HomeMy WebLinkAboutSuWa278Alaska Resources Library & Information Services
Susitna-Watana Hydroelectric Project Document
ARLIS Uniform Cover Page
Title:
SuWa 278
BOC final report #4 with AEA responses
Author(s) – Personal:
Author(s) – Corporate:
Alaska Energy Authority
AEA-identified category, if specified:
Board of Consultant meeting
AEA-identified series, if specified:
Series (ARLIS-assigned report number): Existing numbers on document:
Susitna-Watana Hydroelectric Project document number 278
Published by: Date published:
[Anchorage, Alaska : Alaska Energy Authority, 2014]
Published for: Date or date range of report:
Volume and/or Part numbers:
Final or Draft status, as indicated:
Attachment
Document type: Pagination:
10 p.
Related work(s): Pages added/changed by ARLIS:
Cover letter: Board of Consultant Meeting #4, Susitna-Watana
Hydroelectric Project P-14241-AK. (SuWa 276)
Two additional attachments (SuWa 277 and 279)
Added cover letter (2 pages)
Notes:
All reports in the Susitna-Watana Hydroelectric Project Document series include an ARLIS-
produced cover page and an ARLIS-assigned number for uniformity and citability. All reports
are posted online at http://www.arlis.org/resources/susitna-watana/
BOC Final Report #4 with AEA Responses
AEA Questions , BOC Responses and AEA Comments
1. AEA Question 1: Does the BOC agree that the configuration of the dam is acceptable as a
basis for further design evaluation and optimization (and license application), with the
proviso that the dynamic analysis be revised with foundation mass etc. and (results of) Site
Specific Seismic Hazard Analysis studies, (and) site investigation (for the configuration)?
BOC Response to Question 1: The BOC agrees that if the result of the planned site
investigation program is positive, i.e., that there are no shears, linear features or faults found
that can negatively affect the performance of the dam, the configuration of the dam would be
acceptable as a basis for further design evaluation, analysis and license application. The
BOC feels that confirmation of these site conditions are paramount to the feasibility and
therefore encourages early completion of the planned foundation investigation. The Site
Specific Seismic Hazard Analysis (SSSHA) studies that are being conducted and presented
are reasonable and appropriate for determining project feasibility and to serve as a basis for
further design evaluation and optimization.
The various dynamic analyses need to be further investigated for the Operating Basis
Earthquake (OBE) to show that operation can continue without interruption after an OBE
event. Analysis for the MCE should be conducted as a limit case and evaluated for overall
stability to ensure that the dam is stable during and after the MCE without sudden and
uncontrolled release of the reservoir. Some possible damage and small displacement is
acceptable for the MCE case as long as the water retention capability of the dam is
maintained.
AEA Comment:
There has always been the intention to perform a dynamic structural analysis of the
dam for both the OBE and the MCE. However, using the level of OBE suggested by
the various guidelines in use in the US indicates an event that can “reasonably be
expected to occur within the service life of the project, that is, with a 50-percent probability
of exceedence during the service life”. This would be about a 144-yr occurrence which
is less than 0.25g. AEA will discuss and may choose an OBE that is higher. In any
event, the behavior of the structure under the MCE will essentially govern the
feasibility design of the dam, so AEA has chosen to focus on that criteria for the time
being. The analyses will be performed for that seismic criteria and when the final
configuration has been derived, it will be subject to the chosen OBE - and is expected to
perform satisfactorily.
With respect to follow up dynamic analyses, the BOC suggests the following:
It is important that dynamic analysis with massed foundation uses appropriate
foundation modulus consistent with rock properties at the dam site. A low foundation
modulus may be assigned to a narrow strip of elements along the footprint of the dam
to account for rock fractures and joints that may exist in shallow depths, but a higher
deformation modulus of the rock should be used in the rest of the foundation.
Page 2 of 10
AEA Comment:
Noted.
Appropriate transmitting or non-reflecting boundary conditions should be applied to
the bottom and sides of the foundation model to eliminate reflection of seismic waves
at the boundaries of the model.
AEA Comment:
Intend to include.
Transmitting boundaries (dampers) with no constraint do not permit acceleration time
histories as the seismic input. The ground-surface acceleration time histories
therefore should first be deconvolved and then converted to stress time histories and
applied to bottom and sides of the model. It is important that the deconvolution and
conversion to stress time histories are verified to ensure that they produce similar
ground surface acceleration when applied to a foundation block with the dam in
place.
AEA Comment:
Intend to do this. However, there is, we believe a typographical error. In the last line it
says “block with the dam in place” this seems wrong and that it should be “block
without dam in place”.
2. AEA Question 2: Does the BOC agree that the Site Specific Seismic Hazard Analysis
(SSSHA) studies performed to date are acceptable with the proviso that further crustal
lineament analysis and angled drill holes across the valley under the dam foundation (2014-
15 field program) be completed before final seismic criteria can be verified for final design?
BOC Response to Question 2:
The SSSHA studies accomplished to date cover lineament, probabilistic, and deterministic
seismic hazard analyses and are detailed enough to provide preliminary design ground
motions for the project. Sensitivity analyses show that further analysis of regional crustal
lineaments will not significantly impact the design ground motions in the period range of
significant dam response. However, further analysis of site area lineaments and angled drill
holes are necessary to assess potential for fault rupture under the dam as a result of primary,
secondary, or sympathetic fault displacement, as discussed in more detail under “Additional
Considerations” section below. On this basis, the BOC believes that the main purpose of
further lineament analysis at the dam site including angled drill holes should focus on
assessment of potential fault displacement and not necessarily on the vibratory ground
motion associated with the lineaments. From the BOC perspective the probabilistic and
deterministic ground motions are acceptable and can be finalized giving consideration to the
following:
Based on preliminary measurements, Vs30 at the dam site could be as high as 2,000 m/s,
which is significantly higher than the current value considered (1,080 m/s). The BOC
Page 3 of 10
recommends that ground motions be estimated for Vs30 consistent with the rock
properties at the dam, and if necessary be adjusted for the effect of Kappa (effect of
upper crust damping).
AEA Comment:
Noted. In the absence of any actual site testing, we intend to run the analysis using a
Vs30 that is conservative with respect to ground motions.
The BOC is generally satisfied with the deterministic estimates of Mmax (i.e., 7.5, 7.8,
and 8.0) and the associated level of ground motions (i.e., 84th% for 7.5 and 69th % for
M7.8 and M8.0) for the intraslab events. However, the BOC considers that the
magnitude of 7.5 is a more defensible MCE for the dimensions and historical activity of
the McKinley Block beneath the site; and recommends that a Mmax of 7.5 at the closest
hypocentral distance and 84th percentile deterministic ground motion (pga of 0.76g) be
adopted for the feasibility/design (see “Additional Considerations”).
AEA Comment:
Noted.
An ANSYS dynamic analysis of Layout-4 indicates a fundamental period of vibration of
about 0.55 seconds with sliding and permanent displacements under the MCE ground
motion (intraslab M7.5 at 84th percentile, equivalent to a return period of 5,000 years).
The nonlinear response of the dam is therefore expected to be sensitive to long-period
and long-duration ground motion, typical of interface events. As such, the BOC
recommends that in addition to the intraslab ground motion, the dam also be analyzed for
the interface deterministic ground motion scaled to the 5000-year UHS at the
fundamental period of the dam (0.55 sec).
AEA Comment:
Noted.
3. AEA Question 3: Does the BOC agree that the draft PMP/PMF studies – prior to
completing the report – are acceptable for finalizing the feasibility design and that if there
are no changes in conclusion during the finalization of the report that the conclusions can be
used for the final design of the spillway?
BOC Response to Question 3:
3.1 Probable Maximum Precipitation (PMP)
Applied Weather Associates (AWA) has made subtle but very significant changes to their
modeling approach, many of them based on suggestions from the BOC. It is the BOC’s
opinion that the current configuration of the model, and the results obtained, represent an
accurate, consistent, and defensible estimate of PMP as it would affect the spillway design.
Page 4 of 10
While there are some details which have been identified for modification in the PMP reports,
these are mostly minor clarifications and grammar suggestions. The PMP process itself, and
the resultant rainfall estimates, are excellent.
The most significant change in modeling procedure is in the storm transposition process.
AWA is using a procedure which normalizes precipitation using a comparison with an
extreme precipitation coverage (such as 100-year grids from NOAA Atlas 14). Then this
normalized grid is transposed to the target watershed and multiplied by the extreme coverage
values at the target. The result is a process very similar to the so-called “isopercental”
method, which is known to work well in areas with complex terrain. We believe that the
storm transposition approach currently used by AWA is the most consistent and defensible
method available.
The BOC recommends that report be edited for consistency and clarification prior to final
submittal.
AEA Comment:
Noted. Final editing and review of the PMP report will be performed.
3.2 Probable Maximum Flood (PMF)
The BOC agrees that the PMF inflow hydrograph presented in the draft PMF report
represents generally appropriate assumptions and modeling methodologies and the study is
sufficiently complete to be used in feasibility design. The BOC commends the
thoughtfulness and level of detail shown in the work to date, considering the unusual
challenges posed by two factors in particular: the sparseness of historical
hydrometeorological data and the dominance of snowmelt in many historical and
hypothetical floods.
However, The BOC recommends that additional analyses and investigations be completed
before adopting a final design inflow hydrograph, as follows:
The constant loss rate of 0.02 inch per hour applied to develop the PMF was less than the
calibrated loss function (that is, it resulted in more computed runoff) and was chosen to
provide an additional level of conservatism. The BOC concurs with this decision because
the original loss function calibration for snowmelt-impacted events is somewhat unreliable,
requiring an increase in the precipitation inputs relative to the values provided by Applied
Weather Associates (AWA) for the calibration events. When greater-than-observed rainfall
has to be added to the model in order to achieve the observed runoff volume, the other
inputs affecting volume become very questionable. In fact, the calibrations were not able to
conclusively prove that the loss rate was any greater than zero. In the presentation to the
BOC on April 3, MWH’s hydrologist noted that they had begun the process of applying the
adopted PMF loss rate to the calibration events with generally acceptable results. The BOC
strongly supports this step – along with the elimination of any arbitrary adjustments to the
AWA-provided rainfall data – for the sake of consistency and credibility. The BOC would
view a less perfect fit to the calibration events to be an acceptable sacrifice in order to avoid
adjusting the rainfall data.
Page 5 of 10
AEA Comment:
It is expected that the analysis of a constant loss rate of 0.02 inch per hour for the
spring calibration and verification floods will be added to the PMF report in either a
new or a revised report section.
If the model continues to underestimate snowmelt-impacted events, consider the possibility
that the energy budget loss method is not adequately considering the release of free water
from a compacted snowpack. One reference addressing this condition is the1966 Bureau of
Reclamation Engineering Monograph No. 35.
AEA Comment:
It is anticipated that the energy budget snowmelt method (as recommended in FERC
guidelines) will prove adequate for simulation of historic snowmelt events. It is noted
that USBR Monograph No. 35 states that it is intended for use in in inflow design flood
studies in which a design rain occurs on a fresh snowpack.
The near-record flood of June 2013 raises the possibility of a “sun-on-snow” PMF. In light
of the fact that the PMP rainfall is relatively small and is associated with temperatures
substantially lower than the temperatures that may occur in late spring/early summer with
no cloud cover, the BOC suggests investigating the snowmelt-only event in at least enough
depth to confirm it cannot control the PMF. This investigation would involve two elements:
o Apply the HEC-1 model to the June 2013 event to confirm that it can replicate this
type of flood;
o Consider whether a probable maximum snowpack combined with unusually high
temperatures, with no rain, could produce a controlling PMF. An efficient approach
might be to make multiple model runs to determine what temperature/wind
combinations would be needed to produce a PMF “contender” and then consult with
AWA to evaluate whether such a combination of circumstances is plausible.
AEA Comment:
Noted. The sun-on-snow PMF will be investigated in sufficient detail to determine
whether it could constitute the controlling case for the PMF, including simulation of the
actual June 2013 flood event.
The sensitivity analysis leading to the adopted PMF model used a June
precipitation/snowmelt combination. However, in the “base case” model runs, the August
PMP with no snowmelt controlled the PMF. The final study should either address making
similar conservative loss rate adjustments to the August case, or explain why the adopted
changes apply to rain-on-snow floods but not to summer floods.
AEA Comment:
Noted.
Page 6 of 10
In the PMF report, Table 9.1-3, run M6 (The October PMP, or 80 percent of the all season
PMP, on an October snowpack accumulation) is listed as resulting in a peak inflow of
24,000 cfs. However, in mid-October 1986 the Gold Creek gage recorded a daily flow of
36,000 cfs. The BOC agrees that this case will not control the PMF. Still, to retain
confidence in the model assumptions the discrepancy between the computed extreme flood
and the observed flood needs to be resolved.
AEA Comment:
This case will be revised to present a more plausible October PMF, probably by
investigating an October 1 PMF condition to replace the October 15 condition.
4. AEA Question 4: Given the configuration presented does the BOC consider that the
planned site investigation is appropriate for the provision of data for feasibility/design of the
dam?
BOC Response to Question 4: The proposed and planned Site Investigation Program for the
project was presented at the meeting. This program entails a phased series of field
investigations, conducted over the next three years that respond to several of the BOC
comments and concerns. The primary focus of the program is to characterize and confirm
the geologic/geotechnical conditions in the dam site area. The main objectives include; 1)
investigation and verification of the fracture and shear zones and geologic features, 2)
evaluation of the potential for offset displacements in the foundation due to earthquake
motions; 3) delineation of the frozen ground and groundwater conditions (adits) and 4)
evaluation of the abutment stability. Given the above plan and objectives, which can be
accomplished by the detailed mapping and exploratory adits and borings planned; the BOC
considers the present detailed and phased site investigation appropriate for developing the
data for supporting the feasibility and design of the dam.
AEA Comment:
Noted.
ADDITIONAL CONSIDERATIONS:
The BOC agrees that the Site Specific Seismic Hazard Analysis (SSSHA) studies performed to date
are acceptable for further design evaluation and optimization with the proviso that the proposed site-
specific studies be performed in the 2014-15 field seasons to evaluate the potential for fault rupture
at the dam site.
Potential seismic hazard at the site includes two components: (1) Strong ground motion
(development of design ground motion criteria), and 2) Fault rupture (documenting the absence of
potential for fault displacement through the dam foundation). Fault displacement includes both
tectonic fault displacement (primary, secondary or sympathetic) and non-tectonic displacement (e.g.,
sachung, etc.).
Considerable work has been performed to date to develop probabilistic and deterministic ground
motion estimates for the dam. As described below, the BOC considers that this work is sufficient to
move forward with further design and optimization studies, and the license application, pending
Page 7 of 10
completion of the final SSSHA studies.
Conversely, given site access limitations, only limited assessment has been performed of the
potential for fault displacement at the dam site. Given the importance of this potential hazard to
selection of the RCC dam design and location, the BOC concurs with the prioritization given to this
assessment in the proposed work scope for 2014 field activities, including detailed field mapping,
angle boreholes, and dating of identified shear zones. We reiterate our BOC comment following
Meeting #2 (Comment 10) “It is recommended that the energy of the geologists and the funding be
focused on the mapping, drilling, and adits at the dam site area in a major effort to define the
geometry of the shears in order to locate the dam such that any offsets occurring along these
features during an earthquake do not need to be considered. This activity must be given the highest
priority compared to the lineament studies at significant distance s from the possible dam site”.
Fault Rupture Hazard. In performing the upcoming 2014 site studies, the BOC recommends that
particular attention be given to the following:
1. The Lineament Study provides an excellent basis for further evaluation of geologic features
that may intersect the dam site area, and provides explicit criteria for “including” or
“excluding” lineaments for further study. Currently, all of the criteria are related to the
identification of potential seismic sources for ground motion analysis. The BOC
recommends that one or more criteria be added related to the assessment of lineaments for
potential fault rupture. For example, all lineaments that project toward the dam site,
regardless of length, should be evaluated for potential fault rupture.
2. During the upcoming 2014 field season, the BOC strongly recommends that explicit
attention be given to the evaluation of the Susitna lineament as a potential fault structure, and
the potential relationship of this lineament to the NW-SE trending shears mapped through
the dam foundation (e.g., as potential Reidel shears). This potential association must be
ruled out either by direct observations such as cross cutting relationships or by indirect
arguments such as expected sense of slip in the current tectonic stress/strain regime.
3. In addition to mapping observed lineaments in the dam site area, the BOC recommends that
a large scale lineament and detailed geologic map of the site area (approximately 1 or 2
kilometer radius) be prepared documenting the “absence” of lineaments or geologic features
that may be associated with tectonic or non-tectonic activity (e.g., sachung features or other
deep-seated sympathetic structures) and explicitly stating in the report that these features are
absent at the site.
4. Considerable “legacy” data exist for the site, including the Woodward Clyde lineament and
trenching study. For example, trenches on the Susitna lineament and Talkeetna lineament are
used to conclude that these features are not active faults. The BOC recommends that the site
report include a section or commentary regarding the use of this pre-existing knowledge for
current evaluation of these features.
Ground Motion Hazard.
(1) Seismic Source Model. The seismic source model developed for the site is well documented
and is appropriate for use on both deterministic and probabilistic analysis of ground motion
at the site. Sensitivity analyses have identified those parameters of the model that are
significant to hazard for further evaluation, including the Mmax distribution on the Intraslab
source and selection/weighting of appropriate GMPE models. Sensitivity analyses show that
crustal faults in the site region do not contribute significantly to hazard. Thus, the BOC
recommends that little further effort be given to characterizing the potential activity of
Page 8 of 10
Lineaments in the site region. As needed, any external lineaments “identified for further
study” in the Lineament Report may be addressed by assigning a probability of being
seismogenic (P(s)), a slip rate based on the threshold of detection, and an Mmax based on
lineament length. Sensitivity analyses have shown that incorporating these lineaments as
seismic sources will not contribute to ground motion hazard at the site.
AEA Comment:
Noted.
(2) Mmax of the Intraslab Source. For the deterministic analysis, the BOC supports the use of an
Mmax of 7.5 at the closest hypocentral distance and 84th% deterministic ground motion, or an
Mmax of 7.8 to 8.0 at a hypocentral distance uniformly distributed on the rupture plane and
84th% deterministic ground motion as recommended in Technical Memorandum 14-04-TM.
Alternately, the latter is equivalent to the use of the closest hypocentral distance and 69th %
ground motion as shown in 14-04 TM. The BOC recommends that additional discussion be
provided in 14-04 TM regarding the hypocentral distribution used in the analysis both in
depth (width) and length of potential Intraslab Fault Planes. The BOC recommends that the
project team evaluate the fault rupture dimensions associated with recorded magnitude 8
Intraslab earthquakes in the global data base for comparison to the dimensions of the
McKinley Block beneath the site. Such an evaluation can be used to assess whether the
Intraslab source at the site can support similar large magnitude events, whether a lower
Mmax (e.g., 7.5) is defensible for the MCE deterministic evaluation, and to inform the
weighting given to the magnitude distribution for the PSHA.
AEA Comment:
Noted. For the moment we will continue to use both M7.5 (84th %tile) and M8.0 (69th
%tile) or the higher of the two.
(3) Ground Motion Prediction Equation (GMPE) Model. The BOC recommends that a final
decision be made on the selection of the GMPE model for the Intraslab source. Currently,
the Deterministic and Probabilistic assessments use different GMPE models. For
consistency, a final assessment of the GMPE model should be used in the SSSHA for
development of final seismic criteria.
AEA Comment:
The deterministic and probabilistic assessment use the same GMPE. There is an error
in the Technical Memorandum 14-04-TM that would lead the reader to believe
otherwise. This error will be corrected. The computations included in the Technical
Memorandum 14-04-TM are for comparison purposes only and utilize one GMPE
criteria.
Concluding Remarks
The BOC appreciates that this phase of the Feasibility Study need not get into the final design
details; however, it also feels that there are significant basic conditions that influence the
configuration and performance of the dam. These factors can and may well affect the feasibility and
Page 9 of 10
estimated cost of the Project. Whether the present Feasibility Report addresses or tries to address all
of the potential conditions of the dam and its environs is up to both the Owner and the Engineer.
However, these are serious conditions and considerations that must be addressed. The following are
several considerations identified, that need to be recognized and attended to:
The presence or absence of fault or shear features in the foundations that may affect both the
static and dynamic performance of the RCC dam. The BOC strongly suggests that these issues
be resolved in a timely manner, preferably in the 2014 detailed surface mapping and inclined
cross-river borings planned during first stage of the investigation.
AEA Comment:
Noted.
The existence of permafrost within the foundation rock formations and how it has affected or
will affect the foundation characteristics (i.e. ice jacking, rock block movements, long term
foundation permeability etc.). The BOC strongly suggests that these issues be resolved in a
timely manner, preferably in the first stage of the investigation (2014-15), from the proposed
exploratory adits.
AEA Comment:
Noted.
Knowing that the dam will be subject to sliding during an MCE event, consideration should be
given to shaping of the foundation of each monolith to provide an upstream inclination for
additional shear resistance.
AEA Comment:
Noted.
It is important that upper abutment blocks (thrust blocks) provide adequate support for the arch
thrust to capture or limit movements. This may be accomplished by appropriate orientation of
the thrust blocks and possibly increasing their cross sections.
AEA Comment:
Noted.
With the completion of the PMF and estimate of the design ground motion, the BOC reiterates
its Comment 12.4 from Board Meeting 2 that special attention should be given to potential of
increasing the number of spillway gates thereby reducing the height of piers, which would be
more efficient in transferring the arch thrust into the adjacent blocks and improving cross-valley
performance of the piers. This issue is also of concern to FERC as a result of a present incident
at Wanapum project spillway.
Page 10 of 10
AEA Comment:
Number of gates will be increased to four.
Thermal considerations regarding placement of RCC directly on the cold foundations and
shrinkage.
AEA Comment:
Noted. A thermal analysis is planned.
The transverse joint spacing that is appropriate for the cold climate and the thermal shock
stresses generated by the cold water when the reservoir is impounded.
AEA Comment:
Noted.
Considerations regarding longitudinal cracking from concrete shrinkage and foundation restraint
Consideration of foundation grouting within the extremely cold foundation rocks and
groundwater.
The complications of sequencing of the seasonal placements and the thermal effects on the
internal stress development.
AEA Comment:
The thermal model will address.