1 Resistivity monitoring for leakage and internal erosion detection at Hallby embankment dam Sjodahl, P.; Dahlin, Torleif; Johansson, S.; Loke, M. H. ...
1 Resistivity monitoring for leakage and internal erosion detection at Hallby embankment dam Sjodahl, P.; Dahlin, Torleif; Johansson, S.; Loke, M. H. ...
1 CONSTRUCTION DEWATERING AT SALUDA DAM Paul C. Rizzo, P.E., Ph.D., Jeffrey M. Bair. P.E., John P. Osterle, P.E., Howard W. Gault, P.G. Saluda Dam, ow...
1 Intro to Model United Nations (MUN) MUN Guide 2017 Contents Preface Introduction Basic Vocabulary Etiquette Points of Action Process of a Conference...
1 MGI MUN 2016 CONFERENCE THE NEWSLETTER2 Welcome Speech of the Secretary General My fellow precious Gandhians, It fills me with great pleasure to be ...
1 MUN Rules and Procedures UCIMUN 20052 Forms of Debate General Debate General debate occurs in the beginning of committee. In this period of the conf...
1 Model United Nations (MUN)2 About MUN Simulation of the United Nations and its different organizations You don t represent yourself/your own views y...
1 EUNMI MUN February, South Pleasant Street #202 Amherst College Amherst, MA Phone: (413) EDUCATION Ph.D., Sociology (2011) Seoul National University ...
1 I. DELEGATE GUIDE MUN2 II. III. IV. NIS MUN Preparing Resolutions The MUN conferences issue resolutions, (official statements of intention) at the e...
1 Reservoir Sedimentation Schleiss et al. (Eds) 2014 Taylor & Francis Group, London, ISBN Reservoir sedimentation management at Gebidem Dam (Switz...
Missouri University of Science and Technology
Scholars' Mine International Conference on Case Histories in Geotechnical Engineering
(1993) - Third International Conference on Case Histories in Geotechnical Engineering
Leakage at Upper Mun Dam Nibondh Saihom Royal Irrigation Department, Thailand
Ruongrit Ammawat Royal Irrigation Department, Thailand
Mondhian Kangsasitiam Royal Irrigation Department, Thailand
Follow this and additional works at: http://scholarsmine.mst.edu/icchge Part of the Geotechnical Engineering Commons Recommended Citation Saihom, Nibondh; Ammawat, Ruongrit; and Kangsasitiam, Mondhian, "Leakage at Upper Mun Dam" (1993). International Conference on Case Histories in Geotechnical Engineering. 33. http://scholarsmine.mst.edu/icchge/3icchge/3icchge-session02/33
This Article - Conference proceedings is brought to you for free and open access by Scholars' Mine. It has been accepted for inclusion in International Conference on Case Histories in Geotechnical Engineering by an authorized administrator of Scholars' Mine. This work is protected by U. S. Copyright Law. Unauthorized use including reproduction for redistribution requires the permission of the copyright holder. For more information, please contact [email protected]
Proceedings: Third International Conference on Case Histories in Geotechnical Engineering, St. Louis, Missouri, June 1-4, 1993, Paper No. 2.07
Leakage at Upper Mun Dam Nibondh Saihom, Ruongrit Ammawat, Mondhian Kangsasitiam and Thanu Harnpattanpanich Royal Irrigation Department, Thailand
Abstract: The Upper Mun Dam located In Nakom Ratchaslma, "!he Nortfleastem Province of Thailand, lntercep'ls a catchment area of 454 sq. kms. It Is a zone earth dam with 32.7 meters of maximum height and 880 meters In length. comprises alluvial deposits overlying sandstone and siltstone.
The founda1!on strata
The chute spillway Is designed for probable maximum flood of 950
m3/sec. and the outlet conduit Is 1.84 by 1.84 meters caters for a maximum outflow of 11.2 m3/sec. No foundation treatment was carried out except cement grouting beneath the spillway and the outlet. The dam was completed on May 1989. The maximum storage for the first year Impoundment was 33.7 million m3 with water level at the eleva1!on 212.32 meters. elevation Is 205.00 meters.
The river bed
In October 1990, the water level rose to nearly full reservoir level at the elevation 220.42 meters with
the storage volume of 131.8 million m3 due to heavy rain fall when concentrated seepage was noticed at two loca1!ons downstream of the toe.
At one of the two locations, the leakage caused a cave-In of the down stream slope of the
embankment which pluged and stopped the leakage.
At the other location, concentrated leak reached Its peak of about
5 m3/sec. and resulted In progressive erosion of embankment toe.
Many countermeasures to save the dam from felling were deployed.
Earth moving equlpmen'ls were used to push rlprap as well as embankment material continuously Into the leakage loca1!ons. The reservoir level was lowered through the outlet and slphonnlng over the spHiway crest. before the leakage was brought down to a few litre/sec. of clear water. reslden'ls had to be evacuated.
The opera1!on was continued for 3 weeks
During this Incident 57 villages Involving over 13.000
After the reservoir was emptied In a few months later, an Intensive Investigation program was
carried out to look Into the cause that lead to "!he leakage of the dam.
INTRODUCTION The Upper Mun Dam or Mun Bon Dam located
1.7 Km. south-west of Khonburl district In Naknon Ratchaslma. nortfleastem province of ThaOand (See location Map In Rg.1). The Project was the consequence of a proposal submitted by the Royal
lrr1ga1!on Department for a scheme to develop the Upper Mun and Its eastern tributary the Lam Sae and entails the construction of two
storage dams and canals system serving about 19,100 hectares of agriculture land.
The feaslbDity study(!) was carried
"'I l ')
finished since 1970. lhe details design was completed In 1984 and the construction work was completed by a local contractor In 1989. During "!he first rainy season, the reservoir was Impounded to elevation 213 m. or 39 million m3 • On October 4, 1990 the reservoir level rose from elevation 210.79 m. with storage volume of 21.8 million m.3 to elevation 220.35 m. with storage volume 130.7 mUBon m3• on October 22, before leakage was observed. FoOowlng the Incident. a technical Investigation committee was formed to Investigate the cause of leakage. The authors were designated by RID to serve on this committee.
views expressed In this paper definitely Influenced by the resuHs observed by "!he committee.
Ag. 1 Location map
However. this paper does not
represent the opinion or the policy of RID.
excava11on works. The upstream slope Is covered with riprap over
2. SITE CONDffiONS AND EMBANKMENT
the opera11ng range of the reservoir where as the downstream The Upper Mun Dam as designed and constructed has the
slope Is covered with grass sodding. An Inclined chimney drain, 3 meters In horizontal width
following features. The dam was founded on alluvium deposHs, overlying rocks of Khorat Group comprising sandstones, siltstones, and mudstones. These rocks are exposed on the hills on either side of the valley on the dam abutments.
consisting of fine filter material, Is provided to assure positive control of the phrea11c line and prevents erosion of material from the Impervious zone of the dam. Since the dam Is founded on alluvial deposits and no
The alluvial deposits, as much as 19 meter
cutoff trench was provided except at both abutments. The
thick, primarily consist of clay, silty sand and sandy slit soils of low
underseepage water through the foundation was controlled by a
plasticity. Flg.2 depicts the geologic section along the center nne
sandwich drainage blanket underlying the downstream shoulder
of the dam. The embankment Is a modified homogeneous earthflll dam with .length of 880 meter and
together with a row of pressure relief wells along the downstream toe of the dam. This sandwich drain consists of 0.5 meter thick of
height of 32.7 meter above
"/ ). ~ >-. -·-c-~;,;-~;-o~~-;;~;;,:-T. .
"'. / ' / . / ./
SANDSTONE ( 8rt1l
POORLY GRADED BAND
LEAN CLAY AND CLAYEY BAND
SANDSTONE , SILTSTONE, CLAYSTOIIE
SANDSTONE ( fino &roy l
fAT CLAY AND CLAYEY SAND
SANDSTONE , CLAYSTQNE AND SHALE
SILTY CLAYEY SAND
Fig. 2 Geologic section along center line of the danF 1 coarse filter material In between flne filter layers of which the top the general foundation level.
The upstream slope Is 3H:1V for the
layer Is 0.5 meter thick and the lower layer Is specified for a
upper 10 meters and flatten to 5H:1V over the lower section. Over
minimum thickness of 0.15 meter.
the river section for a length of about 400 meter, the dam has
Flg.3 shows two typical cross sectlons of the dam.
been extended with an outer slope 15H:1V to form a parHal
dam crest Is at Elv.+230.70 meter.
blanket and stabilized berm on the soft alluvium materials. The
level Is set at Elv.+228.9 m. with a capacity of 350 million m. while
downstream slope Is 2.5H:lV which also flatten In the lower part to
the normal storage level Is at
5H:1V for the length of about 250 meters over the river sectlon of
of 141 million m3. The dam consists of a service spillway situated at the right
the dam where the foundatlon Is soft. The zone upstream of the Inclined chimney drain consists
The probable maximum water
with a capacity
water at the rate 95 m /sec.
of Impervious earthflll. The outer zone of the downstream shoulder
for probable maximum flood and an outlet conduit 1.84 m. by 1.84
consists of a random material obtained from spillway and outlet
m. located at the left abutment with discharge capacity at the
Probable maalmun flood Normal max. pool El.!!e.IO !1.221.00
Excoaa topsoil dlspoaol Min. oporatln~ pool E1.208 .oo
~-~--~~~~~~~~~~~~ Up8froam ata. 2+ 25 to ata. !+85 and eta. 8+ 10 to 8+ 80 OC>Wnotroam Sta.2+25to ato.!+85 and oto.e+es to 8+80
Mtn. operation pool Et. 208.00
Exc- topsoil disposal I.
Upstream ala.!+ 81Ho 8+ tO oo ... dotroam ala. !+65 to 8+65
Oalft oreet roadway loaat oourat
Riprafl llop• protection
Upttrtota 1tr111 fill
Rg. 3 Typical cross section
of the dam
The sto'age volume, thus increased from 21.8 million m3 . to 130.7 million m3.
3. HISTORY OF IMPOUNDING
llrneDnes of reservoir elevations, since the
4. HISTORY OF LEAKAGE
of the dam until Immediately following the failure Incident, Is shown
October 22, a. leakage appeared at the toe of the
In Fig. 4. During 1989 wet season the reservoir has only reached
downstream slope at STA 0+856.
elevation 213 m. As may be observed from October 4 to October 23, 1990
em. In diameter.
reservoir elevation rose rapidly from 210.79 m. to
The hole was approximately 15
The leakage water was clouded wHh sediment.
Rock and aggregate were dumped into the hole In order to stop
220.35 m. as a consequence of two consecutive depression storms.
However, the attempt was not succeeded.
leakage amount Increased and the hole enlarged throughout that night. In the momlng of consecutive day, another leakage larger than the first one was found on the downs1ream slope at STA 0+475. This latter leakage was In the dam body approximately 5 m. above the toe level. An attempt was made without success to stop the leakage throughout the night therefore the outlet gate was opened to release the Impounding water. October
downstream slope at STA 0+475, approximately at elevation 226 m.
The sinkhole was 3.5 by 6 meters and extended deep Into the
OATES CAPA1L '80-HOVEM8Efl '1101
dam body. The collapse of the embankment, however, stopped the leakage at the second location. Concen1ratlon was then given
Rg. 4 Plot of Reservoir Elevation
to the leakage at the first location. Materials were dumped Into
VS. time before incident
the hole as water gushing out off lt. The voh.nne of leaking water
waa 2 to 3 m3/&ee. The effort to prevent further damage to the dam waa going on ffvoughout the night. October 25. at day llglt. a vortex was seen at the upstream of the leakage point. Material from upstream part of the dam was pushed Into the reservoir In order to plug the hole at the vortex. The reservoir level rose to El v + 220.42 m. at 6 p.m. an Increase In storage,deaplte all the emergency discharge effort&. The leakage exceeded 5 m3/&ee.
before decreasing to about 1
m /&ee. late In the evening, as much of the top part of the dam
waa cut and pushed Into the leakage area.
Fig. 5 Circular crack at leakage STA. 0+856
October 26, early In the morning, the quantity of the leakage Increased again and reached lfll peak of
m3/&ee. at around 11 a.m . At 8.30 a.m. a tension crack was at
down3fream slope from STA 0+800 to
below the crest.
Before 9 a.m.
1) Leakage was Initiated through dam embankment.
leakage volume was
constantly con11nued throughout the night.
still high. With a large
from upstream, the
pertaining to the design such as geological
had been reviewed. At
the discharge capacity. of
was slowly eroded away with the leaking water, the rneaauree
October 28 to 31, since an amount
112 m /&ee through the outlet. To further reduce storage, .pipe siphons over the spillway crest were
the resevolr level was
In leakage quantity, the
program In order to help determine the charactel1stla and
against the erosive lealdng water. However • the leakage was the reduction
was Initiated through the foundation or
contact between the foundation and embankment.
than 1 m /&ee . .owing to the successful counter measures taken
Folllowlng the failure. two main hypothesis regordlng
the causes of leakage were developed.
leakage decreased to about 1 m3/sec. October
loud cracking sound occurred for about 15 minutes as the collapsed
flnaUy dam In
5. POST FAILURE INVESTIGATION
aiding earth waa ob3erved (Fig.5) . Shortly before 1 p.m. a embankment
cracka appeared on
The cracks on the slope grew and developed
0 +850, just above the leakage point, approxlrnatly 3 m. the
leakage reduced to less than O.Ql
area, the a
taken to counter the down3fream slope collapse were to fl In
concentrated leak through defect In the embankment and
smal amount of rock, grovel and sol at the toe of slope In
order to lncr9089 the .stabily of the downstream slope and.
6. EVALUATION OF PIEZOMETRIC DATA
lncreoeed to 51 eetJ.
31, the -reeervolr
On the and
amount of leakage was reduced to
about 0.5 m3/&ee. Eventualy. downstream
Reading had been made daly since Apfl
of piezometer~ located In the embankment at STA 0+225 and
November 30, the
In accordance with the tlrnelnea of the responeee
lowering of reservoir
had been lnstaled In the embankment and foundation at
0+700, Elv + 210m since
Fig. 6 Plan of Mun Bon Dam Indicated
210.6 m, an
almost constant response
embankment was absent. llmellnes the
Elv + 206m,
Elv + 196m,
Elv + 198m,
Elv + 201 m, of various chalnage appeared reservoir 1990,
Elv +200 m, lags
Elv + 196m,
Elv + 190, STA
1990), achieved data
beyond similar this
of the dam. Indicated that they were Increase
location of leakage). ( see the
However, relief wells
which were at the
vicinity at STA
0+800 and STA 0+820 also Indicated large underseepage.
Elv + 198,
Indicated the presence of permeable members that
Elv + 201m, that
Implying that at these or
During rapid filling of the reservoir In October
20 m. to STA
piezometer response values that may Indicate piping, was Time
Elv + 200m,
seepage forces In the foundation.
llmeDnes of piezometers located In the foundaflon Elv + 190m,
seepage pressure since Initial Impoundment. at
EVALUATION OF RELIEF WELL DATA
value of about 219 m. seepage
8. BORE HOLES AND GEOPHYSICAL INVESTIGATIONS
Elv + 200m
upstream of the
of the bore
cross sections where the
neighborhood of the two cross secffons.
addition, a near
number of crest
·•· OCIODER IST,I980 ......
OCIODER 111~ 1890
O<:IODER22N0,1890 H· OCIOOER 28TII,ISSO·
Fig. 7 Relief well record data
The least 2.5
meters Into the bedrock material.
Penetration Tests were
performed at every
meter depth. Further classification tests were carried out on the
embankment determined In
the bore holes. Based program
geotechnical profiles were drawn up.
They confirmed ihe
original description of the foundation materials and did not reveal
However many example
near the (see
on thick permeable sand deposits
1991, geo-electrlcal surveys
This ihln clay layer might prevent the underseepage water
the dam. In were
Incidents. The survey an
to an between about
Rg. 8 Geologic Secffon at STA. 0+848
of the resuHs of this geophysical
presence of holes In
STA 0+855 and STA 0+880.
of potenffal holes or cclvHies were suspected to be below
Just after the declaration of safe condHlon of the The reservoir was at EJv +
dam on November 30, 1990. the
Elv +208 m. Also,
have been detected between
The cut was made at STA
0+450 near the existing river channel down to the original
The nature of deposits and sedimentary structure found In the foundation at Mun Bon Dam is truly representative of 'meandering river" sedimentary environment. The deposits found In such an environment are traditionally divided Into two broad categories. namely
channel deposits and alluvial deposits. While
channel deposits display an upward fining sequence. with sand deposits. The alluvial deposits. comprising flood basin and back
swamp deposits. are fine-grained sands, silts,and clays. Offen.
switching or avulsion of the river channel results in the formation of an abandoned river channel. where the deposits are similar to flood basin deposits. However. abandoned channel deposits are distinguished from flood basin deposits by their channel shaped
geometry and by the fact that abruptly pass down Into channel floor lag coame
PU:ZOw::rriR HO.: 0Pl6 STAnON: 8+50
materials. The aforementioned features are
Dlii'TII (m): IM
concurrent with those observed at Mun Bon Dam, as indicated
DATES IN OCTOBER
by the result of post-failure geophysical survey. The results of the geo-electric investigation are further
Fig. 14 Comparison between Observed and FEM
corroborated by results of the standard penetration tests. Profiles
Pressure Head In Foundation
of standard penetration number at boring near the 'meandering river" channel, indicate the presence of loose sediments (N less than 10) to a depth between 5m and 10m. While the presence of loose sediments to a depth of 5m Is indicated In the boring
near the abandoned river channel around STA 0+856, other As the reservoir level rose. between October 4, 1990
boring In the vicinity Indicate the presence of medium dense and
sediments (N greater than 20).
gained access to loose and permeable channel deposits, left
13. MOST PROBABLE MECHANISM IN THE DEVELOPMENT
unexcavated. along the 'meandering' and 'abandoned' river
OF LEAKAGE Based
October 22, 1990. a large quantity of water would have
channels. This would created high pressure against the down upon
details available. the following stages In
of the dam. which
would have led to the
appearance of leakages at STA 0+856 and STA 0+475.
the development of leakage are suggested.
removed by the flow, which resulted
formaHon of the cavity
Mobilization of man power and machinery to fight against the
failure of the dam was possible then. If the dam had leaked
beneath the embankment.
It would be
reasonable to assume that first the Inclined chimney drain would have collapsed along the wall of the Impermeable zone.
the material of the horizontal drain was already washed away leaving
At STA 0+475
material In the upper part unsupported.
The aufhors would like to thank the Royal lrrlgaHon
the caved-In of part of the downstream slope was somehow Instrumental
plugging of the
atoremenfloned assistant. fhe consequences would
have been very much more serious.
Department for the permission for preparing this paper. Thanks Is
Meanwhile, the erosion of foundation and embankment materials
also go to Mr. K. Sridhar for the permission to use some of the
figures from his thesis
the cavity, causing a further collapse of the leakage
To a certain extent, the undermined upstream
embankment collapsed such that cracks were developed and provided shorter seepage paths.
Enhanced by the dispersive
nature of the embankment material, the cracks were quickly developed to sinkholes and resulted In a vortex flow In the vicinity
1) NAM MUN PROJECT THAILAND, FEASIBIUTY INVESTIGATION.
of STA 0+856, observed on October 25, 1990, three days after the
Harza Engineering Company, June 1970.
first observaHon of leakage.
2) NAM MUN PROJECT FEASIBIUTY REPORT. Appendix IV,
Continuous dumping of materials Into the holes and the
Geology, USBR. Febuary 1971.
collapse, and the formation of erosion tunnels might have
contributed to the fluctuation of leakage quantity at STA 0+856. The continuous leakage led to the erosion of the downstream filter zone, thus generating a slope faiiiJre. 14. CONCLUSION Evidences from post-failure detaHed lnvestlgaHon affirmed that the leakage could not have been caused by water seeping Into the embankment body from the reservoir through the voids or cracks In the embankment. Thus It should be generally agree that the
leakage must Inmate from the high permeable
foundaHon. From the standpoint of fair conclusion, It Is necessary also to examine the question: 'If fhe horizontal
materials met fhe design criteria or specification, would leakage still occur 1•. question. capable
It Is Impossible to form a unique opinion on this
If the horizontal drainage blanket materials were of
embankment as It was Intended, It Is still not easy to conclude that 1he dam would never have faDed. The materials for horizontal drainage blanket were constructed of a rather uniformed materials throughout the length of the dam. The reservoir pressure subjected to every section of the horizontal blanket drain was almost the same Hence, the posslbllllty for leakage to taking place should be randomly the same everywhere. However, leakage only occurred at two specific locations, I.e. where the foundation composed of river channel deposit. A more fortunate aspect of this Incident was 1hat the leakage started
at fhe time when
Sridhar K On
MS. Thesis, Asian Institute of Technology, No. GT-91-29 , 1991
equlpmen18 were sfiD staHon at a dam not very far away.