265

THE WET SEAM AND THE TETON DAM FAILURE

JACK W. HILF
Box 440080, Gateway Station, Aurora, CO 80044 (U.S.A.)
(Accepted for publication December 1986)

INTRODUCTION

During the discussion of the June 1976 Teton Dam failure at the 13th
ICOLD Congress in New Delhi in October 1979 (1), I concluded t h a t the
most probable trigger mechanism for the failure of that dam was the con-
struction defect represented by the wet seam. This conclusion was based on
the fact that even after intensive post-failure investigations, the previously
postulated trigger mechanisms of hydraulic fracturing and differential settle-
ment could not be substantiated, while the wet seam was f o u n d to occur
throughout the core material of the dam remnant close to and parallel with
the 1974--1975 winter shutdown surface. This wet seam was characterized
by low density, high moisture content, and significantly lower temperature
than the normal compacted silt (6°C cooler in drill hole PR 100) (2).
I stated that the temperature difference implied that frozen ground or
frozen borrow soil was present near the 1974--1975 winter shutdown surface
and that when this layer melted an avenue of easy access of reservoir water
through the core resulted.
Since that time the Final IRG Report of the failure has been published,
and it attributes the wet seam to causes other than frost action. This Interna-
tional Workshop on Dam Failures presents an o p p o r t u n i t y for me to take
another look at the wet seam as a failure mechanism. I find that additional
data and information were available which had not been adequately con-
sidered in the IRG Final Report or which postulated t h a t Report. This discus-
sion is concerned with this new or not-adequately considered data. Also I
feel obliged to respond to some of the criticism of the Teton Dam design,
which I believe to have been unfair, by providing pertinent facts that should
be considered.

CONCLUSIONS OF THE IRG FINAL REPORT, JANUARY 1980

The final report of the U.S. Department of the Interior Teton Dam Failure
Review Group (3) states on p.4-11, that the results of a freeze--thaw analy-
sis; post-failure reports by construction inspectors t h a t no frozen fill existed
at the time of initial Zone 1 placement on May 1, 1975, and that no frozen
266

soil was placed within Zone 1; and information on elevations of wet seams
eliminate frost action, per se, as a possible cause for the wet seams. Rather,
Conclusion 9, on page 8-1 of the Report states: "9. The wet seams are the
result of thin, low<lensity, high permeability layers placed in the embank-
m e n t . Their low density is a function of fill placement procedures and/or
unusual material properties possibly related to varying percentages of
amorphous minerals or calcium carbonate equivalent in the soil's clay size
particles."
This explanation of the cause of the wet seams is not only indefinite, but
is obviously inconsistent with the following facts.
(1) As shown in figure A-12, Embankment Placement Zone 1, May 1975
of the IRG Final Report, placement was in horizontal layers from about
Elev. 5110 to Elev. 5140.
(2) The wet seam occurred just above or at the 1974--75 winter shutdown
surface which contained a 10 horizontal to 1 vertical slope. At least 14 post-
failure drill holes between Stations 23+00 to 19+37 f o u n d wet seams only
near that surface after having penetrated many layers of compacted fill above
(see Fig.l). The placement procedures and/or unusual material properties
could n o t conceivably be limited only to the portion of those horizontal
layers near the existing fill surface.
Thus a credible explanation of the wet seam must take into account its
proximity to the 1974--1975 winter shutdown surface. Since the IRG
explanation of the origin of the wet seam does not meet t h a t criterion, we
must take another look at frost action as a cause of the wet seam.

A. CASAGRANDE'S REPORT OF JUNE 1978

In June, 1978, the members of the Independent Panel who had investigated
the Teton Dam failure (report of December, 1976) (4) were engaged by the
Bureau of Reclamation to visit Teton Dam and to review additional investi-
gations with particular attention to implications of "wet seams". Individual
reports were made to the Bureau by the members of the Panel and each
report of these experts gave individual conclusions. Of particular interest to
me was the report of the late Arthur Casagrande, who for m a n y years had
been a member of the A.S.C.E. Committee on Frost Action in Soils and on
the basis of his early investigation of frost heaving of highway pavements was
uniquely qualified to evaluate the relation of frost action to the wet seams
found in the Teton Dam remnant.
In his 8-page report of June 26, 1978 (5) Dr. Casagrande discussed the
phenomenon of frost action in silty soils with respect to the climatic condi-
tions in late 1974 and early 1975 and the wet seams. On page 3 of his report
he wrote: "In my judgment, not only wet seams below the winter shutdown
1974/75, but also above the shutdown were caused in some form or other by
frost action in the highly frost susceptible soils of Zone 1."
And Dr. Casagrande concluded: "On the basis of the incomplete data and
the limited time which were available for review, m y tentative conclusions
are as follows:
 267

I~3 NI NOIIVA373
Y, Zg
i i i i i , ~ i i i , i ~ i , I i [ ~ i i ,' [ i i .~

%
..=
©

~-d9

o~-ud v7..,¢= ,,, <

Z d ,.--4-- ~

~Z ~

Ix_o, ~

"-'JZ I,,- ~
0 ~ m

N
268

1. The principal cause of the wet seams is probably frost action in the
highly frost-susceptible material of which the Teton Dam is constructed.
2. In the right abutment key trench, wet seams probably extended to rock
contact at approximately the location where piping started. This would
explain w h y piping developed so very rapidly. Therefore, the w e t seams
appear to have served as a contributory factor in the failure of Teton Dam."

C.W. JONES' REPORT OF MARCH, 1981

More recent information of frost action at Teton Dam is contained in

the publication "Closed-System Freezing of Soils in Linings and Earth
Embankment Dams" b y C.W. Jones (6). The author states that closed-system
freezing in a soil is a condition in which no source of water is available dur-
ing the freezing process b e y o n d that originally in the voids of the soil at and
near the zone of freezing, while open-system freezing is a condition in which
pore water in excess of that originally contained in the voids of the soil is
available to be moved to the surface of freezing to form segregated ice in
frost-susceptible soil.
Pertinent information from this publication related to Teton Dam includes
results of frost tube installations in the Teton Dam remnant February/March
1978 and during the winter 1978--79. The report on page 33 notes that
USBR specifications (including that of Teton Dam) state that "no embank-
ment materials shall be placed in the embankment when either the material
or the foundation or embankment on which it would be placed is frozen".
Under the heading "Exploration for Frost" the report states on page 34: " T o
fulfill the specifications requirement it would be necessary before resuming
embankment placement in the Spring, to thoroughly explore for frozen soil.
In the Teton Dam remnant, it was n o t possible b y conventional mechanical
sampling equipment to determine frost depth in the compacted silt nor to
tell b y observation or scratching the side of the hole with a special t o o l . . . "
Amplification of that last statement is contained in a letter from C.W.
Jones to Dr. A. Casagrande dated October 29, 1978 (7) which states: "When
the field crew installed the frost tubes at Teton a b o u t the first of last March,
they were told to look for any signs of ice lenses or striations in the holes
made in the frozen Zone 1 embankment. They reported that t h e y could not
see any such evidence of frost action. In fact, they could n o t tell when t h e y
had reached unfrozen soil, even though we sent them a special tool to scratch
the side of the hole."
Under the foregoing circumstances the post-failure statements b y field
inspectors that there had been no frozen soil in a 5-ft. deep cutoff or key
trench excavated in the Zone I of Teton Dam in early May 1975 cannot be
considered reliable. Yet the IRG and other investigators have accepted such
statements as facts.
Similarly the inability to tell by visual observation that a silty borrow soil
in the loose state is frozen was reported to me b y an experienced soils engi-
neer who had worked extensively with glacial till soils on dams in Canada.
 269

He stated that what had appeared to him to be relatively dry soil was com-
pacted normally, then to his surprise later became extremely wet and unstable
when thawing occurred.
Therefore, it is probable that frost could have been present in the Zone 1
embankment in April--May 1975 and in the borrow soil placed in the fill in
early May 1975, without having been noticed by the Bureau inspectors and
the Contractor.

THE 5-FT. D E E P K E Y T R E N C H IN Z O N E 1

Independent Panel Member E. Montford Fucik in his June 15, 1978 report
(8) on his visit to Teton Dam made a recommendation related to this trench:
" . . . (2) Drill several holes into the area where the 5 f o o t deep cutoff trench
was excavated during May 1975 in the 1 on 10 slope of the winter shut-
down surface. The presence or absence of the wet seam will help decide if
the seam was caused by frost effects. If no wet seam is f o u n d , that will add
weight to the idea that frost effects caused the wet seam."
Unfortunately this recommendation was not implemented as can be seen
in figure A-11, "Surface of 1974--75 winter s h u t d o w n " of the IRG Final
Report (Fig.2). It shows a rectangular area located 40 ft. upstream from the
dam axis and extending from about Station 19+30 to about Station 21+50,
identified as "5' Deep key trench, April 1975 Survey Book 549-X-468".
No drill holes are located within the rectangular area which represents the
b o t t o m width of the trench, but one hole (PZ-4 at Station 19+37.0, 63.0 ft,
upstream from dam centerline) is just outside of the excavation for this
trench. This hole was drilled using 6-inch flight augers w i t h o u t casing from
ground elevation 5218.0 during the period April 5--10, 1978. The log of
this hole on page A 20-54 of the IRG Final Report shows t h a t the wet seam
was found at El. 5139.0 and extended to the b o t t o m of the hole at El. 5138.1.
Since the ground elevation of the 1974--75 winter shutdown was 5138.5
(from Survey Book 549-X-468), at this upper part of the dam remnant the
wet seam was coincident with or below the winter shutdown surface. The log
of this hole, which had been converted to a piezometer, shows water levels
as deep as 80.9 ft. or to El. 5137.1 which is 1.4 ft. below the shutdown
surface.
Let us look at this 5-ft. deep key trench more closely. Apparently it was
excavated in accordance with the following provisions of the Teton Dam
specifications, Paragraph 86 c (2):
"(2) A temporary gap through the dam embankment, for diversion pur-
poses as described in Paragraph 63 will be permitted: Provided, that the
slopes of transverse bonding surfaces (surfaces normal to the centerline crest
of embankment) in Zone I shall not be steeper than 4:1, and in other zones
of dam e m b a n k m e n t , transverse bonding surfaces between previously com-
pleted portions of embankment and embankment to be constructed shall not
be steeper than 2:1. No other transverse bonding surfaces will be permitted
in the dam embankment.
270

32c ~_ ~ zo.E3

24C j S'~'~ Z ~ /
-- 20C "---"- 5109 ...~._..,ZONEI
OF 5110 N

1975 ~(10.5 / v * 10

o
8(]
,c
/ ~ "~' ELEVATI
LAI
4~-0 --~-" l " - - - S DEEPKEYT~EN[HAP~L1975 - ---
K~EYTRENCHI ~- ~ - - - J AP409
80 '----^n~T-- I ( 1 2 .- 6~ ) ' E' L 5 1- 1 2 . -~ 0 ; = ' - _~ - - --=~ ==:-~
" ~ - -.~ - ' ' -"' ' - - " - - ' -

El P~lOOa*0 p 51b
~U.
,~,.. ,EEP*~EO.,*= I \
--FOFCUT,O(TOBEi
~ I / I lPR:°' I /I I"~-7 5155 ,

24 1LIi6'sl • 13

~2eO~-,ED.OC,,'~"'."1 "~ "hl\\Y I/ ~,l,'o,o, ,o,,,

ZONE

360
4001 q I I I
26-1"00 24+00 22+00 20-t"00 18-1-00 , 16+00 14+00
STATION ALONG ~ OF DAM

Fig.2. Surface of 1974--1975 winter shutdown (after fig.A-11 of the IRG Final Report).

"During construction of earthfill, Zone 1, embankment in the opening at

the gap, the contractor shall construct a keyway trench in each 4:1 transverse
bonding surface in the previously placed earthfill, Zone 1. The k e y w a y
trenches shall be excavated in the bonding surfaces to a minimum depth of
5 feet, shall have 4:1 side slopes, and shall have a minimum b o t t o m width of
20 feet. The centerline of the trenches shall be located approximately mid-
way between the upstream and the downstream slopes of the Zone 1. The
trenches shall be refilled with earthfill, Zone 1, material subject to the provi-
sions of Paragraph 87."
In June 1985, I examined the referenced Survey Book. It gives the data
for setting of the stakes for the cut slopes of this trench which was 5-ft.
deep, 20-ft. b o t t o m width, with 4:1 side slopes. The trench extended from
Station 19+25 at the t o p of the 10:1 slope to Station 21+50 at the toe of
that slope. The t o p width of the cut for the trench varied slightly b u t was
staked o u t to be a b o u t 58 ft. I could find no separate record of the construc-
tion of this trench, but the extracts from construction records associated
with the placement of Zone i (App. A2 of the IRG Final Report) for May 1,
1975 swing shift, page A2-8 states:
" . . . One D8 with dozer pusher scraper to excavate trench Sta. 19+50 to
Sta. 21+00, to level fill and to push (material to) left a b u t m e n t 330+ feet
U/S and 275+ feet d/s. One scraper excavating key trench and hauling the
material to left a b u t m e n t . . . "
Hence it appears that this trench was excavated on May 1, 1975, prior to
the start of full-scale placement of fill in Zone 1.
 271

What relation does this trench have with the wet seam? F r o m Fig.2, none
is apparent; however, in that figure the dashed line showing seepage on the
face o f cut, Oct obe r 1977, f r o m Station 24+40 t o Station 24+60 is incorrect
in th at the 48-ft. gap in the wet seam shown clearly in p h o t o g r a p h A-10 of
the IRG Final R e por t (Fig.3) was omitted. On page A-4, Appendix A of the
IRG Final R ep or t the 48-ft. gap is stated t o be located from 29 ft. upstream
t o 77 ft. upstream f r o m dam centerline, bu t a review of the weekly reports
o f the investigation shows that the correct location of the gap was f r o m 20 ft.
upstream to 68 ft. upstream f r om dam centerline. Thus the 5-ft. deep key
trench in the Zone 1 e m b a n k m e n t is almost precisely aligned with the only
place within Zone 1 where t he exposed wet seam was absent! (see Fig.4).
This certainly implies t hat frost was present in the soil at least to a shallow
d e p t h below the winter s hut dow n surface on May 1, 1975. Its removal by
excavation o f the 5-ft. deep key trench would have effectively cut off the
source o f water pressure f r o m the 10:1 slope. This pressure was evidenced by
measurements in the wet seam and by the fact that the wet seam continued
to seep for more than 9 m ont hs after it was exposed in O ct ober 1977. A
pumping test in a test pit in the wet seam indicated a flow of 2.2 gallons per
hour (P. A-18 IRG Report). This gravity flow to the observed wet seam at
Sta. 24+30 explains w hy no wet seam was visible f r o m the right a b u t m e n t
across the failure gap. Since the highest elevation of the wet seam would be
exposed there it would n o t seep, just as there was no seepage in the 48-ft.
gap at Sta. 24+40 where t he c o n t i n u i t y of the wet seam was i nt errupt ed by
the key trench constructed in May 1975.

FREEZE--THAW ANALYSIS -- NEW FACTS

The IRG Final R epor t eliminated frost action as a cause for the wet seam
n o t only on the basis of post-failure interviews with construction inspectors,
bu t on the results of freeze--thaw analysis• I shall n o t a t t e m p t to question
t h e assumptions used in t hat analysis which concluded t hat under the most
conservative conditions frost would have left the fill by May 8, 1975. How-
ever, the p h e n o m e n o n of freezing at a soil--rock interface was n o t considered.
The Jones' r e p o r t (6) states on page 34:
" T h e r e is one aspect of closed-system freezing in an incomplete earth em-
b a n k m e n t dam that could possibly be a cause for failure of the dam, but
which seems to not have been published. This has to do with frost action in
e m b a n k m e n t material in contact with a solid b e d r o c k abutment• The ther-
mal conductivity of r oc k can be an order of magnitude greater than t hat of
soil (23).
T h e r e f o r e it seems logical t hat under certain conditions, frost action could
penetrate much deeper at the soil--rock interface than it would in the em-
b a n k m e n t at some distance f r o m the interface ( f i g . 3 5 ) . . .
• .. The f o r m a t i o n of a critical piping zone at the interface could be accen-
t u ated b y any water f r om snowmelt, precipitation, or c o n t r a c t o r operations,
flowing down the a b u t m e n t and freezing in the embankment• The possibility
272

Plmtogralth A-10.-The location o[ the initial wet ~ a m s encountered in the .south lave of the embankment remnant. The wet
~eams are vo~ered with plastic ~heets. 10/6/7T. PK-12-10 and PR-9- 1 1 .

Photograph A-I I.-The wet seams at Stm. 2 4 + 5 0 ± and 100± feet

d.wn~tream of the dam eenterline. Note the flow~ of ~aturated
fill nmvin K down,lope. 10/7/77. PK-14-12.

Fig.3. Gap in the wet seam (photograph A-10) and the flowing wet seam (photograph
A-11). (After IRG Final Report).

of dam failure is predicated upon the assumption that construction personnel

were not aware of the deep freezing and possible effects and did not exca-
vate the soil deep enough and recompact it to specification l i m i t s . . . "
Note that the effect of external water, as Jones assumed, changes the
closed-system freezing to open-system freezing which is more severe. The
fact that precipitation in the form of snow and rain occurred during May
1975 has been well documented by the IRG Final Report. What about con-
tractor's operations that could have contributed to increase in frost action?
In his report of the June 1978 visit of the Independent Panel members to
examine the wet seam (9), Tom Leps ascertained that water from the grouting
 273

26+00 24 + O 0 22+00 20+00 18+00

STATIONS ALONG (L OF DAM

Fig.4. Portion of 1974--1975 shutdown surface from Sta. 17+00 to left abutment; added
gap in wet seam and cut line of trench in Zone 1 from survey book; adapted from fig.
1-11 of the IRG Final Report.

o p e r a t i o n s o n t h e right a b u t m e n t was f l o w i n g in a d i t c h o n t h e Z o n e 1 dur-

ing t h e w i n t e r o f 1 9 7 4 - - 7 5 at t h e l o c a t i o n o f t h e i n i t i a t i o n o f failure. This
f a c t was r e p o r t e d t o m e i n d e p e n d e n t l y b y L. D a v i d s o n w h o said t h a t t h e
individual in charge o f t h e g r o u t i n g r e c e n t l y t o l d h i m a b o u t this w a t e r f l o w .

RELATION OF WET SEAM TO WINTER SHUTDOWN SURFACE

A final r e a s o n f o r t h e I R G ' s e l i m i n a t i o n o f f r o s t a c t i o n as a possible cause

o f t h e w e t s e a m is i n f o r m a t i o n o n elevations o f w e t seams. I t implies t h a t
since t h e w e t seams w e r e f o u n d above t h e 1 9 7 4 - - 7 5 w i n t e r s h u t d o w n surface
in t h e r e m n a n t , n o f r o s t e x i s t e d b e l o w t h a t surface. I h a v e a l r e a d y s h o w n
t h a t t h e w e t s e a m o c c u r r e d below t h e s h u t d o w n surface in PZ-4. T h e s a m e is
t r u e f o r Drill H o l e PR 1 0 8 w h e r e a w e t s e a m w e n t as l o w as Elev. 5136.1
while t h e s h u t d o w n s u r f a c e was at Elev. 5 1 3 6 . 5 . Also, t h e t h r e e m o s t d o w n -
s t r e a m s u r v e y e d p o i n t s o f t h e e x p o s e d w e t s e a m at Sta. 2 4 + 4 0 , r e p r e s e n t i n g
a d i s t a n c e o f a b o u t 50 ft., are f r o m 0.2 ft. to 1.6 ft. below t h e w i n t e r shut-
d o w n surface. It m u s t b e r e c o g n i z e d t h a t a f r o s t - d e r i v e d w e t s e a m b e l o w
s h u t d o w n g r o u n d s u r f a c e m e a n s t h a t t h e entire surface a n d t h e entire bor-
r o w area s u r f a c e p r o b a b l y c o n t a i n e d frost.
 274

A check on the surveys on which the surface elevations of the 1974--75

winter shutdown were based shows that there probably was no large syste-
matic error in the elevations although the benchmark used for that survey
was destroyed during the failure and could not be used to establish elevations
for the wet seams. Hence it is probable that frost action occurred both in the
shutdown surface and in the borrow material placed in early May 1975, the
latter either as visually undetected frozen soil or subsequently frozen during
the cold nights. There apparently was ample moisture from rainfall and snow
to provide open-system freezing. It is inconceivable that the material f o u n d
in the wet seam was placed in that condition in the fill. It would have been
impossible to traverse such soil with hauling equipment. Hence the wet seam
must be the result of secondary processes wherein the soil was compacted
normally and later softened and became saturated. This could happen by
extremely dry placement conditions resulting in very low dry densities or by
the phenomenon of frost action followed by reservoir inflow. The proximity
of the wet seams to the sloping shutdown surface and its areal content of
more than 5 acres make frost action the more probable mechanism for the
observed wet seams rather than very dry placement.

MOST PROBABLE MECHANISM OF FAILURE

From the foregoing discussion of the wet seam, the most likely trigger
mechanism was frost action resulting from the V-shaped ditch in the Zone 1
used as a waste water channel for the grouting operations on the right abut-
ment between December 1974 and April 1975. This trench was located
about 10 ft. from the downstream side of the key trench then continued
along the downstream right a b u t m e n t of the dam, according to Davidson
(10).
There is no longer any need to postulate mechanisms that, although pos-
sible, have no physical evidence to sustain them when we have a clear case of
some frost action t h r o u g h o u t the Zone 1 in the vicinity of the 1974--75
winter shutdown surface and especially severe frost action conditions at the
right a b u t m e n t (the rock and the water ditch) which fit exactly the location
where the failure was triggered.
In summary, the following new facts should make frost action the most
probable mechanism for triggering the failure of Teton Dam:
(1) The coincidence of the gap in the wet seam with the location (ex-
tended) of the 5-ft. deep key trench in the 1974--75 winter shutdown surface.
(2) Some wet seam elevations below the winter shutdown surface.
(3) The effect of the rock contact on the depth of frost action in soils
(Jones' report) which was not taken into account in the IRG frost action
calculations.
(4) The availability of surface water during the winter months which was
conducive to "open-system freezing" at the location (Station 14+00--15+00)
of the failure.
 275

(5) The unreliability of post-failure statements that no frost existed below

the 1974--75 winter shutdown surface or in borrow material placed in May
1975. This is based on the Jones' Report as well as information from an
experienced soils engineer who was fooled by frost action in borrow soils.

C R I T I C I S M O F T H E T E T O N DAM D E S I G N A N D L E S S O N S L E A R N E D

The design of Teton Dam has been severely criticized for its failure to pro-
vide adequate sealing of the downstream faces of the a b u t m e n t key trenches
and filter material between the silt core and the a b u t m e n t rock. The follow-
ing facts should be considered in deriving lessons to be learned from this
tragic failure:
(1) The use of key trenches backfilled with impervious soil was a design
feature in lieu of attempting to construct a grout curtain in this pervious
rock formation. It should be recognized that neither the Bureau geologists
nor the designers had foreknowledge of the extensive open joints that were
revealed when these key trenches were excavated. This fact is d o c u m e n t e d
by the travel report of March 29, 1974, by a construction liaison engineer, a
designer, and a geologist. After an inspection of two intensely fractured zones
containing extensive voids and fissures in the right a b u t m e n t key trench at
about Station 3+55 and Station 4+30, the report states in part:
"d. Preconstruction investigations indicated that intensely fractured zones
with high permeability existed in this area, but no voids were encountered
during drilling, nor were any observed during television l o g g i n g . . . "
Hence in accordance with long-standing and previously successful practice,
no special surface treatment other than special compaction was included in
the specifications.
(2) Once the initial excavation of the key trenches had been made, exis-
tence of some open joints in the upstream and downstream faces of the key
trenches was revealed. Were these conditions ignored by the designers? The
facts are otherwise. Evidence that the designers instructed the Project forces
to treat these open joints is as follows.
(a) Joint travel report {including a designer) dated October 25, 1972.
After an inspection of the excavated key trench on the left a b u t m e n t and
the partially excavated key trench on the right a b u t m e n t with the construc-
tion engineer and the field engineer, the report states:
" A positive assessment of the jointing system in the key trench walls was
not possible because of the effects of blasting. It appears that only in iso-
lated areas will it be necessary to seal the surface of the downstream key
trench wall to prevent the Zone 1 material in the filled trench from piping
through open j o i n t s . . . "
(b) Joint travel report (of a geologist and a designer) dated November 23,
1973, prior to placement of any Zone 1 material in the key trenches states
that the " . . . large mass of loosely jointed block rock on the downstream
side of the left a b u t m e n t . . , can be treated by sealing and guniting in selected
areas as the fill is placed."
276

Thus designers, geologists, and the construction liaison engineer on several

occasions discussed and participated in the method and timing of treatment
of joints and other openings in the rock on the sides of the key trenches. The
actual treatment was accomplished under the direction of the construction
forces as indicated in fig.9-1 of the Independent Panel's report, "Slush
Grout Distribution and Density-Zone 1 Foundation Sta. 11+41 to 16+00."
(Fig.5). The highest elevation of treatment is interpolated as 5235 ft. Appa-
rently above the elevation the rock did not require treatment in the opinion
of the construction forces.
(3) With respect to the criticism as to lack of a filter between the rock
a b u t m e n t and the core, I understand that the design concept was to construct
an impervious plug within the key trench. It was not contemplated that this
well-compacted soil would crack; nor is there any evidence from the archaeo-
logical-type excavations made in the remnant of the right a b u t m e n t key
trench nor in the completely excavated left a b u t m e n t key trench that any
cracking b y hydraulic fracturing or differential settlement occurred or was
imminent. Without such a crack the piping that occurred must have had
another mechanism. That mechanism was frost action, in m y opinion.
If severe, open-system frost action had not been present in the right abut-
ment key trench it is reasonable to assume that failure would n o t have occur-
red on June 5, 1976, even if the sealing of the downstream side of the key
trench shown in Fig.5, were not complete.
The post-failure excavation of the left abutment key trench was made pri-
marily for the purpose of validating the postulated mechanisms of failure,
since the evidence on the mirror-image key trench on the right a b u t m e n t had
been lost b y the failure. Having found absolutely no evidence of cracks, seep-
age, or piping in the left trench, it is manifestly unfair to say that failure
would nevertheless have occurred there also, perhaps somewhat later. In my
opinion this is unacceptable, "heads I win, tails y o u lose" thinking, that does
not engender confidence in investigative procedures.
It should be noted that any slowly developing leak considered to be
dangerous could have been controlled b y the high-capacity b o t t o m outlet
works which had been provided in the design of the Teton Dam for such a
contingency. A similar outlet saved the Fontanelle Dam from failure in 1965.
The b o t t o m outlet at Teton Dam was not used on June 5, 1976, because it
was n o t quite ready. Even if it had been ready, the few hours available for
lowering the reservoir might not have been sufficient to control this rapidly
developing leak through the dam.
In addition to the healthy emphasis on foundation treatment and protec-
tive filters for the cores of embankment dams that has resulted from the
Reports of the Independent Panel and the IRG, there is another lesson to be
learned from the Teton Dam failure.
More emphasis should be given to the identification and treatment of frost
action in winter shutdown surfaces and in borrow areas, for all impervious
cores of dams located where frost is significant. It is interesting to note that
such provisions, including the use of frost tubes, are n o w being used in
 277

m ~ o SrwS n ~ E
LJrw~" rw
o
~<<
~_>_>_<<> <>- <zm{2~
m
o zo
zz
<O
~-
'"
rm (DO (9 (.9 (D ~< OZ _ .~!
zl~ • m m ~'>- 0
wwDD W ~rn
~moo ~ ~ 6 ~o
--z + |"
r ~ D
o©~ ~ ~ ~ OO ,~'" ~:o~ '
ooooo00 ~w
rw
I--t~ 0
Or},.- I--

~ ~zr~ • "O
m)
~Z
on,-~ +
,.-
w< 0>-~ 0

~z &
..Aw
m~

o O
O
+

2
+

<

0
N

@3 0 0
o

®
L~
@,

o© -%
y o~ 0 0o, OOLC6o -----"0--'-~

A~
278

U.S. Bureau of Reclamation specifications. This appears to be tacit recogni-

tion of the importance of frost action in the Teton Dam failure.

REFERENCES

(1) Proceedings, 13th ICOLD, New Delhi, October 1979 Vol V p.422.
(2) U.S. Bureau of Reclamation, Pacific Northwest Region, Regional Office, "Investiga-
tions for Wet Seam -- Teton Dam Remnant -- Teton Basin Project, Idaho," June
1978, 6pp.
(3) Failure of Teton Dam -- Final Report by U.S. Department of the Interior, Teton
Dam Failure Review Group, January 1980.
(4) Report to U.S. Department of the Interior and State of Idaho on Failure of Teton
Dam, by Independent Panel to Review Cause of Teton Dam Failure, December 1976.
(5) Report of June 26, 1978 to Bureau of Reclamation by Arthur Casagrande, Subject;
Teton Dam, Review of Additional Investigations with Particular Attention to Impli-
cations of "Wet Seams".
(6) REC-ERC-81-1 Closed-System Freezing of Soils in Linings and Earth Embankment
Dams, by C.W. Jones, Engineering and Research Center, Water and Power Resources
Service (now Bureau of Reclamation) March 1981.
(7) Letter from C.W. Jones to A. Casagrande dated October 29, 1978. (Private communi-
cation from C.W. Jones.)
(8) Letter of June 15, 1978 by E. Montford Fucik, subject: Report on Visit to Teton
Dam Site and Review of Technical Information -- Left Abutment Remnant Excava-
tion, Purchase Order 8-07-DP-00249.
(9) Letter on June 25, 1978 to Bureau of Reclamation by Thomas M. Leps Inc. Re:
Teton Dam Review -- IRG Post-1976 Investigations Po. No. 8-07-dp-250.
(10) Notes by L.W. Davidson for possible inclusion in the paper, "Reconsideration of
Failure Initiating Mechanism for Teton Dam", by G.A. Leonards and L.W. Davidson;
presented at the Conference on Case Histories in Geotechnical Engineering, St. Louis,
Mo., May 1984.

(Leonards)
I had intended to summarize what I believe are all the facts about the
wet seams and to point out that it is necessary that a correct hypothesis ex-
plain all the facts, not just a select few of the facts. In m y judgment the
hypotheses that have been proposed can explain some of the things that were
observed, but not all of the things that were observed. There may be more
than one type of wet seam in the dam, as none of the explanations given
thus far are consistent with all the facts. However, in view of the time, I
turn the podium back to the panel for any concluding remarks t h a t t h e y
m a y wish to make.

(Sherard)
There were several points made, and I will try to respond to them in order.
First, the idea that there was no evidence found in the post failure explora-
tions on the right and left sides that there was any piping in progress. The
idea that such evidence is important is for me not valid. Once the leak started
the dam failed within a few hours. Therefore, there is no reason to expect,
regardless of the origin of the initial leak, that there would have been any evi-
 279

dence developing any place else. I can easily imagine that the reservoir could
have been filled and it did not fail, but it would fail later on. I can easily
imagine t h a t if it did not fail on the right side it could have failed a week
later on the left side. There is no reason to think that because there was no
incipient piping found anywhere else that this restricts us in our thinking
about the initial leak.
Second, the idea of the low temperatures. The hydraulic fracturing hypo-
thesis will handle it very well because the water was very cold. It was a snow
melt.
Third, the idea of Mike Duncan's about the different material in the wet
seams. I say that, as I read this thing, I could not be sure whether the conclu-
sion that the material in the wet seams was different was due to a statistical
error based on an insufficient number of samples or whether it is, in fact, a
different material. Clearly, it is a very uniform borrow material. If it was,
for example, a layer or a zone placed with higher permeability, t h a t would
be the first place the water would penetrate in hydraulic fracturing.
Fourth, regarding the idea that it is not important from the standpoint of
explaining the water dribbling out, whether the water went in by hydraulic
fracturing, or whether something happened during construction. In the
hydraulic fracturing hypothesis a thousand times more water could go in
than could get in from any other hypothesis. It could have been going in for
a couple of months, and in my opinion, that is the only way that free water
could have been f o u n d seeping out months later. Anything that happens dur-
ing construction no matter how badly frozen it was, if there were no ice
lenses, is not going to have any influence whatsoever on the condition of the
fill 2 years later under 200 feet of overburden.

(Penman)
I believe that ! am the only person who is sufficiently independent to be
able to make remarks in the present company. I would like to support Dr.
Jim Sherard, but with slight modifications. I believe that the b o t t o m 100--
120 feet of fill that was placed prior to the 1974--75 shut<lown period
could very well be subject to collapse settlement on wetting because we
know that it was placed relatively dry. The Bureau's own tests show this
a m o u n t of collapse settlement taking place when material that had been
placed 3% dry of o p t i m u m was wetted in the laboratory. So we have got
perhaps 4--5% compression, and 4--5% compression on 120 feet gives y o u
the 5 or 6 feet t h a t Jim Sherard was talking about. [Sherard interjection.
You only have 60 psi pressure there, that should be 300 psi.] Well, that
makes the matter even better. So that the gap t h a t would be formed, as Jim
has pointed out to you, would fill as the loose material forming the soffit of
our passage way through the fill, as that collapses and falls down onto it.
Then, as Jim has described to y o u very clearly, when the failure occurred
and the arch was broken the upper part of the dam fell into this loose mater-
ial. Of course, when y o u excavate the loose material and measure it, it has a
much lower density than when it was placed. It is almost impossible for the
280

Bureau's construction people, or the contractor who is doing the work, to

have placed the material in such a loose condition with the machines he was
using. I fully support the notion that the lower part of the fill settled. Of
course, we have evidence t h a t the upper part of the fill did n o t settle. I
believe, there were reference monuments placed in the upper part of the fill
and the measurements showed t h a t the settlements were extremely small
confirming the arching action. So that with the b o t t o m part settling, I think
that this fits in very well with the hydraulic fracture concept and with the
possibility of a 5 or 6 feet loose zone forming which of course would collect
a sufficient a m o u n t of water for it to dribble for 2 years.

(Duncan)
Just one last remark. The samples that were taken from the wet seams
were measured; their densities, their water contents were measured. The aver-
age degree of compaction in those wet seam materials met the compaction
requirements for construction of the dam. They were lower<lensity mater-
ials, but t h e y had a lower-compaction curve. There were some samples from
the wet seams that did not meet the compaction requirements, but on the
average t h e y met the compaction requirements. It is inconceivable to me that
these materials were subjected to drifting down through a 5-ft. cavity filled
with water and coming to the density that it came to as it accumulated on
the base on this 5 ft. horizontal crack filled with pure water. The average
density of those samples met the compaction requirements.

(Sherard)
Mike, when y o u take a sample of the silt as a slurry and if y o u consoli-
dated it in under 300 ft. of fill it will go to more than 98% of Proctor den-
sity. That is what happened after the failure.

(Duncan)
Then it is at most at the saturated water content that it would have at
98% compacted density, right? The water t h a t y o u saw dribbling out had to
be dribbling out of the material which was at most saturated at 98% Proctor
density. This then erases nearly all the history having to do with y o u r mecha-
nism of formation of the seam pumping in, as y o u say, a thousand times as
much water. I think that is a little exaggerated.

(T.M. Leps)
One of the things that has occurred to me in all of this discussion has been
t h a t we are beating a particular speculative item to death. It is a fascinating
thing for those of us in soil mechanics to try and imagine and to wonder
exactly how it could happen and as it has been indicated there may be 3 or 4
or 5 different ways that this seam could have happened. What is the b o t t o m
line? The b o t t o m line is t h a t I hope we all leave here knowing how to design
a dam so that we do not care if this happens. What we want to do from what
we have learned at Teton, is t h a t we want to seal open joints in the rock and
 281

we w a n t t o provide p r o p e r filtering b o t h u p s t r e a m and d o w n s t r e a m f r o m t h e

core zone. When we have t a k e n care o f t h a t the w e t seams, and I t h i n k t h a t
t h e y do o c c u r in e v e r y e m b a n k m e n t t h a t was ever built, just like t h e varia-
tions t h a t we find in d e n s i t y and m o i s t u r e c o n t e n t during p l a c e m e n t d u e to
t h e looseness, s o m e t i m e s , in c o n t r o l . Let t h e m be t h e r e . T h e y are n o t going
t o cause t h e d a m t o fail.

(Fucik)
I do n o t agree w i t h t h a t last s t a t e m e n t b y T o m Leps. If y o u l o o k e d at t h e
picture showing t h a t seam going across a l m o s t the full w i d t h o f t h e core,
which was m u c h m o r e t h a n half of t h e w i d t h o f t h e dam, t o assume t h a t t h e
failure o c c u r r e d f r o m piping in o p e n cracks in t h e r o c k , in m y o p i n i o n , is a
v e r y large a s s u m p t i o n t h a t is n o t justified. We are n o t o v e r e m p h a s i z i n g the
wet seam. I believe t h a t t h e w e t seam was the m o s t significant thing t h a t
came o u t o f T e t o n Dam. T h e idea t h a t t h e r e are wet seams in d a m s like t h e
w e t seam in T e t o n , in m y o p i n i o n , is n o t correct. I have seen a lot o f w e t
areas in dams, b u t I have never seen a n y t h i n g , I have never h e a r d o f a n y t h i n g ,
r e m o t e l y t h e same as t h e w e t seam we saw. I can easily imagine t h a t this was
the cause o f t h e failure because t h e r e was no o t h e r reason e x c e p t t h o s e based
o n p u r e c i r c u m s t a n t i a l evidence. What we have learned, as P e n m a n has
p o i n t e d o u t in t h e m a n y e x a m p l e s he gave, we all p r o b a b l y have m a n y
examples o f r o c k f o u n d a t i o n s t h a t have joints in t h e m t h a t we have t r e a t e d
b y p u t t i n g c o n c r e t e or slurry, or w h a t e v e r , over t h e m . I do n o t t h i n k t h a t
m a n y seasoned d a m designers or e x p e r i e n c e d p e o p l e have learned m u c h f r o m
t h a t part o f i t . . . t h o s e pictures o f t h e r o c k t r e n c h w i t h o u t having a n y treat-
m e n t o f t h e joints, I do n o t t h i n k we have learned m u c h f r o m t h e m . But t h e
wet seam, I do n o t t h i n k a n y o f us have seen a n y t h i n g like it b e f o r e and I
t h i n k t h a t we should t r y t o learn s o m e t h i n g f r o m it.

(Leonards)
We could h a r d l y have h o p e d for a m o r e a n i m a t e d discussion t h a n we have
had this a f t e r n o o n and I am very, very pleased t h a t t h e differing o p i n i o n s
w h i c h have b e e n simmering for m a n y years have finally surfaced. I do h o p e
t h a t m a n y o f y o u will feel m o v e d t o s u b m i t r e t r o s p e c t i v e c o m m e n t s , as
suggested in m y o p e n i n g r e m a r k s , t o say w h a t y o u feel n o w given all of this
discussion. I can tell y o u t h a t I i n t e n d t o do so, as t i m e constraints deprived
me o f an o p p o r t u n i t y to m a k e m y s t a t e m e n t s now.