The National Construction Safety Team Advisory Committee
National Institute of Standards and Technology
Minutes of December 2-3, 2003, Meeting - Gaithersburg, Maryland
The slides from the presentations are embedded as links to PDF files within this document and, thus, are summarized in these minutes. Each presentation was followed by a discussion period. "Q" indicates a question, "A" the corresponding answer, and "C" a comment. All questions and comments, unless otherwise noted, were made by Advisory Committee members. All answers unlessotherwise noted, were by NIST personnel.
The minutes summarize the main points of each discussion; they are not intended to be a verbatim transcript of the meeting.
December 2, 2003
Mr. Paul Fitzgerald opened the National Construction Safety Team (NSCT) Advisory Committee meeting at 8 a.m. and welcomed the members of the Advisory Committee, National Institute of Standards and Technology (NIST) staff, media, and public attendees to the meeting. Mr. Fitzgerald acknowledged the contribution of Dr. Jack Snell for his leadership in initiating the two ongoing investigations and in getting the NCST Advisory Committeeup and running. Dr. Snell retired from NIST in October.
Mr. Fitzgerald reminded the members of the Advisory Committee and NIST staff that the proceedings of NCST Advisory Committee meetings will be audiotaped beginning with this meeting. He encouraged the Advisory Committee members and NIST staff to identify themselves when speaking and to use the microphones so that their remarks would be clearly recorded.
Mr. Fitzgerald also reminded the Advisory Committee to phrase recommendations as a motion for Committee action.
Mr. Fitzgerald then turned the floor over to Dr. Arden L. Bement, Jr., Director of NIST for his opening remarks.
Dr. Arden L. Bement, Jr., Director, National Institute of Standards and Technology
Dr. Arden Bement began by acknowledging Dr. Snell's contributions to the ongoing investigations and implementation of the NCST Act. He noted that Dr. Snell would receive the Department of Commerce Gold Medal for his leadership in responding to the World Trade Center (WTC) disaster. Dr. James Hill, Acting Director of NIST's Building and Fire Research Laboratory, will serve as the Designated Federal Officer for the NCST Advisory Committee.
Dr. Bement stated that NIST has now received all of the essential information it needs to conduct the WTC Investigation. He also stated that NIST had reached agreement with the City of New York to allow NIST access to 9-1-1 records and to transcripts of 500 first responder interviews. NIST has also received all of the approvals it needs to conduct first-person data collection, and the first-person data collection has commenced.
He also noted that a press briefing on The Station Investigation was held in Providence, Rhode Island, on November 25, 2003, and that media coverage of the event was extensive.
With regard to implementation of the NCST Act, Dr. Bement stated that as of November 28, NIST had published all of its NCST implementing procedures.
Finally, Dr. Bement asked the Committee to review the questions he had previously provided and requested that the Committee respond to these questions prior to the next Committee meeting.
- What grade do you give the effort on the investigation to date? Why?
- Where in your view is it most likely to fall short? Why?
- What needs to be done to assure that does not happen?· What must be done to ensure that this a world-class report?
- Do you have any advice on how we can assure that we are doing the best job possible in managing external communications including Web sites, media, and public interactions?
- Within the constraints of the role of a Federal advisory committee, how could we improve our use of the committee?
Planning for the changes needed as a result of the investigation, R&D and DTAP:
- Have we identified all the right things in the WTC R&D and DTAP plans? Are there some things that don't belong on the list?
- Are the right organizations and interests involved?
- Do you have any advice for how we can assure that other organizations are prepared to quickly consider and act upon our reports recommendations? How could we improve?
Status of the World Trade Center Investigation and Discussion
Dr. S. Shyam Sunder, Chief, Materials and Construction Research Division
PRESENTATION Dr. S. Shyam Sunder gave an overview of the WTC Investigation stating that the team was extremely busy and that solid progress had been made since the last Committee meeting. He acknowledged, however, that there was still a lot of hard work that needed to be done and that much of the "heavy lifting" would occur over the next 4 to 6 months.
Dr. Sunder reported that 15 months into the Investigation good solid progress has been made. A public update was issued on December 2, 2003, which provides additional details on the progress of the investigation (http://wtc.nist.gov). The process for the selection of external experts and contractors is completed, and all contracts have been awarded. Four individuals were hired as expert consultants. Dr. Sunder stated that this is an excellent group of contractors and experts to augment the NIST in-house capabilities.
Significant progress has been achieved on data collection since August 2003. Dr. Sunder pointed out that NIST has received all of the essential information it needs for the WTC Investigation. NIST is continuing to seek photographs and videos of the south face of WTC 7. Also, pursuant to an agreement with New York City, NIST will be provided access to 911 tapes and logs as well as transcripts of about 500 first-responder interviews no later than December 31, 2003.
NIST is reviewing available information related to the calculations and analyses that supported the original aircraft impact studies of the WTC towers. Although not normal practice in building designs, the safety of the towers and their occupants in an aircraft collision was a consideration in the original design. Dr. Sunder discussed the different views on the available information.
Dr. Sunder stated that NIST has developed a comprehensive approach to identify the most probable structural collapse sequence, from the point of aircraft impact to collapse initiation. There has been significant progress in defining the requirements for the complex series of analyses and formulating detailed technical modeling approaches. Simplified modeling approaches are also being developed to provide insights and bases to evaluate results for the detailed models. In addition, NIST is developing a rigorous technical approach to evaluate the performance of the fireproofing in the towers and the role it played in maintaining the integrity of the towers but not preventing the subsequent collapse.
NIST is continuing to analyze the WTC steel in its possession and believes the collection of 236 pieces from the WTC towers is adequate for purposes of its Investigation. Preliminary results show that the measured room temperature steel strength met or exceeded relevant ASTM specifications.
Dr. Sunder provided a brief update on the fireproofing and fire rating of the WTC floor system. Project-specific fire endurance testing was not and is not normal practice, but may be conducted when circumstances warrant it, as for example, in a new application. NIST has awarded a contract to Underwriters' Laboratories to determine the fire rating of typical WTC floor systems under both as-built and specified conditions.
The collection of photographic and video images continues to grow. Dr. Sunder stated that from analysis of the images, NIST has determined the exact times for the major events of September 11, 2001 (aircraft strike, building collapse initiation). In addition, NIST has developed detailed mappings for the fire, smoke, and the condition of windows at several specific times for each WTC tower.
Dr. Sunder reported that an important milestone was reached for the project on occupant behavior, evacuation, and emergency response. NIST received all necessary approvals and has commenced the first-person data collection efforts. First-person interviews will provide valuable information to improve public safety during extreme events in tall buildings. There is currently a critical lack of information on which to base evacuation and emergency response practices, standards, and codes in such conditions.
Part of the WTC Investigation is an analysis of building and fire codes and practices. Dr. Sunder described the procedures and practices that are of interest and stated that NIST has completed a preliminary comparison of then-current building regulatory and code requirements. In addition, NIST has completed preliminary documentation of the fuel system for emergency power in WTC 7.
NIST is investigating the active fire protection systems in the WTC towers and WTC 7, and Dr. Sunder identified the specific systems that are being analyzed. These include the sprinkler system and pre-connected hoses, fire alarm systems, and smoke management system.
Q: Referring to the list of contract awards, have some contractors produced final reports? When will the Advisory Committee receive these reports? Will the Advisory Committee have the opportunity to review contractor reports?
A: We have not yet received any reports from contractors that have been approved by NIST. Once they are deemed technically acceptable, the reports will be at a stage where they can be shared with the Committee. Some technical reports may be more appropriate for review by specific experts on the Committee, while others may be appropriate for the entire Committee. At what level would the Advisory Committee want to review the reports?
C: The Advisory Committee should consider procedures for reviewing reports.
C: (Chair) After the first of the year I will work with Dr. Hill to set up a procedure for reviewing consultant reports.
C: With regard to ASTM E 119, it is not project specific and not always used. Projects rarely use E 119.
C: Generally, buildings use assemblages that have had E 119 testing previously done. Testing is not done unless the building is using systems that have not been tested before. The point here is that the floor system was sort of never done before, and the statement here is that maybe it should have been done. But, normally it is not done.
C: (NIST) E 119 was not or is not routinely conducted. However, the builder always has the option to conduct the tests, especially when there are unique aspects involved.
Q: You stated that you have all of the essential information needed for the Investigation. Can you characterize your understanding for all three buildings?
A: A lot of information exists for WTC 1 and 2. However, our feeling is that not a lot of information is kept on older buildings. After 6 to 8 years, very little information is kept on buildings. Sufficient information exists for WTC 1 and 2 for NIST to conduct its investigation. Information for WTC 7 is less complete. Some information is missing. I feel very good about the amount of information we have for the towers and reasonably good about the amount of information for WTC 7.
Project 1 – Analysis of Building and Fire Codes and Practices
Dr. H.S. Lew, WTC Investigation, Project 1 Co-Leader
Mr. Richard Bukowski, WTC Investigation, Project 1 Co-Leader
PRESENTATION Dr. H.S. Lew and Mr. Richard Bukowski presented information on Project 1 of the WTC Investigation, which is an analysis of building and fire codes and practices.
Dr. Lew described the building codes used for WTC 1, 2, and 7 and the five building codes that have been reviewed and compared. The structural requirements reviewed included dead loads, live loads, live load reduction, lateral loads (wind and earthquake), load combination, and progressive collapse resistance. The review is ongoing, but interim findings indicate that the Port Authority adopted the 1968 New York City Building Code for the final design of WTC 1, 2, and 7. It appears that designers were allowed to use "acceptable engineering practices" with the approval of the Port Authority where code provisions were obsolete. Dr. Lew explained additional interim findings regarding the five codes reviewed.
Mr. Bukowski presented the status on the review of codes for fire protection and egress systems. He described the building codes that have been reviewed by NIST and the national fire safety standards adopted by the 1968 New York City building codes. The 1968 New York City Code reduced the fire resistance ratings from the 1938 edition, and the 2001 edition was reduced further from the 1968 edition. The 1968 New York City Code contains provisions on topics not covered in other codes. He explained the New York City Local Laws and their requirements and applicability.
The fuel system for emergency power in WTC 7 is being analyzed as part of the investigation. Mr. Bukowski described the systems, their purpose, and layout.
Q: Mr. Bukowski indicated a difference between the 1967 Chicago and 1968 New York City fire protection codes. Are there any other substantial differences between the New York City code and other codes?
A: Not known as yet, but a side-by-side code comparison is under way.
Q: Regarding revisions to the New York City code, you indicated that prior to the 1968 New York City revision, the code was based on 1938 requirements. These requirements are frequently amended. Was there an amendment that kicked in before 1968?
A: No. The 4-hour fire resistance for columns and 3-hour fire resistance provisions stood until the 1968 revision to the New York City code took effect.
Q: Did you compare NFPA 101 and the building code in effect at the time?
A: Yes, and they were consistent.
Q: A lot of time has been spent on the fuel system in WTC 7. The 1993 WTC bombing incident report indicated a failure of the emergency power system for WTC 1 and 2 due to its lower floor location. Was this an issue during the September 11th attack?
A: It's not thought to be an issue. The 1993 explosion occurred in a garage, and it took out piping for the cooling water.
Q: In the Federal Emergency Management Agency (FEMA) study, several floors were reported as mechanical equipment floors. Were live load reductions different for those floors?
A: Yes, this is common practice. No load reduction was permitted for these floors.
Q: Excellent and to the point presentation. One of the goals of the act is to provide recommendations for improvements to codes, standards, and practices. The Port Authority was not tied to using the New York City code, but chose to do so. This is not a problem in major cities, which tend to follow model codes. In other locations, this is a political decision where concerns are voiced to minimize requirements to keep down the cost of buildings so that additional business can be attracted to a community. The concern is whether recommendations will be adopted by many jurisdictions. In the earthquake community, this was addressed by the Interagency Committee on Seismic Safety in Construction (ICSSC), and Executive Order 12699 on new construction. This needs to be considered.
A: This is an important issue. The Research & Development (R&D) and Dissemination and Technical Assistance Program (DTAP) are intended to address this issue. The suggestion to consider the ICSSC model will be considered.
Q: Fifty pounds per square foot is generally used for live loads. What was used in the WTC towers?
A: Generally, higher live loads of 75 to 100 pounds per square foot were used. Live load requirements for the core area differ from that of the main floor area.
Project 2 – Baseline Structural Performance and Aircraft Impact Damage Analysis
Dr. Fahim Sadek, WTC Investigation, Project 2 Leader
PRESENTATION Dr. Sadek presented information on Project 2 of the WTC Investigation, which is an analysis of baseline structural performance and the extent of aircraft impact damage. The project has two primary objectives. The first is to develop reference structural models of the WTC towers and establish baseline performance under design loading conditions (gravity and wind). After completion, the NIST-approved models will be used as a reference for other significantly more detailed models to be developed in the investigation for aircraft impact analysis, thermal-structural response, and collapse initiation sequence analysis. These detailed models would be traceable to the reference models. He discussed the progress on the models, including the structural databases and the 3-D models of typical floor systems and towers. Final approval of the reference models is expected shortly.
Dr. Sadek reported that work is also under way on the baseline performance analysis. The analysis will start upon approval of the reference models. Progress has been made in defining wind loading on the towers.
The second objective of the project is to simulate aircraft impacts on the towers to estimate probable damage to structural, mechanical, and architectural systems—including fire protection systems. Dr. Sadek said this analysis will also determine how close the buildings were to collapse immediately after aircraft impact. He presented the status of data collection on the aircraft structure; aircraft engine model development, and component level analyses.
C: The work is well thought out. I hope that you get the results you want. The failure of an exterior column is not affected by the attachment.
A: Failure at the exterior wall is governed mainly by momentum transfer and inertia effects due to the high speed of the aircraft. We will need to run a number of simulations to vary the speed and failure criteria to determine the failure mode, for example, by changing the plastic strain to fracture. This will be part of the sensitivity analysis that will be done.
C: The model will be looking at failure criteria possibly based on plastic strain or fracture with or without triaxial effects. You'll need to consider ductility with or without triaxiality.
C: The experience over 15 years with wind tunnel testing of high-rise buildings is that results are more severe today. A test done today produces higher wind pressures than in the past and I'm not sure why. We are questioning the wind tunnel people, but keep this in mind. On another topic, I recall serious welding problems during construction of the towers with the column splices. It made Engineering News Record. Someone should take a look at that.
C: I don't recall the details. Will that affect the analysis results? Need to review the data on welding, material properties, etc., and consider if it is important.
C: My recollection is that it held up the construction.
C: (NIST) We're looking at doing component testing and what happens to the weldment and what happens to the components in general and comparing it to the simulations we are planning to do. We definitely need to consider the welding.
Q: (NIST) Are you referring to the bowtie cracks or elsewhere?
A: (Committee Member) Not sure, I only recall that there were problems with the welding.
C: (NIST) What we know is that there were cracks in the bowties during construction. The explanation we have been given is that these were due to freezing and expansion of water and ice. We saw samples of these bowties with the cracks at JFK and we have some here at NIST. Our initial hypothesis is that this problem played a very small role in the impact and fire regions of the buildings. We are interested in the welding especially in the core columns.
C: Going back to the wind tunnel testing, Battery Park City did not exist at the time of the first wind tunnel tests. It will affect turbulence.
A: Yes. We are considering that.
C: It is very difficult to achieve strain rates higher than one second to maximum load.
C: I second the concern about wind tunnel testing. It has a lot to do with the uniformity of wind tunnel loads as opposed to the variability of actual winds. The actual wind loads vary from wind tunnel loads and can have a significant effect on the loading of a structure. This is particularly true if you look at edge effects and corner effects.
Project 5 – Reconstruction of Thermal and Tenability Environment
Dr. Richard G. Gann, WTC Investigation, Project 5 Leader
Dr. William M. Pitts, WTC Investigation, Research Chemist
PRESENTATION - Pitts Dr. William Pitts presented information on a task of Project 5 of the WTC Investigation, which is an analysis of photographs and videotapes of the attack on the WTC towers and the subsequent collapse of WTC 1, 2, and 7. NIST is using photographs, videos, and other relevant information to develop detailed timelines for the spread and growth of fires at the peripheries of WTC 1, 2, and 7. More than 6,100 photographs from over 185 photographers and about 150 hours of videotape (5,726 clips) from more than 20 individuals, news organizations, and documentaries have been cataloged and entered into a visual image database. Dr. Pitts presented the times for the major events of September 11, 2001 (plane strikes and collapse times). NIST has timed the events to within an accuracy of 3 seconds. Dr. Pitts discussed the relevant visual observations for WTC towers.
There were no questions or comments at the end of Dr. Pitts presentation.
PRESENTATION - Gann Dr. Richard Gann presented information on the status of the reconstruction of the thermal and tenability environment for the WTC towers and WTC 7. The project objective is to reconstruct, with assessed uncertainty limits, the time-evolving temperature, thermal radiation, and smoke fields in WTC 1, 2, and 7 for use in evaluating the behavior and fate of occupants and responders and the structural performance of the buildings. Dr. Gann pointed out that due to a lack of physical evidence for this effort, NIST is relying on computer simulations, experiments, photographic evidence, and eyewitness accounts. This type of reconstruction has never been done before, and NIST is redefining the state-of-the-art in fire and thermo-structural modeling.
NIST has collected and reviewed information on building combustibles, floor geometry, partitions, and spray-on fireproofing for input to experiments and modeling. NIST has conducted shake tests on ceiling systems similar to those used in the World Trade Center towers to determine what may have happened to these systems after the buildings were impacted by the aircraft. In addition, NIST has information on the aircraft contents, including the mass and location of the jet fuel. Interim findings are that the mass of aircraft solid combustibles was significant relative to the mass of the building combustibles in the impact zone. Interior walls blocked the view through many windows so it cannot be determined whether there was a fire inside the building.
Dr. Gann described the workstation fire experiments conducted to validate the Fire Dynamics Simulator (FDS), a 3-D model previously validated for tunnel fire temperatures and sprinkler activation. The results from the experiments showed excellent agreement with FDS. Dr. Gann stated as a provisional finding that FDS can be used with confidence to recreate a given WTC fire event. The series of tests are nearing completion.
Q: Explain the framing system for the ceiling tile you showed in your presentation?
A: The grid system used for testing is almost identical to that of the WTC. The rib failed in the first photo as the result of a single pulse. The second test profile lasted for 10 to 15 seconds with more severe results.
Q: How was a ceiling tile system unique?
A: The tiles in core areas were built to Port Authority specification, but similar to common tiles. Ceiling tiles in tenant spaces were built to an odd sized square. The grid systems were not unique.
Q: What are the heights of the partition and the ceiling in the test you showed—relative to the building?
A: The ceiling height was generally about 8.5 feet in the building (drop ceiling). The partition is 4 to 5 feet tall and consists of a core material with a fabric covering.
Q: Have you correlated flare-ups in the tests with the experimental output?
A: We're not sure that it is a flare-up tied to a single workstation—probably more than that.
Q: The workstation in the test appears tidy, no clutter. To what extent do you have data on tenant practices, inspections, etc?
A: We have talked to Port Authority building managers and tenants. Many had a clean desk policy. We put paper in cabinets and on desktops in the test to assess the effect of additional fuel. We do have accounts that many workstations were kept neat.
Q: In multiple workstation burns—given the ventilation that you have, what heat release would you expect to start to see an impact for those windows?
A: Fourteen megawatts is the highest intensity peak. We expect to see differences between test data and models.
Q: There is overshoot, or rather an undershoot, in the fire test data versus the model prediction at about 500 seconds. Is that real?
A: The area under the curve in this region is negligible compared to the area under the total curve. Fuel is a mixture of chair, computer, carpet, wood, and paper in different locations. I worry less about the spike.
C: I agree there is very good correlation between the model results and the test burns.
Q: There is good agreement in the data. Qualitatively, the model shows a significant drop when the fuel burns off while the experiment doesn't show much drop off. With more fuel, the drop at the end goes away. Is there a qualitative difference in the model?
A: I don't think so. The model has reached zero based on the fuel loaded into the model at that point. We have the actual weight loss from the experiment. If we put the actual weight loss into the model, it picks right up.
C: The weight loss is an output not an input. The weight loss affects the heat release rate.
C: (NIST) We can talk about that separately, but I don't think it is a problem.
C: Agree, it is a minor point.
Q: Regarding mechanical properties of spray-on insulation. Some results are already reported. The results seem to suggest significant insulation was removed by the impact of the aircraft. Is there any way for you to make an evaluation now without waiting for detailed tests?
A: No. We will get criteria for adhesion. We will also get a sense of forces from the impact modeling and at that point we will be able to combine the two and hopefully get a sense of what happened to the insulation.
Q: Will you take into account outside work?
A: We have information on other tests that have been done. Obviously, if others have valid data it would be foolish not to use it.
Q: (NIST) What calculations are you referring to?
A: (Committee Member) For example, some of our structural engineers at the University of California in San Diego have done tests.
C: (NIST) We would love to see the results.
C: (Committee Member) I will try to get those for you.
Q: Regarding the contents of the aircraft. Is there information about the amount of their contents remaining in the building or did most of it leave the building on impact?
A: We have to look at mass and distribution of combustibles and relate it to the photographic data. This is an issue that will be worked. We will need to try different variations to find out.
Q: Going back to the question about the first peak in the chart, I believe it is where the 2-megawatt burner is extinguished?
A: The burner is turned off in 600 seconds. Both pick up the burner being turned off.
Q: I am interested in the integration of projects. You indicated almost a seamless meshing. How did you do it and is anything preventing greater integration?
A: No, people are willing to do whatever is needed to get the job done.
C: (NIST) It's not all ideal. We find problems and then fix them. They require different levels of staff and management attention at different points in time, but everyone has an interest in doing the right thing.
Q: Did you review NAFEC (Federal Aviation Administration Tech Center) information on the aircraft interior and the role that the contents may have had on the fire development?
A: Yes. We have all their input. Some of their insights have been very helpful.
Fire-Induced Thermal States and Structural Failure Analysis
Dr. Emil Simiu, Competence Building Project
PRESENTATION Dr. Emil Simiu presented information on simplified modeling efforts that are ongoing with respect to fire-induced thermal states and structural failure analysis. Dr. Simiu explained that no methodology is currently available for evaluating the capacity of redundant structural systems with fire-induced thermal effects (?). His group is using state-of-the art methods to achieve the most effective and realistic possible investigation of fire-thermal-structural effects. Tools for the analyses include the Fire Dynamics Simulator for fire representations, algorithms developed by NIST for calculating temperatures in structural members due to fire-induced gas temperatures in the interior of the building, and nonlinear finite element methods software (ANSYS) for structural analysis. Dr. Simiu discussed the possible collapse scenarios and the status of the 2-D and simplified 3-D modeling and analysis.
Q: I have a question regarding two-dimensional structural model analysis. The sagging of joists took time; the columns could have heated as well. Around the cavity, if these columns get heated, they shrink or creep. Is there any effort to assess this?
A: Absolutely, there are hot spots in the thermal/structural sense. Some columns may give and the next column may be called upon to take up the slack. This may lead to a progressive column failure scenario. The model helps to provide insight into this issue.
C: It's important to integrate the fire and structural models. I encourage carrying this work further. Breaking windows can have a large effect on fires. You can go back further and say windows will break under certain load conditions.
A: It is not easy, but we're making an effort. You need to stop the model and change boundary conditions. We are continuingto look at these aspects of the analysis.
C: (NIST) While this is a 6-month effort in support of the investigation, the work will continue over a 5-year period as part of a NIST competence project.
Project 6 – Structural Fire Response and Collapse Analysis
Dr. John L. Gross, WTC Investigation, Project 6 Co-Leader
Dr. Therese P. McAllister, WTC Investigation, Project 6 Co-Leader
PRESENTATION Dr. John Gross and Dr. Therese McAllister presented information on Project 6 of the WTC Investigation, which is an analysis of the structural fire response and collapse scenarios. The objectives are to determine the structural response of the WTC towers to internal fires—with and without aircraft impact damage—and the structural response of WTC 7 to debris damage and internal fires. In addition, the Project 6 team will identify the most probable structural collapse mechanisms for the towers and WTC 7.
Dr. Gross presented progress on the analysis of the WTC towers. Work is under way to develop finite element models of components and subsystems for nonlinear thermal-structural analyses. These include a single exterior panel section, a floor section (80 inch wide), and a truss seat connection. NIST is evaluating performance under load and elevated temperature. Dr. Gross presented some results of the preliminary studies with illustrations from the models. NIST has awarded a contract to develop additional finite element models of floor systems and the entire tower structures and to conduct nonlinear thermal-structural analyses. Dr. McAllister discussed preliminary studies of the effect of floor removal, elevated temperatures, and load redistribution on column failure. The effect of fireproofing condition on the time for smaller columns in the impact area to reach elevated temperatures within a 1 to 2 hour timeframe was also presented. She explained that ongoing studies include the effects of load redistribution as individual columns fail, as well as the effect of floor removal and elevated temperatures, to gain insight into component behavior for global studies.
Dr. McAllister described the structural features of WTC 7. There is a lack of evidence of significant damage to structural, fireproofing, or fuel systems. The fuel distribution systems were located between floors 1 to 9. NIST has awarded a contract to develop component, subsystem, and global models and to conduct thermal-structural analyses. She stated that NIST has made progress on collecting and reviewing data such as final design drawings, specifications, and the fuel distribution system. Photographs and videos have been reviewed to develop collapse hypotheses by establishing the initial damage to WTC 7, the timeline and locations of fire and smoke, and the timeline and location of exterior signs of collapse (penthouse, windows, etc.).
C: You keep using the word simplified when describing your modeling efforts. You need to make clear what simplified means. It's important for people to understand that we simplify things after we figure out they don't need to be complicated. It could be misconstrued. It's not simplified until after you have tested certain other things on one end of the extreme and on the other end to get a feel as you move into 3-D modeling. You should identify dead ends and areas that you looked into and that didn't lead anywhere. People will be looking at this closely and will want to know what areas were looked at.
Q: I second the comment about the excellence of this work. Referring to the 600º C figure versus the gravity load, I assume that the columns are plumb. Have you combined expansion of the joints with heating of columns?
A: No, not yet. This is continuing as part of our ongoing work. We will be looking at deflected columns due to heating of thefloor system.
Q: Is there an overlap with Emil's work?
A: There is an overlap in the sense that Emil is looking at simplified methods to rule in or out possibilities. I wouldconsider the two as complementary efforts.
Q: Column failure depends on the dead load. Is there a significant difference in dead load for a column damaged on the 90th floor versus the 80th floor?
A: Column failure depends on the level of stress; columns on the 80th floor had a greater cross-sectional area than thoseon the 90th floor and so the stress was about the same in both.
Q: In the discussion of force levels, column yielding was the lowest load and the floor buckling is the highest load. Did you also look at the bolted connection?
A: We are looking at the column-to-column bolting and determining yield strengths. We need to do a variety of column analyses, and we will be combining our analysis of joists with the workmentioned above.
Q: From what direction was the video of the WTC 7 building listing shot?
A: The video was shot from the north. The east penthouse isfalling.
Q: What were the locations of the diesel fuel tanks?
A: The tanks were located in the southwest corner.
Q: What about the utility substation enclosures?
A: The transformers were housed in transformer vaults. We've read that silicone fluids were used in the transformers. [Subsequentfollow-up found that oil was used in the utility transformers.]
Q: The transfer trusses are susceptible to thermal loads. Is there any evidence that the trusses in WTC 7 received any thermal damage?
A: There is no evidence of heating yet. We need to postulatehypotheses and evaluate their plausibility.
*Recommendation: The Committee recommended NIST investigate the fire exposure on the east side of WTC 7 where the cantilever transfer girders were located because they are susceptible to failure under fire exposure.
Project 4 – Investigation of Active Fire Protection Systems
Dr. David D. Evans, WTC Investigation, Project 4 Leader
PRESENTATION Dr. Evans presented information on Project 4 of the WTC Investigation, which includes an analysis of the active fire protection systems used in WTC 1, 2, and 7. NIST is looking at the fire sprinkler system, including the fire standpipes and pre-connected hoses; the fire alarm system; and the smoke management system. Dr. Evans discussed the approach for documenting the design, installation, and modifications of these systems as well as assessing their probable performance on September 11, 2001. He briefly reviewed the results of the NIST fire history study of 12 significant fires (not including the well-known 1975 fire and the 1993 bombing) that occurred after occupancy in WTC 1, 2, and 7. Dr. Evans presented the fire alarm system monitoring record for WTC 7 on September 11, 2001.
Q: You mentioned that you had collected information on sprinklers in the WTC buildings. I suggest you also look at the One Meridian Plaza fire. It could be helpful for WTC 7. I thought WTC 1 and 2 were not sprinklered originally? Were they retrofitted later?
A: Originally, the WTC 1 and 2 buildings were sprinklered only in the below-ground level areas. However, a program was initiated by the Port Authority to sprinkler the buildings, and we havebeen told that the program was complete.
Q: Is there no information on fires in unsprinklered areas?
A: We have information on all fires responded to by FDNY from FDNY.
Q: The equivalent and actual fire size?
A: We have all of the FDNY fire response records to look at for information. For fires in nonsprinklered areas, we considered other indicators, such as hose line use and fire detector activation, to determine whether the fire was considered significant ornot.
Q: In getting data on fires, did you screen out non-fire responses?
A: Yes. The total record count only includes fires.
Q: In the 1993 bombing, the control room was demolished. What was the design of the evacuation and alarm system in the towers on 9/11?
A: I can't give all the details, but I understand it provided an audible alarm plus voice commands.
C: The interviews with occupants may help to clarify this information. In the past we have seen instances where the alert device can mask announcements.
Q: A remaining question is in the absence of the ability to deliver a signal. What is the default capability?
A: We want to know the default message to understand how thesystem worked.
Q: In WTC 1 and 2, you stated you don't have records of what the alarms did on 9/11, were they destroyed in the collapse?
A: Yes, we have been told that the records recorded by the fire alarm system were destroyed by the building collapse. The fire alarm system records in a building are like "black boxes" on aircraft, but they aren't built likethat.
Q: Could the alarms in WTC 7 have been triggered by dust when the towers collapsed? Did any of the other buildings in the WTC complex have alarms triggered by dust?
A: That is a possibility. Smoke alarm systems are susceptible to activation by dust.
Q: Would alarm records from other nearby buildings subjected to dust from the collapse but not fire be of value in understanding the alarm system performance?
A: That is a good idea. We will investigate that.
Project 7 – Occupant Behavior, Egress, and Emergency Communications
Mr. Jason D. Averill, WTC Investigation, Project 7 Leader
PRESENTATION Mr. Jason Averill presented an update on Project 7 of the WTC Investigation, which includes an analysis of occupant behavior, egress, and emergency communications on 9/11. The focus of the presentation was on collecting data through first-person accounts, including 800 telephone interviews, 350 face-to-face interviews, and 10 focus groups. Mr. Averill indicated that the NIST study, using a multidisciplinary approach bridging expertise in a new and innovative way, is one of the largest evacuation studies ever conducted. The required approvals and training are completed, and the first interviews are under way. He presented the methodology for identifying possible volunteers to be interviewed.
Q: What is driving the limit on the number of face-to-face interviews that will be done? Is it budgetary or something else?
A: It is strictly a technical decision. We have taken a look at some of the areas that we want to conduct face-to-face interviews in, and those were in the white paper that was previously issued. The number has changed somewhat and that is strictly a function of the fact that we probably had overestimated initially the numbers of people that we would need to interview in particular sections of the building. If it were an issue that we needed a few more interviews, we could find a way to make that happen. We looked at the areas and said that if we have x number of interviews of this type is that enough for us to answer the types of questions we have. We feel that we can accomplish our objectives with this number of interviews.
Q: You are gathering observational data. Why wouldn't you talk to everyone if you could?
A: In the interview process, you reach a point of diminishing returns. We are planning to talk to as many people as we canwho were in the impact zone.
Q: If there were an event in the future with a large number of people to interview, like we had at the World Trade Center, how long would it take to procedurally get something in place to support the interviews?
A: In the future, we will view interviews as evidence collection procedures, not as human subject research. We will still have a procedure to protect interviewees. More systematic studies will likely require a pre-cleared human subjects protocol.
C: It is imperative that NIST have a standby protocol in place for a quick response to an event. A task force should be assignedto make that happen.
Q: I'm concerned about telephone interviews. Elaborate on the procedures to get a statistical representation of the population. People have to be selected randomly. You said that a great deal of letters were mailed out. But we don't know in advance how many recipients were in the building. We haven't defined the universe. How do you ensure a random sample with an undefined universe?
A: The buildings were divided into zones using the mechanical floors as the divider so that we have a low, medium, and high strata in WTC 1 and 2. We then divided it up into tenants with 40 percent or more of the occupied floor area and those with many smaller tenants. We have the badge list, which would be a superset of people with access to the zones.
A: The frame is the best surrogate you can get for your population. In most surveys, you are dealing in some way with a frame that is less than perfect. We tried to get as close to perfection as possible. The badge list is the best source of data. There are almost 94,000 records. We were able to eliminate 33,000 records for people that were not in WTC 1 or 2. We also matched known deceased. That left us with 45,000 records. We then divided the list by building. We took 13,000 from each building as a random sample. We matched 75 percent of the 13,000 using a Lexis-Nexis database. Twenty-five hundred from each building were randomly selected and sent a letter. Letters were forwarded where possible; undelivered letters were returned. We are getting calls from persons who received letters, but were not in the buildings on September 11. We eliminate these people from the list of potential interviewees. We will do the same when calls are made. We're taking as much care with samples as possibleand keeping track of the results.
Q: In the past, you have mentioned the integration of the other studies that are ongoing with this study. What is the status?
A: We are going into the field; Columbia University is at a different phase. We'll formally get together and discuss what we're finding. This is a Federal investigation, and we have to figure out how to integrate results.
C: (NIST) The two studies are complementary. The Columbia focus is on sociology and demographics, ours is on evacuation and egress. There's synergy between the two studies, but they have to go forward in and of themselves. The intent wouldbe to share results and findings when they become available.
Project 8 – Fire Service Technologies and Guidelines
Mr. J. R. Lawson, WTC Investigation, Project 8 Leader
PRESENTATION Mr. Randall Lawson presented an update on Project 8 of the WTC Investigation, which includes an analysis of fire service technologies and guidelines. The tasks of this project include data collection, field interviews, and re-creation and analysis. Mr. Lawson discussed the interviews that will be held with the Fire Department of New York City, the New York City Police Department, and the Port Authority police and security staff. NIST has analyzed most of the radio and telephone communications tapes in its possession. Mr. Lawson reviewed some of the noteworthy radio calls, including a timeline for the dispatch/arrival of emergency response units. He stated that although the radio systems appeared to work well during normal operations, roughly one-third to one-half of the radio communications were not completed due to surge load conditions.
Q: You're looking at dispatching and documenting the units arriving on the scene. Many units responded that were not dispatched. Can you quantify the impact this may have had on communications?
A: Many did not show up in the radio traffic. It is hard to document.
Q: Could you get qualitative data from the interviews?
A: Yes, we do have an approach to try to gauge that.
Q: You made a statement about two orders for evacuation. Where did the call come from for evacuation? Did it come from the field?
A: We are trying to determine the location of the person who made the call, and we want to understand what is normally done with such a call.
C: (NIST) We do not want to identify the individual. It wasan officer with responsibility.
Project 3 – Mechanical and Metallurgical Analysis of Structural Steel
Dr. Frank W. Gayle, WTC Investigation, Project 3 Leader
PRESENTATION Dr. Frank Gayle presented an update on Project 3 of the WTC Investigation, which includes a mechanical and metallurgical analysis of structural steel recovered from the WTC towers. The 236 pieces of steel from WTC 1 and 2 have been catalogued and identified; NIST has samples of all 14 grades that were used in the towers. Dr. Gayle reported that the steel has been analyzed to determine local failure mechanisms and repeated patterns of fracture/failure. The photographic evidence was enhanced and compared with the recovered steel. Testing for metallurgical and mechanical properties is under way, and tensile properties have been determined for all perimeter columns.
Q: I'm curious. Using a high number of different steel grades is kind of unusual. What was the reason for all these different grades of steel? Was it for architectural purposes? The slide showed that the stiffness of the spandrel takes the load. It really does distribute the load. What was the reason for the number of grades? Was it to keep the columns the same size for architectural purposes?
A: The exterior part of the columns was the same size. My understanding is that the large number of steel grades was used to keep theweight down.
Q: Usually the dead weight of steel is not that important. First you set the architectural enclosure so you can use the most efficient and cost-effective solution. I suspect fine-tuning. Did they vary the steel weights around the perimeter at a level or does it vary from top to bottom?
A: It varies everywhere and in every which way. It tends to be ¼-inch steel for the perimeter columns in the upperfloors.
Q: Where does the load really go? Are there any flaws in using high-strength steel?
A: We are looking at this issue. We've read that the design tried to make the stresses in the columns and the core equal. This is part of the baseline analysis.
C: (NIST) We haven't yet found the rationale for using the many high-strength steels. We're trying to determinethe rationale.
Q: Is there a difference in weldability? Are there any subtle metallurgical issues?
A: It is an issue with high-strength steel. There has been a lot of work done by the Japanese on weldability [referring to the perimeter columns, where steel was supplied by YawataSteel].
Q: How are you adjusting the ASTM loads with the dead loads of the building?
A: There's a factor, maybe 3.5 kips per square inch or something on that order, and we have one for each steel to take it to a static test. When you have different strain rates, in this case for the estimated properties, we took all the data we had from the literature and accounted for the strain rate used in order to adjust them to the properties that you would have measured at the strain rates they used for mill testing. In that case, we probably have to modify up some test results because they tested a faster strain rate at the steel mills, but you have to have some equal basis to do that. But if you have a static load in a building, you may want to bring that down to account for the strain rate.
C: I wish you had told Bob Hanson or me that you had to do that. We did a lot of work following the Northridge earthquake, and there are reports that would give you that.
C: The report was FEMA 355.
C: (NIST) We have a complete set of the documents here at NISTand are using them.
C: Regardless of what data you are going to generate, I would have a problem using these numbers for the Project 2 analysis. There is too much variability with data from different mills, etc. I encourage using it as the reference point for the analysis. Spend more time doing literature searches and getting the typical properties for steel and use that for materials properties. Take the mean of the error bar.
A: We are using test data generated by NIST as a checkpoint to confirm that we don't have anything out of the ordinary. We have taken your advice on the use of data reported in theliterature.
C: I have a problem justifying the need for triaxiality data on welds. I don't think it is a triggering mechanism and will take too long to complete.
Public Comment Period
Mr. Fitzgerald stated the ground rules for public comments. Each speaker has 5 minutes to address the Committee. Members of the public may submit their comments in writing at the meeting or at any time. Mr. Fitzgerald read into the public record the comments of Mr. Donald Bliss, who submitted a prepared statement.
Donald Bliss, National Association of State Fire Marshals' Partnership for Safer Buildings
Mr. Fitzgerald called the first speaker to the podium, Mr. Arthur Scheuerman.
Arthur Scheuerman, retired FDNY Battalion Chief
I appreciate the opportunity to speak to you today. The main problem as I see it is building codes. I can't believe that they are still reducing the timeframes for how long columns and beams must be protected. You all are doing a great job. I can't believe the job you all are doing.
I have a different opinion on the collapse mechanism. I read a report by a British engineer, Usami. He comes to the opposite conclusion that the columns failed from a sudden release of compression when the columns buckle. In other words, as the columns are being pushed out, they're building up compression and when they buckle, the sudden release of the column pushes them in further and they collapse. His graphs show just the opposite. His graphs show that the floor joists pull the columns inward as they sag. I don't know whether it's from thermal bowing or loss of strength, but his report is very interesting. As shown in the overhead, the columns are very stiff, so floor 4 is under compression and floor 3 is in tension because this is acting just like a lever. The two buckling floors are putting the columns in floor 4 under compression and floor 3 is under tension. There were several reports or pictures of an explosion out of the side of the building. That, I think, was floor 3 or like the 79th floor exploding outward and pulling the floor with it. It may have disconnected from the core columns. Usami did a finite computer element analysis, like you did, and concluded that the buildings failed as the columns collapsed inward. In no case do the columns go outward. Thank you.
Mr. Fitzgerald called the next speaker, Mr. Ray Scott.
Ray Scott, Battalion Chief, Prince William County, Virginia
Mr. Fitzgerald thanked the presenters, attendees, and speakers. He adjourned the meeting at 5:40 p.m.
December 3, 2003
Mr. Fitzgerald called the meeting to order at 8:10 a.m.
Status of The Station Nightclub Investigation
Dr. William L. Grosshandler, Chief, Fire Research Division
PRESENTATION Dr. William Grosshandler is leading the NCST investigation of the fire at The Station nightclub that occurred in West Warwick, Rhode Island on February 20, 2003. He presented information on the status of the investigation. NIST has about 75 percent of the building geometry identified, including the floor plan, vent locations, doors, windows, and ceiling heights. NIST is still seeking more construction drawings of the building and fire alarm system plans as well as photographs or videos of the outside or inside of the building prior to the fire.
Dr. Grosshandler described the tests that have been conducted thus far, including samples of 8-foot by 4-foot wood panels covered with a common non-fire-retarded foam. Videos recorded the flame spread in an exterior and interior corner arrangement and various measurements were taken. NIST is also analyzing the evacuation process by using commercially available software models. Other tasks include documentation of emergency response to determine how the fire department was notified. NIST is collecting emergency response data in cooperation with the local fire department to document procedures and operation of equipment.
C: You indicated that it is difficult to get information on the wall covering. From the viewpoint of improvement to codes, it is of interest to do tests of fire-retardant foam to the extent funds are available. You may be able to learn something by varying the parameters.
C: (NIST) Even if we can't get the exact foam used in the club, it is important to move forward. There is a lot to be learned in any case. There are no funds for additional full-scale tests. There are no codes for use of this type of material as a wall covering (fire-retardant or not); maybe looking at the fire spread is worth doing. We've been doing this for many years. We need to be able to reproduce the fires better than we've shown. We have not tuned the experiments. We're not moving at the pace we would have liked. We have waited for access to data. However, we feel we are still in a position where we will be able to apply findings with confidence.
Q: One of the criteria you mentioned for when NIST would be involved in an investigation was the quality of the analytic team—it will never be the case where a state or local investigation could match NIST's qualifications.
A: This was taken from the implementation rules. There are statewide, county, and local investigations of the Chicago fire. We have confidence they will reach the same conclusions we would. That is one of the reasons we chose not to investigate the Chicago fire.
C: Even if there are other investigations ongoing, they may have different objectives relative to an NCST investigation. I understood that the Chicago decision was mostly a resource-driven decision.
C: (NIST) Resources were not the only driver.
C: (NIST) We want to make a difference and not duplicate other efforts. We want to be able to make a broad recommendation that would affect national practices.
C: I would argue that all of those conditions are met in Chicago. Another point, it should be up to the team to conduct interviews and to gather all the data. Assume that information is not available from other sources in a timeframe that is useful. It should be in the team's mind to obtain information directly in any future investigation.
C: (NIST) You're right, certainly we're learning. We need to start gathering information from day one.
C: From a social science point of view, this (the Rhode Island nightclub fire) is an important event. Evacuation and the thought processes of occupants are important to understand. Panic is a controversial subject and there seems to be an assumption that it played a role here. It will be very important to look into the extent that it did happen. I'm not saying there was panic; I'm not saying there wasn't. As we saw on the film, there was a delay in the response of people as they were normalizing. Maybe they thought fire was part of the act. It is important to have good social science input when people are interviewed. I would encourage NIST to seek out that kind of support when developing the questions that will be asked. Beyond that, NIST should have in-house social science capability for conducting these kinds of investigations.
Q: (NIST) Agreed. However, interviews will be investigative and are not addressing these issues. In this case, we don't have the funds or access to do what you suggest and will be unable to fulfill this desire within the timeframe available. Could such a study be done as a follow-up, say six months from now?
A: (Committee Member) interviewing the survivors? Yes, it could be done and it's essential. This is central to understanding decision processes and group dynamics in these rapid evacuations. As you aptly pointed out in your presentation, current models don't capture this type of behavior. How do we improve them if we don't get a deeper understanding of the behavior involved?
C: We may need to call upon you when we draft that part of the report to make sure we get the right recommendation.
C: (NIST) We are operating on an ad hoc basis. Until we get appropriated funding, this is about where we are. Other capabilities are necessary to conduct investigations. We need to take these recommendations seriously, and of course, we also need the resources to put those recommendations into action.
C: We wouldn't expect this in 2004. These recommendations should be considered in the long range. We have the responsibility to help in the overall competency of future investigations.
*Recommendation: The Committee recommended NIST develop in-house social science capabilities to perform egress studies for future investigations.
Q: There is no competent model for the behavior of people in fire evacuations. This makes Project 7 of the WTC Investigation so important (first-person interviews). Can you tell us the number of survivors in the Rhode Island fire?
A: There were 430 people in the club. One hundred people died. There were 330 survivors.
Q: Do you have information on when the building was constructed and for what purpose?
A: The building was originally a restaurant.
A: Yes. It was converted to a nightclub in 1980 or 1981.
A: (Committee Member) According to the public records, the building was constructed in 1950. Anecdotally, the initial building may date back to World War II.
Q: The building may actually date back to 1941 or 1942. Were there changes to the floor elevation?
A: There was a subfloor added near the front of the building, and the stage was added at a later time. The building permits, which are filed publicly, provide some information, but don't match the structure itself.
Q: The plan is presently for about one investigation per year. Historically, that's probably good. If you have three events in a year that are within what we're supposed to do, how do you choose which one to investigate?
A: The approach was to address upset conditions mostly based on natural phenomena that statistically would average out to be one a year. This does not take into account manmade events such as terrorist acts and other events. There is a lot of concern that NIST would investigate beyond what the act envisioned. If we had three serious events in a given year, we would request the authority to investigate all events within the scope of the act.
C: I would like to see NIST have the funds and manpower to support three investigations, if there are that many in a year, and want to recommend that funds be made available.
C: I have a comment with regard to the safety of first responders that is also part of the charge. Future investigations should address communications criteria. It may have been a problem at The Station. This is an ongoing issue that should be highlighted. In Chicago, there was great confusion on the fire ground.
C: (NIST) NIST has been asked by the Governor of Illinois to assist with the Chicago fire, and we will assist to the extent that our limited resources allow.
NCST Act and Requirements
Dr. James E. Hill, Acting Director, Building and Fire Research Laboratory
Dr. James Hill discussed the NCST rules published in 15 CFR Part 270. When the act was passed, NIST was told by Congress to publish rules within 90 days. The final rule addressing the collection and protection of information was published in May 2003. An interim final rule with additional procedures was published on November 28, 2003, requesting public comment (68 FR 66703 pdf file). Dr. Hill asked the Committee to comment on this rule during the public comment period on the Federal Register notice, and said that NIST would also welcome comments individually. He began the discussion by highlighting four items in the November 28 interim final rule.
1. Section 270.2 adds a definition of credentials. This includes photo identification and badges, deemed appropriate by the Director similar to the National Transportation Safety Board.
2. Subpart B addresses the establishment and deployment of teams generally. The expectation is that NIST will conduct one investigation per year. NIST could investigate more than that.
C: More realistically, I think we are looking at three to four investigations per year of the scale of The Station nightclub fire and one at a fraction of the scale of the World Trade Center. It may be better to put this into the rule to help with public expectations.
C: (NIST) In terms of the procedures, there are things that we can do as NIST and things that you can do as an Advisory Committee. Perhaps these thoughts could be compiled and included in your report.
C: We prefer incorporating that in our report.
C: (NIST) I'm not sure I'm in full agreement with the earlier comment on three to four investigations. The Station nightclub fire is a once in a generation event. It depends upon your interpretation. Three to four may benefit from our investigation, but it's not clear that that was the intent of the act.
3. There is a section called Preliminary Reconnaissance. This section simply gives NIST the flexibility to establish a team before doing preliminary reconnaissance or send staff out before a team is established.
4. On page 66705, Part 270.102 sets the conditions to establish and deploy teams. Paragraph (a) comes directly out of the act. Paragraph (a)(1) deals with major failure of one or more buildings due to a natural event, a fire that resulted in a major building failure, failure of a building at less than its design basis, and an act of terrorism that results in a Presidential declaration of disaster. Then the third element is spelled out in the next paragraph, which states that the investigation will likely result in significant or new knowledge or building code revision recommendations to reduce public risk or economic loss.
Some of the issues that came up as NIST was pulling together this rule are stated, whether there are sufficient funds available to support the investigation, whether the investigation warrants the advanced capabilities and experience of a team, if the cause is readily apparent, whether the investigation is likely to result in relevant knowledge, and whether the investigation will be duplicating some other investigation.
C: Shall we discuss this now?
C: (NIST) The rulemaking schedule versus the report to Congress—how does the report affect rulemaking?
C: The Committee should respond separately on the rule in a letter to Dr. Hill.
C: (NIST) the concern is that the Committee has had the rule for only one week.
C: (NIST) The Advisory Committee as individuals are free to comment on the rule. The diversity of comments may be better than a consensus. As an Advisory Committee, consensus advice must be reached in public session.
The Committee took a break to read the interim final rule. When the meeting resumed, Dr. Hill went through the interim final rule section by section, asking the Committee members for comments.
C: (NIST) Refer to the second page, middle column toward the bottom. Any issues on credentials? None. The third column, one-third of the way down. Are there comments on natural disaster frequency?
C: I suggest replacing the last "than" with "more than." If you look at earthquakes, in Northridge, there were four to five classes of building structures that could warrant investigation. Consider the number of events versus the number of investigations.
C: I second that comment.
C: (NIST) The statement refers to building failures, not investigations.
C: As well as potential for loss of life.
C: Integrate potential. Resource and organizational planning should assume three to four investigations at The Station level over more serious events.
C: (NIST) There was a suggestion to say "more than".
C: Three to four is more than one. That's okay by me. I can live with "more than." Resource planning is a separate issue.
C: Suggest we use the words "more than."
C: (NIST) It may be more appropriate to say two to three.
C: Bound it. Say less than 10.
C: (NIST) Setting expectations is the issue.
C: I would suggest several times per year.
*Recommendation: The Committee recommended that wording in 15 CFR 270.100(a) be changed to read: Historically in the United States building failures from fires, earthquakes, hurricanes, tornadoes, and other disasters that have resulted in a substantial loss of life or that posed significant potential for substantial loss of life have occurred "several" times per year.
The recommendation was agreed to unanimously by the Advisory Committee.
Section270.101, Reconnaissance. No issues.
Section 270.102, Conditions for Deployment.
C: On (iv), the Federal Response Plan, it is about to be superseded by the National Response Plan.
Q: Section 270.101. Meaning of a team?
A: Team is the team to do the reconnaissance.
C: The same team would do the investigation?
A: The section gives the flexibility to either do reconnaissance or establish an investigation team.
C: In section 270.102, paragraph (a), the list does not capture the idea of investigating building fires. Fires may have potential for loss of life without building failures.
C: (NIST) The logic goes hand-in-hand with everything else here. The incident would have to meet the loss of life plus a type of fire criteria.
C: (NIST) Section 100, paragraph (b) defines a building failure.
C: The criteria may be too restrictive on when a team may be active.
C: (NIST) This section is out of the act.
Q: (NIST) Are specific scenarios not listed?
C: The Chicago high-rise had the potential for loss of life without building failure.
C: The earlier suggestion may be the way out.
*Recommendation: The Committee recommended that wording in 15 CFR 270.102(a)(1)(ii) be changed to read: "a fire that resulted in a building failure of the building of origin and/or spreads beyond the building of origin."
C: Under the list of five items that the Director will consider: number three (15 CFR 270.102(b)). If the technical cause is readily apparent, sometimes the cause is not the issue. Understanding is needed to have some impact on codes, standards, and practices. Add the appropriate mitigation measures clause.
C: There were comments during the Rhode Island presentation relative to duplication [of local investigations].
C: (NIST) The intent is not to duplicate state or local investigations.
C: That is highly unlikely. I recommend striking that requirement.
C: I support that.
C: I don't think you want a clause like that in the rule.
*Recommendation: The Committee recommended that 15 CFR 270.102(b)(4) be deleted because it is highly unlikely that the NIST investigation would duplicate local or state capabilities.
*Recommendation: The Committee recommended that a new section 15 CFR 102(b)(4) be added that the Director will consider whether an investigation is likely to result in relevant knowledge for mitigation of the building failure.
C: First page, middle column. Purpose of investigations (15 CFR 270.101(b)(1)). Other purposes. Add: "..., safety of building occupants, and it emergency egress and response measures." Life-safety is a preeminent criterion. NIST is charged with looking into egress issues and activities of first responders.
C: I support that. It is consistent with section 270.100.
C: On page 66705 (15 CFR 270.104(b)(5)): after "civil" add structural, mechanical, and electrical. There is an issue on projects where NIST attempts to respond but has limited resources or there is a criminal investigation under way. NIST could extract data later from litigation to determine if important issues came up. The John Hancock building took years for the real story to come out. The Citicorp building is another good example. On The Station nightclub fire, why not wait until information comes out and see if it warrants changes in the codes?
A: Nothing prevents NIST from reexamination later. The report may state there is a possibility of future reexamination after the criminal investigation is complete.
C: Within the structural engineering profession, word leaks out. Forensic engineering firms are the leading edge. NIST may not be the right group, but I can't think of anyone else.
C: (NIST) The sun doesn't set on our role in code development.
C: (NIST) Our existing authorities are still available. We continue to do investigations that are not NCST investigations.
C: I want to revisit 270.201(a). The rule does not address state and local authorities.
C: (NIST) Paragraph (b)(2) addresses state and local criminal investigations. This is not part of the statute but an exercise of NIST's administrative rulemaking authority.
Q: Does this fall under the Attorney General's purview?
A: (NIST) This is administrative. It is not addressed in the act.
Q: Is it clear that the team would have priority over civil litigants?
A: That is a fair point. When there is litigation, we respect the rights of the litigants to seek legal remedies.
C: I would like the statement to clarify priority relative to civil litigants.
A: We have to be sensitive to the fact that 15 USC 281 removes from the evidentiary stream the contents of NIST reports. We cannot extinguish the value of evidence to litigants.
C: (NIST) We do have subpoena authority if needed.
C: (NIST) An example is the steel at JFK. We found a mechanism to go with the civil litigants to get what we needed.
C: You got all you wanted, but it was time consuming.
C: (NIST) Another example is the negotiation with New York City. The city is facing $12.5 billion dollars in claims. The city is represented by excellent lawyers. We found a solution that protected New York City while meeting our needs. The city is concerned about emotional statements and criminal investigations. Our experience is that there isn't a cookie-cutter approach and we need to be fair to other parties.
C: This is an important area for rulemaking. I encourage NIST to look carefully at this issue.
C: (NIST) We place guidelines in rulemaking on priorities. Rulemaking affects NIST. We're not able to affect efforts at the local level. The NIST Director has the broadest discretion and the ability to get information through the use of subpoenas. We don't want to limit the discretion of the Director.
C: The length of time to complete investigations due to availability of evidence is a concern. Forensics is best done quickly. Civil litigation has to be looked at seriously.
C: In 270.202, a factual correction. The Federal Emergency Management Agency does not have local offices. Suggest saying FEMA urban search and rescue teams, local emergency management agencies, and local emergency response groups.
C: (NIST) We will consider those comments along with the public comments in the final rule.
Dr. Hill had asked members from the NIST NCST Secretariat to reflect on their experiences and lessons learned from the investigations over the past year for use in preparing a final report. He had listed those ideas on a handout and passed it to each Advisory Committee member.
Lessons Learned from Current NCST Investigations
What has worked well
- Hire an administrative subcontractor (i.e., SAIC) for large investigations (i.e., WTC).
- Use the NIST Human Resources and Management Division to hire individual experts as an alternative to contracts whenever possible.
- Conduct training for all NIST staff participating in an investigation so they understand all administrative processes and responsible parties.
- Focus specifically on the information needed from others for an investigation rather than make broad non-specific requests. Use all possible paths to negotiate for the information.
Where can we improve
- Establish and staff an NCST office within BFRL (or NIST) to provide the basic infrastructure to conduct investigations and prepare NIST for responding quickly to future building failures.
- Establish the basic internal decision making processes and tools needed by NIST to conduct investigations promptly and efficiently. Implement faster less conservative decision-making. Make maximum use of small working groups (as opposed to a large NCST Secretariat) to work through critical issues and the NIST Chief of Staff to aid in decision-making.
- Establish a blanket contract to provide technical expertise to support investigations. This would eliminate the burden of the vetting process with multiple contractors and of multiple COTR responsibilities for NIST technical staff.
Dr. David D. Evans, Fire Protection Engineer
Dr. Evans presented information on the community-scale fires and fire concerns associated with increasing urban wildland interface. He explained the characteristics of these fires and stated that 36 percent of U.S. homes are in wildland-urban interface areas. A technical challenge is to resolve the fire spread at the scale of individual land parcels and buildings. NIST has performed physics-based community fire spread models using the Fire Dynamics Simulators. Dr. Evans discussed the modeling, fire tests to measure heat release rates, and NIST participation in community-scale fire investigations.
Q: Is the house standing in the lower left corner of the photograph? Would you study construction and features of that house?
A: Yes, that one house is standing and all others have been destroyed by fire. We would document the construction, but we have to ask more general questions about the event. We need to know about weather, wind, the first property ignited, and fire department decisions during response, etc. Many things in addition to construction can be important in determining which houses survive.
Q: Has NIST worked with the fire insurance companies?
A: We have been approached by the insurance industry about our technical work in modeling community fires.
Q: Did you look at the Saint Lawrence burns?
A: We have the reports in the NIST Fire Research and Information Services collection. We make use of that information in calibrating the model.
Q: If you look at the Southern California fires, are there priorities you would like to look at?
A: Science and calibration. As in the present investigations being conducted by NIST, getting access to the videos that were taken would be useful. If there were videos of isolated trees burning, we could learn immediately if our present correlations obtained in the laboratory for burning time can be extrapolated to larger trees. Video may show how structures were taken in the community, the time for structure-to-structure spread, the fire service response, and where people were lost and why. A lot of that information is probably there and would help us.
C: (NIST) The mechanisms of fire spread for community-scale fires are not well understood. The wind fields that strongly influence fire spread are determined as much by the fires themselves as by local weather conditions. The scale of these interactions is largely unknown. This complex interaction can only be predicted by an organization familiar with both building and outdoor fire dynamics.
C: (NIST) If you are there during the event, you want to understand how these houses are taken and which houses are taken. We have aerial pictures of the houses before the event, in this case, the Oakland fire. You can pick out house with tile roofs, and there were many of those houses that weren't there anymore, just tiles lying on the ground. So, these houses weren't taken by the roofs catching on fire. Quantifying these effects helps us understand these fires better.
C: (NIST) The Forest Service considers only fires in wildland fuels (trees, shrubs, and grasses, for example), not fires in areas where structures and wildland fuels coexist. Fire departments in wildland urban intermix regions recognize that techniques for fighting fires in wildland areas differ from those in wildland urban intermix areas. As a result, they have begun cross training their structure firefighters in techniques to combat fires in both wildland and structures.
Simultaneous Criminal and Safety Investigations
Mr. Craig S. Burkhardt, Chief Counsel for Technology, Department of Commerce
Mr. Michael R. Rubin, Counsel for the National Institute of Standards and Technology
PRESENTATION Messrs. Craig Burkhardt and Michael Rubin presented information on the ability of NIST to conduct NCST investigations at the same time as criminal investigations. The NCST Act requires that the NIST investigation team relinquish investigative priority in the event of a criminal investigation. In the WTC Investigation, NIST has cooperated with the Fire Department of New York, the New York Police Department, and the Port Authority. In the NIST investigation into the fire at The Station, NIST has structured its investigation to ensure a successful outcome with the available resources, and respects the jurisdiction of the State of Rhode Island. NIST is taking steps to establish its investigatory reputation and develop agency relationships.
Q: The question about priority. How does it relate to people speaking to you voluntarily? Is that a violation of priority?
A: There is no conflict—get what you can get. The problem arises when people realize they may be the target of a criminal investigation. Priority does not tie our hands. The evidence is subject to priority. Until another agency asserts priority, we can continue our activity.
Q: A two-part question. The act reads that investigation materials cannot be used in criminal proceedings. Can NIST employees be subpoenaed? What about non-Federal employees?
Q: 15 USC 281a refers to the use of a NIST report in a suit for damages (civil). Anyone can be subpoenaed; the question is if it will stand up. The Department of Commerce will not allow its employees to be expert witnesses except in actions where the government is a party. This is repeating case law (Touhy). Would the Supreme Court decide Touhy the same way today as it did 50 years ago? Time will tell. WTC may change thinking. For example, the New York City negotiation; action in another circuit may find that the 15 USC 281a protections are unconstitutional. It's an interesting time to be a lawyer.
Q: Can protection extend to a contractor?
A: I don't know.
C: (NIST) The contract language says that information gained belongs to the Government.
C: There are four scenarios. If you send someone without authority to a site it may be useless. Much more is needed to document failure. The team needs to be equipped (e.g., cameras, video, tape recorder). The person has to have the authority to enter the site immediately.
C: (NIST) It may be more of a Director or resource issue. When a person shows up, reports, and suggests that the event is worthy of an investigation team, the declaration can be made more speedily. Or the decision may be made that the incident doesn't warrant an NCST investigation, thus avoiding the report-writing requirement of the team.
C: What if there are no unusual factors; not deserving of a team? We would not have to engage in collection of evidence and reporting.
C: There are many cases. I know of specific situations where someone would close a site and you may be blocks away. People sent by NIST have to have the credentials necessary to enter these sites. NIST needs to empower reconnaissance teams.
C: (NIST) That is a management issue.
Q: What NIST is asked to do is important; there is a tradition of research. A new responsibility has to upset the cart. What has management done to balance the traditional function of NIST with these new responsibilities? For example, training, etc.?
A: At the end of next year, we will be at the height of receptivity to get support. Good technical work, good reports will build support. We have the luxury of using staff that have done investigations for some time. We will need additional training, operating procedures, credentials, and to build relationships with other agencies.
C: This is the perfect opportunity. The WTC Investigation is the largest fire/structural investigation ever done. NIST has assembled a great team and is developing capabilities.
C: A public relations campaign is needed to build awareness of NIST capabilities/authorities. NIST needs to educate local authorities.
Q: I can envision the need for security clearance at the site of investigations.
A: Many NIST employees have clearances. The challenge is for contractors to obtain appropriate clearance.
Q: The act requires non-NIST members on teams. That may not be possible without having persons on retainer. Reconnaissance teams may have individual contractors with credentials: We may need to revisit the act to address non-NIST personnel requirement.
A: I would wait to revisit the act until after the current investigations are complete. There is sufficient flexibility in the act that we would not have to wait for a non-NIST person to arrive on the site.
NCST Advisory Committee Report to Congress
Mr. Paul M. Fitzgerald, Committee Chair
During this portion of the meeting, most of the discussion was by the Committee members. NIST staff made a few comments, and asked and answered a few questions.
C: Before we get started on the report, we did not complete an item from this morning. Would community fires be appropriate for deployment of a team?
C: (NIST) The Advisory Committee asked for the presentation. We're looking for advice from the Advisory Committee on this issue.
*Recommendation: The Committee recommended that wildland urban interface and community-scale fires be eligible for future NCST investigations.
C: I agree with the statement.
C: It is the opinion of the Committee that these events should be eligible for NCST investigations.
The Advisory Committee members reviewed and discussed the draft 2003 Annual Report of the National Construction Safety Team Advisory Committee to Congress.
Section 2.0, Introduction
C: In the draft report, there is a difference in the level of detail in the discussion of meetings summaries and there needs to be balance. Section 3.3 is well written, clear, and more detailed than the section on the WTC Investigation and may give people the wrong idea.
Q: Are you referring to the public comments section? Are you suggesting that we condense this section?
A: Yes. Acknowledge only the speaker and the topic.
Q: I will condense per the suggestion. Are there other comments on section 2.0?
C: Section 2.2 2.1. To be parallel with the previous discussion, add a sentence to say that several meetings between individual Advisory Committee members and NIST staff have been held.
Q: In section 2.3, what does cross-representative mean?
A: I will change that sentence to read, "membership is multidisciplinary."
Q: In section 2.3, decisions not to form a team begs the question that we did not give thought to investigating southern California fires.
A: We could add a reference to that now.
C: There is no process to report on.
A: The report would be on today's action.
Q: Could we restate the first sentence at the top of page six?
A: Something along the lines of future funding streams are not available to support the anticipated investigations.
C: Rework language from the rule. Several events per year could warrant investigation.
C: Put a period after investigation in the first sentence.
C: "Potential" should be emphasized and other large-scale investigations.
Q: (NIST) Are you trying to say that after WTC we will have other resources freed up?
A: (Committee Member) No. I am trying to point out that those resources could be easily consumed by another accident.
C: (NIST) WTC was a one-time appropriation. There is no continuation of funding. We have no resources for NCST investigations. We've asked for a continuing appropriation, but we have not received any appropriations at this point for the NCST. Unless there is some appropriation, one-time or continual, we have no resources for future investigations.
C: We should definitely make that point in the report.
C: Add that when the WTC Investigation is completed, no resources will be available to support investigations to the bottom of page 5.
C: That last paragraph is a problem.
C: I will revise that paragraph.
C: Investigations are not the problem.
C: Last paragraph in section 2.4. We've already done quite a bit in that regard.
C: Update that the Committee has worked with NIST.
C: In terms of the outline, section 2.5, DTAP is not mentioned. I added a paragraph based on what was written in the act.
C: In section 2.6, human and financial resources. Add a statement to say that in a typical year, there will be on the order of 10 events per year.
Q: Top of page nine. I understand that the $16 million for the WTC Investigation came through FEMA directly to NIST. In the next sentence, $3.4 million was appropriated to NIST?
A: The response plan consists of three elements. In FY03, the $3 million appropriation supported R&D and DTAP. The $3.4 million in FY02 was for research and preplanning for the WTC Investigation. These funds were redirected within NIST and were not an appropriation. The $4 million was requested for FY04. The $3 million in FY03 was also an increase to NIST's base.
C: (NIST) The overall response plan is $56.4 million.
Q: Is it fair to say that the increase is $7 million?
C: The additional $4 million dollars was requested for R&D and DTAP.
Section 3.0, Current Safety Team Investigations
C: There is a paucity of information on the WTC Investigation versus The Station Investigation in the report. I attribute this to WTC being a 2001 event versus a 2003 event. Congress knows about WTC; it knows less about The Station nightclub fire.
C: I wrote this section per the two-page guidance from the Chair. I could expand the section. Is there something substantive that we want to say?
Q: Could we make The Station section shorter?
A: Information on the Rhode Island fire could be included as an appendix.
C: We should follow parallelism as in section 3.1.2 and stick to headings for projects.
C: The first sentence of this section (3.3.3) should be deleted.
C: List the projects and add a sentence that project details are on the Web site.
C: There is the issue of whether to include meeting minutes in an appendix. I would argue against it. Minutes of the Advisory Committee meetings are available on the Web site and generally available to the public.
Q: (NIST) Could you clarify secondary sources used in section 3.1.3? We received Port Authority and other primary source documents beginning last fall the bulk of the information for the WTC investigation is from primary sources.
A: (Committee Member) Gaps in the information existed through much of 2003.
C: (NIST) In terms of the totality of the information, we had 90 percent of the information. NIST tried to fill gaps with information from reliable secondary sources, and in some cases, there were multiple sources of data.
C: (NIST) I suggest referring to NIST Special Publications rather than Web sites. The Special Publications are permanent records. Written reports are best for references because Web addresses change.
C: This is one in a series of reports to Congress. We are trying to track up to the end of the investigation only.
Section 4.0, Recommendations
Q: In section 18.104.22.168, are we going after a quantitative procedure?
A: We'll just say procedure and delete "quantitative."
C: In section 22.214.171.124, the term geographic distribution may not be appropriate.
C: The National Transportation Safety Board does it that way.
C: The idea was to have persons familiar with local practice to assist NIST staff.
C: Calling for that specific thing is inconsistent with the need for infrastructure.
Q: Do we need to add the wildland urban interface and community fires recommendation?
A: Yes we do.
C: I suggest adding a sentence to section 4.2 that the research and development program under way is on target and reflects the Advisory Committee's input to NIST.
C: (NIST) The funding is still not sufficient to make planned progress.
C: (NIST) Suggest adding the recommendation for a group to look at standing procedures for interviews. We will take that recommendation seriously.
C: It should be covered in section 4.1. The WTC Investigation highlights the difficulty in obtaining information.
C: With respect to civil litigation, NIST should use the subpoena power provided by the act.
C: NIST should continue to explore ways to balance investigation needs with those of litigants.
C: Educate state and local authorities.
Q: In talking about an NCST office, should it be headquartered within BFRL or within NIST? I suggest that an NCST office should be headquartered within BFRL.
Q: In the report, we recommend $2 million to fund the NCST office and $2 million for a reserve fund for future investigations. Do they do that?
A: It is possible. The reserve fund would be available to start an investigation.
C: It would be no year money.
Mr. Fitzgerald volunteered to incorporate the comments and provide the report to NIST by December 20, 2003, for submittal to Congress.
Future Meeting Dates
The Advisory Committee discussed dates for future meetings: April 27-28, June 22-23, October 5-6, and December 14-15.
C: The end of April may be a good time to meet.
Q: Would NIST have all contractor reports by the June meeting date?
A: A large number of reports would be available by that time, but not all.
C: The June meeting is important.
C: We could hold the decision on a December meeting. We would like to schedule the dates for the April, June, and October meetings.
The NCST Advisory Committee requested that April 29-30 be considered as potential meeting dates and that the October dates be moved to the middle of the month. The June dates were acceptable to the members' schedules.
Motion to adjourn the meeting was agreed to.
Meeting was adjourned at 2:00 p.m.
Hill, James, Acting Designated Federal Official
Office of the General Counsel
BFRL WTC Disaster Study Secretariat
The Station Nightclub Investigation
WTC Administrative Support