Idahodarod ski-packraft traverse

     Whenever I propose a new route, I am immediately suspicious of this type of response:

 'That looks like a good idea, I think I'm in!'

    However, even I couldn't shake the notion that a 300+ mile traverse of Idaho wilderness using skis and packrafts, in early springtime, might be a good idea. 

  In the end, we were largely successful from start to finish with our core group of 5 skirafters, though our success may have relied in large part on a miraculous two week long near-perfect weather window. A ski traverse along the spine of the Sawtooth mountains delivered us to Marsh Creek into the Middle and Main Salmon. Following a 6000' climb out of the Main canyon with packrafts in our packs, we skied nearly to the banks of the Selway River, our final leg of the journey. The Sawtooth, Frank Church River of No Return, and Selway-Bitterroot Wildernesses lay underneath our boots, boats, and pillows nearly the entire route length over 14 days.

blue = paddle, red = ski, green = wilderness areas

Sawtooth Range

Marsh Creek

Middle Fork

Climbing out of the Main

Selway River

  My original motivation for this trip, on the heels of a handful of successful skirafting trips the previous year [Grand Canyon, San Juan Mountains, Absaroka Mountains, Henry Mountains], was to tackle a true packrafting expedition with Mike in his new backyard in central Idaho. We settled on the 2nd week of April as our starting point for several reasons: 1. minimizing the dryland mileage (ultimately <10 miles!), 2. springtime snowpack stability featuring high likelihood of overnight refreezing, 3. Marsh Creek flows just high enough, but Selway flows that weren't yet absolutely raging. 

    And so we arrived from Idaho, Colorado, Utah, Montana and Colorado at a point just south of Stanley and saddled up for a slog-to-climb into the Sawtooth range.

  Allen and Brian B would leave the group after this first leg, while Mike, Jesse, Tom, Brian G and I would push on for another 9 days into another climate zone ending in Lowell, ID just above where the Selway joins the Lochsa river. Will later joined us for the Selway, skiing in solo from Montana.

First of thirteen passes on our Sawtooth traverse

Snowgeese heading north

    There are very few places where you can travel for 300 miles in the lower 48 and only cross 3 roads (one highway, two dirt roads). The highway separating the Sawtooth Range from Marsh Creek (Middle Fork Salmon headwaters) was our only resupply for the trip - and here we had stashed our paddling gear in advance. Skis loaded into boats, donuts and fried chicken down the hatch, and a direct launch from snowpack into the creek.

And onto the Middle Fork Salmon, leaving winter behind for the moment. Sort of.

    It was sunny, but it sure wasn't very warm. Spray was still flash freezing to our faces and pfds even after lunch most days.

    Green grass along the Main Salmon felt very out of place, having come straight from the Sawtooth Range and a freezing river leg. And for that matter, maybe some of us felt a bit out of place, packing up boats and skis into giant packs on a backcountry lawn.

    Carrying everything on our backs, even just halfway up the 6000' climb from the main before reaching snowpack, it was hard to not feel like a caravan of clown cars.

nearing the transition to skis, climbing away from the Main

What an exemplary, tight pack Mike!

two full days of ski traverse along Waugh Ridge

Perched camp overlooking the Bitterroot Mtns

    The morning after our final snow camp we descended to the Selway, and we could finally start to relax. Or, at least those of us not fully intimidated by a healthy flow and a juicy set of rapids that lay between us and civilization. For the second time we would be sitting in our boats rather than carrying our packs, now to the finish line. Mercifully, the skittery snowpack covering the road had a fresh few inches of powder padding to make our lives easier with heavy loads.

    At one point on the ski descent, while in the lead, wolf prints appeared on the track in the snow during a brief flurry. I bent down to inspect, and immediately said, 'These are less than 10 minutes old'. I took another look... 'No, probably less than 2 minutes old!'. I skated ahead around the bend, and after 100 feet they veered into the woods.

    Whereas Marsh Creek and the Middle Fork Salmon are barely above base flow mid-April, the Selway is already cranking.

It felt great to end a trip on a river leg - sunny, green, and wet... but still cold. Spring has barely begun.

Leaving camp before shoes thaw, last morning

Straight up out of gas

       It certainly didn't feel like our trip ended too soon, but the sense of accomplishment of pulling off a trip of this magnitude was downright awesome. A fluid link of wilderness snowpack and the waterways it feeds is such a satisfying concept - and thankfully parts of the west contain enough intact wilderness to make that possible.

Paria River Flash Flood

    The Colorado Plateau sports a number of gorgeous sandstone canyons that are routinely to infrequently paddled - Cataract and Labryinth are runnable year round, while Muddy Creek, the San Rafael Black Boxes and the Virgin River system have semi-reliable snowmelt runoff or releases after good winters. Paddlers debate the merits of a predictable spring or fall scrape/walk/float down the Dirty Devil, or occasionally catch a few thousands cfs on a summer rain event. Some small drainages like Dark Canyon, Courthouse Wash, and Cottonwood Wash have been paddled, usually by locals. The Paria River, spanning terrain from Bryce Canyon to the head of the Grand Canyon, though gaged, produces exceptionally fleeting floods given the extent of its watershed.

A 2017 Sorbet Soiree on the Paria

                    And now, from 2021:

    I live between 4 and 5 hours drive from the Paria, but if I wait for a flood to show up on the gage the river is reduced to a trickle by the time I arrive. So, the two types of events that produce marginally predictable flows: summer monsoon rains and winter rain-on-snow, must be anticipated. This is a difficult, frustrating and time consuming process that I don't want to get into the weeds about here. Suffice to say, I've now made it work for 2 out of 4 attempts based in Durango.

     The first success came in the winter of 2017:

(click for old video from 2017)

with Flows ranging from 120 cfs (narrows) to 200 cfs (boulder gardens) to 300 cfs (exit)

My initial trip spanned Feb 10-12. Shown: three diurnal snowmelt pulses followed by rain-on snow

    In mid-August 2021, an active phase of the monsoon created an opportunity. Usually a single dose of rain is too difficult to predict with enough geographic accuracy to justify driving for 5 hours, but the Paria basin had gotten hit with three runoff-producing events in 24 hours and a fourth was on the way. With soils/sand/substrate likely saturated, Tom and I decided to gamble on a trip. We drove out as the fourth rain event was occurring, and it looked like only slightly promising accumulation compared to the prior events. 

    Boy were we surprised when we reached the put-in!

Fourth rain spike arrived around 9am Aug 19. The Paria had only been that high twice since 2014.

    We didn't see that the gage had spiked at over 3000 cfs until after we finished the run (data reporting lag), we only knew the last report was 800+ and rising, but clearly it was going to be high. We weren't quite on the peak, but damn close and rode high water the whole way to the first serious rapids near the end of the narrows around mile 25.

    How high? I went back through my footage from the 2017 winter run to compare water lines to this 2021 summer run, and here are some sketches from roughly the same positions:

Entering the Narrows

2017

2021

---------------------------------------------------

Slideblock Arch

2017

2021

---------------------------------------------------

mid-Narrows rockfall rapid

2017

2021

        These are all relatively wide parts of the narrows, and in the short slot sections I'd guess our 2021 flood had water 8-10+ ft higher than our 2017 run. Even at extreme flows, I anticipated the low river gradient would produce little more than boils and seams - cobble bar riffles in the narrows could be run on the inside of corners and the rockfall only occurred in wide parts of the narrows so holes could be dodged. I'd read reports of successful hardshell descents around 500 cfs, so how different could 800+ be? Well, even given that we were actually riding >2000 cfs, I was mostly right. About 99.9% right. However, one of the roughly 3 slot sections (all 6-10 ft wide) featured an eddyless and narrowing run-in that started with mild laterals but delivered us straight into a set of crashing waves that was downright terrifying (starts at 1:00 in the video below). 

    In retrospect, it was fortunate that we couldn't stop to scout because there was no time get frightened or psyched out... we were just in it and had to execute. A wave hit my face and gopro at video minute mark 1:18 and obscured the view, but there were another few crashing waves followed by a boof off the side of a large submerged boulder into a relatively placid mudpool.

  Here's POV video of the run:

(Click for Video)

    The first overwhelming part of this experience, for me, was the smell. The river emits a heavy odor of wet earth that just penetrates everything; it felt like it was accumulating in my lungs. As we worked our way down the braided approach to the narrows, dodging active log jams, we got our first doses of mud/sand/water droplets. A single drop in the eye was somewhat crippling, and we both came to the conclusion that a swim, or even flip, or even a crashing wave to the face, could be catastrophic. 

    The water was so thick with sediment that it behaved and looked and felt different even compared to other desert floods I've paddled. The San Rafael or LCR at high flows were nothing compared to this, though the lack of foam here that the LCR featured is still curious to me. The combination of both mud and sand made it difficult to grip my paddle shaft - I took to dipping my hand in the river and immediately shaking the sand (not mud) off, then using my muddy hand to wipe the sand off the shaft. Otherwise I had too shaky of a grip. Awful. At every stop I would have to shovel handfuls of sand, foam, sticks, needles and juniper berries out of my cockpit.

    The somewhat predictable nature of the cobble bar rapids in the narrows became routine - strong boils off the walls were at worst meddlesome for our downriver progress. 

    One of the river-running problems at low flow in 2017 was that eddies were mostly filled in with sand so eddy hopping was not entirely easy. In 2021, eddies were much more workable in the vicinity of rapids.

    Working through the Paria narrows, the river cuts progressively deeper into the massive sandstones, exposing more easily erodible units underneath. This has caused the tall sandstone walls to retreat away from the river, but at the same time allows large blocks from the cliffs above to fall into the river channel. This, in combination with the higher gradient (~80ft/mi), forms a 4 mile long crux section of III/IV or IV/V rapids, depending on flow. 

    The first set of rapids we encountered, bedrock ledge plus giant boulder affairs, were mostly flooded out wave trains (!), but due to time spent scouting and portaging the water was quickly draining out. Fortunately, at the crux boulder gardens towards the end the water had already dropped by 3'. We still portaged about 5 rapids overall.

recent high water mark visible on left boulder

    Part of our plan involved not spending time to set shuttle at Lee's Ferry, so before our hike out over Dominguez pass back towards HWY 89 we spent the night. Meaning we had no spare water to clean gear or ourselves until the next day...

    Seven hours out of the canyon, over the pass and down towards the highway the next day.

    Would I do it again? Hell no! I mean, maybe...

    If you made it this far and still thirst for muck, check out MC's post from their winter run in 2019. Small flow for the narrows but similar flows in the lower rapids, and they used real cameras.