— Special Report —
First in a Series

Is Joplin the Next Andrew?
The Devastation Caused by Hurricane Andrew Led to Landmark Code Changes. Will the Recent Tornadoes Lead to More?
by Tara Taffera

Eight hundred tornadoes are reported in the United States in an average year resulting in 80 deaths and more than 1,500 injuries, according to the National Weather Service (NWS). Yet as of June 2011, this tornado season has proven to be anything but average. Among the deadliest U.S. tornados, the “Dixie Outbreak” of April 27, 2011, set a modern mortality record with more than 350 people killed, according to preliminary numbers released by the NWS. That’s four times the average in just one day, and includes the fatalities in Tuscaloosa, Ala. Add to that the EF5 tornado that killed more than 150 people in Joplin, Mo., on May 22, and you have a devastating loss of life and infrastructure.

Dr. David O. Prevatt, PE, assistant professor, department of civil and coastal engineering for the University of Florida, is the principal investigator of a research data collection project funded by the National Science Foundation and the American Society of Civil Engineers to document damage to residential buildings following the tornado that devastated Tuscaloosa and Joplin. Prevatt points out that you never know where a tornado may strike next.

“If these storms passed five miles south of their actual tracks the damage may have been far less than what occurred. But between the two storms you have 11,000 homes damaged or completely destroyed,” he says.
Prevatt’s focus is to minimize that damage.

“If we can improve structural performance and reduce the amount of houses lost from 11,000 to 5,000 then that is an enormous improvement,” he says. “We can improve through use of better engineering details.”

Prevatt’s research focuses on the mitigation of hurricane damage to low-rise construction and the performance of existing residential structures. He is examining whether building practices, such as those adopted in Florida to prevent hurricane damage, could help with protecting structures against tornadoes.

“In this new direction I am asking myself if this [damage] is acceptable community-wide,” says Prevatt. “My personal answer is no but the difficulty is trying to find what to do and what else can be done.”

Prevatt and his team are researching these two EF4 and EF5 storms to study the construction techniques used for those buildings significantly damaged. “For the most part I have not seen buildings that were built to current building code—most were older structures,” he says.

Prevatt traveled to Tuscaloosa with a specific hypothesis in mind: that certain structural retrofits and details that work with hurricane coastlines will work in tornado alley (see box below). “To some extent there must be some benefit we can get from impact glazing—particularly from smaller tornadoes,” he says.

But what will encourage everyone—from homeowners to code officials to product manufacturers—to move toward the use of impact-resistant products in these tornado-prone regions? Or are tougher codes even the answer?

What’s the Baseline?
Before we can attempt to answer these questions, it is important to look at existing standards.

The Federal Emergency Management Agency (FEMA) has addressed the design and construction of safe rooms since 1997. The purpose of a safe room is to provide a space that offers a high level of protection during an extreme weather event such as a tornado. While these rooms may not typically include windows, they can. FEMA 320, Taking Shelter From the Storm: Building a Safe Room Inside Your House, is applicable to the construction of residential safe rooms. FEMA 361 Design & Construction for Community Safe Rooms is applicable to commercial safe rooms and includes design criteria for community safe rooms. These standards offer information about the types of permissible glazed openings in these spaces.

Since 2008, ICC/NSSA 500 Standard for the Design and Construction of Storm Shelters has been available for adoption and use by any jurisdiction. The standard applies to the design, construction, installation and inspection of storm shelters in hurricane- or tornado-prone areas. In 2009, the International Building Code included the ICC/NSSA 500 in the model building code.

But what if homeowners want impact-resistant glazing in their homes, and opt against a safe room? The ICC 500 is the standard for design and construction of storm shelters but it really doesn’t address fenestration products, points out Ken Brenden, technical services manager for the American Architectural Manufacturers Association (AAMA).

To fill that void, AAMA has released a voluntary specification for testing and rating fenestration products that will be exposed to tornadoes and similar extreme wind and rain conditions in April 2011. It worked with the Building Enclosure Technology and Environment Council (BETEC) and the National Institute of Building Sciences (NIBS) to gain input on the document.

“There are a number of test methods that exist and many were developed for hurricane products,” says Brenden. “Tornadoes are on the same level, if not more, than what is expected during a hurricane—but tornadoes are much more extreme wind-wise, etc. So through expertise of our members, we took those things under consideration in development of this document.”

AAMA 512-11, Voluntary Specifications for Tornado Hazard Mitigating Fenestration Products, uses existing test methods and other procedures to qualify windows and other glazed fenestration products for tornado hazard mitigation. The newly released document provides a system for rating the ability of windows to withstand impact, pressure cycling and water penetration, which are generally associated with tornado conditions.

“When it came to fenestration products used in tornado structures there was a void that existed in the architectural community,” says Brenden. “Architects had no document to go to if they wanted a product that exceeded code.”

The specification outlines the different levels of protection that apply to different buildings such as, but not limited to, hospital emergency rooms, community shelters and police/fire headquarters. These levels of protection are specified based on requirements of the authority having jurisdiction, and each level corresponds to different testing requirements. The level of testing required for each of these types of facilities also depends on the FEMA performance zone in which the building is located, as the weather conditions and likelihood of a tornado varies depending on the part of the country. The testing and performance requirements section of the specification outlines the necessary test conditions for adhering to AAMA 512. According to the specification, test specimens should be tested for anchorage, missile impact, water testing and cycling for windows used in hurricane-prone zones only.

Windows will require a minimum AW60 rating in accordance with AAMA/WDMA/CSA 101/I.S.2/A440 to meet the requirements of AAMA 512.

“Most single-family dwellings do not entertain an AW60 product in their homes,” says Brenden.

He clarifies, however, that the spec makes no claim for a specific occupancy.

“It could be applicable to residential or commercial,” says Brenden. “The current North America Fenestration Standard rates AW as the highest rated performance class.”

Code and Cost Concerns
While architects and others may not entertain the idea of upgrading to an impact-resistant product currently, the industry is divided as to whether the recent storms will serve as a catalyst for change.

“Most window companies will use the building code as their guide or at least as the minimum requirements,” says Brenden.

Time will tell if the building codes will move toward higher standards.

“You are seeing some positive movements, some voluntary specs, [and] some initial steps that can lead to code changes, but I
don’t know the amount of hold it will take,” says Frank Lukens, president of WinDoor Inc.

But Prevatt says that after surveying the damage following the recent tornadoes, the answer is simple.

“Making the building codes stronger in itself is a large step but it only addresses vulnerability of new buildings going forward. The more difficult challenge is to determine what should be done to reduce the risk of damage to existing buildings,” he says.

Prevatt believes the codes will change because of a unique factor in the Tuscaloosa and Joplin tornadoes.

“These hit densely populated areas in the middle of the cities,” says Prevatt. “These are towns of about 100,000 or more. It literally cut right through the center. This is what we have to be prepared for—a low-risk event but with very severe consequences and this is a major concern.”

That “low-risk” factor is what make some say the codes won’t change—and shouldn’t.

“One of the reasons why the codes don’t require impact-resistant windows outside of hurricane-prone areas is because the probability that a home will face a storm of this severity is very low compared with the probability a home in certain hurricane-prone regions will experience a hurricane,” says Jeff Inks, vice president, codes and regulatory affairs, for the Window and Door Manufacturers Association.

No matter who you talk to, people are going to make comparisons between tornadoes and hurricanes. And when it comes to hurricanes, most will mention Hurricane Andrew, as it was the impetus for major changes in Florida building codes. While there are similarities between these two weather events, Inks points out there are some pretty significant differences to note.

“In areas like Miami, the likelihood that a home will face several hurricanes is significant,” says Inks. “That is not true with tornadoes for homes located in regions where tornadoes can occur. Tornadoes do not fall into the same repetitive and more predictable patterns that hurricanes do.”

Part of the other requirement related to hurricane resistance is the length of the exposure, adds Inks.

“A tornado is a much more spontaneous and shorter event than a hurricane, often coming and going in a matter of minutes. Hurricanes on the other hand can expose windows to hours of high wind and flying debris in a single event,” he says.

However, Dr. Tim Reinhold, senior vice president of research and chief engineer at the Insurance Institute for Business and Home Safety (IBHS), points out that the low probability of an area being hit by a tornado more than once also will play a role in future building code decisions and that regions that want higher standards will do so on a voluntary basis.

“Deaths most frequently occur in EF3, EF4 and EF5 Tornadoes. These intense storms represent less than six percent of the storms that hit the United States and EF4 and EF5 tornadoes represent less than one percent of U.S. tornadoes. The chances that you are going to get hit by any tornado is less than about one in 5,000 per year. Normally we design residences for a one in 500 to one in 700 event and critical facilities at the one in 1,500 to 2,000 probability level— mandating the design of residences for tornadoes would be a huge change,” he says.

Valerie Block, senior marketing specialist, DuPont Glass Laminating Solutions, suggests homeowners may start to look more at impact-resistant windows for tornado-prone regions, but admits that it will take a while for codes to change.

“From the residential viewpoint, we probably aren’t going to see building codes require this in the near future but you may see people opting for better protection based on their personal concern for safety,” says Block. “Change will come faster if building code requirements for mandatory impact protection are adopted at the state or local code enforcement level.”

Prevatt hopes that new research being initiated now, will result in changes to ASCE-7 by the 2016 version of the codes.

Still, some caution against making code changes based on a few select weather events.

“The adequacy of building codes is often questioned after natural hazard events like this,” says Inks. “But when evaluating damage there are many factors that must be considered including what code requirements the home was built to to begin with, whether those requirements were met and whether installations of components like windows and roofs were done properly. That information is critical to determining whether the current building codes are adequate.”

Lukens says that if you look at when the Florida codes started to change, the demand from homeowners was small until the local codes mandated impact-resistant products. Reinhold says that for the codes to mandate impact-resistant products in tornado areas, the areas affected would have to be expansive as tornadoes “could” occur in a variety of states.

“The probability is one in 5,000 to 10,000 if you are not in tornado alley,” says Reinhold. “So for someone to mandate specific requirements you are going to jack up the price of a house by a large percent, and you are designing for an event that is way out on the probability scale. The homebuilders would fight that tooth and nail.”

“Impact-resistant windows are part of a whole structural system designed to protect the home against hurricanes,” says Inks. “If they are to be installed or required for homes outside of hurricane regions, than all of the other structural requirements for homes in hurricane prone regions would also need to be met in order to provide the same level of protection.

“The increased cost for all of those requirements is significant and because winds in stronger tornadoes can exceed those of hurricanes, they may be of limited benefit,” he adds. “Imposing them on all homes constructed in the vast areas of the United States where tornadoes can occur is not considered necessary or practical.”

When it comes to the cost argument, Block also points to what happened after Hurricane Andrew.

“People talked about the expense of impact-resistant products 20 years ago in Florida,” she says. “To some extent that is true—until the building codes required it people didn’t consider the need as seriously as they do today. But now we have 20 years of missile impact solutions that are an option for builders and homeowners.”

“You can’t pick and choose certain building components,” cautions Inks. “It is so much greater cost. If a builder were to offer to build a home in Joplin in compliance with Miami Dade and showed them the price I bet most homeowners would opt out of that. You can’t make claims as to what this home will protect against.”

Block believes, however, that homeowners who want added protection will be more vocal than perhaps homeowners were in Florida.

“I think you are going to see more people raising their hands and asking, ‘What can I do?’” says Block. “That’s a little different than what we saw 20 years ago from homeowners who did not understand their options for impact protection.”

She adds that the industry offers many more impact-resistant door and window solutions than were in place 20 years ago and that today’s products deliver better performance.

“There is not a lot of product development at this point for just tornadoes and I think that will start to change,” says Block. “It will be led by people who have concerns for safety. We are at the beginning of this conversation, and I believe companies will develop tornado products. They realize this is a new opportunity for them.”

But aside from tornado-specific products, Lukens says the same formulas manufacturers use to build doors and windows for a bomb-blast setting can play a role in a tornado impact situation.

“We started doing research into how we test for Miami-Dade and found that the DP ratings that are required for testing can far exceed the wind speed for an EF5 tornado, which has wind speeds in excess of 200 mph and we have products that can handle DPs up to 200 mph,” he says. “So if we look at the DP requirements for windows and doors in hurricane zones, that exceeds the wind speeds for an EF5.”

And no matter what happens with the codes, there may be an interest on the part of some companies to build units to higher voluntary standards.

“We’ve been working on this AAMA standard for a few years as our members determined this was a gap in the market,” says Brenden. “If manufacturers will test to these standards it is up to them on how they market these products. I would expect that, at least initially, you will see commercial companies adding this to their offering. Whether it gets to residential or not I think that is driven by code.”

Whole-House Approach
While there is a divide as to whether fenestration codes should change, and whether consumers will pay a premium for impact-resistant windows, most agree that installing impact-resistant windows is not enough.

“There needs to be a holistic approach,” says Brenden. “It doesn’t pay to have a window that will withstand the elements when the wall itself can’t withstand the extreme storm. The overall safety and protection of the occupants is key.”

“The biggest problem I see is the entire building envelope,” adds Lukens. “There are products out there today that will exceed the impact requirements and DP ratings but when it is attached to the building in which it is designed it does little or no good. Those are the hard parts and the model used for building envelope construction in hurricane areas should be the same model used in tornado alley. The walls need to be designed to meet the pressure requirements.”

Reinhold adds that in lesser tornadoes such as 1, 2 or 3, impact-resistant doors and windows can provide more protection but it “shouldn’t be your only form of protection.” “It puts a stronger surface between you and outside world. If people are willing to spend money, and some will, then these [doors and windows] will provide additional protection,” he says.

Lukens agrees, saying impact-resistant products can go a long way in offering protection from flying debris in an EF1 or EF2.

“If you can minimize the damage in smaller-scale storms, then you are getting off to a very good start. You can take minimal impacts on frame walls and the product won’t disassemble,” he says.

“Even if the optimal home is designed, if the windows don’t hold up that system can fail,” adds Forrest Masters, assistant professor, Department of Civil and Coastal Engineering, University of Florida. “If the windows break you will have a dominate breech in the building envelope. As a result you will have extreme pressures in the building.”

He adds that it is crucial that the different trades work together.

“The fenestration industry has done a good job at innovating solutions for storm resistance,” says Masters. “The next frontier is to look at tornadoes, but it will take a group effort. It will take all the different trades to work together. In an event like this, if one system doesn’t work it doesn’t matter. The system is only as good as the weakest link.”

Prevatt discovered this first-hand when he visited the tornado-devastated areas.

“Sadly this is not new knowledge,” he says. “This has been said by me and other engineers in many previous past-tornado surveys. The issue of why structural systems fail where they do is the same. This has developed over time through decisions to build non-engineered homes in a particular way that lack an integral vertical load path.”

Still, Masters and Prevatt say that many in the building industry believe that buildings simply can’t survive an EF4 or EF5 tornado—no matter to what standards they are built.

“Someone has to shake it up and see what buildings can withstand. If someone comes up with a structural design that can withstand extreme limits, then it is incumbent upon the industry to withstand the loads and the debris,” says Masters.

Many questions about future construction methods remain. When rebuilding occurs, will building professionals rebuild differently? Prevatt will do his part to see that they do.

“I think the issue has to be driven home by people who know the loads and structural behavior of houses, like myself and other engineers, working with builders and code officials,” he says. “Previously there has been the sense that the building codes are responsible primarily for loss of life and life safety. As a result we have been focusing on storm shelters as the major design element to protect from a tornado. But after witnessing the personal tragedies unfolded when a family has lost everything they possess, there is a compelling argument that we need to also consider building homes somewhat more robust—particularly to protect life and reduce property damage from smaller (less intense) tornadoes.”

He has seen some positive signs of people willing to listen.

“It is already happening,” says Prevatt. “I have had positive feedback from structural engineers and building inspectors in Joplin. They are receptive to some change and waiting for the opportunity to get in front of legislators.”

But continued education of building professionals is crucial.

“I would like to work with organizations to spread the word on how to build better by developing research knowledge with practical applications. Further, I would like to work with high school students from areas affected by these tornadoes, from whom we can find the next generation of engineers working to solve the issues and develop tornado-resistant housing. Let’s see what we can do in terms of learning and then improving designs,” says Prevatt.

"One of the reasons why the codes don’t require impact-resistant windows outside of hurricane-prone areas is because the probability that a home will face a storm of this severity is very low compared with the probability a home in certain hurricane-prone regions will experience a hurricane.”
—Jeff Inks, WDMA

Future Research
Until then, Prevatt, Masters and other researchers, including teams at the University of Iowa and Texas Tech University, will continue looking into these storms for items that can give insights into the building industry, including door and window manufacturers, when designing products to withstand extreme weather events.

Until now research by Prevatt and his colleagues at the University of Florida has focused mainly on hurricanes but that is now changing. And the hurricane research will go a long way in helping researchers as they study tornadoes more closely.

“Given what we have learned through the past 20 to 30 years of hurricane research we can apply some of those research results to structures in tornado-prone regions,” says Prevatt. “The question is how far can we go? It really depends on the community’s tolerance for risk and willingness to pay for a safer community.”

Prevatt’s tornado research will be a major focus, and his first report from Tuscaloosa and Joplin was due to be published in August at press time.

Many agree that the tornado conversation is just beginning. The industry needs to start thinking about increased protection, says Masters.

“We are really looking five years out and ten years out and we need to get the door and window industry to start thinking this way and coming up with cost-effective solutions,” says Masters. “The research community is working on developing the fundamental science and hopefully the fenestration industry is thinking about this and there will be an intersection between science and technology.”

For more on this investigative report about tornadoes as it relates to the door and window industry, see upcoming issues of DWM/Shelter magazine in 2011.

Tara Taffera is the editor/publisher of DWM/Shelter magazine.

DWM
© Copyright 2011 Key Communications Inc. All rights reserved.
No reproduction of any type without expressed written permission.