Document Actions

ijege-13_bs-bonaria-tosatti.pdf

background image
Italian Journal of Engineering Geology and Environment - Book Series (6) www.ijege.uniroma1.it © 2013 Sapienza Università
Editrice
383
DOI: 10.4408/IJEGE.2013-06.B-36
13
TH
AUGUST 1935: A CATASTROPHIC DAM FAILURE
IN THE ORBA VALLEY (PIEDMONT, ITALY)
V
ittorio
BONARIA
(*)
& G
ioVanni
TOSATTI
(**)
(*)
Engineering Geologist - Studio Tecnico Associato BIERREVI - Via San Siro 12/2 - 16124 Genoa (Italy)
(**)
Università di Modena e Reggio Emilia - Dipartimento Scienze Chimiche e Geologiche - Largo S. Eufemia 19 -
41121 Modena (Italy)
ever considering the proper safety of the plant. The
Orba Valley disaster, which occurred nearly 30 years
before the Vaiont catastrophe, shows disturbing analo-
gies with the latter, which will be discussed later. The
dam which remained standing, named “Main Dam of
Bric Zerbino”, still lies within a remote, abandoned
meander of the Torrent Orba which, after the 1935
collapse, found a new course in correspondence with
the rock sill where the failed dam had stood. The Orba
and Vaiont disasters represent two unfortunate exam-
ples of the serious consequences which can be brought
about by underestimating or neglecting the geological
and environmental setting of sites where important en-
gineering works have to be built.
K
ey
words
: Orba Valley disaster, dam, hydroelectric plant,
Italy
FOREWORD
The Torrent Orba is a north-bound stream running
through a small portion of Liguria and southern Pied-
mont (northern Italy) as far as its confluence with the
River Bormida, near the town of Alessandria. In 1926
the construction of a large hydroelectric plant was com-
pleted in the municipality of Molare. The impoundment
was obtained by barring the Torrent Orba at Ortiglieto
(Fig. 1) by means of two dams: the Main Dam of Bric
Zerbino and the Secondary Dam of Sella Zerbino.
After nearly ten years of operation, on 13
th
August
1935, after a very heavy rainstorm the Secondary Dam
ABSTRACT
Between 1919 and 1925 the “Officine Elettriche
Genovesi” (O.E.G.), a large industrial company, built
a hydroelectric plant in the municipality of Molare
(Piedmont, north-western Italy). The reservoir cre-
ated by the barrage of the Torrent Orba (hydrographic
catchment of the River Po) by means of two dams
had a capacity of 18 million m
3
. On the morning of
13
th
August 1935, after nearly ten years of opera-
tion, the rock sill on which one of the two dams had
been constructed - named Secondary Dam of Sella
Zerbino - collapsed following a heavy rainstorm. The
failure caused the sudden emptying of the reservoir
and a large amount of water poured into the under-
lying valley, wreaking havoc along the whole course
of the Torrent Orba as far as its confluence with the
River Bormida, some 50 km away, near the town of
Alessandria. At least 111 people lost their lives in the
disaster. The criminal trial held by the Turin Courts
in 1938 returned a verdict of acquittal for the manag-
ers of the O.E.G. and the designer of the plant. This
article pinpoints the multiple causes which led to the
disaster and which can be mainly ascribed to the lack
of geological investigations within the hydroelectric
plant area. In addition, thorough hydrological and
hydraulic studies were not carried out, which could
have accurately tested the adequacy of the dam’s
discharge system. Finally, the original project was
changed several times in order to reduce building
costs and increase the volume of water stored without
background image
V. BONARIA & G. TOSATTI
384
International Conference Vajont 1963-2013. Thoughts and analyses after 50 years since the catastrophic landslide Padua, Italy - 8-10 October 2013
were available for the implementation of the project
and the works were limited to the construction of a
service road and the levelling of the ground for the
first hydraulic structures.
Finally, in 1916 a new concession was granted
to a company named “Officine Elettriche Genovesi”
(O.E.G.) which was presided over by Mr Zunini since
1918. The O.E.G. was located in Genoa and was con-
trolled by “Edison”, the largest Italian energy company.
Eventually, from 1922 onward construction pro-
ceeded rather quickly after changes where made to
the original project by another engineer, Mr Vittorio
Gianfranceschi, who decided to increase the height
of the Main Dam. Therefore, the Torrent Orba was
barred at Ortiglieto by a 47 m high slightly curved
gravity dam (whereas the maximum height originally
planned for the dam was 34 m) equipped with four
water dischargers. The plant was completed in 1926
and named Main Dam of Bric Zerbino (Fig. 2). This
dam could provide water storage of 18 million m
3
with a 24,000 metric horsepower hydraulic force.
This decision, though, posed a serious problem:
at one point of the perimeter of the planned impound-
ment, some 300 m west of the Main Dam, a saddle
formed by two ridges would have been at a lower
elevation with respect to the maximum storage level.
As a consequence, water could have overflowed the
saddle and poured out into the underlying riverbed.
Therefore, it was held necessary to construct a
secondary barrage, made up of a 110 m long and 14
m high summit wall. This barrage was planned and
built rather hastily, without the support of adequate
geological investigations since, according to the
planners, this saddle “was made up of sound rock”.
The second dam, named Secondary Dam of Sella
collapsed together with a portion of the ground on which
it was founded. A huge wave of water and debris swept
down the valley causing the loss of at least 111 lives and
serious damage to the villages located downstream.
In 1938, at the end of the criminal trial held
against the planners, owners and managers of this
plant, all the defendants were exculpated from any
responsibility for this disaster.
Today, in the Orba Valley the Main Dam of Bric
Zerbino is still standing, although the stream does not
flow any longer in its original riverbed. What is left of
this plant is now located within a protected wetland
surrounded by wood-covered slopes. Like the dams
of Vaiont and Gleno, the dam is now a monument to
the thoughtlessness and lack of accountability of man.
HISTORY OF THE PROJECT
The history of the hydroelectric plant of the Orba
Valley (Piedmont, northern Italy) dates back to the
end of the 19
th
century when a project concerning the
storage and exploitation of the water of the Torrent
Orba was presented by an engineer, Mr Luigi Zunini
(Z
unini
, 1899), to the prefectures of Genoa and Ales-
sandria with the purpose of supplying the Genoa-
Ovada-Alessandria railway with electric power.
In the following years, notwithstanding the
strong opposition of a group of municipalities from
the Orba Valley, the project was enlarged from the
original 8 million m
3
to over 16 million m
3
. This con-
siderable increase would be obtained with the con-
struction of a 40 m high gravity dam.
In 1912 a concession for the construction of a
dam was granted to a society owned by Mr Zunini
himself, who presented the first project in 1914.
During the years of the First World War no funds
Fig. 1 - Map of northern Italy: arrows show locations
of Orba, Gleno and Vaiont dams
Fig. 2 - The Main Dam of Bric Zerbino showing the
twelve dam siphons (O.E.G. photo, 1925)
background image
13
TH
AUGUST 1935: A CATASTROPHIC DAM FAILURE IN THE ORBA VALLEY (PIEDMONT, ITALY)
Italian Journal of Engineering Geology and Environment - Book Series (6) www.ijege.uniroma1.it © 2013 Sapienza Università
Editrice
385
changes. The dam’s plaster shows several network-
like fissures… Nevertheless, the inspections have con-
firmed that the two dams are in normal condition
”.
In November 1926, a considerable high-water
event in the Torrent Orba catchment increased the res-
ervoir surface by nearly 70 cm above the maximum
filling level, with a real overflow risk. Following this
event O.E.G. decided to lower the position of the
floodway to an elevation of 319 m a.s.l. with respect to
the originally planned 322 m a.s.l. Eight years later, in
November 1934, a sudden rainstorm rapidly increased
the level of the impoundment which was satisfactorily
controlled by the newly placed floodway.
GEOMORPHOLOGIC AND GEOLOGIC
SETTING
The upper part of the Orba Valley is character-
ised by the typical features of the north-western Ap-
ennines and Maritime Alps, such as steep slopes and
narrow riverbed, often with an incised-meander pat-
tern. North of the village of Molare, these landforms
give way to the more gentle hills of the Monferrato
district, where the riverbed is wider and follows a
less winding route. Finally, north of Ovada, the
stream flows into the open Po Plain.
This morphological differentiation is due to the
Zerbino (Fig. 3), was a massive wall of Portland ce-
ment on top of which ran the road leading to the ad-
jacent Main Dam.
The reservoir resulting from the two barrages
stretched upstream for some 5 km with an irregular
shape and a maximum width of some 400 m. The
basin’s storage, at the maximum level of 322 m a.s.l.,
was 18 million m
3
. The hydroelectric plant fed by
this reservoir was built 3 km downstream of the two
dams (Fig. 4).
In 1924 the Commission for checking large dams
(named “Commissione Gleno”, since it was estab-
lished after the 1923 hydraulic disaster in the Scalve
Valley
11
gave its approval for filling the reservoir up
to an elevation of 317.40 m a.s.l., corresponding to
a volume of some 13.5 million m
3
of water. Once the
basin was filled, water leaks (up to 30 l/s) were no-
ticed across the rock diaphragm of Sella Zerbino on
which the Secondary Dam stood. Several attempts
were made to make the rock mass impervious with
no satisfactory results. Paradoxically, these leaks were
not recorded during the final inspection carried out by
the Civil Engineers Board in December 1927.
Between 1925 and 1935, several interventions of
ordinary maintenance were carried out on the Second-
ary Dam and the underlying saddle of Sella Zerbino.
In the reports drawn up by the engineers in charge in
1928 and 1929 various problems concerning the dyke
of the Secondary Dam were pointed out. In particular,
in April 1929 it was written that: “The glass fissurem-
eter fixed across the crack on the right hand side of
the Secondary Dam is broken owing to temperature
1
On 1
st
December 1923 a concrete dam collapsed in the Scalve Val-
ley near Gleno (upper Lombardy). In half an hour some 6 million m
3
of water, mud and debris poured out of the artificial basin and swept
down the valley as far as Lake Iseo. At least 356 persons lost their lives
Fig. 3 - The reservoir and Secondary Dam of Sella
Zerbino (photo from the mid-1920s)
Fig. 4 - Map showing plan view of the reservoir with
the position of the two dams (reduced from Top-
ographic Map of Italy, 1:25,000 scale, 1934)
background image
V. BONARIA & G. TOSATTI
386
International Conference Vajont 1963-2013. Thoughts and analyses after 50 years since the catastrophic landslide Padua, Italy - 8-10 October 2013
Zerbino, the tectonic boundary between the antig-
orite serpentinites, which make up the backbone of
Bric Zerbino (right hydrographic side), and the meta-
basites cropping out on the opposite side of the valley
is well exposed (C
apponi
et alii, 1988). Furthermore,
the whole area is characterised by intense cataclasis
due to the presence of numerous fault systems (V
iGo
,
1998) (Fig. 5).
THE DISASTER OF 13
TH
AUGUST 1935
After a long lasting drought, early in the morning
of 13
th
August 1935 a very heavy rainstorm hit the up-
per Orba Valley. The main stream and all its tributaries
were swollen and there was flooding in many places
(B
onaria
et alii, 2005).
The reservoir level had been very low for several
weeks but at 10:00 hrs it had risen by about 8 m, reach-
ing an elevation of 318 m a.s.l. (V
isentini
, 1936). The
plant caretaker started an emergency manoeuvre by
opening the bottom discharge valves, which stopped
working after a few minutes. At this point, the reser-
voir could only discharge by means of the surface spill-
way and the twelve dam siphons (Fig. 2). Neverthe-
less, by 12:30 hrs the basin was full and water started
to flood over the top of the two dams. At 13:15 hrs
the rock saddle of Sella Zerbino, where the Secondary
Dam had been built, gave way causing the collapse of
the overlying dam. Within a few minutes the level of
the reservoir went down by some 25 m while a wave
of more than 20 million m
3
of water poured down into
the valley sweeping away the hydroelectric plant (Fig.
6) which had just been abandoned by technical staff.
A huge wave of water, mud and debris wreaked havoc
diverse geological environments (pre-Tertiary, Ter-
tiary and Quaternary) which characterise the Orba
Valley. The stream which gives its name to the val-
ley originates from the ridges forming the Ligurian-
Adriatic watershed ascribable to the rock formations
of the Voltri Group. This term, which was introduced
late in the 19
th
century (i
ssel
, 1892), designates one of
the largest metaophiolite and metasediment complexes
cropping out in the SW Alps. These rocks were origi-
nally portions of oceanic crust, with their overlying
sediments, belonging to the Jurassic Ligurian-Pied-
mont Ocean. Following Alpine orogenesis, these rocks
were subject to metamorphism, displaced and over-
thrust onto the margin of the European paleocontinent.
The place named Ortiglieto, where the two dams
were built, is characterised by the presence of rock
types ascribable to the Voltri Group such as mostly bro-
ken and displaced serpentinites, serpentine schists and
metabasites affected by numerous joint systems (C
hie
-
sa
et alii, 1975) which influence the slope morphology
and, in particular, the hydrographic network pattern.
Typically, there are E-W and ENE-WSW oriented sub-
vertical faults which sometimes turn to a N-S direction
in correspondence with the main watercourse.
In 1925, the technical review “L’Energia Elettrica”
emphasized the technological magnificence of the new
plant, though it provided a rather biased description of
the local geology: “The most important feature of the
rocks belonging to the Voltri Group is the absence of
deep joints. The layers’ surfaces are soundly cemented
to the bedrock; therefore there seem to be no cavities at
depth
” (O.E.G., 1925).
Today, at the point of the dam breach of Sella
Fig. 5 - Geological section across the Orba Valley reservoir showing the mylonite shear zones in correspondence with the
Secondary Dam (modified after C
apponi
et alii, 1988)
background image
13
TH
AUGUST 1935: A CATASTROPHIC DAM FAILURE IN THE ORBA VALLEY (PIEDMONT, ITALY)
Italian Journal of Engineering Geology and Environment - Book Series (6) www.ijege.uniroma1.it © 2013 Sapienza Università
Editrice
387
disaster had been caused by an unforeseeable event
such as exceptional precipitation: over 100 lives lost
just because of a heavy rainstorm…
CAUSES OF THE DISASTER
Unfortunately, the Orba Valley catastrophic dam
failure was not the only hydraulic disaster that struck
Italy in the 20
th
century:
1) In the morning of 1
st
December 1923, a dam bar-
ring the Torrent Povo in the Scalve Valley (Lombardy)
collapsed after a few days of heavy rainfall. The huge
wave thus generated swept the Scalve and Camonica
valleys as far as Lake Iseo. Along its route of about 10
km, it destroyed entire villages, scattered houses and
electric power stations taking at least 356 lives.
2) Late in the evening of 9
th
October 1963, a large
portion of the northern slope of Mt. Toc (Friuli) slid
into the hydroelectric reservoir of the Vaiont dam,
which withstood the violent impact of the landslide.
Nevertheless, a huge wave resulting from the dis-
placed water hit two villages on the opposite side of
the valley, overflowed the top of the dam and struck
the bottom of the River Piave valley with incredible
violence, destroying most of the village of Longarone
(Veneto). The ascertained lives lost were 1917.
These events, together with the Orba Valley disas-
ter (111 lives lost), make up the three great hydroelec-
tric plant catastrophes which first marked the develop-
ment and then the decline of the Italian hydroelectric
industry in the 20
th
century.
As in the Orba Valley Disaster, several factors
contributed to the failure of the Sella Zerbino rock sill
and the collapse of the overlying dam. They can be
summarised as follows: i) geological causes; ii) hy-
draulic causes; iii) planning causes.
on the valley downstream, bringing death and destruc-
tion to the villages of Molare, Ovada and several other
hamlets (Fig. 7). There were 111 ascertained victims,
half of whom were from Ovada. 90 houses and 4
bridges were destroyed and the farming activities of
the mid-lower Orba Valley severely compromised. The
total damage caused by this disaster was estimated at
over 45 million lire of the time.
The investigations carried out by the members of
a Ministry Commission set up in the aftermath of the
disaster established that a spate wave ran across the
River Po itself for eight days after the failure, recorded
even by the farthermost hydrometers.
The Sella Zerbino disaster shows striking analo-
gies also with the collapse of the Malpasset dam (Var
District, France) of 1959, in which 423 persons lost
their lives (L
uino
&T
reBo
, 2010).
On 6
th
December 1923, a few days after the Scalve
Valley disaster, Mr. Luigi Mangiagalli, an authorita-
tive hydraulic engineer, declared that “Obviously, if
there had not been construction faults, there would
not have been the Scalve Valley catastrophe. Never-
theless, this incident should not arouse apprehension
or hostility among the people living in valleys where
hydroelectric plants stand, since dams can be built
and managed nowadays with a mathematical level of
safety
” (M
anGiaGalli
, 1937). Fifteen years later, the
same engineer coordinated the various technical re-
ports of the legal consultants of O.E.G., which estab-
lished the absolute unavoidability of the Orba Valley
catastrophic failure.
On May 1938, the Appeal Court of Turin pro-
nounced a verdict of acquittal for all the defendants
in the criminal trial concerning the failure of the Sella
Zerbino dam. Therefore, nobody was responsible for
this terrible event since, according to the verdict, the
Fig. 6 - Destruction of the hydroelectric plant during
the emptying of the penstock (August 1935)
Fig. 7 - Aftermath in the Orba Valley by the village of
Ovada (August 1935)
background image
V. BONARIA & G. TOSATTI
388
International Conference Vajont 1963-2013. Thoughts and analyses after 50 years since the catastrophic landslide Padua, Italy - 8-10 October 2013
GEOLOGICAL CAUSES
The planning history of the Orba Valley hydroe-
lectric plant lasted some 30 years, from the first project
proofs in 1898 up to 1926, when the final executive
project was presented to the Civil Engineers Board.
During this long period the only geological docu-
ment attached to the project was the report by Pro-
fessor Francesco Salmoiraghi, from the Milan Poly-
technic. This geological report consisted of a few
pages in which the geological features of the valley
were summarised (S
almoiraGhi
, 1898). No detailed
geomorphological surveys were carried out, no struc-
tural analyses on the rock masses, let alone boreholes.
Nevertheless, this geologist stated that the rocks found
in the Orba Valley, although belonging to the same
formation (i.e., the Voltri Group), showed extremely
variable characteristics. According to the same author,
this morphological diversity was to be considered a
positive factor, since at one point there was a narrow
gorge suitable for a barrage whereas upstream the val-
ley was much wider and therefore suitable for hosting
a large reservoir. Professor Salmoiraghi concluded his
report by asserting that: “…in any part of this area it
is possible to build a dam in full safety conditions
”.
In 1926, when the final project was presented,
the only geological report in attachment was the
“Salmoiraghi report”, although this geologist had
died in 1910. Indeed, O.E.G. had been long aware of
the real conditions of the rocks making up the sill of
Sella Zerbino.
If in-depth and thorough structural-geological
surveys had been conducted in the Ortiglieto-Bric
Zerbino area, where the reservoir was located, they
certainly would have identified the extremely unfa-
vourable geological characteristics of the outcropping
rocks which eventually led to the dam’s complete
failure. For example, in correspondence with Sella
Zerbino (Fig. 4) accentuated lateral fluvial erosion on
the torrent’s right-hand side had shown the presence
of particularly jointed rocks and easily recognizable
shear zones affecting greenish-blue mylonites (Fig. 5).
Geological investigations should have been corrobo-
rated by boreholes and exploration drills, which were
already available when this plant was built. Mr Ettore
Scimemi, a hydraulic engineer, in his work Dighe
(Dams), published in 1928, stated that: “It would have
been advisable to carry out boreholes and drills on
both sides of the valley in order to avoid the geologi-
cal surprise given by the undetected presence of joints
or thinner rock levels. This system of investigation is
rather simple, although fairly expensive”.
After a century, it should be stressed that these
are still the most important in situ investigations in
order to acquire adequate geological knowledge of an
area chosen for the implementation of large engineer-
ing works such as dams. These investigations should
also be accompanied by geophysical prospecting and
remote sensing imagery interpretation.
On the other hand, as regards analytical assess-
ments of slope stability around the reservoir, it would
have been necessary to wait until the 1970s for the
development of correct approaches by B
ieniawsky
(1973) or B
arton
& C
houBey
(1977) applied to rock
slopes in order to evaluate properly rock mass quality
and joint shear strength.
Years later, another geologist, Professor Mario
Airoldi, wrote a report in which he stated that “…the
identification of jointed and faulted rocks should not
have been difficult even before the construction of the
Secondary Dam since they were cropping out on the
upstream flank of the rock sill, where they were subject
to erosion by the Torrent Orba
” (a
iroldi
, 1935). Fur-
ther evidence on the widespread presence of jointed
rocks was offered when a water load tunnel was dug
at a short distance from the rock sill of Sella Zerbino
before the construction of the Secondary Dam. Un-
expectedly, the excavation works intercepted highly
fractured and water-rich rock levels, corresponding to
mylonite bands, and the tunnel walls and vault had to
be strengthened with reinforced concrete (C
annonero
,
1935). These levels played a major role in the collapse
of the rock sill since they are weaker horizons within
the rocky septum.
Therefore, notwithstanding its modest size, it was
the Secondary Dam which collapsed since it had been
built on poorly compact and intensely jointed rocks,
whereas the still standing Main Dam was founded on
sounder serpentinites.
HYDRAULIC CAUSES
The area occupied by the dam impoundment was
not provided with a pluviometric gauge and, for this
reason, errors and speculations in assessing the pre-
cipitation trend were possible (as indeed occurred dur-
ing the trial). After ten years of activities, O.E.G. did
not consider it necessary to install any pluviometric
background image
13
TH
AUGUST 1935: A CATASTROPHIC DAM FAILURE IN THE ORBA VALLEY (PIEDMONT, ITALY)
Italian Journal of Engineering Geology and Environment - Book Series (6) www.ijege.uniroma1.it © 2013 Sapienza Università
Editrice
389
raised well beyond the height approved by the District
Mining Board of Trent; as a consequence, the sudden
collapse of the Stava unstable earth dams caused the
death of 268 people (T
osatti
, 2003).
Similarly, also the Main Dam of Bric Zerbino in
the Orba Valley was augmented in height by over 1/3
with respect to the initial height (from 34 m to 47 m).
This decision was not followed by adequate adjustment
of the water discharge equipment which was unable to
allow rapid emptying of the reservoir during the heavy
rainstorm which preceded failure. Even more impor-
tant, the final increase of the reservoir level revealed the
geological-structural critical state of the rock sill at Sella
Zerbino which, nevertheless, was not properly assessed.
The implementation of the Secondary Dam as a simple
containment work, rather than a proper dam equipped
with a spillway, which would have guaranteed higher
water discharge, was the direct consequence of the ill-
planned changes to the project (B
onaria
et alii, 2005).
FINAL REMARKS
The incident of 13
th
August 1935 in the Orba Val-
ley preceded the Vaiont catastrophe by nearly 30 years;
nevertheless the two events show striking similarities.
In both cases the predisposing factor is to be found
in the critical geological and hydrogeological condi-
tions of the slopes surrounding the reservoirs rather
than in the actual sites where the dams were built.
As for the Orba Valley, the feasibility of the plant
was assessed on the basis of the most suitable site for
constructing the Main Dam, as happened at Vaiont.
Therefore, it is not a coincidence that the Orba Main
Dam and the Vaiont dam survived these disastrous
events undamaged, since they had been built in mor-
phologically and geologically suitable sites.
Mr Zunini considered the noticeable widening
of the Orba Valley upstream the plant as favourable,
since it could host a large reservoir. Nevertheless, this
sharp morphological change should have warned the
planners of the presence of a complex geological-
structural situation, evident also at the rock sill of
Sella Zerbino.
Unfortunately, both Messrs Zunini and Gian-
franceschi in the Orba Valley and Mr Carlo Semenza
in the Vaiont Valley focused their attention nearly ex-
clusively on the engineering works, neglecting the ge-
ological context of the two sites and the consequences
that a large mass of water could have on the stability
gauge around the area occupied by the reservoir. As a
consequence, the meteorological data available for the
technical advisors during the trial were scanty (V
isen
-
tini
, 1936).
Several pluviometric gauges within the Orba Val-
ley and adjacent catchments recorded for 13
th
August
1935 precipitation values exceeding 300 mm within a
time of less than 8 hrs, corresponding to nearly 30% of
the mean annual precipitation (T
ropeano
, 1989).
The technical commission appointed by the Ital-
ian Ministry of Public Works, which visited the site
of the disaster the following day, calculated for the
Torrent Orba a maximum flow rate of 2280 m
3
/s (≈16
m
3
/s each km
2
). Unfortunately, the actual discharge
capacity of the dam was about 850 m
3
/s (6 m
3
/s each
km
2
). In addition, the two submerged discharge valves
stopped working minutes after they were opened fol-
lowing the rapid level increase of the reservoir. At this
point, the reservoir’s discharge was possible only by
means of the surface spillway and the dam’s siphons:
only 1/3 of the actual flow rate could therefore be dis-
charged by the plant.
Due to this severe hydraulic inadequacy, the res-
ervoir level kept rising up to nearly 2.5 m over the
top of the dam until failure occurred, due to the rapid
erosion of the jointed rocks which made up the foun-
dation ground of the Secondary Dam.
PLANNING CAUSES
The changes introduced in the original project by
Mr Gianfranceschi, which consisted of raising the top
of the Main Dam by 13 m, certainly played a major
role in the tragic failure of the Secondary Dam.
With regard to this aspect, it should be pointed out
that inappropriate and unsuitable changes to hydraulic
and geotechnical structures were aggravation causes
in all the great industrial disasters in Italy. The Gleno
dam was initially planned as a gravity dam but, ow-
ing merely to economic reasons, during construction
it was turned into a multiple-arch structure (P
edersoli
,
1973). In the Vaiont dam project several increases of
the dam’s height were introduced (from the original
200 m up to 264.5 m) which heavily reduced the sta-
bility of the left shore of the reservoir (H
endron
&
P
atton
, 1985; S
emenZa
, 2001). In the Stava Valley the
marshy and ill-suited nature of the soil on which the
two tailings dams were built was completely ignored
(R
ossi
, 1973) and the outer dams of the basins were
background image
V. BONARIA & G. TOSATTI
390
International Conference Vajont 1963-2013. Thoughts and analyses after 50 years since the catastrophic landslide Padua, Italy - 8-10 October 2013
animal species.
After years of silence, in the mid-1970s new pro-
posals for a possible re-activation of the plant started
to circulate (C
alVino
& S
iCCardi
, 1980). These trig-
gered new discussions between environmentalists and
supporters of a new hydroelectric plant (among whom
there were local administrators worried about the re-
current summer droughts and lack of water storage),
although nothing was decided for years. In 2005, the
still standing dam (Fig. 8) was definitely cancelled
from the register of Italian dams, thus putting an end
to a decades-long debate and favouring ecological up-
grading of the area.
Only in 2005, in coincidence with the 70
th
anni-
versary of the Orba Valley disaster did new populari-
sation initiatives take place. The “active memory” of
this tragic event was revived, thanks to local publi-
cations, the website “www.molare.net” and various
commemoration events, such as travelling photo-
graphic exhibitions. In 2010, legal procedures were
activated in order to give the State the ownership of
the surviving dam. This fact could be of great impor-
tance in order to organise guided tours of the site, his-
torical-nature trails for hikers, equipped with explana-
tory panels, and the implementation of a remembrance
museum (like the one set up in the Stava Valley, cf.
G
iordani
et alii, 2003). Only through this sort of ini-
tiatives would it be possible to revive the memory of
one of the three greatest hydroelectric plant disasters
of Italy’s troubled industrial history.
of already precarious rock slopes.
In the case of the Orba Valley this shortcoming
was further aggravated by the total lack of geological
investigations even during construction or when, later
on, problems concerning water infiltration through the
rock sept of Sella Zerbino were not properly assessed.
Furthermore, it should be pinpointed that, in order to
save money, the Secondary Dam was poorly and hast-
ily constructed just on the most critical site: the rock
sill of Sella Zerbino.
Also the discharge system of the Main Dam was
completely inadequate, since no meteorological or hy-
draulic models were elaborated before construction.
Since the collapse of the Secondary Dam, the Tor-
rent Orba has found a new route across the rock sill of
Sella Zerbino whereas the Main Dam is still standing.
In what used to be the reservoir a vast wetland has
developed with the presence of numerous vegetal and
Fig. 8 - The surviving Main Dam of Bric Zerbino as it ap-
pears today (June 2012)
REFERENCES
a
iroldi
m. (1935) - Relazione sulle condizioni geologiche in corrispondenza della Diga Secondaria del bacino di Ortiglieto.
Regia Università di Genova, Genoa.
a
udoly
i. (1939) - Considerazioni sulle cause che hanno provocato il crollo della diga di Sella Zerbino dell’impianto
idroelettrico di Molare in Valle Orba. Genoa.
B
arton
n. & C
houBey
V. (1977) - The shear strength of rock joints in theory and practice. Rock Mechanics, 1 (2). Springer,
Vienna.
B
ieniawski
Z.t. (1973) - Engineering classification of jointed rock masses. Trans S. Afr. Inst. Civ. Engrs, 15: 335-344.
B
onaria
V., e
sposito
F
errando
C. & l
aGuZZi
a. (2005) – 13 agosto 1935, il giorno della diga. Memorie dell’Accademia
Urbense, 65.
C
alVino
F. & s
iCCardi
F. (1980) - Avanprogetto di fattibilità per la trasformazione dell’impianto idroelettrico di Molare (Regione
Piemonte). Relazione tecnica, Genoa.
C
annonero
a. (1935) - Il crollo della diga secondaria del serbatoio di Ortiglieto (Molare). Atab. Lit. Pellas, Ovada.
C
apponi
G
., s
CamBelluri
m. & t
allone
s. (1988) - Introduzione all’evoluzione metamorfico-strutturale del Gruppo di Voltri.
Guida all’escursione sul Gruppo di Voltri. Gruppo Strutturale Italiano, CNR, Genoa.
C
hiesa
s
., C
ortesoGno
l., F
orCella
F. & G
alli
M. (1975) - Assetto strutturale ed interpretazione geodinamica del Gruppo di
Voltri. Boll. Soc. Geol. Ital., 94 (3): 555-582.
background image
13
TH
AUGUST 1935: A CATASTROPHIC DAM FAILURE IN THE ORBA VALLEY (PIEDMONT, ITALY)
Italian Journal of Engineering Geology and Environment - Book Series (6) www.ijege.uniroma1.it © 2013 Sapienza Università
Editrice
391
G
aZZetta
del
p
opolo
(5
th
July 1938) – L’assoluzione dei presunti responsabili dell’inondazione in Valle Orba. Turin.
G
iordani
i., l
uCChi
G., s
alGhetti
d
rioli
G. & t
osatti
G. (2003) - Stava 1985 - una documentazione. Centro di Documentazione
della Fondazione Stava 1985, 96 pp., Curcu & Genovese, Trento.
h
endron
a.J. & p
atton
F.d. (1985) - The Vaiont Slide, a geotechnical analysis based on new geologic observations of the failure
surface. 2 Vols., U.S. Army Corps of Engineers, Washington, D.C.
i
ssel
A. (1892) - Liguria geologica e preistorica. Edizioni Donath, Genoa.
l
uino
F. & t
reBò
G. (2010) - La diga di Malpasset (Francia), a cinquant’anni dal crollo del 2 dicembre 1959, con riferimenti
ad analoghi casi italiani. GEAM, 129 (1): 53-80. Pàtron Ed., Bologna.
m
anGiaGalli
L. (1937) - Sulle cause del crollo della diga di Molare. Relazione tecnica di parte nel processo penale, Milan.
o.e.G. - o
FFiCine
e
lettriChe
G
enoVesi
(1925) - L’Energia Elettrica. Fasc. XII (II), Milan.
p
edersoli
G.S. (1973) - Il disastro del Gleno. Grafica Gutenberg, Bergamo.
r
ossi
G. (1973) - I bacini di decantazione dei rifiuti degli impianti di trattamento dei minerali. Aspetti ecologici e tecno-
economici del problema. Industria Mineraria, 10-11, Serie II, Rome.
s
almoiraGhi
A. (1898) - Relazione sulle condizioni geologiche della Diga Zerbino in Valle d’Orba. Milan.
s
Cimeni
E. (1928) - Dighe. Hoepli, 512 pp., Milan.
s
emenZa
E. (2001) - La storia del Vajont. Ed. Tecomproject, 280 pp., Padua.
t
osatti
G. (
ed
.) (2003) - A review of scientific contributions on the Stava Valley disaster (Eastern Italian Alps), 19
th
July 1985.
GNDCI-CNR Special Volume, 480 pp., Pitagora Ed., Bologna.
t
ropeano
D. (1989) - Eventi alluvionali e frane nel bacino della Bormida, studio retrospettivo. Associazione Mineraria
Subalpina, Quaderni di studi e documentazione, 10, Turin.
V
iGo
E. (1998) - Tettonica a thrust nel settore nord-occidentale del Massiccio di Voltri: analisi strutturale nell’alta Valle del Rio
Meri.
Master Degree Thesis, unpublished, Università degli Studi di Genova, Genoa.
V
isentini
M. (1936) - Écroulement d’un barrage sur l’Orba, en Italie, le 13 Août 1935. Revue de Géographie Alpine, XXIV
(II), Grenoble.
Z
unini
L. (1899) - Progetto di derivazione d’acqua dalla Valle d’Orba per creazione di forza motrice. Relazione progettuale,
Milan.
background image
Statistics