-
gered by a rise of the groundwater level with increased
hydrostatic uplift and swelling pressures. His hypoth-
esis and sections were assumed valid in many subse-
quent studies on the Vajont landslide.
A comprehensive work on the geological charac-
teristics and the hydraulic and seismic phenomena that
accompanied the event is written by s
elli
et alii (1964)
on an Italian journal. The authors support the hypothe-
sis that the mass moved with a generally pseudo-plastic
behavior and that the main causes have to be ascribed
to the particular geological structure, to the morphology
of the slope and to the variations in the water level in
the reservoir. The maximum velocity of the landslide is
calculated in 17 m/sec.
After the first detailed study of local stratigraphy
by G
iudiCi
& s
emenza
(1960), many other studies were
carried out by different authors, but their basic lithos-
tratigraphy is still considered valid and confirmed by
other authors (e.g.: C
aRloni
& m
azzanti
, 1964; s
elli
& t
Revisan
, 1964; F
Rattini
et al., 1964; R
ossi
& s
e
-
menza
(1965); n
onveilleR
, 1967; R
ossi
, 1968; m
aRti
-
nis
, 1978; h
endRon
& P
atton
, 1985), except for some
particular chronological considerations. In synthesis,
the landslide involved a complex sequence of cherty
and marly limestones dating from middle Jurassic to
upper Cretaceous.
After the G
iudiCi
& s
emenza
work (1960) and the
landslide event, the first detailed geological and geo-
morphological study on the Vajont area is carried out
by C
aRloni
& m
azzanti
(1964). According to these
authors the deep and narrow Vajont gorge was cut by
the river through the deposits of a previous landslide
background image
THE VAJONT LANDSLIDE: STATE-OF-THE-ART
Italian Journal of Engineering Geology and Environment - Book Series (6) www.ijege.uniroma1.it © 2013 Sapienza Università
Editrice
29
nearby reservoir.
The last contribute of E. Semenza is represented
by his palinspastic reconstruction of the Vajont slide,
from the postglacial mass movement to the 1963
failure (G
hiRotti
, 2006), based on geological and
geomorphological features that led him to define the
shape and the boundary of the ancient landslide.
A new engineering-geological model for inter-
preting the catastrophic event is proposed by P
aRo
-
nuzzi
(2009) on the basis of a detailed geomechanical
field survey of the sliding surface. The model displays
the presence at the landslide base of a 30-70 m shear
zone constituted by coarse angular gravels with blocks
and clay lenses that, in the opinion of the Author, can
explain the fast groundwater level variations caused
by filling and draw down of the reservoir.
In the following years, a great number of papers
have been published on specific aspects of the cata-
strophic landslide differently interpreting and strongly
debating particular aspects of the landslide event.
They range from geotechnical properties to physical
and rheological behavior and to stability analyses, in
the attempt to understand the role of factors involved
in the landslide triggering and development.
All these aspects are, however, strictly connected
to a first primordial question: was the landslide a first-
time one (e.g.: s
elli
et alii, 1964; s
KemPton
, 1966;
B
Roili
, 1967; m
uelleR
, 1968) or the re-activation of an
and their stratigraphy may be summarized as follows:
1. Flysch (Eocene): alternances of arenites and mud-
stones. 2. Erto Marls (Paleocene): marls and marly
limestones, representing the transition between the
Scaglia Rossa and the Flysch. 3. Scaglia Rossa (Upper
Cretaceous): a monotonous succession of marls and red
marly limestones. 4. Soccher Limestones (Lower-Up-
per Cretaceous): alternances of of limestones, calcaren-
ites and calcirudites with conglomerates and breccias
characterised by a great lateral continuity and visible
in the landslide. 5. Rosso Ammonitico (Kimmeridgian
Titonian): reddish and grey massive micrites. 6. Fonza-
so Fm. (Oxfordian): calcarenites and cherty brown
limestones with claystone interbeds. 7. Vajont Lime-
stone (Dogger): massive calcarenites intercalated with
micrites and intraformational.
Two detailed geological maps of the Vajont land-
slide area, before and after the 9 October 1963 catas-
trophe, are published for the first time in 1981 (R
ossi
& s
emenza
). The main aspect of the geological inter-
pretation is the presence of the rock mass located on
right side of the river (Colle Isolato) representing the
remnant of an ancient landslide already hypothesized
by G
iudiCi
& s
emenza
(1959).
With the aim to perform the numerical modeling of
the landslide, the available geological data have been
completed with a geomechanical survey of the outcrop-
ping rock masses (G
hiRotti
, 1993, 1994).
A list of the main interpretative studies under-
taken on the Vajont landslide is reported on the paper
by s
emenza
& G
hiRotti
(2000), in which particular
attention is paid to a landslides occurred in another
Fig. 18 - Stratigraphy at the Vajont gorge (from h
endron
&
P
atton
, 1985)
Fig. 19 - N-S geological sections before 10/9/1963 (1) and
after 10/9/1963 (2). 1) Quaternary (a); stratified
alluvial gravels (b); 2) Scaglia Rossa Fm.; 3) Soc-
chér Fm. sensu lato: Socchér Fm. sensu stricto
(b), Rosso Ammonitico and Fonzaso Fms. (c); 4)
Calcare del Vajont Fm.; 5) Igne Fm.; 6) Soverzene
Fm.; 7) Dolomia Principale; 8) Faults and over-
thrusts; 9) Failure surfaces of landslide (from S
e
-
menza
& G
hirotti
, 2000)
background image
R. GENEVOIS & P.R. TECCA
30
International Conference on Vajont - 1963-2013 - Thoughts and analyses after 50 years since the catastrophic landslide Padua, Italy - 8-10 October 2013
old one (e.g., h
endRon
& P
atton
,1985; P
asuto
& s
ol
-
dati
, 1991; s
emenza
& G
hiRotti
, 2000; s
emenza
, 2001;
v
aRdoulaKis
, 2002)? This controversial aspect was ini-
tially faced up by Hendron and Patton in their technical
reports (1983, 1985), but still now is the object of stud-
ies and discussions (P
aRonuzzi
& B
olla
, 2012).
It is generally thought that the landslide occurred
at least in part, on the slip surface of an old landslide,
but some different hypotheses exist. For instance, the
reassessment of the morphological and structural evi-
dence have suggested to some authors (m
antovani
&
v
ita
-F
inzi
, 2003) that slip surface coincides with a
normal fault plane juxtaposing Cretaceous limestones
and highly fractured rock mass, consistent in strike
with the regional lineament pattern.
h
endRon
& P
atton
work (1985) may be consid-
ered as the first attempt to analyze the Vajont landslide
from a more complete geological, geomorphological
and geotechnical point of view. Based on the works
of R
ossi
& s
emenza
(1965) and s
emenza
(1965), the
landslide is considered a reactivation of a probably
post-glacial one slid over one or more clay levels, act-
ing both as a continuous impermeable layer and a very
weak level with residual friction angle as low as 5°. In
the northern slope of Mt.Toc, two aquifers have been
detected: the upper one (highly fractured and perme-
able landslide mass) was influenced mainly by reser-
voir’s level, while the lower one (Calcare del Vajont
Fm.) was fed by both rainfall in the hydrogeological
basin and reservoir’s water. This hydrogeological
scheme, supported by a new interpretation of the pi-
ezometric levels, may give rise to high water pressures
in the hypothesis that clay strata separating the two
aquifers were continuous.
One, and perhaps the greatest drawback of the
data gathered during the years preceding the collapse
is the lack of reliable water pressure data throughout
the landslide mass and displacement records along
the failure plane, both essential for correlating slide
movements with reservoir levels. Moreover, a good
estimate of the relationship between stress and per-
meability is necessary when analyzing landslides be-
cause the permeability of fractured rock masses is par-
ticularly sensitive to stress changes at shallow depth
(R
utqvist
& s
tePhansson
, 2003).
h
endRon
& P
atton
(1986) re-construct the hydr-
ogeological model of the Vajont slide area on the basis
of water level data measured from the summer of 1961
until October 1963 inside pipes placed in three open
drill-holes. As a result, the water levels could reflect
some average value of the water pressures and hy-
draulic conductivities of the units encountered. They,
however, commented: “As a result, the water levels
recorded inside the casing could reflect some average
value of the different water pressures and hydraulic
conductivities of the units encountered”
and conclude:
‘‘The piezometric data of the Vajont slope is too little
and questionable; it is not sufficient for drawing up a
reliable hydrogeological model, which is necessary in
these cases to make reasonable assumptions about the
pore water pressures for slope stability analysis’’
.
Probably, the first comprehensive hydrogeologi-
cal study of the Mt. Toc area has been carried out by
B
esio
(1986), that realizes a hydrogeological model of
the area on the basis of the census and characterization
of the existing springs. He recognizes the presence of
four aquifers corresponding to: the quaternary depos-
its; the sequence Scaglia Rossa, Calcare di Soccher
and Rosso Ammonitico, whose impervious bedrock is
the Fonzaso Formation; the Calcari del Vajont, widely
karst as shown by the existing sinkholes, whose im-
pervious bedrock is represented by the marly lime-
stones and marles of the Igne Formations; the Calcari
di Soverzene and the Dolomia Principale, constituting
the main aquifer. The springs are numerous, but only a
few have significant discharges (1-10 l/s).
The problem of the existence and continuity of
clay beds in the calcareous sequence, a fundamental
feature for the location of the slip surface, is a rather
discussed aspect. Examining the borehole logs carried
out after the landslide, B
Roili
(1967) concludes for the
absence of clay beds, a conclusion already reached
by m
ülleR
(1964), who asserted that no clay beds
Fig. 20 - A schematic section through the Vajont Slide
showing the estimated regional groundwater
flow system
background image
THE VAJONT LANDSLIDE: STATE-OF-THE-ART
Italian Journal of Engineering Geology and Environment - Book Series (6) www.ijege.uniroma1.it © 2013 Sapienza Università
Editrice
31
and later discussed by, e.g.,: v
oiGht
& F
aust
, 1982;
v
aRdoulaKis
, 2000 and 2002; v
eveaKis
& v
aRdou
-
laKis
, 2007; G
oRen
& a
haRonov
, 2007.
The grassroots idea is that frictional heat, due
to the mechanical energy dissipation inside the slip
zone, and the consequent vaporization of pore water
may lead, if the surface of failure is deep enough, to
a strong increase in pore water pressure that produce
in turn a cushion of very reduced frictional resistance.
A first approximation gives the possibility to calculate
the critical displacement necessary to create vapor in
the slide zone, while a second approximation gives the
relation between the critical displacement and the rate
of shear displacement (h
aBiB
, 1975).
First in a rather long succession, C
iaBatti
(1964)
proposes a pore water pressure rise due to frictional
heating and, considering a variable friction coefficient,
estimates a maximum velocity of 17 m/s.
v
oiGht
& F
aust
(1982), starting analysis from the
model of C
iaBatti
(1964), propose a thermal mecha-
nism as a possible explanation for the low kinetic fric-
tion mobilized by the Vajont landslide. On this basis
they calculate acceleration, velocity (maximum: 26
m/s) and elapsed time as functions of displacements.
The same mechanism is proposed by n
onveilleR
(1978, 1987) who estimates, however, a maximum
velocity of 15 m/s. s
emenza
& m
elidoRo
(1992), con-
sidering the frictional heat developed in the final ac-
celerated phase, try to explain both the high velocity
and the run-out of the landslide, but they conclude that
this mechanism, even if it can really induce very high
velocities due to the shear strength decrease, could not
be effective just from the triggering of the movement,
but only after rather long times.
R
eCently
v
aRdoulaKis
(2002), assuming that
creep is localized in a clay-rich water-saturated layer,
demonstrates that frictional heating can even trig-
ger an explosive pressurization phase after the long-
term phase of accelerating creep. The final total loss
of strength is explained by the onset of thermal pres-
surization, on turn triggered by the temperature rise
within the clay layers. The pore pressure increase, also
enhanced by elevated friction coefficient, porosity and
deformability, can convert the slow sliding into a cata-
strophic failure. With this approach, the author calcu-
lates a velocity of 20 m/s just 8 s after its activation,
corresponding to a slide displacement of only 74 m.
Results of simulations obtained using a similar
exist on the slip surface, if not some films of pelitic
materials, a few mm thick, seldom observed in the
limestone bedding planes; they, however, could not
have played any significant role in the slope failure.
Afterwards, the same m
ülleR
(1968), re-analyzing
available data reaches different conclusions stressing,
instead, the relevance of the “chair-like” shape of the
slip surface. At present days, it is generally accepted
that failure occurred along thin (5–15 cm thick) clayey
(Ca-montmorillonite, smectite, illite, vermiculite) beds
intercalated to the limestones strata and rather continu-
ous over large areas that acted also as an impervious
layer favoring the rising of the groundwater level and,
then, the consequent decrease of shear strength values.
A significant loss of strength is required to explain
the high velocity of the landslide, but many doubts re-
main as regards the mechanisms controlling the rate
of movement and its sudden acceleration. Various in-
terpretations have been given and they mainly differ
in treating the event as a first-time landslide or as the
reactivation of an old prehistoric one (m
enCl
,1966).
As classical analyses for Vajont slide fails to
explain the sudden collapse and the final velocities
achieved, several mechanisms responsible for the fric-
tional evolution of the Vajont slide have been investi-
gated using different models and formulating different
assumptions in order to ascertain the evolution of the
friction coefficient with both time and deformation
(e.g., v
aRdoulaKis
, 2002; K
ilBuRn
& P
etley
, 2003;
h
elmstetteR
et al., 2004; s
oRnette
et alii, 2004; v
e
-
veaKis
et alii, 2007; P
inyol
& a
lonso
, 2010).
The reduction of frictional resistance is related by
m
ülleR
(1968) to creep phenomena and progressive
failure mechanism, noting that the friction angle value,
required in conditions of limit equilibrium, was too
small if compared to the strength properties attribut-
able to the material directly involved in the movement.
Several authors tried to explain the relevant loss
of strength in terms of pore water frictional heating
and thermal pressurization. The thermoplastic soften-
ing behavior of clays has been studied since '70s (e.g.,
h
iCheR
, 1974; d
esPax
, 1976; m
odaRessi
& l
aloui
,
1997) reaching the conclusion that the friction angle at
critical state of some clays may be a decreasing func-
tion of the temperature, depending on clay mineralogy.
The pressurization mechanism, that could explain
the total loss of strength of large slides, is based on the
original "vaporization" concept of h
aBiB
(1967, 1975)
background image
R. GENEVOIS & P.R. TECCA
32
International Conference on Vajont - 1963-2013 - Thoughts and analyses after 50 years since the catastrophic landslide Padua, Italy - 8-10 October 2013
thermo-poro-elastic mechanism at the base of land-
slides (G
oRen
& a
haRonov
, 2007), indicate the de-
velopment of high pore water pressure and reduced
friction resistance, causing large sliding velocities and
run-out. Pore pressure diffusion rates from the shear
zone is mainly controlled by the depth-dependent per-
meability, so that greater landslides are able to main-
tain high pore pressure for longer times, that results in
lower values of the dynamic friction angle.
With the aim to determine appropriate values of
the friction coefficient, t
iKa
& h
utChinson
(1999) car-
ried out ring shear tests on two clay samples from the
slip surface at slow and fast rates of shearing, observ-
ing a relevant loss of strength increasing the shear rate.
While the obtained residual friction angle at slow rate
(9.7°-10.6°) compares rather well with the previously
reported values (8°-11°), at rates greater than 100 mm/
min and after an initial increase above the slow residu-
al value, the dynamic residual value falls to 4.4°. In the
opinion of the Authors, these results may explain the
high acquired velocity of the landslide without invok-
ing other strength loss mechanisms.
Laboratory tests with higher shearing rates have
been carried out by F
eRRi
et alii (2010, 2011), obtain-
ing even smaller values of the friction coefficient. At
room humidity the friction coefficient initially increas-
es from 0.45–0.48 (24.2°-25.6°) at low velocity to 0.68
(34.2°) at 0.04 m/s and, then, falls to 0.09 (5.1°) for slip
rates up to about 1.3 m/s due to the shear heating caus-
ing a temperature increase to about 260°C.
The increase in water content drastically reduces
the shear strength enhancing the velocity weakening
mechanism as already observed by t
iKa
& h
utChinson
(1999) and in agreement with the model proposed by
h
elmstetteR
et alii (2004) and s
oRnette
et alii (2004).
In saturated conditions, presumably existing during the
collapse, obtained results show that thermal and ther-
mo-chemical pressurization are not required to explain
the high slip rates achieved during the final collapse
of the landslide, at least for shear rate lower than 1.31
m/s, that is the maximum velocity investigated.
In order to explain the accelerations preceed-
ing catastrophic landslides, s
oRnette
et alii (2003)
propose a slider-block model, providing a physically
based derivation of the phenomenological model sug-
gested by v
oiGht
(1988). The model gives, for the Va-
jont landslide, good predictions of the critical time-to-
failure up to 20 days before the collapse. h
elmstetteR
et alii (2004) apply to the Vajont landslide the state and
velocity-dependent friction law established and used
to model earthquake friction. They show that observed
displacements can be reproduced with the slider block
friction model, suggesting that the Vajont landslide
belongs to the velocity-weakening unstable regime.
The Authors consider also that friction can be another
possible process able to reproduce the same accelerat-
ing pattern as crack propagation on a basal shear plane
(P
etley
et alii, 2002; K
ilBuRn
& P
etley
, 2003).
The high velocity of the landslide has been attrib-
uted also to other deformation mechanism such a slow
rock cracking process. When clay layers are deformed,
the stresses are concentrated at the tips of existing
micro-cracks that, in the long run, make the cracks to
grow at an accelerating rate and to coalesce in a unique
shear plane or band (K
ilBuRn
& v
oiGht
, 1998). This
process can be also enhanced by the chemical effects
of water, if present (a
tKinson
, 1984). Laboratory stud-
ies (B
uRland
, 1990; P
etley
, 1995) show that brittle
behavior can occur also in undisturbed and water-sat-
urated clays stressed to loads corresponding to depths
more or less coinciding with the depths of the deform-
ing clay layers in Mt Toc.
Slow cracking represents an interesting deforma-
tion mechanism as it can accelerate under constant ap-
plied stress and it is quickly enhanced by circulating
water, chemically attacking molecular bonds at crack
tips (K
ilBuRn
& P
etley
, 2003). The mechanism is ini-
tially dominated by the formation of new cracks that
increase exponentially with time (Main and Meredith,
1991), and, later, by the exponential growth of cracks
length (m
ain
et alii, 1993) until a unique failure plane
is formed (m
C
G
uiRe
& K
ilBuRn
, 1997; K
ilBuRn
&
v
oiGht
, 1998). On the basis of a technique proposed
by s
aito
(1965, 1969), conceptualized by v
oiGht
(1988, 1989) and developed by F
uKozono
(1990), a
new model for the development of progressive failure
in brittle landslides is proposed by K
ilBuRn
& P
etley
(2003). The system is initially driven and controlled by
stress but, once the failure surface reaches the unstable
crack growth stage, it is stress driven while the failure is
strain controlled. The linear inverse-rate velocity trend
with time, obtained by equating the time of failure to
the condition for which rates of deformation become
infinitely large, is representative of the existence of
movements dominated by crack growth, a process that
indicates the approach to catastrophic collapse.
background image
THE VAJONT LANDSLIDE: STATE-OF-THE-ART
Italian Journal of Engineering Geology and Environment - Book Series (6) www.ijege.uniroma1.it © 2013 Sapienza Università
Editrice
33
s
KemPton
, 1966; K
enney
, 1967). In particular, K
enney
(1967) results show that the filling of the reservoir re-
duces the stability by only 5-10%.
l
o
et alii (1971) uses Janbu’s method for noncircu-
lar surfaces, considering a sliding mass formed by two
wedges separated by a vertical discontinuity. The back-
calculated friction angle at limit equilibrium is 13° for
a groundwater level corresponding to the water level in
the reservoir.
Using two-dimensional limit equilibrium, the
friction angle required for stability is back-calculated
by h
endRon
& P
atton
(1985), who obtain values
ranging from 17° to 28°, rather high if compared to
values obtained by laboratory tests that range from 5°
to 16° with an average value of about 12°. These val-
ues are less than those required for stability and, since
the slope was at least marginally stable for a long pe-
riod before the collapse, the Authors concluded that
other factors were not considered. They perform,
then, three-dimensional stability analyses considering
all the available geometrical, geological, geotechnical
and hydraulic data and conclude that the 1963 land-
slide occurred because of the water pressure distribu-
tion on the slip surface consequent to the combined
effects of the reservoir level raise and the piezometric
levels increase resulting from rainfalls; the necessary
resisting force to ensure equilibrium is provided by
the side friction on the Eastern edge of the slide.
C
howdhuRy
(1978), considering the progressive
failure concept, observes that in the landslide mass
exists an upper portion, fundamentally unstable, that
gradually creeps down and thrusts on a lower more
stable portion: forces are progressively increased up
to cause the sudden failure of the lower portion. This
model is consistent with that proposed by J
äeGeR
(1972), in which it is proposed the existence of a non-
uniform weakening zone separating the upper from
the lower sliding mass.
In order to really incorporate kinematics in the
analyses s
itaR
& m
aClauGhlin
(1997) introduced
the DDA (Discontinuous Deformation Analysis)
technique for which the actual mode of failure does
not have to be assumed a priori, and displacements
and velocities are computed as an integral part of the
analysis. Using a simplified cross-section of h
edRon
& P
atton
(1985) and subdividing the landslide mass,
by vertical lines, in a number of blocks from 1 to 105,
they find that the friction angle required for stability
Some researches indicate that clays can behave as a
brittle material under high loads such as those acting on
deep-seated slope failures (e.g.: B
uRland
, 1990; P
et
-
ley
, 1995). However, even if is still rather difficult to
explain the reached velocity in quantitative terms, ob-
servations on slope movements (v
oiGht
, 1988) are con-
sistent with the failure behavior of clays at high pres-
sure: the basal failure mechanism evolves through time
from an essentially ductile process, during the creeping
phase, to a brittle process at collapse phase (P
etley
&
a
llison
, 1997, and P
etley
, 1999).
In the attempt to explain the increasing velocities
of the Vajont lanslide, the possibility of internal rock
strength degradation has been invoked by a
lonso
&
P
inyol
(2010) who state that the Vajont landslide is a
good example of the reduction to minimum values of
strength available along “dormant” sliding surfaces
in high plasticity clays. In their two-blocks model the
cohesion on the separating plane is progressively de-
graded as a function of displacements but, even the full
degradation of the cohesion cannot lead to velocities
higher than 3 m/s, and the Authors conclude that this
process must be accompanied by other mechanisms.
With the aim to predict the response of the Va-
jont landslide, a multi-block sliding model has been
applied by s
tamatoPoulos
& a
neRoussis
(2006):
computed and observed duration of motion and peak
slide velocity, as well as predicted deformation and
back estimated soil strength, reasonably agree with
the measured values.
A large number of stability analyses have been
performed on the Vajont landslide, mainly concentrat-
ing on the evaluation of the friction angle necessary for
stability using limit equilibrium methods and obtaining
values in a wide range, but hardly larger than 12 de-
grees (J
äeGeR
, 1965; n
onveilleR
, 1967; m
enCl
, 1966;
Fig. 21 - Acceleration in slope movement before the cata-
strophic collapse of Mt Toc on 09 October 1963
(from K
ilburn
& P
etley
, 2003)
background image
R. GENEVOIS & P.R. TECCA
34
International Conference on Vajont - 1963-2013 - Thoughts and analyses after 50 years since the catastrophic landslide Padua, Italy - 8-10 October 2013
increases from 7° to 16°, depending on the inter-block
friction and the position of the vertical discontinuity.
They obtain a trend rather similar to that shown by
C
howdhuRy
(1978) using a limit equilibrium method
and modeling the progressive failure.
After that, the DDA method has been used by s
itaR
et alii (2005) to demonstrate that computed velocity are
dependent on the number of used blocks, so indicating
that disintegration of the landslide during failure results
in acceleration of the slide mass. They find, moreover,
that the pore pressure rises, consequent to the frictional
heating, has an effect of the same order of magnitude as
the process of landslide mass disintegration. Consider-
ing both of these processes, they obtain peak velocity
values comparable to that estimated by h
endRon
&
P
atton
(1985), that are 25–30 m/s, suggesting that both
mechanisms might have had a fundamental role.
The problem of velocity determination, and con-
nected energy-lines and Fahrböschung function, was
already faced b
y
e
Rismann
& a
Bele
(2001) who stated
that considering the scientific knowledge at that time
(e.g.: h
eim
, 1932) the Vajont catastrophe, especially as
regards the transition from slow to fast motion, could
have been foreseen.
In the attempt to address the evident inconsistency
between friction angle values in the range of 18º-28º,
as obtained by back-analyses (e.g.: K
enney
, 1967;
h
endRon
& P
atton
, 1985) and those reasonably at-
tributable to the sheared clay beds of the sliding sur-
face (not larger than about 10º-12º), a
lonso
& P
inyol
(2010), analyzing the stability of a simple two-wedge
models, find that the rock shear strength value being
mobilized between the two wedges is practically in
accordance with the expected strength of present Cre-
taceous marls and limestones.
Recently, a new model for the final collapse stage
has been proposed by C
eCinato
et alii (2010) ac-
counting for the pressurization phase due to frictional
heating. Parametric analyses results show that the
thermal friction softening mechanism may be consid-
ered secondary, if compared to mechanisms of static
and dynamic friction softening that mostly affects the
timescale of final collapse. Thermal pressurization
will cause, besides, thicker slides to accelerate faster.
The landslide generated water waves, a particular
type of tsunamis water waves, have been studied in
the past using mathematical theories, physical model
experiments and numerical simulations (e.g., w
atts
et
alii, 2005). The study of these waves is quite complex
as different base models have to be built in order to
describe: (i) the triggering of the landslide, that relies
on quantity and quality of available data and informa-
tion; (ii) the propagation, that needs the knowledge
of the material rheology; iii) the landslide-water in-
teraction, strongly affected by the permeability of the
collapsing material; and (iv) the propagation of waves
along a reservoir, that is usually modeled using depth
integrated models. The water wave generated by the
landslide collapsing at high velocity into reservoir has
been numerically modeled only in the recent years.
P
anizzo
et alii (2005) developed some empirical
formulations to characterize the generated impulse
waves and their application to three real cases (the
Pontisei landslide and the 1960 and 1963 Vajont land-
slides) estimate quite well the values of both the maxi-
mum generated wave height and run-up.
Dealing with an application of the SPH (Smooth
Particle Hydrodynamics), a three-dimensional simula-
tion of the Vajont landslide induced water wave is pre-
sented by R
ouBtsova
&K
ahawita
(2006) technique to
treat free surface problems.
More recently, w
aRd
& d
ay
(2011) simulate the
Vajont landslide and flood disaster applying the new
developed «tsunami ball» method and considering the
Vajont landslide as a semi-coherent slump. The full
sequence of events is captured in three dimensions,
finding that a landslide of that dimension and volume
can effectively splash water up to about 200 m on the
northern slope, push more than 30 million m
3
of water
over the dam and flood the valley below. The authors
stress the consideration that a compelling physical
quantity, emerging from the analysis, is the astounding
rate at which water run away from Vajont reservoir.
The technical literature on Vajont is quite copious,
as a result of the inconsistencies noted in the interpre-
tations of the event. However, a considerable part of
major questions on the characteristics of the landslide,
particularly on the failure mechanism, have not been
adequately explained yet.
Too many contradictory statements and conclu-
sions are present in the literature regarding especially
the existence and the mechanical relevance of signifi-
cant clays or clayey material along the failure surface.
Besides, the not definitively resolved dispute if the
1963 slide was a new slide or the reactivation of a pre-
historic slide has some obscure aspect yet, even if the
background image
THE VAJONT LANDSLIDE: STATE-OF-THE-ART
Italian Journal of Engineering Geology and Environment - Book Series (6) www.ijege.uniroma1.it © 2013 Sapienza Università
Editrice
35
damental. The presence of unknown geological
structures must be verified, continuously updat-
ing models with the acquisition of new data. 3-D
modelling should be upgraded, allowing for a
better representation of three-dimensional spatial
and kinematic effects. In that way, the models will
more accurately represent the real conditions, es-
pecially as regards the frequent complexity of slid-
ing surfaces.
2) The assessment of shear strength of weak fissured
clay rocks is a subject of paramount relevance in
lito-stratigraphical complex sequences. New labo-
ratory experiments and back-analyses of already
occurred landslide should be carried out in order
to assess more accurately change in rock proper-
ties that could be applied in a more continuous
fashion in the predicted model. The brittle behav-
ior of large landslides, including the relationship
between displacement velocity, mechanics, and
shear strength on the failure surface has to be in-
vestigated in detail, using also new techniques and
apparatus.
3) The pre-failure deformation trend in large landslides
must be more accurately observed and analyzed.
In the case of Vajont landslide, in facts, the early
knowledge of a complete set of experimental data
(mainly, slip surface geometry, water pressure
distribution on it and material type and character-
istics) would have been essential to build a more
reliable model of the slide. The only observation
of the decrease of slide velocity with the reservoir
level reduction resulted to be tragically mislead-
ing: it provided a reservoir filling criterion which
however led, at the end, to the landslide collapse.
4) There is a need of a better understanding of the
basic fatigue and progressive failure processes,
to be properly modeled studying the develop-
ment of deep-seated rockslide. Past sliding activ-
ity causes the reduction of the available strength
down to the residual values or less, depending on
the acquired velocity and associated thermal ef-
fects. The possibility that a landslide reaches high
velocities depends on the amount and duration of
the acceleration, in turn depending on the amount
and duration of the disequilibrium between re-
sistant and destabilizing forces and on the actions
applied to the landslide body. In these conditions,
the reduction of strength to residual conditions
deep trenches on the northern slope of M.t Toc seem
to corroborate the Semenza’ hypothesis (1965) of the
presence on an ancient landslide.
The Vajont disaster demonstrated the critical rel-
evance of geological features within and surrounding
the reservoir sites, so that a growing interest in under-
standing and predicting such catastrophic phenomena
after that widely developed. Moreover, the large scien-
tific literature on this landslide and the obtained results,
frequently contrasting, demonstrates that we are only
at the very beginning in the development of concepts
and tools in modeling and predicting these type of
events. On the other hands, the issues emerging from
the Vajont disaster have induced to a changing empha-
sis in researches on catastrophic rock slope failures
that points to a more integration between geoscience
and engineering so a better understanding of landslide
events, triggers, and processes can be obtained.
I know one thing, that (I know nothing).
ἓν οἶδα ὅτι οὐδὲν οἶδα
Socrates
FURTHER RESEARCH AND DEVELOP-
MENT
The catastrophic Vajont landslide demonstrates
the overwhelming need to integrate geology, geomor-
phology and geotechnical engineering analyses in the
study of slope instability. The Vajont event promoted
a large mass of studies and researches that greatly en-
hanced our capacity to understand such phenomena,
especially as regard their precursory activity, causes
of failure, dynamic behavior and triggering and depo-
sition mechanisms. Failure mechanism of large land-
slides is rather complex and difficult to evaluate if all
involved significant factors are not carefully identified,
understood and considered. The Vajont rockslides has
its own geology and characteristics, but similar large
rock masses can be found characterized mainly by un-
favorable structural configurations, rock weakening by
mechanical processes, long creeping behavior and trig-
gering mechanism. Results of past and current studies
and researches on the Vajont event will improve our
capacity to understand and at least partly forecast the
behavior and development of large rockslides.
In the opinion of the authors, however, some areas
still need to be developed.
1) Geological, hydrogeological and engineering geo-
logical investigations should be considered fun-
background image
R. GENEVOIS & P.R. TECCA
36
International Conference on Vajont - 1963-2013 - Thoughts and analyses after 50 years since the catastrophic landslide Padua, Italy - 8-10 October 2013
may be particularly evident in fractured rocks,
where stability is assured mainly by the presence
of rock bridges within discontinuities. Crack
propagation can lead to the rock bridge failure,
resulting into catastrophic failures occurring,
what's more, with very few warning signs. Both
empirical and numerical methods, existing for
the assessment of landslide velocity and propa-
gation, and data gathered by monitoring systems
should be improved. Monitoring data have been
used to predict the time of failure and to assess
the velocity trend. This approach implies that
the mechanism of movement and the boundary
conditions remain unchanged until the failure, a
condition that has to be checked in the study of
different cases. A further problem is represented
by the fact that available monitoring data are
related to specific conditions, that will not nec-
essarily be the same for a future stability crisis,
even of much larger magnitude. Improvements in
this type of analysis will only derive from a deep
understanding of the geological characteristics
and the mechanical behavior of the rock masses.
The last unresolved question is: could the Vajont
landslide have been really avoided? That is, a detailed
geology and a more complete knowledge of the strength
properties of the sliding surface would have given, at
that time, the possibility for a better conceptual and me-
chanical model of the slide that, in turn, would have pro-
vided a reliable criterion to stop the motion? It should
be stressed, however, that even today, to deal with very
large landslide is a formidable task. We are able to ob-
tain field data, such as water pressures, surface and deep
displacement absolute values and trend, rocks and rock
masses characteristics by laboratory experiments and
“insitu” tests, but chiefly in the first more or less 200
of meters. Greater depths require not only sophisticated
and not easily available instrumentation, but their in-
terpretation is often difficult because the need of vast
site investigation, not always feasible, time-consuming
efforts for collecting them and frequent uncertainties
related to the possible number of installed instruments.
The difficulties to handle large landslides con-
tinue to be still present and the words of Eng. Carlo
Semenza, the dam designer (his April 1961 letter),
must considered a vivid testimony of the formidable
challenge he was at the time facing and we have to
face nowadays:
“[…] things are probably bigger than us and
there are no adequate practical measures […] I am in
front of a thing which due to its dimensions seems to
escape from our hands
[…]
ACKNOWLEDGEMENTS
The authors wish to thank the organizers of the Va-
jont 2013 Conference for inviting us to prepare this pa-
per and all people, colleagues, students and friends, for
the scientific and moral support in carried out researches
REFERENCES
a
lonso
e.e., P
in
y
ol
n . m . & P
uzRin
a.m. (2010) - Geomechanics of Failures. Advanced Topics. Springer, Dordrecht., 27 pp.
a
ndeRson
d.l. (1980) - An earthquake induced heat mechanism to explain the loss of strength of large rock and earth slides. Proc.
International Conference on Engineering for Protection From Natural Disasters, 569-580, John Wiley, New York.
a
ndeRson
o.l. (1985) - Calculation of slide velocities. In: h
endRon
a.J. & P
atton
F.D. The Vajont Slide, a geotechnical analysis based
on new geologic observations of the failure surface”, 2, Appendix E and F, Technical Report GL-85-5, U.S. Army Eng. Waterways
Experiment Station, Vicksburg, MS.
a
tKinson
B.K. (1984) - Subcritical crack growth in geological materials. J. Geophys. Res., 89: 4077-4114.
B
elloni
l.G. & s
teFani
R. (1987) - The Vajont slide instrumentation - Past experience and the modern approach. In: l
eonaRds
G.A.
(
ed
). Dam Failures. Eng. Geol., 24 (1-4): 445- 474.
B
istaCChi
a., m
assiRoni
m., s
uPeRChi
l., z
oRzi
l., F
RanCese
R., G
ioRGi
m., G
enevois
R. & C
histolini
F. (2013) - A 3D geological model
of the 1963 Vajont Landslide. (this volume).
B
oyeR
R.a. (1913) - Etude géologique des environs de Longarone (Alpes Vénitiennes). Bull. Soc. Geol. France, 13: 451-485.
B
Roili
l. (1967) - New knoledge on the geomorphology of the Vajont Slide slip surfaces. Rock Mechanics & Eng. Geol., Jour. Int. Soc.
Rock Mechanics., V(1): 38-88.
B
uRland
J.B. (1990) - On the compressibility and shear strength of natural clays. Geotechnique, 40: 329-378.
C
aloi
P. (1966) - L’evento del Vajont nei suoi aspetti geodinamici. Ist. Naz. Geofisica, Roma. In Italian.
C
aRloni
G.C. & m
azzanti
R. (1964a) - Rilevamento geologico della frana del Vajont. Giornale di Geologia, XXXII (I): 105-138.
background image
THE VAJONT LANDSLIDE: STATE-OF-THE-ART
Italian Journal of Engineering Geology and Environment - Book Series (6) www.ijege.uniroma1.it © 2013 Sapienza Università
Editrice
37
C
aRloni
G.C. & m
azzanti
R. (1964b) - Aspetti geomorfologici della frana del Vajont. Riv. Geogr. It., 71 (3): 201-231.
C
howdhuRy
R. (1978) - Analysis of the Vajont Slide - NewApproach. Rock Mechanics, 11: 29-38.
C
iaBatti
m. (1964) - La dinamica della frana del Vajont.Giornale di Geologia, XXXII (I): 139-154.
C
oRByn
J.a. (1982) - Failure of a partially submerged rock slope with particular references to the Vajont Rock Slide. Int. J. Rock Mech.
Min. Sci. & Geomech. Abstr., 19: 99-102.
d
al
P
iaz
G. (1928) - Relazione di massima su due sezioni del Vajont prese in considerazione per progetti di sbarramento idraulico.
Unpublished report for S.A.D.E., 1-4, Venezia, Italy. In Italian.
d
e
n
aRdi
a. (1964) - Il bacino del Vajont e la frana del M. Toc. L’Universo, 21-70.
d
esPax
d. (1976). Etude de l’influence de la température sur les propriétés des argiles saturées. Thèse de doctorat, Université de
Grenoble.
F
eRRi
F., d
i
t
oRo
G., h
iRose
t. & s
himamoto
t. (2010) - Evidence of thermal pressurization in high-velocity friction experiments on
smectite-rich gouges. Terra Nova, 1-6, Blackwell Publishing Ltd.
F
eRRi
F., d
i
t
oRo
G., h
iRose
t., h
an
R., n
oda
h., s
himamoto
t., q
uaResimin
m. & d
e
R
ossi
n. (2011) - Low- to highvelocity frictional
properties of the clay-rich gouges from the slipping zone of the 1963 Vajont slide, northern Italy. Journal of Geophysical Research,
116: B09208.
F
Rattini
m., a
RRedi
F., B
oni
a., F
asso
C. & s
CaRsella
F. (1964) - Relazione sulle cause che hanno determinato la frana nel serbatoio
del Vajont (9 Ottobre 1963). Frattini Commission Report prepared for ENEL. Roma, 92 pp. In Italian.
F
uKuzono
t. (1990) - Recent studies on time prediction of slope failure. Landslides News, 4: 9-12.
e
Rismann
t. h. & a
Bele
G. (2001) - Dynamics of Rockslidesand Rockfalls. Springer-Verlag, Berlin, 316 pp.
G
enevois
R. & G
hiRotti
m. (2005) - The 1963 Vajont landslide. Giornale di Geologia Applicata, 1: 41-53.
G
hiRotti
m (1993) - Nuovi dati sulla frana del Vajont e modellazione numerica. Geol. Rom., 30: 207-215.
G
hiRotti
m. (2006) - Edoardo Semenza: the importance of geological and geomorphological factors for the identification ofthe ancient
Vajont landslide. In: e
vans
s.G., s
CaRasCia
m
uGnozza
G., s
tRom
a. & h
eRmans
R.l. (e
ds
.). Massive Rock Slope Failure. NATO
Series IV, 49, Springer, Dordrecht, 662 pp.
G
iudiCi
F. & s
emenza
e. (1960) - Studio geologico del serbatoio del Vajoint. Unpublished report for S.A.D.E., Part A: 21 pp, text, Part B:
68 photos with discussions, 42 pp, Venezia, Italy. In Italian.
G
oRen
l., & a
haRonov
e. (2007) - Long runout landslides: The role of frictional heating and hydraulic diffusivity. Geophy. Res. Letters,
34. L07301, 1-7.
h
aBiB
P. (1975) - Production of gaseous pore pressure during rock slides. Rock Mechanics, 7: 193-197.
h
eim
a. (1932) - Bergsturz und Menschenleben. Zürich.
h
elmstetteR
a., s
oRnette
d., G
Rasso
J.-R., a
ndeRsen
J. v., G
luzman
s. & P
isaRenKo
v. (2004) - Slider block friction model for
landslides: Application to Vajont and La Clapière landslides, J. Geophys. Res., 109. B0240: 1-15.
h
endRon
a.J. & P
atton
F.d. (1983) - The Vajont slide, a geotechnical analysis based on new geologic observations of the failure
surface. Technical Report GL-83-3, U.S. Army Corps of Engineers Waterways Experiment Station, Vicksburg, MS.
h
endRon
a.J. & P
atton
F.d. (1985) - The Vajont slide, a geotechnical analysis based on new geologic observations of the failure surface.
Technical Report GL-85-5, Department of the Army, U.S. Army Corps of Engineers, Washington, DC.
h
endRon
a.J. & P
atton
F.d. (1987) - The Vajont slide. A geotechnical analysis based on new geologic observations of the f h
iCheR
P. Y.
(1974). Etude des propriétés mecaniques des argiles à l’aide d’essais triaxiaux, influence de la vitesse et de la temperature. Report
of the Soil Mechanics Laboratory. Paris: Ecole Central de Paris.ailure surface. Engineering Geology, 24: 475-491.
h
utChinson
J.n. (1987) - Mechanisms producing large displacements in landslides on pre-existing shears. Proc. 1
st
Sino-British
Geological Conference, Taipei, Mem. Geol. Surv. China, 9: 1805-1841.
J
äeGeR
C. (1965a) - The Vajont rockslide, Part 1. Water Power, 3: 110-111.
J
äeGeR
C. (1965b) - The Vajont rockslide, Part 2. Water Power, 4: 142-144.
J
äeGeR
C. (1972) - The Vajont Slide. Ch. 14, 340-361. In: Rock Mechanics and Engineering. Univ. Press, Cambridge, 417 pp.
K
enney
t.C. (1967) - Stability of the Vajont Valley slope, discussion of paper by L. Müller (1964) on the rock slide in the Vajont Valley.
Rock Mech. And Engr. Geol. Jour. Int. Soc. Rock Mechanics, 5 (5): 10-16.
K
ieRsCh
G.a. (1964) - Vajont Reservoir Disaster. Civil Engineering, 34 (3): 32-39.
K
ieRsCh
G.a. (1965) - Vajont Reservoir Disaster. Geotimes, May-June, 9-12.
background image
R. GENEVOIS & P.R. TECCA
38
International Conference on Vajont - 1963-2013 - Thoughts and analyses after 50 years since the catastrophic landslide Padua, Italy - 8-10 October 2013
K
ilBuRn
C.R.J. & P
etley
d.n. (2003) - Forecasting giant, catastrophic slope collapse: lessons from Vajont, Northern Italy.
Geomorphology, 54 (1-2), 21-32.
K
ilBuRn
C.R.J. & v
oiGht
B. (1998) – SLOW rock fracture aseruption precursor at Soufriere Hills volcano. Montserrat. Geophys. Res.
Lett., 25: 3665-3668.
l
eonaRds
G.a. (1987) - Dam failures. Special Issue. Eng. Geol., 24 (1-4), 1-577.
l
o
K.y., l
ee
C.F. & G
elinas
P. (1972) – Alternative Interpretation of the Vajont Slide. In: C
oRdinG
E.J. (e
d
). Stability of Rock Slopes.
Proc. 13
th
Symp. on Rock Mechanics, University of Illinois, Urbana, 1971, ASCE, New York, 595-623.
m
antovani
F. & v
ita
-F
inzi
C. (2003) - Neotectonics of the Vajont dam site. Geomorphology, 54 (1-2), 33-37.
m
aRtinis
B. (1979) - Contributo alla stratigrafia dei dintorni di Erto-Casso (Pordenone) ed alla conoscenza delle caratteristiche
strutturali e meccaniche della frana del Vajont. Memorie Sc. Geologiche, Univ. Padova, 32: 1-33.
m
enCl
v. (1966) - Mechanics of landslides with non-circular slip surfaces with special reference to the Vajont slide. Géotechnique,
XVI(4): 329-337.
m
odaRessi
, h. & l
aloui
, l. (1997) - A thermo-viscoplastic constitutive model for clays. Int. J. Numer. Anal. Methods Geomech. 21:
313–335.
m
ülleR
l. (1961) - Talsperre Vajont – 15° Baugeologischer Bericht: Die Felsgleitung im Bereich Toc. Unpublished Technical Report
for SADE, Venezia, Italy. In German.
m
ülleR
l (1964) - The Rock Slide in the Vajont Valley. Rock Mech. Eng. Geol., 2: 148-212.
m
ülleR
l. (1968) - New Considerations on the Vajont Slide. Rock Mech. Eng. Geol., 6 (1-2): 1-91.
m
ülleR
l. (1987 a) - The Vajont catastrophe - A personal review. In: l
eonaRds
G.A. (
ed
). Dam Failures. Eng. Geol., 24 (1-4): 423-444.
m
ülleR
l. (1987 b) - The Vajont slide. In: l
eonaRds
G.A. (
ed
). Dam Failures. Eng. Geol., 24 (1-4): 513-523.
n
onveilleR
e
. (1967) - Shear strength of bedded and jointed rock as determined from the Zalesina and Vajont slides. In: On shear strength
properties of natural soils and rocks. Proc. Geotechnical Conference, Oslo, 1967, 289-294.
n
onveilleR
e. (1978) - Untersuchungen zur Stabilität von Böschungen. Freiberg. Forschungsh., A 617 (1): 57-71.
n
onveilleR
e. (1987) - The Vajont reservoir slope failure. In: l
eonaRds
G.A. (
ed
). Dam failures. Eng. Geol., 24(1-4): 493-512.
n
onveilleR
e. (1992) - Vajont Slide - Influence of frictional heat on slip velocity. In: s
emenza
e. & m
elidoRo
G. (e
ds
) (1992) -
Proceedings of the meeting on the 1963 Vajont landslide. IAEG Italian Section, University of Ferrara, Ferrara 1986, 1: 187-197.
P
anizzo
a., d
e
G
iRolamo
P., d
i
R
isio
m., m
aistRi
a. & P
etaCCia
a. (2005) - Great landslide events in Italian artificial reservoirs. Natural
Hazards and Earth System Sciences, 5: 733-740.
P
aRonuzzi
P. & B
olla
a. (2012) - The prehistoric Vajont rockslide: An updated geological model. Geomorphology, 169–170: 165-1914.
P
astoR
m., h
eRReRos
i., F
eRnandez
m
eRodo
J.a., m
iRa
P., h
addad
B., q
ueCedo
m., G
onzales
e., a
lvaRez
-C
edRon
C. & d
RemPetiC
v.
(2008) - Modelling of fast catastrophic landslides and impulse waves induced by them in fjords, lakes and reservoir. Engineering
Geology, 1-11.
P
asuto
m. & s
oldati
a. (1991) - Some cases of deep-seated gravitational deformations in the area of Cortina d’Ampezzo (Dolomites).
Proceedings of European Short Course on Applied Geomorphology, 2, Univ. of Modena, Modena, 91-104.
P
etley
d.n. (1995) - The deformation of mudrocks. PhD thesis, University of London.
P
etley
d.n. (1996) - The mechanics and landforms of deep-seatedlandslides. In: B
RooKs
s. & a
ndeRson
m. (e
ds
.). Advances in
hillslope processes. Wiley, Chichester, 823-835.
P
etley
d.n. (1999) - Failure envelopes of mudrocks at high effective stresses. In: a
Plin
a.C., F
leet
a.J. & m
aCquaKeR
J.h.s. (e
ds
.),
Physical properties of muds and mudstones. Spec. Publ. Geol. Soc. Lond., 158. Geol.Soc. London, London, 61-71.
P
etJey
d.n. (2004) - The evolution of slope failure: mechanisms of rupture propagation. Natural Hazards and Earth System Sciences,
4 : 147-152.
P
etley
d.n.& a
llison
R.J. (1997) - The mechanics of deep-seated landslides. Earth Surf. Proc. Landf., 22: 747-758.
P
etley
d.n., B
ulmeR
n.h. & m
uRPhy
w. (2002) - Patterns of movement in rotational and translational landslides. Geology, 30 (8):
719-722.
P
inyol
n.m. & a
lonso
e.e. (2010) - Criteria for rapid sliding II. Thermo-hydro-mechanical and scale effects in Vajont case. Engineering
Geology, 114: 211-227.
R
iva
m., B
esio
m., m
asetti
d., R
oCCati
F., s
aPiGni
m. & s
emenza
e. (1990) - La geologia delle valli Vajont e Gallina (Dolomiti
orientali). Annali Univ. Ferrara, Sez. Sc. della Terra, 2 (4): 55-76. In Italian.
background image
THE VAJONT LANDSLIDE: STATE-OF-THE-ART
Italian Journal of Engineering Geology and Environment - Book Series (6) www.ijege.uniroma1.it © 2013 Sapienza Università
Editrice
39
R
ossi
d. & s
emenza
e. (1965) - Carte geologiche del versante settentrionale del M. Toc e zone limitrofe, prima e dopo il fenomeno di
scivolamento del 9 ottobre 1963, Scala 1:5000. Ist. Geologia Università di Ferrara, 2 Maps.
R
ouBtsova
v. & K
ahavita
R. (2006) - The SPH technique applied to free surface flows. ScienceDirect, Computers & Fluids, 1359-1371.
R
utqvist
J & s
tePhansson
o. (2003) - The role of hydromechanical coupling in fractured rock engineering. Hydrogeology Journal, 11:
7-40.
s
adReKaRimi
a. & o
lson
P.e. (2007) - Review of the October 9, 1963 failure of the Vajont reservoir slope. Computer Applications in
Geotechnical Engineering, Geo-Denver 2007, 1-10.
s
CheideGGeR
a.e. (1973) - On the prediction of the reach and velocity of catastrophic landslides. Rock Mechanics, 5: 231-236.
s
elli
R., t
Revisan
l., C
aRloni
C.G., m
azzanti
R. & C
iaBatti
m. (1964) - La Frana del Vajont. Giornale di Geologia, serie 20, XXXII
(I): 1-154.
s
elli
R. & t
Revisan
l. (1964) - Caratteri e interpretazione della Frana del Vajont. Giornale di Geologia, serie 20, XXXII (I): 8-104.
s
emenza
E. (1965) - Sintesi degli studi geologici sulla frana del Vajont dal 1959 al 1964. Mem. Museo Tridentino Sc. Nat., 16:1-52. In
Italian.
s
emenza
e. (2000) - La storia del Vajont raccontata dal geologo che ha scoperto la frana. Tecomproject Ed., Ferrara, Italy. In Italian.
s
emenza
e. & G
hiRotti
m. (2000) - History of 1963 Vajont Slide. The importance of the geological factors to recognize the ancient
landslide. Bull. Eng. Geol. Env., 59: 87-97.
s
emenza
e. & m
elidoRo
G. (e
ds
) (1992) - Proceedings of the meeting on the 1963 Vajont landslide. IAEG Italian Section, University of
Ferrara, Ferrara 1986, 1: 1-218.
s
itaR
n. & m
aClauGhlin
m.m. (1997) - Kinematics and discontinuous deformation analysis of landslide movement. II Panam. Symp.
on Landslide, Rio de Janeiro. Invited keynote lecture.
s
KemPton
a.w. (1966) - Bedding-plane slip, residual strength and the Vajont Landslide. Géotechnique, XVI (1), 82-84.
s
oRnette
d., h
elmstetteR
a., a
ndeRsen
J.v., G
luzman
s., G
Rasso
J.R. & P
isaRenKo
, v. (2004) - Towards landslide predictions: two case
studies. Physica, A 338: 605-632.
s
tamatoPoulos
C. & a
neRouissis
s. (2006) - Back analysis of the Vajont slide using a multi-block sliding model. In: Monitoring,
Simulation, Prevention and Remediation of Dense and Debris Flows. WIT Transaction, 90: 261-271.
s
uPeRChi
l. (2012) - The Vajont Rockslide: New Techniques and Traditional Methods to Re-evaluate the Catastrophic Event. Ph.D thesis,
Universita degli Studi di Padova, Italy, 187 pp.
s
uPeRChi
l., F
loRis
m., G
hiRotti
m., G
enevois
R., J
aBoyedoFF
m. & s
tead
d. (2010) - Implementation of a geodatabase of published
and unpublished data on the catastrophic Vajont landslide. Natural Hazards and Earth System Sciences, 10: 865-873.
t
iKa
t
h
.e. & h
utChinson
J.n. (1999) - Ring shear tests on soil from the Vajont landslide slip surface. Geotechnique, 49 (1): 59-74.
t
RolloPe
d.h. (1980) - The Vajont Slope Failure. Rock Mechanics, 13: 71-88.
v
aRdoulaKis
i. (2002) - Dynamic thermo-poro-mechanical analysis of catastrophic landslides. Geotechnique, 52 (3): 157-171.
v
eveaKis
e.,v
aRdoulaKis
i. & d
i
t
oRo
G. (2007) - Thermo-poro-mechanics of creeping landslides: the 1963 Vajont slide, Northern Italy.
Journal of Geoephysical Research.
v
iPaRelli
m. & m
eRla
G. - L’onda di piena seguita alla frana del Vajont. Internal report of the Engineering Faculty, Hydraulics Institutes,
University of Naples (Italy).
v
oiGht
B. (1978) - Rockslides and Avalanches: 1. Natural Phenomena. Elsevier, Amsterdam.
v
oiGht
B. (e
d
) (1988) - A method for prediction of volcanic eruptions. Nature, 332: 125-130.
v
oiGht
B. & F
aust
C. (1982) - Frictional heat and strength loss in same rapid slides. Geotechnique, 32 (1): 43-54.
w
aRd
s.n. & d
ay
s. (2011) – The 1963 landslide and flood at Vajont reservoir Italy. A tsunami ball simulation. Ital.J.Geosci., 130: 16-26.
w
atts
P. (1997) - Water waves generated by underwater landslides, Ph.D. thesis, California Inst. of Tech., Pasadena, CA, 1997.
w
atts
P., G
Rilli
s.t., K
iRBy
J.t., F
RyeR
G.J. & t
aPPin
d.R. (2005) - Landslide tsunami case studies using a Boussinesq model and a fully
nonlinear tsunami generation model. Nat. Hazards Earth Syst. Sci., 3: 391-402.
w
olteR
a., s
tead
d. & C
laGue
J.J. (2013) - A engineering geomorphologic characterisation of the 1963 Vajont Slide, Italy, using long-
range terrestrial photogrammetry. Geomorphology (submitted)
z
aetta
C. (2009) - Disastro del Vajont: conseguenze a lungo termine sulla salute psichica e fisica dei sopravvissuti. Ph.D thesis,
University of Padova, Italy, 107 pp.
z
anFRon
B. (1998) - Vajont, 9 ottobre 1963. Cronaca di una catastrofe. Ed. Agenzia fotografica Zanfron, 160 pp. Belluno.
background image
Statistics