Colorado Firecamp - wildfire training wildland firefighter training Wildfire Blog Engine Boss Apprenticeship Location and Facility About Colorado Firecamp Frequently Asked Questions
Colorado Firecamp - wildland firefighter training

South Canyon Fire

William Teie Report to OSHA — 1994

6 Minutes for Safety — 2009

Fire Behavior Report, 1998

Cover & Dedication

Executive Summary & About the Authors

Preface & Contents


Fire Behavior Overview

Fire Environment

Fire Chronology

Fire Behavior Discussion



Appendix A

Appendix B

Appendix C

Report of the South Canyon Fire Accident Investigation Team, August 17, 1994

USFS shield logoFire Behavior Associated with the 1994 South Canyon Fire on Storm King Mountain, Colorado


The South Canyon Fire tragically demonstrates the fire behavior that can occur given the appropriate combination of weather, topography, fuels, and fire. One of our most frustrating observations was that while fire behavior during the afternoon of July 6, 1994, can be characterized as extreme, it was normal given the environmental conditions. The fire was a direct consequence of the fuel, weather, and topographical factors. Similar alignments of fire environment factors and the resulting fire behavior are not uncommon.

We have summarized the weather, fuel, and topographical information, and presented a brief overview of the firefighter movements relative to the fire. Using the fire environment analysis and chronology, we presented what we believe are the most likely fire behavior scenarios.

Until the afternoon of July 6, the high intensity fire behavior consisted of occasional short duration torching of individual trees and narrow uphill runs in the fire’s interior. At approximately 1600 the wind, slope, and fire location combined to result in a dramatic transition wherein the fire began burning through the live fuel canopy as a continuous flaming front. This report focused on the mechanisms that led to the dramatic transition from a low intensity ground fire to a fire that was burning through the vegetation canopy. The second part of the discussion addressed the specific fire behavior that led to the entrapment of 14 firefighters.

We have drawn a number of discussion points from the analysis. Some of these points will be readily apparent to firefighters. Others may be less evident. We believe that all are important.

  • Topography can dramatically influence local wind patterns. Surface winds in mountainous terrain are highly variable. Areas of low or calm winds can exist while other areas experience dramatically different wind direction, windspeed, or both. These changes can occur without visible warning across relatively short distances, especially when the area is subjected to large-scale weather influences such as frontal passages. Wind information is needed from multiple locations around the fire perimeter. This information should be communicated to all fire personnel.

  • Vegetation and topography can reduce a firefighter’s ability to see a fire or other influencing factors. Complex topography and dense shrub or tree canopies can restrict the ability of firefighters to sense, visually or otherwise, changes in wind, fire behavior, and fire location. This emphasizes the need for adequate observers and lookouts.

  • Current and past fire behavior often does not indicate the potential fire behavior that could occur. Maximum possible fire spread, flame heights, and energy release are determined by comparing present and near-term future fuel, weather, and topography to their state during past demonstrations of extreme fire behavior, possibly at other fires. The fire behavior exhibited by the South Canyon Fire from July 2 through the morning of July 6 consisted of low-intensity downslope spread. At no time previous to the afternoon of July 6 did it exhibit continuous high-intensity burning in the Gambel oak canopy. Previous experience and observations of high intensity fire behavior and training in fire environment assessment are tools that can be used to anticipate potential fire behavior.

  • The longer a fire burns and the larger it gets, the greater the likelihood of high-intensity fire behavior at some location around the perimeter. Not always is a fire ignited in an ideal location for high-intensity burning. However given sufficient time, a low-intensity fire will often reach a position where fuel, weather and terrain combine synergistically to produce high-intensity fire behavior. While we have not evaluated such probabilities analytically, intuitively it seems that the greater the range of conditions, the greater the potential for a rapid change in fire behavior at some location along the fire perimeter. Such was the case for the South Canyon Fire on July 6. Knowledge about the location of a fire perimeter is necessary to adequately assess fire potential.

  • The transition from a slow-spreading, low-intensity fire to a fast-moving, high-intensity fire often occurs rapidly. This seems to surprise firefighters most often in live fuels, possibly because green vegetation is associated with reduced ignition risk. We do not fully understand the exact mechanisms triggering these transitions. But observations of past fire behavior indicate that such transitions often occur when there are significant changes in windspeed or wind direction, fire location (that is at the top of the slope versus the base of the slope), or in the quantity of live and dead components in the vegetation canopy. Live green vegetation can support and even promote high-intensity burning. Monitoring fire position relative to alignment of wind, slope, and live and dead fuels can assist firefighters in recognizing potentially hazardous fire behavior.

  • Escape route transit time is related to both topography and route length. Escape routes should be considered in relation to potential maximum-intensity fire behavior rather than past or present fire behavior. The ideal escape route includes a downhill direction over the shortest possible distance to the safety zone, thereby maximizing firefighter travel rates while minimizing fire spread rates.

  • The underburned Gambel oak was significant in that it did not provide a safety zone. The blowup did not occur in the previously underburned Gambel oak. The blowup occurred when the fire changed from a relatively low-intensity surface fire to a high-intensity fire burning in the canopy of the green (nonunderburned) Gambel oak near the bottom of the West Drainage. As the fire burned up the slopes from the nonunderburned into the previously underburned Gambel oak, energy release rates and spread rates would have increased over their already high values (appendix B, table B-8). Thus, the underburned Gambel oak caused increased fire intensity as the fire burned above the West Flank Fireline, but it was not a factor for the burning below the fireline and therefore did not contribute significantly to the entrapment of the 14 firefighters. However, the amount of unburned vegetation remaining above the West Flank Fireline precluded it from being an adequate safety zone. Firefighters do not have “one foot in the black” when working adjacent to underburned shrub vegetation.

  • Smoke can significantly reduce the firefighter’s abilities to sense changes in fire behavior. This is certainly realized by any firefighter who has had to breath smoke. There is some evidence that the area around the fatality site was quickly engulfed in smoke just prior to or during the burnover. The lack of a clear view of the fire and the loud noise created by the fire would have prevented the group from fully sensing how fast the fire was closing the gap between it and them and may have distracted the group from their objective of reaching safety. Their likely inability to identify the location of the fire is one explanation for the evidence suggesting that the group was surprised by the fire and did not realize how close the fire was or where to go to escape it. Lookouts positioned outside the burn area or overhead can communicate urgency and help firefighters identify the best escape routes relative to the fire position, direction, and rate of spread.

Our analysis emphasizes the often dramatic changes in fire behavior that can occur when fire is exposed to steep slopes, winds, and relatively continuous fuels. Perhaps even more important is the observation that not all of these factors are needed, rather only one or two are needed for a blowup to occur. None of the findings and observations discussed in this study represent new breakthroughs in wildland fire behavior understanding. Rather the findings support the need for increased understanding of the relations between the fire environment and fire behavior. We can also conclude that fire managers must continue to monitor and assess both present fire behavior and potential future fire behavior given the possible range of environmental factors.

During the review process, some of the reviewers commented that they were left with a feeling of “so what” after reading the manuscript. In fact, this is one of the points that can and should be drawn from our study of the fire. While relatively high-intensity fire behavior was demonstrated, it was normal and even ordinary behavior given the combination of environmental factors. Tragically, what was not normal or ordinary was that 14 firefighters were caught in the middle of the fire and could not escape.

As a last note to the readers, we want to say that the most difficult task in this whole process has been achieving a balance between analysis, calculations, and extrapolations on one end of the scale, and heartache, feelings of loss, and even anger on the other end. Peace of mind, if at all possible, can come only by accepting the fact that humans are part of nature, yet understand just parts of it and master even less. Our efforts were directed toward increased understanding, with the hope that the knowledge gained will help to avert similar future incidents.

Figure 42—Panorama of the South Canyon Fire site taken 2 years after the fire from the ridge directly west across the West Drainage from the West Flank
(Photo is a composite of two photographs taken by T. Putnam).

©2014 Colorado Firecamp, Inc. home scheduleblogENGBfacilityabout usFAQ's