USFS having difficulty hiring firefighters to suppress wildfires in area contaminated with asbestos

The U.S. Forest Service is trying to fill positions on a very specialized 10-person wildland firefighting crew. The mission of the crew would be to suppress wildfires that occur near a mining site at Libby, Montana where vermiculite contaminated with asbestos was extracted by the Zonolite Corporation and later by W.R. Grace from 1919 until the mine closed in 1990. The asbestos became airborne and deposited in the adjacent forest and other areas. Wastes from the mine were used throughout Libby in many public places such as school tracks, public parks, baseball fields, as insulation in public buildings and schools.

former vermiculite at Libby, MT
Map showing the location of the former vermiculite at Libby, MT. Click to enlarge.

The area is now designated by the Environmental Protection Agency as a Superfund Site. The first public health emergency ever declared by the EPA was the Libby asbestos site in 2009. Hundreds of asbestos-related disease cases have been documented in the small community, which covers the towns of Libby and Troy.

Below are excerpts from an article at The Western News:

…As part of a previous agreement, the Forest Service is responsible for fire containment and cleanup in the mine area. Libby District Ranger Nate Gassmann said having a team located in the area is critical to containing the threat of airborne asbestos if that case were to happen.

“Both agencies understand that importance if the community of Libby and the surrounding area is affected by fire in someway,” Gassmann said. “We do not take this as a light consideration for the Forest Service.

Christina Progess, remedial project manager for Operable Unit 3, known also as the former W.R. Grace mine site, said that the EPA worked hand-in-hand with the Forest Service to develop the action plan, while state agencies like the Department of Environmental Quality and Department of Natural Resources and Conservation took support roles, providing input while the plan was under development. Progess said the plan released on Tuesday has been in the works since spring of 2016.

The team would be composed of Forest Service firefighters, according to the plan, but filling that roster has already proven difficult. According to the memo, Forest Service Fire Managers have discussed firefighting within OU3 with Forest Service firefighters and “most have indicated that they would refuse to work in OU3 due to the presence of (Libby Amphibole asbestos) in forest duff and tree bark.”

Gassmann said while efforts to build the team has been met with hurdles, some support positions have already been filled and the Forest Service may begin looking to outside sources to compose the 10-person squad.

Gassman also said the Forest Service has provided forest fire containment in the mining area before, including two incidences in 2015, although those fires totaled a .75-acre area.

“On average, we receive four fires a year” in the former mine area, Gassman said. “Sometimes you get more, sometimes you don’t.”

Progess said the EPA and Forest Service conducted a test burn earlier this year to determine the exposure levels found in the smoke and ash of a fire in the former mine area. The test burn was a small fire, she said, but the exposure levels were great.

“We had the test burn and had firefighters do some mop up in the area. We found that their exposures were well above the risk target set by the EPA,” she said.

“Exposures were significant and of concern.”

Progess said that if a large forest fire were to tear through the former mine site, the EPA is currently unable to quantify how far or how concentrated the mobilized asbestos would travel through smoke and ash.

“There’s so many variables that would factor into it, from wind to topography to the relative humidity,” she said. “We don’t have any way of understanding what the concentrations would be to residents in Libby but the best way to minimize exposure is to prepare to stop a fire.”

Gassmann said while the primary objective is to keep area residents safe from such asbestos exposure levels, there’s plenty of concern for the safety of the firefighting team, once that crew is assembled.

“We have a requirement to provide health and safety for our fire fighters. That’s above and beyond what you would consider a normal fire fighting activity,” he said.

Safely training the tactical athlete

EMTAfter reading the Facilitated Learning Analysis (FLA) about the Rhabdomyolysis (Rhabdo) injury that occurred May 2 in South Dakota on the first day of the fire season after running for more than nine miles before doing uphill sprints, I started thinking about, not so much WHAT happened, but how to prevent similar serious injuries.

A couple of weeks before the Rhabdo case, on April 19 a wildland firefighter in the Northwest suffered a heat stroke while running on day 2 of their season. The employee was unconscious for several hours and spent four days in the hospital.

Both of these exercise-induced conditions can be life-threatening; 33 percent percent of patients diagnosed with Rhabdo develop a quick onset of kidney failure, and 8% of all cases are fatal.

Heat stroke can also kill, according to Medscape:

When therapy is delayed, the mortality rate may be as high as 80%; however, with early diagnosis and immediate cooling, the mortality rate can be reduced to 10%.

These two very serious incidents in a two week period that occurred at the beginning of the fire season should be a wakeup call for agencies employing wildland firefighters.

I am not a medical or exercise specialist, but neither were any of the four members of the South Dakota Rhabdo FLA team. It was comprised of a District Fire Management Officer, a Natural Resources Specialist, an Assistant Superintendent on a Hotshot crew, and an Assistant Fire Engine Operator.

A person might expect that for an exercise-induced injury that is fatal in eight percent of the cases, a medical expert and an exercise physiologist would be members of the team. The FLA concentrated on recognizing symptoms of Rhabdo, which is good. Firefighters need to be be informed, again, about what to look for. But the necessity of treating the symptoms could be avoided if the condition was prevented in the first place.

Prevention was not addressed in the document, except to mention availability of water. Dehydration isn’t the leading cause of Rhabdo, which is caused by exertion, but it can be a contributing factor.

With two life-threatening medical conditions on firefighting crews in a two-week period that occurred during mandatory exercise on day one and two of training, medical and exercise professionals perhaps could have evaluated what caused the injuries, and suggested how to design and implement a physical fitness program that would lessen the chances of killing firefighters on their first or second day on the job. But the LEARNING opportunity of the FLA was squandered.

The wildland fire agencies are not alone in hiring people off the streets and throwing them into a very physically demanding job. The military does this every day, as do high school athletic programs. There is probably a large body of research that has determined how to turn a person into an athlete without putting their lives in danger.

While the three firefighters and the natural resources specialist I’m sure meant well and did the best they could to write the FLA within the limits of their training and experience, the firefighting agencies need to get serious about a professional level exercise training program. After all, they are employing TACTICAL ATHLETES.

This issue is serious enough that the NWCG (since there is no National Wildland Firefighting Agency) should hire an exercise physiologist who can design, implement, and monitor a program for turning people off the street into tactical firefighting athletes.

Facilitated Learning Analysis for May 2 Rhabdo injury

33% of patients diagnosed with Rhabdomyolysis develop a quick onset of kidney failure, and 8% of all cases are fatal.

The Wildland Fire Lessons Learned Center has released a Facilitated Learning Analysis for the Rhabdomyolysis injury that occurred May 2, 2016. It does not specify that it was the case that occurred on the Black Hills National Forest, but many of the facts in the document point to it being the same incident.

The short version of what preceded the injury is that on the morning of the first day of the seasonal firefighters reporting for duty this fire season, the crew was directed to complete an 8.8 mile run which they did in 96 minutes. Approximately 1/2 mile into the run one crewmember dropped out and was evaluated by a squad boss and an EMT. The crewmember and the EMT returned to the base. This was not the person later diagnosed with Rhabdo.

After the 8.8 mile run the crew jogged another 3/4 mile to a location where they ran uphill sprints and a “loop run”. From the report, after the 8.8 mile run:

Upon return to station, the remainder of the crew reconfigured and lined-out in “tool-order” to continue PT. It was noted that during this brief lull in activity, the employee who would eventually be diagnosed with Rhabdomyolysis made the comment “It’d be nice to have some water…”, to which another within ear-shot replied “yeah… I know”. The “long, slow run” was followed by three rounds of relatively short uphill sprints interrupted by a “loop-run” within sight of the hot-shot base. This event lasted roughly forty-five minutes.

Although dehydration isn’t the leading cause of Rhabdomyolysis, which is a condition caused by exertion, it can be a contributing factor.

The crewmember did not inform the supervisors that he was having discomfort and cramping, but about an hour after the work day ended he drove himself 41 miles to seek treatment at a medical facility.

At 0745 on the [next] morning of May 3rd, the hotshot superintendent was notified by the injured employee’s family that he was in the hospital with dehydration and were awaiting additional test results. He was subsequently diagnosed with Rhabdomyolysis.

The FLA points out, and this should not be news to wildland firefighters, that Rhabdo and compartment syndrome are extremely rare and difficult for a physician to diagnose.  Therefore it is imperative that wildland firefighters familiarize themselves with what can cause the condition and how to recognize the symptoms.

Not all past cases of rhabdo in wildland firefighters were correctly diagnosed during initial care. Heat illness and dehydration share common signs/symptoms and can lead to a missed diagnosis for rhabdo. In addition, rhabdomyolysis is a very rare occurrence in the general population. Many physicians will go their entire careers without seeing a single case of rhabdomyolysis. Since early detection and treatment can greatly reduce the severity and recovery time, it is important that medical providers understand and test for rhabdo.

If you are a wildland firefighter, and especially if you are a supervisor, read the entire report, make copies of the Handout for Medical Providers, and if someone exhibits the symptoms and needs treatment, accompany them to the medical facility and diplomatically talk to the physician about the possibility of Rhabdo while giving them a copy of the Handout.

In another injury involving early fire season physical training, on April 19 a wildland firefighter suffered a heat stroke on day 2 of their season. The employee was unconscious for several hours and spent four days in the hospital.

 

Another serious injury during PT at beginning of fire season

The firefighter spent four days in the hospital after suffering heat stroke.

Another firefighter has sustained a very serious injury during physical training at the beginning of their fire season. The first one that we are aware of this year occurred on May 2 in South Dakota when on the first day of training the firefighter was diagnosed with Rhabdomyolysis.

The Wildland Fire Lessons Learned Center just released a “72 hour report” from the Pacific Northwest Region about an April 19 heat stroke victim that happened on the second day.


firefighter heat stroke


Both of these conditions are extremely serious and in the worst case, can lead to death.

It must be very difficult to develop a perfect exercise regimen for all 20 people on a hand crew. Especially at the beginning of the season when some of the experienced firefighters may have had an ongoing physical fitness program during the off season, and others are brand new, never having held a shovel, and may have spent the winter on a couch.

Fire engine’s air brake line melts due to heat on a wildfire, locking brakes, disabling the vehicle

When a fire is bearing down on your location, the last thing you need is for the brakes on your fire engine to lock up, disabling the vehicle.

Below is a Rapid Lesson Sharing report recently released by the Wildland Fire Lessons Learned Center. Click on the two images to see larger versions.


 

brake line failure wildfire

brake line failure wildfire

Some fire engines are more fire resistant than others. They don’t have to be completely fire proof, but it is dangerous to have parts, systems, or areas on the vehicle that are extremely vulnerable to moderate, short-term heat or susceptible to be easily ignited from a burning ember. Some of the bush fire engines in Australia have fire curtains that can be deployed on the interior side of the widows, and have permanently installed external nozzles or sprayers that can be activated to protect the truck and it’s occupants.

What are your thoughts about some vulnerable points on fire engines that could be improved in order to lessen the risk to firefighters?

Jerry Williams: can-do, to make-do, to tragedy

This video was posted by Colorado Fire Camp on their Facebook page. It shows less than three minutes of a talk by Jerry Williams, who became Director of Fire and Aviation for the U.S. Forest Service. It was 1995, a year after 14 firefighters were killed on the South Canyon Fire in Colorado.

Here is a partial transcript of Jerry Williams’ talk:

…It’s a proud outfit. It’s a can-do outfit… It’s a real short step between can-do, and make-do, and from make do to tragedy.

If the forests we work in start to come undone and get too hot, we’ll go contain/confine and we’ll make-do. Our budgets drop, we’ll go with three per engine instead of five and we’ll make-do.

We come up short Division [Supervisors], we’ll rob a hotshot crew of a Division and put him on the division, and we’ll make-do.

We need to quit thinking about making do.

Maybe we need to quit thinking about just doing the job, and begin thinking about doing the job right.

I used to tell my firefighters, “If we can’t do it safely, we won’t do it at all.” Mr. Williams said it much more eloquently.