CAIC: Colorado Avalanche Information Center

2014/02/15 - Colorado - Star Mountain, near Twin Lakes

Published 2014/03/05 by Ethan Greene - Colorado Avalanche Information Center

Avalanche Details

  • Location: Star Mountain, near Twin Lakes
  • State: Colorado
  • Date: 2014/02/15 (Estimated)
  • Summary Description: 5 skiers caught, 3 injured, 2 killed
  • Primary Activity: Backcountry Tourer
  • Primary Travel Mode: Ski
  • Location Setting: Backcountry


  • Caught: 5
  • Partially Buried, Non-Critical: 2
  • Partially Buried, Critical: 0
  • Fully Buried: 3
  • Injured: 3
  • Killed: 2


  • Type: HS
  • Trigger: AS - Skier
  • Trigger (subcode): u - An unintentional release
  • Size - Relative to Path: R4
  • Size - Destructive Force: D4
  • Sliding Surface: O - Within Old Snow


  • Slope Aspect: NE
  • Site Elevation: 11606 ft
  • Slope Angle: 40 °
  • Slope Characteristic: Planar Slope

Avalanche Comments

The avalanche released in the most eastern path on the north side of Star Mountain. This group of avalanche paths threatens Colorado Highway 82. In the avalanche atlas for the area, the Star Mountain paths are named A through F, starting on the east side and moving to the west (Star Mountain F does not impact the highway). 

The avalanche released the vast majority of the Star Mountain A path. It was a hard-slab avalanche, unintentionally triggered by a skier, very large relative to what the avalanche path could produce, large enough to destroy a railway car or a substantial amount of forest, and broke into old snow layers (HS-ASu-R4D4-O). The force of the avalanche debris threw large chunks of ice 50 feet uphill as it ran into Lake Creek. The air blast from the avalanche snapped off trees approximately 8 inches in diameter. 


Weather Summary

A dry period during the middle of January ended on the 28th with sustained snowfall through February 4th. A second period of sustained snowfall began on February 8th and continued through the 12th. During this time period, the Snow-Water Equivalent at the Brumley SNOTEL site increased by 3.7". Between January 28th and February 15th, the Ski Cooper ski area reported 79" of new snow and their total snow height rose from 50" to 92". The only site that records wind in the Lake Creek drainage was not operating at the time of the accident, but a site near Cottonwood Pass (17 miles south) recorded westerly winds in the 5 to 20 mph range for two days proceeding the accident.         

Snowpack Summary

Winter came early to the Lake Creek drainage, with the first significant snowfall occurring during the first week of October. The Brumley SNOTEL site shows continuous snow coverage beginning on October 30th. Dry periods in mid November and mid December caused the thin snowpack in the area to metamorphose into a fragile layer of depth hoar. Although there was close to 80 cm of snow in the upper portions of Star Mountain A in mid December, the lower layers were quite weak with snowpit tests producing propagating cracks and Sudden Collapse failures. Snow accumulation in late December and early January did little to affect the lower layers, but continued to build a slab on top of the very weak basal layer. The two snow events (ending February 4th and 12th) each produced significant avalanche cycles. Avalanches caused significant tree damage in the area between Loveland and Fremont Passes and destroyed a mining-era structure near the town of Montezuma. The Monarch Ski Patrol and CAIC/CDOT crews reported several very large (D3) natural and triggered avalanches in the Monarch Pass area. There were several snowmobile avalanche involvements during this time period.  

On the day after the avalanche accident, CAIC investigators found a thick slab resting on very weak layers of depth hoar. The slab in the likely trigger area was 60 to 80 cm thick and composed of mostly rounding snow grains. In this area there was a 4 cm layer of Fist hard depth hoar, 2 mm in size, above a decomposing crust. Extended Column Tests on this layer produced cracks that propagated across the column during taps 20-30, producing a Sudden Collapse fracture or Q1 shear. The avalanche broke into thick areas of snow that were Pencil hard for several meters. The depth hoar in these areas was 1 Finger hard, and the fracture traveled along the top of the depth hoar layer. In portions of the bed surface, the avalanche broke down to the ground, ether removing all of the snow or forming a secondary crown.           

Events Leading to the Avalanche

A group of 7 traveled to the La Plata the trail head on Saturday February 15th, 2014. The group departed from Colorado Highway 82 at about 10:30 AM and followed the trail up the South Fork of Lake Creek. They stopped to check the function of their avalanche transceivers and to practice searching with the devices. They continued following the trail and observed several whumpfs. They turned west and left the trail, ascending southeast-facing terrain to gain the northeast ridge of Star Mountain. The group continued up the ridgeline, stopping in a low-angle section at about 11,700'. This placed the group at the top of the eastern portion of the Star Mountain A avalanche path at about 3:00 PM.

The group decided to dig a snowpit to examine and test the layers in the snowpack. They selected an area near the top of the slope and had two people dig the pit while the others watched from a low-angle portion of the slope. They rotated all of the group members through the snowpit in teams of two so everyone could examine the snowpack. They conducted 8 or 9 Compression Tests. The results were all in the Moderate to Hard categories. They described the results as consistently a "collapse near the ground". They also did a Rutchblock-like test with no results.

The group of 7 decided to descend the avalanche path and to implement four safety measures. 1) They would split into two groups for their descent. 2) The riders in each group would depart in 20 second intervals. 3) They would stay in treed areas and avoid open portions of the slope. They discussed moving downhill and to the skier's right (east) whenever possible. 4) They would regroup in "safe areas" and to keep track of each other's progress.

The first group of four riders left the ridgeline in 20-second intervals. They traveled downhill and to the east, regrouping in two locations, and descending about 400 vertical feet. They signaled to the second group to begin their descent. The three riders left the ridgeline in 20-second intervals. This group began their descent. All three were moving downhill when the avalanche released. 

Accident Summary

Note: The riders are numbered 1 through 7 in the order they left the ridgeline. The first group includes Riders 1 through 4. The second group includes Riders 5 through 7.

Riders 1 through 4 were stopped and within about 20 feet of each other when the avalanche released. They heard yells from above and saw the avalanche move down the hill to their left (the gully in the eastern portion of the Star Mountain A path) before the moving snow washed over them. Riders 2 and 4 did not travel very far down hill, but were both completely buried. Rider 3 traveled about 400 vertical feet through trees and stopped above a steep, rocky portion of the avalanche path. He was partially buried and injured. Rider 1 traveled about 500 vertical feet, through trees and over the rocky area, stopping completely buried in the debris.  

Rider 5 (lead skier in the second group and likely trigger of the avalanche) made about five turns before he saw a cloud of snow in front of him and saw the snow crumble around him. He became enveloped in the flow, which pushed him into several trees, one of which snapped under the force of the avalanche. He traveled less than 300 feet vertically and stopped buried to his waist in debris and injured.

Riders 6 and 7 were skiing when they saw large blocks of snow moving down hill to their right. They stopped and saw more large blocks of snow moving down hill to their left. They yelled "Avalanche! Ski to the right!" as they watched the slide descend to the valley floor. They were left standing on a peninsula in the crown, with the fracture line extending uphill of them on both sides. 

Rescue Summary

Riders 6 and 7 continued to move downhill, yelling for other members of the group. They made voice contact with Rider 5. He had dug himself out of the snow and was injured and had lost both skis and poles, but he was able to participate in the rescue. They made voice contact with Riders 2 and 3. Riders 5, 6 and 7 turned their transceivers to receive and headed downhill.

They reached Rider 2 first. He had been completely buried, but had his head and one arm out of the snow when the others arrived. He was wrapped around a tree and in a lot of pain. Rider 6 dug him out of the snow as Riders 5 and 7 continued downhill to Rider 3. When Rider 2 was free, he turned his transceiver to receive. Rider 6 got a signal on her transceiver with a distance under 10 m. She conducted a fine search, but was not able to get a distance reading under 2.4 m. She used her probe pole to locate a buried victim. Riders 2 and 6 used shovels to dig into the snow, eventually finding Rider 4. He was almost 4 feet under the snow. Rider 2 began CPR while Rider 6 made voice contact with Rider 7. Rider 6 then assisted in the CPR effort. The rescuers eventually decided there was nothing more they could do. They collected equipment, marked Rider 4's location, and began descending to assist in the remainder of the rescue.

Riders 5 and 7 descended to Rider 3. He had been buried to his chest but had dug himself out with his hands. His snowboard was destroyed and he was injured, but he was able to assist with the rescue. Rider 3 did not have any avalanche rescue equipment, but Riders 5 and 7 turned their transceivers to receive and got a signal and distance reading of 43 m. The three of them descended and located Rider 1. He was buried under about 2 feet of snow. All three worked together to dig Rider 1 out of the snow, but they only had two shovels and the injuries sustained by Riders 3 and 5 injuries made it difficult for them to use the tool. They lifted Rider 1 out of the hole and looked for signs of life but found none. They took Rider 1's transceiver and gave it to Rider 3. Riders 3 and 5 continued downhill in search mode (they did not know that Rider 4 had been found), while Rider 7 called 911.

Riders 3 and 5 continued to search with their transceivers as they moved downhill. They did not get another signal. Rider 3 was having a very hard time moving through the debris with his injuries and told Rider 5 he was not going to be able to help much longer. They agreed that Rider 3 would proceed to the road and look for help or try to reach one of their vehicles. Rider 5 worked back uphill doing a transceiver search.

Rider 7 struggled to reach a 911 operator on his cell phone. He made about 10 calls, passing some information to the dispatcher each time. He moved around searching for better reception. Rider 6 reached Rider 1's burial site. She stayed with Rider 7 as he tried to communicate with the 911 dispatcher. Rider 2 reached their location and continued his descent. Riders 6 and 7 watched emergency vehicles pass on the highway below them and then collect below the avalanche path. They made voice contact with Rider 5, who reported his injuries could be life threatening and that he needed to descend. Riders 6 and 7 marked Rider 1's location and they all traveled downhill towards the EMS vehicles on the highway.

Riders 2, 6 and 7 reached EMS staff on the highway and Rider 2 was transported by ambulance to a nearby hospital. Members of the Lake County Sheriff's office and Search and Rescue team brought two sleds across Lake Creek and transported Riders 3 and 5 to the EMS vehicles. They were transported to a nearby hospital. Rider 5 was then transported by air ambulance to a hospital in the Denver area. Rider 3 was moved to another medical facility that could provide more specialized care.

The Lake County Sheriff's office, Lake County Search and Rescue, and supporting agencies returned to the accident site on February 16, 2014. They located and transported Riders 1 and 4 from the field.      


This was a tragic accident in which two young men lost their lives. However, given the size of the avalanche and the number of people caught, or with the potential to be caught, it is fortunate that only two people died. This is partly because the slide resulted in one shallow burial and two partial burials, and because of the good fortune of Riders 6 and 7, who were standing on a 32-37 degree portion of the slope that did not slide. In addition, the group both performed a swift companion rescue and showed a tremendous amount of tenacity as they fought to reach rescue teams that reached the site at sunset. 

 As with most human/avalanche involvements, there are several places where this day of recreation turned into a tragic event.

1) Unusual weather produces unusual avalanches: The accident occurred near the end of a period of heavy snow fall that few in this area have witnessed. This prolonged loading event stressed a very weak snowpack and had produced large and destructive avalanches during many of the 10 days preceding this accident. These avalanches were on slopes that face north, northeast, east, and southeast. They occurred at all elevations. This was not a good time to venture into a large, northeast-facing avalanche path with over 2 acres of terrain steeper than 40 degrees.

2) Recent avalanches are the best predictor of future avalanches: There was a natural avalanche in the Star Mountain B avalanche path, which is next to the Star Mountain A path and has the same aspect and elevation. This avalanche released within the 36 hour period before this accident. The group was not aware of this avalanche and could not see it along their route or from the portion of the ridge where they began their descent. The Star Mountain B path crosses Highway 82 three quarters of a mile west of the La Plata Trailhead. Debris from this avalanche reached Lake Creek and would have been visible from the highway with atmospheric visibility of 500 feet or more.

3) It is difficult to make effective decisions from snow profile observations alone: The group observed a snow profile and conducted several snowpit tests. The results were a significant factor in their decision to descend the path. Although snow profile information can be very useful, it must be weighted for the particular application. In the case of a very large avalanche path, particularly one with non-uniform terrain features and a start zone heavily affected by wind loading, it is unlikely that one profile at the top would yield enough information to make a life and death decision. The variability of snowpack depth and structure within the start zone was both large and very important for determining the potential for skier triggering. A single profile is rarely enough to make this determination, especially in large and complex terrain.

4) Safe travel techniques: The group used travel techniques or terrain management to increase their margin of safety. Terrain management is a very important tool in backcountry avalanche safety. However, in this case the combination of the methods they applied and the terrain they were in did not produce a significant increase in safety. Common practice is to expose only one person to avalanche hazard at a time. The 20-second intervals they used did not accomplish this. They also used areas with sparse trees as Islands of Safety. Traditionally Islands of Safety are areas where, if there is an avalanche, people inside that area will not get caught by the slide debris. Trees on a 40-degree slope need to be very close together (close enough that you cannot easily ski through them) to offer a significant degree of safety.          




Image Missing
Figure 33: A fracture-line profile observed on the downhill side of the portion of the crown that the group descended.