STRUCTURAL ENGINEERING SOCIETY
NEW ZEALAND
BUILDING SEISMIC PERFORMANCE
Submission to the Ministry of Business, Innovation and
Employment
8 March 2013
Executive Summary
The Structural Engineering Society of New Zealand (SESOC) is pleased to provide its response
to the Ministry of Business, Innovation and Employment’s Building Seismic Performance
Consultation document.
The key points of our submission are as follows:
1. Noting that there is inadequate information on the scale and extent of the problem, we
support the recommendation of the Tony Taig report, to prepare a comprehensive Risk
Assessment framework, before finalising a programme for addressing these issues.
However upgrading or removing the highest risk buildings or building elements should
be progressed in parallel.
2. In order to ensure that the most dangerous buildings are identified (many of which may
not be considered earthquake prone under the current system) a new definition of
earthquake prone buildings is required. This can also be tailored to exclude those
buildings which may be currently considered earthquake prone but are demonstrably not
dangerous.
3. Once the Risk Assessment process is completed, a prioritisation can be developed for
further assessment and implementation of risk reduction measures that is effective and
affordable.
4. MBIE should continue to engage with IPENZ and the technical societies to develop a
programme that is focused on the most dangerous buildings and the most effective means
of reducing unacceptable risk.
5. There must be considerable effort given to training and communication, of the
professionals that will be involved in assessment and retrofit, and of the general public, in
their understanding and management of risk.
www.sesoc.org.nz
PO Box 6508, Wellesley St, Auckland 1141, NZ
A Collaborating Technical Society with
Contents
Introduction ..................................................................................................................................... 2
Structural Engineering Society of New Zealand............................................................................. 3
Objectives........................................................................................................................................ 3
Problems with the Current System.................................................................................................. 3
Risk Framework .............................................................................................................................. 6
Proposed Changes to EPB Definition and Regulations .................................................................. 7
Earthquake Prone Building Definition........................................................................................ 8
EPB Regulations ......................................................................................................................... 8
Role of SESOC and other Societies looking forward ..................................................................... 8
Responses to Proposals and Questions............................................................................................ 9
Appendix A
Earthquake Prone Buildings
Discussion Paper
Introduction
This submission has been prepared by the Structural Engineering Society of New Zealand
(SESOC), in response to the Consultation document prepared by the Ministry of Business,
Innovation and Employment (MBIE). The submission firstly discusses a number of issues raised
in the consultation document and then provides specific responses to the proposals and questions
raised.
SESOC has been working collaboratively with IPENZ, the New Zealand Society for Earthquake
Engineering and the New Zealand Geotechnical Society to deliver a comprehensive and
considered response. These groups have made separate submissions which may vary in detail,
but which are in general agreement of the key points, as articulated in the summary of this
submission.
Furthermore, although there was relatively little time to do so, SESOC has consulted with
membership. Those who were able to respond were generally supportive of the draft
submission, with a number of further suggestions, many of which have been adopted.
A discussion document is attached to this submission, providing further background to some of
the issues discussed.
The Consultation document notes that the Building Act covers other structures such as bridges
statues and memorials. These are not addressed specifically in this document, noting that
bridges are generally publically owned and considered in lifelines studies, and that other
structures generally pose a low risk to life. However, many of the principals expressed in this
submission would be equally applicable across all structural forms.
Version 4
Page 2 of 20
Structural Engineering Society of New Zealand
The Structural Engineering Society of New Zealand (SESOC) is a collaborating technical society
of the Institution of Professional Engineers of New Zealand (IPENZ). SESOC has
approximately 1450 members, comprising mostly practising structural engineers. As a non>
profit organization, SESOC is generally managed and staffed only by volunteers.
The vision of SESOC is to be recognised as the authority promoting the highest standards in the
science, art and practice of structural engineering; with our mission as follows:
•
•
•
To uphold and promote the profession of structural engineering
To provide leadership and structural engineering advice to society on key issues
To meet the professional needs of members.
SESOC has a strong interest in the Earthquake Prone Building (EPB) legislation and regulations
in reflection of our members’ interests and structural engineers’ wider role in the community. In
this, it has been working already with IPENZ, the New Zealand Society for Earthquake
Engineering (NZSEE) and New Zealand Geotechnical Society (NZGS) to provide expert
guidance to MBIE.
Objectives
SESOC considers the objectives of the Building Act and regulations in respect of existing
buildings should be as follows:
Primary:
To reduce building>related risk of death and injury from earthquake at reasonable
cost to the country
Secondary:
To ensure, as far as is practicable, that emergency facilities remain functional and
accessible after a moderate to severe earthquake.
To minimise the impact of earthquake prone building legislation on New
Zealand’s built heritage, while achieving an acceptable level of safety.
To provide (accurate and readily understood) information to the public so that
they can better understand the level of risk to which they are exposed when
entering buildings
These objectives are broadly in line with those noted in section 2 of the Consultation document.
Problems with the Current System
There are a number of issues with the current system. Some of these have been highlighted by
the Canterbury Earthquakes Royal Commission (CERC) and MBIE, and are summarised in the
Consultation document:
•
•
•
•
•
•
Too much variation in local practice
Public confusion about risk
Lack of good data
Poor information on individual buildings
Inconsistent market responses
Lack of central guidance
Version 4
Page 3 of 20
SESOC agrees that these issues are of significance. To these should be added some further
issues:
1.
The current system is inappropriate for critical facilities. The current system fails to
adequately consider the more critical requirements of emergency facilities. For facilities
that are required to be operational following a major earthquake1, a lower standard of
performance is not acceptable. Even though such structures are required to be designed and
assessed against a higher load level, the current system requires existing buildings to come
up to only 33% of this level. This level would not ensure such buildings are operational
when most needed.
This needs to be further extended to consideration of adjacent buildings that may impact or
influence the continued use of the critical facility and major arterial access routes, noting
that both of these impacted on the use of critical facilities following the Christchurch
earthquakes.
Consideration should also be given to the application of this to structures housing highly
hazardous contents.
2.
The current system does not address the highest risks. There are some forms of
building that may be earthquake prone under the current system, such as older timber framed
buildings, which are not a collapse hazard and are unlikely to represent a life safety hazard.
Completing detailed assessments and seismically upgrades to these buildings may be poor
use of resources. Conversely, there are many buildings that would not be classified as
earthquake prone under the current system, but which represent a significantly greater life
safety risk. This issue is discussed in more detail in the appended discussion paper.
The conclusion is that a better definition of earthquake prone buildings needs to be
developed that focuses more directly on the most dangerous buildings. Not only will this
provide a higher level of life safety, but it should reduce costs associated with assessing and
retrofitting buildings that are not dangerous.
3.
The cost of fully upgrading all EPBs may be disproportionately high compared to the
benefits derived from doing so. In many smaller cities, towns and suburban areas, there
may be only marginal returns from commercial property. This will not sustain high levels of
expenditure on full seismic upgrades. However, the low occupancy rates and low street
counts may also mean that the overall risk is relatively low, compared to busy city streets,
noting that the majority of death and injuries from low>rise unreinforced masonry buildings
in Christchurch were to people outside the building. It is possible that an intermediate
upgrade approach that concentrates on mitigating parapets, face loaded wall failure and
other exterior life safety threats may be a better use of resource, where the buildings have an
otherwise sound independent gravity structure.
It is recommended that a study is completed that investigates the risk presented by such
buildings and develops simplified methods of mitigation of these hazards. The level of
occupancy may be a factor that could be included in assessment of risk.
1
Typically, emergency facilities are designated Importance Level 4 (IL4). They are designed for a higher seismic
design load of 1.8 times that used for regular (IL2) buildings. In addition, they are subject to an additional
serviceability limit state requirement (SLS2) requiring them to be operational after a 500 year return period event,
corresponding to the ultimate limit state design level for IL2 buildings.
Version 4
Page 4 of 20
4.
Heritage buildings in private ownership are potentially under threat due to the high
cost of compliance. By definition, if a building is regarded as a heritage building, this
recognises its overall high value to the community. However, an upgrading standard that
concentrates only on meeting a life safety standard may result in buildings that are unlikely
to be practically repairable after the event.
A higher level of protection that recognises these buildings’ heritage status is likely to be
considerably more expensive than compliance with the life safety standard implied by the
current system. However there is often little or no public funding available to upgrade these
buildings and so the burden of protection falls only on the owners.
SESOC supports the Historic Places Trust recommendation for the development of a
National Risk Map for New Zealand’s heritage. This may form the basis of a prioritisation
of heritage buildings requiring additional protection; and could also inform an approach to
public funding (or part>funding).
5.
The current system is being misused by some Bankers, Insurers, Lawyers and
Owners. Use of the IEP tool, devised to screen buildings in order to identify those requiring
further evaluation, is out of control. For example:
•
•
•
•
•
Building tenants are unsure if they are breaching Health & Safety obligations to their
staff if they discover their building is less than 100% NBS (based on IEP).
Real estate advertising now frequently quotes that a building is 100% Code or (in at
least one case) 230% NBS based on IEP results.
Lawyers are writing clauses in to leases contracting the owner to warrant that a building
is 80% NBS.
Bankers are requiring an IEP assessment of every building before funding loans on
buildings.
Insurers are limiting or refusing cover on some buildings on the strength of IEP
assessments.
All of the above are examples of undue reliance of a screening tool for inappropriate use.
Some clear guidance on a grading system, along with an explanation of relative risk, is
required to rationalise the developments that have occurred with the property market since
the Canterbury earthquakes. Although this issue is not a direct outcome of the current
system, it is indicative of the broader issues that have resulted from it.
6.
The Building Act must be extended to provide a framework for evaluating and
repairing damaged buildings. Although this is not strictly related to earthquake prone
buildings, we note that the sections dealing with earthquake prone and dangerous buildings
have been the default sections considered for dealing with earthquake>damaged buildings.
There needs to be a legal framework for evaluating and repairing damaged buildings that
sets reasonable targets for interim occupancy, repair and strengthening (if required). This
should be developed in consideration to all contributing factors, including:
•
•
•
•
Version 4
the type and scale of the event causing the damage
the likelihood of further events
the extent and nature of the damage
the impact of the damage
Page 5 of 20
Risk Framework
Lack of good data is one of the most important issues highlighted in the consultation document.
This is a hindrance to the development of robust policies, achievable timeframes and accurate
costs assessment. As part of the preparation of the consultation document, MBIE commissioned
a risk review from TTAC Limited2. A strong recommendation of the TTAC report was that
MBIE develop a risk>based framework to link the Building Code and related documents to the
safety outcomes that they are designed to deliver.
SESOC strongly supports the development of such a framework. This in turn will enable the
development of a programme of assessment, prioritisation and implementation that delivers the
best balance of risk reduction and impact on society. This should not override or interfere with
robust risk reduction programmes that are already underway (for example in Wellington City)
and nor should it delay the mitigation of the most dangerous risks, but SESOC considers it
should be developed before more long>term policies are implemented. A broad framework is
described in Figure 1 below.
Building Vulnerability
Profiling
By age
By construction type
By size
By vulnerability
Occupancy Profiling
By building use
By occupancy –
interior and exterior
Risk
Framework
Consultation
and
Communication
Building Inventory
By Local Authority
By vulnerability class
(refer above)
Hazard
Seismicity
Other hazards (eg
rockfall)
Figure 1: A Proposed Risk Framework for EPBs
Some of the components of the Risk framework are already in existence. For example, a
considerable amount of work has already been completed on the Building vulnerability profiling
(by NZSEE, as inputs to the IEP development3, and the Engineering Advisory Group4.
2
Tony Taig, TTAC Limited and GNS Science, A Risk Framework for Earthquake Prone Building Policy,
November 2012.
3
New Zealand Society for Earthquake Engineering Assessment and Improvement of the Structural Performance of
Buildings in Earthquakes, June 2006
4
Appendix A, Guidance on Detailed Engineering Evaluation of Earthquake Affected Non#residential Buildings in
Canterbury Part 2 Evaluation Procedure, Draft prepared by the Engineering Advisory Group, Rev 7, 16 May 2012.
Available at www.sesoc.org.nz.
Version 4
Page 6 of 20
A suggested sequence for the subsequent development of a robust programme is shown in
simplified form in Figure 2 below. The first stream involves the identification of the greatest
hazards and the implementation of an immediate retrofitting or removal programme.
The second is the longer term development of the Risk Framework with subsequent programme
development and implementation. Step 1, the development (and implementation) of the risk
framework is described above. The following assessment of industry capacity to respond (both
for assessment and implementation) then allows prioritization and finalization of the programme
with a better understanding of the costs and benefits that may be derived.
Figure 2: Flowchart of proposed programme development
There will be a need for consultation and communication at each step. The development of a
practically achievable programme will require consideration of tolerable life risks, taking into
account the improved data that should be available as the risk framework is populated. This
consideration must be spread wider than engineers and policy>makers. Although it is
uncomfortable to consider the relationship between cost and safety, it is a necessary fact that
society needs to be able to balance earthquake risk against other possible sources of risk and the
overall cost of reducing the risk.
It should be considered that some of the best risk reduction measures may not involve physical
risk reduction, noting that an effective education programme to discourage people from running
out of buildings during earthquakes could have saved several lives during the Christchurch
earthquake.
Proposed Changes to EPB Definition and Regulations
SESOC recommends that the EPB definition and supporting regulations are amended to provide
more focus on those buildings that present the greatest risk to life. This requires a dual approach
that places greater emphasis on building behaviour, as outlined below.
Version 4
Page 7 of 20
Earthquake Prone Building Definition
A revised definition is proposed as follows:
An earthquake prone building is a building that either:
• Is likely to have its ultimate capacity exceeded in a moderate earthquake, either
wholly or in part, in a way that may lead to death or injury to persons within or
outside the property; or
• Has significant critical vulnerabilities that could result in catastrophic collapse
in a major earthquake.
The first part of the proposed definition is essentially unchanged, but requires supporting
changes in the regulations (and in supporting documents and with training of assessors)
that provide greater definition of which buildings and/or hazards may be considered
dangerous.
The second part targets buildings such as the CTV building. Currently, even though
MBIE has initiated a process of identification of such buildings, there is no legislative
mechanism to enforce upgrade. The term ‘major earthquake’ may be defined in the
regulations. A possibility that may be considered is to link this to the equivalent new
building seismic design load, i.e. 100%NBS. However, an additional margin is
recommended over and above 100%, in order to achieve a similar level of resilience –
refer to the appended discussion paper.
EPB Regulations
In addition to the definitions of Moderate and Major Earthquake, it is recommended that
the Regulations be expanded to allow specific vulnerabilities to be defined which must be
considered in addressing the second part of the definition noted above. This should be
generally consistent with the building vulnerability profiling developed for the risk
framework.
This may also be the appropriate location to indentify those structures or elements of
structure that do NOT require consideration – for example low>rise timber framed
structures may be exempted from consideration unless they are on significant slopes.
This will save unnecessary expense in the assessment and retrofit of buildings that are not
dangerous.
Role of SESOC and other Societies looking forward
SESOC has already been working (together with IPENZ, NZSEE and NZGS) to provide MBIE
with advice on the EPB legislation and regulations, and their implementation. Recognising that
there are a range of other groups including territorial authorities, building owners, building users
and heritage advocates, the engineering societies are collectively in a unique position to assist
MBIE in the following:
•
•
•
In formulation of appropriate definitions and wording for the Act and regulations that
will ensure that EPBs are clearly defined and reasonable measures for risk reduction set
in place.
In developing technical guidance to deliver consistency and quality of outcomes in
review and in risk reduction measures.
In providing training for practitioners and stakeholders.
Version 4
Page 8 of 20
Responses to Proposals and Questions
Proposal 1: Local authorities would be required to make a seismic capacity
assessment of all non5residential and multi5unit, multi5storey residential buildings in their
districts within five years of the legislation taking effect, using a standard methodology
developed by central government, and to provide the resulting seismic capacity rating to
building owners. An owner could have their building’s seismic capacity rating changed
by commissioning their own engineering assessment.
Proposal 2: Assessments would be prioritised faster for certain buildings (e.g.,
buildings on transport routes identified as critical in an emergency).
1.
Should local authorities be required to assess the seismic capacity of all buildings covered
by the earthquake>prone building system in their areas, and to issue seismic capacity ratings
to owners?
Yes. However, there is a need to develop a Risk Assessment framework first, as noted in
the issues raised above, that provides sufficient clear information to make informed
decisions. This should not stop practical risk reduction programmes already underway.
Equally, the known severe hazards such as unreinforced masonry parapets and non>ductile
concrete columns should be subject to immediate action.
There is no need for a new rating system to be developed. If the IEP is not considered
suitable for use as a sifting tool, it should be modified. However, no more simplistic method
can deliver a more realistic view as the IEP is already a broad filter that is widely understood
by engineers, local authorities and informed building owners.
Ratings assessed using a simplified assessment tool such as the IEP should only be issued to
owners. Such ratings should NOT be publically notified or notified to building
occupants/tenants, as this simplified tool is conservative and therefore potentially
misleading. This could induce users to vacate serviceable buildings unnecessarily.
2.
Do you think five years is a reasonable and practical time to require local authorities to carry
out assessments in their districts?
No. Although a shorter timeframe may be desirable, it is probably impractical. A Risk
Assessment framework should be developed first, as noted above. However, the critical
point is that it must actually happen this time.
3.
Should unreinforced masonry buildings be assessed faster than other buildings?
Priority should be given to assessing the highest risk buildings. However, this should be
qualified further by a better identification of seismic risk than simply assessed capacity –
refer to our separate discussion paper.
4.
What costs and other implications do you see with these proposals to assess the seismic
capacity of buildings?
The allocation of assessment costs (between local authorities and building owners) is a
significant issue. A suggestion is that the local authorities pay for the initial assessment and
then building owners pick up all following detailed assessment and retrofit costs. The cost
Version 4
Page 9 of 20
of assessment could be reduced by limiting assessment to those buildings that are genuinely
at risk, although this may be offset by widening considerations to buildings that may exceed
the EPB threshold but should still be considered dangerous.
Availability of resource will be a serious consideration, and should be included as a factor in
the Risk Assessment framework.
These matters are addressed in our separate discussion paper.
Proposal 3: Building information would be entered into a publicly accessible register
maintained by MBIE.
5.
Do you agree that local authorities should be required to enter information on the seismic
capacity of buildings into a publicly accessible, central register to be managed by MBIE?
Yes, provided that the only information entered and made public are the final assessments as
agreed between local authorities and owners and not the initial (IEP) assessments carried out
by the local authority.
6.
Should information other than a building’s seismic capacity rating be entered into the
register – for example, agreed strengthening actions or information from an agreed building
ratings system?
Yes. It may be possible to enter at least agreed timeframes over which action is to be taken,
so that these dates are likely to be adhered to and are available for prospective tenants of
purchasers to complete longer term planning. However such additional information should
be limited and defined, to prevent “extrapolation” by over>zealous local authority officials.
LIMS and PIMS would presumably include this information where available.
A further extension could be the provision of assessment data that identifies vulnerabilities
in the building so that in the event of an earthquake, likely damage hotspots and/or
performance could be identified. Refer Galloway, 2012.5
7.
Rather than a central register, should local authorities be responsible for both collecting and
publishing this information?
Although the local authorities should be responsible for collecting the data, there should be
only one central source of information, which logically resides in a central database. This is
however not critical, provided that the information is accessible, accurate and consistently
applied across the country.
8.
Should there be any other information disclosure requirements – for example, should
building owners be legally required to display information on the building itself about the
building’s seismic capacity?
Subject to building type and location, building owners may be obliged to publish finally
agreed information regarding the building capacity in a location that is open to public view,
in the same way they currently show a building warrant of fitness. This may be limited to
5
Galloway B.D. & Hare H.J., 2012, "A review of post>earthquake building control policies with respect to the
recovery of the Christchurch CBD", New Zealand Society for Earthquake Engineering Bulletin, Vol. 45 No. 3,
NZSEE, Wellington, New Zealand
Version 4
Page 10 of 20
buildings that are commonly accessible to the public. Such information could include the
assessed building capacity, when it was assessed and future upgrading requirements.
Subject to future development, a rating system such as proposed by Quakestar could be an
effective means of conveying some of this information. However, there must be only one
system in use, or the information will be confusing. As for question 1, displayed
information should not be based on the initial local authority assessment.
It is of note that an effective structural engineering assessment of a building generally relies
upon access to structural plans. Site inspections are of limited value. Records are often held
by Local Authorities, but are not always available. A database of structural engineering
records is critical to successful seismic assessments. Publishing assessments based on poor
information may be over>conservative and potentially punitive to owners.
9.
What costs and other implications do you see resulting from the proposal to put seismic
capacity information in a register?
There will be considerable cost in agreeing the published capacities of the buildings, but this
is unavoidable. The cost of maintaining a database should be reasonable.
Liability issues (from assessors) must be addressed, or the publication of outcomes is likely
to lead to the assessments being over>conservative.
Proposal 4: The current national earthquake5prone building threshold (one5third of the
requirement for new buildings, often referred to as 33 per cent NBS) would not be
changed. However, it is proposed to establish a mandatory national requirement for all
buildings to be strengthened to above the current threshold, or demolished, within a
defined time period.
10. Does the current earthquake>prone building threshold (33 per cent of the requirement for
new buildings) strike a reasonable balance between protecting people from harm and the
costs of upgrading or removing the estimated 15,000 > 25,000 buildings likely to be below
this line?
No. Refer to the notes preceding these questions and answers and the separate SESOC
discussion paper. The current EPB approach is discussed and alternatives are presented that
have greater focus on achieving the desired outcome.
Not addressed in the report is that buildings of higher importance levels should be required
to achieve higher levels of compliance, or to reduce the use. So, for example, it is not
acceptable for an IL4 emergency facility to achieve only 33% > if such a building is below
strength, it should either be re>designated IL2 and its emergency use transferred to another
building, or it should be strengthened to full compliance, so that it is available to fulfil its
function when required. Refer Brook, 20076. Furthermore, it is important to note that such
buildings may be compromised by neighbouring buildings and blocked access>ways if wider
consideration is not given to the surrounding buildings and streets. This requires greater
consideration.
6
Brook R.A., Kelly T.E & Mackenzie C.S.M., “Performance Based Assessment and Design Policy
Recommendations”, Proceedings of the New Zealand Society for Earthquake Engineering Conference, 2007.
Version 4
Page 11 of 20
A further matter to consider is that even if 33% remains as the assessment threshold load,
should it be acceptable to strengthen to only 34% in order to meet the Act. A strong
argument could be made to suggest that in the long run, if any upgrading work is required, it
should increase the building capacity significantly, to provide higher levels of safety. This
would support an assessment level of 33% and a recommended strengthening level of 67%
(which should also allow future change of use without further structural work, provided
other requirements are met). However, consideration should be given to the overall risk
assessment, to ensure that only the truly dangerous buildings are treated this way. This
should probably be covered explicitly in the Regulations.
11. Should the requirement for earthquake>prone buildings to be strengthened or demolished
take precedence over all other legal, regulatory and planning requirements, such as those
designed to protect buildings of heritage or local character?
No. Acting in haste to reduce risk may cause decisions to be made that will be regretted in
time. There is a wider public debate to consider the importance of heritage buildings, after
which this can be revisited. This decision should not be made without that debate, but
heritage buildings should still be subject to minimum life safety standards, within the Risk
Assessment framework
Although ownership of such buildings may be privately held, there is a public good aspect to
this, which may necessitate the formation of a public private partnership approach to
upgrading such buildings. Such an approach may take some time to implement, but it would
be unreasonable to insist on short>term upgrade or demolition.
12. Should local authorities have the power to require higher levels of strengthening than the
earthquake>prone building threshold, or strengthening within shorter timeframes than the
legally defined period?
The EPB Act and Regulations must be set to an appropriate level in reflection of the Risk
Assessment framework that is to be developed. There should be consistency throughout the
country in setting a minimum standard. Communities that wish to reduce their seismic risk
further may choose to do so, but if so, a legal framework must be in place that will support
this and it should only be done on an informed basis, once the actual risks and costs are
understood, i.e. after the Risk Assessment framework is available.
13. Should certain features of unreinforced masonry buildings, such as chimneys and parapets,
be required to be strengthened to a higher level?
Yes. Refer to the SESOC discussion paper. For certain features, the strengthening should
be such as to improve the behaviour, not just to add capacity, if the behaviour results in a
brittle failure regardless of the load level. For masonry walls under face loading for
example, installation of ties at closer centres than required to meet 33% would change the
failure mode from failure in the ties themselves, to failure of the wall itself, which is
generally less catastrophic. Components such as chimneys and parapets are particularly
hazardous features and warrant prompt attention. Thus securing of these hazardous features
should also be required to be on a shorter time frame (say 5 years in lieu of 10).
Version 4
Page 12 of 20
Proposal 5: All buildings would be strengthened to be no longer earthquake5prone, or
be demolished, within 15 years of the legislation taking effect (up to five years for local
authorities to complete seismic capacity ratings, followed by 10 years for owners to
strengthen or demolish buildings).
Proposal 6: Strengthening would be carried out faster for certain buildings (e.g.,
buildings on transport routes identified as critical in an emergency).
Proposal 7: Owners of buildings assessed as earthquake5prone would have to submit
a plan for strengthening or demolition within 12 months.
14. Is it reasonable and practical for owners of earthquake>prone buildings to meet the following
timeframes:
• 12 months to submit plans for either strengthening or demolishing the building?
• 10 years from the date of the seismic capacity rating to strengthen or demolish?
No. Under the current environment, with so much resource tied up in the Christchurch
recovery, this may be impractical, especially for some of the smaller centres. It is suggested
that the timing for submission of plans be relaxed, or otherwise determined on a sliding
scale so that the more critical/dangerous buildings be dealt with as a priority. This would
spread the workload for response and strengthening construction. However the first priority
remains the development of a Risk Assessment framework as outlined above, which will
allow decisions about timeframes to be made on an informed basis.
15. What additional powers would local authorities require to enforce the proposed
requirements?
The Act may need to be strengthened in a number of ways to enable implementation of
policies, but that is a matter for MBIE to determine with the local authorities.
16. Should local authorities be able to require faster action on buildings of strategic importance,
such as those:
• located on transport routes identified as critical in an emergency
• with important public, social and economic functions, such as schools and police stations
• with post>earthquake recovery functions, such as civil defence centres and hospitals.
Yes, in accordance with the significance of the building or its use.
Faster action should be required for any remaining Importance Level 4 essential facilities
that have not been dealt with, and these should be required to achieve a higher level of
compliance as noted above. Similarly, any building that may affect the provision of
emergency resources should be considered with more urgency. This includes both buildings
on essential transport routes and any which are adjacent to IL4 buildings.
Buildings with high value contents under public ownership (IL3) should be considered
separately, with the possibility of re>locating the contents until such time as strengthening is
completed.
Once again, these are issues that must be addressed within the Risk Assessment framework
outlined above.
Version 4
Page 13 of 20
17. Should all unreinforced masonry buildings require strengthening more quickly than other
earthquake>prone buildings?
Consideration may be given to prioritising buildings according to risk, but it could be noted
that there may be other buildings which, by virtue of the number of occupants or
construction type, represent a higher risk that URMs. This should be addressed by the Risk
Assessment framework.
Proposal 8: Certain buildings could be exempted or be given longer time to strengthen,
e.g., low5use rural churches or farm buildings with little passing traffic.
18. Should the owners of certain specified types of earthquake>prone buildings be able to apply
to local authorities for exemptions or time extensions to the requirement to strengthen or
demolish?
Yes, provided this fits within the Risk Assessment framework. If a building has occasional
use only, and is isolated or can be safely cordoned off, a longer timeframe could be
considered, particularly if this means resource will be targeted to higher risk buildings. The
level of occupancy (combination of number and duration) should be a factor included when
determining risk.
19. If yes, what are your views on the following possible criteria:
• the building is used only by the owner, or by persons directly employed by the owner, on
an occasional or infrequent basis
• the building is used only occasionally (less than eight hours per week), and by less than
50 people at any one time
AND in each circumstance above
• all users are notified that the building is likely to collapse in a moderate earthquake
• the building is not a dwelling
• the building is not a school or hospital and does not have a post>disaster recovery
function
• there is no risk of the building partially or fully collapsing onto a public walkway,
transport route or a neighbouring building or public amenity
• effective mitigation measures have been put in place to protect building users from the
risk of collapse in a moderate earthquake?
These criteria are reasonable, except perhaps the last, which may require work that
effectively means the building may as well be brought fully into compliance; and which
requires assessment of a collapse limit state, which is not practical. We note that a building
that is a dwelling will generally not be considered as a potential EPB in any case and so this
criterion is redundant.
However, if a building as a whole is granted an exemption or extension and particularly
hazardous features (e.g. chimneys and parapets) should nevertheless be required to be
secured within the normal time>frame.
More critically, many of the buildings types mentioned e.g., low>use rural churches or farm
buildings (including the illustrating pictures) are likely to be constructed primarily of
lightweight timber framing. Such buildings may have a low assessable capacity but in
reality are highly unlikely to fall down in a moderate earthquake. Consideration should
simply be given to changing the assessment criteria so that these buildings are not
considered earthquake prone.
Version 4
Page 14 of 20
Proposal 9: Central government would have a much greater role in guiding and
supporting local authorities and building owners, as well as in public education and
information.
20. Are the advice, information and education activities proposed for central and local
government agencies sufficient to help ensure effective implementation of the new
earthquake>prone building system?
We would envisage both Central Government and Local Authorities will require support
from the SESOC membership in developing and implementing the new earthquake>prone
building system.
Past iterations of the policy have been ineffective, and at successive changes, the clock has
been reset to zero, so too many buildings have been able to escape detection or mitigation.
The communication of seismic risk has been poor, at least inasmuch as there is poor
understanding of the issues and what a good outcome may be. However, the proposals are
in principle satisfactory.
Proposal 1 refers to local authority assessments using ‘a standard methodology developed by
central government’. The development of this methodology is a key to the successful
implementation of the proposals. Previous DBH Advisory notes, sent to Local Authorities
(e.g. Egress Stairs, Non Ductile Columns) have resulted in the Local Authorities sending out
‘blanket’ letters requiring the building owners to engage consulting engineers to resolve
these matters. The ‘buck’ is efficiently passed from Central Government to Local
Authorities to Building Owners. The ‘standard methodology’ needs to be effective in
screening out high risk buildings, whether that risk is a result of brick construction, non>
ductile columns, non>compliant stairs, inadequate diaphragms, hollowcore floors, liquefiable
soils, non>ductile mesh, weak beam/column joints, inadequate services restraints, beam>
elongation prone buildings and other highly technical matters.
It will be necessary to partner with organisations such as the Property Council, IPENZ,
NZSEE and SESOC; to get their membership on side and to make use of their expertise and
ability to communicate with other key sector groups; and in providing support and training.
Views are sought on whether the current Building Act fire and disability upgrade
requirements are, in practice, a barrier to building owners deciding to carry out
earthquake strengthening work.
21. Are current requirements to upgrade buildings to “as nearly as reasonably practicable” to
Building Code fire and disabled access requirements a disincentive or barrier to owners
planning to earthquake>strengthen existing buildings?
Experience has shown that these requirements can cause seismic upgrade projects to be
scrapped or scaled back. This is particularly the case in heritage buildings, where access
requirements such as ramps conflict with heritage objectives, making it all but impossible to
get through both RMA and Building Consent requirements. At the very least, they can add
considerable cost and time to the design and consenting process with no long>term benefit to
the actual work that is executed. The ideal outcome may be to minimise the lost value of
consulting and consenting effort that is spent on building seismic upgrades to prove that
nothing else needs to be done.
Version 4
Page 15 of 20
22. Should local authorities be able to grant building consents for earthquake strengthening
without triggering the requirement to upgrade the building towards Building Code fire
escape and disabled access and facilities requirements?
Yes. A reasonable approach may allow owners to complete seismic upgrade work where the
building is otherwise going to remain unchanged. It may be possible to add a requirement to
demonstrate undue financial hardship if these had to be complied with, but even completing
the design and costing exercise may cause hardship. This could be related to use and
tenancy e.g. if there is no major change proposed other than the strengthening, the other
upgrade work will not be required. Access and fire upgrade requirements should be assessed
on their own merits and should not be triggered by the entirely unrelated matter of
earthquake strength. However building owners should be encouraged to consider these
matters when determining the strengthening plans in order to avoid un>necessary additional
expense of carrying them out at different periods.
23. Should any change apply to both fire escape and disabled access and facilities requirements,
or to disabled access and facilities requirements only, i.e., retain the current fire escape
upgrade requirements?
Yes, to both. Although there is already discretion in the use of “as nearly as is reasonably
practicable” in S112, experience shows that this is unevenly applied. Further guidance
should be provided on what is intended by “as nearly as is reasonably practicable”.
A significant issue is the diversion of funds away from the actual work that is required. So
even though a building undergoing seismic upgrade (with no other change) may be shown to
not require any other upgrade for fire escape and disabled access, there may be considerable
study required to demonstrate this. This flows through to consenting, where there may be
considerable further review and discussion required to reach agreement on that point. A
possible way through this may be to set some minimum standards (as has been done for
earthquake prone buildings with the 33%NBS standard) that can be simply qualitatively
assessed, in order to minimise this effort and diversion of resource.
A further suggestion is that although fire escape is critical to life safety, if the buildings are
truly dangerous, there are other means of acting upon this independently of whether seismic
work is proposed for the building. It could be assumed that if notice of dangerous fire
conditions has not been served, the building is considered to be adequately safe. Therefore
there may be no need to encumber seismic upgrade (a separate safety issue) with this
requirement, unless the work proposed will in some way reduce the existing level of
compliance.
A further matter for consideration is the risk posed. A significant factor in this is the use of
the building. So for example the sleeping accommodation purpose group may be considered
a particularly high risk group for fire escape and frequently, the most likely to be at risk of
fire escape issues are also the most likely to be earthquake prone. So a full fire study could
be warranted for such buildings, whereas lower (fire escape) risk purpose groups, could be
exempted from a full fire review, possibly with the provision of further criteria to be met. A
broad risk categorisation may therefore form the basis of a more qualitative review.
On balance, it is recommended that each of these issues is dealt with separately.
Version 4
Page 16 of 20
24. What would be the costs and other implications of de>linking earthquake strengthening from
current Building Code fire and disabled access requirements?
This will vary considerably from building to building. Previous reviews prepared for the
Christchurch City Council have observed the cost of possible fire upgrading work to be up
to 10% of the value of the strengthening work. Although disabled access has not been
separately identified in the same way, this could easily reach the same levels, or even more
if for example a lift was required to meet the provisions. At the very least, there is cost
associated with the assessment before any other work can be done.
Views are sought on how important heritage buildings can be preserved while also
being made safer.
25. When considering listing heritage buildings on district plans, what factors should local
authorities consider when balancing heritage values with safety concerns?
There appears to be no valid reason why seismic safety should be a criterion for
consideration in assessing whether a building should be listed. However there is a need to
acknowledge heritage in considering how seismic safety concerns may be resolved.
A desirable outcome is that seismic upgrading of heritage buildings is done in a sympathetic
way, but it is important that seismic upgrading work is not deferred because of perceived
heritage impact. Therefore there needs to be a better mechanism for reaching compromise
between seismic safety and heritage, to be administered by the TAs, without simply
imposing greater and greater expense.
26. What assistance or guidance will be required for owners, local authorities and communities
to make informed decisions on strengthening heritage buildings in their districts?
•
•
•
•
A greater understanding of the nature of seismic risk – not simply the %NBS rating of
buildings. This requires better communication between engineers and non>engineers.
A more consistent application of ICOMOS principles with guidance on what
compromises may be acceptable. A constant source of frustration to engineers is the
inconsistency where ICOMOS allows new work (such as seismic strengthening) to be
exposed and clearly demonstrated as new, whereas many heritage advocates require it to
be hidden, which is often impractical or compromises the outcome too much, and adds
considerable expense.
Engineers working in this field need to have demonstrable expertise. This may entail the
development of more assessment and strengthening tools, and/or further training.
However, there is a finite population of such buildings, so a matching of existing
skillsets to the need is required.
Clarification of the relationship between Building Act and RMA with respect to heritage
provisions – which trumps which? If safety is agreed to be the principal driver, then it
must be enabled by the regulations, not stalemated.
27. What barriers deter heritage building owners from strengthening their buildings?
Principally, cost. There is an underlying question here to be resolved for those heritage
buildings in private ownership – probably the majority. Assuming that the retention of
heritage buildings is for the long term wider public benefit, is it reasonable that private
owners are expected to pay the cost of retention? Or more specifically in this debate, the
Version 4
Page 17 of 20
additional cost of meeting heritage objectives when strengthening their buildings, given that
they may generate no further value by doing so?
Alternative funding sources could be identified and publicised. This links to the points in
the response to qn. 27 below. Although there is little indication that government would
support it, consideration should also be given to the retention of heritage as a public good
and therefore that there should be an element of public funding available. There may be a
variety of ways of achieving this, including tax breaks (local or central), direct grants
(maybe repayable in whole or in part if the building is subsequently sold), or interest>free
loans. This requires public debate and must obviously be balanced against other public good
issues.
The secondary issue is the time spent in meeting additional demands set by RMA
requirements, which are considerably more onerous than for equivalent non>heritage
buildings.
Sometimes, a heritage building may be a “terrace” style building with multiple owners for
individual lots, but not necessarily in a Body Corporate structure. It is not possible to
upgrade the individual properties in isolation nor is it realistic for an individual owner to
prop up its neighbour. A separate solution may be required for these situations.
By the same token, it could easily be argued in most cases that buildings were likely to have
been heritage buildings when they were purchased and so these matters should be reflected
in the value of the building already.
28. Do heritage rules (for example, those in district plans) deter owners from strengthening
heritage buildings?
In most cases, seismic strengthening of buildings cannot be justified on a financial basis –
after spending what may be a considerable sum of money to strengthen, the building remains
as it was with no greater commercial return being generated. (At least this was the case pre>
Christchurch. With more attention to seismic safety by users, the likelihood is that EPB
owners will now have to lower rents, meaning this money may have to be spent just to get
them back to where they were – hardly an attractive economic proposition)
With the already poor economic return from strengthening, if the costs are increased
significantly for heritage reasons, it is a significant deterrent.
Many owners have simply continued to hold buildings at low rents rather than face the time
and cost of upgrade for only modest if any increase in rents. However, others have been
able to achieve good results, finding commercially viable uses for sensitively upgraded
buildings. Such projects could be identified and used as case studies for future
redevelopment.
29. What are the costs and benefits of setting consistent rules across the country for
strengthening heritage buildings?
The major benefit is in owners and advisers knowing what is required before committing to
the process, whereas the current situation provides no such certainty, resulting in many
projects being abandoned before even completing the consent process. Refer also to our
response to question 4.
Version 4
Page 18 of 20
Views are sought on the Royal Commission’s recommendation to allow local
authorities the power, following consultation with their communities, to adopt and
enforce policies to require specific hazardous elements on residential buildings to be
dealt with within a specified timeframe.
30. Should local authorities have the power, following consultation with their communities, to
adopt and enforce policies to require specific hazardous elements on residential buildings to
be dealt with within a specified timeframe?
Enforcement of rules on private residential buildings is not necessarily the best outcome, but
there should be a consultation process. There are a number of hazardous elements, notably
brick chimneys, but including heavy tile roofs, that have caused considerable problems,
including death, and that have been well>known for many years. At the very least, there
should be a significant communication of these issues to the public.
Consideration may be given to whether this might be dealt with separately with respect to
rental accommodation, where tenants may not have the ability to impose safety related
changes.
A means of improving uptake may be to consider co>funding arrangements similar to the
EECA Heat Smart programme, possible in association with EQC and/or building insurers,
with obvious benefit to both in limiting future liabilities.
Other questions
31. What would the proposed changes mean for you?
SESOC represents the structural engineering profession and many of its members are
involved in consultancy and design services for buildings. This impacts directly on the
profession and as such SESOC is directly affected by any changes.
32. Are you aware of any problems with current policy and practice around earthquake>prone
buildings, other than those identified in this document?
The consultation document touches on most of the major issues in respect of earthquake
prone buildings. Any that have not been dealt with are addressed in our separate discussion
paper.
33. Do you agree with the following objectives for changes to the existing earthquake>prone
buildings system:
• reduce the risk – to an acceptable level > of people dying and being injured in or by
buildings that are likely to collapse in moderate to large earthquakes.
• ensure that building owners and users have access to good information on the strength of
buildings they own and use, to help them make good decisions about building resilience
and their use of the building.
This is addressed in the first part of this submission as well as in our separate discussion
paper, appended.
Risk reduction is essential, but it is important that there is an appropriate means of
addressing the real issues, commencing with the implementation of a Risk Assessment
framework as described above. The first bullet addresses this point, but we note that the
Version 4
Page 19 of 20
term “likely to collapse” is a problem in engineering terms, and is probably redundant in this
sentence. SESOC supports the explicit consideration of larger events (than the hypothetical
‘moderate earthquake’) in this bullet.
The second bullet point is more of an issue. SESOC agrees in principle that there should be
better communication and information available to both building owners and users.
However, ‘strength’ is not an appropriate measure of safety. Resilience is certainly the
desired outcome, but a focus on capacity, as currently reflected in %NBS ratings, does not
identify the most dangerous of buildings and frequently brands buildings that will in fact
perform adequately where buildings of higher capacity may not. This is partly an outcome
of the assessment processes in use, partly a communication issue.
Feedback form completed by:
Structural Engineering Society of New Zealand (SESOC)
C/> John Hare, President
Holmes Consulting Group LP
PO Box 6718
Upper Riccarton
Christchurch 8442
Version 4
Page 20 of 20
Appendix A
Earthquake Prone Buildings
Discussion Paper
Executive summary
The need to advance the implementation of seismic upgrading of our most at risk buildings
was clearly demonstrated by the Canterbury earthquakes, in particular the February 22nd 2011
event. However the aftermath of the earthquakes has also demonstrated that the current
approach has shortcomings that mean we may not be focusing on the most dangerous
buildings. Furthermore, society’s current collective obsession with estimations of building
capacity with respect to a theoretical minimum shaking level is misleading the public,
misdirecting our engineering effort and potentially diverting funding away from its best use –
to mitigate the problems.
It has been observed many times that buildings rarely fail because they are below strength.
Instead, it is more typically a combination of specific vulnerabilities that cause them to fail,
sometimes catastrophically. If this is accepted, it follows that we should use an approach that
recognises this. It has also become apparent that even though small frequently occurring
earthquakes may affect the poorest of our building stock, in the event of a larger earthquake in
a populated centre, there are many larger buildings that were not previously considered high
risk, that have caused the most deaths.
The earthquake prone building legislation is now under review, following the publication of
the Canterbury Earthquakes Royal Commission (CERC) reports. In consideration of the
CERC recommendations, the proposed emphasis of the review is on implementation rather
than identification, but it is considered that this may result in some of our worst potential
problems escaping detection, and in many buildings being classified as earthquake prone that
do not present a significant life safety hazard.
A new approach is recommended that changes the emphasis of the assessments from
assessment of capacity to identification of key vulnerabilities. These are the vulnerabilities
that represent the greatest overall risk to building users, in a broader risk consideration than
the current approach considers. It combines elements of the existing approach with a more
qualitative assessment of buildings, based in part on experiences from the Christchurch
earthquakes.
By adopting this approach, it should be possible to exclude from consideration building types
that do not represent a high risk to society, but still to target those which are genuinely
dangerous. Building types that do not present a significant life safety risk but which may be
considered earthquake prone currently include for example older timber framed buildings
with low assessed capacities. Buildings that may be considered dangerous but which are not
currently deemed earthquake prone include for example non ductile concrete buildings with
poorly detailed gravity columns, similar to the CTV building. Although these buildings will
probably survive a moderate earthquake, they may fail catastrophically with greater shaking.
Due to their greater occupancy levels, these buildings collectively may have a greater
probability of causing loss of life.
Further targeting of mitigation methods to typical building types in smaller centres is also
recommended that may reduce the impact of the legislation on these vulnerable communities.
EPBs Feb 27 2013_R4.doc
1
Revision 4, Feb 24, 2013
The difficulties of designing and implementing such a wide set of changes are alleviated in
part by a considerable amount of the necessary development work having already been started
in connection with the Detailed Engineering Evaluations Guidelines, and the possible
incorporation of aspects of similar assessment approaches from the US. A progressive
assessment and roll out of technical guidelines is possible under this approach, provided that
it is supported by the appropriate re structuring of the Act and Regulations.
1
Introduction
This paper has been prepared in response to the MBIE consultation document1 addressing the
current Earthquake Prone Building legislation and regulations. This includes discussion of
appropriate load levels for retrofit and associated costs.
2
What is an Earthquake Prone Building?
An Earthquake Prone Building (EPB) is defined by the Building Act2 (s122). Technically, it
is a building that may collapse, leading to loss of life, in a “moderate” earthquake. Only
residential buildings (except those which are of two storeys or more and contain three of more
units) are exempt from consideration3. A moderate earthquake is defined in the Regulations
as an earthquake of the same duration, but of one third the intensity, as the design earthquake
for an equivalent new building.
Although the definition may sound precise, in engineering terms it is not, for a number of
reasons. Firstly, it is not practically possible to pinpoint ‘collapse’. Hence this has been re
defined by the NZ Society for Earthquake Engineering (NZSEE) as a building which has its
ultimate limit state (ULS) capacity exceeded in a moderate earthquake. ULS capacity is used
in structural design, and in force based design or assessment, is the point at which a structure
has fully ‘yielded’, i.e. it has developed a full structural failure mechanism4. Depending on its
ductility, it may continue to deform without collapsing, while still offering the same
resistance. The ULS capacity is clearly less than the collapse load, but is a reasonably
understood and definable number. A key point is that although the Act requires that the
ultimate capacity is exceeded AND the building would be likely to collapse, this is a highly
subjective assessment, and therefore in practice, is inconsistently applied.
In assessing a potential EPB, using a conventional force based methodology, the designer
must make an estimation of the ductility of the building, use this to calculate the demand
(seismic load) that would be applied to an equivalent new building and then assess the
capacity (strength or displacement) of the building. This is then generally presented as the
‘percentage of new building standard’ (%NBS).
1
MBIE, Building Seismic Performance Consultation Document, December 2012
Building Act 2004, Department of Building and Housing
3
Although the Royal Commission has recommended that consideration be extended to residential buildings with
unreinforced masonry elements
4
This may require further definition, even for engineers. A single element reaching its capacity does not mean
that a full mechanism has developed. Redistribution (subject to reasonable limits) allows the building to
continue to deform within the elastic range of unyielded elements. However this may be limited by the ductility
of the elements, which may become complicated in buildings with mixed systems. While displacement based
assessment is generally a better means of evaluating the performance of such systems, it should be recognised
that most engineers still tend to think in terms of force based assessment.
2
EPBs Feb 27 2013_R4.doc
2
Revision 4, Feb 24, 2013
In common terms, if the building capacity is less than 33%NBS, the building is earthquake
prone. This excludes consideration of collapse, noting as above, that collapse is not a
condition that can be readily estimated in engineering terms. This is not the way it should be.
It should be noted that 100%NBS as calculated for an existing building is not the same as the
actual likely new building capacity. This is because new building design is calibrated to
ensure that new buildings are statistically well in excess of the code minimum values. So the
term %NBS is somewhat misleading. This has also been addressed by the Royal
Commission, which has instead recommended %ULS be used.
3
Why was the term Earthquake Prone Building developed?
Much of our building stock predates mandated seismic design requirements, at least in the
form that we now apply them. Many such buildings are lacking in lateral load resistance, and
have little ductility, meaning that they may fail in even relatively small earthquakes,
sometimes catastrophically. While it is well understood that unreinforced masonry buildings
may be considered dangerous during earthquakes, over the last few decades (And as
illustrated graphically in the Christchurch earthquakes, there has been increased awareness of
the dangers of other forms of building also, particularly non ductile concrete buildings.
Clearly there is a need to identify our most seriously at risk buildings and mitigate the seismic
hazard, either through removal or upgrading. Previously, buildings designed to predecessor
codes or other at risk buildings were informally known as ‘earthquake risk buildings’5. This
term is still used and is generally applied to buildings with less than 67% of the capacity of a
new building. However given the high proportion of such buildings, it was considered that
there was a need to identify the very worst of the poor buildings, hence the development of
the earthquake prone building definition. This was the first step in the legislative process to
identify and mitigate this problem, imposed in the previous 1991 Building Act.
A critical point to consider is that in the (previous) 1991 Building Act, EPB considerations
were limited to buildings constructed “wholly or substantially of unreinforced concrete or
unreinforced masonry”. These were considered the most dangerous of buildings at that time.
These requirements dated originally from the 1968 Municipal Corporations Act. The
extension of these considerations to all building types in 2004, at the same time as the
threshold load levels were increased, has considerably widened the net,.
4
What is wrong with this picture?
Technically, nothing, provided the limitations are understood. But a significant issue is that
buildings rarely collapse solely because they have a low capacity. Collapse is more likely to
result when a building has one or more specific weaknesses (vulnerabilities) which make it
unable to continue to support gravity load when a critical displacement has been exceeded. It
should be considered that the two most lethal buildings in the Christchurch earthquakes (PGC
and CTV) both had capacities exceeding 33%NBS.
5
The earliest fore runner of the current EPB legislation is the Municipal Corporations Act (1968), followed by
the Local Government Act (1974). The Building Act 1991 introduced the earthquake prone building
terminology, but with a considerably lower threshold load limit. It could also be observed that every time the
legislation has changed, the clock has been set back to zero, in many jurisdictions.
EPBs Feb 27 2013_R4.doc
3
Revision 4, Feb 24, 2013
It is also of note that many buildings that have survived the earthquake are now being
assessed as EPBs, in some cases without significant damage. These facts should cause us to
question our assessment approach, which currently relies solely on assessment of capacity
against a low level of loading. The question may be asked, whether the legislation in its
current form and application is serving its purpose.
For example, it is particularly evident that light timber framed buildings have consistently
performed well in earthquakes, but that many older examples have low nominal capacities
based on conventional assessment techniques. Given that most of our homes and many of our
educational facilities are constructed of light timber framing, this is a serious concern.
Under previous legislation (which was typically focused on unreinforced masonry), such
buildings were exempt from consideration. However, the current Act has extended the scope
of consideration to all structural forms. The underlying assumption is presumably that this
provides equal consideration of risk, but that is only the case if the assessment tool (in this
case, ULS capacity) provides a consistent measure, which it cannot. The 2004 legislation
resulted from concerns out of the Northridge and Kobe earthquake, that there could be similar
modern buildings in New Zealand with vulnerabilities that could lead to collapse and loss of
life.
In the more common event of a single large earthquake followed by a relatively short
aftershock period, absence of damage may be explained by matters such as direction of attack
and characteristics of the shaking. But in or near central Christchurch, there have been in
excess of 50 shallow nearby earthquakes of Magnitude 5 or more, that may have approached
or exceeded the EPB threshold level. This should influence our thinking of what an EPB is,
in post earthquake Christchurch.
Some of the reasons for the apparent gap between engineers’ evaluations and the actual
outcomes are simply explicable:
•
Firstly, all quantitative methods of assessment have to make reasonable assumptions
about material strength and make use of algorithms that have been established by
research or first principles. Even though there has been some calibration of this effect,
by assessing average values (instead of lower bound) in the guidelines, they will
generally still give conservative answers. In reality, materials and key details are
likely to vary considerably in older buildings.
•
Secondly, many of the methods used to assess existing buildings are not really suitable
for older buildings that use archaic materials or non engineered systems, so engineers
have to make approximations. Once again, these will tend to be conservative.
•
Thirdly, most engineering methods have to use approximations of the structural form
that, of necessity, will simplify the structure to something that can be readily modelled
and assessed. To try and develop a full model that identified and quantified every
possible load path would be too time consuming and the outcome would not generally
justify the computational effort. However this means that often, the assessment may
not take into account a large proportion of a building’s capacity. Instead, a more
holistic view is needed.
EPBs Feb 27 2013_R4.doc
4
Revision 4, Feb 24, 2013
•
Fourthly, in the case of age based assessment methods (such as the IEP) of light
weight structures, earthquake may not have been the governing design case. In that
situation, the factoring of the earthquake design load as a proxy for the capacity is
likely to have missed the higher (actual) design load input and therefore capacity.
No data exists to determine how conservative these outcomes may be, but my estimate is that
of 100 buildings assessed to be just earthquake prone, less than 10 might actually fail in a
moderate earthquake, and at the upper end, some will be able to withstand actions in excess of
100%NBS. (This may be a contentious opinion, but could be justified by observation. This
information may in the future be backed up by collation of the DEE reports, for which a
substantial research project is required). We should look more closely at what survived, how
well and why.
An observation that goes with this is that most engineers are designers, not analysts. As such,
they are accustomed to meeting or exceeding a target capacity in design, but how much they
exceed it by is irrelevant (except as it relates to cost). Conversely, in assessment, they are
being asked to deliver a specific capacity value, with a narrower margin. Prior to the
earthquakes, only a small minority of engineers were completing assessments of existing
buildings, and a number of these had developed significant IP to assist with the process,
which is technically more demanding than the design of a new building. Now, most
engineers are completing existing building evaluations, many with little experience or
familiarity with the process.
Another possibly more significant factor in this is the estimation of the seismic hazard itself.
The design spectra that engineers work with are nice straight lines and smoothed curves,
applied on an area wide basis. But the reality of earthquake shaking is vastly different and
can vary considerably over short distances. Identical buildings which may be only a short
distance from one another can display completely different performance in the same event,
reflecting the impact of site effects, differing soil profiles and orientation of shaking.
Moreover, the ‘moderate earthquake’ is a fictional event. The idea of an earthquake of one
third the shaking intensity but of the same duration as the design earthquake is a device for
assessment, but only really possible to reproduce in theoretical computer simulations.
Taking all the above into account, it is possible to begin to understand why it is that
engineering assessments are generally, but not always, very conservative. But it raises a
broader question – are these assessments achieving what we need them to, i.e. effective
identification and mitigation of the worst of our seismic hazard, within reasonable cost limits?
Not always and possibly not often. There is often too much emphasis on the numbers rather
than the performance of a building and the improvements that can be made.
5
What Buildings are likely to be earthquake prone?
As noted above, only non residential buildings are currently subject to consideration as
earthquake prone buildings.
Virtually all unreinforced masonry (URM) buildings are likely to be earthquake prone, unless
they have already been retrofitted. URMs commonly drop section of wall, parapets and other
appendages in earthquakes. This form of building was largely abandoned after the Napier
earthquake of 1931.
EPBs Feb 27 2013_R4.doc
5
Revision 4, Feb 24, 2013
2005
2000
1995
1990
1985
1980
1975
1970
1965
1960
1955
1950
1945
1940
1935
1930
1925
1920
1915
1910
1905
1900
1895
1890
1885
1880
1875
1870
1865
1860
1855
1850
It is less easy to generically predict other categories of building according to material type. It
is instead more related to age and form. Milestone years in building design are commonly
considered to be 1965 and 1976, respectively when seismic design was formally adopted and
when capacity design was introduced, but even those dates can sometimes be misleading.
The figure below illustrates a subjective assessment of the most common forms of
construction in New Zealand according to construction type with approximate timelines.
A. B uild ing T y pe
U n re in forc e d Ma s on ry
R ive te d s te el m o m en t fram e s
W eld e d a n d B olte d s te e l m o m e nt fra m es
C o nc re te F ra m e w ith infill
N o n+d u ctile co n c rete m om e n t f ra m e
D u ctile c o n crete m om e n t fra m e s
T itl p a n el s ing le s to rey
T ilt p a n el m u lti+store y
C o nc re te s h ea r wa ll stru c ture s
L igh tly rein fo rced p artially filled c on c re te m a s on ry
F u lly fille d c o nc re te m a s o nry
B . Ele me n t T yp e
P reca s t co n crete floo r sys tem s
H e av y m a son ry o r p la ste r clad d in g
P reca s t Cla dd in g s ys te m s
Pro b ab ly E arth q ua k e Pro ne
Po s sibly E a rth q ua k e Pro n e
Ma y h a ve som e issu e s
Pro b ab ly n ot E arth q ua ke Pro n e
Figure 1: Building Types and Approximate Timelines6
There is a disturbing incidence of relatively modern buildings being assessed as EPBs. This
is possibly in part due to an element of earthquake induced conservatism that pervades much
engineering practice currently; partly a reflection that mistakes can happen in design, where in
effect, every building is a prototype. Additionally, depending on your definition of modern, it
must be remembered that many of our Standards have had significant amendments over the
last 20 years, meaning that buildings that may have been in full compliance with earlier
Standards may now be significantly out of compliance with current Standards.
However a distinction could be drawn between those buildings that are earthquake prone as a
whole (typically buildings that were designed before the advent of modern codes) and those
which may fail to comply in a few details only. The former may require wholesale upgrading,
whereas the latter may require rectification of only a few details, subject to the overall
behaviour being acceptable.
6
From CCC Earthquake Prone Building Review Preliminary Scoping Report – report to Christchurch City
Council, Holmes Consulting Group, June 2009.
EPBs Feb 27 2013_R4.doc
6
Revision 4, Feb 24, 2013
Examples of non URM buildings that may be earthquake prone as a whole include:
•
Non ductile concrete buildings: buildings that may have strength, but little of the
ductility of modern reinforced concrete buildings. This is normally associated with
columns that have few stirrups (confining steel, wrapped around the main longitudinal
reinforcement, but may also impact on other members) and is more typically found
prior to 1976, but also in some buildings designed between 1984 and 1995, due to a
code problem.
•
Partially filled masonry: reinforced concrete block buildings that do not have all cells
filled with grout. This type of building has been a poor performer in the Christchurch
earthquakes and is known to have been plagued with construction quality issues that
cannot be seen.
Examples of detailing issues that may lead to a building being considered earthquake prone,
that is otherwise generally compliant include:
•
Inadequate stair clearances and support. This was highlighted by the failure of the
Forsyth Barr stairs in the Feb 22nd 2011 earthquake. Note however that in most cases,
even though the potential exists for these stairs to fall in a large earthquake, it would
be unusual for these to be strictly earthquake prone7.
•
Lack of adequate seating for precast floor systems (including allowance for beam
elongation and support rotation). Similar to the above.
•
Inadequate roof bracing connection details. This is common in many older portal
frame/warehouse buildings, particularly those with heavy precast cladding panels.
•
Non ductile columns. In many older buildings, all columns may be non ductile, and
the building will be assessed accordingly, but a possibly more significant concern is
with structures like the CTV building that have non ductile columns within a separate
lateral system that is (at least intended to be) ductile.
Further guidance on the likely vulnerabilities of buildings, by type, is given in Appendix A of
the DEE Guidelines8.
Conversely, the question may be asked – which buildings are not earthquake prone? Since
the earthquakes, many lightweight timber framed structures have been assessed with very low
%NBS values. However, unless these buildings have heavy masonry elements (such as
firewalls, chimneys or heavy tile roofs), they have always performed well in earthquakes and
certainly do not present a life safety hazard. Moreover, if they were really so poor, they
would in many cases be prone to failure in severe winds, as this is often the governing design
load case. This is an example of a type of building that is being conservatively assessed; on
account of the inapplicability of most assessment methods and the generally high level of
redundancy in these structures.
7
The Engineering Advisory Group has proposed a multiplier (called the Kd factor) which artificially doubles the
demand on such details, which are considered to be critical safety or hazard elements.
8
Guidance on Detailed Engineering Evaluation of Earthquake Affected Non residential Buildings in Canterbury
Part 2 Evaluation Procedure, Draft prepared by the Engineering Advisory Group, Rev 7, 16 May 2012.
Available at www.sesoc.org.nz.
EPBs Feb 27 2013_R4.doc
7
Revision 4, Feb 24, 2013
In some cases, the assessment of earthquake proneness is because of the assessed likelihood
of liquefaction. However, although liquefaction may present a severe risk of failure through
excessive movement for heavy brittle buildings, it is highly unlikely to cause a life safety
hazard for lightweight buildings. It is understood that no buildings are assessed to have failed
causing life safety hazard due to liquefaction in the earthquakes.
6
Which Buildings should we be most concerned about?
The obvious truism is that we should be most concerned about the buildings that might kill us.
This is not so easy to determine if you consider some of the points raised above. The
difficulty is that %NBS addresses the building capacity with respect to earthquake shaking
intensity, but this is only one factor.
The PGC and CTV buildings have been assessed (since the earthquakes, with extensive
analysis9) as having had between 35 60%NBS and 40%NBS respectively and yet it has been
estimated that they were subjected to something close to 100%NBS demand in the September
4th 2010 earthquake (as were all of central Christchurch buildings), with minimal damage.
However their collapse in the February 22nd 2011 earthquake was sudden and catastrophic,
particularly in the case of the CTV building. In neither case was the September earthquake
concluded to have had a significant impact on their February collapse. This illustrates the
point made above, that the vulnerabilities of a specific building have more influence on its
safety than its assessed capacity.
Another significant contributing factor is the influence of structural redundancy. The more
alternative load paths that can develop, even in a building with limited capacity, the more
chance the building has to survive an earthquake. Conversely, too much reliance on a single
system or element may place a building at severe risk, particularly if the element then
underperforms.
Part of the difficulty of this is that focusing solely on a “moderate” earthquake may not be
very appropriate. This directly relates only to return period (hazard) and should be considered
in a wider risk context. The shortcoming is that consequence is not very well addressed by
this method.
In conventional risk management, risk is the product of probability and consequence.
However, risk needs a context. This requires us (society) to consider what outcomes we need
to achieve, within practical constraints that may include cost, the impact of wholesale change
and our changing views with the passage of time. In particular, we need to consider whether
we should maintain a focus on a relatively low level of demand (the moderate earthquake),
and exclude consideration of the potential for more catastrophic collapse of buildings that
may exceed the moderate earthquake threshold by only a nominal amount.
An observation is that we have had a number of moderate earthquakes in the last few
decades10, but they have caused relatively little damage and few deaths. This may be
consistent with the objectives of the Act, although it is questionable whether it is in any way
related to the Act or its predecessors, as most of these events have been in low population
9
Canterbury Earthquakes Royal Commission Final Report, Vols 2 & 6
Without attempting to be comprehensive, these might include Edgecumbe (1987), Weber (1990), Gisborne
(1993, 2007). Prior to the Canterbury earthquakes, the last fatalities in a New Zealand earthquake were 2 in the
Inangahua earthquake (1968).
10
EPBs Feb 27 2013_R4.doc
8
Revision 4, Feb 24, 2013
areas. However the earthquakes that have caused death tend to be of significantly greater
intensity. In such events, the buildings that may present the greatest threat to life are not
necessarily earthquake prone (under the current definition) but nor are they necessarily
difficult to identify and mitigate, if we were to amend the assessment process.
One of the significant pitfalls of an approach that exclusively considers capacity relative to a
moderate earthquake is that actual performance of the building is neglected beyond that point.
In other words, the argument can be made that provided a building exceeds 33%NBS, we
need not worry how it behaves beyond that. This is the most important lesson from
Canterbury. This point is illustrated in the figure below.
ULS
(100%NBS)
for ductile
structure
1
4
Load
6
3
5
2
Displacement
Figure 2: Load Displacement relationships for buildings11
Notes:
Line 1 represents a fully linear elastic approach, that is, the building has been designed to
simply resist the full applied load in proportion to the imposed displacement.
Line 2 represents a high ductility level. The required strength is reduced according to the
ductility, and capacity design is used to ensure that the building yields in a controlled fashion.
The design detailing provisions of the standards should ensure in the majority of cases that the
buildings will displace to significantly greater levels of displacement with acceptably low risk
of collapse.
11
From Guidance on Detailed Engineering Evaluation of Earthquake Affected Non residential Buildings in
Canterbury Part 2 Evaluation Procedure, Draft prepared by the Engineering Advisory Group, Rev 7, 16 May
2012. Available at www.sesoc.org.nz.
EPBs Feb 27 2013_R4.doc
9
Revision 4, Feb 24, 2013
Line 3 represents a building of limited ductility. If higher strengths are provided, designers
may reduce the detailing standards. However, this may mean that the margin between ULS
and collapse is reduced. This is explicitly checked in the concrete standard, at least in respect
of soft storey mechanisms, but is implicit in the steel standard.
Line 4 represents a structure that is designed to remain fully elastic for the ULS. Such
buildings are penalised (a higher Sp factor is specified) and are therefore required to have a
higher design capacity than a ductile structure. However, because there are no implicit or
explicit checks, there is no guarantee that they do not contain a critical structural weakness
beyond the design capacity.
Line 5 represents a building that may just exceed the EPB threshold. Even if similar margins
between ULS and collapse available in a new building are maintained it is apparent that there
may be little capability to survive anything other than a moderate earthquake, which is only a
little greater than a SLS event for a modern building.
Line 6 represents a building that may have been strengthened to 67%NBS. Because there is no
requirement to add ductility, the onset of collapse is still only marginally above the design
load.
For example, is it better to have a building that may collapse catastrophically at 60%, or one
that gradually grinds its way down, starting at 30%, but eventually exceeding 100% without
full collapse? The latter is what many URM buildings were observed to have done in
Christchurch. It could be argued that with a better education programme for the public on
what to do in an earthquake, more lives may have been saved by people not running out of
buildings which were dropping masonry on the outside, but otherwise remaining upright.
The concern is that the focus on %NBS and the moderate earthquake brings the assessment
directly to a numbers game. There is a misconception that all engineers use the same methods
and should get the same results, but that is not the case. Just like most other professions, there
is much ‘art’ and subjectivity involved. So one engineer’s 10% may be another engineer’s
50%. Training and peer review may compensate for this to a degree.
The problem with this is that while attention is focused on finding the highest number, it is
taken from the more critical activity, of identifying specific vulnerabilities that may lead to
potential collapse, regardless of load input. While there may still need to be a threshold
limiting level of shaking for this, if the emphasis is shifted from the capacity to the
vulnerability, we would get closer to knowing which buildings may be most dangerous in the
future.
Christchurch has also taught a further lesson, that a focus on life safety may not be the only
measure that should be considered. The loss of 185 lives and the injuries to a further 6,500
people were a deep human tragedy, but in time may be overshadowed by the severe impact on
the local economy and the loss of amenity for the city. The silver lining for Christchurch was
the extraordinarily high level of insurance cover, meaning that there will be an influx of funds
to the city that will offset a considerable portion of this loss, but this level of cover may not be
available in the future for other centres. New Zealand could afford an uninsured loss to a
smaller centre, but would struggle to carry a similar level of loss to the Christchurch
earthquake.
7
How are buildings assessed?
Further uncertainty arises in consideration of the methods of assessment, some of which are
intended simply to determine if a building may be an EPB. For example the ‘desktop’
EPBs Feb 27 2013_R4.doc
10
Revision 4, Feb 24, 2013
method of assessment that may be used over the next five years under the MBIE “Building
Seismic Performance” Consultation Document will be a qualitative assessment based on
council records, likely to be primarily age related.
The next level of assessment is the Initial Evaluation Procedure (IEP), developed by the
NZSEE. This is also primarily a qualitative procedure based on known building attributes,
but has been specifically developed with more attention to the form of the building. Again,
this is intended primarily as a sifting method, to better verify if a building may be earthquake
prone – it should not be used to confirm that a building IS an EPB. However, it is debateable
as to whether it will successfully trap all buildings that actually ARE EPBs, as there are some
types of building that may be able to escape detection. This will be a more critical
consideration if the criteria for assessment are changed from a simple capacity test (against
33%NBS) to another measure – refer below.
More detailed assessments are generally quantitative. The NZSEE has published a guideline
known as ‘the Red Book’12 which offers a range of methods for evaluating buildings of
different materials and forms. The methods offered may be applicable for a range of analysis
approaches, but are typically more geared to linear analysis (such as is used for design of new
structures) or simplified non linear static analysis.
Other methods of analysis are available for consideration, that were produced in other parts of
the world. Some are already in use in New Zealand, typically when assessing buildings that
are outside the scope of the Red Book, or when more sophisticated levels of analysis are
being implemented. Among the most commonly used are the companion documents ASCE
3113 & ASCE4114, from the American Society of Civil Engineers in the USA. The latter is a
performance based document that is particularly useful when using more sophisticated forms
of assessment such as non linear time history analysis. This is typically outside the capability
of most structural engineering consultants and requires considerable experience and
judgement, even when familiar with its use.
Updating of the Red Book is a topic currently under discussion by the NZSEE (with potential
involvement of other technical societies). This is a considerable task if it is to be done
comprehensively, as there has been significant research on the performance of existing
buildings since this document was conceived. Similarly, although ASCE41 post dates the
Red Book, it also is being updated, with a revised document understood to have been slated
for later in 2013.
8
What should be done with an EPB?
Broadly speaking there are two options – fix it, or demolish. The timeframe over which an
EPB may continue to be occupied until this is done is dependent on the local Territorial
Authority’s EPB policy, but the recent consultation document from MBIE implies that this
may soon be regulated by central government.
The ideal time to upgrade an EPB is obviously before there is an earthquake –now too late for
most of Canterbury. But there are several important matters to consider
12
New Zealand Society for Earthquake Engineering Assessment and Improvement of the Structural Performance
of Buildings in Earthquakes, June 2006
13
American Society of Civil Engineers, Seismic Evaluation of Existing Buildings (31 03)
14
American Society of Civil Engineers, Seismic Rehabilitation of Existing Buildings (41 06)
EPBs Feb 27 2013_R4.doc
11
Revision 4, Feb 24, 2013
The first is the upgrade load level. The Christchurch City Council EPB policy requires
owners to target 67%NBS, but implies that exceeding 33%NBS is the minimum that can be
accepted, provided that the owner can demonstrate why 67%NBS is not achievable. It is
silent on what criteria may be used to determine this, but typically this may include cost,
practicality and impact on heritage.
Other councils have differing policies, some requiring upgrade to 67%, others insisting on
33% only. There is also a mix between passive policies (as used to be the case in
Christchurch) and active policies. However, the recommendations of the Royal
Commission15 and the recommendations of the Discussion Document clearly point to an
active approach country wide.
From a ‘safety’ perspective, the probability of damage in the future is not directly related to
target load level. For example, consider the following (from the Red Book):
%NBS
>100
80 100
67 80
33 67
20 33
<20
Relative risk
<1 time
1 2 times
2 5 times
5 10 times
10 25 times
>25 times
These numbers are applicable in cases where there is a ‘normal’ seismic hazard level. In
Canterbury, because there is an elevated level of seismic hazard related to the increased
activity near Christchurch, the risk of exceeding the lower levels of shaking is higher again.
In Christchurch, the relative risk at 33% NBS is around 20 times the risk at 100%NBS.
The next related factor is cost. The difficulty for most people is that this is all ‘dead’ cost,
that is, after spending this money, an owner is left with the same building and the same use,
so there may be no way of generating any additional rent. A change of use may allow a
building owner to generate increased rental return, but this will also trigger S11516 of the
Building Act, requiring upgrade to at least 67%NBS.
The relationship between cost and strengthening load level is not linear. That is, if the cost of
achieving a certain level of improvement is ‘x’, achieving double the improvement is not two
times ‘x’.
Reports prepared for the Christchurch City Council by Holmes Consulting Group (HCG)
prior to the earthquakes put the cost of strengthening heritage buildings in Christchurch to
33%NBS, at $400/m2; and to 67%NBS, at $1,000/m2(noting that these costs are subject to
extreme variability. These figures were based on completed projects and/or quantity
surveyors’ estimates for proposed work over a reasonable range of buildings.
More recently, HCG completed a further review of costs for the Royal Commission, although
that work is not published. But it concluded that there were no grounds for adjusting those
figures, noting that (at that time) four buildings undergoing strengthening and repair had been
15
Canterbury Earthquakes Royal Commission, Final Report, Vol 4,
This requires buildings to comply, as nearly as is practicably possible as if it were a new building, with the
Building Code for provisions relating to structure, fire, sanitary facilities and disabled access.
16
EPBs Feb 27 2013_R4.doc
12
Revision 4, Feb 24, 2013
costed at levels remarkably close to the figures above. It also noted that the cost of repair, as
opposed to strengthening, was considered to be of a similar order to the strengthening cost.
The last will of course vary considerably with the level of damage, but it could be reasoned
that the lower the capacity of the building, the more damage will be suffered and hence the
more repair will be required.
Increasing cost
The general relationship between cost and strengthening level can be represented by a generic
chart, as below:
3
1
2
Figure 3: Relationship between costs and strengthening level
Increasing target capacity
The reason for the shape of the curve is that typically, the addition of capacity of a building
beyond its optimum level requires the addition of significant new elements, to compensate for
inherent weaknesses. For example, in the chart, point 1 may represent the cost to achieve
33%NBS capacity, with no major new elements added. But for relatively modest increase in
expenditure, additional capacity may be added to get to point 2. The addition of a new
element (such as a wall or frame) may be required to achieve significant additional capacity in
order to get to point 3. Once a significant new element is introduced, there are often some
relatively simple secondary items such as further ties and connections that can then maximise
the benefit of that element, before the next milestone load level is reached requiring the
insertion of further primary elements.
For this reason, a dogmatic approach to seismically upgrading buildings should be avoided.
A better approach is to optimise the proposed work to achieve the best value between the
minimum acceptable level and the maximum practical level. Further criteria to consider in
this are the possibility of change of use (to generate higher rental rates) and the potential for
configuration changes to allow greater lettable area.
Although the general shape of the curve may be similar for all buildings, each individual
building will generate different cost vs. capacity relationships, so the line does not follow a
fixed formula.
EPBs Feb 27 2013_R4.doc
13
Revision 4, Feb 24, 2013
9
Risk Analysis
A comprehensive risk study was completed for MBIE prior to the completion of the
consultation document17. This study concluded (in part) that there should be a risk based
framework developed for Earthquake Policy setting as part of a longer term process, but that
in the meantime, priority should be assigned to the identification of EPBs, reduction of the
time taken to implement upgrading works, and extending the building related measures to
provide equivalent levels of risk control over other earthquake life safety hazard.
In classical risk management terminology, risk is the product of consequence and probability.
Under the strict threshold based consideration of risk implicit in the current EPB (moderate
earthquake) definition, this becomes in effect a conditional probability, i.e. given an
earthquake of this intensity, what are the probability and consequence of failure? This may
have been appropriate when we were considering a limited population of buildings (URMs)
that are particularly vulnerable at even modest levels of ground shaking. However without
any limitation to the seismic load input, these buildings do not necessarily present the greatest
risk.
Experience in Christchurch shows that some of the buildings that are not earthquake prone
may be even more dangerous than the recognised EPBs, when subjected to higher levels of
earthquake shaking. This is either by virtue of poor behaviour over a higher threshold load
level or in the exposure of greater numbers of occupants (in the case of larger buildings). A
non conditional risk analysis could therefore conclude that without limitation to the seismic
load level, we would achieve a greater level of risk reduction by taking a more focused view
of which buildings we assess and mitigate.
Consequently, the stated objective (of the Act) of preserving life safety may be better met by
reconsidering the risk equation according to the nature of the buildings being assessed. .
10 An Alternative Approach
The discussion above is of limited value if there is not an alternative offered that may
generate a better outcome. So this section puts forward a conceptual approach that may offer
greater security to building users, given our observations from the Christchurch earthquakes
and subsequent engineering reviews around the country. It is put forward for discussion,
acknowledging that the proposal will need considerable further development and consultation.
10.1 Context
In order to consider an alternative approach, the context must be reviewed.
It is estimated that there may be as many as 25,000 earthquake prone buildings in NZ18,
using the current approach. This number is probably heavily skewed towards the larger
centres, but it must also be considered that many of the smaller cities and towns that
serve rural communities were established before the advent of current seismic design
standards and have not had the same degree of urban renewal as our larger cities, over
the last thirty years.
17
Tony Taig, TTAC Limited and GNS Science, A Risk Framework for Earthquake Prone Building Policy,
November 2012.
18
MBIE, Building Seismic Performance Consultation Document, December 2012
EPBs Feb 27 2013_R4.doc
14
Revision 4, Feb 24, 2013
An approach is required that will serve both needs, with the following characteristics:
•
The larger cities still have many older URM buildings, typically of 2 3 storeys,
but some higher. However, most of these centres have been more actively
following an upgrading process, so there has been some strengthening of the
worst of these buildings. Smaller concentration of these buildings may typically
exist in the fringe CBD and outer suburban centres. Larger cities will also have
the greatest proportion of non ductile concrete buildings and other ‘at risk’
buildings constructed in the mid to late 1900s, typically closer to the CBD and
other satellite urban centres. These buildings may have greater occupancy loads
and although less likely to be earthquake prone as currently defined, represent a
higher risk of heavy loss of life in larger events.
Finding an economic use for URMs is difficult, although there has been (pre
Christchurch earthquakes at least) a slight heritage resurgence, increasing
demand in some locations. Demand for category B and C (or uncategorised)
office space may be fickle according to local economies and location, but for
reasonably serviceable buildings, there is a market. Apartment conversions
were prevalent for a time in the 90s, with a subsequent legacy of possible under
capacity buildings with difficult ownership structures.
•
Smaller cities, towns and rural service centres were typically developed in the
late 1800s and early 1900s and have a familiar pattern of 2 3 storey brick
buildings lined up either side of the main street and some side streets. As stated
above, these centres have typically not had the same extent of urban renewal as
there has been internal migration to the larger centres and as the relative wealth
of the rural communities has declined. Many of these buildings are in use only
at ground floor level (retail) with little demand for upper levels.
A significant concern in such communities must be that in the event of
significant numbers of EPBs being identified, the cost of mitigation coupled
with the limited demand for space will result either in demolition or dereliction
for a disproportionately large number of buildings that otherwise have at least
some useful function. This could have a devastating impact on these centres.
10.2 Objectives
Broadly speaking, the objectives of any approach should be to reduce risk of death from
building failure, but this needs to be considered with respect to its impact, primarily
cost, but also in consideration of the overall effect on communities that may result for
example from overzealous removal of hazardous buildings and loss of heritage.
A dual approach may be needed, that allows for the impact of wider loss to be
considered. An underlying life safety objective could be applied to the country as a
whole, with other designated centres to consider the impact of wider loss, such as:
•
•
•
Impact on key transportation corridors
Maintenance of key central or regional facilities servicing much wider
communities
Maintenance of the local economy.
EPBs Feb 27 2013_R4.doc
15
Revision 4, Feb 24, 2013
10.3 Methodology
A dual approach is recommended, with the determination of which to follow being
dependent on the building material and construction type. A significant factor in this is
the identification of key vulnerabilities, and whether the vulnerabilities relate to global
collapse or key element failure.
The recommended approach is to prepare a matrix which classifies buildings by type
and identifies key potential vulnerabilities. This allows us then to rank the most critical
vulnerabilities and building types and to generate potential retrofit measures that can be
applied, in parallel. Life safety issues may be classified as either collapse issues or
element failure issues as this will impact on decisions as to whether complete upgrade
and replacement is required, or whether remediation of the vulnerability only may be
effective.
1. For buildings where there is a global collapse issue, the approach used should be
generally consistent with the current approach, i.e. assessment against the current
moderate earthquake level, but modified to include consideration of performance at
higher levels of load also. Where a building fails to meet the assessment criteria, it
must be upgraded or demolished.
2. For buildings with key element failure issues, the building should be assessed
qualitatively for key vulnerabilities which are applicable to that building type.
Where a key vulnerability exists, it must be upgraded, unless it can be demonstrated
by analysis that the vulnerability will not be triggered in events below an agreed
ceiling load level. The ceiling load level may vary according to the nature of the
vulnerability and the building type.
In this case, there is no need to complete a full quantitative analysis of the building
as a whole, provided that it otherwise complies with requirements to be determined,
unless the owner and engineer consider that a more complete analysis may allow a
reduced level of mitigation.
For simplicity, ceiling levels and/or mitigation levels may be pre determined
according to building type, unless by specific analysis.
This may achieve several things. Firstly, if it is deemed that a building has a simple
element vulnerability, it is possible to proceed quickly with mitigating the vulnerability
rather than complete lengthy assessments of global capacity, saving time and cost of
assessment, and allowing funds to be efficiently directed to the mitigation. An example
of such vulnerabilities is stairs with short seatings.
Secondly, it allows the most dangerous life safety issues to be targeted so that we can
achieve the maximum overall benefit in the shortest time.
10.4 Further Considerations
The following require some additional consideration:
EPBs Feb 27 2013_R4.doc
16
Revision 4, Feb 24, 2013
1. The behaviour of unreinforced masonry buildings should be considered and
addressed with specific reference to the small community issues raised in 10.1
above. Although these buildings are commonly considered collapse hazards, a
review of their performance in the Christchurch earthquakes shows that there
are other factors that may be considered.
In most 2 or 3 storey URMs (as typified in smaller centres or outer CBD and
suburban centres), there is a full or partial independent timber or steel gravity
support system. In the event of earthquake, this structure has generally been
effective in preventing complete collapse. However, toppling of parapets and
appendages, and out of plane failure of upper storey walls were responsible
for the majority of the deaths associated with such buildings, generally either
in the adjacent building or on the street.
Conversely, in URM buildings without such internal structure (such as
churches or halls), full or partial collapse was not preventable, and this type of
structure accounted for the only deaths inside a masonry building.
It is suggested that a further modification of the approach could be considered
to address this. For buildings with independent gravity structure in smaller
centres (possibly included outer CBD and suburban centres of larger cities)
that satisfy certain criteria, a qualitative assessment may be used that then
focuses more on mitigation of the hazards. This is a variation of the US
‘Bolts plus’ approach.
2. There may be concern over the time that it would take to produce a fully
detailed procedure as discussed above. However, we note that the appendices
of the DEE Guidelines already have considerable development of the building
types and vulnerabilities, as a starting point. Also that this approach has much
in common with ASCE 31 03, which may form the basis of the methodology.
This would reduce the time for implementation considerably.
3. Coupled with the timing question is flexibility for future application. The
approach outlined above potentially allows for the addition of further key
vulnerabilities in the future, provided that the identification of vulnerabilities is
contained in the Regulations. The Act should be limited to describing the
desired outcome, life safety.
10.5 Examples
The following are some examples of how the approach may work for some of the
common building types or vulnerabilities.
10.5.1 Lightweight Timber0framed buildings
These buildings are known to be a low risk category of structure, unless they have
heavy masonry elements such as brick fire walls or chimneys, or a heavy tile roof.
Even then, the major hazard is generally restricted to the immediate adjacent areas
(for heavy single elements such as chimneys) or the exterior (for the tiles).
Typically, these buildings are not a significant collapse hazard, even under severe
seismic loading. They are not vulnerable to liquefaction or settlement in the way
EPBs Feb 27 2013_R4.doc
17
Revision 4, Feb 24, 2013
that heavier structures are. Older timber framed buildings were generally not
specifically designed, and frequently lack well defined load paths. However, they
generally have considerable redundancy in the structure, and their lightness means
that it is often wind loading that governs their design or assessment.
The most common deficiency of these buildings is the lack of connectivity of the
main structure to the foundations, which may often be poorly embedded and/or
brittle in nature. However, unless they are on a severely sloping site, this does not
constitute a collapse hazard.
However, there have been numerous incidences of timber framed buildings being
assessed as earthquake prone, in some cases with capacity less than 10%NBS,
while remaining undamaged.
This form of building is widespread among residential buildings (in which case
they are exempt from consideration) and amongst schools, where they are a
significant proportion of the 17,000 or so school buildings in New Zealand.
The recommendation in this case is that timber framed buildings may be subject
to a simplified vulnerability assessment based on a limited number of key
vulnerabilities. Timber buildings that conform to certain size, regularity and risk
factors should be exempt from further review.
10.5.2 Limited Scale Unreinforced Masonry Buildings
Unreinforced masonry buildings are an acknowledged seismic risk, but few were
observed to have fully collapsed in the Christchurch earthquakes. Instead, the
major hazard was from falling masonry into either the street or into neighbouring
buildings.
A purely force based approach that seeks to simply lift the building to above the
threshold (33%NBS) does not change the nature of the vulnerabilities, even
though it may increase these buildings’ capacity slightly.
Instead it is recommended that for buildings within certain constraints (for
example, two or three storeys, independent gravity support structure, flexible
timber diaphragm, and overall plan dimension limits), that a more prescriptive
retrofit process be implemented. Key features of this (depending on the observed
vulnerabilities) might include:
1. Reduction of parapets, or replacement in lightweight material.
2. Tying of all diaphragms with tie spacing of no greater than twice the
wall thickness (in order to ensure failure in the wall, rather than the
anchor, at higher levels of load)
3. Construction of a timber framed wall against the exterior wall in the
upper storey (to provide secondary support to the roof) and possibly
below.
4. Removal of any specific hazard such as chimneys.
5. Install veranda posts and remove veranda tie backs from the face of
URM buildings. (they pull the front walls off the buildings particularly
if the parapet falls on the veranda)
EPBs Feb 27 2013_R4.doc
18
Revision 4, Feb 24, 2013
Upgrading of the shopfronts for in plane resistance may only be required on
corner sites, or not at all, depending on scale and level of redundancy of
remainder of structure.
The intention of this is to provide a means of achieving a limited upgrade
addressing the most dangerous features, that may be applicable for small town
New Zealand (noting that upgrading to a higher level is always permissible at the
owners discretion, or if required by change of use).
10.5.3 Non0ductile columns
This was a major concern identified from the CTV building investigation. This is
an obvious vulnerability that would be identified for concrete structures, whether
the primary system is moment frames or shear walls.
As this is a critical vulnerability, the recommendation is simply to retrofit the
columns to provide confinement in the vulnerable regions. As this is a drift
related effect, it is conceivable that a ceiling limit may be set based on drift – for
example to require all non ductile columns to be upgraded unless the drift at
100%ULS (for an equivalent new building) is less than 0.25%, or the columns are
required to undergo plastic rotation of less than 0.003 radians (subject to axial
load ratio etc, tbc).
This leaves it over to the owner and engineer to determine if it is worthwhile to
consider a complex analysis, or simply to retrofit the columns, the preferred
outcome.
10.5.4 Stairs with sliding supports
Designers of concrete buildings designed prior to the 1995 code are likely to have
underestimated the drift at ULS levels. In these cases, if the stairs are on sliding
supports of limited bearing length, it is possible that these stairs may fall of their
support in some cases. In this case it is recommended that if a stair does not have
enough residual seating at the drift corresponding to 100%ULS (to be formulated
against height and building form, for example, 2.5% of the storey height +30mm
for a conventional moment frame structure, down to 0.5% + 30mm for a squat
shear wall structure, then the stair must have the supports retrofitted, to sustain a
drift 50% higher than this; unless the building can be shown by analysis to be able
to provide sufficient residual support to the stairs at 100%NBS. Further
consideration may be given to increasing the residual seating in cases of scissor
stairs, where the failure of one will block the other.
Once again, this leaves it to the engineers and owners to determine if it is
worthwhile to attempt to justify what is there, or having identified the
vulnerability, simply to fix it.
EPBs Feb 27 2013_R4.doc
19
Revision 4, Feb 24, 2013
11 Conclusion
This discussion paper has highlighted some of the most significant issues with the current
approach to earthquake prone buildings that result from the practice of assessing capacity
against the new building design load or displacement. An alternative approach has been
developed that is reasonably straightforward to implement and which would be more effective
in mitigating the seismic risk associate with our most dangerous buildings. This is currently
not
John Hare
24 February, 2013
EPBs Feb 27 2013_R4.doc
20
Revision 4, Feb 24, 2013