Rethinking When, Where, and Why We Carry Fire Shelters

By Brent Woffinden, Fire Management Officer

U.S. Fish and Wildlife Service/National Park Service – New England States

The pending 2021 fire season will be my last as a permanent federal employee. Over the years of my service I often had pause in carrying a fire shelter when I knew that it would not be a useful component to my Personal Protective Equipment.

The thought of being involved in a fire entrapment or burnover incident is a real and serious concern for all wildland fire fighters. The ubiquitous presence of fire shelters as a significant component of our PPE numbs us to its importance. Our most important tool in entrapment avoidance is the collective knowledge of weather and fire behavior that should guide how we interact with the dynamic nature of wildfire.

While I am not an advocate for eliminating the use of fire shelters, when we understand and train for their proper application and importance, we will be able to move beyond our reliance on shelters when their presence is either unnecessary or dangerous.

Sound Decisions on Proper Engagement and Disengagement of Tactics

My career began in 1985 working as part of the Cal Poly Max 2 Crew, a 20-person wildland fire crew comprised of college students. Our first shift of our first fire of the season was on the Wheeler Fire near Ojai, California.

We had started to cut handline off a road going up Matilija Canyon when the fire blew up and made a run down the canyon. As engines momentarily disengaged from structure protection, we did the same, leaving a partially completed fireline to escape the advancing fire. By the time we had reached the crew buggies, the fire was on both sides of the road and engulfing the vehicles as they made a slow egress to safety. As this was my first fire, I thought all this was normal.

We continued to engage in suppressing the Wheeler Fire for the next 19 days as it moved onto the Los Padres National Forest. Three years later, as a member of a 20-person Alaskan hotshot crew, while engaged in suppressing a large tundra fire near Alaska’s Warring Mountains, a wind switch occurred, rapidly changing direction of the fire and forcing our crew to use an escape route to a previously identified safety zone.

In both of these situations—in California and Alaska—crew leadership made sound decisions on proper engagement and disengagement of tactics based on fire behavior that remain with me to this day.

Photo of the Cal Poly Max 2 Crew taken just after they were chased out of Matilija Canyon on the Wheeler Fire in 1985. (Blue arrow identifies Brent Woffinden.)

Actions Taken in a Compromised Situation

In 2005, while serving as a Fire Management Officer (FMO) based on the Texas coast we were applying fire on Matagorda Island. A trainee from the Fire Use Training Center and I were doing some interior ignitions when we temporarily lost radio communications with the primary ignition crew and were in a compromised situation.

The trainee asked if we should find a location to deploy fire shelters. With the average fuel loading exceeding 8 tons per acre this was not a viable option. He did not have a familiarity with the terrain and fuel type. I assured him that we would find a suitable spot where deployment of shelters would not be needed and if no such spot could be located we would use one of the several different types of firing devices to create a suitable safety zone. We found a suitable area where the fire worked its way around us without the need to exercise the emergency firing option.

The Questioning of Carrying a Fire Shelter

I share these three examples because, while in all three instances fire shelters were carried, they definitely would have led to a much worse outcome if a shelter had been deployed in fuel conditions that would not have been survivable.

Since that auspicious initial fire and the 36 years of wild and prescribed fire since, like the vast majority of U.S. fire personnel, I have never exercised the option of deploying a fire shelter. As much as I would like to attribute this to the application of sound fire practices and tactics, luck has also played a role as there are times when circumstances dictate the need for such use. Having the ready option to deploy a fire shelter should therefore be available. That being said, there are also times when carrying a fire shelter has a neutral or negative consequence for wildland firefighters.

Consider Eliminating Fire Shelters on Prescribed Fires

For instance, when planning for and implementing prescribed fires, the need to carry fire shelters in most, if not all, cases should be eliminated through proper planning. The planning process for prescribed fires requires us (federal employees) to assess the risk to firefighters when completing the “complexity analysis” of all prescribed fire plans. If on the high end of the complexity analysis, we must ask ourselves why we are engaging in the implementation of a high-risk operation.

Furthermore, if the risk is both required and accepted by line personnel and leadership, then fire shelters should be considered and discussed in the operational briefing. If on the moderate to low end of the complexity analysis, then fire shelters should not be carried or provide the option to not carry one as they are unnecessary.

This spring my current home unit plans on implementing prescribed fires all with low to moderate risk based on the completed complexity analysis. Those on the moderate end of risk are due to other factors not related to fire behavior. One example is a planned burn on a small island in the middle of a river in which we plan to burn three acres of grass. This risk is on the boat ride to the island in which having the extra bulk and weight of fire shelters would reduce safety rather than enhance it. The other planned fires are on coastal islands, in urban areas or open fields with insufficient vegetation to allow for Rx fire to get out of control sufficient to require shelter deployment.

If a local unit accepts the rationale for implementing high complexity prescribed fire, then by all means fire shelters should be carried. During the briefing, it should be explained why such a tool is needed and reinforce the measures taken to avoid its use. In the same line of thinking, if a high level of risk is taken, then that risk should be acknowledged via hazardous duty pay for those that qualify (line assignments and employment status).

When, Where, and Why

In summary, as the risk of engagement with wildfire will continue to increase, we should rethink when, where, and why we carry fire shelters.

WHEN: When engaged in initial attack or certain large fires. Fire managers and incident commanders should be able to decide when there is a need to carry fire shelters. (For example, when mopping-up in the rain the need to carry a fire shelter is not necessary and having it in one’s gear subjects it to undue wear and tear as well as adding weight and reduced efficiency to individual firefighters.)

WHERE: Where the fuels would allow for rapid safe deployment. (For instance, not in boreal or palmetto forests or other fuel types where deployment is detrimental to survival.)

WHY: If there is valid rational for fire personnel to carry a fire shelter then the only reason for use would be as a last resort when all other options are not viable (retreating to the black/safety zone, etc.).

By thinking about When, Where, and Why to carry fire shelters—rather than having our shelters be a heavy, bulky and expensive resource that sits in our packs being sat on and subject to abrasion by dirt, dust and smoke—will make us all better students of fire and how we engage with it.

14 thoughts on “Rethinking When, Where, and Why We Carry Fire Shelters

  1. I couldn’t disagree more on the assessment in this article that we should be able to choose when to carry a fire shelter. There are too many variables involved and whether you are on a prescribed fire, a wildfire or in training, it should be part of the equipment you have on hand all the time. As has happened in the past, firefighters have forgotten to replace their shelter after training or project work and then needed it later. This also sets a bad precedence. Other fire fighters will see those who decide not to carry their shelter and do the same. Some fire fighters will remove the shelter and then plead ignorance later that they don’t have it or forgot to bring it. There is a reason we carry the fire shelter, it is the last ditch protective device we have when everything goes wrong and it has saved many lives.

    • I’d also point out that other parts of the world that don’t carry shelters seem to have less burnover fatalities because they are forced to rely solely on situational awareness. I note also the part about not being required for mopping up. There is some old research that states one of the four common factors in burnovers is small fires or quiet sections of larger fires supporting complacency as a key factor. Certainly when we are assigned in good black a shelter is hard to justify.

  2. I have to say, after thinking about the ideas in this post, I have to agree with Brian Kliesen.
    In my 42 (and counting…) years in the fire service, I have been blessed never to have been in a situation where a fire shelter was employed, nor needed to be. This is not to say that I haven’t experienced situations where they may have been if good decisions and serious situational awareness were not employed. I have walked my people out of many situations prior to the need to run!
    Nonetheless, I am not aware of ANY situation where those entrapped firefighters willingly exposed themselves to conditions that they knew in advance would require the use of a shelter. (Heck, no person I know even exposed themselves if there was a significant chance that they would have to!). No one does, shelters are popped once everything else has gone to hell, and one’s options have diminished to basically none.
    When watching training videos and reading after action reviews, it is common to see those who had to deploy are very commonly highly experienced firefighters. Not new folks who don’t know better, but good, quality personnel possessing a multitude of experience who somehow ended up with no other option. The reasons those situations occur are probably about as numerous as the number of deployment reports.
    The bottom line is that you don’t know when you will need a shelter. I will guarantee though, that 30 or so minutes before you deploy, you were not thinking that deployment would be in your immediate future.
    One of the first rules of firefighting I learned was: “Better to have it and not need it, than need it and not have it.” Of all my gear, this probably applies to shelters more than anything else. Just because I’ve come home every night for 42 years with an unused shelter, that doesn’t mean tonight will be the same. I am not so foolish to think that because I’m so “good”, I don’t have to worry.
    I get that the shelter is heavy and expensive. It’s certainly not a piece of equipment I enjoy carrying around. However, that’s not really the point.
    I pray that none of us will ever need to use a shelter. If you’re very good (and, possibly, very lucky!), you won’t, but there is no doubt in my mind that when all else fails, you will be better served if it’s securely attached to you and available when your choices dwindle to just two.

  3. I couldn’t agree more with this article. We often do things in the wildland fire environment because that’s what we’ve always done, or the assumption we can’t train folks to make good decisions. I could compare our decision making and fire response to Australia, where they don’t carry fire shelters, but I’ll let you experience that for yourself one day. Having been on hotshot crews for 15 years or so, in the FS for 24 years, there was more than one situation where 8 pounds lighter could have aided me scouting, aiding injured firefighters, escaping hazard trees, and prevented the repetitive banging away on my C5/6 disc which ultimately ruptured and is now fake.

    We have examples of shelters deployed in dense fuel types that didn’t do anything to save the poor souls and may have led to a false sense of security. Only once have I had a shelter in hand, deep in the Bob in 2000 with my saw partner and supt and was saved by an old Vietnam veteran in a bell 205, who shouted “this is just like Vietnam!!” as he pulled up hard and fast with flames not far behind. If this is you reading this type 2 helicopter pilot, I owe you a beer and would love to hear your side of the story. When the superintendent went back up he couldn’t even find the LZ where we preparing to deploy in a dense lodgepole forest.

    We need more of this hard thinking and questioning. We need to question the old answers and keep pushing for new technologies, new ways of assessing risk, new ways of analyzing our decisions. There have been endless debates on the rules or guidelines or requirements of the 10 & 18, which have saved more lives than shelters ever will. It is often the questioning and discussion that are most important, generating new ideas and ways of thinking, even if we collectively decide the shelter is a required piece of PPE forever.

    • Absolutely agree Marcus. Well presented. My Australia roll certainly presented the additional rationale of fighting fire without them and utilizing the most important 6 inches on the fire ground more and more sound.

  4. “In the same line of thinking, if a high level of risk is taken, then that risk should be acknowledged via hazardous duty pay for those that qualify (line assignments and employment status)” no hazard pay unless it is deemed needed? Who get to make that decision? I know there have been few times in my 10 years I would not want to have my shelter with me. Start giving that decision to someone else and I am out is the CRWB, DIVS or who making the call? Also can bet that no shelter means no H pay and this decision is done for anyone trying to live the year on a seasons pay. If it was an easy job everybody would be doing it.

  5. I would suggest that the wide range of comments already posted demonstrates the power of opening up a space to discuss/question practices and assumptions. Good job, folks!

    Started my fire career with the FS in 1985 for what it’s worth, 14 years of it hotshot crews in WA, UT, AK, WY.

    My utopia would be no fire shelters, which research from all over the globe shows would result in a reduced level of risk taking (both big picture/strategic and operational). Hard to see us ever implementing that in the US, as our system is very supportive of litigation.

    However, understanding how the mandatory carrying of fire shelters at the very least subconsciously influences our risk tolerance/risk taking decisions is worth talking about. Cool article from career risk management dude, with a list of the 10 factors that influence risk tolerance; here’s #8:

    “8. Overconfidence in personal protective equipment
    Risks become more common when the limitations of the personal protective equipment (PPE) are not understood. Higher levels of risk are accepted when workers are overconfident in the safety equipment they use and their belief that if something does go wrong, the PPE will keep them from harm.” https://www.westernenergy.org/news-resources/personal-risk-and-safety-decisions/

    I know at this point in my career, I am fairly aware of what I’m factoring into my own risk tolerance and risk decisions, but that’s been gained by decades of personal experiences, tales from peers, LLC, working accident investigations, etc. I think there is still a lot of room for improvement in creating a higher level of knowledge around risk tolerance/decision factors in all of fire, and amongst the agency administrators who set the fire strategies and objectives into motion with their leader’s intent.

  6. Ah yes, lets replicate historic fire in un-historic fuel beds and by all means lets plan on it to a gnats ass degree. Anyone who has done this to any complex level knows differently. As it is now, litigation and CYA management are making burn boss a no win and why bother position. When writing burn plans articulate the risk better and make sure its a shared risk worth taking on and by shared I mean all levels of the organization and effected interests. If the conditions are so deteriorated that fire is the only fix then a potential fix needs to consider that risk wisely. When you are trying to address issues in say Mixed Conifer forests and trying to replicate a stand replacing regime without actually replacing the stand then why bother? If it makes sense then implementing it means you are on the ragged edge if successful or folks are going to seriously not like my control lines in order to prevent escapes and enhance safety. Try getting high mortality in Juniper and tell me how that works out without using the precautions you use on any wildfire. This is a prescription for setting back prescribed burning, the first time we lose someone to a prescribed fire that a shelter could have been used and rest assured we will they will stand down everything and ultimately require them for every fire. .

  7. Thank you for posting this article and opening it up for discussion. I started in 1972 when there were no fire shelters, no Safety Zone mentioned in the Fire Orders (only Escape Routes until late 1980’s), and 13 Watch Out Situations. Fire shelters used to be “optional” on certain wildfires and finally became mandatory after the fatal 1976 Battlement Creek Fire in CO. Those of us working on the August 1985 Butte Fire in ID, feel that it started what we call the “fire shelter movement.” Contrary to NWCG approved training materials, the wooden sign, hand-tool handles,and the grass never burned on Tin Cup Hill where a Hot Shot Crew deployed shelters. The hand-tools, drip torches, etc. that are referred to in the video burned on a spot fire elsewhere.

    Another story for another day.

    The mandatory fire shelter training video claims that they have “saved hundreds of lives and prevented hundreds of serious burn injuries.” This may very well be true with exceptions. Several have died BECAUSE they deployed their fire shelters. It is a fact that tens of thousands of WFs are “saved” every fire season because of the wildland firefighting rules and guidelines.

    All WFs are trained in specific rules, crucial to follow to ensure good direction, leader-ship, safety, and vigilance. The strict Standard Firefighting Orders, organized purposely and sequentially, are to be carried out sensibly on all wildfires. The 18 Watch Out Situations, (i.e. guidelines), are faced on all fires, more to warn of impending dangers. Experienced WFs contend that knowing and abiding by the wildland firefighting rules works. They urge sound leadership and safe decisions. There are no documented wildfire fatalities when the Standard Firefighting Orders are followed and the cautionary 18 Watch Out Situations (“10 & 18”) are mitigated. Sadly, after the 2013 Yarnell Hill Fire, there is a crusade afoot by current and former WFs and Managers to discredit these based on the SAIT conclusion, (i.e. “they did everything right and still died”).

    The most critical of the established Wildland Fighting Rules are listed in the (NWCG) Incident Response Pocket Guide (IRPG). Again, if WFs follow the Standard Firefighting Orders and are alerted to the 18 Watch Out Situations, much of the risk of firefighting can be reduced, saving tens of thousands of WF lives each fire season. It is well-known and accepted in the WF community that these tried and tested rules work when applied consistently and with sound, decisive judgement.

    The fire always signals its intentions.

    I agree with kevinpfister’s claim about the danger of “8. Overconfidence in personal protective equipment …” and anonymous’s claim that the “USA [is the] only country in the world to use fire shelters and coincidentally has the Number 1 burnover rate.”

    Marcus Cromwell, a highly experienced and knowledgeable WF and supervisor makes excellent points, (e.g. more of this hard thinking and questioning; questioning the old answers while pushing for new technologies; new ways of assessing risk; new ways of analyzing our decisions; the rules or guidelines or requirements of the 10 & 18, have saved more lives than shelters ever will”).

    I wish that we would get rid of fire shelters (or make them optional) but the Thirty Mile Fire all but ensured they will remain.

    However, generating new ways of thinking (i.e. “consensus management” and “Forest leadership teams”) have caused more indecision and trouble than they’re worth. And how about the alleged panel of “experts” discussions in the “Honor the Fallen” video discrediting the Fire Orders, while instead relying on “luck decision conversations” is a huge mistake. It’s only a matter of time before the outcome of this manifests itself.

    By the way, how does that “luck” thing work? Do you get so many when you’re born? Can you buy or trade for them” I refuse to believe in luck per se.

    New technology? There will NEVER be a reasonably weighted fire shelter that we can carry on the firelines. The only man-made product that would have saved WFs (that died in their fire shelters) is a literal, metal shipping container made of “black box” material for aircraft communication and data equipment.

    One of the main changes we need is forthright serious accident investigations that give us the human factors of “how” and “why” of what happened. We are instead learning incomplete lessons. Fatalities in ALL work groups is inevitable, so all we can do is lessen them.

    Ponder on this quote: “If, as teachers of history will tell us, failing to learn the lessons of the past dooms us to reliving those lessons, then we must either impress indelibly into the minds of fire-fighters the lessons of the South Canyon Fire or we will again experience its tragic outcome.” (Gleason and Robinson, 1994)

    • I wish to clarify this assertion: “New technology? There will NEVER be a reasonably weighted fire shelter that we can carry on the firelines.”

      What I meant to say was: There will NEVER be a reasonably weighted fire shelter that we can carry on the firelines, that can withstand the temperatures and fire behavior that occurs in these entrapments, burnovers, and fatalities.

  8. Several comparisons to Australia in the discussion. As a remote area firefighter in Australia, with 18 years of fire, i would have to say that the main differences, and why we don’t carry shelters (apart from cost), is tactics.
    We fight fire differently, most of Australia’s firefighters are engine based (all vehicles have burn over protection features). Only a very small number of Australian wildland firefighters hike or fly in to work remotely, and never on ‘going’ fires with aggressive flame heights as it’s simply not safe. Remote crews here, work close containment on active fire edges so we are literally seconds from the black when actively fighting.

    US ground crews sometimes work in a much higher risk environment. Whether that risk is worth the results obviously depends on the fire, but here, management jump through a lot of health and safety hoops before deploying crews on foot away from support, Our risks are often lesser than US hand crews and shelters are simply not a consideration. That does not mean we are complacent and ignorant of fire behaviour, we monitor and assess constantly and will pull out without a second thought if the situation dictates.

  9. Some really great discussion here. Thanks for writing this article, it gets us all thinking and talking and that is good. I’m a fire behavior researcher, and also a firefighter for 25 years (quals up to DIVS). I was also involved in the development of the current fire shelter.
    I would say my personal experiences and thoughts align with much of what has already been said. I DO think we need to put more emphasis on training firefighters to understand and predict fire behavior better on the ground. Nearly every entrapment has the common theme that the fire behavior that cut them off was not anticipated. Are there cues that were missed that could have helped predict the fire behavior? What are these cues experienced firefighters use and can they be taught, improved, documented. Or at least were there cues that the potential and risk was high? One of the main keys to all this is the weather of course. Understanding the uncertainty associated with the weather forecast, how quickly and likely it is that things could change for the worse (for example a dramatic increase in winds). Are there essentially random thunderstorms with outflows possible today, a cold front with unknown timing, or a regular high pressure sunny day? The other key piece of all of this is how it relates to our subsequent actions… where we put ourselves in the terrain related to the fire location, escape route/safety zone locations, potential future spot fire locations, etc. How comfortable are we being far from our safety zones under different conditions? How much margin is built into our escape plans? As has been said before, we are trying to make these decisions based on incomplete information, some of which are very subjective, in a time crunched situation.

    I often wonder how we could improve these safety decisions. It really hurts me every time I read another entrapment report. Can our training be improved? Should our culture in the US of accepting a certain amount of risk be changed?

    I my capacity as a researcher I’ve been working on something I think can improve one small piece of the big puzzle. It will be an app called the Fire Weather Alert System (hopefully released in the next year or so). The basic idea is that the mobile app will warn firefighters of incoming hazardous weather using all the information that is out there (weather stations, radar, satellite, forecast models, lightning detection, etc.). Some examples: a thunderstorm 5 miles south of you was just detected on radar, or a RAWS station 20 miles west just measured a 40 mph wind gust from an incoming cold front, or the RAWS just measured a 10% RH. All of the alert thresholds can be custom set by the user, and the app watches the weather at their location and a user set radius around (say 50 miles). In case you were wondering, no, I don’t think technology is the answer to all the entrapment problems… but I’m hoping this can help a little, at least in some cases where unforeseen weather contributed.

    I think another path toward improvement could be in our fire behavior training. We don’t do a great job right now in my opinion. Much of our training is based on and geared toward a computer model (Behave). That is great if you’re in the office in the winter writing a burn plan and need to identify and document anticipated fire behavior and acceptable conditions, but not so great if you’re standing on a ridge above a fire during the southwest monsoon season trying to decide if it’s safe dig downhill indirect line. Instead could our training be more geared toward the on-the-ground firefighter making decisions based on what they see? Possibly incorporating the idea of recognition primed decision making into this training could help. The idea of teaching cues to look for that can warn of impending changes. This idea is not completely new, the Campbell Prediction System and FLAME used some of these ideas.

    Regarding our culture of risk in the US, maybe we can learn from others. The Australian firefighter above seemed to suggest that they will not work under the level of risk that we do, and pull out sooner. I was on a fire in Oregon years ago and ended up working with a Canadian Task Force Leader. I think he was from BC or Alberta. We were getting spot fires over our line and the crews were picking them up, fairly standard stuff in the US. He commented that in Canada where he works, if they get just one spot fire they are required to disengage and let aircraft knock it down before they can re-engage. A much lower level of risk acceptance.

    One last point I’d like to make is about fire shelters. The discussion above about fire shelters actually being a detriment in some situations is correct. The example I like to use is the Thirty Mile Fire. Remember in some of the fire shelter training material where you are told to not get out of your shelter once you get in, no matter what happens or how hot it gets. This may be true in most cases, but for the Thirty Mile Fire doing the opposite likely saved one firefighter’s life. In that situation 5 firefighters (if I remember correctly), deployed in the rocks above the road and river, one eventually decided this location was bad and too hot and left his deployment location and moved down to the river. He survived and the others that stayed in their shelters died. So yes, I think it is very important that firefighters know the limitations of their shelters for different fuel types, and when running is a better option than deploying in a bad spot. Personally, I do still think all firefighters should carry fire shelters though. There are just too many entrapments, too many misjudgments of fire behavior. It happens many times each year.

  10. I believe that making fire shelters required PPE does lead to assumptions by some, that they can push the envelope and depend on a shelter for protection. As others have pointed out, the US requires this and has the highest rate of burnovers – you have to wonder why. Second, I believe that the level of instruction and training regarding fire behavior and weather is completely inadequate. We also do not have near the weather prediction capability we should have – like many other parts of the fire management program, funding is not remotely keeping up with the need. Finally, why are we even putting people out on most Rx fires? The use of at least PSD firing operations and the best, drones can nearly eliminate much of the risk that is talked about in this thread. Again, we are way behind the curve in truly using technology to keep folks safe.

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.