Local Colorado newspaper investigates management of Waldo Canyon Fire

Waldo Canyon Fire, June 26, 2012
Waldo Canyon Fire, June 26, 2012, the day hundreds of homes burned in Colorado Springs. Credit: Keystoneridin

The Colorado Springs Independent conducted an exhaustive study into the management of the Waldo Canyon Fire, which in June, 2012 killed two people, destroyed 346 homes, and burned 18,247 acres — with some of those acres and most of the homes being within the city limits of Colorado Springs. The results of their investigation, published in a lengthy article on December 12, are extremely interesting. More about that later.

On October 23, the City of Colorado Springs released what they called an “Initial After Action Report”, but in the document they admitted the AAR was not the final, comprehensive report on the fire, but was considered preliminary. An in-depth analysis, they said, would occur over the next several months to fully explore Colorado Spring’s management of the fire.

The report, as we wrote then, was unusual. It listed some strengths and recommendations, but omitted information about the issues that caused the recommendations. Therefore, it was not always clear WHY the recommendations were made, forcing an observer to read between the lines. This limited the opportunities for lessons learned and may not in all cases have the desired result of preventing mistakes. There was no indication that they consulted experts outside the city for unbiased opinions or recommendations.

In reviewing the recommendations in the ARR, it appeared that many of the issues would be mitigated with adequate training and experience in the Incident Command System.

After reading the article in yesterday’s Colorado Springs Independent, I am left stunned. Regarding the management of the fire within the city of Colorado Springs, I have never heard of a wildland fire with such a huge impact that was so utterly, catastrophically mismanaged.

The fire started Saturday, June 23, 2012. A Type 1 Incident Management Team, Great Basin Team 2, with Incident Commander Rich Harvey, assumed command Monday morning, June 25. Except, according to the article:

[Colorado Springs FD Fire Chief Rich] Brown said he specified in the city’s delegation of authority that the city, and no one else, would have control if the fire crossed into Colorado Springs.

The newspaper examined hundreds of documents including reports written by firefighters working on the fire. The article is astounding. If it is correct, and I have no reason to believe it is not, it exposes complete failures in pre-incident planning, qualification and training of fire department personnel, evacuation planning and execution, logistics, daily incident planning, strategy, and tactics. Apparently this large, modern city with an extensive, very vulnerable wildland-urban interface was completely unprepared to manage a large wildland fire or evacuations.

The article does not criticize firefighters. It points out the failures in preparedness and management of the fire by upper level officials, before, during, and after the incident. Some mid-level and upper level firefighters were forced into positions of great responsibility on the fire without having been provided the necessary training and experience. That was not their fault, it was just the way it was done in Colorado Springs.

You should read the article. It’s long, but worth it from a lessons learned perspective. The comments at the bottom of the article are interesting as well.

Here are a few excerpts from the article, copied and assembled into bullets:

  • When the fire swept into Mountain Shadows, the city had a mere four firefighting vehicles, or apparatus, assigned to that subdivision and all other land north to the Air Force Academy.
  • The evacuation plan had been drafted only that morning, and was enacted minutes before the first homes burned.
  • Local firefighters found themselves outgunned, and much of the help from other fire departments was nowhere close, because leaders sought those resources only after flames came into the city. Their chief staging area wasn’t set up and equipped until houses were ablaze, and they didn’t have a mobile command post until eight hours into Tuesday’s firefight [when most of the homes burned].
  • And, as readily admitted by city firefighters leading efforts on the ground that night, the fire could have charged further eastward for miles had it not been for the unanticipated arrival of U.S. Forest Service engines and their hot shot crews.
  • [A Colorado Springs FD Captain], a heavy rescue expert with no current wildland certification, was in charge of the city’s deployed resources on Tuesday [when most of the homes burned].
  • And because the department hadn’t made maps in advance for out-of-town engines, crew’s like Company Officer P.J. Langmaid’s were told to “split all our companies and place one member on each Denver Fire rig as a point of contact and communications/operations liaison.”
  • The next day, the city did sign its own delegation of authority — but apparently made it clear its fire personnel would remain separate from Harvey’s [Type 1 Incident Management] team, which arrived Sunday night. In a July 16 interview, [Colorado Springs FD Fire Chief Rich] Brown said he specified in the city’s delegation of authority that the city, and no one else, would have control if the fire crossed into Colorado Springs.
  • As adamant as the city was that its personnel operate independently during the fire, it looks equally adamant now that its personnel alone examine the response. No third party has been engaged to review the city’s performance, which often is requested when a fire kills people.

The map below shows the progression of the Waldo Canyon Fire. Colorado Springs is on the eastern edge below the blue box that says “July 1-6”.
Waldo Canyon Fire progression map

The Colorado Springs Independent, and the reporter, Pam Zubeck, deserve a great deal of praise for their investigation and the writing of this excellent article. Pulitzer Prize-worthy maybe?

Differences between military and Forest Service accident investigations

The accident report on the fatal crash of the military C-130 MAFFS air tanker which was released yesterday illustrated one very important difference between accident investigations conducted by the military and the U.S. Forest Service. A notice on page two of the report points out that the findings of military aviation accident investigations are regulated by law, 10 U.S.C. 2254(d), which states:

Use of Information in Civil Proceedings.—For purposes of any civil or criminal proceeding arising from an aircraft accident, any opinion of the accident investigators as to the cause of, or the factors contributing to, the accident set forth in the accident investigation report may not be considered as evidence in such proceeding, nor may such information be considered an admission of liability by the United States or by any person referred to in those conclusions or statements.

C-130 MAFFS crash, July 1, 2012
C-130 MAFFS air tanker crash, July 1, 2012. US Air Force photo

For fatal wildfire burnovers or entrapments of U.S. Forest Service employees, a law provides for just the opposite, thanks to a bill that was sponsored by Senator Maria Cantwell and U.S. Representative Doc Hastings, which became Public Law 107-203 in 2002:

In the case of each fatality of an officer or employee of the Forest Service that occurs due to wildfire entrapment or burnover, the Inspector General of the Department of Agriculture shall conduct an investigation of the fatality. The investigation shall not rely on, and shall be completely independent of, any investigation of the fatality that is conducted by the Forest Service.

The Cantwell-Hastings bill that was signed into law in 2002 was a knee-jerk reaction to the fatalities on the Thirtymile fire the previous year. The Department of Agriculture’s Inspector General’s office had no experience or training in the suppression or investigation of wildland fires. They are much more likely to be investigating violations at a chicken ranch than evaluating fire behavior and tactical decisions at a wildfire. The goal of the Inspector General investigation would be to determine if any crimes were committed, so that a firefighter could be charged and possibly sent to prison.

After the trainee wildland fire investigator for the OIG finished looking at the Thirtymile fire, on January 30, 2007 the crew boss of the four firefighters that died was charged with 11 felonies, including four counts of manslaughter. The charges were later reduced to two counts of making false statements to which the crew boss pleaded guilty on August 20, 2008. He was sentenced to three years of probation and 90 days of work release.

The criminal charges brought against the firefighter who may or may not have made some mistakes on the fire had a serious, chilling effect on wildland firefighters. Not only does it make them reluctant to speak to anyone about what happened on an accident, some even had second thoughts about their willingness to continue working in a professional they loved because potential criminal charges or convictions could ruin their lives and the livelihood of their families.

In addition, firefighters lawyering-up after an accident makes it difficult to discover the causes of an accident and to learn lessons which could save lives by preventing similar fatalities.

The four-fatality MAFFS accident was a complex chain of events involving many individuals and firefighting resources. But in spite of the complexity, the report was released to the public only four months after the accident, making it possible for lessons to be learned while reducing the chances of a similar accident taking more lives.

This short turnaround is unheard of in the wildland fire agencies in part due to the potential civil and criminal implications down the road.

This is literally a life and death issue — Senator Maria Cantwell’s and Representative Doc Hastings’ hastily conceived Public Law 107-203 must be repealed and replaced by one similar to 10 U.S.C. 2254(d), which serves the military very well. The Cantwell-Hastings law serves no useful purpose. Accidents are investigated, with or without the ridiculous law. It had unintended consequences and needs to be fixed.

Report released on Colorado Springs’ Waldo Canyon Fire

Waldo Canyon After Action ReportAn After Action Report (AAR) was released today about the Waldo Canyon Fire that burned into Colorado Springs June 26, destroying 345 homes and ultimately blackening over 18,000 acres. This AAR is not the final, comprehensive report on the fire, but is considered preliminary. An in-depth analysis will occur over the next several months to fully explore Colorado Spring’s management of the fire.

The organization of the report is a little unusual, listing strengths and recommendations, but omitting many of the issues that caused the recommendations. Therefore, it is not always clear WHY the recommendations were made, forcing an observer to read between the lines. This limits the opportunities for lessons learned and may not in all cases have the desired result of preventing mistakes. Maybe the final report will fill in these gaps.

The report listed several areas identified as “major strengths”, including interagency cooperation, dedicated personnel, pre-incident training and exercises, planning, and the fact that they saved 82 percent of the homes in the direct impact area.

Some of the recommendations:

  • A system needs to be designed to provide immediate notification to first responders and key agency representatives as decisions are made.
  • Real-time documentation. Use “scribes” to track real-time information for record keeping and serve as a communication link between locations when primary staff is busy with their duties.
  • Train staff and volunteers who can serve in the Emergency Operations Center to staff a more robust Logistics Section.
  • Provide additional training on the use of the Incident Command System (ICS).
  • A Communications Unit Leader should be assigned to ensure that an incident-wide Communications Plan is developed.
  • Exercise the numerous emergency management plans. The report listed seven of them.
  • For the Emergency Operations Center, develop an organization chart early in the incident, and train the personnel on their roles and how to interface with the Incident Command Post.
  • Develop a plan on how to provide adequate food to incident personnel.
  • Establish procedures for handling large quantities of donated food and water. Provide incident management training for non-profits and agency personnel to improve management of volunteers and donations.
  • Develop a plan to ensure incident personnel work consistent shifts and receive adequate rest, breaks, and rehabilitation (food and supplies).
  • Evaluate the need for post-incident critical stress debriefings.
  • Develop an ICS organization chart to ensure that a Safety Officer and Accountability Officer are assigned.
  • Ensure that span of control policies are followed.
  • Utilize Staging Areas to assist in accountability of personnel and resources.
  • Use street names rather than neighborhood names to define evacuation boundaries.
  • Provide maps of areas that are being evacuated to first responders.

In reviewing the recommendations in the ARR, it appears that many of them would be mitigated with adequate training and experience in the Incident Command System, NIMS, or NIIMS. A Type 1 Incident Management Team assumed command of the fire at the end of Day 2, June 24; most of the homes were destroyed on June 26. The team was no doubt very fluent in the use of ICS, but if they were interfacing with multiple agencies who had limited knowledge in the management system, there could have been some inefficiencies and a lack of adherence to ICS protocols.

You can download the entire 1.7 MB After Action Report.

HERE is a link to articles on Wildfire Today that mention both “Waldo Canyon” and “Colorado Springs”.

Reviews of Pagami Creek Fire, and FLA for canoe entrapments

The U.S. Forest Service has released two additional reports about last year’s Pagami Creek Fire which was managed, rather than suppressed, for 25 days, until it ran 16 miles on September 12, eventually consuming over 92,000 acres of the Boundary Waters Canoe Area Wilderness in Minnesota. We also remind you of the facilitated learning analysis of the eight USFS employees caught out in front of the fire in canoes.

Policy review

The objective of one of the reviews was to determine if the major decisions made by the incident management teams and the staff of the Superior National Forest were consistent with official USFS policy. The review was conducted by one person, Tom Zimmerman, a program manager for the USFS’ Wildland Fire Management Research, Development, and Application Program in Boise. Mr. Zimmerman analyzed the decisions and compared them with 21 policy statements, manuals, directives, and Forest level planning documents. He concluded that the decisions “appear consistent with all levels of policy and process direction”.

Decisions review

There was another review, “looking at decisions made by line officers and Incident Management teams based on the Delegation of Authority from the Forest Supervisor”. The individuals involved in this review were Jim Thomas, Fire and Emergency Operation Specialist for the Eastern Region of the USFS, and Jim Bertelsen, a Superior NF employee acting in his capacity as President of local NFFE Union 2138. This review also found no fault with how the fire was managed, saying no information was overlooked that would have predicted the unprecedented movement of the fire on September 12.

While we don’t dispute the qualifications of Mr. Zimmerman and Mr. Thomas, a person has to wonder if these internal reviews, each conducted basically by one person, would have reached different conclusions had they been completed by a panel of neutral subject matter experts.

Entrapment and near-miss facilitated learning analysis

Pagami fire shelters
Deployed fire shelters on the Pagami fire. USFS photo from the facilitated learning analysis.

In addition to those two reviews, released earlier was an excellent facilitated learning analysis (FLA) of the near misses and entrapments of eight USFS employees who were caught out in front of the rapidly spreading fire in canoes while they were trying to evacuate the recreating public from the area. At one point when they were fleeing the fire, the smoke was so thick they could not see the fronts of their canoes. Two people left a canoe and took refuge in the cold water, deploying a single fire shelter over their heads as they floated, suspended by their life jackets. Two others were flown out at the last minute by a float plane when it somehow found a hole in the smoke and was able to find them and land on the lake. Four people, after paddling furiously in the strong winds, dense smoke, and darkness, unable to find a fire shelter deployment site on the heavily forested islands, finally found a small, one-eighth acre barren island where they climbed inside their shelters as they were being pounded with burning embers.

The very well done FLA is a must read. Someone should make a movie about this.

 

Thanks go out to Dick

Clearing the Air: Perspectives on the Large Cost Fire Review

Below is a guest post written by Mike DeGrosky, the CEO of Guidance Group, Inc.

================================================

I want to thank Bill for allowing me to blog as a guest at Wildfire Today.

In 2010, my company, Guidance Group, Inc. coordinated the work of the Secretary of Agriculture’s Independent Large Cost Fire Review Panel, which reviewed the six fiscal year 2009 wildland fires whose suppression costs exceeded $10 million. The six fires included the Backbone, Big Meadow, Knight, La Brea, and Station fires in California and the Williams Creek fire in Oregon.

Phil Schaenman (of TriData Corporation) and I presented the Panel’s final report in a briefing to the U.S. Forest Service on August 13, 2010 followed by a briefing with the Secretary of Agriculture’s Chief of Staff later that week.  Apparently, sometime after these briefings, but before the Departments of Agriculture and Interior had completed their review and transmitted the report to Congress, someone, who remains unknown, leaked the report.  The report found its way to a group of Forest Service retirees as well as Los Angeles Times reporter Paul Pringle, both who were critics of the Angeles National Forest’s handling of the Station Fire.

Large Fire Cost ReviewWhen excerpts from the report began showing up in LA Times articles critical of the Forest Service and in Congressional panel hearings, a commenter on Wildfire Today accused me of leaking our own report.  Not only was this accusation false, this person offered neither justification for their accusation or evidence to support it.  In fact, it would not have been in our interest to leak the report and endanger our reputations and working relationship with the Forest Service, but you never know why people get the ideas they do.  At the time, in deference to the Forest Service and their review process, I felt it best to say little.  However, now that the report is out in the public domain, I would like to clear the air.

In reality, we first became aware that the report had found its way outside agency circles in early October, when I received a call from a member of the Forest Service retirees’ group challenging the Angeles Forest’s handling of the Station Fire.  The caller complimented our work, commended the report, and asked me to verify its authenticity.  When I enquired as to how he had come to be in possession of the report, he told me that the group had received the report “from a contractor’s association.”  I can only speculate as to how it found its way to Paul Pringle at the LA Times.

It is interesting to note that Mr. Pringle never contacted either the panelists or I.  I would have loved the opportunity to help him put his story in context.  Interestingly, the members of Congress who conducted panel hearings on the Station Fire in October never contacted us either, nor have the various organizations investigating the Forest Service’s action regarding the Station Fire. That is, in part, my purpose for my entry on Bill’s blog.  Those who want to understand how the 2009 Large Fire Cost Review does (and does not) relate to the Station Fire need to know a few things that have gotten lost as the controversy took on a life of its own..

First, having the report in the public sphere was not troubling to us.  We are proud of our work and do not fear public scrutiny of it.  However, a few people focused on a single paragraph taken from a six-page section discussing the Station Fire.  Wanting it to support their point of view, they stretched a few phrases beyond their intended purpose and presented these passages outside the context in which the Panel made them.  For example, the Panel’s findings included the following:

Incident Management – The Station fire represented a very large, complex incident, in rugged terrain, involving multiple jurisdictions at the edge of the City of Los Angeles. Fire personnel faced extraordinary challenges. However, the agency personnel, including agency administrators who were actively engaged, handled the situation as well as one might expect given the circumstances. The fact that the IMT came from southern California and had experience with this type of high profile fire proved advantageous” (p. 26).

and:

Initial Response – Controversy continues over whether Forest personnel could have stopped the fire on the morning of August 27 (day 2). Critics claim that if the Forest had airtankers and heavy helicopters on station over the incident at first light, they may have stopped the fire’s spread. If true, more than $90M in cost could have been avoided. However, the Forest Service, Los Angeles County, and CAL FIRE jointly reviewed the initial and extended attack. Their report, issued on November 13, 2009, found that the initial attack ICs acted appropriately and made prudent decisions regarding the safety of firefighters, including those involved in air operations. Further, the report determined that aggressive air operations in the early daylight hours of day 2, without necessary ground support, would not have been effective. The matter remains under investigation and, therefore, beyond the scope of this Panels’ review” (p. 26).

In short, the Panel was largely complimentary of the Forest Service’s incident management under nightmare conditions and, more importantly, the Large Fire Cost Review for FY2009 purposefully avoided the initial or extended attack of the Station Fire.  However, these facts remained largely unreported.

Unfortunately, the Panel’s report included an unintended choice of words, causing confusion.  Citing factors that increased fire costs, in referencing troubles with ordering federal resources, the report described how, in early 2009, the Regional Forester issued a letter providing budget guidance for the region’s fire preparedness funds.  In the course of our fieldwork, it became obvious that field personnel interpreted the letter to mean that the Forests should order Forest Service personnel and equipment before ordering state or local resources; and that this interpretation had delayed, on occasion, the arrival of critical resources.  As an example, the report recounted a situation in which the nearby Morris fire released a strike team of CAL FIRE engines who returned to San Diego while an order for a Federal strike team of engines for the Station fire remained unfilled.

Unfortunately, we inadvertently included the word “initially” in the description of events, leading some to believe that this example had bearing on the controversy concerning the extended attack of the Station Fire, despite the Panel’s statement that the initial attack of the Station Fire was beyond its scope.  Some even called it the “smoking gun” that they had been seeking.

In reality, this strike team (and another that was reassigned) were released on August 29th, three days after the start of the Station Fire, not during initial attack.  However, it is important to note that, way back in October, as the Paul Pringle referenced report passages in the LA times and the report came up in Congressional panel hearings, we acknowledged to the Forest Service that this section of text described resource orders made early in the fire, but not during initial or extended attack.  I still contend that other text in the Station Fire section of the report made that context clear.  Inclusion of the word “initially” was inadvertent, and the Panel was aware that the situation occurred days after the fire’s start.

I am pleased that the report is finally out in the public eye, where people can read it for themselves rather than speculating on its contents or allowing others to interpret it for them.  I hope that these remarks clarify the relationship between the Large Fire Cost Review for FY2009 and the controversy surrounding the Station Fire.