Tornado-resistant construction,

Tornado safety,

And reconstruction after disasters




Chuck Doswell

Page created: 9 September 1999 last updated: 13 October 1999 ... updated the links to the BPAT final report.

This page contains my opinions and since it is on my home Website, no disclaimer is needed. If you want to discuss the material contained herein, e-mail me at


The events of 03 May 1999 and their aftermath have created some controversy regarding the construction, and the reconstruction of homes in the path of the tornadoes. This controversy invites some elaboration in view of the responses engendered from the contractors in the newspapers. It's difficult to carry on a meaningful dialog by going through the media.

During the review of the damage tracks by qualified engineering teams, including the Building Performance Assessment Team (BPAT) sent in by the Federal Emergency Management Agency (FEMA), a number of engineering observations were collected that confirm earlier findings as seen by, for example, the wind engineers at Texas Tech. University. In no particular order, these include such observations as:

  1. Building code violations of various sorts are often clearly seen in the parts of the path where damage has made them visible.
  2. Considerable damage is associated with either building code violations or construction practices that fall within existing codes but which create weak links along the "load path" connecting roofs to walls and walls to foundations.
  3. Considerable damage is caused by flying debris from nearby homes in situations where a home might otherwise have experienced only minor damage.
  4. Most of the homes exhibiting damage from tornadoes have not experienced windspeeds that would cause total destruction of the home. In other words, by far the majority of damage is not of the sort that leaves a home uninhabitable.


Issues about Construction

Building codes currently in effect over most of the Midwest are aimed at building homes to withstand winds of up to 90 mph. Beyond that, it becomes increasingly likely that some structural damage will occur. Clearly, winds up to 90 mph should cause at most only minor damage (shingle and siding damage, broken windows, etc.), provided the building codes are followed, so the structural integrity of the home is not be likely to be compromised.

By far the majority (around 60-70 percent) of tornadoes, while not ever to be taken lightly, are considered to be "weak" ... producing winds on the order of 100 mph and no more. Of those that exceed the "weak" rating, tornadoes might attain "strong" intensity (around 30-35 percent of all tornadoes), producing winds capable of some structural damage, up to and including failure of the roof and most of the exterior walls, but typically leaving interior walls still standing. Only a few percent of all tornadoes reach the "violent" intensity, capable of removing all the above-ground structures of a typical, well-constructed (code compliant) frame home.

Further, within any given tornado's total path, only a small percentage (about 10 percent) of that path experiences the strongest winds produced by that storm. Tornadoes are characterized by a column of violently rotating air, which may or may not exhibit a visible funnel cloud ... see my discussion about what is a tornado ... damaging winds extend for some distance around that core of rotating winds, so even if your home is not hit directly by the core of the tornado, it might be affected by the winds that extend outward from that core. With distance from that core, the windspeeds tend to decrease, so as you get farther from that core containing the highest winds, the winds become correspondingly less damaging. On the very margins of the damaging winds that surround a tornado's core, the damage becomes spotty and pretty minimal. This is discussed and illustrated in FEMA's BPAT report about the 3 May 1999 tornadoes. In strong and violent tornadoes, this means that as much as 90 percent of the affected area experiences winds on the order of 100 mph or less. In weak tornadoes, virtually all of the area sees such windspeeds or less.

What does this all mean? First of all, it means that a considerable amount of damage caused by tornadoes is preventable!! No one is suggesting that it is practical or even worth doing to try to make a house tornado-proof, especially if you're unlucky enough to experience the strongest winds of a strong or violent tornado. Rather, the suggestion is that if your home is affected by a tornado, the odds are that the winds you'll experience will be less than totally devastating and that it is possible to prevent structural damage if the tornado is weak or you're on the edges of a strong-to-violent tornado; you still might lose roof shingles, or siding, and certainly some windows, but the house can came through more or less intact.

Second, it's important to understand that if you build your house to prevent structural damage in weak tornadoes (or on the periphery of more intense tornadoes), the cost of the necessary modifications isn't all that high. Adding the clips, anchor bolts, and straps necessary to tie together the roof, the walls, and the foundation adds about 2-4 percent to the overall cost of the house. When this is spread out over the duration of a 30-year mortgage, this is not a big change to the monthly payment. If contractors would offer this enhancement option when a home is built, then the prospective homeowners could choose for themselves whether or not to make this additional investment.

Third, if you choose this option ... to build or retrofit your home to exceed the standard requirements of existing building codes ... you still might suffer structural damage if your neighbors don't also improve the construction of their homes. Considerable damage in tornadoes is done by flying debris; if your neighbor's roof takes off in a 90 mph wind and crashes into your home, it will almost certainly do structural damage, even if your home is capable of resisting the wind! Thus, you increase your odds of coming through a nearby tornado event if your whole community builds to a higher standard than is imposed on construction as of this writing. The idea is to follow FEMA's advice to develop disaster-resistant communities. Nothing can make a community disaster-proof, but it is quite feasible to reduce the damage caused by natural disasters like tornadoes. Every bit of damage we can prevent represents a savings to ourselves and our communities.

Of course, the odds are that you will never experience a tornado, even if you live in the heart of "Tornado Alley," especially a violent tornado, like that in the southern and eastern parts of the Oklahoma City metroplex on 3 May 1999. It's a gamble where the odds are in your favor and when considering costs versus the benefits, many people might not choose to spend the extra money unless their builders are required to by a change to the building code. In my opinion, this is something that communities should consider and if they choose collectively not to make this investment, then perhaps the insurance companies might choose to raise the rates they charge for homeowner's insurance, in comparison to communities that choose to improve their resistance to natural disasters like tornadoes. Conversely, it should be possible for insurance companies to reduce the premiums for those who choose to invest in the structural improvements, with even lower rates being made available when whole communities participate.

Under the current system, with building codes that are, in my opinion, inadequate for those who live in tornado-prone areas of the nation, we are all being asked to pay higher insurance premiums and taxes to support the unlucky few who get hit. Given that most people have no idea that their homes will experience unnecessary damage in the unlikely event that they experience a tornado, it's not worthwhile blaming the homeowners for the problem! Should we blame the homebuilders? I'll have more to say on this, later. If after a tornado, reconstructed homes are built to the same or perhaps even lower standards, then all that does is recreate the same level of risk of structural damage, or even increases it. We need to come up with a fair way to fix this dilemma, rather than seeking to establish blame.

Another related issue is code compliance. It's a demonstrable fact that code violations, even under existing codes, are not all that rare. It's obvious that the contractors have a financial incentive to build homes quickly. Doing every structural connection to an exacting set of requirements takes time and care, which limits the number of homes that can be built. The pressures to be profitable are likely to make code violations inevitable, in spite of self-serving denials by the construction industry. Home construction is a competitive business, where there is considerable pressure to avoid passing on costs to the consumer. The structural details that are of concern are not visible to the homeowner and so the homeowner (who may well be more concerned about more visible construction issues, like the kitchen cabinets or the bathroom fixtures) is trusting that the "team" of the contractor and the building inspector will be sufficient to ensure that the home complies with the codes. When that trust is violated, the home becomes a sort of "time bomb," ready to fail in conditions that should not produce structural damage.

Most homeowners haven't a clue how to identify structural details that would suggest a failure of code compliance. Moreover, many structural details are sealed up behind plasterboard or paneling, and hard to review after the home is built, even if the homeowner becomes aware of the things for which to look. In order to be absolutely certain that every structural detail has been completed in compliance with any standard, a homeowner would have to be knowledgeable about construction and be there to watch every bit of the construction. This is an unlikely situation and so most homeowners simply must trust their builder. Regrettably, there simply are not enough inspectors to be certain of every detail! The result of all of this? Code violations! Any study of damage from storms across the United States makes it clear that these violations are occurring and this fact simply cannot be denied.

To take the word of a contractor on this issue is quite comparable to taking the word of the executives of a tobacco company in response to charges of health hazards associated with smoking. When someone has a financial stake in the outcome of an investigation, it's quite obvious that they are going to deny that any wrongdoing ever occurs. Would anyone expect a contractor to admit, "Sure, the homes I build are full of code violations!" Rather, it is quite predictable that they'd claim that code violations simply don't occur. But is it plausible to believe them? Only if their statements can be validated by someone doing an independent investigation, with no financial incentives flowing from the findings! The findings of engineering teams when disasters strike show without question that code violations do occur.

In spite of appearances, however, it serves no useful purpose to cast the contractors and building inspectors as villains in this situation. The homebuilders are simply responding to market pressures in ways that allow them to stay afloat. The inspectors are likely to be doing the best they can under circumstances of limited resources. There may well be "bad apples" among the homebuilders and inspectors who are bilking the customers and taking kickbacks in return for overlooking code violations, but these surely are a minority that most responsible homebuilders and inspectors would be glad to see drummed out of the system. In fact, I am not interested in finding any villains in this exercise, although they may well be present. I'm trying to consider solutions to what is, in fact, a serious problem for the tornado- and hurricane-prone parts of the United States.

It's my belief that there can be no solutions if we maintain the status quo and deny that problems exist, but I certainly have no wish to engage in a witch hunt, seeking scapegoats, or finger-pointing. Americans seem to be prone to seeking to establish blame (perhaps in hopes of making a killing via litigation) and appear not to be interested in finding workable solutions that allow everyone to live and let live. It is natural to respond defensively when it appears to you that someone is trying to blame you for something, and it is common to misrepresent the position of the "other side" in such a polarized dialog about the issues. I really want to avoid this, so I want to make it clear that although I have just associated the homebuilders and inspectors with tobacco company executives, the purpose of this comparison is only to illustrate the point that it would be unrealistic to expect them to admit the slightest possibility of wrongdoing, even though it certainly exists. I am by no means equating homebuilders and code inspectors to tobacco company executives!


Issues about tornado safety

I'm not inclined to go through a detailed discussion of this topic here, since I've done so at some length elsewhere. However, in the context of construction, there are some clear findings from the engineering studies of tornado (and hurricane) damage.

  1. Since the vast majority of tornadoes are observed to be of the weak (F0 and F1 on the so-called Fujita Scale) intensity, most folks can avoid serious injury and death in most tornadoes simply by following the standard rules for tornado safety.
  2. When tornado intensity increases, and structural damage approaches the high end of the strong intensity category (i.e., the upper limits of F3 damage, where only core remnants of homes are left standing) and move into the violent intensity category (F4-F5), it becomes increasingly likely that serious injury or death can occur when the occupants of a building have no underground shelter or an approved above-ground "safe room" in which to take shelter.
  3. Since it's impossible to look at a tornado and be certain of its intensity, it's necessary for citizens to assume the worst about any approaching tornado and take whatever precautions are recommended and available. The widespread lack of underground shelters and the relative rarity of approved above-ground safe rooms puts everyone lacking such shelters in the tornado-prone part of the United States (i.e., where strong-to-violent tornadoes are frequent enough to be a credible threat ... roughly everywhere east of the Continental Divide) at risk. Thus, if someone has to choose between investing in construction enhancements versus investing in a proper shelter (i.e., an underground shelter or an approved above-ground "safe room"), the clear choice is to invest in a proper shelter first!
  4. It is always possible to be lucky, and the lack of a proper shelter (as defined above) is not a death sentence, but it seems clear that it's probably a poor strategy to trust in blind luck. Many people in the 3 May 1999 storms survived in spite of having their homes swept completely away. On the other hand, a substantial fraction of the residents in Jarrel, Texas were killed in a comparable event in 1997 when their homes were completely destroyed and swept off the foundation slabs. Luck is a factor, of course, but not one that should be relied upon. Even if you never experience a devastating tornado, surely the peace of mind associated with knowing you and your family have a proper shelter is worth something to a homeowner.
  5. It's also obvious that manufactured homes are generally less secure from tornado damage than site-built homes, properly constructed. There is some limited evidence to suggest that a new manufactured home, securely fastened to a permanent foundation, is not dissimilar from a "stick-built" frame home ... but any manufactured home less well-constructed and secured to its foundation than this relatively high standard is simply not a secure place to ride out a tornado. For residents of manufactured homes, the ordinary rules about seeking shelter in bathtubs or closets are simply futile gestures, that don't add significantly to occupant safety. Access to a proper shelter is basically the only option that makes sense for residents of mobile homes, and such a shelter should be easily reached and entered from anywhere in a mobile home park (not more than about 100 yards from any resident).

There is considerable evidence that, apart from residences, many businesses and public facilities have less than adequate plans for their occupants should a tornado threaten them. This is not the venue for a complete discussion of this, but citizens should demand that their workplaces and public facilities (including schools, nursing homes, hospitals, sporting and recreational facilities, etc.) have worked out plans and made them known to their occupants in the event of tornadoes (or other natural disasters). This is part of making communities disaster-resistant.

It is well-known that most casualties in tornadoes are the result of flying debris. It's not the wind that kills and injures ... it's what's in the wind! Anything we can do to reduce damage from tornadoes cuts down on flying debris, so a side benefit is that disaster-resistant communities should suffer fewer casualties, as well as less property damage, in comparison to those communities that do nothing to enhance the structural integrity of their buildings.


Thoughts about the reconstruction

I've characterized the reaction of the construction industry and building inspectors as being one of public denial of the existence of a problem. This is certainly validated by appearances in some of my recent experience, but it does not mean that homebuilders and their inspectors are entirely venal and without concerns for their customers. What I believe this reaction (to revelations about the very real code violations) to be is a natural defensive response to what they probably perceive as an unfair representation of their industry. I have no evidence to suggest that the industry is thoroughly riddled with (a) violators and (b) those who permit violations to go unchallenged, and I don't believe that anyone responsible is making such a charge. I have no idea what the percentage of violators might be ... I only know that shoddy construction practices are found in every engineering survey conducted after natural disasters.

Hence, it is equally unproductive for the industry to pretend that no code violations occur or to characterize anyone suggesting enhancements to building construction as a starry-eyed idealist who has no comprehension of the realities of the construction industry. The "us versus them" mentality is a clear barrier to working out solutions.

I perceive that :

What I am really hoping we can accomplish with a discussion of these issues is that as we go through the reconstruction after natural disasters like the 3 May 1999 tornadoes, we can accomplish:

  1. The creation of an option offered by homebuilders to include tornado-resisting structural enhancements to their customers. Presumably, the added cost would be passed on to those customers who choose this option.
  2. The growth of understanding within the public as a whole that tornado damage can be mitigated and that the resulting structures need not be tornado-proof fortresses that cost a fortune to build.
  3. The proliferation of proper shelters within new construction in tornado-prone regions of the United States and, to whatever extent is feasible, the retrofitting of proper shelters.
  4. The success of the FEMA program to encourage disaster-resistant communities, such that many communities understand the value of encouraging better construction throughout their communities, and take actions to make those good things happen.
  5. To whatever extent is possible, government at all levels should support programs encouraging public and private cooperation in the development of resistance to natural disasters of all sorts. This might well include financial assistance or tax incentives to residents and businesses that make an effort to mitigate the hazards of natural disasters in approved ways. It should also include support for building code enforcement and a general review of codes and standards in light of post-disaster assessments.
  6. A commitment by the insurance industry to provide premium reductions to citizens that would encourage them to invest in enhanced construction and the inclusion of proper shelters in their homes, at least within the tornado-prone areas of the United States

Basically, I want to seek solutions that do not have as their core an effort

  1. to establish blame or cast some limited group as the scapegoats,
  2. to force one sector or individual to bear the entire burden of financial responsibility,
  3. to maintain the status quo.

It's my belief that a mutual accommodation of the relevant business and governmental participants in community safety can be found that needn't amount to a "free lunch" for someone and a "bloodbath" for someone else. Most of the participants in this discussion are well-intentioned and only reacting defensively to what they perceive as threats from "outside" of themselves ... if we can avoid creating defensive reactions, perhaps we can have the needed discussions and arrive at satisfactory answers to the problems as they now exist.

No solution has to be perfect before it can even be considered, and all sides have to be willing to listen and compromise where they can. If we begin in a spirit of accommodation, then perhaps we will have built a foundation from which our problems can be analyzed and solved from all relevant perspectives. No one need be a "loser" in such arrangements, except perhaps the few, isolated individuals who are acting purely selfishly and without regard for the good of the communities as a whole. As noted in my essay on the events of 3 May 1999, I believe we owe the victims of these disasters a good faith effort to fix the problems that currently exist within the tornado-prone communities of the United States. I am confident it can be done and am only uncertain of whether or not it will be done.