Integrated Design | Integrated design examples | Ecological design photos | Consulting | Ecological design Q&AFree Newsletter

Optimal, Integrated Design for Fire Safety:
Fire-Resistant Construction, Fire Bunkers, Water Systems, Etc.

Shifts in fire strategy

Prepare, Stay and Defend vs Ready, Set Go

Terrifying, educational first hand accounts

Fire bunker research

Fire-armored water supply

etc. etc.


Summary: Articles and information about fire prevention and preparedness, including extensive, raw research notes on fire bunkers.


On this page:



"Our current codes are as effective at preventing the best practices
as they are for precluding the worst ones."

—David Eisenberg


Integrated Design for Fire

Our work is to find areas where existing systems seem insufficient, research and develop new systems, then publish them on our web site, books and in presentations.

At any given point we have many lines of research going, some short term, some long term. Some are slam dunks, some are long shots. Sometimes we start pulling on a seemingly small thread, and and it just keeps going and going, branching into all sorts of other things we are already working on that we thought were unrelated, pulling us into new areas of research, upending received wisdom...

Fire safety turns out to be an example of the latter.

We're about 15 years down a dual track of 1) securing our own home against wildfire, and 2) doing research and development for a possible future book about integrated design for fire safety.

We've had innumerable conversations with fire marshals, building officials, planners, builders, collected first hand accounts, inspected various fire sites, delved into the politics and economics of building standards for roads, water systems, and flame retardants; explored ASTM standards for testing building assemblies and windows, researched fire bunkers, designed fireproof water systems for wild land properties, and participated in the stakeholder code development standard for California's Wildland Urban Interface Building Standard. This research and our conversations with Don Oaks, former Fire Marshall of SB county, led us to focus on architecture and construction as a main, under-appreciated factor in residential fire safety. The consequent rise in the building standards for our own house left the WUIBS in the dust 15 years ago and have led us to rip out work we did not that long ago and redo it. After 15 years, just this last month (July 2013) for the first time we felt that our house was ready: to passively resist a fuel-driven fire, with windows and doors shut, and, if actively defended, resist a full blown wind driven fire.

A portion of our research notes are reproduced here. On materials testing, for example:

There's not much published on the fire resistance of public domain materials such as adobe, and we heard that some fire ratings for manufactured products are influenced by industry, and not necessarily a reliable guide. We double checked with some tests of our own, on various building materials such as raw wood, painted wood, earth plastered wood, hardipanel, cement stucco, several kinds of insulation, a wide range of adobe (sand-clay-straw) mixes, cedar shingles, asphalt shingles, and metal roofing, as well as checking how a few dozen common native and landscaping plants burn.

We're we surprised by the results? That's an understatement…suggest you check out this video of "class a" rated asphalt shingles burning like napalm which we made for our presentation at the Chief Building Officials of California Conference (text in this version messed up but you can get the idea):


We couldn't get the asphalt roofing off our house fast enough after seeing how it burns with our own eyes. I felt betrayed by my reliance on this misleading rating system. Class A fire rating? More like class action lawsuit. If you think about it for two seconds, even without having to see the stuff burning vigorously, flowing like lava, and making drips of flaming napalm, you realize that a legit rating system would not give same rating to asphalt as metal or 6" of concrete. Evidently manufacturers of asphalt shingles have influenced the standards so that consumers have a false sense of safety about their product. (This is certainly the case with harmful flame retardant standards. Before we leave the topic of roofing, here's another test, of metal roofing vs asphalt shingles flame resistance).




Prepare, Stay and Defend vs Ready, Set Go-It Depends

As with the rest of Integrated Design, the key is to do what makes sense in the context.

Wildfires are all different, and local conditions within the same wildfire can differ radically over inches, or in seconds.

A hugely complicating factor for fire safety design, vs, say, integrated design for building a house, is that since it is harder to think clearly about nuances of context in the midst of a fire. There is a legitimate advantage to simplifying, to having a plan and sticking to it.

As the debate over "Prepare, Stay and Defend" vs "Ready, Set, Go" rages in Southern California, I can see that if I were responsible for tens of millions of people's fire safety in Southern California, and I could only have one policy, I'd say: Get Out. This simple policy would save most people, possibly at the expense of the lives of those few people whose ability to evacuate is least certain.

The more chance of having short warning time, and the more chance of being trapped, the more necessary it is to prepare for the scenario in which people can't get out, whether they want to or not.

We personally live in a context where it is quite possible that we will only have seconds between when a fire starts and our home is engulfed in flames...flames that could quite likely be blocking our only exit route, a long, narrow drive up into the prevailing wind.

For this reason, we should have a contingency plan for weathering a wildfire in place, regardless of official policy, or even if our first choice is ready set go...what if we can't? Thousands of our neighbors are in similar positions. This page contains research that should be of intense interest to people like us, in the wildland urban interface.

There isn't much reliable information for people living on this edge. There is a statewide detailed building code for houses that take longer to catch fire. There are statewide, uniform requirements for wide, fire truck-friendly roads that may be useless for escape because of smoke limited visibility (and are nearly impossible to build in our mountainous terrain).

But if you want the security of a well-designed and built shelter in which your family would be guaranteed to survive a firestorm, there is precious little in the way of standards or testing, even though it is intuitively obvious that it is possible (and there are real examples) showing that such structures can be built.

Supposing you have such a shelter, the calculus shifts dramatically. Instead of simply Get Out, you simply Get In. You don't need much, or any of the Ready Set Go preparations (a working car, wide roads, distant clearing, heroic fire suppression...and enough visibility to drive). In a steep, brushy, sparsely populated area like ours each of these elements are more expensive and of more dubious effectiveness.


Worst Case Fire Conditions are Worse Than What We've Been Preparing For

Black Saturday, 400 Australian brushfires on and around Saturday 7 February 2009, killed 173 people and shifted the political winds against Prepare Stay and Defend. However, this was an extraordinary firestorm; perhaps the most extreme in recorded history. It seems that global climate disruption is raising the extremity of conditions that must be planned for, and the necessary margin of safety. From Wikipedia:

The majority of the fires ignited and spread on a day of some of the worst bushfire-weather conditions ever recorded. Temperatures in the mid to high 40s (°C, approx. 110–120°F) and wind speeds in excess of 100 kilometres per hour (62 mph), precipitated by an intense heat wave, and almost two months of little or no rain fanned the fires over large distances and areas, creating several large firestorms and pyrocumulus systems, particularly north-east of Melbourne, where a single firestorm accounted for 120 of the 173 deaths. A cool change hit the state in the early evening, bringing with it gale-force south-westerly winds in excess of 120 kilometres per hour (75 mph). This change in wind direction caused the long eastern flanks of the fires to become massive fire fronts that burned with incredible speed and ferocity towards towns that had earlier escaped the fires.

Humidity was as low as 6%, and much of the vegetation is oily Eucalyptus. Doesn't get much gnarlier than that. Classical (pre-Black Saturday) Australian Prepare Stay and Defend hinges on keeping a somewhat beefed up but essentially standard structure from burning, through active intervention of the residents. The margin of safety in this approach is slim (though it still might be better than making a run for it). What I'm proposing here is something considerably beyond that--


Impregnable Fire Shelter

I'm thinking that the key element for places where escape might not be possible is an impregnable fire shelter, designed to passively save the lives of the people inside, even in Black Saturday level conditions, with no active protection from the inhabitants.

This would have saved many Australian lives, and could save many Californians when our own Black Saturday happens. Once this element is in place, the calculus shifts. Even if you are stuck, you will probably live. Also, even if you might be able to get out and you can elect to stay to try and save your property, with the fire shelter as the fall back position. In that case, what you do could include active defense of other structures (the fire shelter shouldn't need it).


What would this fire shelter look like? That's not a settled design discussion, and it probably won't be until/ unless governments and researchers undertake some experimentation, or a lot of time passes and some lives are lost. Some DIY fire shelters look more like ovens. Large areas of exposed metal transmit heat, and are subject to buckling. An effective shelter would have likely have these elements—

Optional elements...

Retrofitted or purpose-built water tanks may be the cheapest, most effective fire shelters.

Water for Fire Safety-pdf, updated in the 2009 printing of Water Storage-book. Water systems for fire; fire bunkers; water tanks as fire bunkers.


Case study: Santa Barbara

Shortly after the 2008 Tea Fire in Santa Barbara, Governor Shwarzenegger said that due to climate disruption, we are now looking at a 10-month fire season in Southern California (followed by a 2 month flood season—only if it is a good water year).

With California's economy tanking, it doesn’t seem like we can count on heroic fire suppression efforts continuing indefinitely. It may be just a matter of time before most of the state adopts the Mexican fire suppression philosophy: let it burn.


Interestingly, Mexico is reputed to lose no more land or structures than we do. In the absence of suppression, the more frequent fires burn less intensely (as was the case in Santa Barbara for ten or twenty thousand years up until the modern era). Also, most construction in Mexico is either masonry, or simple and inexpensive to replace.

If current trends continue, the only viable strategy in the Santa Barbara wildland interface may be to plan on fire washing over the site fairly regularly and to build a mixture of structures that 1) can't burn, or 2) burn clean.

With defensible (and/ or low cost) structures, you can relax and let frequent, mild fires take care of most of the clearing.

Both of these strategies—500-year structures, and 10-20 year structures—make sense for Mountain Drive and other wildland interface areas. Neither is contemplated in the building code. Code requires super expensive structures that are more fire-resistant, but far from fire proof.

During the Gap Fire last summer, there were dozens of fire trucks on San Marcos Road, but none (except the local volunteers) would come into our community, because there was no safe haven from the fire should the wind change.

The fire proof Gymnasium at Westmont College kept 800 students safe, and off the road. (Imagine the chaos with another 800 evacuees).

Fireproof safe havens should be a feature of every mountain community. They are not mentioned in the code, though wide roads—which may or may not be useful in a fire, but certainly have high ecological and social costs—are required.

Cob (monolithic adobe fibercomposite) is a material which is more fire-resistant than concrete, because it conducts heat much more slowly. It is also earthquake and termite resistant, and exceptionally owner-builder friendly. A cob cottage with a ferrocement roof and metal shutters over metal dual-glazed windows and doors is a structure which could passively resist the most intense fire storm, and still be pleasant to live in the rest of the time.

A bio-regionally appropriate adaptation of rainwater harvesting for less than perfectly fireproof structures would have the rainwater harvesting tank plumbed to perforated metal water lines along the roof ridges. When the fire comes, flick on the pump and the stored water floods over the roof, into the gutters…and back in the tank for recirculation.


Current Trends

I've been predicting that within a few decades at most that:

1) Economics will drive a retreat from super expensive suppression and clearing around wildland homes in favor of a strategy allowing frequent, less intense fires around fire-armored wildland houses.

2) The expectation that residents will be the primary protectors of these homes will be widespread

2) Building standards for the wildland interface will call for structures that cannot burn, (or ones that burn clean and can be cheaply replaced)

3) Each area will be required to be able to house its population in fire safe havens, by law, practical necessity, or both

4) Fire insurance for combustible homes will be way less affordable even than it is now

If you have a lot of brush and not many houses, it is simply cheaper to have non-flammable houses and frequent fires than to have flammable houses, a vigorous clearing program, and a billon in suppression costs (see below). This transition will be facilitated by the progressive disappearance of flammable homes in the interface.


Interesting Articles

Note: this is a non-commercial, fair use compilation that includes copyrighted material from other sites, with credit to the source site. The research was motivated by our own desire for fire safety, after four close calls with wildfires in the past two years. We're posting our research notes in the hopes that others will learn from it and be safer as a result.

The first hand accounts of survival are particularly powerful and educational.



Southern California fire chiefs debate stay-and-defend program

Fire chiefs in tinder-dry Southern California, faced with lean budgets while more people squeeze into the region, are starting to rethink long-standing policies on ordering mass evacuations in a wildfire, debating whether it may be wiser in some situations to let residents stay and defend their homes.


Borrowing from tactics used in Australia for nearly two decades, top officials from fire agencies in seven Southern California counties started last fall to discuss moving toward an evacuation policy that makes allowances for people who want to try to save their homes. They will take the matter up again Wednesday at a meeting of Firescope, an advisory panel representing fire services statewide, said Roper, vice chairman of Firescope and a member of Gov. Arnold Schwarzenegger's 2005 blue-ribbon fire commission. The new approach recognizes that residents who have made their homes fire resistant, have cleared the brush around the house and have learned how to extinguish spot fires might be able to save property that would otherwise go up in flames because firefighters are overwhelmed.


'First you're fighting for your home, then you're fighting for your life'

A survivor describes his experience of the fires in Victoria, Australia

...when he saw the flames crashing through the trees Mr Lynn knew he was in for the fight of his life. "You do fire guard meetings and they show you what it (fire) is like", Mr Lynn said. "It comes along the ground but this thing just came in the sky on the trees. "When I was in the fire I thought 'this is just like hell'."

As the fire advanced Mr Lynn tried desperately to save his home, throwing buckets of water on the flames after his fire pump and hose exploded in the heat. Gas tanks at his house exploded around him before the fire finally reached his house and caused it to erupt in a ball of flames.

With nothing left to save, Mr Lynn made a break for the safety of the roof of his shed. "I burnt my hands, even with gloves, just opening up the door to get in the shed," Mr Lynn said. "I got a ladder and climbed up with buckets of water. The fire was all around and I was trying to put that out, but the shed caught on fire from inside. It just exploded." ...

A wall of flame pierced the roof of the shed and blocked Mr Lynn from escaping off the roof and down the ladder. "I just hung on to the spouting and the spouting fell down and I just collapsed down on the ground," Mr Lynn said. "I still had enough energy, so I just started going towards the paddock.

"I just got to that point when you are running and running and you can't run anymore. There was just no air, it was like you were breathing vapour." Crawling on his hands and knees through thick smoke Mr Lynn felt his way along the ground to the dam at the back of his property... "I just kept going and I could feel the bank of the dam," he said. "I felt like passing out so I didn't go in the dam very far, I just lay in it and put my head in the mud on the side."

Historian Stephen J. Pyne on the Australian Fires


What saddens many of us is that Australia knows better. It developed many key concepts of fire ecology and models of bushfire behavior. It pioneered landscape-scale prescribed burning as a method of bushfire management. It devised the protocol for structure protection in the bush, especially, the ingenious stratagem of leaving early or staying, preparing, and defending. In recent decades, it has beefed up active suppression capabilities and emergency response services.

Almost uniquely, Australia seemed to have gotten the basics right, certainly better than the muscle-bound, paramilitary response of North America. That approach only set up an ecological insurgency which summer surges of hardware and firefighters could never quell. Americans looked to Australia especially as a cognate country that knew how to replace feral fire with tame fire.

Yet Australia keeps enduring the same Sisyphean cycle of calamitous conflagrations in the same places. It isn't getting what it knows into its practices. It seems to be abandoning its historic solutions for precisely the kind of telegenic suppression operations and political theater that have failed elsewhere. Even when controlled burning is accepted "in principle," there always seems a reason not to burn in this place or at this time. The burning gets outsourced to lightning, accident, and arson.



Spending to fight California wildfires tops $1 billion

In fiscal 2008, half of the $1.4 billion that the U.S. Forest Service spent nationally on wildfire suppression was spent in California alone. State fire expenditures topped $1 billion.


"There is an absolute disconnect between requiring state taxpayers to take on the ever increasing burden of fighting fires when it's the decisions at local levels to put more homes and people in harm's way,"


Grijalva said fire officials also are considering experimenting with a version of Australia's stay-and-defend program: Rather than evacuating, homeowners are trained to protect their residences from the shower of embers that are typically more of a threat during a wildfire than encroaching flames.

"You'll start seeing pilot programs of what they do in Australia, with some modifications in California -- huge education programs," Grijalva said.


Australia's radically different philosophy about interface fires

According to an Aussie fire expert who stayed with us a few years back, Australia is one of the few places in the world where the peak fire front intensity exceeds our own.

'Stay or go' policy puts Australian families on front lines of firefighting



'A continent of fire'

Like those in Southern California, Australian bushfires are driven by wind. Hot, arid blasts out of the northwest rake the nation's desert interior, hurtling toward Victoria's eucalyptus forests and grasslands.

"Here reside the fires that give Australia its special notoriety, not merely as a continent of fire but as a place of vicious, unquenchable conflagrations," American fire historian Stephen J. Pyne has written. "In the fire flume lurk the great, the irresistible fires of Australia."

One of the worst blazes in the country's recorded history occurred in February 1983. One hundred fires started on what came to be known as Ash Wednesday. Eighty-three people were killed, 2,600 injured and more than 2,500 homes destroyed in Victoria and South Australia.

The tragedy shook the nation and prompted governmental and academic inquiries. One study found that occupied houses survived at twice the rate of those left unattended. If the occupants were able-bodied, the survival rate for houses was 90%.

Teams of researchers interviewed survivors and compiled their recollections. A major finding: Most of those killed by the fires were in vehicles or out in the open. In one instance, a group overtaken on the road by a wall of flames huddled in their car as the windows exploded and the seats began to melt. One woman screamed that she couldn't stand it any longer, then climbed out and ran for it. Her body was later found nearby, her rubber thongs fused to the blacktop.

Similar tragedies have occurred in Southern California. Nine people died fleeing the massive Cedar fire in San Diego County in October 2003. In Australia, Ash Wednesday prompted a rethinking of fire preparedness. Federal researchers pored over more than 100 years of data recorded by rural fire brigades. Their findings dispelled a host of calcified myths, notably the belief that bushfires incinerated homes in a wall of flames. In fact, scientists concluded, more than 90% of the houses lost were never exposed to direct flames or radiant heat. Rather, structures typically were ignited by embers. "That all instantly clicked with the fundamental observation that when people are around these structures, there's a massive improvement in the statistical survivability of the house," said fire researcher Justin Leonard. "It's small, relatively insignificant ignitions that, in isolation, are relatively simple to put out. But you just have to be there at the time."


IAFF: abandon "stay and defend" program


Most organizations call the program "Prepare, Stay, and Defend". It can only work if the PREPARE phase is complete. If a residence is not fire safe it can't be defended and SHOULD be evacuated.


Survival Forum: Stay and defend your home from a wildfire?

"Which brings me back to the bunker/shelter. One or two canny people had shelters, and they survived. These weren't even professionally designed. But they gave that vital protection, just for that 15 or so minutes that it take for a fire front to pass.

"There's a guy who lives a few miles from me [south of Sydney ] who has built two reinforced concrete dome shelters on his rural property. He's a Rural Fire Service veteran and says he would obviously trust his shelters with his life.

"Simple, proved fire protection, which gives him the option to fight the ember attack against his home until the last minute before retreating into shelter."

Nagging Wife Saves Family from Bushfire

Source: The Advertiser

The bunker, dug into an earth embankment with 15-centimetre concrete walls and a $1000 fireproof door, saved their lives and that of their son, Raphael, 14 months. "It was like a firestorm, it was like a raging inferno. It's a cliche, but that is what it was like," Ms Berry said.

When flames engulfed their home they wrapped themselves in wet towels and sprinted to the bunker. "We couldn't shut the door of the bunker, it was that buckled and warped," Mr Berry said. "The embers were coming through the gap, it was like the fire was coming to get us."

Wildland Fire Forum: Stay and Defend in SoCal

Re: Stay and Defend in SoCal
From : Californiadisasters_discussion

Hi Kim, Patricia and all, on the various subjects of "leave early or
stay and defend", as one who fights Wildfires here in Australia as well
being Senior Day shift Incident Controller for the the North Shore of
Sydney during the '94 fires, there are a few things about the Victorian

1. The majority of Towns/villages/houses totally or near totally
destroyed appear to have been located in areas totally surrounded by
tall timber forest or drought dried long grass paddocks and fields.

2. Most were perched on top of ridges with access up and down the sides
of the hills on roads with the forest right up to the safety barriers.

3. Most had the forest run basically through the houses to the main
roads, no fire breaks between forest and township.

4. Those who died in their vehicles trying to evacuate had left it to
late and had to drive down through the fire in very limited visibility
to try to reach safety. Very few had a large cleared area in the middle
of the village to shelter on.

5. The "Leave early" didn't work in some areas as before they had
realised, it most roads out had been cut off miles away by fires.

As all Wildfire fighters will tell you, the place not to be in a
wildfire is uphill of it, a fire travels up hill very fast compared to
traveling down hill, it also creates its own wind which pushes embers
and super dry air in front of it, pre heating the "Fuel".

None of the many burnt out vehicles fuel tanks "Exploded", this is one
thing that didn't appear to happen. Those that burnt out may have had a
window partly open or went stationary on top of burning matter with
plastic fuel tanks , which is now the "Norm" with most modern cars,
plastic melts and then dumps the boiling fuel to burn! Or the rubber
tyres caught fire.

We are taught to shelter in a vehicle in a "Cleared dirt or sealed Area"
with all the windows up and air vents on recirculate with blankets over
us to prevent radiant heat. But driving, partly blind from smoke, in
sheer panic down a mountain road with trees burning and falling around
you is not the place to be. Also if you stop on the road to not hit a
fallen burning tree, you risk being hit from behind by someone else
driving blind!

In one community all the women and children sheltered in the Volunteer
Rural Fire Station whilst the men sheltered in their vehicles in the
cleared car park outside the station. The Rural Fire Fighters stood and
defended these people, in the open with what water they had whilst under
continual ember attack. One description of the ember attack was "It was
like burning hail stones!". The men helped put out spot fires until it
was deemed unsafe for them due to the burning embers burning them and
their clothes and only the Volunteer Rural Fire Fighter wearing fire
proof turnout gear continued to protect everyone until the fire was
past. Most rural fire stations are either "Tin Sheds" or Metal and
brick, so they don't tend to burn.

There was also a couple of people who had put "bunkers" on their
properties, 40' containers buried under berms of earth to keep away
radiant heat, who survived the firestorms which left no buildings
standing. Maybe an idea in wildfire prone areas hmmm, an underground
tool shed.



Water Storage for Fire Prevention

Check out  Water for Fire Safety-pdf, an excerpt which describes how you can fire armor your water system, and turn your water tank into an impregnable fire bunker. This info is from Water Storage-book.


Home fire sprinklers may make massive losses of homes in fire storms more likely.


Fire Lists


More Resources


Tip jar