Research Paper RMRS-RP-9
Fire Behavior Associated with the 1994 South Canyon Fire on Storm
King Mountain, Colorado
Research Paper RMRS-RP-9
Fire Behavior Associated with the 1994 South Canyon Fire on Storm King Mountain, Colorado
Fire Environment
This section is a short summary of the fire environment on the afternoon of July 6, 1994. First we discuss the topography. Next we discuss the condition of the vegetation or fuels in the area. Finally, we summarize the long-term and short-term weather related factors that influenced the fire behavior.
For the purposes of this study, we have attempted to use the same location names as were used in the original investigation report. However, we have identified a few previously unnamed areas. In most cases we identify the specific locations by a dominant geographical feature.
Topography
The fire occurred in a topographically complex area. The highest peak in the area, Storm King Mountain, is over 8,700 feet in elevation. The terrain south of the peak falls in a series of steep rugged broken ravines, gullies, and ridges to the Colorado River (elevation approximately 5,680 feet). Figures 4 and 5 present an aerial photograph and topographical map of the fire area. Figures 3, 6, 7, and 8 are aerial oblique photographs of the fire site from different view angles.
Main Ridge -- Up to the afternoon of July 6, 1994, the South Canyon Fire burned on the southwest side of Storm King Mountain. One of the main ridges between the mountain and the Colorado River Gorge is locally known as Hellsgate Ridge, we refer to it as the Main Ridge. The northeastern end of the Main Ridge begins as a saddle at the southern base of Storm King Mountain. The two firefighters working helitack were killed in a steep rocky chute northwest of this area. From the saddle the Main Ridge runs to the southwest for about 3,700 feet where it abruptly ends at the Colorado River Gorge.
Ignition Point -- A lightning strike to the top of the southwest-most knob of the Main Ridge (elevation 6,980 feet) ignited the South Canyon Fire. To the south from the Ignition Point the terrain falls steeply (about a 55 percent slope) to Interstate 70 and the Colorado River.
East and West Drainages -- The Main Ridge is flanked on the east and west by two north-south oriented drainages. These drainages are referred to as the East and West Drainages. Both drainages were formed by erosion of the sandstone rock and soil. Fine sand fills the bottom of the drainages with intermittent large sandstone boulders. The West Drainage follows a twisting path north from the Colorado River Gorge. The bottom of the West Drainage is a steep-sided (80 percent slope) ravine. The West Drainage channeled air from the Colorado River Gorge north to the base of Storm King Mountain. The fire changed from a backing surface fire moving down the slope to a fast spreading fire burning upslope and upcanyon through the shrub and tree canopies in the West Drainage. Surviving firefighters exited via the East Drainage. The fire burned through the East Drainage about 30 minutes after the firefighters exited it.
Lunch Spot and Spur Ridges -- Two spur ridges that are important to the narrative extend west from the Main Ridge down into the West Drainage. The southernmost ridge, referred to as the Lunch Spot Ridge, was where the firefighters who had been building fireline down the West Flank of the fire ate lunch on the afternoon of July 6 (fig. 3 through 6 and 8). Eight smokejumpers deployed and survived in fire shelters on the upper portion of the Lunch Spot Ridge during the afternoon of July 6, 1994. One smokejumper survived near the Lunch Spot without deploying a fire shelter.
A second smaller ridge, located about 1,400 feet north of the Lunch Spot Ridge, we call the Spur Ridge. The Spur Ridge serves as a point of reference when discussing firefighter, helicopter, and fire location and movement.
Helispots H-1 and H-2 -- Two helispots were constructed on the Main Ridge. The first helispot (designated H-1) was constructed on July 5 and was at the junction of the Main Ridge and Lunch Spot Ridge at about 7,000 feet elevation on an unnamed knob. It consisted of an area approximately 50 feet in diameter that was cleared of brush and trees. Stumps and ground litter remained. A second helispot (H-2) was constructed on the morning of July 6. H-2 was at 6,760 feet elevation and was about 1/4 mile north of H-1. This helispot was smaller than H-1 and was situated in a small saddle from which the brush had been cleared away.
Double Draws -- The southwest face of the Main Ridge near H-1 falls steeply (approximately 55 percent slope) dropping more than 1,000 feet in elevation to the bottom of the West Drainage. The Double Draws are on this steep west-facing slope. They consist of two steep chimneylike gullies leading into the West Drainage. They originate approximately one-third the way down the slope below H-1 and were formed by erosion of the loose soil (fig. 5 and 8). Several crown fire runs occurred south of the Double Draws shortly before the fire suddenly began burning up the West Drainage on the afternoon of July 6, 1994.
Bowl -- A small bowl (approximately 1/2 acre) is in the bottom of the West Drainage about 250 feet north of the intersection of the Double Draws and the West Drainage. This area is referred to as the Bowl. The accumulation of live and dead fuel in the Bowl contributed to the development of the fire that spread up the West Drainage.
West Bench -- Northeast of the junction of the West Drainage and the Lunch Spot Ridge is the West Bench. This area is approximately 80 feet above the bottom of the West Drainage. The West Bench is approximately 150 feet wide and 450 feet long with the long side oriented parallel to the West Drainage. While not flat (approximately 15 percent slope), the West Bench is significantly less steep than the slopes on either side of it. Gambel oak with grassy openings was the dominant vegetation on the West Bench. Gambel oak completely covered the slopes between the West Bench and the Main Ridge.
West Flank Fireline -- This fireline is approximately 2,100 feet long with the Lunch Spot at its southernmost point. From the Main Ridge, the West Flank Fireline leads west down the steep (about 55 percent) slope for 300 feet. From the base of this first steep pitch the fireline continues, at a much flatter slope (about 10 percent), another 380 feet down and across the slope to the Spur Ridge. It leads over the Spur Ridge, turns to the southwest, and follows a contour route across the West Flank slope for about 800 feet. From this point it contours 400 feet diagonally across and down the slope, leading generally southwest, and then climbs 200 feet up a 65 percent slope to the Lunch Spot. Several points of interest on the West Flank Fireline are identified by their major topographical features. Distance along the fireline from the junction of the Main Ridge and the Fireline are given in parentheses. The points are: the Zero Point (0 feet); the Fatality Site (120 to 280 feet); the Tree (200 feet); the Draw at the base of the first steep pitch in the fireline (445 feet); the Spur Ridge (680 feet); the Stump (850 feet); two other locations (1,450 and 1,880 feet) are important to the fire chronology presented later. All of these points are shown in figure 3. All distances are detailed in appendix B, table B-2.
Zero Point -- The junction of the Main Ridge and the top of the fireline constructed on the West Flank is an important geographical location when specifying specific firefighter locations and movement. For the purposes of this report, it is referred to as the Zero Point. We found it helpful to reference distances along the Main Ridge and the fireline from this point. For example, helispot H-1 is 790 feet south along the Main Ridge from the Zero Point and helispot H-2 is 560 feet north along the Main Ridge from the Zero point.
Drop Zone -- The smokejumper Drop Zone was near the Saddle on the north end of the Main Ridge (approximately 1,000 feet north of H-2).
Fuels
Vegetation in the area consisted primarily of Gambel oak (Quercus gambelii) on the north- and west-facing slopes and open, mixed pinyon-juniper (Pinus edulis and Juniperus sp.) with a cured grass understory on the south-, southwest-, and east-facing slopes. A few "pockets" of Douglas-fir (Pseudotsuga menziesii) existed in the area of the Double Draws, the Bowl, and on the lower east-facing slope of the West Drainage. Continuous cured grasses and some shrubs and small stands of pinyon-juniper covered the bottom of the north-south oriented West Drainage (Anon 1994). The Bowl contained a heavier concentration of grasses, mixed shrubs, woody debris, and trees. Figure 9 shows the approximate distribution of the vegetation over the fire site.
The Gambel oak was more than 50 years old and did not contain much dead material (South Canyon Report; Biastoch 1995). It formed a dense, green, continuous closed canopy, 6 to 12 feet tall and appeared to be unaffected by spring frosts (fig. 10). Visibility within the stand was limited. The surface fuels beneath the canopy consisted of a 3 to 6 inch layer of leaf litter (Husari 1996).
Fine dead fuel moisture content was estimated to range from 2 to 5 percent on the afternoon of July 6 (South Canyon Report). Both the large diameter (1000 hour) dead fuels and the live fuel moisture contents in the area were several weeks ahead of their historical summer drying trends (South Canyon Report). No fuel moisture content samples were collected at the time of the fire. However, live foliar moisture contents were measured on July 12, 1994, at a site near the South Canyon Fire with an aspect and elevation similar to the west flank of the Main Ridge. Foliar moisture in underburned Gambel oak was about 60 percent while that in green unburned Gambel oak was 125 percent (South Canyon Report). Typically, live fuel moisture levels change relatively slowly in undamaged shrubs. Therefore, we expect the foliar moisture content of the unburned Gambel oak on July 6 would have been similar to the values measured on July 12. No samples were collected in the pinyon-juniper fuels.
Weather
Long Term -- The South Canyon Fire Investigation report states that "weather significantly contributed to the blowup of the fire." Below normal precipitation (as compared to the 30 year average) during the winter and spring of 1994 had pushed western Colorado into a severe drought (South Canyon Report). Precipitation levels at Glenwood Springs from October 1, 1993, through July 6, 1994, were 58 percent of normal. Accompanying the below-normal precipitation were much warmer than normal temperatures through May and June. This pattern persisted into July. Fires in western Colorado during the previous weeks had exhibited rapid spread rates and long range spotting, both characteristic of drier than average conditions.
July 5, 1994 -- On this day, weak high pressure aloft and a hot, dry air mass covered western Colorado. Upper level winds were light from the southwest through 14,000 feet and then increased to 30 miles per hour at 16,000 feet. At Grand Junction, CO, the lower atmosphere was dry and unstable as indicated by a Haines Index of 6 (Haines 1988). That morning a strong cold front developed in western Idaho. The front was associated with an unseasonably cold upper level low pressure system, centered over northern Oregon. By late afternoon the cold front extended from eastern Idaho into central Nevada.
We are not aware of any weather observations taken by firefighters working on the fire. The closest remote automatic weather station (RAWS) was southwest of the airport at Rifle, CO, approximately 10 miles west of the fire site. We assume that the measurements from Rifle are representative of the conditions over the South Canyon Fire.
Between 0100 and 0600 on July 5 the relative humidity at the Rifle RAWS ranged from 10 to 40 percent. Relative humidity remained near 11 percent until after midnight (fig. 11). Winds at the Rifle RAWS were light and variable through the morning and late afternoon of July 5, 1994 (fig. 12). By 1800 on July 5, 1994, winds over the South Canyon Fire area were blowing generally from the south at 10 to 15 miles per hour with gusts to 20 miles per hour (South Canyon Report; Bell 1994).
July 6, 1994 -- At 0600 this day, the cold front extended from central Wyoming across northwest Colorado into southwestern Utah (fig. 13). Relative humidity at Rifle reached a high of 29 percent at about 0700 as compared to 40 percent the previous date (fig. 11). The lower relative humidity during the early hours of July 6 enhanced burning during the night. This is supported by witness statements that the fire was more active through the night on July 5 than on previous nights.
The front passed over Grand Junction around 1300 and caused the 10 to 15 mile per hour winds to increase to approximately 30 miles per hour (South Canyon Report). The front passed over the Rifle RAWS station between 1400 and 1500, as indicated by increased winds out of the west-southwest. They reached their peak at about 1600 with gusts to 45 miles per hour. The relative humidity at Rifle reached a minimum of 8 percent at 1700 on July 6. The winds at Rifle changed to the northwest at 1900. They gradually decreased through the evening and were blowing less than 6 miles per hour by 2300 (fig. 12).
Firefighters on the fire and at Canyon Creek Estates, a residential subdivision about 2 miles west of the fire, noted an increase in winds between 1400 and 1430 (Scholz 1995). We surmise these were prefrontal winds associated with the approaching cold front. The Accident Investigation Team meteorologist estimated that the cold front passed over the South Canyon Fire at approximately 1520 (South Canyon Report). Witness statements characterized the winds as continuing to increase in speed, reaching their peak strength shortly after 1600. The original investigation report estimated southerly winds in the West Drainage of 20 to 35 miles per hour and westerly winds blowing across the ridges of 45 miles per hour at this time. Statements made by firefighters and the helicopter pilot suggest that it is likely that gusts over 50 miles per hour occurred in the chimneys and saddles along the ridges. Smokejumpers near the Lunch Spot at 1606 estimated upslope winds of 30 to 45 miles per hour (Petrilli 1995). Winds over the fire area remained strong until 2000 (South Canyon Report).
Topography and Wind -- There is inconsistent testimony in witness statements regarding winds in the area of the fire. While the general flow was from the west, the direction and intensity of the winds varied significantly from one location to another (South Canyon Report; Shepard 1995). For example, firefighters working on the Main Ridge stated that as they cleared away brush from the fireline they could throw it into the air and the wind would carry it over the side of the ridge. At nearly the same time firefighters working on the northeast portion of the West Flank Fire Line (a location topographically sheltered from the wind) and at an exposed rocky outcropping near H-1 stated they experienced almost no wind at all. Though inconsistent with each other, these observations are consistent with the character of surface winds in complex mountainous terrain, particularly when influenced by strong, synoptic-scale weather patterns.
We feel topography was particularly influential on the surface winds the afternoon of July 6. West of the fire site the Colorado River flows through a relatively open and broad canyon. The low level westerly winds accompanying the cold front passed through this canyon. The canyon narrows dramatically about 2 miles west of the mouth of the West Drainage then opens to the east near Glenwood Springs (fig. 7 and 14). As the wind flowed from the broad canyon into the narrow Colorado River Gorge it increased in speed (fig. 15).
Several factors combined to create unusually strong upcanyon flow in the bottom of the West Drainage. First was the topography near the mouth of the West Drainage. The orientation of the ridge running from the ignition point, west to the Colorado River, acted as a "scoop" redirecting a portion of the westerly flow in the Colorado River Gorge northward up the bottom of the West Drainage (fig. 4 and 5). Second, normal daytime heating of the upper slopes in the West Drainage induced general upcanyon flow. This upcanyon flow resulted in a low pressure region at the entrance to the West Drainage, causing even more air to flow from the gorge into the drainage. Third, the increasingly strong westerly winds associated with the cold front evacuated air from the top of the West Drainage (fig. 16). This resulted in pressure-induced flow from the Colorado Gorge into the West Drainage. Some photographs and video footage taken during the blowup show smoke movement (Bell 1994) and clearly illustrate the complexity and turbulence of the surface winds shortly after 1610. High resolution meteorological modeling (see appendix C) supports the mesoscale wind analysis by the Accident Investigation Team meteorologist.
Title: Fire Environment: RMRS-RP-9 -
Fire Behavior Associated with the 1994 South Canyon Fire on Storm
King Mountain, Colorado
Publish Date: February 5, 1999
Last Update: December
22, 2005