Accident Database

Report ID# 522

  • Flush Drowning
  • Hypothermia
  • Other

Accident Description

On October 5, 1998, four American kayakers launched boats onto the Yarlung Tsangpo River in East Tibet . This was the beginning of a long-planned expedition to attempt to travel one of the world’s great unrun rivers. At the town of Pei the water was medium-brown, flowing swiftly, about 1/3 mile wide. Considering high water marks left recently on the banks, the river had been higher ten days before. Now it was at a medium-high level, 10-20 feet lower. Over the next few weeks, water levels would continue to drop 2-4 inches daily.

We ran the stretch from Pei to Gyala as a warm-up cruise before entering the heart of the gorge. The gradient was not as steep as it would be downstream, and well-traveled trail following the river allowed the support team to walk to Gyala, and provide re-supply. The four paddlers, Jamie McEwan, Roger Zbel, Doug Gordon, and Tom McEwan, were able to assess the “Himalayan” magnitude of river and terrain, and developed appropriate methods for dealing with it. Paddling and carrying, usually avoiding the main flow and scouting far ahead, they took four days to arrive at Gyala.

Leaving Gyala, the group expected to spend significant time on foot scouting the river. They carried 15 days of food. Their plan was to meet up with the Support Team of Wick Walker and Dave Phillips near Rainbow Falls , about 26 miles downstream,  They planned to hike up the gorge with porters and supplies for the Expedition’s next segment.

On October 16, at about 11 a.m. , as the River Team made their way down the left side of the river, avoiding the main current out in the middle. They stopped to scout the rocky edge of a large rapid. Doug, Jamie, and Roger considered several possible routes, while Tom set up downstream on a boulder to do video and to hold a safety rope. Doug went first, choosing a line over an 8-ft. waterfall hugging the side of the river. Aiming for an eddy just below, he would boof over the rock and land himself in the left-hand eddy. However, he was unable to clear the hydraulic at the foot of the falls, causing him to be caught and pushed toward the main part of the river.

There, freed from the hydraulic and still in his upturned boat, he had an opportunity to recover. Doug attempted two rolls, which were both unsuccessful, and continued to drift further out into the middle of the current. He and his boat were now well out of range of the safety rope. His three teammates watched helplessly as he was swept into the rapid below - a certainly fatal series of huge recirculating hydraulics - and out of sight.

The search for Doug began immediately. Tom and Roger raced over rocks down the shore, while Jamie unloaded his boat, carried stretches of difficult rapids, and paddled down the river. The next four days were devoted to moving downstream and scanning the shores for any remains of Doug or his equipment. The Support Team, alerted by satellite phone, reached the river and began a search downstream. On October 20, 8½  miles below the accident site, the two groups met, and the search was called off. Doug was presumed dead. All local and national authorities, as well as families concerned, were notified of the accident.

At this point, we decided to end the expedition and return home by the most direct route. They were still seven days of hiking and three days of driving from Lhasa . A small ceremony was held alongside the river, with the local Tibetan porters and the Americans participating. Tibetan and American songs were sung, and a square stone with Doug’s name written on it was cast into the river. The next day, as we were preparing to leave, the porters, decided to bargain for more money by threatening to abandon the Americans and their equipment. They packed their bags and marched out of camp, only to turn a few hours later when this maneuver did not work. The Expedition arrived in Lhasa November 3.

The River Team had traveled 35 miles down the Tsangpo, out of 140-mile gorge originally intended. We passed between the 25,000 and 23,000 foot-high peaks of Namcha Barwa and Galiperi respectively, through the deepest canyon in the world. New methods of long-range scouting were developed to prevent the team’s becoming trapped, and this allowed us to make the progress that we did. The Support Team was able to meet up with the River Team deep within the gorge after journeying over some of the most extreme, Himalayan terrain. 

The expedition members deeply regret the death of Doug Gordon, our loyal friend and team mate, an expert kayaker who lost his life in this challenging expedition.

SOURCE: Written by Jamie McEwan, as published in the CCA Cruiser

2. (McEwan) There is no moral to be drawn from Doug’s death, other than that what he himself wrote in reference to Richie Weiss’ death the summer before: Running hard whitewater is dangerous, and that those doing so must accept that risks as the price of pursuing their sport at a high level.

3. (Walbridge) An expedition member told me that Gordon was paddling a large kayak, which was necessary to carry their gear. He hypothesized that the boat may not have fitted him well. This might explain why a world-class paddler might miss a straightforward roll attempt.



On October 16th 1998, Douglas Gordon lost his life on The Tsang Po River in Tibet in the vicinity 29 Degrees N 45min 94 N 58min E. Witnesses observed him being swept into an almost lethal set of rapids from which he was not seen to emerge. Five days of diligent searcg by members of his party assisted by eleven Monpa Hunters failed to find Mr. Gordon or any of his equipment on either shore for a distance of approximately 16 kilometers downstream. Due to the nature of this river and surrounding terrain, his survival a greater distance than that is certainly impossible. Mr. Gordon is presumed to have died by drowning.

Elements of the search team were assembled on 21st October and will begin trekking out of The Tsang Po Gorge via Mendung Village Pailung on Lhasa-Chengdu Highway. Estimated travel time is 10 days.The Tsangpo River in Southeastern Tibet flows through what may be the deepest canyon on earth. The stretch between Pei and Medong is one of the great remaining whitewater challenges on this planet. On September 24,1998 a 12-man expedition organized by Wick Walker assembled in Kathmandu to begin the trek to the river.

Jamie McEwan, one of the river team members, reported that they found the river running at over 30,000 cfs, far more than the 5000-15,000 cfs they had hoped for. However, the team was encouraged to hear that the water had been dropping steadily for weeks. The group took four days to warm up on the easier, more accessible 18 mile stretch from Pei to Gayla. After more discussions the group then preceded cautiously downstream, carrying 15 days worth of food. They planned to work the edges of the rapid, staying out of the main flow, running what they could, and carry the rest. If it turned into "boat assisted hiking", so be it! During this time the Land Crew would trek into the gorge to meet them. This would also serve as limited emergency back-up.

On October 16 the group paused at the top of a rapid to scout. Doug Gordon, 41, a former U.S. Team member with several expeditions to his credit, elected to run an 8' ledge near shore while others set safety. The boat failed to clear the hydraulic and flipped. Gordon missed two roll attempts and was pushed towards the middle of the river. Still upside down, he was swept into the maw of the rapid below and washed into a series of terminal hydraulics. For the next four days the land and river crews, assisted by native hunters, searched the river downstream. They found no sign of Gordon or his gear. On October 20th the search was called off. The group notified authorities and began an arduous 7-day hike back to civilization.


1. (McEwan) Were we foolish to be on the river at that level?  In light of what happened to Doug, it’s easy to say yes to that question. Knowing the consequences, I wish we had never put on. And yet, without the benefit of hindsight, I believe we would make he same decision.

When we arrived at Pei and first looked at the river’s flow not far above the gorge, Roger at once voiced some doubts about the enterprise, suggesting that it might be wise to run an easier canyon upstream, or perhaps the tributary Po Tsangpo. Tom suggested a long scout, giving the river a chance to drop, or perhaps to paddle down to the first major obstacle, where we might camp and wait for the water to further recede. We had a report from ten days earlier, from German kayaker Lukas Blucher, that the river was 70-90,000 cfs; we estimated the flow at our arrival at between 25 and 50,000 cfs. The banks showed evidence of a 10-20 foot vertical drop within the last few weeks, and we knew, from placing markers, that it was dropping still. It was no longer flood level – the river was well within its banks, probably thirty vertical feet below its vegetation line. Still, it was considerably higher than the 5 to 15,000 we had hoped for.

Doug and I were consistently on the side of putting in and working our way downstream in something approaching our normal river-running mode:  paddle what we could, scout and portage where we must. Yes, Doug admitted, it might quickly turn into “boat-assisted hiking,” rather than river running, and we would not cover anything close to the full mileage planned. But why not paddle down and at least begin our hike from a point further downstream? If need be we could always retreat to Pei, but if we could get anywhere close to Rainbow Falls, we could hike out with Wick and “complete the loop.”

We had all been in places on other rivers, above unrunnable waterfalls, huge holes, or rock sieves, where a missed eddy or a swim would have caused great danger or even death. The difference was not in degree, but rather in the constancy of the danger. It was hard to get used to the idea that we could seldom find a place where we could safely ferry the river. The current in the river’s center was so inexorable that we rarely dared venture there, for fear of being swept into the next unrunnable portion. However, we felt that by taking the same sort of precautions we took, we would only tackle those risks that we felt we could handle. Even if this meant hiking most of the way. But we could have been affected by the “Tsangpo difference”. We might have become used to the danger because it was always with us. Judging from Doug’s comments that morning  I don’t think that was the case.

Join AW and support river stewardship nationwide!