Tower Mountain and Azurite Peak Winter Ascents

Tower Mountain (8,444ft) and Azurite Peak (8,400ft)

Feb 4-5, 2023

26 miles skiing, 40 miles snowmobiling, 11,000ft gain

Eric, Nick, Talon

The east north zone looked to have stable snow this weekend, with the safest conditions farther east. I was excited to try out the snowmobile approach on highway 20 from the west, starting at the Ross dam trailhead. Tower and Azurite can be accessed from highway 20 and are two Bulger peaks I still needed in winter, so we decided to go for those. I had previously climbed Tower and Azurite in July with Matthew, so was familiar with both routes.

Winter routes aren’t always the same as summer routes, though. In summer 2018 I had hiked the PCT to approach both peaks, but in winter it makes much more sense to reduce mileage and bushwack in up swamp creek. I had read on Turns All Year that in May 2022 Tim Gibson had approached via Swamp Creek and skied the w face of Tower. His snow conditions looked similar to what we expected in winter, so we decided to follow the same route he took. I have seen on peakbagger Chris Buchta climbed Tower in March 2010, but I’m not sure which route he took.

For Azurite peak I couldn’t find any accounts of previous winter ascents, so we decided to give the summer route a try. If visibility turned out bad at least we would be familiar with the route so navigation would be easier.

It would seem to make sense to include Golden Horn with these peaks, since it is another Bulger right next to Tower. But I’ve already climbed Golden Horn in winter, so didn’t need to repeat it. On that trip I had skied in from the Yellow Jacket sno park in Mazama up the Methow River. Avy conditions were not good on that trip so we had skipped Tower and Azurite.

In November 2021 I had scouted the mazama approach with my snowmobile, and was able to make it through the liberty bell slide path since it was still early season. I had parked the sled at swamp creek and skied up the creek a few miles. This allowed me to get an idea of the bushwhack conditions, which were pretty tame, even though that road approach might not be viable later in winter.

The Ross dam approach for these peaks is appealing because it requires about half the driving time from seattle as the mazama approach. Also, there are no major avalanche slide paths that cross that western portion of highway 20. Coming from the Mazama side would require crossing the Liberty Bell slide paths, which can require tricky side hilling later in winter.

The Ross dam approach is interesting because it starts at a gate at 2200ft elevation, but then drops to 1800ft a few miles later, before rising again all the way to rainy pass. This can be problematic because the first few miles might be snow covered, but a few intermediate miles at lower elevation can be melted out. It can be difficult to get through long pavement sections on a snowmobile. Most snowmobiles (including mine) require snow to cool the engine, and the engine can easily overheat if you ride on pavement for too long. Also, pavement can destroy the skis and carbides on the front of the snowmobile, and it is very difficult to steer. I have installed retractable wheels on the front of my snowmobile, so I’m able to steer on pavement and won’t destroy the skis. But there is still a risk of overheating, I’ve found, on stretches more than a mile or so.

Based on historical satellite images the meltout seems to be problematic in late winter, so this approach is generally most realistic in mid winter when the road is completely snow covered. In fact, in mid march 2021 I had planned to take this same approach to climb Tower, but when I got to the gate the night before starting I discovered that WSDOT had plowed the first five miles of road! They usually announce this in advance each year, but hadn’t that time. I didn’t have the retractable wheels then and had to abort the trip.

A few days before this trip I checked the daily satellite image and on Wednesday it showed total and continuous snow coverage from the Ross dam trailhead all the way to swamp creek at the base of Tower Mountain. There was forecast to be a heavy rain event Friday at the lower elevation below 2000ft, but we were optimistic the snow might still be there Saturday.

Friday evening we met up at the trailhead, slept a few hours, then started up 6am Saturday morning. There were no other snowmobilers there, and this is a huge difference from the Mazama side of highway 20. On that side there are around 50 snowmobiles that are parked at the end of the plowed road, ready for use all winter. And many people ride that stretch of road all winter. The difference is that side is  reliably snow covered, while the west side is warmer and has less reliable coverage.

That morning there was forecast to be a bit of light rain at our starting elevation, which would change to snow as we got higher. I don’t often snowmobile in the rain, but it sounded like a risk of getting soaked. So we each wore a separate outfit for the ride up, then we would change to climbing clothes at the trailhead so we could be dry.

We loaded two packs, skis, and chainsaw on the snowmobile, then I squeezed it through the small gap to the side of the gate. Nick got on and rode passenger while Talon wore his pack, strapped on his snowboard, and prepared to get towed. I rigged up a bike-tube belt with caribeaner clipped to a 20m section of old climbing rope rigged up with a 3-1 pully system. The belt gives the person a smoother ride if the snowmobile jerks a bit, which is hard to avoid. The pully system allows the person to hold on with less force. Talon then hooked up his ascender to the rope so he could grip more easily. The system is rigged so that if he lets go the rope pulls out and he isn’t dragged at all.

I started out slowly and gradually worked up to a 20mph speed, which I’ve found is the fastest reliable speed for a person not to fall off. The snow started a bit icy but Talon stayed on no problem. After about 2 miles, though, we started hitting sections of ice and pavement. I had to stop and let Talon walk around while I deployed the wheels and got through. There was also a major slide across the road around mp 137. I was able to carefully ride over the jumble of trees and rocks, but it was tricky.

We soon decided that was very inefficient to have Talon walk all the pavement sections, and I was worried about losing too much time for the approach. So I decided we would just all three ride on the snowmobile until we got to a high enough elevation that the snow was continuous. We would have to ride “meatloaf” style, which means one person sits on the seat and the other two stand on the sides. I would hold the handle bars to steer while the other standing person (Nick) held onto the middle bar.

I had never tried this before and was a bit nervous of overworking the snowmobile. The problem with that much weight is that it compresses the tail onto the track. This can make the engine work much harder with the track pushing against the tail, and it runs the risk of breaking a drive belt. We gave it a shot, though.

This method saved a lot of time. At every pavement section I would stop, leave the engine running, jump off to deploy the wheels, then start up again. Then back on the snow/ice on the other side I’d retract the wheels, still with the engine going. I managed to not overheat at all and was very efficient. Eventually around mp 142 we hit what appeared to be the end of the meltout pavement and ice. I was still nervous about overworking the engine so Talon got off and got towed then.

As we climbed higher the snow got deeper and more powdery. Nick rode backwards so he could be a spotter in case Talon fell off. After a few miles Talon yelled to go faster, so I gradually ramped up to 40mph. That proved to be too fast, and Talon wiped out, but not before getting some good gopro footage.

We regrouped and continued at a more reasonable 20mph, eventually reaching swamp creek at mp 154, 20 miles from our starting point. Luckily there were no major trees across the road, so we didn’t need the chainsaw. Nick and Talon jumped off while I did a few loops revving the engine to high rpm to ensure the spark plugs didn’t get fouled. This can be a problem if going at slow speeds for too many miles, like I generally do on these trips.

We repacked, stashed the chainsaw and spare gear with the sled, then started up around 8am. I was happy to be in a group of three, since we could frequently rotate out trailbreaking duties and make more efficient time. The woods were generally open with easy bushwhacking. We stayed on the right side of the creek for the first two miles, then crossed over to more open terrain. We eventually started ascending more steeply, and the woods got kind of dense.

It got better around 5600ft and we made good time up to the edge of treeline below the west face of Tower by 1pm. We ditched overnight gear at a boulder at 7200ft and continued up light. We zig zagged up higher until a point just below the cave. There the conditions got a bit too slippery, even with ski crampons. It was about 4 inches of powder on crust, and got too steep. We ditched skis there and changed to ascent plates.

These conditions were actually perfect for ascent plates. The snow under the crust was soft enough that we would sink pretty deep in crampons, but it was too steep and crusty for skis to make sense. Ascent plates are rectangular pieces that sandwhich between the crampon and boot with the crampon front points sticking out. They are designed for climb up steep deep snow. I had a pair of aluminum billy goat plates I bought, and Nick had made his own pair out of carbon fiber sheets he cut with a jig saw. Both were effective. Talon had ordered a pair but didn’t have them yet, so he would follow in crampons in our steps. He carried his split board up but I was too nervous to try to ski the chute since it has a cliff under it.

I led the way up to the cave, then we traversed above a cliff over to the base of the west face chute. Nick and I alternated kicking steps up the steep slope. The conditions were 4-8 inches of powder on crust, and it was well-bonded and stable. We made it to around 8100ft when we encountered an ice and rock cliff. The summer route continued higher then traversed along the top of a cliff. That looked sketchy, so we decided to diagonal up and right to try to intersect the route on the right skyline ridge.

Step kicking got tough and we had to redirect course a few times. We eventually reached a ledge near the right skyline. It was too steep to go up, so I belayed Nick across a mixed rock and snow ledge to see if it connected to easier terrain. He got about 15m across but the route didn’t go. However, he noticed a snow gully beyond that might work if we could get to it from below our ledge.

Talon caught up by then and downclimbed a short rock step to a snow bowl below. I was too nervous to solo the downclimb so I slung the rope around a horn and made a short rappel. Then Nick followed. We debated leaving the rope there to jug back up on the return, but decided to bring it to keep our options open. We had two 30m ropes and might need them both for the climb. Also, we wanted the flexibilty to rap back down a different way if needed.

In the bowl we traversed below the cliff and then reached the snow gully Nick had seen. Miraculously the gully connected all the way up to the summer route elevation. Talon led the way up kicking steps. At the top of the gully we went up and right to hit the right skyline. There was one more tricky mixed rock step to get up. Talon went first. Then I followed. I had to to an underhook with my ice axe and it was delicate but I made it up.

We then marched up the windy and exposed southwest ridge to the summit by 4:45pm. The huge summit cairn seemed a bit shorter than I remembered, probably because I was standing on a few feet of snow and ice. Unfortunately we were in a windy whiteout with no visibility so didn’t get any views. We spent about 5 minutes on top before retreating. It was already very close to sunset.

We downclimbed to the tricky rock step, then I slung the rope over a horn and we took turns rapping down. We then downclimbed the steep snow and traversed the snow bowl. Now I wished we had left our rope on the next rock step. It was too sketchy to climb back up unroped, so Nick led up to the top then belayed Talon and I back up. From there we traversed back to the main gully, reaching it just when it got dark enough to need headlamps.

Unfortunately in the low visibility it was too sketchy for Talon to snowboard down, so we all plunge stepped back down to the skis. From there we skied and rode down in the dark to our ditched overnight gear, then packed up and continued to snow lakes.

We skinned up to the lakes, which were very much frozen and snowed over. They looked a lot different than in July when I had swam in them. On the west end of the lake we skied and rode down through the trees aiming for the PCT. We soon found a fun creek bed to follow that was clear of trees and well filled in. We soon intersected with the PCT, then made fast time down the PCT to our planned bivy site at the Methow River bridge.

We reached the bridge by 10pm, a bit later than hoped for. It was actively snowing, so not ideal for an open bivy. I immediately started setting up my mega mid tent while Talon and Nick started cooking with our two stoves. I’ve previously slept three people in the mega mid by digging out the snow underneath to make a lot of space. However, I noticed the bridge had a huge sheltered place underneath. So I volunteered to bivy under the bridge so we didn’t have to spend time digging out the mega mid. It was important to get to sleep as soon as possible to be ready for Azurite the next morning.

I ate my ramen with cheese and salami, and we all boiled a few liters of water before going to bed around 11pm.

Sunday morning we were up and moving by 4:30am. Talon decided to rest in camp to conserve energy for the long tow out. Nick and I skinned a few hundred feet past the bridge then left the trail and started diagonalling up. We followed an open avy slope for a while, then cut into the trees heading up and right. We kept the slope angle low except for one short step at 6200ft. That was too steep to skin so I kicked steps up for 50ft. From there we could skin all the way to Azurite Pass by sunrise.

At the pass we got above one of the cloud layers and actually had great views over to Golden Horn and Holliway. From the pass we diagonaled up the ridge through open trees to the ridge crest at 7400ft. The terrain was too rocky to skin farther, so we ditched skis there and continued in crampons. We followed the ridge to 7900ft, then traversed following the standard summer route.

We crossed a long snow slope to a notch in the ridge, then across to the next gully. I recalled this distinct gully from the summer, since there is a constriction that you can almost touch both sides at once with arms extended. The snow in this gully was well-consolidated and made for very efficient step kicking. This was as we expected, since the gully faces southwest and the prevailing winds during the Friday snow event were from the southwest, meaning the gully shouldn’t get wind loaded.

I was feeling strong and ended up leading the way all the way up the snow gully, then up the next gully when it turned left. I had recalled some third class scrambling here in July, but that was now all buried in snow. The last gully ended 50ft below the summit, and from there we made a short rock scramble to the top by 10:30am.

Nick had carried the 30m rope and rock rack up the whole way, though we never ended up needing it. This is hard to predict in winter. Both Tower and Azurite are third class in summer, but the summer route doesn’t always go in winter, and with no known previous winter ascents of Azurite we didn’t quite know what to expect.

Unfortunately there was no view, so we snapped some quick pictures and headed back down. We carefully downclimbed the rock bit, then made quick progress downclimbing the snow slope to the base of the gully. We traversed back across the west face to our skis, and put the skis on in the cold and wind. My boots were a bit messed up and one couldn’t switch to ski mode, so I had to ski leaning backwards in walk mode. I was careful and made it work, though.

We made fun turns down the icy slope to the pass, then had fun powder skiing below that. We followed our up tracks back to the avy swath, then had more fun turns in the slide path. By 12:45pm, approximately on schedule, we arrived back in camp.

Talon had had a nice rest and we quickly broke down camp and headed out. The temperature had risen to close to freezing and our ski skins were starting to glop. But we applied skin wax and eventually the snow got powdery as we got higher. It was nice to be able to follow our previously-broken trail getting out.

After a few hours we made it back to snow lakes, and skinned across to the other side. It was 5:30pm by then, and it looked like the ski out would be in the dark. We skied a few turns down, then, surprsingly, we encountered other ski tracks. They looked like they were from that day.

I hadn’t realized that bowl was a popular place to ski, since it is so remote. Down lower at the edge of treeline I saw two orange flags in the ground where the tracks ended. That meant it was heli-skiers. I guess they are allowed to ski pretty much anywhere in the national forest, so that area was actually not too hard to get too. It’s funny that, after 57 Bulger peaks in winter so far, the only other people I’ve ever seen on the mountains have been heli skiers (on silver star and big snagtooth).

We made fun turns down in the woods, but things got difficult when our tracks started traversing. We wanted to follow our up tracks to reduce risk of getting cliffed out or running into dense bushwhacking on an unknown route down. But the up tracks occasionally went up and down or traversed. That is no problem on skis, but is very difficult on a snowboard. Talon had a rough time getting down, at times postholing, skinning, and boarding.

Eventually we reached the flats, and we all skinned from there. The slopes were too gradual to ski and ride out, so skinning was the only option. We took our time and eventually reached the snowmobile by 9:30pm.

Interestingly, there were no new snowmobile tracks over ours, meaning we were the only ones on that stretch of highway 20 all weekend. I assume the pavement turns a lot of people around coming from the west, and the liberty bell avy slides turn people around from the east. Also, it’s pretty far in from either side (at least 20 miles) so probably isn’t a very popular destination.

We got packed up and rolling by 10pm. Talon was feeling strong enough to get towed, and I kept it to a constant 20mph. We made it to about mp 140 before we hit patches of pavement. To increase efficiency I decided we would just all three ride on the snowmobile for the rest of the way back to the truck from there. Talon and Nick got on and I deployed the wheels. We continued that way for the rest of the road, stopping at all the pavement sections to deploy the wheels then retracting them on the other sides.

This time we were much more efficient, making it to the gate at 11:15pm. I made a few turns to rev the engine, then rode back into the truck. We headed out by 11:45pm, and after dropping Nick off and dropping the snowmobile off back at storage I made it home by 4am. That gave me enough time to get a one hour nap before getting up to give my morning lecture.

Movie of the trip:

© 2023, egilbert@alum.mit.edu. All rights reserved.