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

Fire Behavior Discussion

The chronology presented in the previous section describes firefighter locations, movement, and actions. The scenario presented in the chronology and the fire behavior analysis presented in this section represents our best estimate of the sequence of events given the available information.

This study would be incomplete without an analysis of the physical factors that caused the change from a relatively low-intensity, slow-moving fire, backing downslope in the leaves and sticks on the ground, to a high-intensity, fast-moving fire, burning through the entire vegetation complex. In the following discussion we attempt to identify the most significant factors leading to the dramatic transition in fire behavior.

We concentrate on two events: the blowup or transition from surface fire to a fire burning through the shrub canopy, and the fire behavior in the area identified as the West Flank that resulted in the entrapment and deaths of 14 firefighters.

We identify three major factors that contributed to the blowup on the afternoon of July 6, 1994. The first factor was the presence of fire in the bottom of a steep narrow canyon. Second, strong upcanyon winds pushing the fire up the canyon. Third, the fire moving into the green (not previously underburned) Gambel oak canopy.

Fire in South End of West Drainage

The presence of fire in the West Drainage at the base of the Double Draws is important to the later fire behavior because it places fire at the bottom of a steep narrow canyon. After the original investigation report was published, various theories continued to circulate regarding the source of the fire in the West Drainage. These theories ranged from burning logs rolling down the slope to possible arson. The available evidence most strongly suggests that this fire originated from one or both of the following sources: (1) fire spreading downslope through the previous night and morning of July 6 and (2) fire brands lofted into the drainage from the crown fire runs that occurred south of the Double Draws. We discuss both.

Witnesses report that the fire remained active through the night of July 5. On the July 6 morning reconnaissance flight, smoke was visible low in the West Drainage. This condition continued through the day (Good 1996). On July 2 through 6, fire burned downslope in the surface litter beneath both Gambel oak and pinyon-juniper. The total burned area approximately doubled each day. This rate of area growth is consistent with an approximately constant rate of fire spread.

By midmorning on July 6 the fire had burned into the Double Draws and was approximately 75 percent of the way down the slope between H-1 and the bottom of the West Drainage (fig. 19). As the relative humidity dropped and the sun heated the slopes through the day, the fire continued to spread downhill. Photographs taken at the time of the crown fire reburn south of the Double Draws show smoke near the bottom of the West Drainage (fig. 22).

It is easily shown that while total burned area increased exponentially, the actual rate of spread remained remarkably constant after July 3. We evaluated fire spread through July 6 by projecting the fire spread for the day based on the fire area data from the previous days. The last measured fire perimeter before the blowup was made during the morning reconnaissance flight on July 6. Assuming continued spread at the rate exhibited during the previous 2 days, the fire would have been within 100 feet of the bottom of the West Drainage by 1600 on July 6 (fig. 38). The original accident investigators estimated a downslope fire spread rate of 70 feet per hour during the night of July 5 and early morning of July 6. Our calculations indicate a rate of spread of approximately 32 feet per hour. While, this analysis includes some uncertainty, it clearly supports the possibility that fire reached the bottom of the West Drainage by 1600 on July 6, 1994. The downhill spread, the location of the fire at midmorning, and the presence of smoke relatively low in the West Drainage make it probable that the fire reached a point in or near the bottom of the West Drainage by early afternoon.

Fire spotting occurs when burning embers are lofted into the air by the buoyant smoke column above a flame, carried by the wind, and then redistributed on the ground causing new fire starts. Short distance fire spotting occurred throughout the day on July 6, 1994, as individual trees

Figure 38—Projected fire location on afternoon of July 6, 1994, based on fire perimeter maps from previous days. The fire spread distances were estimated measuring the distance down the slope on a line running from a point midway between the ignition point and H-1 to a point in the bottom of the West Drainage just to the south of the Double Draws. A least squares linear approximation was then fitted to the data after July 3; this is represented by the heavy shaded line. All distances are increased by 14 percent (assumes 55 percent slope) to account for the actual distance down the slope. The burned area data are included in the table shown in the figure. Dates and some critical times are also shown on the horizontal axis to assist the reader in relating fire growth to the chronology.

burned and the fire made short runs. Given the wind flow patterns in the West Drainage, it is probable that a shear layer formed where the upcanyon (southerly) flow met the westerly flow blowing over the ridges (fig. 16 and 39). Smoldering and burning embers lofted into this turbulent air mass by the crown fire south of the Double Draws would have been distributed generally northward along the bottom of the West Drainage. The original investigators reported 90 to 100 percent probability of ignition based on information from area National Fire Danger Rating System Stations.

Witness statements and later interviews suggest that the attention of the smokejumpers was focused on the crown fire runs rather than the source of smoke farther down the slope. However, shortly after the crown fires south of the Double Draws, the smokejumpers saw fires starting to burn actively near the bottom of the east-facing slope across the West Drainage (Petrilli 1996). This suggests that burning embers from the crown fires may have ignited the fire in the bottom of the drainage.

Figure 39—Schematic showing interaction of westerly flow over ridgetops and northerly flow up bottom of West Drainage forming a shear layer (dashed line). The turbulence generated by this shear layer enhanced the spread of burning embers up the West Drainage and surface wind turbulence in the area of the Double Draws and the West Bench. J. Kautz, U.S. Forest Service, Missoula, MT.

While it is not possible to identify with absolute certainty the exact ignition mechanism for the fire in the bottom of the West Drainage, the evidence suggests that the fire resulted from one or a combination of the two mechanisms discussed above.

Winds Push Fire into Bowl

Relying on witness statements and fire behavior knowledge, we suspect that the area identified as the Bowl contributed to the blowup on the afternoon of July 6. Postfire investigation of the site revealed nearly complete consumption of the surface fuels in the Bowl. Scorch marks caused by increased burning on the north side of the trees in this area suggest the presence of strong upcanyon (southerly) winds during the fire. We surmise that the concentration of debris on the ground carried the fire into the crowns of the conifers in the Bowl. This increased the size and height of the convection column over the fire. Several witnesses observed the smoke column build rapidly over the area identified as the Bowl. We believe that strong vertical momentum associated with the fire in the Bowl lofted embers up and onto the slopes on both sides of the drainage (South Canyon Report). These embers ignited spot fires.

Fire Transitions to Gambel Oak Canopy

General wind direction and topography caused the fire to spread up the West Drainage. Witness statements support this. Pushed by the winds up the steep slopes, the fire burned past the junction of the Lunch Spot Ridge and West Drainage and up onto the West Bench (see fig. 4).

Although the fire area was exposed to wind on July 4, 5, and 6, the Gambel oak canopy did not sustain continuous fire spread, even in previously underburned areas. As the fire burned downslope in the litter fuel beneath the pinyon, juniper and oak canopies, it was generally sheltered from the wind by the vegetation canopy. Upslope fire spread was confined to unburned islands within the fire perimeter or initiated by ladder fuel concentrations under individual trees (for example, the tree on the West Flank Fireline that became the Stump). Significant change in fire behavior occurred only after fire burned into fine fuels at the base of steep slopes and was exposed to strong winds. Such a transition occurred on the west-facing slope south of the Double Draws where the surface fire burned into the conifer crowns spreading upslope in several high-intensity crown fire runs.

Following the crown fire runs, the fire burned in litter and cured grass fuels along the bottom of the West Drainage spreading up the steep east- and west-facing slopes and up the West Drainage past the Lunch Spot Ridge onto the south end of the West Bench. The vegetation canopy was less dense on the east-facing slopes and along the West Bench. This exposed the surface fire to the strong winds. The steep slopes and exposure to strong winds resulted in significant increases in the size of the flames and energy release rates. This resulted in ignition of the pinyon-juniper canopy on the east-facing slopes and the green Gambel oak canopy on the West Bench. The following discussion focuses on the physical mechanisms that resulted in the fire spreading into and through the live fuel canopy as a continuous fire front.

The mechanisms driving the transition from surface to crown fire are not fully understood. In general, fire in the vegetation canopy follows an increase in the amount of energy entering the canopy, or a decrease in the amount of energy necessary to ignite the complex, or both. Increased slope, wind exposure, or decreased moisture status of live or associated dead fuels may individually, or in combination, result in such transitions.

Fire spread from the surface into the vegetation canopies often occurs rapidly; however, the factors leading up to the transition may develop relatively slowly. For example, fires often burn downslope relatively slowly, but when a backing fire reaches a position where an upslope run in unburned fuels is possible, the transition from backing to a fast-moving upslope fire may happen suddenly. Another example is fire burning through an area where it is sheltered from the wind into a location where it is more exposed to wind. The increased wind exposure can lead to a sudden change in fire behavior with little or no apparent change in the environment. In both of these examples the fire burned from one area to another resulting in an abrupt change in the slope or wind exposure. Solar heating can also influence the tendency for a fire to spread into the vegetation canopy. Exposure to the sun can cause a decrease in relative humidity and subsequent decreases in dead fuel moisture levels. This effect may occur both under and within the live vegetation complex. Decreased fine dead fuel moisture reduces the amount of energy needed for ignition. This drying may occur throughout the day or, as in the examples above, be caused by the fire burning into a sun-exposed aspect. When the fire reaches an area of drier fine dead fuel, the flaming zone can increase in size and intensity, again leading to sustained combustion in the vegetation canopy.

When a fire begins burning in the vegetation canopy, the flaming zone often significantly increases in height and depth. This increase is linked to the overall increase in total burning fuel load and decrease in bulk density (mass of fuel per unit volume). Increased fuel load leads to larger flames and energy release. Decreased bulk density often results in faster fire spread rates (Catchpole and others 1998). These two factors contribute to sustained burning in the live vegetation.

Another factor contributing to fire spread in vegetation canopies is live fuel moisture content. In an effort to assess the impact of fuel moisture on the blowup of the South Canyon Fire, we compare the conditions present on the Battlement Creek and South Canyon Fires.

A high intensity fire run occurred on the Battlement Creek Fire on July 17, 1976. This fire was approximately 30 miles west of the site of the South Canyon Fire, both burned in similar terrain and vegetation. The live Gambel oak foliar moisture content at the Battlement Creek Fire was 167 percent (USDI 1976). A killing frost on June 14, 1976, followed by dry weather significantly increased the quantity of fine dead fuel in the oak canopy over historical levels. Surface winds were light and consisted of normal upslope convective flow characteristic of summertime conditions in the area. Winds aloft were 5 to 15 miles per hour from the southwest. Slopes ranged from 10 percent near the bottom to 75 percent near the ridgetop. The fire burned on slopes that were generally west-facing and fully exposed to solar heating from about 1100 (USDI 1976).

In contrast, the Gambel oak at the South Canyon Fire site was not frost damaged, and consequently the canopy did not contain an abnormally high amount of dead leaves and stems. However, low precipitation levels during the previous 8 months had pushed the area into an extreme drought. Green, nonunderburned Gambel oak vegetation was sampled on July 12, 1994, at two sites located east of the South Canyon Fire area. The sites were at a similar aspect and elevation to the area identified as the West Flank. The measured live fuel moisture contents were 125 percent. The live fuel moisture content would not have changed significantly between July 4 and July 12, 1994. The West Flank was west-facing with 10 to 60 percent slopes and was exposed to solar radiation from about midmorning. The South Canyon Fire site was exposed to strong winds on July 4 and 5, 1994, and for some time prior to and during the blowup on July 6.

Both the Battlement and the South Canyon Fires experienced similar flame sizes, energy release, and spread rates. Fire reaching the base of steep slopes was the triggering mechanism to ignition of the canopy at the Battlement Creek Fire. Large quantities of dead matter in the otherwise relatively high live moisture canopy contributed to fire spread into the canopy on the steep slopes. Strong winds were not a contributing factor. In contrast, the transition in fire behavior on July 6, 1994, on the South Canyon Fire can be linked to strong winds pushing the surface fire into fuels of sufficient quantity that the green Gambel oak began burning. Sustained fire spread through the green Gambel oak canopy was supported by steep slopes, wind, and moderately low live fuel moisture. Crown fire spread continued with reportedly much reduced windspeeds on the steep slopes of the West Flank Fireline. Once the canopy was ignited, the increase in energy release rates substantially contributed to continued crowning on both the Battlement Creek and the South Canyon Fires. It was only after the fire began burning in the nonunderburned green Gambel oak that it spread into the previously underburned Gambel oak.

Conclusions based on only two samples cannot be considered definitive. But this comparison suggests that sudden transitions from surface fire to fire in live vegetation canopies can be linked to a combination of factors, including but not limited to: live and dead vegetation moisture content, the spatial distribution and quantity of live and dead components in the canopy, exposure to wind, fire site aspect and slope, and the intensity of an initiating fire burning within, adjacent to, or under the vegetation canopy. Not all of these factors are necessary for a surface fire to spread into the vegetation canopy.

<<< continue reading—Fire Behavior at South Canyon Fire, Fire Behavior Discussion: West Flank>>>


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