01-17-2013 Avalanche Accident in Central Gully

 

On Thursday, January 17, 2013 a party with a total group size of 12 was ascending Central Gully in Huntington Ravine when one rope team triggered a soft slab avalanche from the top of the route. The avalanche swept over the three other rope teams, carrying one team of three to the bottom of the gully. This team was not buried, but sustained injuries. The remaining three teams were able to rappel the route.

Weather Summary:

The weekend prior to the incident was incredibly warm. Mt. Washington set an all-time record high temperature for the month of January during this time, at 48 degrees Fahrenheit. On Monday, temperatures across the mountain began to fall back below freezing and by Tuesday morning, all snow surfaces in Huntington had frozen into a very firm crust. On Wednesday, snow began to fall with strong W and NW winds. The Mt. Washington Observatory reported 2.3” of light density snow from this weather system. On Thursday morning, the Observatory forecasted a trace to 2″ of new snow with isolated higher amounts possible, and W and WNW winds increasing from 60mph to 80+mph with higher gusts. Thursday’s wind and snow played out as forecasted. Most of the snow fell between 7am and 1pm; total snow accumulations of 3.6”  exceeded the forecasted amount.

Snowpack Summary:

The melt freeze crust that developed Monday and Tuesday created a slick bed surface for future avalanche activity. This was noted in avalanche advisories Wednesday and Thursday. On top of this icy layer, new soft slabs began to form on Wednesday while winds were blowing 30-40mph. As additional snow fell Thursday with increasing wind speeds, slightly denser slabs were deposited above the weaker slab and the crust. The climber who likely triggered the avalanche stated that, at the time of the avalanche, he was climbing through soft snow about thigh-deep or waist-deep. However, other reports were that the slab that released was only 8” deep and between 25-35ft across. We believe that failure occurred in a weak layer interface somewhere within the new snow, rather than at the crust.

Avalanche Summary:

The avalanche was a soft slab, artificially triggered by foot penetration, which in the professional avalanche lexicon means that it was triggered by a person climbing or hiking, not by a person traveling on skis, snowboard, etc. The slide is further classified as D1.5, R2 (hyperlink to SWAG PDF). This is a measure of the destructive force of the avalanche and the size of the avalanche relative to the specific avalanche path’s potential. Compared to the size of avalanches Central Gully can produce, this was on the smaller side. The debris was examined by a Snow Ranger, who estimated its size as 5-7 meters wide, 60 meters long, and 30-60cm deep.

Events Leading to the Incident:

An organized group of twelve climbers planned a promotional climb to draw awareness to their organization’s mission. They had been training for the climb in the days preceding the event, which included ice climbing in Crawford Notch. The group was organized with a variety of experience and skills, from novice to experienced mountaineers. In addition, a film crew was included in the group.

The group of twelve arrived at the Harvard Mountaineering Club cabin on Wednesday afternoon and spent the night in the cabin. Thursday morning, after receiving the weather forecast from the Mt. Washington Observatory they had decided they would climb Central Gully. Before departing, a USFS Snow Ranger arrived and talked with the group about weather and avalanche conditions. Despite this discussion and warnings about increasing avalanche danger through the day and that Moderate avalanche danger means that “human triggered avalanches are possible,” the group decided to stick with their plan. They departed from the cabin at 8:30am.

Four hours after leaving the cabin, they arrived at the start of the climb. The group split into four rope teams of three people each. They ascended to the ice bulge in the gully, then one by one they climbed the bulge on belay. Above the ice bulge, the teams began simul-climbing. They reported that they had been skirting the newly deposited snow and trying to stay on the older crust. Just prior to the avalanche, the lead team allowed the second team to pass them, so that they could get better set up for filming. At the time of the avalanche, there was one team nearing the top of the gully, another was slightly below them and positioned in the center of the gully. The other two teams were lower, hugging the climbers left side of the rock wall. During the time the teams were in avalanche terrain, snow continued to load into many areas, including the top of Central Gully.

picture of approximate location of climbers at time of avalancheThe Avalanche:

The details we received about who was where and what happened when the avalanche hit don’t give us a 100% clear view. The picture indicates our best estimates of where the rope teams were located at the time of the avalanche. It was approximately 4:30pm when the avalanche was triggered. The party at the top was not caught or carried, though they may have slid a short distance. The second-highest team was caught and carried over the ice bulge to the base of the gully. They came to rest in the debris, which terminated at roughly the elevation of the base of Pinnacle Gully. A third team, located to the side and away from the path of the greatest debris flow had started to be carried, but was able to avoid being carried downslope by the bottom climber arresting the fall with his ice axes. The fourth team was carried downslope, but they stopped moving when their rope was caught on an exposed rock.

After the accident happened, the three teams remaining on the route took a quick inventory of who was present. It quickly became apparent that one rope team, including the lead guide, had been swept down off the route below all the others. At this time, the remaining members of the group reorganized and began to descend on rappel. At all times, all members of the descending party were either clipped into a rock or ice anchor or were actively on rappel. They stated they were unable to make contact with the three people who were carried down with the avalanche, either by voice, visual, or their family-band radios. They attempted to call for help via cell phone, but were unable to do so because their batteries had died. They also had a satellite phone, but were unable to sufficiently connect with satellites.

The team that was caught and carried down to the base sustained some injuries. Of the three, two had lower leg injuries and the third initially complained of pain in his shoulder. They were carrying a radio that operates on the same frequency as the Mt. Washington Observatory, Appalachian Mountain Club, and HMC cabin. With this radio, the lead guide was able to contact Rich, the caretaker at the HMC cabin. While Rich worked with the AMC Hermit Lake caretaker to notify USFS Snow Rangers, the injured climbers began sliding along the snow, working their way down the fan to toward the base of Huntington Ravine.

The Rescue:

USFS Snow Rangers were notified of the incident at approximately 5:22. In addition to the USFS, AMC, and HMC, the volunteer Mountain Rescue Service (MRS) was called for assistance. They responded with 18 skilled mountaineers for a total rescue team of 25 people. The first two Rangers arrived at Pinkham Notch at 6:00pm. One immediately left on snowmobile for the ravine while the other stayed behind to organize other rescuers who began arriving shortly afterward.

The first Snow Ranger and HMC caretaker parked the snowmobile near the first aid cache at the base of Huntington. At 6:20pm, approximately 200 yards uphill from the cache, they encountered the injured climbers slowly working their way down the trail. They briefly questioned the group about what had happened and if they had any information about the rest of the team. Knowing there were more rescuers who would be arriving soon, they did not want first aid at this time. At the request of the lead guide, the hasty team continued up into Huntington where they could see headlamps slowly descending the gully. They climbed up the fan, careful to avoid the avalanche runout path from Odell, Pinnacle, or Central Gully, until they were able to make contact with the remaining climbers and determined that they were doing OK. The group continued to rappel out of technical terrain.

The second Snow Ranger on scene and one member from MRS arrived and began treating the team’s injuries. The two most seriously injured climbers were treated and packaged into rescue litters. As they did this, more MRS members arrived and began to transport them to the Harvard Cabin where the USFS snow tractor was waiting to transport them to ambulances while the third waited for rescuers to return and transport him in a litter. This group arrived at Pinkham Notch Visitor Center at about 9:15pm. Meanwhile, the remaining MRS members and the hasty team assisted the group of 9 uninjured climbers’ rappel to the talus and then down from the top of the fan to the Harvard Cabin. Of this group, one sustained minor frostbite injuries to his toes. The entire group was transported from the Harvard Cabin 2 miles to Pinkham on the snow tractor, arriving at the base approximately 11:30pm.

Analysis:

The information and timeline described above are the facts as best as we have been able to gather them. The facts presented are as accurate and objective as possible. The discussion that follows below is our analysis and interpretation of the situation. It is a subjective analysis of what took place the day of the incident and represents the collective professional knowledge and experience of our team.

With the value of 20/20 hindsight, any accident can be picked apart by someone looking to place blame or find mistakes that may have been made. This is not our intent here. The purpose is to try to determine what lessons can be learned from the decisions that led to the accident so others can learn from these experiences and avoid making similar choices.

Every accident in the mountains is unique, so understanding the context surrounding decisions and actions is an important component. Doing this helps us understand the “decision crossroads” that led to this incident and other similar historical mountain mishaps. Whether as Snow Rangers or as recreational climbers, we have each faced decisions points where we need to assess the interaction between a wide variety of factors and choose a course of action. Things such as turnaround times, changing weather, changing avalanche hazard, group decision making dynamics, evaluation of the group’s skill and experience, etc. all provide context from which we can reflect and learn.

In this incident, there were many factors involved that added risk to the overall situation. When considered individually, each one may not seem like a catastrophic error or miscalculation. However, we believe the accumulation of these overlapping factors led this group to being in a dangerous situation, and moreover, to continue moving forward with their plan when other groups may have chosen another course of action. We believe this incident was not a freak natural event completely outside of the control of the party. Avalanches are a common natural phenomenon in Huntington Ravine and this event became an incident because the group made decisions and took actions that placed them in a precarious position.

Motivation and Commitment:

A frequent contributor to avalanche incidents worldwide is the motivation and commitment level of a group. Once a group invests themselves into an objective, it becomes more difficult for the group to retreat from the objective or alter their plan. This is a heuristic trap that is commonly taught in basic avalanche classes. No person, from the novice to the avalanche professional, is immune from it entirely. The trick is to know how to recognize its influence on one’s decision making and try hard to minimize the effect.

This group was heavily invested in success in many ways. They were organized as a charity for a very worthwhile cause. The team members had all donated significant amounts of time. The climb was being filmed by a professional filmmaker for a documentary. There was a strong media campaign to draw attention to the climb…these all increase the level of commitment beyond what might be normal for a purely recreational climb. An increased acceptance of risk comes often comes with an increased level of commitment. There is no way for us to know how much of a role this factor played in the incident, if it played a role at all. It is our assumption that for at least some members of the team this was a contributing factor to their acceptance of the risks they faced.

They also had pre-arranged to spend the night at the Mt. Washington Observatory. Whereas for most climbers the summit is the halfway point for their entire climb, in this situation the group had extra incentive to push through to their final destination. When groups are planning to return to their starting point, they will often set a turnaround time. Regardless of where they are when the time comes, they will stop climbing and head back down. Establishing protocols such as these are a time-tested method for helping keep climbers out of trouble and mitigating risk. Staying flexible and watching for reason to turn around earlier, as an example, is an excellent decision, but sticking to predetermined protocols is essential. On a one-way trip, deciding to turn around and descend is a very difficult decision to make.

Avalanche hazard:

A fundamental challenge for avalanche forecasters is to convey the meaning of each different rating level. Understanding the rating scale is a critical first step in understanding how much risk you are accepting. Often people think that Moderate conditions equate to a risk level that they are comfortable with since moderate ranks second on a scale that goes from one to five. It’s easy enough to understand the degree of risk from Extreme or High avalanche danger. The risk of traveling in the lesser-rated terrain drops from there. Read the definitions carefully and you’ll see that even a Low rating indicates some risk of encountering pockets of unstable snow. A “moderate” rating means that “human triggered avalanches are possible.” Not only should people think about the probability of an avalanche, but the consequences of such an event must not be ignored. In Huntington, particularly in lean snow cover, avalanches run out into boulder fields. Within the United States, New Hampshire has the highest percentage of avalanche fatalities due to trauma as opposed to asphyxiation due to being buried.

In this incident, the group made the decision to climb Central Gully after receiving the weather report at the Harvard Cabin. When a Snow Ranger arrived at the Harvard Cabin shortly thereafter, the group had already decided they would climb Central. The Snow Ranger attempted to discuss snow stability with a gathering of several group members, but the group deferred judgment to the group leaders who were inside the cabin at the time. He then went inside and discussed the rating, the incoming snow, and the increasing danger with the leaders, who confirmed that they would move forward with their plan to climb Central. Later, when one of the injured climbers recognized the Snow Ranger rendering first aid as the one who had spoken with the group in the morning, he stated that this was the Snow Ranger “that thought we were idiots for climbing Central” that day. Of course these aren’t the words that were used, but the statement demonstrates that at least one member of the group understood the risks described by the Snow Ranger.

The avalanche hazard was known to be on the rise during the day. This was described in the morning avalanche advisory and as snow was forecasted to fall heavily at times. By early afternoon, hours before the avalanche occurred, snow accumulations had exceeded the weather forecasted totals by 1.6”. This snowfall event brought 0.4” of snow-water equivalent (SWE) to the summit, 90% of which was recorded between 6am and 12pm. In afternoon hours, snow continued to fall at a much lighter rate, but snow was being actively transported into Central Gully due to high winds and forming soft slabs. These slabs were recognized by the group leader, as he stated he had been trying to avoid them all afternoon. This evidence indicates increasing avalanche hazard, and is commonly considered to be “bulls-eye data” or a “red flag.”

Regardless of the forecasted rating, it is very important to be capable of assessing snow stability during a climb. In this case, the lead guide had been doing this. He stated he had been “skirting a slab all afternoon.” Indeed, avoiding areas of unstable snow and staying on hard old surfaces is a recommended way to avoid triggering an avalanche. However, it was not the lead guide’s rope team that triggered the avalanche. Another team had moved out above this team to get better set up for filming. This group had an experienced climber in the lead for most of the climb, but just before the final pitch they “swung leads,” so that the person who had been at the bottom of the rope was now leading. This person initially stated that the snow he was climbing through was thigh or waist-deep. The depth and softness of the snow would be another “red flag,” which should trigger another decision point where the climbing team can reassess the plan to move forward. Even at this point near the top of the gully, descending was still a viable option, albeit a challenging one.

We believe that the overall confidence in the leader’s ability and experience may have led to some group members withholding from the entire group avalanche concerns they may have had. This confidence was stated by one group member as a reason for not carrying avalanche rescue gear (i.e. beacons, shovels, and probes). While we don’t condone the practice, it is not uncommon for climbers in Huntington to travel without avalanche rescue gear. We understand that there are times when the risk of being buried in an avalanche in Huntington is much less than the risk of being severely injured or killed by the fall itself. However, leaving this equipment behind significantly reduces your safety margin should an avalanche occur. This life-saving equipment should be seen as an important part of an overall safety system. It’s the final defense, to be used only when objective hazards are not avoided through decision-making. Without it, the chances of rescuing a buried victim in time are reduced to unreasonable odds. We recommend carrying avalanche rescue gear when traveling in avalanche terrain, because we believe it is the right thing to do.

With the benefit of hindsight, we do not think climbing Central Gully would have been a poor choice for every group on this day. Given the weather conditions and increasing avalanche hazard, an early-rising, fast-moving team of climbers comfortable with the terrain could have climbed through the gully before instabilities developed very far. If snow stability during the climb had deteriorated too much, they could have downclimbed, rappelled, or traversed out of the gully into the rocks on the right before they developed to the point where they might naturally release. This group’s pace certainly contributed to the accident, as they arrived in avalanche terrain four hours after leaving the Harvard Cabin. It was during these hours that most of the snow had fallen, and the group continued to climb into worsening avalanche conditions.

Group Size:

Twelve people on a climb such as Central is not completely unreasonable, but it does create some challenges and risks. Managing avalanche hazard, choosing appropriate technical climbing techniques and the pace of travel are all affected by the large group size.

One of the fundamental concepts of traveling with others in avalanche terrain is to minimize the exposure to avalanche hazard at any time. For skiers, this most often equates to skiing a slope one person at a time. For climbers in Huntington, the one-at-time maxim is very difficult since the gullies are fairly narrow slide paths without many “safe zones” between which a group can move. In such cases we often advise roped parties moving through potentially unstable snow to protect their route with rock and ice gear. With the exception of descent this is one of the only ways for climbers to mitigate avalanche risk when ascending narrow steep slopes. Three distinct ways the group size added to their exposure to the hazard are 1) the sheer number of people on the same slope at the same time, 2) it slows the pace and therefore lengthens the duration of exposure, which is particularly a problem during increasing instability, and 3) more people on a slope increases the likelihood that someone will climb over a weak point and trigger a slide.

The pace of climbing is also related to group size. Generally, larger groups move more slowly than smaller groups. Other factors can slow a group down. With this group, one climber was using a prosthetic device that had a smaller footprint than a standard boot. This slowed the climbing greatly, as he would break through the crust where others would not. There is no doubt about this climber’s physical fitness and endurance, it is simply more difficult for anyone to move fast when he or she is breaking through an established boot pack. The temperatures on Thursday dropped down to around 0F during the afternoon in the ravine and -10F on the summit. In temperatures such as these, speed and efficiency are important safety measures.

Related to the pace is the choice of how to travel as a group in steep terrain. There are many techniques available to climbing teams and no one way is right for every situation. In this situation, the group was divided into four separate teams, each tied together with 60 meter ropes with one climber tied to the middle. At times earlier in the climb, the teams had used protection and anchors to belay climbers over the ice bulge. Sometime after this, most teams had begun climbing without the benefit of snow, ice, or rock protection. They were belaying at times, using “snow thrones” backed up with ice axes planted in the snow as their anchors, but otherwise there was no protection between anchors. This technique exposes climbing teams to a significant amount of risk. If one climber falls, the other two climbers must arrest the fall to prevent the entire team from falling. In steeper terrain and on icy surfaces, arresting falls becomes increasingly difficult. If one team falls together or is caught in an avalanche, there is a chance that their rope will catch other climbing teams and cause them to fall as well. Here, the topmost rope team triggered the avalanche but fortunately did not get carried downslope. The team that was caught and fell +/-800ft was located farther out into the center of the gully than the others. The other two teams did get carried at least a short distance. One team was able to arrest their fall, but the fourth did indeed fall until their rope became hung up on an exposed rock just above the ice bulge. It could be argued that they would have fallen all the way if they weren’t tied to a rope, but the rock essentially served the same function as ice, snow, or rock protection would have in this instance. We believe using protection is a safer option when using roped techniques in this terrain. Of all the options available, the chosen method for this climb on this day would be among the least desirable techniques.

Lastly, related to the group’s pace, is the method of descent. Once the avalanche passed, the group was able to account for those still on the slope and knew that one team of three had been swept downslope. The team reorganized and made the decision to descend the route which we believe was the correct thing to do. However, when dealing with an avalanche accident you are in a race against time because statistics show you have 15-30 minutes before most incidents move from rescue to recovery in the case of full burial. The speed of the companion rescue is a key factor in preventing fatalities. Although no one was fully buried in this incident, the remaining teams in the gully were unaware of the fate of the others until rescuers arrived. With 9 people in the group, descending on rappel one at a time is a very slow process, though it is also a very safe method. Had the fallen team been buried, received more serious injuries, or not been intercepted by rescue teams, the delay in treatment would have been life-threatening. Because the terrain in Central is not overly technical it is commonly used as a descent route for parties who have climbed another route. In a group of 9 skilled and experienced climbers, it would be reasonable for some in the party to downclimb more quickly to initiate a rescue, while the others continue to rappel.

In conclusion, this is clearly a complex situation where a lot decisions needed to be made as the day unfolded. We believe that this was an avoidable accident that fortunately resulted in very minor injuries considering the magnitude of the incident. We have the benefit of hindsight and were not involved in the group’s decision making process, so it’s impossible to know all the factors and how they were considered. Again, the intent of this analysis is not to place blame, but to allow others to learn from the experiences of their fellow climbers. We wish group members the best in their admirable cause and in their future mountaineering endeavors. We look forward to seeing them again in the hills pursuing climbs with new lessons learned under their belt.