Ila Jones, Florida Delegate & Chair, Regulatory Affairs Committee
Claude Cooper, Richmond and AMCBO Chairman
Cindy Bocz, Texas
Greg Burgoon, Akron
Dr. Shyam Sunder, NIST
Dick Thomson, New York
Rob Austin, New Jersey
Marcel Eglesia, New Jersey
Brian Ferris, Wisconsin
Robert Wible, NCSBCS

As chair of the Regulatory Affairs Committee, Ila Jones welcomed AMCBO and NCSBCS public sector members to the monthly important issues call and thanked them for participating. She introduced the guest speaker, Dr. Shyam Sunder, Acting Deputy Director of the National Institute of Standards and Technology's Building and Fire Research Laboratory and Lead Investigator in the Federal Building and Fire Safety Investigation of the World Trade Center Disaster. (Dr. Sunder had previously addressed NCSBCS and AMCBO members last year on the progress of the NIST World Trade Center Investigation, first at the AMCBO/NCSBCS Building Retrofit Conference in Wilmington, DE in March 2004 and then at the Joint 37^th Annual Conference in Salt Lake City in October 2004.)

On April 5, 2005, NIST had released an update report on NIST World Trade Center disaster findings. During this call, Dr. Sunder would address that and other NIST reports.

Dr. Shyam Sunder thanked Ms. Jones for the introduction and for the opportunity to address state and local building officials.

Dr. Sunder noted that NIST on April 5 did not issue any recommendations regarding their findings thus far in their investigations. On April 5 NIST issued 15 draft reports, including 3 project reports and 12 supporting technical topic reports. These included project reports on Codes and Practices, Evacuation and Occupant Behavior, and Emergency Response and a PowerPoint presentation which showed the probable sequence of collapse of each of the two World Trade Center Towers based upon the impact of the planes, the fire and the aftermath.

Around the third week of June, NIST will release the approximately 200-page final investigation report for the WTC towers, containing principal findings and recommendations for changes to codes, standards, and practices. NIST also will release the remaining 27 reports, including 5 project reports and 22 supporting technical topic reports. All totaled there will be over 10,000 pages of reports out of this investigation. The reports in June will cover other 5 projects, including those documenting the collapse analysis. All 43 reports will be open for a six-week public comment period.

OVERALL SUMMARY:

In undertaking the World Trade Center Disaster Study, NIST had 4 objectives:

1. Why and how did twin towers collapsed after terrorists flew large jet-fuel laden commercial airliners into the buildings?
2. What actually happened during the evacuation and emergency response phases of the event?
3. Determine what procedures and practices were used in the design, construction, operation and maintenance of the buildings?
4. From the above make specific recommendations for improvements to building and fire codes, standards, and practices that warrant revision.

While our recommendations have not yet been issued, some of our preliminary reports have raised some issues for the public’s consideration.

One of the main questions we were trying to address in our work was, how and why the North Tower (Building 1) stood nearly twice as long after its impact as did the South Tower (#2)? The North tower stood for 102 minutes after impact and the South Tower for 56 minutes after impact.

We also are asking in our research:

• What factors related to normal fire safety could have explained the above difference or helped either tower stand longer?
• Would the undamaged towers have remained standing in a large conventional building fire?
• What factors could have minimized loss of life among occupants and emergency responders?
• Did the procedures and practices used in the design, construction, operation, and maintenance of the WTC towers conform to then accepted procedures and practices?

Looking at the Collapse side of the event, NIST did a thorough airplane impact analysis – we looked at analysis results, video tapes, and photographs and tried to estimate the number of columns that were: destroyed, severely damaged, moderately damaged, and slightly damaged – which was the way we graded them.

NIST also predicted the path of jet fuel distribution through the buildings, and looked at how workstations and partitions were damaged by the planes and their fuel paths.

Workstations and partitions were an important energy absorber of the airplane impact. Using the predicted path of debris, NIST was able to determine where fire proofing may have been dislodged.

Fireproofing was knocked off in the direct path of debris. Fire proofing also was knocked off just by the shear shock of the impact of the plane. The second building for example vibrated for 4 minutes after the impact, its roof top swayed 20 inches which was about 33% to 50% of the sway under hurricane wind design conditions.

As result of the above analysis we did not include fire proofing that was dislodged in our evaluation of the building performance during the fire.

NIST then did fire dynamics calculations through both buildings, carefully modeling the building contents and the dispersion of aviation fuel. Once we calculated the air temperatures on the floors as a function of time, we then could do a thermal analysis to determine the temperatures on the steel (with or without fireproofing).

Once we got the temperatures in the buildings we put them into a structural analysis program to analyze the progression of component level failures that took place over the 102/56 minutes after each airplane’s impact up to the point of collapse initiation.

PRIMARY FINDINGS:

There were striking similarities for two buildings. While Building #1 (the North Tower) was struck on north side and Building #2 (the South Tower) was struck from south, inward bowing of the exterior columns was observed on one of the faces of both buildings.

In the North Tower, the inward bowing was observed on the south face. The extent of the bowing was 40 to 50 inches and it spanned multiple floors and the entire width of the south wall.

There were major reasons for the observed bowing:

1. Columns on south face had heated up and had additional axial load.
2. Floors to which they were connected were pulling the column inward – since they were sagging under the heat to which they were subjected.

Then as the fires burned the building reached a point where it became unstable on south side. At that time the building tried to redistribute the load to the core through the hat truss at the top of the building. Since the core was already thermally weakened, the instability on the south face spread across the entire face and then to the adjacent east and west walls.

As a result the North Tower then tilted towards the south and started to collapse.

The differences in the collapse sequence between the North and South Towers were not very significant.

In the South Tower photos showed us that there was inward bowing on the east face (not north) of that building. The fires on east face affected the columns and the inward pull forces on the east face because the floors were sagging.

We found pictures showing that this bowing occurred within 18 minutes of impact on the second tower.

NIST found there were differences in the fires between Building #1 & #2. In building #2, the internal building content debris accumulated on the northeast corner on 80-81 floors and the fires tended to persist on the east face of the building from the beginning.

In the first building (the North Tower) the fires stared in north side of the building and took much longer (an hour or so) to move around to rest of the building to the south face. In general it took 20 minutes for that fire to burn through each section.

This explained why the second building collapsed first. In addition there was more severe damage to core columns on the southeast corner of the South building than was the case in the North Tower.

WHAT DO WE KNOW ABOUT THE FIRES?

NIST reviewed over 7,000 photographs and over 150 hours of videotape. Then we looked at the overall picture.

North Tower - There were fires in all 4 faces of first building on multiple floors. The worst fires were on the east and west faces. The building plan was about 208 feet on a side. The building’s footprint was square but the cores were rectangular. The north and south sides had long span floors, 60 feet long. The east and west sides had short spans of 35 feet. Since the north and south sides had longer span, greater floor sagging was possible on those sides to cause inward pull and bowing in the exterior columns.

Why did sagging not happen on the north side? Because a large number of columns were severed by the airplane impact and the floors were no longer connected to those columns.

South Tower - Here the south & north faces were short span and the east and west were the long span. Hence the building’s tilting was on the east side where the fires occurred and not on the west side where there were no fires observed.

CONCLUSIONS & IMPLICATIONS:

FIRST FINDING – Overall the collapse of the two towers occurred from a combination of factors: airplanes and their impact, fires and the effect of jet fuel, the core weakened, floors sagged, and exterior walls bowed inward.

No single factor triggered collapse. The collapse resulted from a combination of factors.

We found here that the collapse wasn’t governed by the light frame construction or the design of the core.

SECOND FINDING – If the fire proofing had not been dislodged by impact of planes then we could not predict the collapse of the building. Not in a conventional large building fire or in an unconventional jet fuel ignited fires.

Even if the planes hit the building and did all damage and initiated large fires as they did on September 11, 2001, the building would not collapse unless the fire proofing was dislodged. We don’t design buildings to withstand airplane impacts. It is more effective to keep terrorists away from airplanes and airplanes away from buildings.

ISSUES AND PENDING RECOMMENDATIONS:

First, these buildings are unique buildings and therefore some of the issues and recommendations are not universally applicable.

On the other hand, the procedures and practices used to design the buildings and carry out the evacuation and emergency response are based on commonly used procedures and practices and there are several lessons to be learned here, especially in the areas of structural integrity, fire resistance of structures, building evacuation, emergency response, other procedures and practices, and education and training.

Our investigation studied the procedures and practices used in the design, construction, operation, and maintenance of the WTC towers in great detail, and we have identified a number of issues that will form the basis for recommendations in the final report.

Here are a few of our observations upon which we will base our recommendations:

We looked at commonly used procedures and practices and compared them with the following codes: 1968 NYC, Chicago, BOCA model code, NY State, and the current NYC Building Code.

Our recommendations will be based on the above code comparisons, and our recommendations will place an emphasis on tall buildings.

Some of our recommendations will be for all tall buildings and some to select tall buildings that may be at higher risk of being a terrorism target – because of their iconic status, critical function, or design.

Most of our recommendations, however, will focus on all tall buildings

We also will address selected other types of buildings where issues that we identified, such as those related to fire protection/rating/test methods, are relevant.

For example, we have some important findings related to the fire rating of the floor system in the towers that were built as a composite system with concrete slab and steel floor trusses using open-web bar joist elements:

3. We found that ½ in. of fire proofing that was specified had a rating of ¾ hour as compared with a rating of 2 hours with the ¾ in. fireproofing thickness that was-installed. We found out that this large difference in rating was not due to the difference in the fireproof thickness on the truss but instead may have been due to differences in over-spray on the underside of metal or the moisture content of the concrete slab. In either case, it appears that the thickness of fire proofing on trusses didn’t make much difference.
   
4. All four tests, however, supported the maximum design loads for at least the 2 hour rating period without failure of the specimen.

EVAUCATION AND EMERGENCY RESPONSE COMMENTS:

In our work NIST conducted over 1000 interviews. We had 900 with occupants and 116 with emergency responders, including fire fighters, police, and Port Authority personnel. We also used formal procedures in our interview questions so that the findings are statistically valid and can be generalized.

There were 17,000 people in the two buildings that day (9/11/01).

The buildings were designed for 25,000 people each. Thus, on 9/11 at the time of the attack they were only 1/3 occupied.

6% of the people in those buildings that day had some kind of mobility impairment (heart condition, recovering from recent surgery or injury, pregnancy, asthma, etc.). 1,000 people is a large number.

We calculated that the rate of evacuation of Building #1 in the first 16 minutes was half that in Building #2. About 3,000 people self-evacuated the second building once the first building was hit and that was major factor in enhancing life safety in second building (especially for those on the upper floors).

During those minutes, all of the elevators in Building Two were functioning, so those on the upper floors were able to take advantage of that. After the planes hit each tower only one elevator in each building worked. The other 98 were non-functional.

Projections were that during the last 20 minutes before each building came down the rate of evacuation had reduced to 20 percent of what it was prior to that time. The stairwell capacity was sufficient for occupancy that day especially for those who were able bodied and could reach undamaged exits. Some were helping others out of the stairwells.

What would have happened, however, if each building had been fully occupied with 25,000 people that day and the buildings had to be evacuated?

We ran a few available egress models and found that it would have taken 4 hours to evacuate and some 14,000 people would have lost lives instead of 2,100. The egress capacity wouldn’t have been adequate for full building evacuation within the time that was available.

The evacuation rate on 9/11: People took 48 seconds per floor once they were in stair well, which is half the slowest speed reported in SFPE handbook for non-emergency evacuations. This was a surprising finding.

EMERGENCY RESPONSE:

There are several findings here:

Of first importance is communications within the building. We have a problem with radio communications in steel or reinforced concrete since they pose challenging radio frequency propagation environments.

Second is the scale of the event. Most emergency response takes place on a much smaller scale. On 9/11 we had about 1,000 emergency responders onsite using point-to-point communication. In point-to-point communications, there can be only one communication on a frequency at a given time. The events on 9/11 caused a surge in traffic. This led to doubling, where people were trying to talk simultaneously, which in turn led to incomplete messages. Many people possibly gave up using the system because of these difficulties and so important communications probably didn’t occur.

Third then is the need for interoperability between fire and police and other responders. (This includes those coming from other jurisdictions under mutual aid.)

SUMMARY:

We have covered a lot of ground in this call and we will have recommendations coming out in June covering such areas as: progressive collapse; better test methods for fire ratings; better fire proofing inspections; better building evacuation; enhanced emergency response; and new methods for fire resistance design of structures.

Question: In your report will you have code related recommendations?

Answer: Yes, it will but the most important thing to remember is that there must be the adoption and enforcement of codes by state and local governments. No matter what our recommendations are, they mean nothing unless they are adopted and enforced.

An example of a code-related recommendation may be in ASTM E-119. This may involve a significant rethinking of current test methods due to issues of scale and restraint and also due to the fact that there are no ratings for connections in buildings. Codes currently are silent here for connections between columns and beams. What should be the fireproofing at their connection?

Also need to look at issues like: if you have certain fireproofing thickness on beams and columns, would the same thickness get you the required rating for the connection?

To a degree some of our codes address these issues. For example, the Structural Frame concept is already in the ICC’s International Building Code. Under that concept if you have beams, joists, and trusses that directly connect to a column, then they should have the same fire rating as the column. At present, however, we are not certain as to what extent states and locals adopt and enforce that provision.

Evacuation provisions in codes are another area. Most stairwell capacity in buildings is based upon phased evacuation. That’s how we set our stair width. We evacuate occupants from the one or two affected floors, not the full building.

The issue is how to develop adequate capacity for full building evacuation (which may occur under any number of normal threats, not necessarily terrorist threats) and what performance criteria do you use in design for evacuation time? 1 hour? 2 hours? 5 hours? Should the evacuation time be threat specific? What design tools do we have to ensure stairwell capacity is adequate?

Our recommendations will not be prescriptive. We will make performance-based recommendations, and then we will work with ICC and NFPA (also ASTM, ASCE, and other standards development organizations) to develop an approach towards getting them implemented.

Threshold levels will be developed by the private sector. From our research it appears that there clearly needs to be a relationship between fire rating and evacuation times for a building.

Inspection of fireproofing during construction where spray-on fireproofing is applied is another code area. We need to have procedures that are robust and take into account the fact that other trades come in after fireproofing is installed, and their work may damage the already-installed fireproofing. There is a need to inspect the fireproofing in the building again after other trades complete their work.

Also there is the need for procedures for in-service performance expectations and inspections of the fireproofing during the life of the building.

Question: Will elevators play a more important role in the evacuation from the building?

Answer: Yes, I believe so. In our report we will talk about fire-protected elevators and structurally hardened elevators. We probably also will talk about the core of a building – elevator and stairwell shafts should have structural integrity. Now they are only required to have a fire rating.

Our current thought process is to call for a structural integrity requirement. What should it be for the core, which contains the stairwells, elevators, and utility shafts since the core is the last line of defense for life safety? What should be the structural loads? Accidental structural loads (e.g., gas explosions, debris impact from hurricanes, etc.) for most buildings, more stringent requirements (e.g., bombs) for buildings subject to terrorist risk?

We already see elevators being used in UK for firefighters (2 or 3 member units). An Asian model involves service elevators that are fire protected also may be looked at.

We possibly should consider such elevators, at a minimum, for fire fighter access and for evacuation of mobility impaired. Also for very tall buildings (say 70-80 stories and higher), such elevators also should be considered for occupants.

Again we are working towards having our recommendations released during the third week of June. These will be released as draft for public comment. There will be a 6-week public comment period from the date the reports are released on the NIST website.

Question: How do we access your report?

Robert Wible and Claude Cooper thanked Dr. Sunder for his time and his detailed presentation. Mr. Wible noted that NCSBCS will issue an "ALERT" to its members when the NIST reports are posted in June.

Mr. Wible thanked everyone for their participation and noted that June 16, 2005, has been tentatively set as the date for the next Important Issues Call. The topic will be "Safe Harbor – HUD Fair Housing Requirements and the International Building Code." Speakers for the call are being invited from both HUD and ICC. Additional details will be available in the June Members -Bulletin.

There being no further items for discussion, Claude Cooper, AMCBO Chairman, thanked everyone for being on the call. Mr. Wible indicate a summary of the call would be sent to everyone on the call for their review and comment prior to its being posted on the Members Only portion of the NCSBCS/AMCBO website.