Tuesday, January 11, 2011 10:18 PM
If you dont make dams higher when the sun shines so you can't store water when the rain falls !!!!!!!!!!!
Currently the Wivenhoe storage above full supply level is 67m
The original DFL is 80m (ie 13m wet freeboard which is a lot) with an annual exceedence probability of 1:22,000 (1 in 22,000 year event)
The secondary spillway (fuse plugs) maintains both the FSL & DFL but now gives an AEP of 1:100,000 which is below recommended for such a dam but reasonable as a stage 1 event.
Brisbane river seems to have peaked at Wivenhoe at 74 metres, as of 6PM, 11 Jan 2001. 7 Metres above its FSL. Currently its at 192% full, 100% FSL and 92% flood storage level. Its a very nervous state.
The unsettling thing is whats in Appendix H Somerset Dam Risk Assesment Failure Modes and Consequence Assesments which I shall add at the end.
http://www.bom.gov.au/fwo/IDQ65389/IDQ65389.540177.plt.shtml
South East Queensland Water now acnowlege Wivenhoe Dam is releasing about 576,000 megalitres (ML) per day and still rising.
http://www.seqwater.com.au/public/home
Section 4 gives a pretty good account of the Wivenhoe Asset.
A report writen in 2007 but not acted on as the Dam FSL is still 67 M. Had they chosen to implement recommendation W3 with FSL to 75M then there wouldnt be anywhere near the flood that Brisbane is now prepairing for.
DEPARTMENT OF NATURAL RESOURCES AND WATER
PROVISION OF CONTINGENCY STORAGE IN WIVENHOE & SOMERSET
DAMS
http://www.qwc.qld.gov.au/planning/pdf/support-docs/provision-of-contingency-storage-in-wivenhoe-and-somerset-da.pdf
1. Executive Summary
This report has been prepared in conjunction with the Queensland Department
of Natural Resources and Water (NRW) to investigate options to provide
contingency storage as part of the South East Queensland Regional Water
Supply Strategy (SEQRWSS). As part of these investigations it is proposed to
look at options for the provision of an additional 200 to 600 GL of contingency
storage in the Brisbane River catchment. The two options for this report are:-
• Raising Wivenhoe Dam Full Supply Level (FSL)
• Raising Somerset Dam FSL
These two options are being compared with other storage options in South
East Queensland.
1.1 Scope of Work
This scope of work for this report includes the following options for the
provision of the contingency storage:-
Option W1 - Raise Wivenhoe Dam FSL by 2m to EL69.0
Option W2 - Raise Wivenhoe Dam FSL by 4m to EL71.0
Option W3 - Raise Wivenhoe Dam FSL by 8m to EL75.0
Option S1 - Raise Somerset Dam FSL by 2m to EL101.0
Option S2 - Raise Somerset Dam FSL by 4m to EL103.0
Option S3 - Raise Somerset Dam FSL by 6m to EL105.0
This report provides:-
• Background data for each dam including risk profiles.
• A broad description of the works required to raise each dam to the
nominated FSL.
• Feasibility cost estimates for each option.
• A preliminary assessment of the environmental and social impacts of
each option.
• Risks and opportunities associated with each option.
It can be seen from the table that the most attractive option for the provision of
contingency storage would be a 2m raising of Wivenhoe Dam as an
operational change eliminating the need for expensive capital works.
Intuitively, Wivenhoe would be the most logical option for contingency storage
given the size of the catchment and the corresponding probability of capturing
the additional flows.
The provision of contingency storage in Somerset will be difficult due to the
upstream flooding issues associated with Kilcoy and land owners.
1.2 Flood Security Costs
Neither Wivenhoe nor Somerset currently satisfies the ANCOLD Guidelines on
Acceptable Flood Capacity (2003). SEQWater is committed to an agreed
program of works to allow the dams to comply with both ANCOLD and the
Spillway Adequacy Guidelines (NRW 2005) in the timeframe specified by
NRW. Given the assumptions for this study that the dams will be required to
pass the current estimate of the PMF, a substantial portion of the costs to
raise the FSL is associated with the long term works to increase flood security.
It is arguable whether these costs should be included for the provision of
contingency storage as SEQWater is likely to incur these costs in the future
even if the storage is not raised. An attempt has been made to separate out
the costs associated with the provision of additional storage from the costs
required to upgrade the current dams. These costs are presented in Table
1-2.
For Wivenhoe it can be seen that the incremental cost associated with the
small increase in the storage capacity is much less than the cost required to
upgrade the dam to full PMF Capacity. For Somerset the cost of increasing the
storage capacity is much less than the cost to upgrade to full PMF capacity in
all cases.
1.3 Limitations
This report is intended to be a preliminary feasibility investigation for options to
raise Wivenhoe or Somerset Dam. The investigations carried out for the
report have focused on the engineering aspects of raising Wivenhoe and
Somerset. There has been no attempt to quantify:-
• The potential impacts of the raising on the end of systems flows.
• The frequency and volumes of the storage to be held above FSL at
either or both of the dams.
• The potential benefit of raising Wivenhoe or Somerset on the
downstream flood impacts.
• Major environmental impacts.
• Impacts of the additional storage on the levels of service.
1.4 Flood Operational Procedures.
The proposed raising options investigated for Wivenhoe are capable of
producing similar outflow hydrographs to the current configuration, thereby
preserving the flood mitigation benefits downstream of the dam.
The proposed options for the raising of Somerset reduce the flood mitigation
capacity of the storage for downstream stakeholders (impacts on the flood
mitigation capacity of both Wivenhoe and Somerset) to limit the impacts of the
raised storage levels on Kilcoy and upstream areas. These options would
require a substantial revision of the flood operational procedures.
Option W1A has impacts on the flood capacity of the dam for events greater
than the 1 in 1,000 AEP event. Given the rarity of this event it considered that
this option has potential to be acceptable to the downstream stakeholders as a
short term (10 to 15 years) option to capture additional storage in Wivenhoe.
It has been assumed that minor changes to the flood operational procedures
and works to the downstream bridges may reduce the adverse impact of this
operational change even further. It is proposed that this assumption be
investigated further by SunWater, to provide a detailed assessment of the
impacts of the raised storage on the downstream flood levels.
1.5 Wivenhoe Raisings.
The raising options W1, W2 and W3 considered involve:-
• Complex work in the spillway which could only proceed one bay at a time
and probably only in the dry season months.
• The cost of such complex work with limited time windows is difficult to
estimate with reasonable certainty.
Options W2 and W3 involve raising the embankments and a temporary
relocation of the Brisbane Valley Highway causing major disruption to traffic.
Less significant disruption would be caused to the Wivenhoe - Somerset
Road. The indirect cost of these disruptions has not been estimated.
For Option W1A, the increase in downstream flooding is relatively minor but
its acceptability would be dependent on consultation with stakeholders. A
raising of Kholo Bridge and possibly of Burtons Bridge and Savages Crossing
could be required to deal with possible concerns.
For Option W1A, the existing fuse plug will be triggered more frequently
(existing 1:5,000 AEP flood). The frequency and consequences will need to
be examined in further detail.
1.6 Somerset Raisings
Issues associated with the raising of Somerset include:-
• Flood Mitigation. Each of the options investigated for the raising of
Somerset impact on the existing flood mitigation performance. This
impact is greater as the proposed raising increases. This is due to
constraints on the upstream flood levels imposed by Kilcoy and other
upstream development.
• Equipment age. The gates and hoist equipment at Somerset Dam are of
considerable age. There is some uncertainty whether it can be adapted
as proposed.
• Dam condition. Cracking in a number of the dam monoliths and other
stability concerns will be addressed concurrently with the raising
proposals.
• Community opposition to the higher raising proposals is likely to be very
strong.
• The indirect costs associated with the increased frequency of highway
disruption have not been estimated.
1.7 Recommendations
It is recommended that:-
• Raising of the FSL level of Somerset Dam be rejected due to the impacts
on the upstream population during flood events. Major flood events
already result in inundation of the Kilcoy and surrounding private
properties and infrastructure.
• The provision of contingency storage in Wivenhoe is investigated further.
A 2m raising in the FSL could be achieved with minimal capital costs
subject to addressing regulator and stakeholder issues.
• A detailed flood assessment is carried out to develop and asses changes
to the flood manual to allow the storage of the additional 2m in Wivenhoe.
The impact of the changes should be assessed for the full range of
Annual Exceedance Probabilities and Storm Durations. This assessment
should also link with the Brisbane River Flood Damages Assessment
currently being carried out by Brisbane City Council.
• A detailed review of the structural adequacy of the various components of
the dam is carried out to confirm the assumptions of this report. The review will provide design detail to refine the cost estimates and confirm
the feasibility of the proposed increase in storage level.
• A program of consultation with the downstream stakeholders is carried
out with the proposed changes to the flood manual once the assessment
of flood events is completed.
• SEQWater be provided with the opportunity to instigate a public
consultation process prior to the public release of options to raise the
storage levels of Wivenhoe.
End Executive Summary
Appendix H.
Somerset Dam Risk Assessment,Failure Modes and Consequence Assessments
Risk Assessment Studies
A number of studies have been undertaken in recent years relating to
various aspects of Somerset Dam. These include:
• A preliminary risk assessment of Wivenhoe, Somerset and North
Pine Dams by SKM, reported at SKM, 2000;
• A dam safety review of Somerset Dam by GHD, reported at GHD,
2000;
• A detailed risk assessment for Somerset Dam by SMEC. This risk
assessment was undertaken in two stages. The initial stage entailed
a review of information and identification of deficiencies. Stage 2
provided a detailed assessment of the likelihood of failure of the
identified deficiencies. This work is reported at SMEC, 2004.
• This study included an assessment of the reliability of the spillway
gates.
• A short review of dam safety issues, based on the above Reports
was carried out by Commerce in December 2004 and is reported at
Commerce, 2004:
• Further stability assessments of abutment monoliths were carried out
by Commerce and are reported at Commerce, 2005.
• The above Commerce Reports were based on a hydrology study by
WRM Water and Environment, WRM (October 2004). This Report
has been revised (WRM, September 2005) but these revisions have
not been incorporated in to the Commerce, 2005 conclusions
Failure Modes
The following is taken mainly for Commerce (2004) but includes
information from all sources referenced above, particularly SMEC (2004).
The detailed risk analysis for Somerset Dam, SMEC (2004), identified
three basic failure modes:
• Erosion of the downstream toe due to flood discharges passing
through the open sections of the dam abutments and impacting on
the foundation at the downstream toe of the dam;
• Structural failure of the dam under extreme water load. The dam
was considered stable at the foundation interface for the PMF (albeit
approaching the limit of its stability) but liable to failure at two higher
locations for smaller flood events;
At the change of slope in the back face of the non-overflow
sections;
At the Upper Gallery.
• Structural failure of the dam under earthquake.
The results obtained from the event tree analyses are summarised at
Table 11-1. Structural failure of the non-overflow units at the change in
slope of the back face was the dominant failure mechanism followed by
failure at the Upper Gallery. Gate reliability was assessed and included in
the event trees and had a significant effect on the results.
Table 11-1 - Result of Event Tree Analyses
Failure Mode Probability of Failure (/year)
Failure at Change of Slope under Flood 110*10-7
Failure at Upper Gallery; under Flood 80*10-7
Failure under Earthquake 80*10-7
Failure due to Toe Erosion 5*10-7
Total for Somerset Dam 275*10-7
Reference SMEC (2004)
Failure due to toe erosion at the toe of the dam was not considered to be
a major factor. The foundation was assessed as a hard strong andesite
with jointing the major defect. While erosion of the surface rock is
expected under low to medium flows, the rock mass was judged to be
“tighter” at depth and have a high resistance to erosion that is unlikely to
lead to dam failure.
Moderate earthquake events are likely to cause distress at the change of
slope, but as this is above Full Supply Level, it had no impact on the risk analysis. Stability analyses, GHD (2000), indicate the dam is unstable at
the Upper Gallery for the Maximum Design Earthquake.
Structural Investigation Studies
The critical flood levels adopted for the risk analysis were:
• EL109.7 for the Change of Slope failure;
• EL110.0 for the Upper Gallery failure.
These levels adopted by SMEC (2004) were based on separate stability
analyses by GHD (2000) and SKM (2000). SMEC (2004) noted that “the
results from the two analyses are at odds” and that “the reasons for the
differences are not apparent”. In addition, the Report in Appendix 3.6
extracts from DPI (1994) quotes a Ben Russo conclusion that differs from
both of these studies.
“Russo also recommends that to ensure the survival of the two
portions of two non-overflow monoliths above EL100.0 , the reservoir
should not exceed EL111.7 . He adds that the structural integrity of
the spillway gates(if used) would have to be checked for the loads
such a reservoir level would impose.”
The variations in these three stability assessments cover a range that
could have a significant impact on the event trees developed by SMEC
and on the overall risk assessment. The differences are presumably due
to different assumptions for uplift and for the extent of cracking in the
concrete at the Upper Gallery.
Commerce reviewed the stability assessments and concluded that
stability criteria were satisfied for:
• Storage levels up to EL111.0 at the change of slope;
• Storage levels up to EL110.9 at the Upper Gallery;
However, if extensive cracking exists above or below the gallery. The
dam just satisfies stability criteria for a storage at EL109.7.
Hydrological studies (WRM, 2004) assess the storage level for flood with
an AEP if 1 in 100,000 at EL109.75. The above studies indicate that the
dam would satisfy normal stability criteria at this level, although there
would be little margin if cracked concrete exists above or below the Upper
Gallery.
This conclusion needs to be reviewed following the revised hydrology
study at WRM, 2005.
Impact of a Somerset Dam Failure on Wivenhoe Dam
The impact of a Somerset Dam failure on Wivenhoe Dam was detailed at
Commerce (2004) and summarised below.
The consequences of failure of Somerset Dam are largely dependent on
whether it can cause a cascade failure of Wivenhoe Dam. Wivenhoe
Dam, with Stage 1 Upgrade works now completed, is designed to handle
a 1 in 100,000 flood event centred on the Wivenhoe catchment, assuming
that Somerset Dam does not fail.
Somerset Dam, on the basis of its known condition, satisfies stability
criteria for a storage level of EL109.75 and will safely handle the 1 in
100,000 AEP flood event. This in turn ensures that the Stage 1 upgrade
works for Wivenhoe Dam are not compromised by any Somerset Dam
deficiencies.
On this basis upgrade work at Somerset Dam, if required at all, would
reasonably attract the same degree of urgency as Stage 2 Wivenhoe
works. It is recommended that any upgrading of Somerset Dam be
considered at the time that Stage 2 Wivenhoe works are assessed.
Commerce, 2005 raises several issues in relation to the above:
• Cracking observed in the Upper Gallery walls may also exist above
or below the Gallery. While such cracked concrete would just satisfy
stability criteria for a storage level of EL109.75, stability reduces
rapidly for higher storage levels and failure could occur at EL110.1.
It was recommended that some exploratory drilling be carried out to
determine whether such cracks do exist. A similar recommendation
was made in GHD (2000);
• The WIVOPS flood operation program at one time required that the
Somerset spillway gates be lowered if Wivenhoe Dam is in danger of
being overtopped. This is a difficult procedure that would raise a
number of operational and safety issues and require a review of the
stability conclusions given above.
• Stability analyses assume that the gallery systems are not flooded by
water overtopping the abutment monoliths. The dam layout should
be reviewed to ensure this is the case and waterproof doors installed
where necessary.
Consequences of Failure for Somerset Dam
Loss of Life Assessments
If Somerset Dam fails without causing a cascade failure of Wivenhoe
Dam, the consequences are limited to the area between the two dams. The SKM (2000) Report predicted no loss of life would occur from a
Somerset failure for the following reasons.
• The small population at risk for flood failures;
• Adequate warning times for flood failure;
• The location of the population at risk above peak flood levels caused
by normal operational failure.
If failure of Somerset causes a cascade failure of Wivenhoe Dam, then
the loss of life figures are substantially increased. This could only occur
during an extreme flood event as Wivenhoe reservoir has sufficient
capacity to store the normal Somerset storage without initiating the
secondary spillway fuse plug.
SKM (2000) provided loss of life estimates for both day and night failures
of Wivenhoe Dam for a variety of load cases. SMEC (2004) has used the
SKM data for total loss of life at night and adopted the following loss of
life figures for the risk assessment:
• IFF Failure (Main Embankment) 89
• Earthquake 36
• Normal Operating Condition 77
Financial Loss Assessments
SKM (2000) has assessed the financial consequences associated with
the failure of Somerset Dam under three broad categories; third party
damages, SEQWater direct damages and SEQWater loss of revenue. A
major failure of Somerset Dam, involving failure of the spillway gates and
partial failure of the abutment units was valued at $20M, with $18M of this
classed as SEQWater direct damages.
SMEC (2004) quote a far higher cost of $200M to repair Somerset,
including environmental impacts.
These estimates depend heavily on the type of failure and extent of the
damage. Failure of several abutment units at the change of slope would
incur a relatively low repair cost, while major damage to the gated
spillway would involve substantially higher repair costs. No detailed
estimates are available but the SKM (2000) estimates appear low,
particularly as they involve spillway gate failure.
Similarly, a major flood failure of Wivenhoe Dam is estimated at $12B to
$25B by SKM (2000).
Environmental & Intangible Consequences
The SKM (2000) study included an assessment of environmental and
intangible consequences. SKM assessed the incremental environmental
consequences for both Somerset and Wivenhoe dams as low. The
incremental intangible consequences were also assessed as low for
Somerset although high for Wivenhoe. It concluded that:
“These environmental and intangible consequences were far
outweighed by the significant life loss and financial consequences for
this portfolio. As such they did not play a significant role in the
development of the risk reduction strategy.”
Risk Analysis
No Failure of Wivenhoe Dam
SMEC, 2004 notes that for zero loss of life, the ANCOLD life safety
criteria do not apply.
The ANCOLD fallback criteria however, would require either PMF security
for an “Extreme Category” or PMPDF security for a “High A Category.”
Somerset Dam does not satisfy PMF and is unlikely to satisfy PMPDF.
This reflects the overall importance of the dam to SEQWater.
SMEC, 2004 also notes that the risk of failure could be reduced by
around 3 orders of magnitude by:
• Installation of anchors to increase the structural adequacy at the
upper gallery and change of slope;
• Construction of a concrete slab/cutoff at the toe of the dam to protect
against erosion and undermining.
The above works have not been costed, SKM nominated costs between
$1M and $2M (now dated) and SMEC “judged that costs are likely to be
higher, but still in the millions of dollars range”.
SMEC noted that the cost of anchors could be justified, even if
consequential failure of Wivenhoe did not occur. The value of erosion
protection was more difficult to justify and that “:it would need to be
determined whether its cost is grossly disproportionate to the
improvement gained”.
Upgrading of Somerset Dam, as a stand alone structure is an ALARP
issue under the ANCOLD Guidelines. As noted by SMEC, SEQWater
needs to determine their acceptable level of risk in order to assess the
need for risk reduction measures.
Cascade Failure of Wivenhoe Dam
The FN Chart produced by SMEC, 2004 is shown at Figure 5, and is
based on the risk assessment of Somerset Dam with the assumption of a
conditional probability of failure of Wivenhoe Dam of 1.0. This Report did
not assess the likelihood of a failure of Somerset Dam resulting in a
failure of Wivenhoe Dam. The FN Chart plots above the Limit of
Tolerability and as such the risk would be deemed intolerable.
The original risk analysis for Wivenhoe Dam was developed by SKM. The
Wivenhoe Alliance revised this work to incorporate the risks associated
with a Somerset failure. The FN Charts for total loss of life is shown at
Figure 6 and indicates that:
• The original Wivenhoe Dam plots well above the ANCOLD Limit
Line;
• The Stage 1 Upgrade for Wivenhoe brings the risk below the
ANCOLD Limit Line provided Somerset does not fail;
• If allowance is included for Somerset Dam failure case, the plot rises
just above the Limit Line;
• The Stage 2 Upgrade brings the risk well below the Limit Line.
The total risk to Wivenhoe Dam as a stand-alone construction following
the Stage 1 Upgrade works is assessed at 0.84*10-5. Introducing the risks
associated with a Somerset failure increases these risks by a factor of 2.4
to 2.0*10-5.
The risk to life matrix (F-N Chart) using the incremental loss of life figures
is reproduced at Figure 7. This shows the Wivenhoe risks plotting below
the ANCOLD Limit Line.
Limitations of Risk Studies
The SMEC, 2004 study of Somerset Dam is considered a detail risk
assessment, with the limitation that it does not consider the likelihood of a
failure of Somerset Dam resulting in a failure of Wivenhoe Dam. The
Report uses the SKM loss of life figures. The SKM Report was a
preliminary assessment and as SMEC notes the consequence study is
not developed to the same standard as the failure analysis.
The Wivenhoe Alliance study is a modification of the SKM study and as
such is a Preliminary Risk Assessment. If the risk profile is a concern, a
detailed risk analysis should be carried out, that includes a detailed
assessment of the consequences, particular loss of life.
Hazard Category
The Dam Safety Management Plan, SEQWater (2005) at Section 6.1
states that “The Corporation’s dams are classified under the ANCOLD
classification guidelines as HIGH hazard because of the significant
consequences of a dam failure”. These are presumably the 1986
ANCOLD Guidelines.
The hazard classification was determined by GHD and the following
statement included in GHD (2000).
“A hazard assessment was conducted in accordance with the DPI (DNR)
Dam Safety Guidelines Procedure DS003 and the June 2000 ANCOLD
Guidelines on Assessment of Consequences of Dam Failure. Both
methods indicate that the dam should be classified as having a High
Hazard Category.”
No discussion of the hazard classification was provided at GHD (2000). It
is assumed that the hazard classification allows for the possibility of a
cascade failure of Wivenhoe Dam and, given the financial loss
assessments noted at 0, that this would be a High A classification for
flood under ANCOLD.
It is recommended that the Hazard Classification be given further
consideration on the basis that:
• The PAR from a cascade failure would be in excess of 1,000, and
the ANCOLD Guidelines would indicate a an Extreme Classification;
• The Hazard Classification for a sunny day failure would be lower,
possibly High B or High C.
Conclusions
Somerset Dam as a stand alone structure satisfies the ANCOLD risk to
life criteria. There is scope for substantially reducing the risk of failure, but
the value of this work needs to be assessed in terms of the SEQWater
risk management procedures.
A cascade failure of Somerset and Wivenhoe Dams is possible and
stability is marginal for the 1 in 100,000 AEP event. Preliminary risk
assessments indicate the cascade failure is close to the ANCOLD Limit of
Tolerability.
SEQWater has completed Stage 1 of an upgrade program and Wivenhoe
Dam now is now capable of handling a flood with an AEP of 1 in 100,000.
Stage 2 would provide full PMF security. This would satisfy the ANCOLD
Limit of Tolerability and the ALARP principle.
80M = a 1:20,000 We are at 74.5 Metres so its most likley a 1:10,000 year event (The second one in Australia in the last 12 months.)