DOCUMENTATION OF DISCUSSION WITH NALCOR

ON DESIGN ISSUES ASSOCIATED WITH THE

NORTH SPUR

A list of 20 issues was attached to a letter to the Muskrat Falls Oversight Committee (OC)
from Ron Penny and David Vardy on January 16, 2017. The OC advised us in letters dated March
31, 2017 and June 29, 2017 that our letter and attached list of issues had been forwarded by
the OC to Nalcor. A response was received from Nalcor routed through Karen ONeill on July
2017.
This document contains the issues list, the Nalcor response, and finally our comments.

The document is number-formatted to facilitate reading as follows

1. Subject of issue.
1-A Issue.
1-B Nalcor response.
1-C Comment on Nalcor response.

2. Subject of issue. Etc.

Note there are several references to the slope of the dam sides in the document. In all
such references, the normal hydro methodology has been used, namely horizontal distance to
vertical distance. Hence a 3:1 slope is flatter than a 2:1 slope.

The comments have been prepared mainly by Jim Gordon,

with assistance from Maurice Adams and Phil Helwig.

1
 1. Precedents for dams on sensitive clay.
1-A Issue.

This is the first use of a "natural dam" containing quick clay in a hydro facility. Other
precedents at Ontario hydro and Hydro Quebec are far smaller, and have a totally different dam
section with far flatter slopes, constructed with selected materials, and therefore are not
"natural".

1-B Nalcor response.

The North Spur Dam is not the first use of a natural dam built on sensitive clays. The edge
or rim of a reservoir is considered to be a natural dam and there are many examples (of) sensitive
clays being part of the reservoir rim. There are also several examples in Quebec and Ontario of
dams built partially or completely on sensitive clays. There are two examples of dam built on
sensitive clays on the Chicoutimi river: Chute Garneu and Pont Arnaud dams. These dams are
partially built on sensitive clay deposit. In areas where sensitive clay is present, like in Saguenay-
Lac St-Jean, similar cases to the North Spur can occur. The reservoir rim can contain natural
sensitive clay deposits like the North Spur. This is the situation for all dams on the Chicoutimi,
Aux Sables, Shipshaw, Peribonka Rivers and Kenogami, Lac St.-Jean Shipshaw Chute and du
Diable reservoirs. The Waba Dam in Ontario is another example. Design and construction
methodologies have to be adapted with regards of soil properties for any embankment structure,
this is true for every natural material and it is true also for the soil and rock present in the North
Spur.
1-C Comment on response.

The question referred to dams on sensitive clays, not reservoir rim stability.

The dams referenced as precedents are all small dams compared to the large North Spur
natural dam. This is shown on the following photographs.

Pont Arnaud Dam photos showing height of the two dykes.

2
 Pont Arnaud and Chute Garneau Dams. Google Earth images.

Comparable photo of downstream face of North Spur.

Note size of pickup shown by red arrow.

Both the Pont Arneau and Chute Garneau dams are very small dykes with wide berms as
can be seen in the above photos.

The small 11m high Waba Dam in Ontario is founded on a 66m deep deposit of sensitive
clay. The cross-section through the Waba Dam bears no resemblance to the North Spur natural
dam. The Waba Dam sideslopes are much flatter, and there are 60m wide berms on both sides,
as shown in the following section.

3
 Source - GEO-SLOPE International Ltd, Calgary, Alberta, Canada www.geo-slope.com
QUAKE/W Example File: Waba Dam permanent deformation (pdf)(gsz)

There are several dams founded on sensitive clays in the James Bay projects. However,
they were very conservatively designed according to the following extract from page 75 of the
Hydro Quebec publication La Grande Rivire Hydroelectric Complex, published in 1988.
Clay.

Clay material was also encountered in the deep (20m) rock depressions along Dykes D-5
to D-14, located south of the main dam at La Grande 2. This soil with low resistance and high
sensitivity, was completely removed beneath the highest dykes or left in place beneath the low
ones. In the latter case, (Figure 6.7), the filter zones were enlarged, because of the greater risk
of the core cracking from inevitable differential settlement. Moreover, berms were provided on
either side of the central zone, to ensure the stability of the dyke.

The low dykes were less than about 10m high, with 20m wide berms on both sides, as
shown in the following Figure 6-7 illustration.

The same design criteria was used on the Eastmain River developments in 2002-5, where
sensitive clay was found in the foundation for Dyke 24-A. Even though the foundation
excavation required cofferdams and pumping out a small lake, all sensitive clay was removed
down to sound rock as shown in the following panorama.

4
 Figure 6.7 showing Hydro Quebec designs for dams on sensitive clay.
Note large berms on both sides of Dyke D-14

Panorama of excavation to sound rock, removing all sensitive clay at 32m high Eastmain
Dyke 24-A.

The above panorama clearly illustrates the concerns that Hydro Quebec have for
building a large dam on sensitive soils. The cost of cofferdams, pumping out a lake, excavation

5
of the sensitive clay down to bedrock, and re-shaping the rock abutments, was more than the
moneys spent on building the 24-A dyke.

A comparison pf the geotechnical data for the foundation clay at the North Spur and
Dyke 24-A follows.

Property <------------ Dyke 24-A --------------> < ------------ North Spur ----------- >
Range Average Range Average
min/max min/max
Plastic limit 16.5 / 33 24.2 13/27 21
Liquid limit 30.0/ 54.0 37.6 22/48 37
Plasticity index 7.0/21.0 13.5 7/25 16
Liquidity index 0.7/2.5 1.4 0.1/2 0.6
Water content 24/71 42 17/25 29

Data in the above table shows that the foundation sensitive clays at Dyke 24-A and the
North Spur are similar. But Hydro Quebec elected to remove all the Clay in the foundation of
the dyke.

Section through the North Dyke at LG-1. Note very large berms.
Source SEBJ/HQ: Le Complexe Hydrolectrique de La Grande Rivire deuxime phase.

Another indication of the questionable design of the North Spur, is the dam base width
(including berms) to height ratio compared to other dams on sensitive clay. This ratio is Waba
18:1, North Dyke 24:1, Gardiner Dam 23:1, and the North Spur at 11:1. This indicates that the
base width of the North Spur is about half of what it should be.

From the foregoing, it is obvious that the referenced dykes cannot be used as a
precedent for the North Spur natural dam, for three reasons

6
 1 The North Spur has a height more than 4 times the height of the referenced dykes.
2 All the high dykes (over 11m) on the James Bay development with sensitive clay
foundations, had the sensitive clay removed down to bedrock prior to building the dam.
3 All the James Bay low dykes have homogeneous sections of clay core, wide sand filters
and gravel shoulders, with very wide berms on both sides, with the berm width being
more than twice the dam height.

On the other hand, the North Spur has - (1) A height of over 50m. (2) Sensitive clay in
the foundation. (3) A natural heterogeneous mixture of sand, silt and clay within the structure,
and (4) No berms.

Our statement to the effect that this is the first use of a natural dam in a hydro facility
is an accurate statement, given that the examples cited by Nalcor are not comparable to the
North Spur.

2. Dam safety factors.

2-A issue

The recent comment by Phil Helwig in the use of the same safety criteria as a dam
constructed with selected materials all compacted and tested to CDA standards is very
pertinent. The natural dam materials have not been compacted nor tested sufficiently to
warrant the same safety factor - it should be higher.

2-B Nalcor response

The safety factor required for a natural dam is not different from a constructed dam.
Safety factors are calculated based on the properties of the soils and on the conditions at the
site whether the slopes are natural or part of a dam being constructed. The Engineer of Record
has the responsibility to define these parameters accordingly with the state of the art to make
sure results and evaluation of the stability are in accordance with the recognized
practice. Safety factors in geotechnical industry are calculated based in the physical and
mechanical properties of the soils/embankments and in the hydrogeological conditions of the
site/embankment. A model of the structure is then prepared and safety factors are
calculated. This was the case for determining stability of the North Spur and is used for
determining stability 2 of constructed embankment structures, like dams, or determining if work
is required to protect or stabilize natural slopes.

7
 2-C Comment on response
There has to be a difference in the safety factor in natural dams, simply because the dam
is formed with a heterogeneous mixture of sand, silt and clay, as opposed to a constructed dam
with homogeneous sections of carefully placed and compacted gravel, sand, and clay. It is
unfortunate that Nalcor engineers apparently do not appreciate the difference.
The factors of safety recommended by the CDA are for conventional earth dam
construction and assume normal quality control for the selection, placement and compaction
of dam earthworks. In the case of the North Spur there is much more uncertainty about the
properties of in-situ materials that comprise the body of the "dam"; therefore more
conservative factors of safety should be applied in the design.

Our statement that the safety factor for a natural dam should be higher than for an
engineered dam remains accurate.

3. CDA Dam safety guidelines.

3-A Issue

The Canadian Dam Association was contacted to comment on the application of their
guidelines to a natural dam. They replied within 48 hours advising that at a directors meeting
over the phone they had come to the conclusion that the CDA publications are Guidelines
only, to be interpreted as required by geotechnical engineers. This means the guidelines may
not be relevant to the North Spur.

3-B Nalcor response.

The CDA guidelines include the reservoir rim (or edge) in the overall dam system. The North
Spur natural dam is part of the reservoir rim and thus considered in the CDA guidelines. As per
the CDA 2007 Dam Safety Guidelines the reservoir rim is included in the overall dam system. The
North Spur natural dam is part of the reservoir rim and thus is considered by the CDA guidelines.
The North Spur has been correctly considered as a dam and analyses performed have
demonstrated that the North Spur will remain stable under all hazards and loading conditions.

3-C Comment on response.

There is a difference between the reservoir rim and a dam. If there is a landslide in the
rim, there is still higher ground available to contain the reservoir. This is not the case for a dam.
8
 The CDA standards are simply guidelines and their relevance to the North Spur is
questionable because the Spur is much more than a rim.

4. Sloping foundation.

4-A Issue.

The North Spur strata are all sloping slightly downstream. The clay surfaces slope
downstream. Water will accumulate on the top of the clay strata, reducing the friction and
inducing a slide. This is why Phil Helwig's comments are so pertinent. A constructed dam does
not have sloping downstream surfaces.

4-B Nalcor response.

The construction work at the North Spur addresses this concern through the construction
of filters and drains so that water does not accumulate. It was demonstrated during the
investigations and during the construction works that the stratigraphy of the North Spur is highly
complex, it slopes in different directions in different areas and the strata are sometimes not
continuous from the upstream to the downstream. In addition, the stratified drift deposit in the
North Spur is already saturated.
Regarding potential induced slides it is important to note that the important factor in this
case is the seepage forces (gradient) acting inside the dam which are controlled by the inverted
filters and are therefore drastically reduced in the body of the North Spur. Dams are constructed
with filter systems capable of dealing with seepage forces which is also the case in the design of
the North Spur.

4-3 Comments on response.

The difference is the large horizontal force induced by the full reservoir acting on a
natural dam with a downward sloping sensitive clay foundation. This issue has not been
addressed in the Nalcor analysis. It is essential to undertake the stability analysis for the worst
case scenario. Drains cannot be relied upon to remove all water at the sand/clay interface.
The LEM analysis was used to determine the stresses in the lower clay foundation when
the Spur was subjected to the horizontal reservoir load. However, the section used assumed
horizontal strata.

Our comment on the downward sloping surfaces at the North Spur remains valid.
9
 5. Dr. Bernander methodology.

5-A Issue.

Experts such as Dr. Bernander, a professor at Lule Technical University, Lule, (Sweden),
with extensive experience investigating quick clay slides, has criticised the design, but has been
dismissed almost as a crackpot for questioning the design. Instead, his opinions should be
thoroughly investigated to determine whether his different approach to the calculation of the
dam stability factor is correct, requiring a different design or whether it verifies the current
design.

5-B Nalcor response.

Dr. Bernanders opinions about the North Spur have been taken into consideration and
the type of analysis he recommended was done with the assistance of a worldwide recognized
expert in landslide on sensitive clay. This expert, Dr. Leroueil, is very familiar with Dr. Bernanders
approach and methodology, having coauthor papers on sensitive clays with Dr. Bernander
("Progressive failures in eastern Canadian and Scandinavian sensitive clays" and "Study of a
lateral spread failure in an eastern Canada clay deposit in relation with progressive failure: The
Saint BarnabeNord slide."). The expert undertook a thorough review of all information and data
available and reviewed all the design criteria. He also looked at the approach and to the design
philosophy adopted for the stabilization work.
Bank stability studies of the Muskrat Falls reservoir were performed in 2008 and 2010
(LCPAM CD0000EARP000301, Bank Stability Study for the Proposed Lower Churchill
Hydroelectrical Generation Project, June 2008 and LCPAMCD2110CVRP000101, Bank
Stability and Fish Habitat, 2010 Field Investigation Reports. Vol 1 Bank Stability Assessment,
June 2011).
The Lower Churchill River valley shows numerous landslides scarps suggesting a flowslide
type. To address the potential for occurrence of progressive failure landslides (both downward
and upward) a specialized study was performed (MFASNCD2800GTRP000101, North Spur
Stabilization Works Progressive Failure Study, December 2015). The reports referenced above
are available on the Muskrat Falls website:
https://muskratfalls.nalcorenergy.com/newsroom/reports/ It is understood that Dr.
Bernander has based his comments on his 40 years of experience in sensitive clay in Sweden. He
has also stated that he did not go through all the data available and his calculations did not
consider it. The design team is very experienced and there are engineers who also have more
than 40 years of experience in sensitive clay, both in research and in construction. These
10
engineers performed the investigations in 2010 and in 2013, made the calculations, and
prepared the technical specification and drawings.

5-C Comment on response.

There is no mention of the type of analysis recently undertaken by Mr. Dury, where the
Bernander methodology was used to determine the safety factor in the North Spur, resulting in
a factor of 0.38, indicating that the North Spur will fail when subjected to the reservoir pressure.
It is recognised that the Dury approach was not available during the design, since it has only
recently (June 2017) been available.
The Bernander methodology looks at the stability of the dam mass when subjected to
the large horizontal force from the reservoir water.
A review of the report North Spur Stabilization Works Progressive Failure Study 21
Dec. 2015 indicates that the LEM analysis was attempted for the landslides (Page 8). Dr.
Bernander has shown that this type of analysis will produce incorrect results. Also, the analysis
was confined to the upstream and downstream slopes using a methodology developed by Dr.
Locat which is similar to that used by Dr. Bernander. Again, the analysis was applied exclusively
to the upstream and downstream slopes.
The shear strength has only been determined for the weight of the Spur as shown in
Figure 7-10, page 61. Where the reservoir effect has been added, the slip surface is assumed to
be horizontal, not inclined downstream, as shown in Figure 7-11, page 62.
Mr. Dury has shown that using a development of the Bernander methodology on the
entire Spur section, results in a safety factor well below 1.0, indicating failure of the Spur. This
type of analysis was not undertaken by Nalcor. This contradicts any analysis undertaken with
the help of Dr. Leroueil, and should be investigated further.

In conclusion, the analysis needs to be updated using Bernanders dynamic approach

to non-linearity in estimating stability and safety factors. Dr. Bernanders comments on the
need for redesign should be explored.

6. Cut-off walls.

6-A Issue.

Dr. Bernander has indicated that the use of cut-off walls where the horizontal force will
be concentrated, may have a detrimental effect on the stability. This has not been investigated.
11
 6-B Nalcor response.

The presence of the cutoff walls and their implication during and after the stabilization
works has been thoroughly studied as part of the design. Factors of safety for stability greater
than those recommended by the Canadian Dam Association were obtained. During the
construction work, no instability occurred.

6-C Comments on response.

Cut-off walls are commonly used to seal off a pervious section. The type of wall used at
the North Spur usually has many defects due to sand sloughing off the vertical wall down into
the impervious material being used to fill the wall excavation. There is no method of detecting
this deficiency during construction. The deficiency can be countered by building 2 walls, or fully
grouting the area before constructing the wall.
The cut-off wall concentrates the reservoir horizontal force to a section within the dam.
This becomes an issue when the Bernander methodology is applied to the North Spur.

Application of Bernanders analysis is essential if the most accurate estimates are to

be obtained. Bernanders comments about cut-off walls need to be investigated.

7. Steep upstream slope on North Spur.

7-A Issue.

The upstream slope is too steep as demonstrated by the "slips" detected during
construction. At 2.5:1, it has the same slope as the downstream face, but there are no clear
photos of the downstream face to determine whether it also has "slips". Yes, the lower half of
the upstream face is flatter at 3:1, but it is not known whether this is flat enough.

7-B Nalcor response.

Its important to note that no slips occurred during construction. During construction,
there were no landslide events that could be referred to as slips. There was minor surficial
movement of material and erosion in areas that had not been stabilized yet, both of these are
normal occurrences during any construction project and was addressed during construction.
For clarification on the upstream geometry and slopes and the associated safety factors
for the different loading conditions of the structure, typical crosssections of the stabilization

12
works are presented in document MFASNCD2800GTRP000401, North Spur Stabilization
Works Design Report, January, 2016. IFC drawings also show the different crosssections.
Below elevation 29.0 m the final slope of the works is 2.5H:1.0V, above this elevation up
to elevation 43.0 m the slopes vary between 2.5 to 4.5 H:1.0V, finally between elevation 43.0 m
and the top of the North Spur the slopes were regraded to 2.0H:1.0V.
Stability analysis performed for the different load conditions (temporary excavations,
end of construction, partial pool, steady state at full supply level static and dynamic conditions,
rapid drawdown static and dynamic conditions and permanent slopes static and dynamic
conditions) shown that the slopes as designed are stable and safe and comply with the safety
factors suggested by the Canadian Dam Association (CDA).

7-C Comments on response.

Source Nalcor photo gallery, 2016.

The yellow arrow in the above photo clearly shows a vertical scarp
where the slip has commenced.

The foregoing photo taken during construction of the re-shaping work on the upstream
side, apparently shows several slips as indicated in the photo. They appear to occur in the
2.5:1 slopes. Red arrows show locations of slips. The two slips on the left show where the soil
has slipped down onto the roadway berm, with a shovel at work removing the slide material.
The same effect can be seen, but to a lesser extent on the two right hand slips. The slips have
all occurred on the flattened upstream slope during construction.

This needs to be investigated further.

13
 8. Back calculations.
8-A Issue.

The "slips" should have been used to back-calculate the average friction angle used in
the dam design. It is suspected that it will result in a lower friction, requiring flatter slopes.

8-B Nalcor response.

Previous landslides were used to backcalculate how flat the slopes would need to be to
prevent landslides from occurring. The last major landslide at the North Spur occurred in 1978
and information from that landslide was used in the analysis to determine how flat the slopes
would have to be. The landslide from 1978 was used during design works to calibrate the
selected parameter. The calibrated properties were used in all the analyses performed in the
North Spur.
Backanalyses were performed using this information in representative upstream and
downstream slopes of the North Spur to confirm the measured and interpreted physical and
mechanical properties of the soils and the hydrogeological conditions prevailing in the North
Spur.

8-C Comments on response.

No back-calculations were made for the upstream slips encountered during

construction, of the North Spur upstream slope, only for the landslides in the Spur. The point
here, is that using the data from the landslide back-calculation, the upstream slope still
contained slips in a slope of 2.5:1, indicating that the data used in the LEM calculation was not
sufficiently conservative.

We continue to believe that flatter slopes are required.

9. Downstream erosion hole.

9-A Issue.

The west slope in the deep hole downstream of the Spur is very steep, at a slope of 1.5:1,
is steeper than the downstream natural slopes in the Spur where the factor of safety is about

14
1.0. No geotechnical analysis of this West slope has been undertaken. Any failure in this slope is
likely to migrate upstream into the Spur.

9-B Nalcor response.

The deep hole underwater at the downstream side of the North Spur was considered and
analyzed and found to be stable. No stabilization work was required in this area. Stability
analyses and assessment of potential progressive failure were performed for a crosssection
through the depression in the downstream side of the North Spur. Results of these analyses
showed that the slope was stable and safe for the different loading conditions and the
stabilization works are adequate to avoid a potential progressive failure.

9-C Comments on response.

The analysis of the erosion hole stability is shown in Figure 7-13, page 65 of the 2015
report. Shear stress analysis showed that stress is well within the acceptable limits. However,
when the sensitivity of the analysis was tested by assuming no load from the Spur, there was no
change in the stress level. Report Section 7.8, page 64. This is very strange, perhaps indicating a
flaw in the methodology.
The following figure shows an alternative analysis of the erosion hole stability. As with
all landslide scarps, the slope of the scarp is at a maximum near the top, decreasing at lower
levels. A close examination of the scarp slope, indicates that it is very steep near the top, with a
horizontal distance of only 25m for the top 20m, resulting in a slope of 20/25 = 80%. This chart
from Scott clearly indicates that such a slope is unstable.

The 2015 report on Progressive Failures, page 27, contains the following
Flowslides in sensitive clay result from a succession of slides. There must be an initial
slide. If the .. slump is large enough to remould the clay, and flows out of the crater if the
liquidity index is large enough, or the remoulded shear strength is low enough. The backscarp
stands . Without being supported by debris. If this backscarp is unstable . There will be
another failure instantaneously or after some time . And the process may go on and on. . This
is possible when . (there is) a liquidity index greater than 1.2

Nalcor has reported that 123 tests of the lower clay show a liquidity index ranging from
0.1 to 2.0 with an average of 0.6. Nalcor uses the average liquidity index of 0.6 to conclude that
a flow slide will not occur. This does not follow the principle that a dam must be designed for
the worst case conditions. The analysis of this is shown in the second following figure.
15
Source for the Height - inclination relationship graph for unstable slopes in eastern Canada is
Scott, J.S., 2003, Fig. 26, page 58 [ https://books.google.ca/books?

It is interesting to note that the upper clay layers have a liquidity index varying from 0.6
up to 2.8 with an average of 1.3, or just above the value where a flow slide can occur. (2016
report, page 87). This has not been discussed in any of the reports by Nalcor.

16
 In the latest engineering study, only one section through the North Spur shows the deep
hole, namely Section HH on page 75 of the 2016 report. Unfortunately, the steep upper slope
of the scarp does not appear, probably due to the small scale of the section, hence the inherent
instability of the scarp is not clearly evident.
Please provide a copy of the geotechnical analysis of the West Slope of the deep hole
with the steeper slope entered into the calculations.

17
 10. Steep upstream slope.

10-A Issue.

Below the low-drawdown reservoir level, the upstream slope in the re-shaped North
Spur increases from 3:1 to 2.5:1, where it is not necessary to design for a "rapid drawdown".
This indicates that the designer lacks experience - a steeper lower slope is never used - it reduces
the safety factor.

10-B Nalcor response.

Its important to note that the slope below elevation 29.0m is designed for the
construction condition with the water level at elevation 25.0m (Winter Headpond). Above
elevation 29.0m, the design is for the final reservoir level at elevation 39.0m, and a flatter slope
is required.
The embankment below elevation 29.0m (final slope of 2.5H:1.0V) was built in order to
counteract uplift water pressure during the construction works due to the nature of the
foundation behind the till embankment. This feature has driven the decision to build the
embankment with two different slopes. The slope of the till embankment, which has a greater
impact in this loading condition doesnt change.
The calculated factor of safety of the stabilization works shows that the designed slopes
are stable and safe for the rapid drawdown condition. For clarification, the stabilization works
and all construction dam works in general are planned taking into account all load conditions,
construction, shortterm and longterm conditions. The North Spur Stabilization Works design
took into account all these scenarios. This is outlined in document MFASNCD2800GTRP
000401, North Spur Stabilization Works Design Report, January, 2016.

10-C Comments on response.

With respect to the steeper slope in the upstream face below El. 25, there is no known
precedent for an embankment dam having steeper slopes in the lower level. This reduces the
dam slope safety factor.
The response that The embankment below elevation 29.0m (final slope of 2.5H:1.0V)
was built in order to counteract uplift water pressure during the construction works due to the
nature of the foundation behind the till embankment, below El. 29, the soil behind the till
embankment is mostly stratified drift (Figure 4-43, page 142, 2016 report) or sensitive clay. This
would require a flatter slope, not steeper.

The lower steep slope needs to be flattened to correspond with the slope above.
18
 11. Pump tests.

11-A Issue.

There is no mention of any pump tests to determine whether the lower aquifer is not
connected to the upstream reservoir. In at least one Upstream-Downstream cross section
through the Spur, the lower aquifer slopes upward in the upstream direction, indicating that a
connection to the upstream river water is likely. Several pump tests at different locations are
required to verify the assumed lack of a connection. It is essential for the safety of the dam that
there is no connection.

11-B Nalcor response.

For clarification, pump test were completed in previous studies and the results were
integrated into the 3D hydrogeologic model which was specially developed for the purpose of
examining the behaviour of the lower and upper aquifers.
Potential connection between the lower aquifer and the upstream reservoir was taken
into account during the stabilization works design. Instrumentation and stabilization measures
are already in place to counteract this potential feature.

11-C Comments on response.

The interpretation of pump tests is highly unreliable, with many tests required to
determine the average seepage coefficient. How many tests were undertaken?

Please confirm that there is no connection between the upstream reservoir and the
lower aquifer.

12. Core barrel drop.

12-A Issue.

The AMEC report on the test holes undertaken by a vibratory drill across the North Spur
indicated that the drill casing dropped under its own weight through very soft clay in several
holes. This is very troubling. Yet, there is no mention of the implications in the dam design
document prepared by SNC.

19
 12-B Nalcor response.

In a deep investigation holes in sensitive clays it is common to locally exceed the bearing
capacity with the casing or with any metal stem. Experienced engineers undertaking
investigation done in sensitive clays are aware of this and recommend that measures be
undertaken to retain the casing and rod so that this does not occur.
The Geotechnical Investigation Report, 2013 Field Investigations (MFAAMSD2800GT
A99 000201, November 2013) reports that only in borehole NS0213 the core barrel dropped
two times over a depth varying between 3 and 5 m and in borehole NS0913 the Shelby tube
dropped 9 cm.

12-C Comment on response.

Does the dropping of the drill casing indicate the need for more testing in the areas
where the dropping occurred?

13. Geotechnical investigations.

13-A Issue

The geotechnical investigation relies, in part, on past data extending back to 1965. This
data should have been verified since test equipment and methodologies may have changed over
the last 50 years. Also, according to Dr. Bernander, several types of tests were not undertaken,
all necessary in determining dam safety.

13-B Nalcor response

For clarification, it is the North Spur design, not the geotechnical investigation, which
relies on data dating back to 1965. The design relies on all the geotechnical, geological and
hydrogeological data collected over the years since 1965 up to and including the current day.
Equipment, methodologies of investigation and tests were performed with the latest technology
available at the time of the investigation campaigns and in accordance with the available state
oftheart for this type of activities and for this type of industry at the moment of the
investigations. All data has been reviewed for relevance and reliability for use in the final
design.

20
 13-C Comments on response

Sufficient geotechnical investigations may have been undertaken for a LEM analysis.
However, the LEM analysis assumes a linear relationship between stress and strain in the clay.
According to Dr. Bernander, the characteristics of the clay indicate that this is unlikely, and that
the linear relationship breaks down at lower strains than assumed in the LEM analysis.

Further geotechnical tests are required as recommended by Dr. Bernander.

14. Liquidity index shear strength anomaly.

14-A Issue

In the geotechnical data there is an anomaly in the relationship between tested shear
strength and the liquidity index, as discovered by Maurice Adams. This relationship is well
outside the normal range, indicating that one or the other is incorrect. This anomaly needs to
be investigated since any incorrect data will affect the dam safety. If the liquidity index is correct,
then the shear strength should be lower by about one magnitude. If the shear strength is
correct, then the liquidity index is too high by about one magnitude.

14-B Nalcor response

The writer mentions that the value coming from test results on the North Spur clay are
not in agreement with a relation between the liquidity index and the undrained shear strength
resistance. His conclusion is that there is a mistake in the result for one or the other parameter.
The writer refers to values of intact undrained shear strength of the clay. The relation
between the Liquidity Index and the Undrained Remolded Shear Strength Resistance has been
known for many years (Leroueil, 1983; Demers and Locat, 1988; and others). This relation is valid
for the remolded clay, not for the intact clay.
The results of the tests performed in the clay are in agreement with the relation proposed
in the literature. These results were taken into account in the design of the stabilization works of
the North Spur.
Locat, J. and D. Demers, 1988. "Viscosity, yield stress, remolded strength, and liquidity
index relationships for sensitive clays." Can. Geotech. J., 25, 799806 (1988).
Leroueil, S., F. Tavenas and JP Le Bihan, 1983. "Proprietes caracteristiques des argiles de
l'est du Canada." Can. Geotech. J., 20, 681705 (1983). ("Characteristic properties of the clays of
eastern Canada.")

21
 14-C Comments on response.
Nalcor's Response states that the "The relation between the Liquidity Index and the
Undrained Remolded Shear Strength Resistance ... (as shown by Lerouell, Demers and Locat and
others) ... is valid for the remolded clay...".
Refer Nalcor to Fig. "A" (Source: http://www.vision2041.com/north-spur.html ). This
shows that when Nalcor's reported lower clay layer Liquidity Index values are above the lower
liquidity index boundary of 0.8 (and up to 2.0) or when remolded shear strength values are
below 1 kPa, flowslides (successive slides/multiple surface failures) can occur, and that Liquidity
Index values at 2.0 should have a corresponding remolded shear strength ranging from about
0.1 - 0.3 kPa, about 10 to 25 times lower than Nalcor's reported lowest lower clay remolded
shear strength value of 8 kPa (see graphic below):

A review of the liquidity-undrained shear strength is required.

22
 15. Previous designs.
15-A Issue

There have been at least three designs for the dam, including a downstream groin
extending out into the water and infilling of the deep downstream hole. Why should the current
design not address these issues and particularly the slope of the Spur as it extends to the bottom
of the hole.

15-B Nalcor response

All the engineering studies performed since 1965 and their evolution was reviewed by
the design team. The design evolved as more information became available. All previous
designs are presented in the document MFASNCD2800GTRP000401, North Spur
Stabilization Works Design Report, January, 2016.
Outcomes of the field works and interpretations presented in the different engineering
studies shaped the basis of the current design. The final design took into account the basis and
assumptions for the previous designs and all the geotechnical information collected.

15-C Comment on response

Previous studies have recommended filling in the hole (1965, page 20 in 2016 report) or
building a groin to counter toe erosion (1977 study, page 27 in 2016 report).
In the latest studies, there is no mention of any remedial work for the hole despite the
fact that the top portion of the scarp is unstable and could cause a succession of flow slides
reaching upstream into the Spur.
The extent of downstream erosion is clearly indicated in the disappearance due to
erosion of the downstream debris from the 1979 landslide as shown in Figure 4-39 on page 133
of the 2016 report. In the current design, the erosion is countered by a small berm with rip-rap
on the downstream face. This is totally inadequate, since erosion will continue and undermine
the berm.

Further study of the erosion hole stability and the effect of erosion on the downstream
shoreline of the Spur is required.

16. Hydrogeological model.

16-A Issue

Acres, now Hatch, have reviewed the geotechnical design. Their main concern was found
to be the natural water table level as it changes with the introduction of the upstream reservoir.
23
Apparently, the water table affects the stability more than expected and Hatch had to develop
a 3D computer model to calculate the new water table levels. This effect of the water level was
verified by Dr. Bernander in his report. The water table problem was so complex, that the work
was undertaken by geotechnical engineers with postgraduate degrees. There has been no
review of this very complex work.

16-B Nalcor response

All recommendations issued from the Cold Eyes Review performed by Hatch were
addressed. As part of these recommendations, a 3D Hydrogeological Study was performed
(MFAHECD2800GTRP000301, Three Dimension (3D) Hydrogeological Study for the North
Spur, October 2015).
The assumption regarding effect of water table level changing the stability more than
expected and as a consequence a 3D hydrogeological model was developed is not
correct. Ground water pressure is very important in soil mechanics and its impact on stability is
recognized by experienced geotechnical engineers. The 3D hydrogeologic model was developed
to help the designer follow the impact of construction and impoundment on the stability of the
North Spur. The model was developed by Hatch with a weekly supervision of the design team.

16-C Comment on Nalcor response

This is the first time the hydrogeological model has been used. There has been no testing
of the model on other sites, and it is recognised that such testing would be extremely difficult.
Hence the model must be viewed with suspicion, until such time as the model can be tested on
another development.

The hydrogeological model needs to be tested.

17. Quick clay.

17-A Issue

Quick clay has been detected at two locations on the downstream slope. But the extent
has not been determined. If extensive, it will have serious implications for the dam safety. This
will require more drilling of boreholes.

17-B Nalcor response

The presence of high sensitivity clay in the North Spur was known from the first
investigation works performed in 1965, which found it in some samples coming from two
24
boreholes. All investigation works done since then confirmed the same fact: some samples
coming from the upper clay (part of the stratified drift deposit) show high sensitivity.
The design of the stabilization works was done with consideration of the presence of high
sensitive clays in the North Spur.
During the construction works (2015 and 2016 seasons), more than 1 million m3 of soils
were excavated from the spur, including about 600,000 m3 of soils excavated in the stratified
drift unit. From this volume, less than 0.1% was observed as being very high sensitivity clay and
it occurred in two small, localized areas.
In summary, the presence of sensitive clay was known and expected. The design and
construction, including all the activities, methods and operations during and after construction
have taken this into account.

17-C comments on response

Only 0.01% of the soil removed from the flanks of the Spur was found to be quick clay.
However, this does not exclude the possibility that more and larger deposits will be continue to
be found within the Spur. Also, the presence of quick clay deposits indicates that the clay is very
sensitive.

Further investigation is required for areas of the North Spur where fewer holes have
been bored.

18. Reviews of the North Spur design

18-A Issue

NALCOR insist that the geotechnical design has been reviewed by two independent
experts, namely Dr. Leroueil and Dr. Idriss.
Dr. Leroueil is currently a professor at Laval University in Quebec City. He advised that
he is not qualified to undertake such a review since he is not an expert in dynamic analysis, was
not provided with all the data (no report appendices) and hence confined his review to only
Chapter 2 in the voluminous geotechnical report. His conclusions, on less than two pages, were
that the stabilization works increase the factor of safety from about 1.0 to 1.6, which is very
significant. Not a full endorsement of the design.
Dr. Idriss is a retired earthquake specialist, from the University of California at Davis, who
attended several meetings, made some comments, but never issued a report.

18-B Nalcor response

Technical reviews of the current engineering were performed by the Advisory Board (April
and October 2013), by the Independent Engineer (September 2013) and by Hatch (September
25
2013). During the reviews, additional specialized studies (Dynamic and Hydrogeological) were
recommended by these reviewers.
Additional assessment of the potential liquefaction of granular soils and cyclic softening
of cohesive soils under a seismic event was performed by Drs. Leroueil and Idriss. Results of this
assessment are presented in section 5.12 of the design report (MFASNCD2800GTRP0004
01, North Spur Stabilization Works Design Report, January, 2016).
Dr. Leroueil also participated in the progressive failure study. Results can be found in
section 5.13 of the design report (MFASNCD2800GTRP000401, North Spur Stabilization
Works Design Report, January, 2016).
Geotechnical design was not reviewed by Leroueil and Idriss. These two external experts
were part of the study to ensure the design is in accordance with the expected seismic loading.
Leroueil is a worldwide expert in sensitive clay behaviour and Idriss is the same in impact of
dynamic on the soil behaviour. Both experts gave advice while the study was underway. Idriss
gave a presentation and report during and after the study was completed.

18-C Comments on response

It is noted that neither Dr. Leroueil nor Professor Idriss have reviewed the geotechnical
design. The question then becomes Who has reviewed and approved the design?

Review and approval of the geotechnical design is still required.

19. Approvals of the Nalcor design for the North Spur.

19-A Issue

There is a reference (NALCOR 2016 report on the spur design, page 120) to the
Independent Engineer (MWH) having reviewed and approved the design of the North Spur
quote - The Independent Engineer commented that the stabilization works have been designed
in accordance with currently accepted geotechnical design practices and effectively stabilizes the
North Spur when the reservoir is impounded. end quote. An email to the Independent
engineer asking for confirmation of this statement resulted in the following reply - MWH have
never at any stage been involved in the design of the North Spur. We act as Lenders Engineer
to the Federal Government and have never at any point been actively involved in the design of
any of the project components. This contradiction needs to be resolved.

26
 19-B Nalcor response

It is not the mandate for the Independent Engineer to approve any aspect of the design.
They are engaged by the federal government to review aspects of the Muskrat Falls 10 Project.
They present their comments, observations and recommendations to the federal government.
The provincial government is permitted to rely on same. Nalcor also is privy to these comments,
observations, and recommendations and incorporates that input when and where appropriate.
The Independent Engineer stated on page 9 of in its Interim Independent Engineers
Report Lower Churchill Project, The stabilization works have been designed in accordance with
currently accepted geotechnical design practices and will effectively stabilize the north spur
when the reservoir is impounded. The report is located at the following link:
http://muskratfalls.nalcorenergy.com/wpcontent/uploads/2013/03/November292013
IndependentEngineerReport1.pdf. There is no mention of approval by the Independent
Engineer or by any other expert reviewers or advisors.
For clarification, the reference stated above in the writers reference should be page 210
(Section 5.3 Current Engineering Technical Reviews).

19-C comments on response

There is a vast difference between a review and an approval of the design. As mentioned,
the MWH review merely stated that the design was undertaken in accordance with current
geotechnical design standards. Neither MWH nor Hatch has approved the design.

Who has given unqualified approval to the design of the North Spur?

20. Insurance.

20-A Issue

Finally, there is the question of insurance for the dam. This can be purchased, but if there
is a failure, the insurance company will deny payment since this was a known and unacceptable
risk assumed by NALCOR.

20-B Nalcor response

Construction All Risk as well as Wrap Up Liability and Pollution Liability has been
purchased for the Muskrat Falls Project. Detailed discussions with the current underwriters as
27
well as their engineers were undertaken so they could fully understand the risks associated with
the project including the North Spur before the project even started. The insurance underwriters
made their own assessment of the risk before accepting to underwrite the project. The insurance
companies also continue to monitor the project and have their engineers visit the site twice per
year and have had no issues with the design or construction of the North Spur infrastructure.

20-C Comments on response

The insurance company does not have the level of geotechnical experience to assess the
risk associated with the North Spur design. If a failure occurs, the claim will be denied on the
caveat that the risk was known to Nalcor, as mentioned in the SNC risk report, but not divulged
to the insurance company.

Discussions with the insurance company are required to ascertain that the failure risk
is accepted and covered in the policy.

CONCLUSION

It is essential that the dam design be reviewed by a panel of geotechnical

experts. It is not too late to undertake such a review, since any changes resulting
from the review can still be engineered and constructed. If there are no changes
required, then there is the added assurance that the dam is safe. Since the North
Spur relies on an untested hydrogeological program, and constitutes a major risk
to both life and cost, it is absolutely essential that a geotechnical review of the
design, by an eminent review panel, be commenced immediately.

28