CTV+Building+Collapse+–+Christchurch+NZ+Earthquake

=CTV Building Failure=

Christchurch, New Zealand February 22, 2011
Josh Jaskowiak | BAE/MAE | Pennsylvania State University 2016

toc =**Introduction**= The Canterbury Television Building serves as a case study into failures of seismic design and the demands of an ethical and competent practice of structural engineering. On September 4, 2010 the 7.1 Magnitude Darfield Earthquake and on December 26, 2010 a magnitude 4.9 aftershock struck the Canterbury, New Zealand region causing significant damage, but resulting in no loss of life of major structural failures. Unfortunately, on February 22, 2011, the Magnitude 6.3 aftershock caused the failure of four major buildings in Christchurch, the most significant being the collapse of the Canterbury Television Building. The resulting loss of 115 lives and total destruction of the CTV Building resulted in an investigation led by the Department of Building and Housing that found that the original design and lack of oversight led to the construction of the unsafe structure.

=Key Words=

CTV Building, Canterbury Television Building, Christchurch, Gerald Shirtcliff, Will Fisher, Alan Reay, David Harding

=Brief Summary of Lead-Up Events=

In 1986, Alan Reay was principal of Alan M. Reay Consulting Engineering in Christchurch, New Zealand. His small firm received the CTV Building project, to be designed by David Harding. Harding’s previous design experience had been with civil engineering project and low-rise construction. Upon completion of the design, Reay did not complete a full review of the work. In total, David Harding spent 305 hours on the project while Dr. Reay spent 3.5 hours reviewing the work.

When submitted to the deputy building engineer, Graeme Tapper, for approval, additional information was requested. Eventually, Reay went to the city engineer and Tapper’s boss Bryan Bluck for approval without the additional requested review of design documents and calculations (Macfie 2014). Tapper’s reservations to the original submitted documents were related to insufficient seismically-detailed connections between lateral force resisting elements and the building diaphragm. As a check, Tapper wrote to Reay's firm asking for the complete set of structural engineering calculations for his review. The investigative committee would were not able to find the complete set of calculations, foundation report, and revised drawings requested by Tapper. These findings were reported in the final report (IPENZ 2014).

Harding used the structural engineering software ETABS to produce a model of the building’s response to loading. Seismic loads were to be resisted by concrete shear walls at the north edge of the building and eccentric with respect to the building’s center of mass. This was Harding’s first use of ETABS software (IPENZ 2014). == The CTV building would initially be built by Williams Construction in 1986, and Gerald Shirtcliff was promoted to construction manager of the project. During the completion of the building, two former employees broke off of Williams Construction to found Union Construction. The CTV project re-awarded to Union Construction, and Gerald Shirtcliff was retained as the as the construction manager of the project (Beynen, 2012).

In 1990, major design defects were first brought to light. Holmes Consulting conducted a review of the existing structural system for a potential tenant looking to evaluate the safety of the building. They found that there was inadequate strength in the connection between the floor slab and shear walls, especially the walls at the north shear wall core. When alerted to the issue, Reay took 21 months to implement a fix. During the retrofit process, however, permit applications were never filled and the local building council was not alerted to the structural renovation that was being done. The lack of disclosure to the council prevented their involvement in the consultation and review process (Macfie, 2014).

=Previous Seismic Events=

On September 4, 2010, the Magnitude 7.1 Darfield Earthquake hit the Christchurch region of New Zealand. That event and the following aftershocks caused damage, mainly to un reinforced masonry structures and non-structural components, but there was no loss of life or major structural failures. The CTV building remained standing, with minor damage observed. Damage observed included:


 * “Minor cracking to the south wall and adjacent floors
 * Minor structural damage including fine shear cracks in the North walls
 * Fine cracking of several perimeter columns in the upper floors
 * Several cracked or broken windows
 * Floor to ceiling cracks at the junction of the lift doors wall and return walls on Level 6 “Structural Performance of Christchurch CBD Buildings in the 22 February 2011 Aftershock” 2012)

=February 22 Aftershocks=

The February 22, 2011 Magnitude 6.3 aftershock had a very different outcome from the Darfield Earthquake. A total of 182 lives were lost and large-scale structural failures were observed. The February 22 aftershock had such a large impact due to several characteristics of the quake. The epicenter was located 6 miles Southeast of downtown Christchurch at a depth of approximately 3 miles. The proximity to downtown Christchurch and shallow depth resulted in recorded peak vertical accelerations of 1.8g to 2.2g, which are some of the highest measured in a highly-populated environment. Although the short peak ground accelerations of 0.5g to 0.8 helped reduce some effects, the significant resulting liquefaction further increased the severity of the event.

16 eyewitness reports were collected following the CTV collapse. While specific details from those interviewed were varied, there was a consensus among the reports that the building in a duration set in terms of seconds, not minutes. Rapid pancaking of the floors and the lack of structural elements left standing were clues to the investigative teams that brittle failures occurred, especially at the slab to wall connections. This hypothesized behavior was in direct contrast to the ductile failure modes preferred in seismically-controlled design (“Technical Investigation into the Structural performance of Buildings in Christchurch – Final Report, 2012).

In total, 115 people were killed in the collapse of the CTV building including 71 students and 9 staff at an international school located within the building. While 15 survivors were rescued by first responders, many more were lost in the collapse. The stories of those who did survive helped inform the investigative team as to the cause and sequence of collapse. Nilgun Kulpe and Angela Osborne were on the top floor of the building when at 12:51 in the afternoon they felt the beginning of the earthquake. In their attempt to escape, they ran to the door of their office when the building gave way. They would recount the sensation of riding the building down with the wreckage and later being rescued by first responders (Macafie, 2014). Their location on the 6th floor provided them a greater chance of surviving the collapse. Only one fatality was recorded of those individuals who were on the 6th floor. This result contrasts with the story of Maryanne Jackson, a receptionist with CTV who was the only person on the first floor to survive (Wright, 2012).

=Investigation= The resulting investigation into the cause of collapse of the CTV building was led by the New Zealand Department of Building and Housing. Given eyewitness reports and the damage observed, the main initial hypothesis was that the lack of cover concrete and adequate confinement reinforcing steel for the perimeter columns resulted in the failure of one or more columns and therefore triggering the full collapse. Furthermore, the eccentricity in the Lateral Force Resisting Systems, lack of proper connection between the diaphragm and shear wall, and incompatible deflection between spandrel beams and columns were identified as potential contributing factors.

A three-pronged approach was taken by the investigation team. The original design documents and later structural modification documents were reviewed for compliance with the original code. Samples from the debris were taken and tested for structural properties. Computer structural models were created to simulate potential modes of failure to determine a likely scenario (“Structural Performance of Christchurch CBD Buildings in the 22 February 2011 Aftershock” 2012).

Given the liquefaction observed at various sites ar ound Christchurch, failures due to foundation collapse or soil conditions were an initial hypothesis. The investigative team found that the original 1986 was conducted in a manner consistent with the means and methods of the time. Furthermore, the engineer noted that there would be cause to investigate the site further based on the limited stiffness of the soils. While under today’s standard of care, this would have certainly signaled the need for further study, at the time, the engineer proceeded in a professional manner. There was some evidence that liquefaction could have occurred, but was not critical in the collapse. The foundations built were typical of the time and appropriate. Foundations and soils were not a contributing factor to collapse (Sinclair, 2011).

Structural debris found at the site were examined and tested as evidence in the investigation. There was no evidence that the foundations failed or were inadequate. The concrete used in the shear walls and slabs met the specified strength. The concrete specimens tested had a mean strength of 20 MPa, less than the specified 35 MPA, 30 MPa, and 25 MPa for columns at level 1, level 2, and levels 3-6, respectively (Stannard, 2012). Figure 5.3 from the expert panel final report shows the measured strength of concrete versus that specified.

Contributing Factors of Failure

In their final report, the investigative team stated that there were three main fields where inconsistencies occurred between the stated design and that necessary to resist the seismic loads.

__Horizontal Ground Motion__
In the aftermath of the February 22 aftershocks, many asked why this event caused so much damage while the September 4th event was less severe. The following figure from 5.9 of the final report shows the response spectra of the design earthquakes and those that hit in September and February. At a one second period corresponding with the natural period of the CTV building, the building response was approximately 0.9g on February 22nd, 0.5g on September 4th, and 0.7g for elastic design as specified by the 1984 code.



In a more telling infographic, the drift demand versus design capacity for drift in the North-South and East-West directions is shown in figures 5.10 and 5.11 for the September 4th and December 22nd events. During the September 4th event, the elastic strain in the columns slightly exceeded that of the 1% inter-story drift limitation if there was no gap between the spandrel precast panels and columns. The range between the solid orange line and solid red line represents the upper and lower inter-story drift limitations as the precast panels were designed and if no gap was provided. As seen in the comparison of drift and capacity in Figure 5.11, the resulting drift far exceeded the limitations. These drifts were analyzed using computer modeling of the building with the response spectra of both seismic events. The column in question, F2 at level 3, is one that was believed to have failed early in the collapse sequence.





__Vertical Accelerations__
While there were already significant design flaws capable of triggering a collapse without the added influence of the vertical accelerations, the historically high values recorded helped to expedite the collapse. The computer model analysis found that vertical forces could increase upwards of a factor of 2 when considering the effects of vertical acceleration.

__Column Design Flaws__

 * Drift limitations for columns along column lines F, 1, and 2 do not appear to have been met when computer models were run accounting for vertical accelerations, the observed material strengths of the concrete, and other variables.
 * Beam-to-columns required but were not designed for a ductile connection.
 * Less than adequate shear reinforcement was provided along the height of the column.
 * Inadequate concrete cover on the columns could not prevent spalling of the concrete and therefore left the already undersized reinforcement potentially susceptible to buckling.

__Precast Spandrel Panels__
As shown in Figure 5.5 from the Expert Panel Report, the precast spandrel panels were shown in the original design documents to have a 10mm gap between the column and edge of panel. This 10mm tolerance was less than the 19mm required to provide adequate seismic separation. When combined with the already limited drift restraint provided for by the frames, the expert report concluded that it was highly probable that interaction between the panels and columns in combination at the upper floors was a probable cause of initial stress and eventually failure.

__Eccentric Lateral Force Resisting System__
The center of rigidity of the lateral force resisting system was located outside of the floor plan of the CTV building, and therefore was highly eccentric with respect to the center of mass. As shown in Figure 5.6 from the Expert Panel Report, the center of rigidity was located alongside the northern shear wall elements that provided the majority of the lateral resistance, thereby distributing torsional shear to the rest of the building elements. In lieu of adequate lateral resistance at other locations, this meant that frames not designed for lateral load felt the initial effects of the earthquake. These findings correlated with the earlier eyewitness reports stating observed twisting of the upper floors in the moments before collapse.



__Lack of Adequate Diaphragm Connection__
Originally identified as a weak component in the original early 1990s retrofit, the connection between the floor diaphragm and the North shear wall core was repeatedly described as subject to failure and tenuous. In figure 5.7 bellow, notations in the expert panel report highlight the shear wall core in purple. In the highlighted details, there is a lack of continuity between the reinforcement in the wall to the floor system. Additional details provided in the original drawings distributed to the public within Appendix L of the Investigative Report provide additional evidence as to the lack of reinforcement (“Structural Performance of Christchurch CBD Buildings in the 22 February 2011 Aftershock” 2012).



The photograph bellow, originally Figure 2 in the Site Examination and Materials Test report shows the north core standing after the collapse and subsequent fire. Looking at the 4th and 5th floors, the rebar that was supposed to connect into the diaphragm can be seen still connected to the wall yet pulled out of the slab. Evident of pull-out failure, these rebar provide visual evidence that the original connections were inadequate of holding together the diaphragm and core (Stannard, 2012).



3. Construction Flaws
During the construction of the CTV building, a number of quality control issues and lack of oversight contributed to the poor performance of the building. Joints between concrete pours and precast elements were not roughened before concrete pours to create a better adhesive surface between the two elements. A number of rebar elements were found to have not been placed correctly throughout the building; in combination with the already inadequate rebar specifications, the required design strength could not be achieved. There was also a lack of adequate development length in the rebar arrangements along the shear wall core as found in the investigation (Sinclair, 2011).

While the original design emphasized the lateral resistance provided by the North shear wall core, there were unintended lateral elements in the building structure. Wall A on the Western edge of the building was not designed to be a lateral wall and was supposed to separate from the diaphragm under conditions where lateral load was distributed to the building. This concrete masonry wall did not rise throughout the full height of the building, thereby representing a significant lateral discontinuity. The investigation found that the connection between these walls and the slab floor provided an unintended transfer of lateral forces to the wall. This unpredicted behavior in combination with the irregularity between the lower 3 stories and upper three stories helped contribute to the torsional effects observed during the collapse.

=Contributing Factors of Failure=

With all of this information complied, the Royal Commission investigative panel developed four theories of the collapse. They were

1. The interaction between the spandrel precast panels and exterior edge columns along either column lines F or 1 at the upper stories.

2. Axial and shear failure of a low story column triggered by high vertical accelerations, inter-story drift, and low concrete strength.

3. Discontinuity between the north shear wall core and slab resulting in the loss of lateral resistance, especially at the 2nd and 3rd floors where no drag bars were installed as part of the early 1990s retrofit.

4. Failure of the drag bars at the 4th, 5th, and 6th floors resulting in discontinuity between the slab and shear wall.

In their conclusion, the expert panel determined that scenario one was most likely, but that contributing effects from the other 3 scenarios also contributed to the total collapse of the building. The deciding factor was the lack of consistent spacing between the precast panels and columns at the upper stories. At some locations, it was believed that the designed gap of 10mm varied from 0mm to 16mm in the built condition based on evidence found and reports. The initial “pinching” action between the spandrel precast panels and columns caused loss of the inadequate concrete cover, exposed the under-reinforced confining steel, and left the column subject to buckling action under the already higher-than-anticipated drift conditions. Once the loss of column F2 occurred, then the adjacent columns were overloaded with both axial and drift demand and were subject to failure at the lower stories. The lack of support at the ground floor allowed for a quick progressive pancake-style collapse to occur. On top of all of this is the fact that the upper floors, which were already subject to high inter-story drift limitations, had limited lateral resistance due to the lack of block wall along column line A and were induced to torsional shear due to eccentricity. These effects increased the drift demands and did not allow for adequate transfer of force into the North shear wall core. At the end of the collapse, only the Northern shear wall core was left standing – an indication that it did not ever see the full load it was intended to see and was able to break free from the diaphragm falling around it (“Structural Performance of Christchurch CBD Buildings in the 22 February 2011 Aftershock” 2012).



=Legal Investigation= Following the publication of the New Zealand Department of Building and Housing’s findings and final reports, the Institution of Professional Engineers (IPENZ) brought complaints against Dr. Alan Reay and David Harding for their role as designers in the collapse in the building. Dr. Reay resigned his credentials as a professional engineer with IPENZ and therefore was not subject to the disciplinary jurisdiction of the organization. As of June 17, 2015, he was cleared of misleading IPENZ and not disclosing his knowledge regarding the design of the CTV building, his knowledge of Harding’s work experience, or his lack of oversight on the project. The investigation into his role was dropped by IPENZ to the chagrin of the Department of Building and Housing (Bayer, 2015). The investigation appears to be ongoing.

Two complaints were brought forth to IPENZ by the investigators of the collapse regarding the role of David Harding. Tim Elms brought forth his case on behalf of those who lost their lives in the CTV collapse. His complaint stated that Harding was incompetent, should have reached out for more help or denied to work on the project, and did not protect the public against risk of collapse, bodily injury, and loss of life. Mike Stannard, on behalf of the Department of Building and Housing, stated that the Royal Commission’s Investigation found that Harding was negligent in his design and unethical in his work.

The IPENZ disciplinary committee found that Harding broke the IPENZ code of ethics and misrepresented his competence by signing the drawings. Harding’s registration from IPENZ limited the disciplinary committee to punishment by releasing their findings and making the compliant public (The Institution of Professional Engineers of New Zealand, 2014).

In the investigation, police work and public media investigations brought to light the duplicity of Shirtcliff’s representation of himself throughout his career. Using the name of an old colleague William Fisher, Shirtcliff used Fisher’s engineering degree to earn a Master degree in an engineering program. In 2005, Shirtcliff was arrested in Brisbane on charges of fraud and convicted in Christchurch for his role in selling a failing automotive business to a New Zealand couple without disclosing the true details of the business. When interviewed by the Royal Commission, he denied that he had heavy involvement with the construction of the CTV building despite his job as the construction manager (Beynen, 2012). Shirtcliff was cleared of criminal wrongdoing by an Australian police investigation, but he lost his Australian engineering license. An investigation by the New Zealand police is ongoing (Bayer, 2013).

=Industry Response and Lessons Learned=

The New Zealand Department of Building and Housing determined that there were many contributing issues to the widespread devastation caused by the February 22 aftershocks. First and foremost, the horiztonal and vertical accelerations were greater than anticipated. In combination with the widespread liquidefacation, this prompted a review of the current code and knowledge of New Zealand earthquake conditions. The Natural Hazards Platform with the Ministry of Science and Innovation was charged with investigating the response spectrum and geotechnical conditions around Christchurch with the goal of presenting modifications to the building code.

The Ministry of Civil Defense and Emergency Management in coordination with the New Zealand Society of Earthquake Engineers was charged with the evaluation of post-earthquake inspection procedures, educating the public and building owners about the importance of inspections, and creating a database where the plans of public buildings were available for review following a seismic event for review. These changes and increased transparency would have hopefully prevented the collapse of the CTV Building because if a licensed structural engineer could have seen the CTV drawings following the September 4th event, the connections at risk could have been identified. In response to the failed structural members within the CTV Building and other locations, the investigation highlighted the importance of educating engineers in the design for resiliency and going beyond the code minimum. In the interim between the publication of the report and the eventual findings, the Department of Building and Housing stated that engineers should go beyond the code and educate building owners on the importance of resilient design (“Technical Investigation into the Structural Performance of Buildings in Christchurch – Final Report, 2012).

IPENZ recognized that changes were necessary to the ethics guidelines. In collaboration with the Association Representing Consulting and Engineering in New Zealand (ACENZ), IPENZ stated that a comprehensive review of the office practice procedures published as guidelines to engineering officers was required. Rules and guidelines for engineering peer-review, both in-house and by a 3rd party, were required in light of the lack of oversight identified in the CTV design. Continuing education and guidance was necessary for all engineering professionals, but especially those in smaller offices where the capabilities for design review might be limited (IPENZ, 2014).

=Conclusion=

The collapse of the CTV Building on February 22, 2011 represents one of the biggest ethical and engineering failures in recent history. While the aftershocks on that date created a response spectrum not predicted by the code in place at the time, there was no greater location were a loss of life occurred on that date. Of the 185 people who passed away, 115 died in the collapse of the CTV Building. The collapse of this building in such quick and violent fashion was the result of inadequate engineering, ethical breaches in the design process, and lack of proper oversight during the occupancy and inspection of the building. While retrofits were attempted in the 1990s with the introduction of drag bars, the unfortunate reality is that there were multiple attempts to catch the mistakes that caused the collapse. The combination of inadequate reinforcement, lack of ductile connections, inadequate spandrel panel spacing, eccentric lateral resistance design, poor concrete quality, and lack of construction oversight combined to create the perfect storm of a building inadequate to protect its occupants and prevent a loss of life. These events serve as a sobering reminder to the engineering community about the necessity of ethics within the profession and the responsibility incurred by the design community.

=Annotated Bibliography=


 * [|Bayer, Kurt. (July 17, 2015). “CTV Building Engineer Cleared.” New Zealand Herald]**


 * After an investigation by a Royal Commission of Inquiry, Dr Reay’s registration for the Institution of Professional Engineers (IPENZ) was allowed to stand, thereby allowing Dr Reay to avoid investigation for his role in the collapse of the CTV building. Dr Reay was principal of Alan Reay Consultants and was responsible for giving David Harding the responsibility of designing the CTV building.


 * [|Bayer, Kurt. (May 20, 2013). “No Case against CTV Man.” New Zealand Herald]**


 * Gerald Shirtcliff was cleared of criminal wrongdoing by an Australian investigation into his alleged negligence in misrepresenting himself as an engineer under the pseudonym William Fisher. While he lost his engineering license and degree from an Australian University, he will not face charges by the Australian police. An investigation by the New Zealand police is ongoing.


 * [|Beynen, Martin Van. (September 15, 2012). ‘How a Fraudster Engineered His Past.” The Sydney Morning Herald.]**


 * The construction manager of the CTV Building, Gerald Shirtcliff, was investigated by the Royal Commission for the Christchurch Earthquake Investigation regarding his responsibility in the construction process. Shirtcliff was found to have forged his resume and engineering credentials, using the identity of a British engineer.


 * [|Macfie, Rebecca. (September 2, 2014). “Collapse of Accountability.” The New Zealand Listener]**


 * Overview of the personal stories of those who escaped the collapse, lives lost, and the chain of responsibility in the design and building of the CTV Building.


 * [|Sinclair, T J E. (July 11, 2011). “CTV Building Geotechnical Advice.” Tonkin and Taylor. New Zealand.]**


 * At the request of the Department of Building and Housing and the Hyland/StructureSmith consultants working on the CTV investigation, Tonkin and Taylor investigated the geotechnical reports from the CTV design and the conditions of the site at the time of collapse. They found that the original report and foundation design was adequate and did not contribute to collapse.

[|**Stannard, Mike. (January 16, 2012). “CTV Building Site Examination and Materials Test.” Hyland Fatigue and Earthquake Engineering and The Department of Building and Housing. Auckland, New Zealand**.]


 * Forensic analysis of the remains of the CTV building for strength capacity and comparison to the original design documents.


 * [[file:dc-determination-elms-and-stannard-v-harding.pdf|The Institution of Professional Engineers of New Zealand. (October 23, 2014). ‘Disciplinary Committee Elms/Stannard v Harding.” The Institution of Professional Engineers of New Zealand, New Zealand.]]**


 * In review of David Harding, the Engineer of Record for the CTV Building, the Disciplinary Committee determined that since Harding had resigned his professional registration, they were left with limited recourse other than to publish their findings of fact and accept the resignation. The decision, though, chronicles the history of the CTV Building design and lack of professional oversight.

[|The New Zealand Department of Building and Housing. (2012). “Appendix L Structural Drawings CTV Building 249 Madras Street Chirstchurch.” The New Zealand Department of Building and Housing, New Zealand.]


 * All of the original CTV Building Documents were collected during the investigation and published with the final report. These drawings show the gravity and lateral element designs, rebar details, and precast elements used.

[|The New Zealand Department of Building and Housing Final Expert Panel. (2012). “Structural Performance of Christchurch CBD Buildings in the 22 February 2011 Aftershock.” The New Zealand Department of Building and Housing, New Zealand.]


 * As part of the investigation into the probable causes of collapse, the investigative team analyzed the lateral force resisting system and resulting effects from eccentricity within the building frames. In addition to the aforementioned lack of ductility and poor detailing, the unintended consequence of the spandrel beams not having adequate spacing resulted in contact between the spandrel precast panels and beams, further increasing the stresses in the already weak columns.


 * [|The New Zealand Department of Building and Housing. (January 31, 2012). “Technical Investigation into the Structural Performance of Buildings in Chirstchurch – Final Report.” The New Zealand Department of Building and Housing, New Zealand.]**


 * Upon review failure causes in the Chirstchurch earthquake, a series of recommendations were made by the Department of Building and Housing. These recommendations focuses on changes to the building code, design requirements, process of granting permit approval, and the post-seismic event inspection requirements.


 * [|Wright, Michael. (June 6, 2012). "Hearing Told of Survival after CTV Plunge." The Timaru Herald]**


 * As part of the Royal Commission investigation into the CTV building collapse, 6 survivors gave testimony as to their experiences during and in the immediate aftermath of the February 22 earthquake.

=Additional Sources Used=


 * [|Bradley, Derek and Tony Stuart. Dr. Barry Davidson, Ed. (February 2012). “CTV Building Non-Linear Seismic Analysis Report.” Compusoft Engineering for StructureSmith Ltd and The Department of Building and Housing, Aukland, New Zealand.]**


 * Compusoft created a non-linear seismic analysis to determine the probable sequence of failure and causes of failure. Computer models were generated with data from material testing and the original structural drawings.