The US Geological Survey publishes LIDAR scans of almost the entire United States. These LIDAR scans are taken from planes flying overhead, shooting millions of laser points per second at the ground and precisely timing their return to map ground elevation with high accuracy. A small proportion of laser pulses make it through vegetation on the surface and to the ground, allowing LIDAR to map the true ground elevation, ignoring trees and other vegetation. The USGS publishes 1 meter resolution elevation maps, freely available for anyone to download and use. The 1 meter resolution elevation maps are an astounding achievement of modern technology when you think about it: look around you at any single square meter of space outside, and the USGS knows (and will tell you!) the elevation of that little 1m square to within 50cm.

This high-res elevation data is very useful for finding cave entrances. Here’s an example of a small section of point cloud data in the Guads. What do you think this is?

There are several open source tools created by cavers to assist and automate the process of finding potential cave entrances with LIDAR. I created a script that uses depression filling to automatically identify sinkholes in these high-res elevation maps published by the USGS. This script identifies sinkholes and exports them to a list viewable in programs like QGIS and GaiaGPS, and it makes a hillshade map with sinkholes highlighted and color coded by depth for easy identification. The hillshade map looks like this:

The circle point and text label is for one of the automatically extracted sinkholes. What do you think the above highlighted area is? Do any of the readers of this blog know what that is? Have any of you been here?

Automated LIDAR analysis dramatically increases the efficiency of searching for cave entrances; however, it’s still a very manual process. After sinkholes are automatically identified, I still frequently spend hours poring over the data, judging what are the most promising leads based on LIDAR and satellite data, and planning routes to get to these remote places. And of course the hardest part of finding caves comes after, when you actually have to get yourself to the leads that the computer tells you to go to.

Another useful caver-made program for finding caves with LIDAR is FLEX by Felipe Roz Barscevicius. His program lets you directly view in 3D the raw point clouds from LIDAR, instead of hillshade maps derived from 1m elevation maps derived from the LIDAR. But it’s entirely manual—no automated entrance location identification. Something both of us should do is combine these programs to automatically identify potential cave entrances from raw point cloud data. But both of us are busy, with day jobs and caving and cave-finding to do!

Anyways, enough backstory. A few weekends ago, I checked some LIDAR leads deep in a remote canyon in the Guadalupe Mountains. I had been eying these leads for many months, but only recently got the time to backpack in there and check them out. The canyon, North McKittrick Canyon, is one of the few canyons in the Guads with year-round water, which allowed me to do this as an overnight trip instead of a heroic day trip.

I arrived in the Guads late Friday night, after midnight, after a long day at work. My intent was to enter North McKittrick Canyon from the west side, from below near the end of the Queen’s Highway as it enters Guadalupe Mountains National Park. After arriving there late Friday night, I camped just off the Queen’s Highway, then slept in til the late hour of 9am, when I was awoken by the already blistering hot sun.

Saturday, March 21

I got to my intended parking spot just north of the Guadalupe Mountains National Park boundary and started hiking at the properly alpine time of 10:42am. I carried 6 liters of water, as it was already crazy hot out by mid morning, and I wasn’t sure that the alleged water source in McKittrick Canyon would exist or be flowing, so I brought enough to make it back safely in case there was no water. The heat was telling me that the season for ridgewalking in the Guadalupe Mountains was rapidly coming to an end, but I was determined to get in a few more LIDAR lead checking trips this year.

From where I parked, I went due east, following the park/BLM boundary, until I got to right below the Guadalupe escarpment. Right below the escarpment, I started traversing north to work my way up the many cliff bands guarding the ridge top. Interestingly, I ran into a cairn and a trail here. After a few little scrambles up short cliff bands, I made it to the top of the escarpment. I followed this escarpment due north towards my first LIDAR point.

On my way north, I saw two Barbary sheep just below the ridge about 100 ft away from me. They ran away when we saw each other, so I didn’t get any pictures. Such graceful animals. I thought that this would bode well for the upcoming leads I was to check.

This good luck seemed to pan out, as the first LIDAR hit I checked was in fact a pit. It had a nice rectangular entrance, about 2ft by 5 ft.

I started downclimbing it and saw a floor beneath me, but it belled out a few ft above the bottom in an interesting rectangular shape. I was pretty sure I could see the entire room at the bottom, but I wasn’t 100% sure. I hemmed and hawed about downclimbing the last bit as it belled out and kept shifting around trying to find a position that would make the last few moves feel secure, until I gave up and decided to rig it. I brought 100ft of ultralight 8mm rope and my ultralight homemade no-buckles harness, so I might as well use it for a climb I was hesitating about. I rigged the pit using a bomber boulder just outside the pit and rappelled in.

I rappelled in and confirmed that it was, in fact, just a blind pit. Oddly, there were 2 Coors beer cans in the bottom. The pit is on a ridge and not somewhere where things would naturally drain into it, so I think these were intentionally thrown in. I did a quick survey and it’s 25 ft deep. I named it ‘Barbary Pit’ after the 2 Barbary sheep I saw on the way there.

After my quick survey, I descended into the North McKittrick Canyon. I looked forward to getting into the canyon bottom, as there were trees visible there that would provide shade from the hot midday sun. There was allegedly a trail in this canyon that takes you all the way to the Virgin Cave 2WD parking lot, although I was unsurprised to find that this trail totally does not exist. It was unidentifiable when I first got into the canyon bottom, and I did not find it anywhere along the way down the canyon except for one 20 ft section that bypasses a pouroff in the canyon. Whoever traverses this canyon seems to be using the same one spot to get around this pouroff.

I followed the canyon bottom downstream for about 2 miles as it gradually got better and better (wider and less bushwhacky). After those 2 miles I saw my first cliff lead of the day. I spotted one small cliff lead in my entire time in the canyon:

I moved on and continued frolicking down the now-wide-and-flat arroyo. I saw my first water near that cliff lead. This was a gross brown pool that I definitely wouldn’t drink from, but it was a good sign that more water was likely to come.

The arroyo eventually turned to bedrock, not gravel, and had many small pools embedded in the bedrock in a row. There were also many micro-springs seeping from the walls creating colorful zebra stripes down the limestone pavement. I saw many sets of these lines of micro-springs all at the same elevation above a pool These lines of springs must represent some sort of water table. Very interesting.

There were also some cool-looking recently dried pools. Some of these had a strange dry film/crust on them. From whatever life was growing in there before the pool dried up?

At 4:30pm, I got to right below the second LIDAR point I intended on checking. I abandoned the comforting embrace of the wide, flat, un-vegetated arroyo for the harsh world of the steep vegetated slope above me. There was a brief scrambly rocky section right as I got out of the arroyo, but after that it was smooth sailing up the bushy slope, at least as smooth as thrashing with an overnight pack can be.

That second LIDAR point did indeed turn out to be a cave. When I got there, I was excited by the clearly cavey entrance, although it seemed to end immediately.

There were neat eroded formations right at the entrance. It seemed like there was nothing more than just the entrance visible above, until I saw a hole between the leaves on the floor and the bedrock wall. The hole was smaller than me but I had a feeling the hole would be larger if I stuffed myself in there and compressed the leaves out of the way. That was indeed true, as the leaves compressed down when I stuck myself in the hole, and I found myself sliding down a small crawlway.

After a few feet of this crawlway, I was surrounded by huge stalactites that I had to carefully navigate to not break as I stepped down onto the floor of a standing room. I was psyched to be standing up surrounded by a hanging garden of stalactites in a walking passage, although just past me it ended. The passage slopes down with a breakdown floor until it meets a vertical wall where it ends. It’s likely a pit that got plugged up with breakdown.

Surprisingly many of the stalactites, even huge ones, were naturally broken. I think it’s from breakdown falling down the slope. Interestingly, there was a bat sleeping while hanging from the very tip of one of the stalactites. I’d never seen that before. For that reason, I named the cave ‘Just the Tip Cave’, after the bat’s sleeping location, and because the accessible portion of the cave is surely just the tip of the larger cave beyond the breakdown plug. The breakdown plug at the bottom has no air and is not a promising dig. There were 2 bats in total.

I surveyed it to 47 ft. After that, I was running low on daylight, so I rushed over to my next LIDAR point, knowing full well that I would be bushwhacking down and finding water in the dark. From Just the Tip Cave, I traversed east to the next sinkhole. This was nothing—just a jumble of rocks in between 2 big fins that made a depression.

From this LIDAR hit, I descended more-or-less straight down the SW-trending gully just to the west of the LIDAR hit. This gully, though brushy, was much more pleasant than the bushwhacking earlier in the day, as it was finally cool enough to wear pants. It was pitch black out by this point.

I was relieved once I made it back down into the large arroyo of McKittrick Canyon, as the route I took up out of the arroyo was a little sketchy at the start, and I didn’t want to repeat that. Thankfully, the route I took down was easier. Also, scrambling and bushwhacking around in the Guads always feels a little creepy to me when it’s pitch black out and I’m alone. So getting out of that zone was a nice feeling. Although my day was not over yet, as I still had to find a good water source. I was momentarily scared for a bit in the arroyo when I encountered strange loud animal sounds at a pool. They were super loud, but strangely I couldn’t see anything where the sounds were clearly coming from at the far end of the pool. After a few moments, I determined that these sounds were frogs. I didn’t spot any frogs, but there were many tadpoles in the pools visible. Listen to the video below to hear the sound. They were much louder in real life.

I continued downstream and kept encountering gross stagnant brown pools, which was a little disconcerting. I could drink from them if I really had to (I had treatment chemicals), but I didn’t want to. A dinner of oatmeal with a side of pond scum did not sound appetizing to me. Thankfully, I found one pool that was clean and clear. Serendipitously, it turned out to be the last pool before the water disappeared for a while. After finding this one clear clean pool, I continued down canyon for a bit, hoping to find water that was not only clean and clear but also flowing vigorously. I did not find that, as the water disappeared at the end of this pool and didn’t return in the quarter mile I continued down canyon. After later looking on the satellite maps at home, it appears that water would have come back after another few hundred ft, and likely would have been flowing. No matter; I was satisfied with the water source I had.

I settled for the clean and clear yet stagnant pool and set up camp. I officially stopped hiking and took off my pack for the night at 9:16pm. With my nerves finally calm after getting out of the death-scrambling-in-the-dark zone and finding satisfactory water, I set up camp. I played some Sabrina Carpenter to set the mood1 as I set up my bougie campsite on the pavement beside the pool. I had a luxurious canopy of blooming madrones and clear stars above me. Just to my side was a still clear pool. I imagined I was at a beachside resort on the Mediterranean. Maybe if it was warm enough I would go swimming the next morning.

I drank about 3.5 liters of water that day before making it to camp. So my 6 would have been enough to make it back if I had to. I made my dinner of oatmeal without extra pond scum flavor, which I much appreciated. Bringing dried food that can be rehydrated with cold water (and still tastes good cold) is nice on trips like this when it’s hot so I don’t really need hot food, and I’d rather save the weight of a stove. I went to bed just before 10pm, satisfied with a productive day of knocking off LIDAR hits, but slightly disappointed that they all didn’t go.

Sunday, March 22

I woke up at another positively alpine time of 8:30am. Man, the previous day must have really worked me to make me sleep for 10+ hours. I made an oatmeal breakfast identical to last night’s dinner, gathered and treated some water, and packed up camp. It was nice that packing up camp took all of 5 minutes, as I didn’t have a tent.

My only objective that day, besides making it back to the car safely, was checking a LIDAR point low down on the ridge separating North McKittrick Canyon from the adjacent Devils Den Canyon to the south. This point was only a little upstream of camp, so I got a brief warmup for my legs before having to charge up yet another evil steep vegetated Guads slope.

As I walked upstream in the arroyo, back towards the way I came, something caught my eye that I didn’t see the day before in darkness. There was a large shelter overhang right next to the arroyo. These large overhangs near water (that was likely flowing right here in the anthropologically recent past) are prime spots for Native American archeology, so I went up to it to see what I could find. The first thing I saw under the overhang was a large fire ring and some camp furniture, but upon closer inspection of the walls of the shelter overhang I was psyched to find pictographs! I spent some time admiring and photographing them while thinking about what it would be like to live here. I bet this place would seem luxurious in a time before such technological marvels as shingle roofs.

A few minutes past the pictographs, it was time for me to charge up the hill to my south. I was dreading this moment, as it was sunny and hot by this time (10:15am). Also, I would be going up a north(-ish) facing slope, unlike the south-facing slopes I ascended yesterday. North-facing slopes in the Guads are generally worse, vegetation-wise, than south-facing slopes, as the denser and more expansive plants seem to like it more there.

I sucked it up and started up the hill. It was even worse than I expected, as some parts of it were cliffy and scrambly, and I had to pull a few rock climbing moves while crashing into dense Gambel oak that was hell-bent on repelling me. Also, oddly, I kept kicking up fine dry dust from the ground when I disturbed it, and the dust would get in my mouth and give me this horrible bitter taste. A few times I had to stop and gargle some water to get the taste of the dust out of my mouth. I’d never encountered that before.

After 400ft of elevation gain that took about an hour, I arrived at the LIDAR lead of the day. It was another pit! I was a bit beaten down, and short on time if I wanted to get back to Albuquerque at a reasonable time, so I considered leaving the pit for another day. It would be nice to have an excuse to camp at that wonderful campsite again and bring more people there However, I talked myself out of that silly position when I thought about doing that horrific bushwhack up the hill again.

I tied the rope to a good tree right next to the pit and went on down. Alas, it did not go. There was a little bit of passage at the bottom. I surveyed the whole cave to 58ft. There was one bat. I named it ‘The Tannin Pit’, after the strong bitter taste of the dirt/dust I kicked up hiking to the cave.

The Tannin pit was my last objective of the trip, so it was time to go home now. I found a slightly better route down the slope to the arroyo (at least, it felt better, possibly because I was going downhill so I could use gravity to assist me in crashing through the bushes). The hike out was mostly uneventful, besides the crushing heat of course. I felt like I had to stop for water every 20 minutes or so. I did find this cool javelina jawbone:

On the way out, because I didn’t have to go by the LIDAR hit now known as Barbary Pit that I checked at the start of my trip, I took a more direct route to the main escarpment dividing the west and east sides of the Guads. I was a little nervous that my more direct route would make me get cliffed out when going back up to, or back down from, the very top of the escarpment, as there were many little cliff bands guarding the escarpment there. However, I had no trouble crossing over the escarpment further south than where I started the trip. My last stressor of the trip was gone when I descended below the lowest cliff band guarding the western slope of the escarpment, at which point I took the following panorama:

I made it back to the car just past 4pm, in time to get to Albuquerque at a reasonable hour before work the next morning. As I drove back, I took a quick swim in Brantley Lake State Park. The water was cold and refreshing. I also got some ice cream in Roswell which was similarly cold and refreshing and hit the spot.

The Guadalupe Mountains (or “the Guads” as they’re often called), which have some of the nicest caves in New Mexico1, are a rugged, wild, and unforgiving mountain range. They are full of fins and spires, layered in a colorful maze that makes traveling in the range difficult, but always a worthy challenge.

Getting around the Guads typically means sticking to an arroyo in a canyon bottom if approaching from the east side/from below, or sticking to a broad ridge if approaching from the west side/from above. One can drive around to the top from the west, where the mountains gently slope up from below. But on the east side, the mountains dramatically plunge down towards the canyons, forming forbidding series of cliffs that make traveling between two different elevations not straightforward. Getting from a high ridge to a low arroyo can be a puzzle that can be solved with any combination of walking, scrambling, or rappels.

Another factor that makes connecting the ridges and the arroyos complicated is the vegetation. The ridges and arroyos are generally not heavily vegetated, but venturing off either of those paths into the slopes between them brings on bushwhacking. The desert environment of the Guads means the vegetation isn’t incredibly thick, but it is punishing in other ways. Cacti, agave, and other plants have vicious spines and thorns to guard against animals that would eat them for their water. Bushwhacking through these spiky plants requires some care.

Back in March, I found a cave in the Guads using LIDAR. It showed up as a series of deep sinks at the base of the cliff, which suggested a cliff entrance more than a sinkhole entrance, which was exactly what we found when I brought Ben, Adia, Liam, and Rosie Tobin out for a 10 mile round trip hike to check it out. We found a 30 ft wide arch full of breakdown. Ben Tobin found a way through the breakdown and into a walking-sized space where we were greeted with massive dogtooth spar crystals.

Upon returning to civilization, we learned that we had refound an old lost cave, Big Canyon Boneyard Cave, that Donald Davis and Rick Rhinehart found in the 80s. It never got mapped, so I returned to continue surveying the next November.

On the next trip this past November, I dragged 5 people out on the 10 mile round trip hike to the cave. Many 5-10 shots into the breakdown meant we didn’t rack up that much survey footage. Since we had 5 people, one too many for a survey team, one of them, Corey Hackley, ditched us to check a nearby LIDAR point. When he rejoined us several hours later in the cave, I was shocked and excited to hear that he found multiple deep pools of water in Big Canyon, just upstream of where we left the canyon bottom to go to the cave at the base of the cliffs in the canyon. We hadn’t yet finished the cave, so I knew that on my next trip out here, I would camp overnight at that water instead of doing a marathon day trip with 10 miles of hiking and several hours of surveying.

And that’s exactly what I did. Two weeks later on Saturday December 13th, I brought six people out to the cave. Travis Broadhurst, Kayla Navarro, Bram Lovelace, Evan Peairs, and Anastasiya Razumovska. I was determined to survey until we killed the cave, as seemed likely, or if the cave actually ended up going, then survey for many hours, not saving any energy for a long hike back.

When we got to the cave, we split into two survey teams to work on the two areas in the caves with leads (really, gaps in the breakdown I hadn’t gone in). All of the leads in this area ended almost immediately, unfortunately. So the cave was dead! We spent another hour or so poking and prodding into every hole in the breakdown/boneyard until we were totally confident that the cave doesn’t go.

We got out of the cave, packed up our stuff, end enjoyed the Geminids meteor shower for a few moments before moving on to our next objective. On the last trip out here, I had refound another old cave that Donald Davis found in the 80s, based on someone’s description of where it should be relative to Big Canyon Boneyard Cave. This cave, Little Darling Cave, had a lead in it past a tight flowstone constriction. I brought everyone 300 ft north along the cliff band to this new cave, and we started surveying. Little Darling Cave was a much easier to survey and sketch, as it was actually one clear passage instead of a jumble of breakdown and boneyard passage.

Little Darling Cave starts as a walking tube that turns into a large decorated flowstone chamber. As often happens in the Guads, I was surprised to find water streaks and drips within 100 ft of the arid desert air outside. Where does this water come from!?

When poking around all the walls, we actually found two leads beyond tight flowstone constrictions. These took some special care to pass through without breaking the formations that made the constriction. Both leads led to some wonderfully decorated passage beyond, but they didn’t go. The cave totaled 155 ft.

With both of the cave objectives done for the night, it was time to pack up, head back down to the bottom of the north fork of Big Canyon, and head upstream to our alleged water source. I hadn’t seen this water source myself, and all of us were getting to the end of the water we packed out here, so I was slightly nervous about actually finding water for everyone.

We got out of the cave around 10pm, just in time to make it to camp at the peak of the Geminids that night at 11pm. The hike down the steep vegetated slope below the cave was unpleasant as always, but made much more pleasant by the thought that we wouldn’t have to hike 5 miles back to the car after this. Just a half mile or so upstream in a flat arroyo to some water, hopefully.

After 10 minutes of walking upstream from where we regained the arroyo below the cave, we were delighted to find a pool of water. This pool was brown and stagnant, so we wouldn’t want to drink it, but it was a promising sign that more and better water was to come.

A few more minutes upstream of that first pool, we started seeing more frequent pools. Then the pools started to be connected by a small flowing stream, which made the water clear and clean looking. Once we found a series of flat spots fit for six people to sleep (at 32.04087, -104.75269), we triumphantly dropped our packs and began setting up camp. Just beyond this spot, the water got deep and the canyon walls got tall and narrow, which would have made advancing further without getting wet tricky. It seemed like this campsite was put here perfectly for us.

We cooked our dinners by the water while the Geminids raged above us. Falling asleep to the meteor shower was an excellent end to our long day.

We slept in the next day and woke up when the sun rose over Lonesome Ridge and hit our campsite in mid morning. We took our time packing up camp and making breakfast, as we wanted to enjoy every last moment in this unique place in the Guads. When it was finally time to go, we began down the arroyo towards some cave leads I wanted to check near the mouth of Big Canyon.

When we got to right below the points I wanted to check, we split up. I went up the hill to check a dark spot I could see from below, and the others checked two dark spots below that. We reconvened a few hours later after no one found anything that goes. One of the points I had in mind ended up being a 15 ft long cave with formations, but it didn’t go.

From there we continued on out of Big Canyon. We made it back to the cars right around sunset. On the drive back Sunday night, we saw a pack of wild javelinas cross the road. I even got pictures and a video of one of them. Look at him go!

I’ve since heard that there is reliable water in the north, middle, and south forks of Big Canyon, and in the north and south forks of McKittrick Canyon. I’ll certainly be going to those canyons to ridgewalk and enjoy some more nights out in the Guads.

Subscribe now

The objective of this trip was to explore in the MK Ultra area of the cave, where many leads remained after the previous trip to this area in 2024. The MK Ultra area is what the Borderlands passage becomes when you go past MK100 or so. This whole area is the MK survey, and the far end earned the creative name MK Ultra on that 2024 trip.

On my last Fort Stanton trip back in June, we spent our third day in the cave exploring a lead at the start of Borderlands at MK64. This lead went for 2000 ft before we turned around at a four-way intersection where three walking passages continued off into the unknown in different directions. The rest of Borderlands, and the MK Ultra area beyond Borderlands, was apparently full of leads like that one. And the La Culebra Passage, which is the large pleasant borehole between the Midnight Junction camp and Borderlands, is wonderful passage that I love traveling in, so I was eager to see the passages that continued beyond La Culebra into Borderlands and MK Ultra.

This trip was with Garrett Jorgenson Olauge, Adam Weaver, and Rene Ohms. I had caved with Garrett previously on a three-day camp trip in Lechuguilla, but I had never met Adam and Rene. I met them in Albuquerque after they flew in from South Dakota, and I had one of those absurd moments that are very unique to caving: “hi, we don’t know each other, nice to meet you, now let’s go spend four days extremely far underground exploring the end of the world together”. I love moments like that.

The far reaches of Fort Stanton really do feel like the end of the world. We would travel over 12 miles one way underground to get to the MK Ultra area, where the original exploration would begin. There is no GPS to tell you where you are, and no way to contact the outside world except by spending a full day traveling underground to get to the cave’s single entrance.

Garrett, Adam and Rene do tons of Fort Stanton caving, and I was very much looking forward to caving with them, to see how the pros do it. I knew they travel fast, and they get a lot of survey (mapping new, unexplored passages) done. I like to do all of those things, and I wanted to see how much they/we could get done on a trip like this where there is so much travel to get to the far reaches of the cave and there are incredible leads to survey.

We got to the Fort Stanton bunk house in the evening on Wednesday, August 13th, the night before our trip. As usual, we frantically packed our bags that evening, spent the whole night wondering and worrying about what we might have forgotten, then unpacked our bags the next morning and looked at everything and packed them again, having not forgotten anything. Then we were ready to enter the cave Thursday morning, confident we were prepared with everything we needed for the next four days we would spend underground.

Thursday, August 14th

We entered the cave at the bright-and-early alpine start time of 9:31am. Man, I love that it’s always dark in the cave so there’s no need to start early. I saw firsthand that Garrett Adam and Rene really are fast and efficient cavers, as we made it to Midnight Junction very quickly, wasting little time and making few mistakes. We made it to Midnight Junction at 5:22pm, 7 hours and 51 minutes after entering the cave. That’s much faster than the 13 hours it took on my first Snowy River trip and the 11 hours 40 minutes it took on my second trip. Minimizing time spent at the changing stations, turning stops for food into quick snack breaks instead of proper meal time, power walking down the easy parts of Snowy, and generally having a sense of urgency all helped us make light work of the trip to Midnight Junction in under eight hours.

I didn’t take any pictures on the way there, so I’ll save you the glamor Snowy River shots and the explanation of how this travel works, which is unique and bizarre even by caving standards. See my previous Fort Stanton trip reports to read all about and see pictures that demonstrate why traveling on Snowy River is such a surreal experience with its own unique rewards and challenges.

Midnight Junction is the standard camp in the cave. It is far out there by caving standards, at 8 hours if you’re fast or 12+ hours for slow teams. It’s a good thing we made quick time to Midnight Junction, because at that point our day of traveling to camp wasn’t even close to over! We were to bivy that night at Camp Wolf, a bivy spot 2-3 hours beyond Midnight Junction. I say bivy, not camp, because Camp Wolf has no fixed camp gear left there. Right now it’s just an empty passage. This is unlike Midnight Junction, which is a proper camp that features permanently stashed sleeping bags, sleeping pads, ground tarps, stoves, among many other miscellaneous camp supplies.

Because Midnight Junction has camp gear stashed there, trips camping there can go fairly lightweight. There’s very little camp-specific gear you need to carry. Stove fuel and your personal mess kit and toiletries are just about it. You don’t even need to pack separate camp clothes, as is common when caving, as your clean-mode clothes will be dry and clean enough for camp, and the cave is warm enough (57 F) such that extra warm clothes are not necessary.

But the far out exploration trips bivy beyond Midnight Junction nowadays, as the frontiers of the cave have gone so far beyond Midnight Junction (4+ hours) that you need to camp further out if you want to get much done. This means much more to carry: a sleeping bag, pad, ground tarp, and stove. Thankfully, extras of these supplies are stashed at Midnight Junction, so you only have to carry them from there. At Midnight Junction, you can also ditch your clean pack and clean clothes, as all the travel is in dirty mode from there. That at least frees up space in your cave bag for all the bivy gear you’ll have to carry. Carrying all the bivy gear was going to be a challenge, but it was a challenge I was looking forward to. I was excited to do a really far out trip in Fort Stanton that necessitated bivying.

At Midnight Junction, we ditched our clean packs, which were the innermost packs that all our stuff was packed inside of. This meant we had to completely unpack our clean packs and repack everything into our dirty packs, which previously had been either rolled up inside our clean packs or covering the outside of our clean packs in dirty mode. We left our clean packs, and various small items we wouldn’t need until the way out several days later, on the camp ground tarps at Midnight Junction. After we spent a little under an hour eating, resting, repacking, and getting water at Midnight Junction, we were off, ready for several more hours of caving before we could sleep.

From Midnight Junction, we turned left down the non-Snowy passage and abandoned Snowy River until it was time to exit the cave three days later. A minute later we were at a T intersection, Harmony Hall. On the left from this T intersection is Midnight Creek, a faint stream that is the water source for Midnight Junction. The This Is Ridiculous passage, which we continued exploring on my last trip in Fort Stanton, branches off Midnight Creek on the left from Harmony Hall. To the right is the La Culebra passage, an impressive borehole that was our way on this trip.

We followed the La Culebra borehole from MK1 to MK65. Here, the lower level borehole ends in a breakdown plug and we enter Borderlands. This spot is also where we started surveying the low belly crawl that eventually turned into three walking leads on my last trip here with James, Beth and Ben. It was also the furthest I had been down this section of the cave, so I was excited to finally be in new-to-me territory and see what lie beyond La Culebra in Borderlands and MK Ultra.

Past MK65, the passage character changed significantly. As we climbed up above the breakdown, we abandoned the pleasant flat sandy floor of La Culebra and gained a series of multi-level canyons and confusing mazes. Many ups and downs and turnarounds followed. The route was lightly flagged, as there had been only two trips out here ever. We added some more flagging as we went, for our own good, although Garrett had the feeling that we wouldn’t come out this way, that we would connect to one of the three walking leads we left from the previous trip. Those walking leads were all right underneath Borderlands, and there were lower-level leads further out in MK Ultra heading this way, so connection seemed like a real possibility.

After eight stations that felt like forever due to the confusing level changes and turnarounds, we found ourselves at our first lead and survey objective of the trip, at MK73. There were many mop-up1 leads in Borderlands on the way to Camp Wolf, and we were planning on surveying them on the way to camp. We ditched our bags and got in survey mode. I took the inventory and photography roles. From MK73, two leads branched off on opposite sides of the main travel route. First, we took the one heading back towards where we came from, north as we were pretty sure it would connect back into known passage. For a mop-up lead, it was pretty fantastic: a huge walking canyon, 5-10 ft wide and 30+ ft tall.

I’m glad I got the photography role, because the passage quickly transitioned from big walking passage to beautiful, sparkly, decorated walking passage. Where we started, we could see nice velvet flowstone and columns down to our side. Then the passage quickly put us right in the middle of the velvet garden.

At MK73C we donned shoe covers and ditched our gloves as we had no choice but to walk on the gorgeous, colorful velvet cascades. Here the velvet became thick, streaky, brilliant yellow, and full of other formations like chenille spar and helictites. We surveyed carefully, minding what surfaces we touched and set stations on.

We were stopped after six stations by a huge cliff in the floor. We could see a big booming canyon beyond the cliff, and I got excited at the prospect of finding some way around the obstacle in front of us. But we then saw a station on the other side, and we realized that this passage was one of the leads we saw earlier in Borderlands. I even remember, as we were doing a U turn and climb up around, looking straight ahead and wondering why we weren’t continuing into the huge wide canyon passage ahead of us. Then I saw the chasm separating us from that obvious way on. “That’s a lead” Garrett told me before we continued on our sinuous route full of climbs and turns. From the other side, I recognized that lead, that we had now just killed.

With one of the two leads at MK73 killed, it was time to take the other one, heading west. This one was another huge walking canyon, which would be an incredible lead in any other cave than Fort Stanton, where it was merely a mop-up lead as it seemed likely to connect to known passage. Garrett was confident enough that it would connect to known passage further along the route to Camp Wolf that he even told us to wear our camp packs as we surveyed. That was a unique experience for me: I had never surveyed while wearing all my camp gear, expecting to stop for the night and sleep somewhere beyond the virgin passage we were exploring.

This canyon quickly turned left and started heading south, paralleling Borderlands. Eventually, breakdown in the floor forced us up onto a ledge in the side of the canyon walls, where we encountered awesome thick mud cracks. In the previous MK73 lead we got amazing flowstone formations, and now we were getting amazing mud formations—what a combo!

This pleasant large walking canyon connected back to main Borderlands at MK75, and we tied in, but our side canyon kept going, paralleling main Borderlands to the west as nice decorated walking passage. We continued surveying with camp packs on until we came to an intersection at MK73O. An intersection, in this big walking passage that was supposed to be a mop-up lead! Maybe this passage really would go somewhere significant, I thought.

From the intersection at MK73O, we went left down an attractive walking canyon (again!). I quickly saw a station and a stream below us, and I knew what was about to happen: we were about to connect to the walking stream passage from my previous trip! We ended up tying in to ML20, not ML40 where we turned around at the complex of three walking leads. Once I got down there, I remembered that lead crystal clear from 1.5 months ago: as we were walking in the stream passage, we spotted an extremely clean flowstone covered canyon heading up and right away from us. We wrote it off as a lead requiring a clean set of clothes (i.e., unlikely for anyone to do any time soon). However, it turns out that lead was actually two leads: the one heading south that we saw and marked on the sketch, and the same passage but heading north, that we didn’t see because we were heading south when surveying that passage. We connected in from the north side, which was decorated but not so pristinely clean as to require a full outfit change to clean clothes. The lead continuing south was definitely much cleaner, although it’s possible that one might be able to get it done with just clean shoe covers and gloves. After tying in to ML20, we left that continuing passage for another day, as we didn’t feel like all getting into (semi-)clean mode.

We returned to the intersection at MK73O and went right. This went for a few more stations before connecting back into main Borderlands again at MK73V. Garrett was right that this lead would connect back in to the way on to Camp Wolf! We were all glad we brought our camp packs through there.

Our next lead was at MK81, just a short ways on in Borderlands. This was really two leads, one going left (north, back the way we came) and right (south). We went left first, which was nice decorated walking passage, although it quickly connected back into Borderlands. The passage going right/south also quickly ended.

The next lead we did was at MK84, which was really a complex of leads where the right/west wall of the passage seemed undefined for a bit. One of these led to a crawl along a beautiful shiny flowstone river. This one was smooth and polished and had a clear direction of flow, unlike Snowy which is more of a pool and has a rough cauliflower texture. We donned our clean shoe covers and began down the flowstone river. Micro-pools full of chenille spar dotted the flowstone, and we soon encountered bigger pools and a formation forest we had to weave our way around.

This decorated flowstone river quickly turned into a crawl, and we had an awkward maneuver getting off the river and onto a dirt bank while in a 2 ft high crawl. While planking in a superman position to keep our dirty clothes off the pristine white flowstone, we had to put on gloves, get our hands on the dirt banks, superman forwards, and reach behind us to take our shoe covers off to get our shoes on the dirt. All while keeping our shoe covers clean, and all while wearing a camp pack and carrying survey gear. It was quite the athletic sequence of moves. Took some core strength.

Right after that maneuver, we reconnected back into Borderlands. We had one last mop-up lead to do at MK94. This one started out as walking passage, although it quickly turned into a crawl. One long shot in this crawl led to a big collapse rotunda room. Breakdown filled the entire room, although there was one spot that seemed most obvious as a way on. We could see void space beyond the breakdown there, and we spent about 10 minutes moving rocks to try to get in there. We gave up when we got to a rock that’s unmovable without a crowbar.

At that point Garrett decided we were done surveying, and it was time to head straight to camp. It was 11:43pm at that point. We had been caving for 14 hours with camp packs. We continued onwards in Borderlands, which at this point was still largely walking, although not the easiest as there were many ups and downs in the breakdown. At MK98 there’s beautiful flowstone slope that requires shoe covers. Around MK115, we left the decorated walking passage and climbed up into an obscure hole in the wall that took us to a crawl. Here the travel to Camp Wolf changed from mostly walking in large decorated borehole to more typical higher-effort caving.

We rigged a handline for that climb, as it required a balancey rockover move that would be extra tricky if we had to downclimb it. After this crawl the caving became a blur and blended together in my mind, because we had been caving for so long. I recall it was smaller passage with a mix of crawling and stemming.

We popped out of a popcorny canyon into a dramatic, interesting four-way intersection at MK135, and Garrett was eager to narrate what each of the three new directions did. Straight ahead went to Camp Wolf, which was just a few minutes away. Right led to MK Ultra, to the most remote point in the cave. Left went back towards Borderlands and was full of leads. Tomorrow, the plan was to go right towards MK Ultra and survey the many leads there. The day after, we would go left towards the passage with tons of leads heading back towards Borderlands. Garrett suspected that some leads there would connect with the dry walking tube beneath Borderlands that we turned around in at the end of James’ trip back in June. But tonight, it was time to head to camp, so we went straight after taking a moment to label the four-way intersection with flagging tape. We got to camp at 12:55am, after over 15 hours of caving.

Camp Wolf is a small walking tube with a flat dirt floor. It’s just barely tall enough to stand up in, and you even have to duck to go between the different sleeping spots. Still, it’s a pleasant, comfortable camp, and I knew I was going to enjoy my stay there as soon as we rounded a corner and Garrett announced we were sleeping here.

The smooth dirt floor was much flatter than that of Midnight Junction, which slopes ever so slightly towards Snowy and constantly tries to roll you out of your sleeping spot. Another feature Camp Wolf has that Midnight Junction lacks is rock furniture: shelving to organize your stuff on, and rock protrusions you can sit down on. I can never sit comfortably on a flat ground, so little rocks to use as chairs at Camp Wolf made the experience much more comfortable. And tables to cook and pack on made all the little camp tasks a little more ergonomic and pleasant.

After everyone scouted the camp and picked their spots, we all unpacked and hastily got ready for bed after our 15+ hour day of commuting and surveying on our way to camp. The only thing standing in between us and bedtime at that point was making dinner. Unfortunately, one of the two alcohol stoves we brought, the one that Garrett and I ended up using, always took forever to heat water. After waiting an eternity to get hot water for my dinner mix, I slammed a huge bowl of it and went to bed at 2:40am.

Here are my times and other miscellaneous notes I recorded on the way in:

Friday, August 15th

Garrett woke us up nine hours after bedtime, at 11:40am. We made haste in the morning and were out of camp at 1:06pm, after I finished choking down my huge bowl of dinner mix that was my breakfast. I ate dinner mix for every meal this trip, as I often do when cave camping.

I had heard that the trip out to the far reaches of MK Ultra was fairly hard, by Fort Stanton standards at least, so I was ready for some real caving. We had a few hours of crawling in low bedding planes and cobble crawls, stemming and climbing in grabby popcorn canyons, tricky route finding, and some nice walking borehole to top it off. It felt like Kentucky caving, as the cave would change character to something totally different, say from stemming to crawling, then stay that way for a long time. And the cave was much wetter than anywhere else in Fort Stanton: ceiling drips and flowing water covered much of the passage out here, at least when we had a floor beneath us.

On the way out, we rigged one handline at a tricky climb around MKA25-30. This handline was for route finding as much as it was for helping with the climb, as the spot where you climb up into the keyhole at the top of a narrow fissure is a little obscure and easy to miss. Near this handline, we surveyed one small “lead” that was really just a lower level of the keyhole passage that we traveled in above. After the handline we had a long stretch of low belly crawling before popping up into the Glorieta Borehole.

At another complicated intersection area (near the Hall of Sacred Mirrors on the map), we turned right/south and continued down the Midnight Climax borehole. This passage hits a climb up into a small crawl at MU12, which was the last station of the survey in Midnight Climax. It was marked on the map as a lead with caption “? Goes 100+ ft Stoops & Crawls”.

Garrett thought this passage would end soon, so we fueled up, drank some water, and ditched our packs2 at the start of the lead. We started the MV survey here at 4:02pm as we surveyed down a breakdown-filled stoopwalking and crawling passage, just as it was marked on the map. At MV4 we encountered our first lead, a low crawl off to our right that looked like one of those leads you almost wished you didn’t see, because now you’re gonna have to do it.

After a dozen or so stations of crawling and stooping in standard mud passage, the passage changed character and became covered in gypsum crust and other gypsum formations. I had the photography/inventory role on the survey team (i.e., the useless role) so I enjoyed my time looking at the pretties and trying to photograph them. They weren’t quite spectacular enough to make for good photos, but they did look cool in person.

This passage kept going further, but unfortunately I’m not yet at liberty to discuss what we found here. What lay beyond in the MV survey has to remain under wraps, for now. Check back later and maybe it’ll be public…

We went back to the low crawling lead at MV4. This quickly turned into a too-tight bedding plane crawl, although Adam thought he saw space ahead, and it was clearly diggable. High off excitement from our previous find, I launched myself at the bedding plane and began scooping dirt and rocks out of the floor, lowering the floor just enough for me to move further with enough space for my arms to dig in front of me. The dig would clearly go, as all I had to do was trench out the dirt-and-rocks floor, although it would take some work, as the rocks were tightly packed in the dirt which made lowering the floor more effortful than just scooping chunks of dirt out.

After half an hour or so, Garrett asked how it was going, and I shouted “six feet!” in an attempt to quell the impatience I could sense. There was definitely a void space beyond, and I was making progress towards it.

“Four feet!”

“Two feet!”

“One foot!”

After an hour or so of lowering the floor, I squeezed my shoulders underneath the sharp corner that was the ceiling and popped up into a low stoopway. No booming borehole, but it was certainly going passage. The rest of the team surveyed in and I let them go ahead, as photography and inventory generally goes last.

It almost seemed like we would have to dig again, as the stoopway lowered to a crawlway and eventually a tight squeeze. But at the squeeze, we popped upwards into a huge space! I squeezed through last, eager to see the borehole everyone was shouting about.

It was a nice walking passage, 10+ ft tall, 20+ ft wide, that went left and right from that obscure hole in the floor. We went right/straight first, and it tragically only lasted at those dimensions for 2 stations before turning back into a stoopway then a crawl. We followed this for 100-200 ft before it became a very low belly crawl. When it got to the point where some people on the team wouldn’t fit down the belly crawl without some major try-hard, we turned around and returned to the branch of the borehole going left.

After two stations in the borehole going left, we noticed a prominent hole in the wall. Air was pouring out of this hole, blowing in our faces, so we abandoned the borehole and did an awkward belly flop up into the hole. The air tragically disappeared after that hole, as it popped us up into a complex of tall fissures and breakdown-filled domes that were all too big to feel any air in. We scrambled up into the fissures and domes and couldn’t find a way on. The big breakdown-filled dome at MV38 still goes, as I didn’t get all the way to the top of it, although it was sketchy wide stemming in between huge breakdown boulders that seemed magically suspended in mid-air. It was not super safe to be questing up into that from below. Hopefully we can find our way into that dome from above, where we could presumably better assess stability as we go down.

After failing to follow the air past the hole at MV33, we returned to the borehole. We set a few more stations down the borehole, which unsurprisingly turned into a crawl. We turned around at a good going crawl. The crawl lead was pretty good, but we had been away from our packs for many hours now, and were all getting hungry and thirsty and full of pee.

We backtracked through the obscure hole in the wall/floor of the borehole, through the pancake dirt crawl I trenched out, and back to the start of our survey at MU12. It was 10pm then, and we had been away from our packs and all our supplies for six hours. We chugged water and scarfed down some food, and readied ourselves to begin traveling back to camp, hitting some mop-up leads on the way back.

The first mop-up lead at MU10 went for only three stations. The next one, at MV7, went for a bit longer, although it ended in breakdown without any reasonable prospects for digging. When we plotted it out later on the surface, we found out that this breakdown plug is directly beneath the going crawl the continues from the branch of the borehole that went left. Interesting.

We did one more mop-up lead at MU6. This one was nice and decorated, and a little mazey, but we killed all branches off that one. At 12:38am we were done with surveying for the day, and began our journey back to camp. It was an uneventful journey, with one exception. At MKA10, we took a right turn instead of a left, and realized that we were accidentally scooping virgin passage. This was a lead! And a very good one at that. It was a walking/stemming canyon, decorated, with a stream in the floor, 4w15h. We realized after 100-200 ft that we were in virgin passage and turned around, excited to continue exploration in that lead next year. I think that passage will connect to the This is Ridiculous passage, which I continued exploration in on my previous Fort Stanton trip back in June. The lead looks similar to This is Ridiculous: a tall, wide-but-not-too-wide-to-stem canyon, with big blobby formations lining the walls and a faint stream in the floor. The lead goes downstream towards This is Ridiculous (NE) at 4w15. This is Ridiculous ended as a 5w30h stemming canyon going upstream. They’re separated by 1800 ft, which means there’s probably 2000-3000 ft of survey in there, never mind if there are side leads. Something to do next year!

We made it back to camp at 2:23am. At not quite 15 hours, it was a fairly long day out from camp, especially considering that we took over 15 hours to get to camp the previous day. After waiting another eternity for the alcohol stove to boil water for my dinner mix, I was in bed at 3:45am.

Saturday August 16th

Garrett woke us up at 12:20pm. Waking up after noon—that means we’re on a real caving trip now! We made good time packing up camp, and we were out and moving at 1:42pm. From the four-way intersection right before camp, we went right. This would have been a left on our way in; it was the passage that Garrett labeled “To Lower Borderlands” on our way in, because he thought that passage would eventually connect into the wet stream passage below Borderlands we found on the previous trip with James Hunter. Today our task was to survey leads out there and see if that was the case. We traveled down this mostly nice walking passage, fully-loaded with camp packs as we thought we would connect back into Borderlands this way. At 2:30pm we arrived at our first lead, at MK120.

We started the MKB survey there at MK120. This was a pleasant walking passage heading down and right, while the surveyed way on stayed up and left. We mixed up the roles from last time: I sketched profile and took frontsights, while Rene did photography and inventory. Garrett stuck with plan sketching and Adam stuck with point/backsight. I was excited to be sketching some nice virgin walking passage in Fort Stanton!

After four stations, we tied in to MK114. Garrett was right that we would connect to other Borderlands area passages! Although this wasn’t a very big loop. Thankfully MK114 was really a pair of leads; one going south that we just came from, and one going north that was our way on. We continued surveying northwards, hoping to make more connections closer to Midnight Junction.

The passage became more decorated and gypsumy as we went on. The passage lowered to a crawl at one point before we popped past a big breakdown rock back into big borehole. Then it really hit me, I’m surveying virgin borehole in Fort Stanton! It was a pretty incredible time, and I was again shocked that Garrett and the others considered these leads suitable for a mop-up trip.

Around MKB12, the passage changed character and started to look quite familiar. It was a rectangular walking canyon, with a flat bedrock-ish floor, and with walls lined with white popcorn and gypsum. It looked quite familiar to the dry walking passage we turned around in during Jame’s trip last June. I began to sense that a connection was near.

We rounded a few more corners before Adam shouted that we had made the connection. After a few minutes for the sketch to catch up, I saw the blue ML40 flagging tape station, just as we had left it a few months ago. Garrett was right! This would be a much easier route back to Midnight Junction and back to the entrance that the way we came through Borderlands and through MK Ultra. It was 4pm when the made the connection; the night was young and we had plenty of time to continue surveying, especially since we now had an easier route back to Midnight Junction, where we planned on sleeping that night.

This spot where we turned around in June had three leads, and we had just killed one of them. Now it was time to do the other two! We took a moment to eat, drink, and ditch our heavy camp packs as we got into day-pack surveying mode. The first lead we did was the other dry upper-level passage, up a short flowstone climb from ML38.

This went as a nice walking tube for a few stations up a beautiful flowstone slope, although thankfully it was muddy enough that we didn’t have to use shoe covers. In fact there were beautiful mud cracks in the flowstone. Mud cracks on top of flowstone is not a combo I see very often.

As the passage turned around from going south to going north, we encountered another flowstone climb. This one Garrett did, and none of the others wanted to follow him, as the climb was tricky and on not-very-solid flowstone. I started up the climb, but was stumped after trying it a few different ways, so I let Garrett do his thing on his own. He surveyed just a few stations past the tricky climb and reported that is still goes in two directions as a decent crawl. Worth going back to at some point.

While Garrett was up past the tricky flowstone climb, we looked at a drain hole in the flowstone at the bottom of the room. It belled out below a small, barely-human-sized hole. Below that we could see a floor of running water. We figured we would be going there next, via the third and final lead from the ML40 area that was a wet walking stream passage.

After Garrett came down from the flowstone climb, we went to the wet lead at ML39. This was awesome walking passage, of a similar size to the rectangular-shaped walking canyon we had surveyed in from, although the floor was mud and deep water. Here we continued the ML survey past 40. We carefully stemmed and tiptoed across the water for a few stations before realizing that resistance was futile, and we were just gonna have to get our feet wet.

Once we all stepped into the water, it felt like we were doing Kentucky Caving. Walking through a muddy stream passage, surveying virgin walking passage very far into a horizontal cave. This passage went on for quite a while like this. There was not much floor detail to draw, so the survey went very quickly. In fact I struggled to keep up sketching profile. I had to remind myself that no one will ever make a profile map of this horizontal maze cave, so the profile sketcher role is kind of fake, so I shouldn’t sweat the detail too much and I should just sketch whatever I can while making sure to not slow the team down.

All good things must come to an end, and this passage was no exception. Eventually the ceiling started to lower on us, and we were crawling in the mud. At ML53, Adam encountered a sump. We were all disappointed at that, as the low water crawling hadn’t gone on nearly enough for us to want the passage to end. However, he found a way around, although it did involve a superman crawl over a section of stream that made keeping the survey book and Distos dry a bit of an athletic feat. We all got it done and were back into walking passage, for a moment at least.

We continued in more excellent straightforward walking passage until a pair of leads at ML68 and and ML70. We charged past the one at ML68, but the lead at ML70 was enticing, so we hopped up onto a ledge on our left above the water to check it out. This was a beautiful chamber full of shelfstone, pool spar, and pool fingers. There was a high lead coming out of the wall where it met the ceiling, and a low lead going into a very decorated wet shelfstone crawl. Both were maybe worth doing another day, not today. We set some good permanent stations in this room and photographed the pretties before continuing on in out wet walking passage.

Further on in the ML70s, our walking streamway once again lowered to a crawl, although this time the crawl was here to stay, rather than a brief duck before popping back up into walking passage. We surveyed many 70+ ft straight shots through the wet, muddy pancake, struggling to keep our survey supplies and hands out of the water. I did some advanced crawling on my elbows to keep my paper notes dry.

At ML81, we got to another intersection/set of two leads. On our left, a dry crawlway went up into a ledge above the water. The wet muddy pancake crawl continued to our right Air poured out of the dry crawlway, so naturally we went there first. Air was ripping through the low wet crawl right before this intersection, and we were eager to follow it somewhere dry.

Following the air into this dry crawl didn’t exactly warm us up, as we had hoped. Instead we got breaded by the dry dirt and dust, which stuck to us after we had rolled around in the muddy water and gotten all sticky. We were a bit cold from surveying in a muddy wet crawl without much for warm clothes, and getting breaded didn’t exactly help that. We turned around at a 2w5h lead with some air after a few stations. It’s a decent lead, although it’s somewhat trying to get out there. We’ll be back on another trip.

Then we went right up the continuation of the stream passage. We put a few more stations of belly crawling before calling it and turning around at ML86. This lead is 1 ft tall, many ft wide, with lots of water and mud in the floor to sap your body heat, but also with some air to entice you (and sap your body heat). We’ll be back at some point!

We traveled back out via the ML survey, stopping only to do the lead we left at ML68. This went for a few stations of boot-sucking mud before lowering to a grim crawl—not a lead for the next trip, but the next generation of cavers maybe. At 10:35pm, we were back at our packs, done with the new wet passage we had surveyed that day. It was time to head back to Midnight Junction, out the route we had surveyed from La Culebra back in June, the back down La Culebra to Midnight Junction.

I took the others down that route, as I was the only one that had done ML1-38 on that team. There was a lot more crawling that I remembered in that passage. Classic; you always remember the nice parts of any trip long after the suffering has faded from memory. Not that the crawling was long/intense enough to call suffering, but you know what I mean.

It took us under an hour to get from our packs at ML38 back to Harmony Hall, the water spot two minutes from Midnight Junction. After topping off our water, we were at Midnight Junction at 11:59pm, very fitting. As we ate our dinner, Garrett added up the survey footage from the trip. Just over 7000 ft, not bad for a mop-up trip! The nature of our trip, with many small sections of new passage instead of a few leads that went for a long ways, made us think we totaled much less footage; Adam was particularly worried that we might not have hit a mile. But it was no matter, as we ended up with a respectable one and a third mile of cave surveyed.

We took a little while to unpack our camp packs and start cleaning and packing a little bit for the next day’s journey out of the cave, although we saved most of that for the next day. We were in bed at 1:26am.

Sunday August 17th

The trip out of the cave was uneventful. The most eventful thing that happened that day was that everyone had tons of extra food, so they left a bunch at Midnight Junction for me to eat on my next trip in the cave starting Thursday. There’s a real pro to doing two trips in a row like that: you can leave things at camp because you’ll be back there a few days later!

There was so much food the others left at Midnight Junction that I made sure to note that I didn’t need to pack any day food for the next trip. Just breakfasts and dinners, and food for traveling in the cave to camp, and some food to stash for the way out. But all my day food for two days of exploration out of Midnight Junction would be there waiting for me in a few days.

After our leisurely 9:30am wake up time, we were out of Midnight Junction at 11:45am, and out of the cave just over eight hours later at 7:59pm. Garrett was so kind as to clean and put away my gear for me the next morning, so I could do the three hour drive home that night and go to work the next morning. Thanks Garrett!

In addition to Garrett for leading this trip, thanks to Adam and Rene for coming and making it happen, and thanks to the Bureau of Land Management and the Fort Stanton Cave Study Project for managing the cave and its exploration so trips like this can happen. Fort Stanton is a very special place, and I’m honored to take part in far out exploration trips like these!

Subscribe now

I had been to Lechuguilla Cave once before, last Memorial Day weekend. That was a three day camp trip at the Deep Seas Camp where we hauled in a bunch of rope to replace the ropes on the deepest pit series in the cave. The trip this past weekend was just a day trip, and our objectives were similar albeit smaller: replace another portion of fixed rope that had been left rigged in the cave for an unknown number of decades, and flag a trail through one of the more delicate, unique areas of the cave, where some trips are upcoming and the park wants impact minimized.

So this trip wasn’t an exploration trip per se, but we did see some of the most spectacular areas of one of the most visually stunning caves anywhere in the world, and we got some great pictures, so I wanted to share. Enjoy the pictures. These kinds of caves certainly make me think “this is fake, this doesn’t happen naturally on planet Earth; aliens and/or CGI must be involved”. Maybe the pictures will make you think the same! Feel free to scroll down to the pretty pictures and skip my extended monologue if you want.

Our day started out well, as we did the hike to the cave under clouds and with cool temps. Usually early September still feels like summer in Carlsbad, but the clouds and breeze made the hike to the cave pleasant. We even got lightly rained on during the approach. The cave is extremely hot at 68 degrees Fahrenheit and 100% humidity, so we welcomed the reprieve from the New Mexico summer heat while we awaited the cave heat.

We entered the cave at 10:45am. Our team was Hunter Klein, Georgia Schneider, and Clancy Montemayor. Hunter is the Park Service employee who ran the trip and is organizing the effort to replace all the aging ropes rigged in the cave. On the way in, I dropped off a rope I was carrying at The Rift, a large fissure complex about 45 minutes into the cave. The main travel route continues right/south, but we were to use that rope to replace an old one going left/north, off the main travel route into the rest of the cave.

It took us just over an hour to make it to EF Junction, the first major junction in the cave where multiple travel routes to important parts of the cave branch off in different directions. That was a fast time, compared to our last trip at least. Despite that, I was still plenty cool and wasn’t burning through my water.

From EF Junction, we took the right branch, instead of the left branch we took last time that leads to the Great White Way and to the Deep Seas Camp. From this point on every part of the cave I experienced was totally new to me. The passage going right from EF Junction was pleasant walking passage coated in sparkly gypsum. It continued as easy walking for about 10 minutes until it ended at the first rope obstacle: the Little White Bastard. This small, low-angle, tubular rope drop is a lesser version of the Great White Way, which is the first rope you encounter if you go left at EF Junction instead of right.

The Little White Bastard led to some more walking passage at the bottom, then some more rope drops and traverses. The shoestring traverse, the last rope traverse before Lake Lebarge, was particularly fun and sporting. After the shoestring traverse, two more down ropes led to the spot where we could ditch our vertical gear for the rest of the trip in. It was also the first spot this trip where I was truly blown away by the cave’s grandeur: Lake Lebarge.

Alright, now for the good stuff

You made it this far, so now you get to see some pretty pictures.

Lake Lebarge is really a complex of multiple lakes that feed into each other. The first major lake presents a unique obstacle: it stretches from wall to wall, and you have to carefully step along shelfstone ledges at the edge of the lake to avoid the water, lest you get your feet wet or contaminate the pristine pool that lies undisturbed hundreds of feet below the Earth’s surface. We all carefully sidestepped along the edge of the pool, holding on to the grippy popcorn-covered formations on the wall to avoid falling in the water.

When our heels did briefly touch the water and disturb it, the ripples that propagated along the pool reflected a beautiful wave pattern on the ceiling above the water. We got our own little laser light show for entertainment while waiting for others to cross the pool.

When this first massive lake ends, the path takes you into a garden of boulders covered in pearly white aragonite bushes. Here you have to be very careful not to touch anything as you weave in between the boulders, lest you break one of these delicate formations.

Just past this first major lake is the lake we collect water from. The path doesn’t go right next to this lake, and there’s no accidentally stepping in it and contaminating it, so we drink this water untreated. I still hadn’t drank any of the three liters of water I had hauled in, at two and a half hours into the cave, so I was a little disappointed that I didn’t have the chance to fill up a bottle full of pristine Lake Lebarge water. I at least had a sip from someone else’s bottle, to sample the goods.

There’s one last major lake in the Lake Lebarge complex. At the edge of this lake, the flowstone slope that the water follows begins to descend steeply into the passage, and we get off the flowstone slope and onto a higher ledge of what becomes a massive round borehole that extends straight into blackness as far as you can see. This aesthetic tubular passage is the Lebarge Borehole.

We lost the water and its flowstone slope as we continued into the Lebarge Borehole. We walked along a flagged trail through the breakdown in the floor of the borehole, always wondering what lay beyond the edge of what we could see as it faded into blackness. While looking at the borehole might make you think it would continue forever and ever, it did eventually transition into small tubular passage and begin branching off into small crawly passages going all sorts of directions. This mazey complex of small passages is the Tinsel Town Maze.

The Tinsel Town maze would be confusing and even downright scary to navigate if there were not one clear flagged path through the whole maze. We followed the path of orange flagging, and eventually the maze turned into large passage again. Where the maze of small crawlways turns back into large passage was an even more wondrous, outrageous, out-of-this-world room in the cave called the Chandelier Ballroom.

The Chandelier Ballroom is named for massive complexes of gypsum crystals growing out of the ceiling, called gypsum chandeliers. The Chandelier Ballroom has the largest gypsum chandelier formations in the world, and they are spectacular not only for their size, but also for their dramatic, jagged structure. They look like growing, branching lightning bolts that froze in place then widened to several feet thick. Some are frosty, like the thick clumps of ice that condense out of the air and freeze onto plants and other structures on a cold winter day. Others are translucent and crystalline. Words do not approach doing justice to these formations, and the pictures only get a little bit closer.

Hunter followed a tradition that someone did the first time he went to the Ballroom. When we all exited the small crawlways of the Tinsel Town Maze and popped up into big passage, he had us turn our lights off and instructed us to only look at the floor. He guided us with an extremely low light so we couldn’t see anything besides what was immediately in front of our feet. He walked us over to the most spectacular part of the room, instructed us to close our eyes, and turned his light on high. We all opened our eyes on the count of three, and feasted our eyes upon one of the most glorious sights to be seen on our beloved planet Earth. Scroll down to get the same effect we did…

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

Hunter flagged a trail that weaved through the formations for some upcoming photogrammetry trips. Once we were done with that task, it was time to enjoy the show and take some pictures. Hunter got out his real camera and flashes and got some quality pictures of all of us, seen above.

Besides the chandeliers, there were other incredible gypsum formations in this room. We spotted huge gypsum needles, clusters of gypsum hair, and gypsum flowers. What a great mineral gypsum is!

After we got our fix of photos and admiration time in the Chandelier Ballroom, it was time to head out of the cave. We backtracked through the Tinsel Town Maze, through the Lebarge Borehole, and back across Lake Lebarge. We donned our vertical gear and continued up the ropes above Lake Lebarge. We stopped briefly at the top of the Little White Bastard, then continued on and took a more substantial snack break at EF Junction. I volunteered to do the rigging task of replacing that rope in the north rift, the one where I dropped off the replacement rope early on in the trip. I sped through the rift while the others continued their snack break so I could get a head start on the rigging task.

I was to replace the first rope in the North Rift, a traverse line. This rope was tied to a rock projection in horizontal passage right before the rift, and the rope snaked through a crawl before popping out of a hole into the gaping floorless chasm that is the Rift. The situation was slightly complicated by the fact that Hunter tried to replace this rope previously, but had too short a rope, and left the too short rope rigged. So there were three ropes I would have to deal with: the old rope of unknown age and origin that maybe shouldn’t be trusted too much, the new rope which was only attached to anything on one side of the traverse and therefore definitely not to be trusted at all, and the rope I was rigging. After dealing with the awkward crawl that pops you out into a bottomless chasm, I quickly stemmed my way across to the other end of the traverse. There I found a massive boulder the old rope was tied to. I anchored to the same boulder, this time slightly better as I got the rope around the whole boulder with constrictions blocking the rope from moving on either side, instead of just wrapping the rope around the tip of the boulder a few times.

From there, another old rope of unknown age and origin led straight down into the Rift. I checked to see whether the old new rope (the one that wasn’t long enough for the traverse) was long enough to replace this down rope by rigging it to the same boulder and rappelling down the Rift. Even while using some of the leftover rope I had from the traverse rope, I still didn’t reach the bottom of this drop. I hit the backup knot about 30 ft off the ground (“ground”, as the Rift is full of false floors), and looked down to see the old rope continuing downwards to a ledge with a large boulder, where a spider web of three other ropes branched off in different directions. Hunter was mildly stressed to hear about this spider web of other ropes below that rope, as none of these ropes in the North Rift were documented in the park’s rope log the way permanently rigged ropes are supposed to be.

That scene of a spider web of ropes continuing off in unknown directions in a vertical maze was intriguing to me. The North Rift is a very interesting area, and I look forward to doing more rope hauling and replacement trips to learn the area better. Hunter says there are leads in this neglected area. I can’t wait to return and coax more secrets out of this cave!

Although I took a while rigging and derigging the old ropes, I still caught up with the others at the bottom of Boulder Falls, the 150 ft pit that is often a bottleneck for parties leaving the cave. So I’m glad I ditched the others at EF Junction to get a small head start on that task. We all climbed Boulder Falls without incident and made it out of the cave just before 11pm, which made a 12 hour trip.

The skies had cleared when we were in the cave, and we were greeted by a brilliant full moon when we got out of the cave. It was the perfect temperature to hike back to the cars.

The next day, Sunday, I got up to another, different kind of cave adventure in the Carlsbad area. Less visually spectacular than this cave, but more off the beaten track and wild. Subscribe to hear about it:

Subscribe now

Special thanks to Hunter for organizing this trip, and to Georgia and Clancy for coming and making it tons of fun.

The current known end of Snowy River is one of the most remote points you can get to in any cave in the world, in terms of travel distance from the nearest entrance. 13 miles of underground travel from daylight is a long way to go underground. And it’s still going! Not only the Snowy River passage, but many equally outstanding side passages branch off Snowy near the far end of the cave and go off in all sorts of directions underneath the Sierra Blanca mountains. Leads (unexplored passages where no human has ever set foot or has any idea what is there) abound in the far reaches of the cave. Exploration trips out to the remote parts of Snowy River and beyond are fantastic adventures, but they are also very serious. There’s no communication with the outside world when you’re underground, as cell service and all other feasible forms of wireless communication can’t penetrate the bedrock. When you’re that far from any vestiges of human civilization, you must take your actions seriously and be extremely deliberate and cautious with your every move. If anything goes wrong while you’re in the more remote areas of the cave, you’ll have to send part of your team out 10+ miles and 12+ hours of travel to the surface, where they will have to recruit more people to repeat the same journey to help you. Expect to be on your own for at least a full day, if not more, before any help can arrive.

The sense of extreme isolation and remoteness you get in the outermost limits of a cave like Fort Stanton can feel crushing and intense, yet also invigorating; it certainly draws me to caving. Fort Stanton trips are also special because the cave has excellent leads, due to both the remote nature of the cave and difficulty of getting out there, and the tightly controlled access to the cave from the Bureau of Land Management. I can think of few caves with so many walking-sized leads going off the map into the unknown.

This past June, James Hunter invited me on an exploration trip he was leading to the Midnight Junction area. At Midnight Junction there’s an established camp, a little over 10 miles of travel from the entrance. It was named because it was discovered at midnight, on a long day trip, before anyone was camping in the cave. After finding that intersection at midnight after a long day, they still had 10+ miles to travel to get back out of the cave! That was the last day trip anyone did that far in the cave; now anyone exploring the far reaches of the cave camps at Midnight Junction or bivvies beyond there, closer to where they’re exploring.

Midnight Junction is an ideal location for a camp, because that’s where the cave changes character from pretty much just one passage going forever in one direction (Snowy River) to a maze of passages branching out in all different directions. At Midnight Junction, one passage branches left off Snowy River. A few minutes of travel later, you get to a T intersection, Harmony Hall, where passages go both left and right. Both of those branches have many different passages branching off them, including many unexplored leads.

I had been on one exploration camp trip to the Midnight Junction area before. It was one of the most incredible caving trips I had ever done, and I was eagerly awaiting returning to this area. There is plenty more exploration to be done in the Midnight Junction area, as the maze is nowhere near having given up all its secrets.

This trip would be a four day camp trip: one day to travel to the Midnight Junction camp, two days to survey new passages, then one day to travel out to the surface. We would enter the cave Thursday June 26, and exit Sunday June 29. The four of us on the trip, James Hunter, Beth Cortright, Ben Tobin, and myself, all met at the Fort Stanton bunkhouse on Wednesday evening. There we packed our bags, discussed the plan and timeline for the trip one last time (after many such discussions in the days and weeks prior), and went to bed early in preparation for our long day the next day.

Thursday, June 26

We got up bright and early at 6am, early enough to leave the bunkhouse at 7:30am and get in the cave at 8am. The first hour of caving brings you through the historic section of the cave, which is mostly huge mud-filled borehole. There is a short crawl into Don Sawyer Memorial Hall, where a dig shaft supported with stainless steel ladders and supports descends straight down into the bedrock. That’s the site of the second dig shaft that cavers used to break into Snowy River. They first accessed Snowy via a breakdown dig called Priority Seven. They explored Snowy for many years via Priority Seven, but the dangerous, unstable breakdown in that dig eventually led them to find a new way in, by digging in Don Sawyer Memorial Hall.

At the bottom of the dig shaft is the Mud Turtle Passage, a 15 minute hands-and-knees crawl. The crawl ends at Turtle Junction, where a plastic tarp lets you change into your clean Snowy River clothes before stepping onto the formation. The Snowy River formation is a clean, pearly white standout in the middle of a cave full of mud, and we go to great lengths not to get Snowy dirty to maintain its pristine appearance. Changing clothes when switching between dirty mode caving and caving on Snowy is one way we keep it clean.

After a few minutes of walking down gorgeous Snowy passage, we arrive at another course of action required to keep Snowy clean: a shoe cover crossing. At Independence Hall, Snowy River ducks under a low ceiling, and we step off the formation onto the dirt banks on the side, to walk where the passage is large instead of crawling on Snowy underneath a low ceiling. To keep our shoes clean, we put on shoe covers for this short dirty section. Putting on and taking off shoe covers at spots like this is a somewhat delicate maneuver. When putting shoe covers on, you have to stand on one leg, put a shoe cover on the other foot, while sticking your leg out over the dirt section so as not to spill dirt from your shoe covers on Snowy, step down onto the dirty section, and do the same for the other foot. All while standing on a potentially uneven surface, and while carrying a 40+ pound pack. Very balancey.

In some spots where you have to step off Snowy, there are plastic tarps we call “magic carpets” that let you walk off snowy while keeping your shoes clean. There are several hundred of these on the way out to Midnight Junction, and they make the trip out much easier, as it takes much less time and effort to use them than it does to do the shoe cover dance.

It’s not just your feet that need to avoid touching dirt. The walls and ceiling of the Snowy River Passage are usually covered in dirt, and the slightest nudge on them will knock off dark brown debris onto the formation. This is not a problem when Snowy is in the center of a large walking borehole, as in the first picture, but it presents a challenge when Snowy weaves its way to the edges of the passage and passes under roofs and other obstacles. There you play a strange game of “The walls, ceiling, and some of the floor are lava”, like the childhood game “The floor is lava” but much stranger and more interesting.

The trip out to Midnight Junction is a little over 10 miles of travel underground. Much of that distance is easy, glorious walking on Snowy. However, there is a lot of less glamorous travel ranging anywhere from slightly hunched over stoopwalking, to low belly crawling, to perilous weaving around dirty obstacles while trying not to touch anything and get Snowy dirty. And that’s just on Snowy. In several sections you get off Snowy, either because it’s covered in breakdown (talus rocks) you have to scramble over, or the route on Snowy is so small and difficult that the standard route is via side passages. When you’re off Snowy, you’re often doing exposed breakdown scrambling above cliffs, or maneuvering around obstacles in small passage that feels more like normal caving. Depending on the length and character of the parts of the route that go off Snowy, you either change into full dirty mode (at the Black Rock Bypass, an hour long route through small, moderately strenuous passage that bypasses the worst crawls on Snowy), or just change your shoes and kneepads (at Rough Country, a long section of breakdown borehole that is huge but exposed and requires some scrambling), or just put on shoe covers (in many places, but in particular at a long section between Rough Country and the Midnight Junction camp). Doing a less-than-full outfit change saves time and inconvenience, but it makes the travel in that section more involved, as you have to ensure that your clean clothes and pack don’t get dirty. Breakdown scrambling in shoe covers can feel particularly high-stakes, as the shoe covers are slippery and your feet love to slide around in them.

We did have one task assigned to us for our trip out to Midnight Junction. There were two large first aid kits stashed at Higher Hopes (a little over six hours in, just past Mount Airy and the Eggshell Trail). We were to pick up both of them and transport one to Finger Lakes and one to Midnight Junction. I was able to fit a whole kit in my pack after giving someone else an empty water bottle of mine to make space. We didn’t have to unpack them and stuff the components in various nooks and crannies, as we thought we would.

I won’t write too much more about the trip to Midnight Junction, as I summarized that leg of the trek in my post about my last Fort Stanton trip. The trip in felt similar to the previous time, but just a bit easier, as it often does the second time around. I definitely did better with my water consumption. When I got to Finger Lakes, the first water source on the trip, eight miles and about eight hours into the cave, I had only drunk 1.2 L of water. Reducing how much water you drink is very useful, as you have to carry all your pee out of the cave. Four days of piss is a lot to carry! I made sure to only drink water after I had cooled down a few minutes after stopping. I find that this tactic greatly reduces my water consumption; I used it to great effect in Lechuguilla a few weeks ago.

We got to Midnight Junction at 7:40pm, in a little under 12 hours. That was slightly faster than the time from my previous trip, 13 hours. I felt great on the way in. I changed up my clothing and shoes a little bit from last time and was pretty happy with those choices. I wore lightweight, thin-soled, zero drop barefoot-style shoes for both my clean mode and dirty mode shoes, instead of the Walmart slip-ons that are a popular choice for Snowy shoes and that I wore last time. The barefoot-style shoes aren’t much lighter, but they do save a lot of pack space, as the Walmart shoes have thick foam soles. I even took the insoles out of my clean mode shoes to save a little pack space. That was honestly an excellent idea if only because it let me feel the fine-grained texture of Snowy through the thin, flexible soles of my shoes as I walked on it. It was wonderful to be reminded with every step that I was walking on an incredible cave formation; walking on a cloud. Snowy has just enough features to make walking on it with little sole perceptually interesting, but not so many features that it hurts your feet. I didn’t get sick of it even after 15+ miles of walking on Snowy over the course of the trip.

I also wore shorts in clean mode. That helped with both pack space and heat management. I definitely got less hot, and thus drank less water, and thus peed less, while wearing shorts. The cave is 55 degrees, and not incredibly humid like most caves are, but I still get hot when moving fast there while wearing long pants. I also wore a short-sleeve shirt in dirty mode, with elbow pads. That probably resulted in net more pack volume (elbow pads take up lots of space), but it helped with heat management, which is much more critical in dirty mode, so it was worth it. With elbow pads, I didn’t have any trouble keeping my arms sufficiently clean for travel on Snowy. My clean and dirty mode shirts were both Hawaiian shirts, and my clean mode shorts were Hawaiian patterned, fitting with a style I’ve been into lately. And fitting with a mood I think is appropriate for any Fort Stanton trip: party time! We get to go in one of the most incredible caves in the world, better dress for the special occasion!

The Midnight Junction camp is, in contrast to the rest of the cave, just OK. Not the best cave camp I’ve stayed in, but not the worst. Maybe second worst? Stay tuned for the upcoming Tears of the Turtle trip report part two to hear about that…

It’s a series of four tarps laid down in a line on a dirt bank on the side of Snowy River. Each tarp is your living space for the next three nights. They’re fairly small, about 5 ft by 12 ft. You want to keep most or all of your stuff on the tarp, as it’s the only clean surface in the sea of dirt, and you’ll need to keep the inside of your camp pack clean for the trip back down Snowy. So anything that isn’t exposed to the cave for dirty-mode caving stays there, leaving you with just enough room to sleep. When you’re on the tarp, you have to be in clean mode, to keep the tarp and all your stuff on it clean. So you can’t go wandering aimlessly around the camp area, sitting down on things or hanging out next to your friends while also conveniently accessing your stuff. You can get off your tarp of course, but you have to put on shoe covers to do so, and if you want to sit down on anything you need to be in dirty mode, which means you can’t get on your tarp. I call this configuration “tarp jail”. Again, it’s not the worst cave camp in the world, because it’s warm, dry, has pre-stashed sleeping bags and pads and stoves and such, has enough space to stand up fully, has a good water source, and has enough space to be able to go take a dump with privacy, without all your friends watching you. All luxuries that are not available in some of the cave camps I’ve camped at!

We collected water from just upstream of Harmony Hall, a two minute walk from camp. Normally there’s a flowing water at Harmony Hall, the intersection just beyond Midnight Junction, but it was dry now. John Dunham noted the low water on his Midnight Junction trip, the most recent trip there, last month. He was worried that the water might totally dry up over the next month before we got there, so he made sure to leave some water stashed at camp so we wouldn’t be totally screwed if we got to camp the water source had dried up. But it was fine when we got there; the water level seemed unchanged from what I saw in the photos from John’s last trip.

Although we had plenty of water, it did taste funny. It had a minerally taste, like tap water, but much stronger and kind of bad. The pool being stagnant instead of flowing probably contributed to the off taste. The others opted to treat their water, which people rarely do in Fort Stanton. I figured maybe I should as well, but I never got around to treating any of my water. Oh well. I’m writing this trip report two weeks later and I don’t seem to be dying of dysentery, so probably it’s fine?

Here’s a table with times and quantity of water consumed at various landmarks on the way to Midnight Junction, for my own personal reference and/or for the reference of anyone else heading out to Midnight Junction. There’s water at Finger Lakes, so I only record water consumption up to Finger Lakes.

Friday, June 27

We got up at 6:30 to James’ alarm. Ben and I, still sharing our cooking spot, immediately started making hot water for breakfast. I made a bowl of hot Dinner Mix, the same as last night and the same as every other breakfast and dinner I have eaten in the cave. I also made a small cup of coffee, about a third of my usual dose. That tiny dose was enough to make me poop that morning (a critical checklist item at cave camp!) without making me generate lots of pee that I would have to carry out.

As much as I complain about the Midnight Junction camp, I will admit that its suboptimal nature makes you not want to hang out at camp too much; it makes you want to go to sleep immediately at night, and get up and get out of camp and go caving immediately in the morning. Plush cave camps, especially in unpleasant caves (the opposite situation to Midnight Junction/Fort Stanton) can make it a little too easy to lounge around in camp, doing anything but what you came there for: to go caving and explore new passage. I was especially motivated to get out of camp and get to the leads that morning, and we were off by 8:28am. Two hours to get out of camp and start caving isn’t super fast, but it’s certainly quicker than many other times I’ve been caving.

To get to This is Ridiculous, you take the dirty passage from Midnight Junction to Harmony Hall, then turn left towards the pools where we got water. You follow that passage, Midnight Creek, for 27 stations that are a mix of crawling (including some precarious superman crawling above pools) and walking in huge passage. At MJ27, you ascend a dirt slope up and right into an upper level of the Midnight Creek passage. After a few stations this upper level branches off Midnight Creek and becomes its own separate canyon passage.

Here, where the canyon first branches off Midnight Creek, it is tight, grabby, and annoying; yet also densely decorated, beautiful, and delicate. The need to not sully the clean formations and not break the delicate stalactites and helictites makes travel through the tight snaggy canyon even more tricky. Here I understood why they named the passage This is Ridiculous: it’s quite the stark contrast to be thrashing and struggling through a narrow canyon that grabs you and all of your stuff, only to be confronted with clean velvet flowstone and other pretties that you must delicately maneuver past. As I was exerting myself in the canyon, I was continually entertained by the colorful displays of brilliant striped flowstone and dense, featured helictite forests. I didn’t take many pictures as I was focused on the task at hand, getting through this passage. I did tear my vibrant red Hawaiian shirt in this passage as it snagged on some popcorn, which was unfortunate. Oh well, I guess it’s a cave shirt now.

I did stop to take pictures at one spot of something particularly notable. I saw a fully articulated skeleton of some sort of mammal sitting on a ledge in This is Ridiculous, at MJA18. The skull was missing, but the rest of the skeleton was remarkably articulated. Gary Morgan from the New Mexico Museum of Natural History & Science identified it as a likely ringtail, although with the skull missing it’s impossible to say for sure.

It was sitting on a ledge high above the floor, 10 ft perhaps? If it was washed in via a flood, it would be one hell of a flood to reach a ledge that high. Still, there is reason to believe it may have been a flood that brought this animal here. It was very far into the cave, over 10 miles via human-sized passage. While there may be other entrances accessible to small animals, there are several hundred feet of overburden at this spot, so any animal would still have a long travel time to get here even if there were a direct small-animal-sized conduit from the surface above. There were some small bones right at the (missing) skull. Perhaps part of this animal’s last meal?

Besides a quick stop to document that skeleton, we all powered through This is Ridiculous to the continuation of the passage that was our lead. The passage followed stations MJ27A-MJ27Z, then continued as MJA1-MJA47. After ~70 stations (not quite 74, as a few shots were splays) we arrived at MJA47 and our lead. Here the passage abruptly changed character from the tight canyon we traveled through to a comfortable keyhole, with a 10 ft diameter tube at the top and that tight canyon in the floor. Either the passage changed character, or this is the first time it made sense to get up into the top of the keyhole that may have been there the whole time. What a spot for the last team to turn around at!

We got to the lead at 9:51am. Under an hour and a half from camp. I stashed most of a liter of water at the lead, as it was clear I wouldn’t need most of the three liters I hauled in there. We got set up to survey; Ben would sketch plan and cross sections, I would sketch profile and do front sights, Beth would make stations and do back sights, and James would do inventory and take pictures (which is actually required at every station when surveying in Fort Stanton). I hadn’t sketched in Fort Stanton before. Although sketching profile doesn’t really count—it’s unlikely that anyone will ever make a profile map using the profile sketches of this horizontal cave, so the profile sketches will probably remain forever archived, never to be made into a map.

As we started surveying, we were able to continue along that comfortable 10 ft diameter tube at the top of the keyhole passage. Stemming over the gaping crack in the floor got a little exposed, but was never difficult. The dense forests of formations continued to line the walls and ceiling of the passage, providing obstacles to avoid, although it was much easier to avoid them in this large tube. Even when they got huge.

And boy, did they get huge. Our first shot brought us down that comfy 10 ft diameter keyhole tube where rich velvet flowstone poured down the floors of the tube into the canyon below, dense hanging gardens of soda straws dangled from the ceiling, and deformed, chaotic helictites grew out of the walls. After a few stations, we came across one of the more impressive clusters of gypsum needles I have ever seen, even more spectacular than those from the Porcupine Passage we found on my previous trip. This was classic Fort Stanton passage.

The tall (~20 ft) canyon in the floor added an interesting obstacle that made it feel like real caving. We had to mind the slick flowstone sloping down to the keyhole canyon while we traversed the small ledges in the tube atop the keyhole. Sometimes the canyon got quite wide and we had to do some creative stemming moves to cross from side to side. You must always be vigilant when doing exposed, high-consequence movement like that in a cave, but in virgin cave like this, where zero humans have ever come through and knocked any loose stuff off, you must be even more on guard than usual. Rock that looks solid may dislodge with the slightest touch, as nothing has been there to disturb it from its resting place for thousands of years until you come along. The delicate, pristine formations that we were trying to avoid in order to not damage them added an extra challenge to this fun, sporty canyon traverse.

I kept expecting for the keyhole to end and force us downwards, and for the passage to return to the character it started at: tight and grabby. But that never happened. The tube atop the keyhole never really ended, except for some short sections where we lost the tube but quickly regained it. In a few spots, the canyon at the bottom of the keyhole got wide enough where we could get down to the floor and walk in it, instead of stemming in the tube above it. That made this already nice passage feel extra luxurious, although we did lose the formation gardens when we dropped down to the floor.

When we dropped down to the floor, we could see that there’s a small, beige/white, Snowy-River-like formation in this floor. Not as thick nor impressive as Snowy proper, and without the distinct cauliflower texture, but it was notable nonetheless. We also eventually started seeing pools in the floor. Ben said he could see them faintly flowing, although I couldn’t tell. They just seemed like standing water to me.

Another interesting change in the character of the passage occurred at MJA79, where we climbed into an upper level that was the continuation of the keyhole tube, as the canyon branched off right and ditched the tube. In that upper level, we picked up another series of pools in the floor, although we lost the water again, as that minor canyon again branched off our upper-level tube at MJA84. Here, our tube turned dry, and we gained a pleasant soft sand floor. It was a nice change to walk on soft sand after stemming above a floorless canyon for most of the day.

At MJA87, our dry sand tube turned into a crawl. We thought that the good times might be ending, and that we would have to go back to either of the canyons that this tube lost as our leads. However, at MJA92 we regained a tall wet canyon with a Snowy-like formation in the floor, and we were back in our usual keyhole passage. The canyon we lost at MJA79, and the minor canyon we lost at MJA84, and the canyon we gained at MJA92, are all leads, although they’re not super promising, as it’s likely they’ll all connect through to MJA92. But maybe not—there’s always a chance one of them could intersect another passage or go off in another direction.

After regaining the lower canyon, the passage changed character yet again. Instead of the classic keyhole cross section with a large tube sitting atop a tall narrow canyon, we instead had a tall, wider canyon sitting atop a tall, narrower canyon. There was a clear ledge system that marked the boundary of the upper wide canyon and lower narrow canyon that we stemmed across the rest of the way. We started marking leads in the ceiling when we couldn’t get a view of the top of the upper canyon, although I don’t think any of them are good leads. They’re just little alcoves where the passage geometry doesn’t let you definitively see the whole ceiling.

The passage stayed intensely decorated after regaining the canyon in the floor. There were formations with wind-grown formations growing on them—meta-formations as I sometimes call them. This indicates that the passage had directional airflow for a long period of time in the past. That’s a good sign that this passage goes somewhere (“if it blows, it goes!”). Ben claimed he felt air, although I never felt any. The velvet flowstone also became more common, and it became impossible to avoid in spots as it plastered the whole walls where we had to stem. It was not really feasible to use shoe covers in involved stemming passage like this, so we just picked our spots to step on and concentrated our impact in those few spots. We climbed across some of the velvety areas with bare hands, to keep it clean. There I learned that velvet flowstone is full of tiny, sharp, grabby spines, like the glochids on a Prickly Pear fruit, or sharp velcro. It kind of hurt my hands, although I didn’t mind, as it was one of the rare occasions when it made sense, from an impact-minimization standpoint, to touch pristine flowstone with bare hands. Forbidden fruit vibes. As I stemmed across the velvety canyon, carefully picking spots for my feet and feeling little prickly jabs of pain on the palms of my hands as I carefully touched the forbidden fruit, I felt very in touch with the medium.

At MJA108, 60 stations and 2000 ft of survey in, we decided to call it. It was not quite 8pm then. We had done just about 10 hours of survey. I felt pretty good about my efficiency when sketching profile and doing front sights. 2000 ft in 10 hours, not bad! And it was all nice, significant passage too, trending south towards the Barefoot Boulevard area of the cave. We turned around at more of the same: 30 ft tall canyon passage, with a clear ledge system partway up allowing easy travel at the interface between narrow and wide parts of the canyon. At 5 ft wide, it was slightly narrower than the passage right before that point, but clearly continuing off into the distance. A great lead that I’m excited to return to!

We departed for camp at 8:06pm. Travel through our survey was fast and easy; it probably took no more than half an hour. Travel through the first part of This is Ridiculous felt similarly ridiculous as it did in the morning, but with some care and finesse we were back in Midnight Creek. On the way back, I took everyone through the alternate route back to camp that we discovered on my trip last year: up into the MA borehole at MJ20, left at the MA borehole, then right into our little shortcut that drops you into La Culebra at MK4. I thought this route would be longer but less effort, as it cuts off some low wet crawling in Midnight Creek around MK10-20. Plus, we got to see the massive MA borehole, which is the largest cave passage I’ve ever been in that doesn’t have a name (MA is the survey junction). However, the others thought it was more effort, as it involved hiking up some huge breakdown hills in the MA borehole. I always feel way more at home hiking up steep breakdown than crawling, so perhaps I discounted the effort required to go that route. Oh well. We made it back to camp at 9:51pm, a very reasonable time. I was in bed before 11pm.

I drank 1.25 L of water and peed 825 mL while I was out. 3 L of water was way too much to pack. I ate 3 of my cheese and pepperoni wraps, about 1/2 a cup of pepitas, and a bunch of dried mangoes. That was enough food, although it may have been nice to pack more to have more margin for error.

Saturday, June 28

We woke up at 7:15am, to an alarm that James had set to give us just over eight hours of sleep. As much as the lead we turned around at yesterday was excellent, and very much worth returning to, James wanted to go to a different part of the cave, and had different leads in mind. The leads were in the Borderlands area, MK60 and beyond. There the map was full of question marks branching out from the known passage walls. It looked as if most of them were going in a direction that would make them quickly connect with known passage, although they were worth checking out, as surely a place where the passages had such porous walls would do something interesting.

We left camp at 8:55am. The route to Borderlands turns right at Harmony Hall, instead of left up Midnight Creek as we did the day before. There we followed the La Culebra passage down 60 stations of walking down huge borehole, with only a few short ducks under boulders or other low ceilings. We got to the first leads at MK60. The first few leads starting at MK60 seemed uninteresting, almost certain to connect into known passage. At MK64, however, James had his interest piqued by a lead. It was a low crawl underneath the right (west) wall of the borehole. Although it looked pretty terrible at 1 ft tall and 30 ft wide, it’s also where the dry streambed in the passage appeared from, so it was interesting. In the borehole beyond they never again saw that (presumably not always) dry streambed, so following it could be promising. Following the water and following the air are always good strategies for cave exploration.

I will admit that I was somewhat skeptical of this lead, a 1 ft tall belly crawl underneath the wall of a borehole. But apparently they were unable to follow the lower level of the borehole much further, so this could be a promising way on. I knew that James knows this cave quite well, so I would see where it would take us.

We were at our lead at MK64 at 9:56am, just an hour out from camp. We mostly stuck to the same roles as yesterday, with a few changes. I just sketched profile instead of doing profile and front sights. Ben did plan and cross sections, although I picked up a few cross sections at the very end. Beth did front sights, James did photography (Fort Stanton is the only cave exploration project I know of that requires photographs at every station), and James and Beth split up inventory.

The passage continued as belly crawling for at least 300 ft. It was easy survey, as we got long shots in the flat pancake crawl, and there was very little detail to sketch. It kinda felt like cheating just drawing straight lines connecting the floors and ceiling on two stations 30+ ft apart, but that’s really what it was like. James was able to set stations at spots where there was enough room to sit (not stand) up, which made the survey more comfortable, as the passage was otherwise entirely belly crawl. At least we had some nice fossils in the ceiling to entertain us. For that reason, Beth named the passage the Beachcomber’s Crawl. Fitting, not only because of the marine fossils, but because of the damp sand in the floor.

Interestingly, the sand in the floor was damp, in places damp enough that when you pushed a small hole into the sand, it filled with water. I made sure to keep the survey book out of the sand, and planked hard to keep my torso out of the water. Ben announced that without a fleece, he wouldn’t last long in these conditions, temperature-wise. I had a fleece, but I certainly wouldn’t complain if our team called it and moved on to a better lead.

However, after 7 stations and 300 ft, we reached a nice standing room. A flowstone slope led up and left (kneepads off to keep it clean) into another standing room. Here we tied in to MK68, which was in the same La Culebra passage we had just departed a few stations ago at MK64. We had surveyed a small loop.

Where we returned to La Culebra at MK68, there was another lead. Probably the original team (which included James among others) left this lead because it was exquisitely clean and decorated. We donned our shoe covers, took off our gloves and carefully entered a dried-up pool basin, full of exotic formations which were all a deep, vibrant yellow.

The floor was coated in translucent yellow pointy crystals that grew like clusters of bushes out of the ground. These had a very different shape and texture than the cauliflower-like coating of Snowy River, although they nominally formed from the same process—standing water in pools slowly evaporating away and precipitating out calcite crystals. I wonder what made them so different? Little stubby pool fingers grew out of the walls just below what was an obvious old water line. Shiny yellow velvet flowstone poured down the walls at the entrance to the room, and from a too-tight hole in the ceiling, and turned into draperies as they spilled over overhangs. Long ribbons of striped, colorful cave bacon snaked their way down the sloping ceiling. If these words don’t make any sense, just look at the pictures of it all. All these are at MJA12-13:

The room was intensely decorated and beautiful, but kind of small, so there wasn’t a ton to sketch. So we took plenty of time to take pictures and document this find. The passage ends there, so it’s very possible that zero humans will ever return. Which is a little sad if you ask me!

We returned to where we tied in to La Culebra at MK68 and had a lunch break. It was about 12:30pm then. We had the option of continuing on to Borderlands to look for better leads (and there were certainly good leads there), or continuing down the low belly crawl that went southwest from ML7, instead of east into La Culebra. James decided to continue down our belly crawl. La Culebra ended in breakdown there, and the way on was via the upper level Borderlands passage, so if this low stream crawl connected to whatever existed of La Culebra on the other side of the breakdown, we would be off to the races surveying big borehole. So down the crawl we went.

Two stations and 100 ft later at ML15, the passage opened up into a nice standing room with a deep pool. It had a breakdown wall on the right (east) side that was probably the only thing nominally separating it from the continuation of La Culebra, although I couldn’t find any gaps in it. So we continued down yet another wet sandy crawl. At ML20, another 200 ft of crawling away, we were in a stand-up walking stream passage with bedrock walls all around that was clearly a distinct passage from La Culebra. There was even a lead on our right (southwest), going away from La Culebra. It’s a pretty good looking lead—you’d scramble up an easy flowstone slope (maybe bring a handline for everyone after the first person), although it’s clean enough that you’d probably want a full clothing change, or at the very least clean shoe covers and gloves. By now, we had consistent pools and/or a flowing stream in the floor, and it took a little finesse to avoid stepping or crawling in the water and getting wet.

Now we were in passage that seemed like it would be at home in Kentucky, not New Mexico: a muddy walking tube (~10-15 ft diameter) with a stream in the floor. James was right! The dry streambed he saw coming out of the wall/floor near the end of La Culebra really was a promising feature to follow.

The passage continued this way for quite a while: pleasant walking passage, with a flowing stream, mud on the floor, and some sporting moves to get across the stream when the walkway or ledge on one side ended. The passage was large enough to be pleasant, but small enough to make for easy and fast sketching. I took a few cross sections from Ben, although he did almost all of the cross sections that day.

I think this was the most flowing water I’ve ever seen in Fort Stanton, with the possible exception of a Snowy crossing in the far south (SRS770) that had tons of flowing water the previous May. Harmony Hall didn’t have this much water when I was there and it was flowing last May.

Right around the time we were starting to feel like it was time to turn back, near 6pm, the passage changed character from the familiar muddy tube with a stream. It became a keyhole, with the stream in a canyon below a ledge system we were walking on. The keyhole with the stream branched off the keyhole to the right at ML39. The keyhole turned left and continued as a dry passage full of gypsum decorations, which we hadn’t seen up to this point today. Where the stream went right and the tube went left was a high lead going straight ahead up a mud-covered flowstone slope (easy climb, but maybe bring a handline for everyone after the first person). We had reached a four-way intersection! We figured this was a good time to turn around and head back to camp. As much as it is tempting to push onwards in moments like these when the passage gets really good, those moments can also be good times to end the day and leave the remaining survey for the next trip, because the excellent lead(s) will entice you and others to return. We wrapped up the day with 40 stations and 2022 ft of survey. Pretty good for 8.5 hours of survey.

I sketched one last cross section at the dry walking lead at ML40, and we took a group photo there. After a quick snack/water/pee break, we departed for camp at 6:30pm. We left a 7 ft wide 10 ft tall dry walking lead (ML40), a 16 ft wide 12 ft tall inverted T-shaped wet stream passage with great airflow (ML39), and a 10 ft wide 5 ft tall stoopwalking lead up that short flowstone climb (ML38). I am excited to return to all of those leads! The airflow at ML38 is especially interesting. You can clearly see the flagging station ML39 blow in the wind; it’s not some phantom air that cavers always feel because they want it to be there. Although, it’s possible that the air there is just caused by flowing water moving air along with it. I’ll just have to return to see! Fort Stanton has very little airflow, since the airlock at the start of the Mud Turtle crawl seals up the air pretty good, so any passage with obvious airflow is significant and interesting. I often feel air in the Black Rock Bypass, but I didn’t feel it going in on Thursday. Besides that and the airlock when it’s open, I’ve never felt air in the cave.

The trip back was uneventful. Our surveyed passage went by surprisingly fast, probably 20 minutes. The walk back down the La Culebra borehole was as serene as ever. I ate a bunch of dates before our commute back to camp and got a sugar high, so by the time I got to La Culebra I was buzzing and ready to power walk, and James had to politely remind me that we should stick together, that I shouldn’t sprint off to camp as fast as I could. After a stop to fill up water at MJ10, we were all back at camp at 7:49pm. Not counting the water stop, it’s about an hour of travel time to get to the excellent complex of leads at ML38-40.

We spent the rest of the night packing up camp as much as we could, as we would be getting up bright and early the next day for our 12ish hours of travel back to the surface. Ben and I had work Monday morning, so we wanted to get an early start and get back to Carlsbad/Albuquerque at a reasonable hour Sunday night. I used a plastic putty knife that Beth brought to scrape all the mud off my dirty kneepads, elbow pads, and pack, to make them lighter and lower volume for the trip out. Not only was it extremely effective in getting debris off those items, it was also extremely satisfying to use. It kind of felt like doing an ASMR-esque activity like cutting up soap in this video, but without the satisfying sound, only the satisfying tactile and visual sensations. See also this pressure washing ASMR video to get a sense of the vibes. It’s like that, but a plastic putty knife instead of a power washer. Thanks Beth for bringing it! I’ll bring one next time for sure. And maybe stash it at Midnight Junction, if that’s allowed?

We decided to get up at 4:30am the next morning. That should get us out of camp by 6:30am, which would mean the surface by 6:30pm if we stuck to the same pace going in, or maybe a little later due to our tiredness from several days of caving and the extra weight we’d have from the several days of pee and poop we’d be carrying. I was in bed by 9:35pm. Early enough, I guess.

For my reference—I drank 850 mL of water and peed 750 mL while out of camp for the day.

Sunday, June 29

We woke up to James’ alarm at 4:30am and were out of camp by 6:15am, pretty fast. My pack started out pretty full, but that was with 2 pairs of elbow and knee pads stuffed in there, as you don’t need them in between Midnight Junction and Rough Country. With one pair of kneepads and elbow pads out of the pack, I knew I would have enough space for the dirty pack cover that was waiting for me at Black Rock Bypass, and the two small bottles of pee I left at Finger Lakes and Mount Airy.

The trip out was fairly uneventful. We got to the start of Rough Country at 7:49am. At first we didn’t notice, and started doing it in shoe covers. We had already been in and out of shoe covers a bunch so far, so it was easy to miss the spot where the dirty breakdown starts and doesn’t stop for a while. But I noticed a tricky climb-up that I distinctly remembered as one of the cruxes of Rough Country, and we stopped to change. Well, most of us at least. James decided to keep his clean shoes and dirty shoe covers on. I changed into dirty shoes because there is some kinda serious exposed scrambling in Rough Country, and I didn’t want to do it in shoe covers. As a reminder, James Hunter is the guy who did the entire Kansas Twister climb in Lechuguilla with about a dozen bolts, all drilled by hand. He’s hardcore!

We stopped in the middle of Rough Country to install a webbing handline at one of the sketchier exposed traverses. In this spot, you traverse across a sloping dirt ledge above a ~30 ft drop. It’s secure enough that plenty of people have done it and no one’s complained too much, but I figured a handline would still be nice to have, so I suggested we do it. We took a ~50 ft piece of webbing stashed at camp in the morning and left it at the traverse, tied to boulders on either side and tied to a rock protrusion in the middle. That should make the traverse a little more chill, but still serious, because no one will have a harness to clip into the traverse line. It’s just there to grab on to in case you want it.

We got to Finger Lakes at 8:50am. Similar time so far to the way in. I didn’t even fill up water there, as I hardly drank any of my liter that I took from Midnight Junction, and I had a liter and a half stashed at various points for the way out. Higher Hopes was our next major stop; we did a lunch break there at 11:40am. Again, about the same time as on the way in. The near side of Mount Airy, where I stashed some food and water, was closer than we remembered, and we were there at 12:28pm. I had some of my water I stashed there but didn’t eat any of the food. At 1:35pm we were all at the far side of the Black Rock Bypass. James and Beth went first, and Ben and I hung out and cooled off while they changed. When Ben and I got to the Black Rock Bypass, I noticed that it was distinctly windy. It did not have this air on the way in! But I do remember it having air like that on a prior trip. At 2:45pm, we were all on the near side of Black Rock Bypass. We powered through the last crawling section of Snowy, which is also probably the worst crawling section on Snowy on the current route, and were quickly back in the nice walking passage with plenty of shoe cover spots near Turtle Junction.

Overall, the trip out was fairly uneventful. There was one spot where a magic carpet slipped out from under me on a slope and I stepped onto the mud in my clean shoes. I did have to take 10 minutes after that to clean the mud out of my shoes, but that was no big deal. It happened on a magic carpet, so Snowy didn’t get dirty at all, and I kept the magic carpet clean. In fact, while I was cleaning my shoe, Ben staked down that tarp so it shouldn’t slip again. Many of the magic carpets need maintenance work like that—either staking down, or lengthening to get them away from obstacles, or cleaning. I’m definitely gonna submit some proposals to lead some route maintenance trips when they open up trip proposals for next year.

At 4pm we got to Turtle Junction and had to bid adieu to Snowy, for now at least. James, Beth, and Ben will surely be returning within the next year, and they might even do another trip this season before the BLM closes the cave for bat hibernation in November. I will be doing a few more trips this season, so my farewell to Snowy was only temporary.

Since I no longer had to care about keeping myself or any of the cave clean from this point, I didn’t even bother changing out of my clean Snowy clothes. In fact I took my shirt off to do the last part of the cave, which involves some strenuous crawling. I caved out in wellies, flowery short shorts, and no shirt, which might just be the strangest caving outfit I’ve known. But it was certainly a good/effective outfit given the circumstances. We were out of the cave at 5:15pm, 11 hours after we left Midnight Junction. A little under an hour faster than the way in. Not bad given that we were carrying extra weight and were tired from 4 days straight of caving and from waking up at 4:30am.

When we got to the surface, Ron Lipinski, Steve Peerman, and Kathy Peerman were waiting at the bunkhouse for us, and they had pizza! Thanks so much guys! We regaled them with tales of our epic trip and showed them photos while we ate and cleaned off our gear. They were psyched to hear that we surveyed 4000 ft, and left a great walking lead or leads both days! Our trip increased the cave’s surveyed length from 52.77 miles to 53.49 miles, we later found out.

When I dumped my pee out while emptying my pack and cleaning my gear, I realized that I only ended up peeing under four liters over four days. Much less than the five liters I peed in three days on my last Snowy River camp trip. I didn’t feel particularly dehydrated this trip. I attribute this low pee volume to three things I did differently than last time. First, I drank very little coffee. I slept well enough that I didn’t need it to feel awake and energized in the morning, and I pooped regularly in the morning without it. Second, I ate lots of salt. Apparently this can actually make you drink less water and pee less, which is unintuitive. Third, I followed a habit I’ve been building around water consumption in hot caves: when stopping to drink water, I force myself to wait a few minutes to cool down before drinking any water. I find that right when I stop for a break, I’m hot and will drink tons of water if it’s given to me. But if I wait even a minute or two, I cool off and don’t feel the need to drink that much water—a few sips will do it for me, rather than half a liter.

I felt pretty good about what I packed this trip. I’ll make just some minor changes next time. One thing I would not bring next time is a third liter of water. Two Nalgenes for drinking and one for peeing during the day (and a dromedary to carry it all out the last day) is plenty. I could have brought more food, although what I had was adequate. I never wore my fleece survey layer, although I like to always have a warm layer in case someone gets hurt and needs to stay put for a while. Maybe a thinner one would work next time. A long sleeve shirt and no elbow pads would save pack space over a short sleeve shirt and elbow pads, but it may not be worth the heat cost. I definitely want a plastic putty knife next time. Maybe I can pare down my personal first aid kit, since there are stashed first aid kits at various points in the cave up to Midnight Junction.

After cleaning up all my gear and leaving it to dry in the gear shed, I started the drive back to Albuquerque. I forget exactly what time I left, but I made it back at a reasonable hour, early enough that work the next day felt great. I look forward to returning to the cave and all of its excellent leads! Thanks to James, Beth, Ben, our surface watch Ron, the BLM, and the Fort Stanton Cave Study project for making this trip happen.

Now that I’ve been on two far Snowy camp trips, I am allowed to lead my own exploration trips. I’ll certainly be doing that next year, after the next cycle of trip proposals this winter. Stay tuned to see what I get up to!

Subscribe now

This past Memorial Day weekend, I did a 3-day camp trip in Lechuguilla Cave. Lechuguilla Cave, or ‘Lech’ as cavers call it for short, is in Carlsbad Caverns National Park, but it is not connected to Carlsbad Caverns (the subject of my last post). Lech is longer, deeper, and in many ways, more spectacular than Carlsbad Caverns.

Lech has 152 miles of surveyed passage that forms a dense maze of huge boreholes, tiny crawls, pits that bore deep into the earth, and everything in between. There are still leads remaining in the cave, so that 152 mile number will surely continue to grow. At 1589 ft deep, it was the deepest cave in the US until Tears of the Turtle Cave in Montana overtook it in 2014.

The cave is not known for its length and depth so much as the incredibly decorated, beautiful nature of much of its passage. Lech is filled with massive, pristine, and awe-inspiring rock formations found only in caves, such as stalactites, stalagmites, and many even more bizarre features. For that reason, access to the cave is strictly controlled by the park, and people can only go for approved purposes such as exploration, scientific research, or other tasks that support these goals.

Both exploration and scientific research trips in the cave have slowed down in recent years due to various policy and personnel changes in the park. But this past Memorial Day, I got invited on a trip there. With 6 people/2 teams, there would be multiple objectives. The task for our 3-person team was to help replace aging ropes in a deep pit series in the cave. Because this pit, called the Kansas Twister, is in a somewhat remote section of the cave, about 6 hours of travel from the entrance, we would do it as a 3 day camp trip. 1 day to travel to camp, haul a bunch of rope into the cave, and do some minor objectives near camp; one day to haul the ropes out to the Kansas Twister and replace them; and one day to exit the cave and haul all the old rope out of the cave.

The other thing that makes this cave noteworthy besides its large, beautiful decorations and besides its long, mazy character is its hot temperature. It’s 68 degrees, which may not sound hot, but the 100% humidity makes it very difficult to cool down once you get warm, as sweating isn’t very effective at cooling you down in those conditions. So the cave is a very comfortable temperature when not moving, but the slightest bit of exertion gets you very hot. I run pretty hot when caving. In fact, I consider it my Achilles’ heel of caving; it’s the best way to defeat me on a long hard trip! So I made it a point to drink tons of water in the days leading up to the trip, to front load my hydration as much as possible. I was definitely a little bit nervous about the heat leading up to the trip.

After work on Friday the 23rd, I drove straight down to Carlsbad and stayed at my friend Georgia’s house. She would be on the trip too. The third person on our team, Michael, arrived at 2am after a marathon late-night drive, long after I had fallen asleep. I went to bed excited to spend 3 days camping in one of the world’s most spectacular caves, but also a little bit anxious to see what the heat would do to me.

Saturday, May 24th

We took our time leaving the house Saturday morning, and didn’t get to the trailhead until around 11. Hunter, Garret, and Karla, the other 3 on our trip, arrived much earlier, around 8am, and had the benefit of the cool morning air for the approach hike. By now it was quite hot out, in the 80s with full overhead sun, although a faint but steady breeze made it feel relatively pleasant. I chugged as much water as I could before starting the hike, as I didn’t want to lose my well-hydrated state that I had spent so much effort cultivating the previous few days. After packing the ropes that the other team left us to carry to and into the cave, it was 11:15 and we were off to the cave.

The hike starts out following the Guadalupe Ridge Trail from the Walnut Canyon Desert Drive route in the park. After about a mile on the ridge trail, you leave the trail and go up a ridge on your right (North). There is allegedly a trail here, although we never found it on the way in. After a while spent walking around in circles looking for the trail, I decided to skip that and just walk in a straight line in the direction of the cave and ditch the others (sorry). This is a classic move I pull: I’m not very good at finding nor following trails, but I don’t mind bushwhacking and am always down to trudge through the shortest-but-probably-not-most-efficient route to get from point A to point B.

I also took all the rope so that the whole group would get there faster. I got to the cave around 1pm. Despite hiking through the midday sun on a hot late May day in Carlsbad, I felt like I was at a pretty good temperature. I figured maybe this would bode well for the cave, which is notorious for being very hot, and would be the hottest cave I’ve ever been in. Although the dry desert heat, which I typically do very well in as I sweat a lot, is certainly a different beast than humid cave heat, which is less affected by sweating and which I typically do worse in. I was still anxiously waiting to see what the cave air would feel like.

I drank half a liter of water and left the other half of that bottle on the surface, stashed there for the exit 2 days later. I hung out for a bit in the shade provided by the entrance sinkhole and cooled off while waiting for the others. They arrived shortly after me, and we were all in the cave by 2pm.

After dropping the entrance pit to a ledge partway down, my first impression was that it was not obvious which way was the way on, and I wondered how confident the original diggers were that they were digging in the right spot. Lechuguilla was famously just a surface pit with a dig blowing air for several decades after it was discovered. Cavers dug at it for many years, following the air, but apparently never completely sure that they were digging in the right spot, as air was coming from several spots in the entrance pit. Clearly they were digging in a good spot, as this 150+ mile cave is certainly the greatest cave digging success story ever told.

After squirrelling down into a little hole in the large chamber in the middle of the entrance pit, past another 10’ drop, I arrived at the airlock. My second impression of the cave was that the airlock and subsequent shoring around the old dig was an extremely impressive engineering/construction project. I was expecting that and had heard much about it before, but seeing it in person still imparts more awe than anyone’s description can. The airlock is a series of two stainless-steel, standing-height doors on opposite sides of a room that 3+ people can stand in. On the side going into the cave, a 3 ft diameter stainless steel tube bores down into the Earth at a 70 degree angle for about 40 ft, with a ladder going down it. All of this was dug open, then reinforced with stainless steel tubing, by cavers over many years. I thought that some of my dig projects got a little ridiculous when they lasted a month or two; but this multi-year effort was clearly in its own class!

The tubing pops you out into a small walking passage. There I followed a trail flagged with orange flagging tape, which would be there pretty much the entire way to camp, and then some. At first I thought the flagged trail was a little excessive, but it later became apparent how useful it was when the mazy character of the cave asserted itself. The cave is full of rooms where multiple similar-looking tubes shoot off in many different directions, and there would be no obvious way to tell the way on if not for the flagging.

The small walking passage continues for a few minutes to the first of many rooms with pools and formations. I’m always amazed when I see water in caves in the Guadalupe Mountains, given how dry it is on the surface, but it often hides just beneath the surface, and Lech was no exception. The temperature here felt quite reasonable; not much hotter than other Guads caves I’ve been it.

2 short, low-angle ropes interrupt that small walking passage before you get to the first big pit in the cave, Boulder Falls. Boulder Falls is a 150 ft pit, freehanging most of the way that starts out large and bells out to become huge at the bottom. At the bottom of the pit, I stashed a half-liter of water for the way out and waited for Georgia and Michael while taking in the views of the massive chamber I was sitting in the center of. The others stashed some water there, and we continued walking/scrambling down the trail, as the huge chamber we just dropped into continued downwards as a similarly huge borehole.

Shortly below the bottom of Boulder Falls, I noticed a spot where the temperature suddenly changed. Above this one spot, I thought that the temperature wasn’t so bad, and this was gonna be easy. But for some reason, there was one spot where the air suddenly changed from cool, not-totally-humid aid that seemed fresh, like it was from the surface, to warm, humid, thick cave air. The heat hit me like a ton of bricks, and I could tell that I was gonna be in for a hot, sweaty 3 days. The dry desert in the 80s with full sun at high noon felt cooler than this.

I was caving with no shirt and short shorts at this point, because the cave passage was so friendly and wouldn’t tear up my skin, like most caves would with that (lack of) outfit. So while I was sweating profusely, I felt comfortable enough.

The borehole below Boulder Falls eventually becomes full of white gypsum blocks, at which point it’s called Glacier Bay. There, probably because gypsum is more water soluble than limestone, the passage opened up into even bigger borehole, at least 100 ft tall. It also started branching off into several boreholes that all continued in different directions into the Earth, ominously fading into blackness. The branching boreholes were inviting (I couldn’t help but wonder where all those passages go!), but also ominous: they had a distinct mazy character to them, and the way they faded into blackness as far as your headlamp could illuminate them suggested that it would be very easy to get lost by wandering around in there. This is where the flagged trail running through the passage became essential.

The trail snakes between huge gypsum blocks, and scrambling around and between them makes Glacier Bay fairly high-effort for a large borehole. The gypsum forms distinct shapes and features rarely seen in limestone that were fun to admire on the way down, such as huge fluting, sharp fins, shiny polished rock, and circular holes going straight down from ceiling drops eroding channels into the rock.

Glacier Bay ends at another notable feature I had heard much about: The Rift. The Rift is a huge crack in the Earth that goes down at a 70 degree angle and continues left and right from the end of Glacier Bay. A short rope leads up to a ledge in the middle of The Rift that continues to the right, towards the rest of the cave. For much of The Rift, you can walk on ledges that continue as easy walking, although just as often the floor disappears and it is apparent that you are right in the middle of a several hundred foot crack. Much like the branching tunnels of Glacier Bay, when you can see down The Rift it fades away ominously yet invitingly into blackness as far as your light will illuminate. At these exposed sections, fixed traverse ropes make for casual travel, as long as the exposure doesn’t get to you. Getting across the first time to get the ropes set up would have been interesting, although probably not too challenging, as the diagonal nature of The Rift means you could probably slab traverse across without too much difficulty. The Rift was some of the more sporting section of the cave on this trip, as it had a few tight spots where I had to take my pack off and finagle it through. I had two coils of rope, 390 ft of 11mm pit rope in total, tied to the outside of my 42 L camp pack, which loved to get stuck in the few sections of the cave that got a little tight. With all the rope I was carrying, and the excessive amount of water I had, my pack probably weighed 40 pounds. Taking it off and shoving it through a few tight spots was fun, and reminded me that I was in a cave, not just a scenic hike. The exposure and spots where you wouldn’t want to fall also added to the fun, although the traverse ropes make it so that a fall would be merely undesirable, but not dangerous.

After The Rift, you get to the first junction in the cave, where the orange flagged trail splits and goes 2 ways. This was EF Junction. I left another half-liter of water for the way out and took stock of the cave so far. With the most strenuous part of the journey to camp done without me overheating, I was feeling pretty good about the temperature of the cave. We took a quick jaunt down the right branch from EF Junction to look at a pretty flowstone room, then continued down the left branch, which was the way on to camp.

A few more minutes of walking and one more traverse ropes led to the next junction in the cave. A pit rope went down a pit on our right, while walking passage continued straight ahead. The pit had flagging with a label indicating that the Western Borehole was down that way. The pit is The Great White Way, and it was our way on. Georgia went first, I followed, and Michael brought up the rear. The Great White Way is a series of low-angle, somewhat tight ropes that were kind of annoying. But they went quickly, and after several ropes we found ourselves once again in a huge borehole. We stopped at the bottom of the pit for food and water, as this would be the last good spot to stop before camp. While it’s possible to stop in the borehole between us and camp, it’s full of large, uneven breakdown, without many good flat spaces where three people can hang out, so Georgia (the only one who had been here before) said we should fuel up and hydrate here and power through the rest of the way to camp. I finished my second liter of water, finished the second sandwich I packed in, and continued onwards to camp.

This next section of cave had several sequential names that I forget, because it all seemed like one continuous passage. It was a large borehole, although the trail takes you around, in between, and sometimes even underneath large boulders that make travel very scrambly and somewhat high effort. Going between and underneath the boulders often makes it seem like you’re in small passage, even when that space between boulders is in the middle of a huge chamber.

In this section is the first gloves-off part of the cave, called Deep Secrets. A small plastic sign marks the start and end of the gloves off section, where you have to stem and use your hands to push off pristine, clean flowstone. Even with gloves off, it’s difficult to keep everything clean. You have to take your pack off and move it with your bare hands, and the pack and rope are somewhat dirty and rub against the walls. At least the bag is likely cleaner than gloves. Touching smooth, clean, wet flowstone felt a bit like eating forbidden fruit, as that is typically a no-no when caving. However, in this spot there is no other way to get through the passage, so you just take your dirty gloves off and get it done. Touching it with bare hands too made it feel exciting; I felt very connected to the cave.

We got to camp, which is in the middle of this borehole where it finally gets to a section with flat dirt floors, at 7pm. Hunter, Garret, and Karla were all lying down at the highest camp spot up on a terrace, patiently waiting for us. We rushed to greet them all and rave about what a fun trip in it was. I had met Garrett and Karla briefly at Fort Stanton, but didn’t know them well, so I reintroduced myself. I knew Hunter quite well, but I hadn’t actually caved with him before. I was excited to finally cave with him, on his first camp trip to boot.

At roughly five hours from the entrance to Deep Seas Camp, we were a little on the slow side, but within the normal range of travel times. The slow travel time definitely helped with not overheating. I made it to camp while only drinking two liters of the 5.5 liters of water I planned on bringing! Hunter, who like myself tends to overheat badly in caves, was jealous, as he had to drink 5.5 liters of water on the way in.

We unpacked a bit and started to set up camp, although not all the way, as we wanted to continue caving that night. Garret and Karla took a somewhat secluded couple’s spot. Georgia took her own private spot. The rest of The Boys (Michael, Hunter, and myself) took the biggest spot high up on that terrace and declared that terrace the Dude Bro camp spot. We all bro’d out up there on that spot the whole trip. As did the other 3; our spot was the central hang out spot when we were at camp and could easily fit all 6 of us.

I unpacked my bag and made everyone chuckle when I unearthed a full 3.5 L plastic jug of water that I didn’t touch on the way in. At least I was fully prepared with plenty of water! I surprised even myself when I unearthed yet another full 1 liter Nalgene that I had forgotten about! I was supposed to stash that somewhere for the way out, but I forgot about it and carried it all the way to camp. So I ended up carrying 6.5 liters of water to camp, and only drank two of them. That’s 13 pounds! Oh well. Since I only drank 2 liters of water on the way in, and had one liter stashed on the way out and a half liter at the entrance, I wouldn’t have to carry much water for the way out at least.

After unpacking and partially setting up camp, it was time to continue caving. The night was young! One objective close to camp that Hunter, a cave manager at the park, wanted to do was to inventory some sites in the cave where there was science equipment stashed from long ago in unknown condition. This was in the Pellucidar area of the cave, about 10 minutes from camp.

Michael was pretty worked from the trip in, so he decided to stay back while we went to Pellucidar. He wanted to make sure he had energy for our big day tomorrow, replacing all the ropes in the Kansas Twister. So the five of us followed Garrett, who knew the way to Pellucidar, out of camp back the way we came.

After five minutes on the trail, we exited off the flagged trail to the right and got on an unmarked but still obvious and impacted trail of brown dirt going through white gypsum. As the flagged trail descended down a slope in the main borehole, this side trail stayed high and angled towards the passage walls, where an attractive-looking tube shot up into the ceiling. The tube was not too steep, 60-70 degrees, and stepped out, so Garrett easily scrambled up it, and we all followed to a ledge. The scramble was was easy, but it was on loose popcorn-and-dirt ledges that all seemed like they were about to break off.

From that ledge, a rope led up a move vertical part of the tube. While Garrett climbed the rope, I looked at the tube, which was likely free-climbed first to get the rope up there. Lechuguilla Cave is in a federally designated wilderness area, which means no power drilling. Hand drilling a bolt is arduous and time-consuming (takes about a half-hour), which means that climbs to access new passage often require a combination of bold free-climbing, sketchy aid climbing with creative trad gear and pitons, and laborious hand-drilling. Much of Lechuguilla was explored this way, including the Kansas Twister that I was to see tomorrow. I tend to think of myself as a fairly bold climber, but I’ve done very little of that kind of bold climbing for cave exploration purposes—all my exploration climbing in caves has been with a power drill. I know there are more leads like this in Lechuguilla, and I would like to explore them! So I am always thinking to myself when I encounter a passage that was explored via climbing, “could I have climbed this”? For this tube leading up to Pellucidar, I thought the answer was yes, but it would be a bold climb for sure.

Atop that rope was yet another large Lechuguilla borehole, this time filled with much more decorations than the previous ones. Aragonite bushes and stalagmites lined the trail that weaved through the borehole. While Hunter checked out some science station, I simply wandered around the trail and admired the formations, which I rarely have time to do on cave trips.

Garrett called our attention to a unique formation at one end of the hall that was apparently known about in only a few places in the world, this being one of the best examples. Subaqueous helictites. They were in a pool that was just beyond a small space that only one person could look through at a time. After Garrett found it, he called Hunter and everyone else over and had us look at them one at a time. I didn’t expect much, as I have seen a ton of helictites, but these were astounding and I audibly gasped when I saw them. They were tendrils of spaghetti-like rock growing at all sorts of funny angles, from the surface of a pool downwards, as the water in the deep pool turned rich, vibrant turquoise.

Those subaqueous helictites looked only superficially similar to the helictites I’ve seen in other caves. Although they both resemble spaghetti-like strands of rock growing in all sorts of funny directions, standard helictites are more curvy and bulbous, while these have a distinct angular look and feel to them.

Although the subaqueous helictites looked very different than standard cave helictites, they did remind me of one cave formation I had seen before. The Flying Spaghetti Monster in Helms Deep, a cave in Tennessee. The Flying Spaghetti Monster kind of looks like helictites, but is definitely distinct. It consists of straighter segments of rock that shoot out at several different angles and mostly stay at the same direction with a few sharp angular bends.

Helms Deep is clearly phreatic and likely formed while submerged. So it’s possible that the Flying Spaghetti Monster and similar nearby formations could have formed while submerged via the same process that forms the subaqueous helictites in Pellucidar in Lechuguilla.

Garrett spent some time looking for another room called the Diamond Room while I sat around admiring the pretties. He couldn’t find it, so it was time to go back to camp. Hunter didn’t want to do the loose free climb down from the ledge below the rope that goes up into Pellucidar (the one with the popcorn-and-dirt ledges that all seem like they’re about to crumble), so I volunteered to go back to camp, grab one of the ropes there, and bring it back up. I downclimbed the steep tube, which was much less pleasant on the way down than up, then power walked to camp to retrieve a rope. I grabbed a short 100 ft rope and brought it back up the tube to the ledge. I rigged it to a boulder, and those still left on the ledge used it to descend safely. I was the last one down, and did the downclimb one last time without a rope after throwing it down so we could use it elsewhere in the cave. I don’t blame him—that climb is a little sketchy, and if people return to Pellucidar with any regularity, we should leave that climb rigged.

There were no more short, nearby tasks to do, so it was time to head back to camp. I finished setting up my camp spot, made my dinner (my classic Dinner Mix, this time with extra bacon), and did a run to the water supply to have plenty of water for the night and morning.

Georgia took me to the water source, a beautiful pool called Lake Louise, five minutes away from camp. At this point I was just caving in my camp shorts (very short) and my barefoot-style shoes; no kneepads nor harness nor shirt. I was caving as close to naked as was reasonable! It actually felt pretty nice. Normally when caving I’m armored up, with long pants, stiff shoes, kneepads, and other protection. Caving with minimal stuff on me in a cave friendly enough where that was feasible felt refreshing and (I’m being a bit melodramatic here) liberating. Especially with my barefoot-style shoes, I could feel all the details of the 3d structure and texture of the ground through my feet. Whenever my bare skin touched rock, I felt a pleasant frisson of cold on my skin, much stronger than it would be if mediated by a layer of clothing. I felt very in touch with the medium.

Despite all the pretties I had seen that day, I was not prepared for Lake Louise. I was astonished yet again by the cave’s beauty when I got there! It was incredibly decorated in a totally different way than Pellucidar. Instead of aragonite bushes and stalagmites, it was colorful mammilary formations. Cave mammilaries are smooth, spherical blobs of rock. The ones here were huge, watermelon-sized, and they abruptly switched from pearly white to a vivid yellow color at what must have been an old waterline. The current waterline of Lake Louise is 10 ft or so below that old waterline. The mammilaries at Lake Louise are so classic, in fact, that the NSS website’s page on mammilary formations has photos of cavers collecting water from Lake Louise.

Georgia showed me the water collection protocol (there is a whole sequence of steps you take to avoid potential contamination of the pool), and we filled up several bottles that would last us all through the morning. When we got back to camp, there was one last surprise waiting for us. Garrett and Karla packed Nacho Libre masks in a bunch of different colors into the cave for us to wear as a little joke. They looked cool, and we played some music, danced, and ran around wearing our masks for a bit, until the foam on our faces made us much too hot. I like when people bring little Easter eggs on cave trips like that—see my Silent River Cave post for an example when Sam Marks brought some colorful patterned ties as novelty items to wear while partying it up on a cave trip. Maybe I should bring the novelty items on my next long cave trip. I’ll take suggestions in the comments…

We kept the music going a little while longer and kept dancing long after the heat forced us to shed the masks. Around 11 pm, we started to approach the going-to-bed state, and we decided on an 8 am wake-up time the next day. As various little tasks kept us up and midnight approached, I suggested a 9 am wake-up time. I wanted to give us 9 hours of sleep time, which I like to have if I want to perform at a high level the next day. Georgia loudly and enthusiastically agreed as soon as I finished my sentence; I figured out this trip that she is not exactly a morning person. Hunter changed his alarm from 8 to 9 and we were off to bed.

Sunday, May 25th

I woke up to Hunter’s alarm on a mostly-flat sleeping pad. Oops; I should have checked to make sure my pad was in good shape before packing it on the trip. Oh well, at least I slept well on it. The cave is warm enough that a sleeping pad isn’t super necessary as insulation from the ground; it just adds comfort when sleeping.

I ate my breakfast of oatmeal with peanut butter and dates, which was different from my usual choice of more Dinner Mix. It was way too much food and I had to try kind of hard to finish it. Despite drinking only 2 liters of water when getting to camp the previous day, I packed 5.5 liters of water for the day. My 3.5 liter plastic jug, and 2 Nalgenes. It was very possible that today could have 15 hours of caving instead of 5. My snacks for the day included a smaller amount of cheese and fatty cured meats (a staple caving food), and more sugary snacks like dried mangoes and peanut M&M’s. I was trying a more sugar-heavy snack strategy in the cave for heat management purposes. Fatty foods tend to fuel me for longer, but they make me hot, as fat takes energy to digest. So I figured more frequent smaller sugary snacks, and being slightly calorie-deficient, would help me stay a bit cooler.

After packing my snacks, my camp bag felt very empty still. I would just have to take more rope! I took 2 cave coils of rope, and off we went, around noon. I started out wearing the cave coils around my shoulder, but I quickly got annoyed with how they dug into my shoulder and interfered with my walking and vision, so I tied them to my cave bag instead.

We left camp around 11am, as one group of six, because we were all going the same way for most of our commute. All of our intended destinations were at least an hour down the Western Borehole, so we all headed that way together. From camp, we started by going towards Lake Louise, then turning right instead of taking the left that leads to the water source. Just after the right turn is a series of two short up ropes (‘the ABC ropes’). I remember thinking that these ropes definitely seem like something I could free climb to explore/get the rope up the first time, before touching a fin of rock and hearing a big reverb sound, like the rock was about to fall off. Loose rock will get you!

After those ropes we popped through a hole that brought us into a huge spherical room coated in pristine white gypsum. There was a perfect large flat boulder that I could lie down on and admire the room from while waiting for the others. If I even slightly changed the angle of my head while looking around the room, the walls would sparkle brilliantly. I felt like I was sitting down in the middle of an IMAX theater, lying down on my comfy flat-topped boulder and enjoying the sparkly light show as I looked around.

When the others finished the ABC ropes and we kept walking, this sparkly chamber got even better. As we followed the trail down the chamber, it grew even bigger, from 50 ft across to probably 100. As we crested a hill in the trail, I got a good long view down the chamber and saw it continue indefinitely as far as my headlamp could illuminate it, and beyond that as it faded into blackness. This was not just one room, but a whole passage with this huge circular cross section and sparkly gypsum lining all the walls and floor. Garrett announced that this was the Western Borehole, and I now knew why I had heard about this passage from several different sources before. It was one of the most incredible cave passages I had ever seen.

As we continued down the Western Borehole, it stayed blissfully pleasant and friendly. It had mercifully small breakdown in its floor, which made travel through it feel like actual walking, not climbing/scrambling as was the case with the previous large boreholes that were full of huge breakdown (Glacier Bay, and all the passage between the Great White Way and Deep Seas Camp). It was consistently coated with sparkly gypsum the whole way, with beautiful formations such as aragonite bushes, huge drip holes, even an aragonite bush growing out of a drip hole like it’s a potted plant, and chandeliers growing out of the ceiling. We took one break for food/water/enjoying the scenic view at a notable formation called the Three Amigos. We figured that a formation with that name was a perfect time to don our Nacho Libre masks and take a silly group picture.

There were two spots in the Western Borehole that I thought had strange and suboptimal rigging. One is a mandatory traverse line on a very exposed walk down a narrow sloping ledge lining a pit that only has one-bolt anchors on each end. Not cool! Another was a rope that ended right in the middle of the exposed section, without protecting the parts of the ledge walk with the highest fall exposure and likelihood. Whenever I lead a trip back there, I’ll be beefing up that rigging with something better.

After two splendid hours of scenic walking down the Western Borehole, it was time for our teams to split up. We were at the Leaning Tower, where the trail branches off to the right towards Emerald City and the Kansas Twister (and many others), and left towards the Far West and Oasis (and many others). The Leaning Tower at that intersection is another awesome and unique formation. It’s a thick (3 ft) pair of stalagmite and stalactite that don’t quite meet to form a column and are covered in bulbous popcorn.

From here, the teams were nominally Garrett, Karla, and Hunter; and Georgia, Michael, and I; but Garrett came with us at first to show us the way to the Kansas Twister and climb up it with us, as the route to there was not clearly flagged the whole way like the main travel route down the Western Borehole was. Karla and Hunter did some smaller objectives close to the junction while Garrett was with us.

It was about 2pm then. We told the other team to expect us back at camp around 2am, but not to worry until 6am. I was excited that I was finally going to have a long, morning-to-morning day of caving! Even if it wasn’t gonna be a super difficult day, it would be a long day, which I like. My caving in New Mexico lately has been lots of short trips, which are still fun, but definitely leave something to be desired in me.

The right turn from the Leaning Tower took us down some small but still stand-up walking passage, then another right turn had us in some real cave passage, with some crawling and shoving the bag required. Nothing too hard, but at 68 degrees with a heavy bag full of rope to shove around, I definitely worked up a real sweat in there. It was intensely mazy, and the route was flagged with little breadcrumbs of flagging instead of the super obvious double-lined trail of the main routes, although it was easy enough to follow.

After a short length of crawling and otherwise more physical caving, we were back in small walking passage. It was full of vibrant blue-green colors, which I don’t see too often in caves. And gypsum hair, that sparkled more intensely than anything in the Western Borehole. Gypsum hair is always hard to get pictures of, as the real wow factor comes when you move around and it lights up with sparkles, but Michael did a decent job of capturing what it’s like.

Eventually, Emerald City opened up from small walking passage into a huge chamber, and we were at the Kansas Twister. At the top of a gravel slope lay the first of several ropes that would take us over 500 ft up, closer to the surface yet also more distant from the outside world. The previous weekend, at Virgin Cave also in the Guadalupe Mountains, James Hunter told me about how he got the rope up that first pitch, many years ago. He free climbed it, without a bolt kit or any other anchoring material, assuming he would find something to tie the rope to or just come back down. After realizing partway up that the latter was definitely not an option, he was relieved to find a small rock wedged in a crack above the difficulties that he could tie off to. I kept that description in mind as I climbed the rope.

And when I climbed that rope, oh boy, all I could think was, “James Hunter is a hardcore motherfucker”. It was a low-angle slope, but it was covered in loose dirt and didn’t have much in the way of good foot ledges, just slopey, dirty, sandy features barely protruding out of the slab. That would be a hairy climb for sure. James said he was young (25), dumb, and full of irrational exuberance when he free climbed it. I definitely had an immortality complex going earlier in my 20s and did some dumb shit, but I don’t know if I was ever bold enough to free climb something like that. Good work James.

Above that rope is a lower-angle gravel slope, and a good alcove in the wall where you can hide out of the rockfall zone below the next rope. We all went to the alcove, snacked and drank some water, and redistributed rope, because we left a 100 ft rope at the top of the slope we just climbed. I also stashed a liter of water there, as I had drunk barely any of the 5.5 liters I had hauled out there.

The next rope goes up about 100 ft to a huge arch bridging the gap between opposing walls of the pit. The wall is sheer vertical and would be unclimbable without a protracted 100 ft hand-drilling effort. And even then, possibly still unclimbable because the rock is total choss. So they got up there by—wait for it—using a slingshot to get a rope over the arch 100 ft up. My understanding is that nailing that slingshot took many attempts over several trips over the years, and that the operation that finally worked involved slingshotting very lightweight, narrow cord over the arch; using that to pull some paracord over the arch; then using that to get some real cave rope over the arch, that someone then climbed.

We all knocked rocks down on that pitch and the next few, so we went one at a time while the others waited above or below in that alcove. When I went, I was not super happy with the exposed walk across a steep gravel-and-dirt ledge that is necessary to get to the bottom of the rope. For most of the walk out to the rope, there’s a trail kicked into the gravel, as you can see below. But right before the rope, the trail disappears, and you have to slab across some bare rock covered with a fine layer of dirt, all above an exposed, steeper slope. You’ll totally die if you fall. I didn’t like that, although everyone else on the trip seemed totally fine with it. Maybe I went too low or too high, or I’m just getting soft.

At the arch is a rebelay, then another long freehanging pitch, then another rebelay right below a big ledge where everyone was hanging out, off rope, waiting. I got up there and enjoyed the view while waiting for Michael to join us. While he was on rope below us, I walked down the stoopwalking passage that departed from the ledge. It was actually really nice! Pristine sparkly white gypsum lined the walls, floor, and ceiling of the circular cross section tube. I stopped when the trail through the passage became not very impacted, just a few faint brown footsteps in fragile, mostly-white gypsum crust. I didn’t feel the need to go any further and impact a mostly-pristine area for some sightseeing that was not our main mission.

It was wild to see the arch in person, even up close and personal as I passed the rebelay on it. I had heard so much about that aid climb. When I was very new to caving, and didn’t even know what Lechuguilla was or that there were caves in New Mexico, I remember my good friend Christian who got me into caving telling me stories, that I kind of thought were fake, about people aid climbing up 500 ft domes via slingshot in long and deep caves out west. Now I was finally there, and it was definitely not fake.

I had also heard stories about how bad the rock was in the dome. That’s totally true! I knocked down several rocks at the arch, as did the others. And that’s after a good amount of people have traveled that route and cleaned things up. When Derek Bristol first went up to the top of the arch, so I’m told, the rock was so chossy that he declared the rest of the dome unclimbable and was ready to give up. But James Hunter, glory be to him, got it done and quested up the sea of choss, hand-drilling only a small handful of bolts in what has got to be one of the boldest yet also most successful aid climbs in American caving history.

From the ledge with the stoopwalking gypsum passage, another short low-angle rope led up to a ledge, then a longer freehanging rope led up to a bigger ledge. We did those pair of ropes one at a time. Then we went up the last rope, which has one rebelay (one at a time on that two-pitch rope, also go right at the rebelay as there are two ropes). After that, we were off rope and officially above the Kansas Twister, one of the most epic pits/domes in the US. I felt somewhat accomplished for getting up 500+ ft of rope without dying of heat stroke, although our real work in the pit had yet to begin: we still had to replace all the ropes in the pit, which we had yet to start, as we would do that on the way down. But first, it was time to see Oz! There are several miles of amazing passage above the Kansas Twister, and we wanted to see it.

Garrett showed us the way up some small tubes above the Kansas Twister, which on their own seemed not very impressive, seemed like they were about to end. I can’t imagine what James and Derek felt after getting up there, only to at first find some small stoopwalking and crawling tubes that constantly seemed like they were going to end. One of them was a crawl with bedrock on your right and a huge pile of rocks on your left that reminded me of the dig into Don Sawyer Memorial Hall in Fort Stanton Cave. With just a little more breakdown, this passage would have ended, and all that effort climbing the dome would have been for not much. But this breakdown-filled tube, thankfully, was just barely open, and just beyond it lie something incredible.

At the end of the small breakdown-filled crawl, you pop into a huge breakdown room at the bottom of a slope leading up into more passage. The abruptness of the change was especially cool: you immediately change from hands-and-knees crawling to a room so big you can’t see all the walls as you pop through a hole in a wall of the latter.

We followed a trail up the slope and realized that this huge room was just a depression in the floor of an even bigger borehole. This borehole was probably 200–300 ft in diameter, which I’m pretty sure is the largest passage I’ve ever been in. I think it’s bigger than the Triple Perspective Vortex in the Peña Negra section of Cheve. A handline at the top of that slope deposited us right in the center of the borehole, and all I could do was look around and be in awe of how far the walls and ceiling were from us, and how far down the tunnel we could see. You could quite literally fly an airplane in there. This passage is called Munchkinland, fitting with the Wizard of Oz theme, and because it makes you feel like a Munchkin in there.

Garrett told us at the top of the handline that Oz was to the left, and The Wizard, a great looking aid lead I wanted to look at and scope out for climbing, was to our right. Then he left; it was time for him to rejoin Hunter and Karla. Georgia, Michael and I were off to our own devices to wander around Oz, but not for too long as we still had our main task ahead of us: replacing 730 ft of rope on the Kansas Twister.

We all went left towards Oz, stopping at one spot to try to get some pictures that capture the scale (see above). Munchkinland ends and Oz begins when the ceiling drops hundreds of feet nearly to the floor, you have to duck under it for a few steps, and you pop back up into merely good-sized walking passage, about 10 ft high and 30 ft wide. After the ceiling duck is a T intersection; Oz goes both ways. Both ways looked decorated with more beautiful, sparkly gypsum and other formations. A trail of dirt weaved its way through the sparkly white floor.

I went right first. After a few minutes, this passage ended in a pit/dome with ropes going up and down. I started by going down, passed a knot in the rope at a bad rub spot, and got to a spot where the rope ran down smooth, clean flowstone. I didn’t want to dirty it up without knowing what the protocol was for that level of clean formations while on rope, so I went back up. I could have put on shoe covers and gone gloveless, but the rope was muddy, and it would be difficult to impossible to keep things clean while on rope. I didn’t want to create much impact for a side trip that was not our main objective in the cave, especially without knowing the standard accepted practice in that spot, so I went back up.

The rope going up from there went diagonally up and left in a J-hang like feature, although you could kind of walk across a small ledge in the lower part of the J-hang. I started up that, but ran into more clean flowstone that I didn’t want to dirty up, so I went back. Reversing the J-hang while in the middle of it was tricky, not something I had done before. I ended up changing over onto a Munter hitch (I couldn’t use my descender because it was in use to descend the other side of the J hang) then moving my ascenders to the other direction on the rope.

We went back to the T intersection and went left. Michael met us there after attempting some photogrammetry in Munchkinland. I don’t think it worked; the passage was too big. This passage off to the left was a similar-size, moderately-sized borehole, but brilliantly decorated, with aragonite bushes, pools, rimstone dams, velvet flowstone, and dense hanging gardens of stalactites lining the passage. We went extra slow down this passage, taking our time to admire the colors and take pictures.

The others stopped at a webbing handline. I went just a bit further, but knew at that time that it was getting close to the point where we had to turn around and start our main task for the day, replacing ropes. Just past that handline was an impressive display of velvet flowstone covering a wall. Just past the velvet, the trail weaved through a forest of huge aragonite bushes. Instead of a double-flagged trail to walk through, the trail was a series of individual footsteps circled with flagging tape, all precariously between delicate aragonite bushes on a steep slope. Walking through there would be like playing Operation with your feet. I figured that would be a good spot to turn around; there was a very real chance I’d break something trying to walk through there, and such impact wouldn’t be worth it for a non-exploration trip.

So, back to Munchkinland and back to the Kansas Twister we went. Before departing Munchkinland, I made sure to check out The Wizard, an excellent-looking aid lead at the end of Munchkinland on the other side from Oz. I saw a photo of this lead in the Lechuguilla book before the trip and when I heard it was on the way to where we were going, I had to check it out. It does indeed look excellent—a ~40 ft diameter hole in the ceiling with huge globs of flowstone oozing out of it. But it would be a tremendous effort to get up there: ~100 ft of hand-drilling up an overhanging wall. That will not be my first priority when exploration begins again in the cave.

At the top of the Kansas Twister, Michael descended first ahead of us. Georgia and I rerigged as a team. I rigged all the new ropes going down, and Georgia carried the old ropes down. It all went smoothly, if a bit slower than I expected. But rigging always takes longer than you think it will. One bolt was a perma-spinner and wouldn’t tighten down (the bolt shaft spun when I tried to tighten it). It’s the right-hand bolt on the two-bolt anchor on the last ledge before the two pitches that take you to the top of the Twister. It should be replaced on the next trip out there—hopefully by me!

I decided that the walk on the sketchy gravel slope out to the rope below the arch needs a traverse line, and I left enough old rope there, tied to the new rope at the bottom of the pitch below the arch, for a traverse line. I especially appreciated having that extra rope to grab onto while swinging over to the alcove where we wait out of the rockfall zone, as I had tons of old rope clipped to me and dangling from me, and it would be easy for me to misstep and fall there. I would like to do those rigging improvements on the next trip, in addition to fixing up the two ropes in the Western Borehole I didn’t love.

The rope on the arch, which goes from the ground below the arch to the ledge above the arch with the stoopwalking gypsum passage, could be a few feet longer. It reaches the ground, but just barely. You could get a few feet back by using a more compact knot at the top of that rope—I’ll also make sure to do that next time. Right now it’s a bunny ears knot on 2 bolts with super long ears, and a simple bowline and butterfly would use much less rope.

The one rope we couldn’t replace was the one on the low-angle slope, one pitch below the arch pitch. The one James free climbed. The rock that rope was tied to was wedged in a crack and buried under tons of small rocks and gravel, maybe by accident as cavers walked up the gravel slope above it, or maybe on purpose to make it more secure. I started digging it out while waiting for the others to rappel down, but it would have taken me at least an hour of digging, so I figured it was not worth the effort, and that next time someone would just place some bolts to make an easier-to-access anchor. I made sure to rebury the rock after deciding I wouldn’t be able to uncover the whole thing—Chesterton’s Fence, you know?

I rigged every rebelay with butterflies for maximum adjustability, and even rigged the rebelay on the arch with my signature three-butterfly style1. One butterfly for each bolt, and a third to have as an equalized clip-in point, because the bolts were spaced far apart. That kind of rigging—everything butterflies, and a third at Y hangs if it’s useful as an equalized clipping point—is my signature rigging style. Michael was a huge fan, although Georgia was not convinced of its utility over standard knots like bunny ears and fusion knots. I still have some work to do to convert my fellow cavers to The Butterflies Cult.

I stuffed as much old rope into my bag as I could and tied the rest of it to the outside of my bag. That was much better than on the way in, when I had a nearly empty, super floppy camp pack with two cave coils tied to the outside. Putting half the rope inside the bag made the bag full and stiff, which made it easier to tie one coil to the outside. And one coil snagged much less often than two coils.

We got to the bottom of the Kansas Twister at 11:24pm, and we were back at the Leaning Tower at our turn-off from the Western Borehole at midnight. Emerald City and the small passage below it were just as hot and sweaty as before. Climbing 500+ vertical ft of rope got me much less hot than that passage. I climbed fairly fast; maybe that breeze cooled me off!

The other team left a note at the Leaning Tower that they had departed for camp not long before us (their time on their note was something like 11:30pm). The hike back through the Western Borehole was long and scenic. We must have been at least a mile down the borehole, maybe two? We were back at Deep Seas around 2am. It was a 14-hour day out from camp. I drank about 3.75 liters of water out of the 5.5 I packed, and I ate most of my food but still had some from the day left over, and my one-liter pee bottle was well within capacity2. Perfect planning.

At camp, we had a pleasant surprise waiting for us: the other team had a laser light show going on the ceiling above camp, that they were all mesmerized by as they lied down on the Dude Bro Camp Spot below the illuminated ceiling. Garrett brought this compact, lightweight laser light show generator he got on Amazon into the cave, and it was so worth it!

The others at camp were pleasantly surprised to hear that we replaced all the ropes in the Kansas Twister and hauled all the old ones back, with the exception of the bottom rope because we couldn’t access the buried anchor. They weren’t sure if we were going to be able to get it all done in one day. We sure did, while doing lots of sightseeing too!Hunter and Garrett and Karla’s team had some objectives that included going to the Far Western Camp to assess it for reopening (I’m not sure why it was ever closed), going to Oasis, Red Lakes (or maybe it was Red Seas or Red Tides, one of those), and assessing some of the abandoned science stations there. Their trip went well, although I confess to not knowing the exact details. Hunter took some amazing photographs that blew me away. I’ll put one here because it’s astounding, but I won’t share all the good ones because I wasn’t there. Check out Hunter's Facebook post for all his best pictures.

After some time spent passively admiring the laser light show while making dinner, it was time to party. Georgia broke out the speaker she hauled in to camp, and the party was started! We all danced under the lights and enjoyed the scenery and the satisfaction of a job well done. The heat stopped me from dancing a few times, and I just layed down staring at the ceiling before cooling off enough to continue boogying. It was definitely the most fun I’ve ever had at cave camp. It’s nice when your camp is warm and spacious and flat and you can lounge around and hang out and party!

We went to bed around 4am. Totally worth it for the long, fun day and the good times at camp.

Monday, May 26th

Today was our day to leave for the surface. Although Hunter set his alarm for 11am, everyone was up before then, either naturally or from the sound of the others packing. Getting ready to leave and packing up camp took a while, as it was 1:35pm before everyone was gone from camp and moving towards the entrance.

My pack was a fair bit lighter than on the way in. We did have collectively ~150 ft less rope than we carried on the way in, due to leaving the 100 ft new rope at the pitch where I couldn’t access the buried anchor and 50 ft of old rope at the spot in the Kansas Twister that I thought wanted a traverse line. But the old rope was damp and dirty, which made it heavier per foot than the new rope. I also had much less food, although I had two mornings worth of poop to carry out. That always weighs more than you think, often more than the food it was made from, because it’s hydrated, unlike dried camp food. The biggest weight reduction was from me carrying two liters of water out instead of the 6.5 I carried in. That alone saved nine pounds. With the liter I had stashed on the way out in various places and the half liter I had at the entrance, that should be plenty. I crushed the 3.5 liter plastic jug of water to save space and only carried water in my two Nalgenes. My pack was also a little less unwieldy and snaggy, as I only had one rope tied to the outside, and the rest stuffed into the bag. I had something like 250 ft of old rope on the way out, less than the 390 ft of rope I carried in.

We would cave out in two teams: Georgia, Michael, and I would go first, and Hunter, Garrett, and Karla would go behind us. Caving in smaller groups would be nice to prevent traffic jams on rope, although that wasn’t a huge deal here as there was only one really long rope, Boulder Falls. And our team didn’t stay exactly together on the way out; we all separated and caved at our own pace, although I waited for them at the bottom of the Great White Way and at EF Junction.

The way out has more hiking up steep breakdown slopes than any other caving on the trip so far, so I was eagerly awaiting that to see if that heat did me in. While I was certainly plenty hot the whole time, I felt very manageably hot, not like I was overheating. Overall, I was pleasantly surprised with how I managed my temperature this trip. That was what I was most nervous about (really, the only thing I was nervous about) going into the trip, and it was fine. I didn’t even get the forearm cramps that I often do when climbing rope in a hot sweaty cave. I think that’s because I made it a point to eat tons of salt on this trip. The spicy dried mangoes I brought were very salty with Tajin, and I also ate a bunch of straight salt in the form of bouillon cubes, those dried blocks of soup broth that are actually just mostly salt, but flavorful salt that’s edible in large quantities.

I didn’t notice the sudden temperature change above Glacier Bay near the bottom of Boulder Falls, probably because I was exerting myself more and just maximally hot the whole time. When I got to Boulder Falls, I took a few minutes to cool off and drink the half liter of water I had stashed there before the big climb up. I allowed myself a few stops on rope to take it all in and look around the huge chamber I was suspended in the middle of. I got to the top right as I heard Georgia and Michael approach from below: perfect timing.

Although I did not notice a sudden temperature change around Boulder Falls, it was definitely there, because waiting at the top of Boulder Falls was the only time the whole trip where I got cold. I still had no clothes except short shorts then, and I stopped to wait for the others for a while. When Georgia got to the top I announced I was heading out of the cave, as I was actually chilly. It was probably about 60 degrees.

It’s only about 20 minutes to the entrance from there. One of the short ropes between Boulder Falls and the entrance (the one on a low-angle flowstone slope below the Liberty Bell room) is coreshot, tied off with a butterfly, and awkwardly shortened in a way that pulls you off to the side of a slippery flowstone slope. That spot could use some bolts to anchor the rope in a better place, maybe.

At the entrance pit, my chest was chafing pretty bad from my chest harness. Thankfully that was the only time it really affected me—on the last pit of the trip. Although if it hit earlier in the trip, I could have just put on a shirt. I think next time I’ll try climbing rope with a crop top—something to protect my chest from chafing against the chest harness, but with maximum ventilation.

I was on the surface around 5:30pm. I waited for a while for Georgia and Michael, while chugging water and enjoying the sunshine. It was buggy out, which is rare for the Guads. That and the fact that the plants looked extra green and all the cacti were blooming made me think it must have rained while we were in there. It was warm and sunny, but not hot at that time in the late afternoon. A comfortable temperature to lounge around on the surface and sunbathe.

Our team, Georgia and Michael and I, started hiking back right as Karla got to the entrance. We said goodbye, then ditched them, as I had work the next morning and didn’t want to wait much longer to start driving back. Although I did stop to get some Taco Bell with the three of us after the trip. How could you not do a post-caving meal together! Thanks Michael for getting us all TBell.

The drive back to Albuquerque was long, and especially tedious because I didn’t have my phone to play music. My phone mysteriously bricked itself in the cave, which is why I don’t have too many pictures in this post. Maybe the humidity caused water damage? I listened to the radio the entire drive home, which was terrible. I am totally addicted to Spotify.

Anyways, it was a super fun trip in an awesome cave, and I can’t wait to go back. There’s more work and exploration to do in Lechuguilla! Thanks to everyone who was on the trip and made it happen.

We were scheduled to get to Carlsbad Caverns at the leisurely time of 10:30am. So of course we stayed up late the previous night hanging out/partying; what else to do? Everyone helping out with Riannon’s fieldwork who traveled from outside Carlsbad—Jason Ballensky, Maya Robles, and I—was staying at Georgia Schneider’s house, along with Riannon. Georgia is a Carlsbad local and caver who generously hosted us this week. Everyone staying at Georgia’s house went to a bonfire at Ben Tobin’s house on Thursday night where we socialized with more cavers, then stayed up later at her house where we all hung out/listened to music/danced/etc.

The rescue

Instead I woke up at 3am to Jason telling me “there’s a cave rescue happening, wanna help?” I could only say “what?” in that moment as I was disoriented after being awoken so early and so suddenly. I remember thinking then that this must be a dream, and shifting from confusion to realization as I woke up and shed the just-woke-up mindset.

I got out of bed and joined everyone in the common room in Georgia’s house and learned more details about the situation. The story was that a local kid went in Parks Ranch Cave the previous afternoon, alone, and told his parents that he expected to be out of the cave by 11pm. When midnight rolled around and there was still no sign of him, his parents called 911. Georgia, a cave manager with the Carlsbad BLM field office, got the call from her boss around 2am that there was a rescue and that all hands on deck were needed to search the cave for the patient. After some planning with her boss, she started to wake up all the cavers staying with her.

Parks Ranch Cave has four miles of passage and is somewhat maze-like; see the map here. The mazy character of the cave would make it a scary place to get lost, if not for the fact that there are about a dozen entrances (I say “about” because some of them close/open up occasionally as water moves around sediment). So if you wander around in the cave for a while, you are likely to make your way to one of the entrances.

The fact that the patient never found his way to an entrance after many hours was worrying. I figured he was probably hurt or trapped. This unpleasant thought lingered on my mind as Georgia drove us over to the cave entrance. I also kept thinking about my only other cave rescue experience, the 2024 New Year’s Eve Fisher Ridge rescue (which I would like to write a post about one day). There, Paul Walko broke his leg (tibia and fibula) in a very remote place deep into Fisher Ridge Cave, and a large team of cavers including myself helped get him out over 4 painstaking days. I hoped that whatever state the patient was in would not be as bad as Paul’s, but I also hoped that if he was in bad shape then the rescue would go as smoothly as Paul’s did; Paul’s rescue went remarkably well given the challenges involved. When I helped out with Paul’s rescue, everyone already knew where he was; this would be my first time participating in the search part of a search and rescue operation.

We arrived at the cave entrance at around 4:30am. There, I and the other cavers staying with Georgia who came together were greeted by Tom Evans, a Carlsbad caver who was running the rescue. At least, the caving side of the rescue. There was a fire truck and an ambulance there also, doing their own stuff.

Tom split us up into groups to search various parts of the cave. Georgia and I were put on a team sent out to search the eastern part of the cave. Tom gave us a map with our route highlighted and told us to get into the cave at our earliest convenience. He also gave us a gas monitor that detects hydrogen sulfide, as the section of cave we were going in was known to have bad air at times. I don’t even know where one can buy a gas monitor like that, so I was glad that Tom (or someone else?) had one on hand.

I quickly packed a bag that included supplies for a potentially injured, hypothermic, thirsty and/or hungry patient in addition to my personal supplies. Knowing that the patient would likely be hypothermic if he spent all night in a cave not moving due to injury, I packed 2 dry fleeces, and a space blanket/candle/lighter for a Palmer furnace setup. I also packed extra food, and some basic first aid supplies.

Coincidentally, I had just finished my Wilderness First Responder (WFR) course with the National Outdoor Leadership School (NOLS) 2 days earlier. The WFR course aims to teach you basic medical skills that can be useful in the wilderness with the very limited medical supplies you are likely to carry on your back into the backcountry. I took a small amount of solace in the fact that those basic medical skills (things like evaluating for injuries, protecting a potentially injured spine, and improvising splints) were very fresh in my mind; as fresh as they would ever be.

Georgia and I entered the cave at 4:50am. We entered the fenced off entrance on the west side of the cave (see the boxed entrance “Parks’ Ranch Cave #1” on the left/west side of the map). Our route took the SE passage from that entrance, turned left/NE at the next intersection, then right/E at the next intersection. The cave started out as pleasant, mostly stand-up walking passage with some easily avoidable water in the floor. I had never been in Parks Ranch before, but had heard much about it, so it was nice to finally see it, even if it was under stressful circumstances.

We followed our assigned route til there and got to the intersection just west of Chocolate Swirl Falls (we thought). There, we took some compass readings of the passages heading in both directions from that intersection, and found they disagreed with the passage directions drawn on the map. Also, the passage heights on the map (circled numbers) disagreed with the dimensions of the passage we just went through. The passage we just took to that intersection was significantly worse than the pleasant walking passage we encountered at the start of the cave: it was small, tight, and strenuous to move bags through at points. 2 ft wide and 3 ft tall was a common cross section. I had to crawl and push off the walls, which hurt my bad wrist1 a little bit. Pushing on my wrist via a closed fist or fingers pointing outwards was fine, but pushing on my palm hurt depending on the angle and how much force I put on it. So I had to move carefully and deliberately through this passage, varying my wrist/hand position, in order to not irritate my wrist too much.

I started out going through it head first because it was mostly flat, but as the passage slowly tilted downwards until it was steep enough where going down head first was no longer advisable, I stopped to switch to feet first. Because I was in too tight of a spot to turn around, I had to squirm backwards for a bit, feet first going uphill, until I got to a turn around spot. There I continued going down feet first (the correct way to go downwards in tight spaces) until the passage opened up a bit. Georgia and I got a little stressed squeezing around in this unpleasant passage, and were glad when we popped out at the intersection with larger passage that we thought was the intersection just west of Chocolate Swirl Falls.

Because our compass readings of the passage directions didn’t agree with the map directions, we realized that we might not be where we thought we were. So we decided to just search around the various passages we found, sticking to the largest, most obvious passages, as we figured those were where a lost patient would most likely end up. Because we didn’t know exactly where we were at this point, we left notes indicating where we came from, which passage we went in, and which passage we didn’t go in, at every intersection, to make sure we could get back to the previous intersection where we knew how to get back. We certainly wouldn’t want to get lost now, as a search team. That would be awkward. Although, I supposed this might be the least bad time to get lost in Parks Ranch Cave, as there were multiple other search teams in the cave simultaneously.

After a while of searching around branching, mazy passages, Georgia suggested that we should check with the surface to see if the patient had been found yet. We were not given an out time, so it was up to us when to leave the cave and seek updates. I agreed, and once we started seeing notes that another team left marking a way to an exit, we started following them. I was relieved to find those notes, as it meant we wouldn’t have to reverse the unpleasant squeezy passage we took before.

This passage we took to the exit had much more water; I got wet up to my knees. The cave was warm and humid (somewhere in the 60-70 degree range), so I was totally comfortable being wet in the cave, but I knew that I would get cold fast when I got to the surface, where it was around freezing. This wet passage also had some foul sulfur smell in some sections, although this was not the eastern passage that Tom mentioned was known to have bad air sometimes. We got to the surface after about an hour and 45 minutes in the cave, and Georgia was immediately vindicated as we found out when talking to a firefighter that the patient had been located!

The firefighter told us that the patient was located on the surface, where he fell down a 40 ft ravine. He had 2 broken legs, and the others were in the process of hauling him out at the moment. We asked if we could be helpful at all, and he told us that they could use people to carry the litter from the ravine back to the parking lot. He encouraged us to first hydrate and fuel up before heading over to the ravine to help, but we were both doing fine after a not-too-difficult few hours in the cave, and were eager to keep moving given that our wet legs were rapidly chilling. The firefighter pointed in the direction of the ravine and assured us that we would find the rescue party if we walked in that direction, and off we went.

5 minutes of walking brought us to the rescue party, where we saw that the ravine was actually another cave entrance. Apparently what happened was the patient found his way to the one vertical entrance from inside the cave. Instead of heading back the way he came, he attempted to climb up the 40 ft pit entrance, and fell, breaking both of his legs. When we got there, he was packaged in a stretcher and had already been hauled up about halfway up the pit. The Texas SAR team that responded to the call was handling the haul, and didn’t need any more people with that, so Georgia and I, along with the rest of the cavers that were all out of the cave and done with their search missions at that point, went to the other side of the pit to watch the haul.

The patient seemed to be in good spirits. Even as the rope dislodged small chunks of rocks from the chossy wall above the stretcher, which then rained down on the patient (who still had his helmet on), he maintained good spirits and kept thanking everyone profusely. He had been alone in the cave, with 2 broken legs for 10 hours. He fell and broke his legs fairly early in his afternoon caving trip, and because he was alone, he had to wait for 10 excruciating hours after his 11pm callout time before anyone knew something was wrong and thought to call for help. So by that point he presumably just felt thankful to be alive and to be in the hands of competent rescuers.

When the patient/stretcher got to the lip of the pit, the stretcher got stuck on a small roof, which you can see in the above photo. Yarding on the haul ropes with tons of force just got it stuck further, so they had to try another strategy. Tom Evans got on rope, lowered himself to be right above the patient, and yarded the stretcher out to get clear of the roof while the haulers pulled it over the lip. This maneuver jostled the patient around a bit, and he groaned in pain, although it was over fast and he was then on flat ground, ready to be carried back to the parking lot.

Now that the patient was on flat ground, it was easy mode back to civilization. Teams of 8 people, 4 on each side, carried him for 5-10 minutes back to the parking lot, where an ambulance and helicopter was waiting for him. I took a short shift carrying the litter for a few minutes, which was my only direct interaction with the patient. The entire time he was being carried back, he maintained good spirits, joking and conversing with all the rescuers. I didn’t have anything to say as I was still blown away by witnessing someone in dire circumstances having his life saved, but thankfully the other rescuers kept chatting with him, keeping his spirits up.

Once we got the stretcher back to the parking lot, we did one last painful maneuver: transferring him from the Ferno stretcher to the ambulance stretcher. There, the paramedics took over and took him into the ambulance to check his vitals and ensure he was stable. While they were doing that, we hung around and chatted with the firefighters. They brought some hot food (McDonald’s breakfast sandwiches!) and coffee, which we all indulged in. The firefighters gave us a tour of the inside of their fire truck, and we hung out there while waiting for the patient to depart via helicopter, as it was heated. It was cool to see the inside of the fire truck; I had never done that before.

Eventually, we saw the paramedics wheel the patient out on a stretcher into a helicopter, and we braced to watch it take off. We were glad to be inside the truck when it did, as the powerful rotors kicked up a ton of dust. A helicopter taking off about 30 ft from us was quite a sight to see and quite a sound to hear.

After that, everyone gathered outside for a quick debrief. Tom Evans led the debrief, where he and others summarized exactly what happened so everyone was on the same page, and opened up time for questions/discussions. All the different responding agencies (Texas SAR, Eddy County Sheriff’s department, the BLM, the fire department and emergency medicine people, and volunteer cavers such as myself [and probably other agencies; it was early in the morning and my memory wasn’t 100% functioning]) thanked each other for helping out. There wasn’t too much to discuss, as everything went more-or-less perfectly. I suppose the search teams, such as Georgia and I’s team, could have had clearer communication about how long to stay in the cave before coming out to check with the surface, although we were fine making a judgement call about that on our own. Tom said, and we all agreed, that while caving alone is never a great choice, the patient did one thing very right: he had a callout. His parents knew where he was and when he was supposed to be back, so they were able to initiate the search and rescue process when the time came.

After thanking everyone from all the other agencies one last time, we changed out of our wet, cold cave clothes and drove back to Carlsbad. It was about 8:30am at that point. After the 45 minute drive back to Georgia’s house, we only had about an hour and a half before we all had to leave to be at Carlsbad Caverns (Riannon, who was in charge of her fieldwork, had to be there even earlier). Some people tried to get in a quick nap, but I didn’t bother—I just packed my bags and went with Riannon on the earlier side to help her out with organizing packing her scientific supplies for the day.

Edit: the Ruidoso News now has an article about the accident.

The fieldwork

The first stop was the resources office at Carlsbad Caverns National Park, where Riannon had some of her scientific supplies stashed. I drove there with her and we started unpacking and repacking. After my tasks helping her were done, I took a quick nap right there in the parking lot. It was nice to recharge a little bit after our long morning. It was late morning then, and the sun was strong and the asphalt was pleasantly warm.

We were all ready around noon, just a little behind schedule. I volunteered to carry the 55 lb car battery that powered her airflow monitoring stations for a whole month, as I am always down to carry heavy things. Riannon gave us all group gear to carry (most of it lighter than the car battery) and high-vis volunteer vests, and we went off into the cave.

We had 3 trips into the cave planned for that day. First, we would enter the cave via the elevator that descends 700 ft beneath the surface of the earth, and head into a passage called the Left-Hand Tunnel. The Left-Hand Tunnel is a nice walking borehole that used to be open to tours but has been closed for a while. It is largely undeveloped, with some trail markings, but no lighting, except for some spots visible from the Big Room above. Riannon had one air monitoring station there; she would download its data and swap the battery. Each air monitoring station she made monitoried radon and CO2 concentrations.

While Riannnon downloaded the data and swapped the battery, the rest of us were free to do what we want—take pictures, wander around the Left-Hand Tunnel (so long as we stayed on the marked trail, and were guided around by Gabe, who used to do tours in this area), or watch Riannon do her fieldwork.

I joined Gabe and others for a quick impromptu tour of the Left-Hand tunnel tour. Although the size of the Left-Hand Tunnel pales in comparison to that of the main entrance passage, it is still large, pleasant borehole by most any cave’s standards, and I enjoyed the passage. Even more remarkable was the flat, even dirt floor of the passage. Even nice borehole in a cave will often have an uneven breakdown floor that can make even the largest tunnels annoying to travel across. Only the occasional low ceiling that forced me to duck or put a knee down reminded me that we were in an actual wild cave passage.

Eventually the Left-Hand Tunnel took us to the lower level of the Big Room, which you can see from way above from some spots on the Big Room trail. Riannon had an air monitoring station here too. While she worked I wandered around the marked trail below the Big Room, looking up and looking forward to being up there later in the day.

After Riannon was done, we took a quick walk around the lower level of the Big Room before heading out. Here the passage opened up from merely-large walking passage to a massive chamber. We also started seeing massive cave formations (stalactites, stalagmites, etc) scattered across the floor, much bigger than the smaller ones hugging the walls of the previous section of the Left-Hand Tunnel. It was also lit up in a few spots that tourists in the Big Room could see from above. A few tourists above saw us and waved to us. It was awesome to see what was essentially another Big Room, but with no paved trails (just flagged trails in the dirt and rocky surfaces) and no other people around.

Once Riannon was done with her air monitoring stations in the Left-Hand Tunnel and the lower part of the Big Room, we walked back to the developed part of the cave and took the elevator to the surface. That was one out of three trips into the cave we were scheduled to do that day. We hung out on the surface for a bit, eating food and enjoying the warmth and sunshine. I heated up a can of chili as my lunch, which is a classic camping meal for me.

The second trip into the cave today would take us into the natural entrance of the cave, down the main passage, and to the elevators, which we would take back up to the surface. Riannon had an air-monitoring station near the trail on the main passage. I had never been to Carlsbad Caverns before, despite having caved several times in the Carlsbad area, so I was psyched at this opportunity to see so much of the cave, including that not open to the public, for my first trip into the cave.

After our leisurely lunch break, we gathered in the parking lot once again and walked over to the natural entrance. It was late afternoon then, after the natural entrance had closed to visitors, so the ranger there checked our permit to make sure we had permission to be in the cave after hours. We started down the paved switchbacks that descended into the steep, massive entrance passage. In the entrance area, swallows flew all around us, presumably startled by the later-than-usual-arriving humans walking underneath their nests. The massive entrance and the swallows flying in formation around us made me feel like I was about to begin a journey in some Lord-of-the-Rings-esque fantasy adventure. It was otherworldly.

Riannon’s station to check on this trip was not far off the paved trail. I enjoyed the massive, decorated borehole that was the main passage as I crisscrossed down the many switchbacks. Because no one was around, we played music and sang loudly during parts of our descent. The cave had excellent acoustics. I wondered whether, and how, early explorers made it down this far without paved switchbacks making the steep descent casual. The passage was consistently 100 ft in diameter, often more, and descended steeply into the Earth. While I have experienced passage this large (e.g., the Total Perspective Vortex in the Pena Negra section of Sistema Cheve is 100 ft in diameter in places), none of it was consistently this large for such a long distance. Carlsbad Caverns truly is one of the great wonders of New Mexico, and the USA/the world as a whole. I highly recommend stopping by and visiting for anyone who finds themself in New Mexico or the El Paso area.

We made it to the elevators and took them back up to the surface. We hung out on the surface for one more short period of time before heading back down for our last, and longest and most interesting trip into the cave. In this last trip we would go up to Spirit World, a passage that branches off from the ceiling of the Big Room. To get to Spirit World, you climb a rope hanging from the ceiling of the Big Room. From certain spots on the Big Room trail, you can see the rope hanging down in the center of the room, going 220 ft up into a gaping hole in the ceiling.

The story of how they first got the rope up there is one for the ages. The entrance to Spirit World is a huge hole in the ceiling that one can see super obviously from the floor. Cavers had eyed that super visible but totally inaccessible large hole in the ceiling for years, and dreamed about ways to get there. The strategy that ended up working was something straight out of a cartoon: they connected many helium balloons to a large wooden frame, let the balloons pull the wooden frame up to the ceiling, used some ropes attached to the frame to guide it sideways into the hole, and pulled it down til it hooked a stalagmite in the passage in the hole in the ceiling. Without seeing or knowing what the frame was hooked onto, someone climbed the rope attached to the helium balloon laded frame up 220 ft into the ceiling. That hole had some passage that kept going. After re-anchoring the rope to something more solid, more cavers went up to explore and survey that passage that they later named Spirit World. The NPS has a great interview about it with Jim Goodbar, who participated in the climb in 1985. There is about a mile of passage up there, and there aren’t good leads left. But Riannon has an airflow monitoring station there, which means that we all got to go up there.

After taking the elevators down into the cave one last time, we walked the section of the main passage from the elevators to the Big Room, which is just as massive and decorated as the passage from the natural entrance to the elevator. Riannon had an airflow monitoring station in the Big Room off the trail, and we unpacked our bags and set up a hang out spot there. After Riannon did her thing at that air monitoring station, it would be a while before we got everyone up the rope to Spirit World, as the 222 ft ascent took some people a while.

Riannon went up the rope first, as she had work to do at the top. She had to carry the 55 lb car battery up there, so her climb was an especially good workout. While she and everyone else climbed, I walked the trail around the Big Room. The Big Room’s footprint is 4000 ft long, 625 ft wide, and 255 ft tall at its tallest point, near where the Spirit World rope goes up into the ceiling. The trail around the perimeter of the room is 1.25 miles long and weaves its way through more massive formations and around big pits in the floor that drop out of sight. In a few spots I saw down the pits to lit up spaces in the lower level where we wandered around earlier in the day. It was pretty cool to come full circle and see those spaces from above, and to be in the spots where we saw tourists waving down at us earlier. I also walked the longer route back to the elevators and back, making the best use of my free time in the cave while others climbed.

Georgia brought a speaker down into the cave and blasted music while people climbed. She knew that we both loved Bruno Mars (❤️), so she made sure to play some of his hits while I climbed. Bruno’s angelic voice reverberating off the cave walls motivated me to climb extra fast—thanks Georgia! I timed myself once I started ascending; I took 5 minutes 36 seconds to get to the rebelay at ~180 ft, and 8 minutes 13 seconds to get off rope at the top, which included passing the rebelay and putting on shoe covers while on rope. My rope-climbing fitness is not too bad right now! This is my first time caving since Tears of the Turtle in August, and I am still recovering from my broken arm/wrist from December and have been less physically active lately, so I felt pretty good about climbing 222 ft of rope in a good time.

When I got to the top of Spirit World, Riannon was already done with her fieldwork, which was the last we needed to do that day. So it was time to see Spirit World then head out of the cave. We were only allowed to go in the main Spirit World passage; we couldn’t go down any of the side passages nor climb any of the ropes hanging mysteriously from higher holes in the walls. Still, we saw some cool fossils and bat skeletons in that nice walking passage.

After we all saw a small bit of Spirit World, we started rappelling back down to the big room. This went much faster as no one took more than a few minutes to get back down. On the rappel, I was able to appreciate the scale of the Big Room much more so than when I was climbing or on the ground in the Big Room. When climbing I was trying pretty hard and focused only on climbing rope as fast as I could. On the way down I had plenty of bandwidth to look around and take in the grandeur of the position at the top of the Big Room at Spirit World. Seeing the room illuminated hundreds of feet away while dangling in space was epic. The massive stalagmites and stalactites, whose scale inspired wonder when standing right beneath them, seemed like minor textural details on the walls from way up high. I couldn’t make out the details of the people beneath us; they were just point sources of light (coming from their headlamps) as far as I could see.

With all of Riannon’s fieldwork done, we packed up and walked back to the elevators while savoring our last 10 minutes in the cave walking through huge decorated borehole. Everyone was tired from our long eventful day, but happy that we got all our work done without incident, especially after the rescue this morning. We got out of the cave a little before midnight, and I was in bed before 1am. We all got some much needed sleep that night.

Thanks to Riannon for inviting me on this fun trip! This was the best possible first time in Carlsbad Caverns. And thanks for everyone else who helped out with the fieldwork and with the rescue that morning, and made everything go so smoothly.

Tears of the Turtle Cave, or just "Tears" as we call it (this should give you a sense of what the cave is like), is remote in many ways. The entrance is a 22 mile hike from the nearest road, deep in the Bob Marshall Wilderness area (“the Bob”) in northwestern Montana. And that road is 3 hours of dirt road driving into the mountains of Montana from the nearest town. Once you get to the entrance, the real remoteness begins: the bottom of the cave, the frontier at which exploration is ongoing, is so many hours of hard, grueling travel to get to from the entrance that we have established a camp partway down so cavers can sleep near the bottom. This lets us live in the cave for days at a time while we conduct the hard yet glorious work of exploration. And this year, for the first time ever in a US cave, we established a camp 2 beyond camp 1, close to the current bottom of the cave, as the frontier of exploration had been pushed to many hours of hard travel beyond camp 1.

This expedition, the first return to the cave in 2 years since the summer 2022 expedition that I was on, was chaotic, but wildly successful. Every day it seemed like something major went wrong that changed the plans for the whole rest of the trip, yet somehow our sequence of last-minute changed plans ended up working well and enabling some exciting and successful exploration. Read on to hear about all that went wrong and all that went so, so right!

Friday, July 19

I woke up at 5:30am in Thorne Bay, a small town on Prince of Wales Island, a large temperate rainforest island off the coast of southern Alaska. I quickly hopped in the car, and my friend Christian, whose house I was staying at that week, drove me to a ferry terminal in Hollis, Alaska. Hollis is only 15 miles away, but a rugged mountain range separating Thorne Bay and Hollis forces the route to take a circuitous, hour-and-15-minute course, crossing between the east and west sides of the island twice. At Hollis, I boarded a ferry that took me on a leisurely 3 hour scenic ride to Ketchikan, a small city on Revillagigedo Island, just to the east of Prince of Wales Island. From there I took another short ferry to the airport. From the Ketchikan airport, I flew to Seattle then to Kalispell, Montana, landing after just over 12 hours of travel. At the Kalispell airport I was greeted by Hans Bodenhamer, a local caver and veteran Bob Marshall Wilderness caver, and Derek Von Nieda, a young east coast caver. We went back to Hans' house to spend the night before the expedition officially began the next day. When we got to Hans' house, Tommy Cleckner, a Denver-based caver that I had caved with in Tears back in 2022 and was looking forward to caving with again, had just arrived. We were supposed to have 2 more people already at that time, which brings me to the first incident that threw a wrench in our plans for the expedition: the CrowdStrike outage.

That day, CrowdStrike, a cybersecurity company, sent out a bad software update to some of its security software that crashed Windows computers all over the world. All sorts of industries that used computers, such as hospitals, banks, and internet services, went down, but airlines were affected particularly severely. United, Delta, and American Airlines cancelled many flights that Friday and that weekend. Thankfully, my Alaska Airlines flight was unaffected. But not everyone on the expedition was so lucky. Jason Ballensky, the expedition leader whose support and planning we needed to make everything happen, had his flight get cancelled. We would be pretty screwed without his leadership, but thankfully he was able to find another flight on a different airline from San Diego the next day, the day we were scheduled to start hiking into the Bob. Eric Pekley had his flight cancelled and was unable to find another flight that would get him here in time, so he had to bail on the whole trip. Losing Eric was a real bummer—I caved with him in the 2022 expedition, and was looking forward to caving with him again, as he was scheduled to be on the first exploration trip with me pushing the bottom of the cave, along with Tommy Cleckner and Dustin Kisner. Without Eric, we would have to swap in someone else to be the fourth person on that first exploration trip.

Saturday, July 20: convene in Kalispell, start hiking in to basecamp

Everyone else on the trip arrived without incident that morning, as no one except Jason, Eric, and myself were flying in. I spent the morning shopping for 2 weeks worth of food and organizing my bags, as all my caving and camping gear was scattered in my checked and carry-on luggage. I already had the main portion of all my dinner and breakfast meals, Dinner Mix, packed in drums and ready to go, but I needed to get 2 weeks worth of day food and various add-ins for the dinner mix. Dinner mix is a dehydrated, ultralight food mix that I make for backpacking/cave camping trips where I need to carry lots of food a long ways. My recipe, which is a modified version of the dinner mix they make at the Cheve cave expeditions in Mexico, is:

• Instant mashed potatoes

• Dehydrated cooked beans

• Chia seeds

• Nuts (currently cashews, walnuts, and pepitas)

• Textured vegetable protein

To turn it into a meal, you add hot water to it, and all the dried ingredients rehydrate and greatly expand in volume. The nuts add a nice crunch, so it’s not entirely wet slop. I also add a generous quantity of butter for taste and nutrition, and typically some other flavorful add-ins like sun dried tomatoes, bouillon power, Parmesan cheese powder, or various spice mixes.

I ate Dinner Mix for every breakfast and dinner meal over the 2 weeks, with the exception of 3 meals where I ate oatmeal. All the remaining food for during the days, when I was away from camp, I had to purchase in Columbia Falls (the town where Hans lives, just outside Kalispell). Highlights from my grocery shopping for 2 weeks of expedition food include:

• 5 pounds of cheese

• 2 pounds of butter

• 1.25 pounds of cooked bacon bits (I actually brought these from Albuquerque, where I bought that at Costco)

• About 20 candy bars of various types (Snickers, Hershey’s, etc)

I bought $120 worth of food. Not bad for 2 weeks of day food during a time when I was going to be eating a LOT. I spent the rest of the morning packing my personal backpack that I would carry into surface basecamp, and other bags full of my food and caving gear that we would put on the mules that we were paying to haul most of our gear into the Bob.

I had all my bags packed by late morning, and everyone had arrived except our Jason, whose flight was delayed til 2:15pm that afternoon. We wanted to leave town by 2pm in order to get to the trailhead by 6pm, when we were supposed to meet the mule packers. We decided to leave for the trailhead without our dear leader Jason, except for one person who would stay behind to pick him up from the airport. I rode with a driver Jason hired to haul all our gear from Columbia Falls to the trailhead. We thought we would need an extra car to get all of our stuff to the trailhead, but after we lost Eric, we had enough car space for all of us and our gear. But the driver was already here, so we put some gear in his truck and I accompanied him to the trailhead.

It’s about 3 hours from Columbia Falls to the Meadow Creek Trailhead, where we were to meet the packers who would haul in most of our gear using mules. The drive is quite scenic. After passing the Hungry Horse Dam, you spend the next 2 hours right alongside the crystal clear turquoise water of the Hungry Horse Reservoir. After the reservoir ends, a dirt road takes you another hour upstream along the South Fork Flathead River until you get to the trailhead and the adjacent packer camp.

We had way too much stuff to carry 2 weeks worth of food; personal caving gear including heavy cave suits, metal vertical gear, and many warm layers, and group caving gear including survey equipment, cave camping supplies, and hundreds of feet of rope. So mules or some other form of pack animal were a practical necessity. I got there first and greeted the packers, who were just getting back from with some packrafters they hauled for. I unloaded all the gear from the driver’s truck, got 2 piles going—one for stuff we were getting packed in, and one for personal backpacking stuff we would carry in for the 2 day approach to basecamp.

The others arrived shortly after me, including Jason who finally made it after his cancelled original flight. After giving the packers all of our gear to be packed in, and packing the rest in our backpacks, we set off for basecamp, a measly 21 miles away. The plan was to hike in about a third of the way tonight and the rest of the way tomorrow, Sunday. On Monday, the mules would bring our gear to Pagoda Pass, 20 miles from the trailhead at the end of the portion of the approach hike that follows established trail. From there, we would shuttle gear in multiple loads the last mile to basecamp. This last mile is off-trail, on terrain that the mules can’t handle.

We started hiking at 7:15pm. This was later than our planned start time of 6pm, although none of us minded, due to the second major complication of the expedition after the CrowdStrike fiasco: record high heat. It hit 98 in Kalispell that day and stayed in the high 90s the next few days when we would be backpacking and shuttling gear. The trailhead was only 1000 ft higher so not much cooler. Sunset was at 9:30pm, so it had hardly cooled down at all when we started hiking at 7:15pm.

The first half of the approach follows the South Fork Flathead River upstream on a flat, dry, dusty trail. Most of the trees on this trail have burned recently, so there was no shade for the first few hours of our hiking that night while we roasted in the strong, high elevation sun. We power walked along the trail, our bags only moderately heavy, as we only needed to carry 3 days worth of food until our planned rendezvous with the mules at Pagoda Pass on Monday. We were all relieved when the sun dipped below the mountain tops across the Flathead River and we could hike without being scorched.

At 10:30pm, after an hour of blissful post-sunset hiking, we arrived at a good spot to camp just past the Black Bear Creek trail. This was 8 miles from the trailhead. We stopped here for the night and set up camp, which was pretty minimalist for most people. I didn’t even bring a tent on this expedition, just my hammock rain fly that I would use as a lightweight ceiling-only tarp shelter, which I didn’t even set up that night cause it wasn’t gonna rain. I like to cowboy camp with no tent/shelter when the weather allows. It certainly didn’t allow in the temperate rainforest that is Prince of Wales Island, Alaska. So I took this opportunity to fall asleep under the stars.

The campsite was immediately adjacent to the river, which was nice. We were all treated to a faint sound of lightly running water as white noise while falling asleep. I took a quick skinny dip in the river before heading to bed, as I had a lot of sweat to wash off and I hadn’t showered since before the previous caving day before leaving Thorne Bay. The river water was surprisingly not that cold; it merely cooled me down comfortably after a short but still scorching evening in the heat.

Sunday, July 21: onwards to basecamp

We woke up with the sun and made a point to pack up camp quickly so we could get started hiking before the heat became too intense. Right as we left camp, we ran into Hans and his group of high schoolers. Hans lives nearby in Columbia Falls and I stayed in his house Friday night. He teaches at the local high school, and every summer he brings some students in the school’s caving club out into the Bob Marshall Wilderness for a week long caving and backpacking trip. This year they were tagging along with us. They would check out some of the easier caves near Tears of the Turtle, and maybe some of them would check out the top of Tears if they felt like it. Anyways, they had started their hike from the at 3am that day to beat the heat. We joined them then overtook them, continuing upstream along the Flathead River,

After 4 more miles along the Flathead, it was time to turn to the east up the Helen Creek trail. Here the trail leaves the gentle flats of the river and begins a gradual but long climb up the Helen Creek drainage. The bottom of the Helen Creek drainage is wet and vegetated, with plenty of tree cover for shade. The greenery and decent, if intermittent, shade provided a nice contrast from the dry, dusty trail on the banks above the Flathead River, where the vegetation was thoroughly burned and there were no trees for shade.

The trail crisscrossed the creek as we climbed up the drainage, and we enjoyed frequent water stops at these crossings. As we made it up higher in the drainage, we entered more recently burned terrain and we lost our precious tree cover. By now we were in the midst of the midday heat and we were roasting again. Now when we encountered water, I made sure to scoop up a few liters and dump them all over me to cool off. The water in these small mountain streams was quite cold, unlike the lukewarm Flathead River, so these little bursts of water I would dump on myself were quite effective in cooling me off and keeping the high-90’s heat and intense high-elevation sun at bay.

We passed the last water source before basecamp at 12:30pm, when the sun was blazing as strong as it would that day. I filled up several liters of cold mountain stream water, dumped several more liters all over myself, and powered up the steepest section of trail ahead to Pagoda Pass. At this point the heat was taking a lot out of me, and I was ready to be done exerting myself for the day. The intense sunny dry heat felt kind of nice the day before, mainly as a contrast against the rainy, cold, soggy rainforest that I just spent 2 weeks in in Alaska. But by today I was already over it. The heat was draining me and I was ready to be out of it.

After last water, Jason and Tommy and I pulled ahead of the rest of the group and gunned it to basecamp. We hit Pagoda pass one very hot hour and a half later, at 2pm. There we left the established trail, quested up to a saddle separating our current drainage from another, then descended down the trail-less drainage on the other side. Camp was in this drainage. The final descent from the saddle towards basecamp takes you through a pleasant, gently sloping, grassy meadow teeming with wildflowers. For a moment we forgot about all the heat, and all the effort it took to get here, as we were finally going downhill through a friendly, idyllic meadow with grand alpine scenery all around us. Sublime.

When we arrived at basecamp at 2:45pm, Alex Lambie was already there. Earlier that day he hiked over from Silvertip, another caving area to the north where a parallel expedition was taking place, Him, and later Dustin Kisner and Matteo Bucalossi, joined us at Tears of the Turtle in the middle of their Silvertip trip. Lambie took the standard route from Silvertip to Tears, while Dustin and Matteo opted for a more roundabout, scenic route, so Lambie got there first. I had never met Lambie before, so we chatted a bit before I began setting up my camp spot. I had heard Lambie described as extremely fun-loving and lighthearted and silly while also highly competent, and my initial impressions matched that description. I could tell we would get along well.

As the rest of the crew trickled in, I set up my humble camp spot. After looking around a bit, I settled on the same spot I used 2 years ago. Then I had my hammock. This year I went even lighter and only brought my hammock kit’s rain fly as a tarp shelter in case of rain. Once I had my rain fly rigged where I liked it, I terraformed a flat spot beneath it, blew up my sleeping pad, and laid out my ground sheet and quilt. With my quick and easy camp setup done, I was free to spend the rest of the afternoon lounging around and resting up before the intense multi-day bout of effort that was about to ensue.

That afternoon, Sam Marks broke out one of her signature surprise items. She likes to bring fun novelty items on caving trips—you may recall that on a Silent River Cave trip in the Grand Canyon earlier this year, she brought fancy ties for everyone to don and some small wine boxes and threw everyone a little fancy party. This time she brought kites! Several colorful butterfly-shaped ones. We amused ourselves that afternoon (and on several other days) flying kites in the gentle breeze.

Monday, July 22: meeting the mule packers at Pagoda Pass, allegedly…

Today was our day to meet the mule packers at Pagoda Pass to get the rest of our supplies and shuttle them in several loads across the final off-trail stretch to basecamp. The agreed-upon meeting time was noon, although we planned on getting to the pass early in case the packers made good time. After a leisurely morning, including a big breakfast to fuel our upcoming portering task, we made it to the pass at 11am. No one was there yet, so we gladly took some time to lounge around in the shade while we waited for the packers to show up. It was hot and the sun was beating down on us again, so even the easy (but uphill) off-trail mile from basecamp to Pagoda Pass with no loads took something out of me. It was going to be a lot harder doing this fully loaded with supplies, multiple times, in the more intense afternoon heat, so we savored some more nap time while we waited for the mules.

Noon passed with no sign of the mules. We waited for several more hours, not moving much from our nap spots except to shift around following the shade as the sun moved around in the sky. Eventually we started losing people at the pass, as some of us started heading back to camp for lack of water or food. I finally cracked and left for camp when I started getting hungry around 3:30pm. I didn’t bring any food, and little water, to the pass, as I expected to quickly grab a load of gear from the mules and take it back to camp.

Irina and I took a scenic route back to camp then. Instead of going off trail right at Pagoda Pass, we followed the Helen Creek trail down the Pagoda Creek drainage a ways before breaking off trail and traversing much lower towards camp. This was the old route they took to basecamp when they first started caving at Turtlehead Mountain. This route had no elevation gain, but it was longer (took an hour) and had more bushwhacking, so the other route is clearly better. But it was nice to take a different route since we had nothing but time.

Eventually everyone else including Jason made it back to basecamp after a few more hours waiting at the pass. Right now we were all in wait-and-see-what-happens mode, in no particular rush to do anything other than wait for the mules. And dissipate heat by not doing anything—the heat was definitely making some of us, including me, kind of lazy. But we knew this mode couldn’t last long. We had already lost one day of caving, as one team was supposed to enter the cave this afternoon to set up camp 1 (“Mud Camp”) in the cave right after getting their gear off the mules. But that was not gonna happen, as no one had any caving gear on them. We had all spent lots of resources (time off, money for plane tickets, etc) to get here, and we didn’t want to waste any of the precious 2 weeks we had here.

There was another reason this mode couldn’t last long. We didn’t have much food. The plan was to only carry 3 days of food to basecamp, and have the rest of our food for the 2 weeks delivered by the mules. Most of us took a little more food than this. But not a lot more. Most people could only last an extra day or so. I, for one, brought tons of extra food—I brought a 6 liter Darren Drum of dinner mix that had 20+ meals worth of food in it, for no other reason than it was already packed in the drum when I flew to Kalispell and I didn’t want to repack it into smaller containers to save food weight. I also brought my 5 pounds of cheese and 2 pounds of butter, so I could keep them fresher by putting them in cold mountain streams at night, instead of having them sit around on the mules in the heat. So I had plenty of extra food to share with others, just incidentally, not because I was planning for any unforeseen circumstances like the current ones. That night I shared some of my dinner mix and cheese with those who were running low on food.

At 6:30pm, Jason announced that it was time for him to leave basecamp and go investigate what was going on with the mule packers. The trailhead was 21 miles away, so he might be gone for quite some time. He took one of the inReaches (satellite messaging devices) and left one with us so he could communicate whatever he found. He packed up his tent and camping stuff, as he could be gone for more than the evening if the issue was all the way at the trailhead, 21 miles away. Jason insisted that no one else go with him, as that would just mean more people hiking unnecessary miles and wasting energy. So we all stayed at camp that night, lounging around and frequently checking the inReach for updates.

At 9pm, while we all surrounded the campfire, the inReach lit up. Pete opened it up and read it aloud as everyone crowded around him with nervous anticipation. The message read…wait for it…

Please confirm that you heard the plan.

Okay, anticlimactic. inReach messages have long delays and can sometimes come out of order. So we knew that this meant there was another message incoming with a status update that would tell us some sort of plan. We returned to lounging around by the fire, even more full of nervous anticipation than before. Several minutes later, the first message came through. He said that there was a “terrible wreck” halfway down the Helen Creek trail, and requested that we arrive the following day bright and early at 6am to fix it. It would take all day, but no longer than a full day, so he instructed us to not bring camp gear. This terrible wreck was the third major complication of the expedition, and the most serious so far.

We estimated that halfway down the Helen Creek trail was about 2 hours from basecamp, at least when going downhill with empty packs and when we had the benefit of the cool morning temps. So we planned on leaving basecamp at 4am the next day. We confirmed via inReach that Hans and his high schoolers, who were camping at Pagoda Pass rather than at the main basecamp, had received the plan and were in. We smothered the fire and quickly prepped for bed, as we had a long day ahead of us. We were all a little nervous, as we had no idea what “terrible wreck” meant. Did this mean some of the people with the mules were hurt? Or just the mules? Was the expedition over, or was whatever situation we would meet tomorrow salvageable? We would find out tomorrow.

I set my alarm for 3:40am. I had my bag all packed with water, clothes, and a little bit of food, so I could get going immediately after waking up. I was in bed by 10:30pm, although I had trouble falling asleep, as I spent the night wondering what bizarre and terrible circumstances I would meet the next day.

Tuesday, July 23rd: the terrible wreck

“Alex, wake up” I heard from my sleeping bag. I looked out and saw Mark Wingard staring me down, and I looked at my watch and read 4:09am. Oops, I slept through my alarm. My watch has a fairly quiet alarm, and this has happened to me before. I should get a watch with a louder alarm.

I flew out of my sleeping bag, quickly threw on some clothes, shoes, a headlamp, and my backpack, and started hiking. I was hiking by 4:16am, just 7 minutes after I woke up—that’s definitely a record for me. We quickly warmed ourselves up charging uphill off-trail to Pagoda Pass, then cooled off as we speed walked downhill down the gentler Helen Creek trail. The stars disappeared and the sun rose as we were on the trail. I got to Jason and the packers at 6:20am, just a bit late. I met them where they camped for the night, at a small flat area on the trail about a quarter mile before the site of the wreck.

They told us what’s up. On a section of trail that gets very narrow (1-2 ft wide) and is cut into a steep hillside, the mule at the front of a 5-mule train misstepped and stepped onto the steep hillside just off the trail. It then fell down the steep hillside the trail was cut into and tumbled a few hundred feet downhill into Helen Creek at the bottom of the valley below. All 5 mules were tied together and tumbled down the hillside at the same time. 1 mule died when it hit the bottom; another was alive but badly hurt and clearly dying, so the packers shot it; the remaining 3 survived the tumble down the hillside and ended up standing up in Helen Creek below. They were trapped in the creek by the steep, densely vegetated hillside on either side, and they spent the night down there. Thankfully the packers were off of the mules then, walking in front of and behind them, as they knew this section of trail was a bit sketchy. So none of the packers were hurt.

A bunch of our gear was strapped to these mules. Most of it was flung off the mules while they tumbled down the hillside and was now scattered in the vegetation or sitting in the creek, flooded with water. Some of it remained attached to the mules. The 2 dead mules were stacked on top of each other, as they both fell on the same spot. Some of our gear was still strapped to those carcasses, pinned under the weight of 2 fully grown mules.

When Jason met the packers there last night, the packers thought that the wreck would take at least 3 days to clean up. But Jason had a small army he could recruit, and he figured that with all 17 of us from his team and from Hans’ team of high schoolers, we should be able to get the situation resolved within a day, hence his instructions to only come with supplies for a day and not bring camp gear. Here, getting the situation resolved means (1) gathering all our gear that was scattered throughout the hillside and in Helen Creek and pinned underneath mule carcasses, and getting it back up the hillside to the trail; and (2) cutting a trail up the hillside for the 3 remaining live mules to use to get back up to the trail from the creek. This was going to take quite a bit of work, but with 17 strong young men and women with nothing else to do, we would get it done. So we went to work. I ate a small breakfast of cheese and pepperoni that I packed from camp before heading down the hillside to the site of the wreck.

The first task was to gather up all of the gear that was scattered around the hillside and in the creek (but not pinned underneath the mules), and haul it back up to the trail. From the mule carcasses at the bottom of the hill, we sent people upstream and downstream to search for bags. Once we had it all gathered in one spot, we set up an assembly line of people to pass the bags up the hill, one after the other. With all 17 of us plus the mule packers, we could move bags about 60-100 ft uphill with one assembly line instance. The first section we moved bags over was a steeply sloping fallen log that we used to get down the hillside. We stationed a bunch of people on the log and right below it in the creek, and passed bags up til they were all in a big pile besides the log.

It took 3 rounds of the assembly line to move all the gear up the hill to the trail. Once we finished that task, we split into 2 teams for the remaining 2 tasks: retrieving the gear that was pinned underneath the mule carcasses, and cutting a trail for the remaining mules to get back up the hill to the trail.

I joined the team working on getting the remaining gear. The mule carcasses, which we had all seen at that point in the creek and which were in quite bad shape (deformed, bloody), didn’t really bother me, so I hopped in on the team that would require burrowing underneath the carcasses.

The situation at the mule carcasses was pretty ugly. The 2 mules were stacked on top of each other. There was also a huge log on top of both of them, weighing them down. They probably dislodged a fallen log as they tumbled down the hill, and it up on top of both of them. Some bags were pinned underneath the bottom mule, with the weight of 2 mules and a log on top of them, some were pinned in between the 2 mules, and some were pinned in between the top mule and the log. Much of our gear was covered in mule blood that had been leaking out of the carcasses and soaking our stuff all night.

Fair warning: some of the pictures below are mildly graphic. Dead animals.

The first order of business was to get all of the bags we could before moving things around too much. We cut any remaining rope and pack cloth holding the bags in place, and removed any of the bags that weren’t too stuck. A few of the bags we got out with only minor pushing and shoving. All the gear besides that (of which there was plenty) would require more serious means to retrieve.

We started by cutting away the log on top of the carcasses, and some other brush that was blocking access to our stuff. The main log was quite thick, at least a foot, but the packers had a handsaw on them so we made quick work of it. With the log and some other brush gone, we could access the carcasses and our gear much better

We got some more gear out after the log was gone, but some was still pinned by the weight of 1 or 2 mules. Dustin was the first one to dive into the pile of carcasses and get all bloody in order to pull on some of the bags that were really deep in there. With much wiggling them back and forth, he was able to wrestle a few more bags free, at the cost of diving in to the carcass pile and getting a little bit bloody and gross. I took a shift in the carcass pile after Dustin got out of there and got at least one more bag. Dustin deserves credit for going first and diving in there: I wasn’t even thinking of burrowing into the carcass pile until Dustin dove in and showed us all it wasn’t that bad. I got a little mule blood on my hands and shoes, but nowhere else—not too bad.

The last few bags of gear seemed totally stuck. Grown mules weigh about 1000 pounds, and some strong shoving back and forth was not gonna free anything pinned under that much weight. After several of us took shifts in the carcass pile and made no progress in freeing them, Jason wondered aloud whether we were gonna have to cut up the mule carcasses with the handsaw we had. None of us wanted to do that, as it would be way bloodier, way more gruesome than it was now, so we took a moment to think of another strategy.

We opted to get some rope around the neck of the top mule and get 3 guys pulling it upwards as hard as we could. Even the head and neck of a fully grown mule is quite heavy. But with 3 of us yarding on it in unison, we were able to move the top mule enough to free the remaining few bags. Mission accomplished.

We hauled those last few bags we freed up to the trail, then joined the other team cutting a trail for the other mules. It was much too steep right where the wreck was to get a good trail in, but they found a section of hillside just downstream that was less steep and they were cutting a trail in there. When I joined them, they were mostly done, and I only cut a few small branches in the way at the very top of their proposed trail. Once that was done, we all switched to the last remaining task: move all the gear from the spot on the trail directly above the wreck to the spot where Jason and the packers camped the night before, a quarter mile ahead. There the trail was much wider, and there was flat space off the trail that wasn’t densely vegetated, so we could spread out all the gear and take inventory of what was damaged. We all ran laps with as many bags as we could carry on our backs and fronts between those 2 spots until all the gear was nicely laid out at an open area on the trail.

Shortly after that task was done, 3 mules sauntered by us on the trail, having just come up the improvised trail we cut. They appeared unfazed, with a few minor skin wounds from the fall but otherwise unharmed as far as I could tell. It was only 9:30am—just a little over 3 hours since we started the cleanup process. Not bad for a task that the packers thought would take 3 days, and that Jason would take a whole day even with the small army that was his and Hans’ team.

The mules passed us, and the packers told us to let them wander on, so they could go graze on plants on the trail and work off all the nervous energy they had surely built up while stuck in the creek all night. We then took a group photo with all the cavers who helped out and some of the mule packers, including our signature “Surrender the Booty” pirate flag that we bring on these expeditions as our team flag.

Now that we had all the gear gathered on the trail, it was time to take inventory and see what was damaged. Many of the gear bags spent the night soaking in Helen Creek or in mule blood, after getting holes on them during the tumble down the hill. Most of our caving gear (ropes, clothing, vertical gear) was fairly robust and would probably be fine under those conditions. But our food was another story.

I found my food bag and quickly determined that it was more or less fine. The outside looked roughed up, and had a small hole in it, but it didn’t have water in it, so it must have not tumbled all the way down to the creek. When I opened it, I saw that several things were crushed, but it was almost entirely all salvageable. My jar of peanut butter was crushed which resulted in peanut butter spilling everywhere and coating the inside of my food bag, which was kind of annoying, but fine. I could scrape it from the inside of the bag and eat it still. Also my bag of peanut M&M’s had ruptured, and many of my M&M’s were crushed ans scattered as crumbs in the bag/embedded in the peanut butter mess. Again, kind of annoying, but totally salvageable, and I ended up eating all of it. After cleanup was complete, I made myself a breakfast consisting of oatmeal with the peanut-butter-and-crushed-peanut-M&M’s gloop that I scraped from the inside of my food bag. Pretty good actually.

Others were not so lucky. Several people had some of their food soaked in creek water and/or mule blood overnight and had to ditch it. Irina had it the worst—a large portion of her food was soaked and destroyed, and she announced that she was gonna have to hike out. Enough of us told her that we had extra food that we could share that we talked her out of it. I had tons of extra dinner mix and made sure it was known that her or anyone could have plenty of mine. I had a 6 liter and a 3.6 liter Darren drum full of dinner mix, which was probably around 33 meals, conservatively, as well as 2 meals of oatmeal. For 2 weeks I only needed 28 dinner/breakfast meals, so I had a good amount to spare. I was less sure about the quantities of food involved for my day food, as I just eyeballed that, but I brought a lot and would be fine sharing some. Irina also had some parts of her vertical gear damaged via crushing force, but Jason had spares in the relevant pieces of gear that he loaned her.

The packers were amazed that we had been able to fix the mess so effectively and so quickly. We could tell that they were a little shell-shocked when we first got to them this morning, so we quickly took over the operation and made our own plans and teams to get the cleanup done. They were grateful that we were able to get 3 of their mules back to the trail, as they are quite expensive. Grown and trained pack mules cost around $15,000. They had just lost 2 of them, and were at least thankful that they wouldn’t lose the other 3 due to our getting them back to the trail in a timely manner. We thought they were gonna turn around and head home after they retrieved their remaining live mules, but to our surprise they told us they would get all our gear to Pagoda Pass by noon the next day. Just 2 days behind schedule.

Now it was time for some planning about schedule changing (something that would happen frequently in the upcoming days). Our default option was the expedition as planned just on a 2 day delay. This option was not bad, but it would mean we would probably have to can the second stay at the yet-to-be-established camp 2 in the cave. One no-brainer option we could do to gain an extra day would be to carry the caving gear and food to basecamp today for those on the first scheduled trip in the cave. This would let the first scheduled trip go in the cave today, only one day late, instead of tomorrow after the mules brought their gear to Pagoda Pass. See the original planned schedule below if this doesn’t make sense.

Today was July 23. The first team scheduled to enter the cave was “Setup Team” on July 22, to haul sleeping bags and other camp gear to camp 1 in the cave and set it all up. That trip had to happen and was not something we could cancel. We decided to carry the gear of the 4 people on that team to basecamp today, so they could enter the cave later in the afternoon and start their trip only 1 day behind schedule. I grabbed one of the loaded camp packs with supplies for camp 1, and Pete’s vertical gear (he was on that setup team), and my own food bag, and packed it all in my backpack so I could bring it back to basecamp.

The “Lower Rig Team” was supposed to do a 2 day trip that would enter the cave, go below camp 1, fix a bunch of bad rigging there, then return to camp 1 on day 1; and exit the cave on day 2. The reason for this trip was that much of the cave between camp 1 and the site planned for camp 2 had horrible, unsafe rigging. There were many exposed stemming traverses without any floor visible below you as you stem above the void, and without ropes for safety. And there were some traverses like that that had very minimal, loose rigging that did not offer much safety. The rigging or lack thereof in these places got the job done in past years when we only went through these places with day packs, not camp gear. But this year when people would be going through there loaded down with heavy camp packs, we thought it was just too unsafe.

Another reason we wanted to beef up the rigging between camps 1 and 2 is that the travel through some of these spots got more difficult over the years as it gets tracked up with mud. Some of the traverses were fine without ropes when they were clean and dry. But as more cavers track mud through there from other parts of the cave that are extremely muddy, the stemming gets slick and insecure. Exposed stemming high above the out-of-sight floor without safety ropes is one thing when you’re stemming on dry rock, but when you’re on slick muddy rock, it’s even less of a good idea.

We decided to cancel the trip to beef up the rigging there before Push Team A came through with heavy camp packs. This would allow Push Team A, which would establish camp 2 and do the first exploration of the expedition, to start their trip on schedule, tomorrow. I was fine with this change, as I had been through this section before and found it a little hairy but generally fine. I would certainly be OK going through there with heavier, bigger camp packs instead of the lighter day packs I had carried through there before.

Still, I found it a little disconcerting that we were cancelling this trip for the following reason. 2 years ago, at the end of the 2024 expedition when we were discussing the possibility of setting up camp 2 in the planned spot that we discovered that year, Jason asked the group whether it would be feasible to go through the terrain below camp 1 with the sketchy muddy stemming, with heavy camp packs, without major modifications to the rigging. I said yes while everyone else in the group said no. So I knew that me being comfortable hauling camp packs all the way to the proposed camp 2 site is not something that we should necessarily expect of everyone else.

Still, that was the decision we made. We decided that “Mud Camp Team A” would take over the rigging work that was planned for Lower Rig Team, so that Push Team B would get the benefit of the better rigging, as would Push Team A when exiting the cave. This would mean that Mud Camp Team A wouldn’t get to do any survey at the bottom of the cave as originally planned, but that was the tradeoff we would take to get both Push Teams A and B going in the cave on time and therefore with there full allotted time for exploration.

At least, this plan is what ended up happening. I know the planned schedule changed several times in the next 2 days, and I forget exactly what the intermediate plans were and when they were hatched. This was the final plan that we had settled on by the time that the relevant teams were going in the cave.

With our new plan hatched, my bag loaded down with gear for people caving today, and the mule packers dedicated to getting us the rest of our gear tomorrow, we set off from the site of the wreck to basecamp. Dustin and I offered to take handsaws from the packers and cut away brush on the Helen Creek trail to make their job easier tomorrow. The trail gets very little use and is generally quite overgrown. Mules can generally handle brushy, overgrown trail just fine, but after the accident we wanted to make the trail as nice as possible. So Dustin and I set off with our saws, stopping every few minutes to cut away brush and letting everyone else pass us as we worked.

Cutting brush ended up being much more work than we thought. Much of the brush was thick and woody and had likely been growing over the trail for several years. Back in the 2024 expedition we ran into a forest service trail crew doing work on the Helen Creek trail, so I know it gets maintained occasionally, but I wouldn’t be surprised if it hadn’t gotten any maintenance since then. Once Dustin and I started working, the cool morning temps were gone and we were baking in the same 90’s-and-sunny heat we were familiar with from the previous few days. The sun was especially harsh as most of our work was above the part of the trail with denser vegetation and shade, and we worked through high noon. We now knew we were both caving tomorrow, so we made sure to drink plenty of water to stay hydrated in order to be ready for a long hard day the next day.

After several hours of work, the Helen Creek trail above the site of the wreck was totally clear of brush and we were quite proud of our work. I could tell that all that labor in the heat had taken a lot out of me, and I was ready to be in a 37 degree cold wet muddy cave instead. Still, Dustin and I were proud of the job we did and of how much easier the trail would be for the packers the next day.

When we were partway up the trail and still cutting brush, Jason passed us after spending some time talking with the packers and some forest service employees who had hiked in. Jason told us that the trail was noticeably better and that we were doing great work. He also told us some bad news: that the forest service was closing the Helen Creek trail due to grizzly bear danger from the mule carcasses. We would finish our day on the trail and the packers would take our stuff up on it tomorrow, but after that they would close the trail for several weeks or until they could dispose of the carcasses.

The Bob Marshall Wilderness in northern Montana has the highest number of grizzly bears (the dangerous ones, unlike black bears which are much more common and much less dangerous) anywhere in the lower 48 in the US. Grizzlies have very sensitive smell, and they were very likely to smell the mule carcasses as they started to fester and decay, especially in this heat. The forest service told Jason, and he told us, that their peak smelliness would likely come after the carcasses bloated then exploded, which would take about 2 weeks. 2 weeks was when we were planning on hiking out down that trail, and was now when there was the highest probability of running into a grizzly bear on the trail. Grizzly bears fiercely defend their food stashes, and they would likely smell us if several (smelly, not-recently-showered) humans came down the trail and passed a few hundred feet from the carcasses. Thus if we hiked out on our planned route in 2 weeks, we were quite likely to get attacked by a grizzly bear fiercely defending its food stash, at least according to the forest service. Hence their decision to close the trail after tomorrow and recommend we take a different way out. They recommended an alternate route to Jason that we would take. This alternate route was an abandoned, unmaintained trail that followed the ridge south of Helen Creek. They didn’t know what kind of shape this former trail was in, but they said it should be passable, if less pleasant than Helen Creek. Oh well, complication number…??… of the expedition. I had lost count.

Dustin and I were the last of the group to get back to basecamp, around mid-afternoon. The first haul team was planning on entering the cave later that afternoon. I volunteered to carry some of their gear to the entrance, so save them a little bit of effort before their long in-cave hauling effort they would face tonight. Thankfully I had a little bit of time to make some food and relax a little bit before getting back into work mode. Because Pete (who was on this haul team) had some extra space in his pack, he volunteered to haul a small amount of gear to camp 1 for Push Team A (which I was one). I gave him a 2 pound block of cheddar cheese and thanked him profusely for taking that to camp 1 for us.

A bunch of us hiked over to the entrance carrying gear for the haul team around 4pm. Hans’ high schoolers also came, as they were planning on going in the cave just a little bit, to the first rope, to get some sense for what it was like. After yet another hot, sweaty hike to the entrance, we all took a group photo of the first group entering the cave this year. The haul team entered the cave, and shortly after, the high schoolers. The expedition was officially back on track.

I hiked back to basecamp and spent the rest of the evening resting and thinking about what a crazy day it had been. I woke up at 4 in the morning, cleaned up a mule wreck including burrowing in between some mule carcasses, did some major trail work, and made it back in time to bring gear to the cave entrance for a team going all the way to camp 1. The team going to camp 1 had an especially long day, and theirs would not be over for several more hours. Tomorrow would be a smaller, but still substantial day for me, as I would haul a few loads between Pagoda Pass and basecamp then haul some more gear to camp 1. We made a fire at camp that night and basked in the glory of our successful day for a little while before going to bed relatively early.

Wednesday, July 24th: Pagoda Pass hauling, and in the cave to camp 1

The packers were right on time at Pagoda Pass at noon. I did 2 laps hauling heavy loads (probably around 60 pounds) from Pagoda Pass to basecamp. The heat and sun were as strong as ever, but at least today I had the luxury of carrying the heavy loads mostly downhill and going back uphill with an empty pack. The route between the pass gains about 300 ft and loses about 800 ft of elevation, all off-trail. I also had the luxury of knowing that soon I would be in a cold, dark wet cave, out of the sun (where I would probably soon miss the scorching heat anyways).

After my 2 laps between the pass and basecamp, I loaded up on food in preparation for my trip to camp 1 in the cave later today, and started packing my cave bag. Jason decided that the 4th person on our Push Team A would be Derek Von Nieda. So the team would be Dustin, Tommy, Derek, and I. I had never caved with Derek, and had never met him before him and Hans picked me up at the airport on Friday, so I was looking forward to getting to know him. The 4 of us packed our camp bags together, taking group gear we would all need for our trip from a pile and distributing it among our bags. With all our food for our scheduled 7 day camp trip in our bags, we had no space for all the group gear we would need. This was a problem.

First we of course unpacked and repacked our cave packs several times, each time fitting a little more stuff in them. But we never came close to getting all the personal gear (food, extra clothes) and group gear (rope, survey equipment, sleeping bags and sleeping pads) we needed into the cave bags. This led to us changing the plan yet again. Push Team A (me and 3 others) would take only 2 days worth of food down to camp 1 then camp 2 the next day. Mud Camp Team A, entering the next day, would take our remaining food, and some extra rope, and bring it to camp 1 then halfway between camps 1 and 2 on that first day or their trip which would also be the second day of our trip. On the second day of our trip, after we reached camp 2, we would head back towards camp 1 and meet Mud Camp Team A halfway between camps 1 and 2. There we would pick up the gear and haul it down to camp 2, where we would then have all the food and gear necessary for our 7 day trip.

With yet another schedule change solidified, we packed the subset of our personal gear and group gear that we would need, and headed to the cave. Just like I did yesterday, we had some of the others still on the surface haul some of our gear to the entrance for us, to save us just a little bit of effort. This was the first expedition that we had ever gone caving on the same day that we hauled gear from Pagoda Pass. Our Push Team A was doing this today, and the Setup Team yesterday did something similar and even harder when they went to camp 1 the same day they cleaned up the mule wreck.

I was trying a slightly unconventional and bold cave suit strategy this year, and I was eager to see how my choice would work out. The standard choice of cave suit for Tears is one made out of a fully waterproof, non-breathable fabric like PVC or TPU. These suits are appropriate for the wettest, coldest caves, because they keep all water from getting through the fabric. This also means that you get soaked from your own sweat when working hard, as none of your own moisture can work its way outside the suit. This tradeoff is worth it for very cold, wet caves, as you’re gonna get wet from something anyways, and you’d rather it be from your own sweat than frigid cave water.

Nylon suits, in contrast, are only moderately water-resistant, and they are moderately breathable, meaning water inside the suit from your own sweat or otherwise can work its way outside via evaporation and diffusion through the suit fabric. These suits are better for caves that aren’t too wet, as you can dry out the interior over time, which makes you more comfortable, and possibly warmer if you stop and generate less body heat in a not-too-wet section of cave.

Most people wear PVC or TPU suits in Tears, such as the yellow PVC suits you can see above, but this year a few brave souls were trying nylon. Pete and I were rocking nylon suits on our long camp trips. You can see my orange nylon Landjoff suit above. Adrian also had a nylon suit, but he wasn’t planning on going below camp 1, where there’s wet, slimy, sticky, sloppy mud that chills you to the bone and makes travel difficult in so many other ways that make the cave infamous. The idea was that this year it might make sense to switch to nylon since we were camping below the slimy mud layer for the first time. Instead of commuting through the slime below camp 1 every day in order to get to the bottom of the cave where there was exploration to be done, this year each exploration team would only pass through the slime once on the way to camp 2 and once on the way out of the cave. So we opted to made a trade that made for a colder, less pleasant trip between camps 1 and 2 but hopefully for a warmer, more pleasant time below camp 2, which was drier and not muddy at least as far as we had gotten last year.

Of course, we had no idea whether the cave below camp 2 would remain dry(-ish) and free of mud. We had only explored about 20 minutes beyond camp 2, and we had truly no idea what lie beyond. That 20 minutes of travel was a slightly narrow but really “tight” sideways walking canyon with a stream in the floor, and walls that were damp but clean-washed and not muddy. It was blissful in comparison to the nasty tight series of squeezes fully of wet slimy mud that characterized the cave above camp 2. But the cave could easily change character beyond what we explored 2 years ago, and me and Pete were both taking a gamble that the cave would remain appropriate for nylon suits beyond the current frontier of exploration.

Either way, nylon suits were better suited for the cave above camp 1. That’s before the serious mud hits you, so you don’t get too wet from the cave. And now that all anyone does in that section of the cave is move fast and haul heavy bags, it’s rare to get cold in that section, so cooler, more breathable suits are best. I would see if nylon was a good choice when going through the slime below camp 1 and when exploring new cave beyond camp 2.

The part of the cave above camp 1 can be divided into 2 sections. The second half, closer to camp 1, is called the Bunga Bunga Meanders. We don’t have a name for the half above the Bunga Bunga, we’ll just call it the upper part of the cave. Travel time from the entrance to camp 1 generally takes 5-8 hours, depending on the group’s fitness and how heavy their bags are. In 2022, our team took 7.5 hours on the way down, as most of us were new to the cave and we tried to avoid getting sweaty/wet, by going slow. We then took 5.5 hours on the way out, even though we were going up instead of down, as we had smaller, lighter bags and we didn’t care so much about sweating and getting soaked. I was curious how my time this year would compare to summer 2022. I expected it to be a bit slower, because now I had a desk job instead of being a funemployed dirtbag that had been doing nothing but hard labor and cardio for months in a row.

How did the actual caving part of this expedition go? Did we get tons of amazing exploration done, and push the cave deeper? Or was it full of mishaps, as the trip had been so far? All that will be in the next post—subscribe and wait and see to find out what happens!

Subscribe now

Prince of Wales Island is a large island (larger than Delaware) off the coast of Alaska, at the southern tip of its panhandle. It is a lush, wet, green temperate rainforest, not unlike the rainforests of the Olympic Peninsula in Washington. Not full of barren snow-capped mountains as many people imagine when they think of Alaska. To get there, you first fly into Ketchikan, Alaska, a small city on Revillagigedo Island, adjacent to Prince of Wales. Plenty of flights there from Seattle. Then you take a 3.5 hour ferry ride from Ketchikan to Prince of Wales Island, operated once a day every day by the state of Alaska's Inter-Island Ferry Authority. Technically you also have to take a short (5 minute) ferry ride from the Ketchikan airport to Ketchikan itself, as the airport is on a tiny island right next to Revillagigedo Island.

I flew to Ketchikan on Saturday, July 6, intending to stay in a hotel in Ketchikan with fellow team members Amelia Fatykhova and John Dunham before taking the Ferry to Prince of Wales the next day. And then began the first of several minor mishaps that happened on this trip: Alaska/American airlines lost my checked bag! This was particularly bad because once we took the ferry from Ketchikan to Prince of Wales Island the next day, we were planning on driving several hours from the ferry terminal (where we could pick up the bag if/when it arrived) to where we would start our approach to the caves, and then hike several hours up a mountain to setup basecamp. So it would be infeasible to go all the way back to the ferry terminal to get the bag midweek. Additionally, we would have no cell service to get notified if/when they found my bag. So I was worried about being stuck in limbo without my essential caving gear all week.

Thankfully, American Airlines found it the next morning, Sunday July 7 (stuck in Phoenix, where I had a layover). They flew it to Seattle then Ketchikan, although it was scheduled to arrive in Ketchikan after I was to board the ferry to Prince of Wales at 3 pm, so I would be unable to pick up the bag before leaving. Thankfully, Alaska Airlines, which was in custody of the bag at this point, said they could send it over to Prince of Wales Island via Island Air, a small local airline that operates seaplane flights between Ketchikan and Prince of Wales Island. However, there was still the problem that the bag would arrive on the Island Air flight in Klawock, AK, on Prince of Wales Island, on the last flight of the day, shortly before the airline closed at 7 pm. We wouldn't be able to get from the ferry to Klawock until after 7 pm. Klawock was 2.5 hours from where we planned on driving to and sleeping that night, and we really didn't want to have to drive back the next morning, when we were supposed to be backpacking up to basecamp. After several calls to Island Air, I convinced them to leave my bag outside, hidden away in a corner, so I could pick it up that evening after they closed on the way on our drive from the ferry terminal to where we were staying that night.

That all worked, and I retrieved my bag and all of us that took the ferry started the 3 hour drive from Hollis, where the ferry terminal is, to the El Capitan Cave guide cabin, where we would all be spending the night. Anna and Christian, US Forest Service employees who live on Prince of Wales Island and make this trip happen, picked us up and transported us and all of our gear in Forest Service vehicles, and arranged for us to stay at the guide cabin that the Forest Service has at El Cap Cave. This is part of the incredible amount of support we get from the Forest Service for these trips. The Tongass National Forest is very interested in knowing about the caves on its land, and supports our cave exploration and mapping efforts in exchange for locations and data for all the caves we find and explore.

That night we stayed at the guide cabin for El Capitan Cave, a cave right by the ocean where the Forest Service offers tours to the public. Every summer they hire 2 interns to be tour guides, and they live at that cabin all summer. They had that week off so that we could stay at the cabin, although they were still there when we got there and stuck around for a while to hang out. Living in a cabin in the middle of nowhere, far from any town, with only 1 other person as your human interaction most days, must get lonely, and thus they seemed eager to hang out and chat.

Monday, July 8

The next morning, Monday July 8, was our day to backpack up to basecamp, setup camp, and start caving. We ate a luxurious breakfast of eggs and salmon that Anna caught before driving a short ways from the guide cabin to where we would start our hike up to basecamp. Thankfully it was dry and sunny that day, not rainy as it usually is and as it was supposed to be the rest of the week, which meant that we could setup camp without getting all of our stuff wet. The hike up to basecamp started out on an old overgrown logging road, which the Forest Service fire crew on the island had recently helped clear a bit, before going off trail and up a steep hillside. We all got quite hot while bushwhacking up the steep hillside with the sun out, but we knew this relatively pleasant weather wouldn't last so we didn't complain. After about 2 miles and 1200 ft of elevation gain, we arrived at basecamp, which was a huge (~500 ft wide), open, grassy sinkhole enclosed by rocky mountains and a sparse pine forest rising above camp, a bit north of El Capitan Peak. Serene.

Me and Christian led the hike up and got there shortly after 11am, not quite 2 hours after we started hiking. We met Anna and Alex Lyles at basecamp, who had already hiked up a day earlier. We took a few hours to setup our tents and make some food before setting off to go caving that day. The forest service paid for a helicopter (!!!) to fly up much of the group caving gear such as rope, bolts, and drills, and all of our food, so base camp was fully stocked when we got there without us having to carry super heavy bags. The kitchen area had a nice large folding table we could use to prep food, and a Coleman stove, so it would be no issue boiling water for 12 people's meals. It was covered with large, standing-height tarps, so we could all hang out and be dry in the kitchen. We also had an battery-powered electric bear fence that enclosed the 10 ft by 20 ft kitchen area, so we wouldn't have to bear hang our food at nights. There are no grizzly bears on Prince of Wales Island (thankfully), but there are black bears that will mess with your food if you don't store it properly. It was a nice convenience not having to deal with any bear hangs, especially since we had tons of food for 12 people for a week and it would have been a pain to hang it all up and take it all down twice a day.

After setting up our camps, we started getting ready for a half day of caving that day. That entailed packing up some gear, namely rope, bolts, a drill, among other things, all of which the Forest Service had helicoptered up to basecamp. I also taped up my big toes, because they had developed some rub spots due to hiking up to basecamp in my rain boots. I tried that somewhat bold strategy of using my rain boots, or wellies as cavers call them, as my hiking shoes in addition to my caving shoes, to save weight and to save the inconvenience of having soaking wet hiking shoes after it started raining on the surface, which it surely would. However, my wellies abraded my big (and later, little) toes pretty bad, and I had to tape them up to reduce the sensitivity of the rub spots.

For the 4 teams we had up there, we had 3 Milwaukee M12 drills (which are small, lightweight drills that drill pretty slowly but are easy to pack and carry far) and one Bosch 36V (which is the opposite—large, heavy, but drills like a champ). Naturally, I wanted our team to take the 36V since we had a lot of bolts to place, and I love carrying heavy things anyways. The lead our team selected to work on was in -560 ft in Snowhole Cave, a known cave, 652 ft deep, that had last been visited by Kevin Allred, Mark Fritzke, and Pete Smith in 2010. They left off a wide open pit lead marked on the map as "unexplored pit about 70 ft deep, with murmuring water sounds from below". What an excellent lead! But, to get to that lead at -560 ft in Snowhole, we figured we would have to do lots of rigging which would probably entail placing a bunch of bolts. The previous crew who had been to Snowhole used more natural anchors and used much fewer bolts, preferring to use thicker, stiffer ropes and rope pads so the ropes could stand up to the wear and tear resulting from rigging that is less bolt-intensive and less impactful to the cave. Cavers call this style of rigging "indestructible rope technique" (IRT). The rigging style we preferred is what cavers call "alpine style"—carefully rigging ropes so they stay away from the cave walls and don't abrade. This lets you get away with much thinner, lighter weight rope, and is generally more suitable for long-term rigging, although it means you have to place more bolts and thus impact the cave more. Anyways, we took the 36V drill because we figured our team had the most bolts to place, and because no one else wanted to carry it.

Our team, which was Christian, Reilly, and myself, left camp at 3 pm and set off for the cave. We had a GPS point from the last crew that went there, which was only 20 minutes from camp. The approach started out in the sparse pine forest above the grassy sinkhole that was camp before breaking out into some impressive karst terrain above the trees. The karst was steep exposed limestone pavement, largely very smooth and monolithic, but punctuated by gaping earth cracks and razor-sharp rillenkarren (prominent solutional channels carved into steep surfaces on the limestone).

The approach hike to the GPS point we had for Snowhole was short, scenic, and pleasant; the only problem was there was no cave when we got there. We spent about 2 hours looking around, wandering in concentric circles around the GPS point, hoping it would be not far off what we were told. We were told the entrance was a 2 ft by 50 ft long crack in the earth, so we figured there's no way we would miss it if we walked over it. After doing enough concentric circles for me to be sure that this was futile, we stopped and thought about what to do next. Thankfully Christian realized he had some old cave documents downloaded on his phone (no cell service up here) that had some info about Snowhole. One of the documents had a location that was different from the GPS pin we had (which was from the Forest Service database of caves). That location was not far away and turned out to be the cave entrance. Probably the location we had from the Forest Service was an estimate from the pre-GPS era—hand drawn on a topo map, then entered into a database of GPS points without noting that it was a rough approximation of the location.

We brought all our gear, which we had ditched at the wrong GPS point when we started doing concentric circles, over to the entrance. It was exactly as described: a 2 ft wide crack that ran up a slope for about 50 ft. It's the type of feature that you would often ignore when out ridgewalking, as these earth cracks rarely go. And apparently many cavers did ignore it over the years—for many years people just stepped over the crack on the approach to Blowing in the Wind Cave. Until one day someone decided to drop some rocks down it to see what they did, and they head the rocks tumble for quite a ways. Then they realized this crack was actually a cave and started exploring it.

When gearing up at the entrance, I realized that my helmet had a big crack in the foam. I'm pretty sure I cracked it by stuffing it into an overhead compartment on a plane, as this crack was not there the last time I caved with it. Helmets with some exposed foam, and a hard plastic shell only over the top of the helmet, are a fairly lightweight style of helmet and questionably durable enough for caving. I had used this helmet for plenty of harsh caving and hadn't noticed any issues with the foam, then of course broke it not from caving but from flying with it. Not good! I needed to use this helmet for the next month, including at Tears of the Turtle Cave, which is notorious for destroying gear. I would see what repairs I could do when back at basecamp, but for now I would just have to use the cracked helmet and try not to make it worse by bashing or dragging my head against a wall.

It was just before 6 pm by the time I was ready to go in the cave. The plan was for me to go down with all our rope and the drill and bolting stuff, and rig down as far as I could in a few hours. The map showed the cave going pretty much straight down one pit complex until -450 ft, perhaps with some small landings which may or may not fit multiple people. I wasn't expecting to get to the good flat area at -450 ft in one day, so Christian and Reilly would follow only if I got to a good spot for multiple people to hang out and get off rope, otherwise they would just hang out on the surface waiting for me to call to them. There were no options for natural anchors on the surface, so I put in 2 bolts to start a lead-in line to the entrance. I put in a rebelay right below the lip, another partway down the crack, and a redirect before getting to a small ledge with some snow where I saw a rebelay bolt from previous parties, the first so far. I could tell that previous parties were definitely rigging this IRT-style. I backed up the single bolt with another at that rebelay ledge, and continued down the pit. Here the pit changed shape from a long skinny earth crack to a more classic circular cross section pit; it now looked like I was in a real cave, not an earth crack. I set another rebelay at another sloping snowy ledge before deciding it was about time to head back up to the surface. I had killed lots of time trundling loose rocks from the walls of the pit to remove rockfall danger on my way down. The walls of the pit were generally good hard bedrock, but clearly the pit had collected a lot of small-medium rocks and other debris over the years, much of which was precariously perched in various nooks in the crannies in the walls, waiting for the slightest disturbance from a caver's foot or swinging cave bag to knock it loose and send it crashing down on whoever was unlucky enough to be below. I wanted all of that debris gone before anyone else came into the cave, and it took quite a bit of effort trundling everything that looked remotely dangerous.

I packed up the drill in a dry bag, and hung it and the rest of the rope and bolts (I had used 11 so far—2 for a lead-in line at the entrance, 4 2-bolt rebelays, and a redirect) at the last rebelay I set before heading back up to the surface. I got to the surface at 8:30 pm, where Christian and Reilly were sunbathing on the rock, enjoying the last bits of warm sunny weather before the rain was supposed to roll in. From looking at the map, I guessed that I got to around -200 ft that day, although the map has very little detail in the entrance series so it was hard to tell. We found a super easy route down from the cave and made it back to camp in a mere 17 minutes. We made dinner at our luxurious camp kitchen and were in bed by 11 pm.

Dinner that night, as with every night, was rehydrated Dinner Mix—a lightweight and calorie-dense mix of dehydrated food ingredients that you only have to add hot water to in order to make it into a hot meal. I was in charge of acquiring all the food for everyone on the trip. Naturally, I, a physicist, was much more concerned with things like energy-density, nutrition, ease of cooking, and cost-effectiveness of the food rather than aesthetic factors such as the food tasting good or not eating the same thing every meal. So I decided to just order a bunch of dinner mix ingredients in bulk. I first found out about dinner mix as cave food when I went on the Cueva Cheve expedition in Mexico in 2022; dinner mix is standard cave food there too. The dinner mix I make is somewhat modified from the Cheve mix, but the idea is the same. The ingredients that go in the dry mix are:

• Instant mashed potatoes powder

• Chia seeds

• Cashews

• Walnuts

• Textured vegetable protein

To turn it into a meal, you add boiling water and some butter and let it rehydrate for just a few minutes. The powder absorbs all the water and expands a bunch and turns into a filling hot meal. The nuts provide a nice crunch, so the texture is not just slop. We did have some add-ins that people could use to spruce up their meals, including cheese, sun-dried tomatoes, bouillon cubes, and various spice blends. Once you add in a few of those, dinner mix can actually be quite good, and everyone on the trip seemed to like it, at least enough to not get sick of it after eating it for every dinner and breakfast for a week. I even had a few people ask me for the recipe so they can make it for themselves.

The big advantage of dinner mix over what we had done in previous years (trying to cook normal meals, and designating one team per night to be the cook team) is that anyone can easily and quickly cook a full meal, whenever they want, in whatever quantity they want, without relying on anyone else. In previous years, one team would have to get back to camp early to cook dinner, which would mean they would have a shorter day of caving. They would inevitably make too little or too much, and when other teams would get back from long days at strange hours in the middle of the night, they come back to cold food, if there was any left at all. Dinner mix solved those problems. It also had the big advantage that all the ingredients could be purchased in bulk on Amazon and shipped to Prince of Wales Island (Amazon Prime works there!). No need to pay expensive Alaska prices. For lunch/snacks during the day, I bought fairly standard cave fare in bulk—cheese, tortillas, nuts and seeds, dried fruit, chocolate, and some other candy. The only things we had to buy on the island instead of online in bulk were the cheese and butter. The food situation worked great, and I plan on doing more-or-less the same food plan next year. All of the food was helicoptered up to basecamp by the Forest Service.

One ominous note: this day, Michael Ketzner started feeling sick. The week before, he had been at the NSS convention (national convention for caving in the US), where apparently COVID was running around rampant. Could Michael be spreading COVID to all of us? Oh well, not worth worrying about it now that we had already spent a bunch of time together…

Tuesday, July 9

The next day, Tuesday July 9, the sun woke me up at 7 am. Before everyone else got up, I took some time to do a few miscellaneous tasks around camp. I dug a deep latrine so people wouldn't have to dig their own cat holes every day. I also flagged a trail to it from camp so people could find it easily even when stumbling around in the dark at night. Then I repaired my flip flops (useful for walking around camp), which I had accidentally torn some of the straps on the day before. I brought a speedy stitcher, a tool that lets you sew with heavy-duty thread into thick fabric, up to camp as an emergency repair tool, and I'm glad I did because I already needed to use it by day 2. With the speedy stitcher and some 1" webbing we had at camp for redirects, I was able to re-attach the torn flip-flop straps.

After a hearty breakfast of (you guessed it) dinner mix, we were out of camp by 10:30 am and at the cave in a short 20 minutes, carrying even more rope and bolts that we brought the previous day. I told Christian and Reilly (who were my teammates the entire week) to give me an hour and a half head start while I continue rigging, figuring that I would get to a landing marked on the map at -328 ft before they caught up to me. 2 more rebelays and 2 more redirects got me to that landing, which was a 3 ft by 8 ft ledge where you can get off rope and 3 people can fit. The timing worked out nicely, as Christian and Reilly caught up to me there as I was rigging the next pit that continues down from the edge of the landing. That pit had a short pitch with a tight awkward entry to a rebelay, then a long pitch that starts out in a tight crack where I put in 2 redirects and turns into a huge circular cross section pit, then another short pitch that I put in 2 bolts for although we ended up just free that last short pitch. Here we finally gained a waterfall which lightly sprays you as you descend the big part of the long pitch, the first flowing water in the cave.

Now we were at a fairly large flat area at the bottom of the pit with a flowing stream in the floor, with the passage continuing horizontally to the south and rapidly narrowing. We were pretty sure that we were at the spot on the map at -485 ft immediately before "The Stripper", a tight vertical squeeze we had been warned about. The passage continuing south exactly matched that description: continuing tight, then getting even tighter as it steepens and you want a rope right before the lip of the larger steeper pit visible beyond. I rigged a rope through the squeeze, struggled moderately down it, and was quickly beyond the squeeze into a nicer, bigger pit. I shouted triumphantly to Christian and Reilly that The Stripper wasn't even that bad.

After The Stripper, the map indicated that I was supposed to not go all the way to the bottom of the pit, but instead find a window into some horizontal passage heading north. I did indeed find such a window, but it was extremely tight, and I was surprised that The Stripper would be labelled on the map but this tight, nasty-looking window heading north would not be named something worse. I got into it, set 2 bolts to get the rope in the right spot at the window, and started squeezing down it. I realized this couldn't be the way on when I saw fragile, easily broken protrusions in the rock right at the widest spot where people would go. Surely if anyone had been down there, they would have broken them off. This must not be the window.

I derigged the rope from that window and went down to the bottom of the pit just to check it out. It was not plugged with rocks as the map indicated; instead I saw survey stations continuing down a tight passage that also visibly constricted right at the lip of a pit. I realized that we had not yet gone through The Stripper—I was staring at it right now. The ledge we landed on previously was actually the ledge on the map at -450 ft, not the ledge before The Stripper at -485 ft. Now we were actually at -485 ft.

OK, so now it was time to face down The Stripper. I bolted an anchor right at the start of the tight part of the passage to get the rope in the right spot, and rappelled down it. Really I rappelled sideways through it, thru passage that you wouldn't even want a rope for if there weren't a pit right at the other end that you can't see because it's so tight. It was slightly trickier than the previous squeeze, but not that much worse. It was a little less pleasant because the stream ran thru the floor that you push your arm and leg off of, so that arm and leg do get quite wet.

Now that I was actually thru The Stripper, I saw an obvious window in the wall of the pit that continued to the north as hands-and-knees crawling horizontal passage. This was clearly the window that we were supposed to find. I rappelled to the level of the horizontal passage, swung into it (in the process swinging myself right into the waterfall going down the pit, getting myself even more soaked than I was after going thru The Stripper), and mantled into the hands-and-knees crawl passage. I crawled a bit on the breakdown floor to a comfy spot then anchored the other end of the rope there, so the others would not have to repeat the tricky mantle move I did to get into the window. There were 2 interesting and surprising features of this passage that I noticed as soon as I got into it. First, the passage had ridiculous air moving through it. The cave was sucking here, in contrast to the entrance which blew a much smaller amount of air. Not the best thing to run into when you're soaking wet, but I didn't mind that much because all this air was very promising. Airflow is generally a good sign that a cave is going to keep going—in fact, there's a common saying in caving "if it blows, it goes".

The second interesting feature that I noticed was that the ceiling was a flat bedrock layer that was distinctly different from the limestone the cave was formed in—probably an insoluble shale layer. The fact that the ceiling, rather than the floor, was an insoluble layer means that this passage likely formed by water being pushed upwards from below, running into the insoluble layer which stops upwards progress of the water and forms a ceiling. This is in contrast to what more often happens when caves form: water flows downwards, stopping at insoluble layers that stop downward progress and form a floors. What would cause water to be pushed upwards like this? Probably this entire section of cave was flooded, underneath some sort of water table, and the high pressure from deep water forced the water to flow in which ever direction is could find weaknesses in the rock, which could be any direction, up down or sideways. When cave passages forms this way, fully submerged in water, we call such passage "phreatic". Cave passages that form not submerged bur rather at or above the level of the water table are called "vadose". We don't see much phreatic passage in Alaska caves, so we were all excited to see where this one would go and what it would do.

We followed this passage for a few minutes of hands-and-knees crawling, excitedly power-crawling as the strong air egged us on. The dirt in the floor of the passage was bone dry, thanks to the air ripping thru the passage, and we were able to dry out our soaking wet suits by dragging them along the dirt, which was nice. The next obstacle a few minutes down this passage was a large pit in the floor that the crawlway spilled out into. The pit was about 20 ft in diameter, and deep enough that we couldn't see the bottom. A traverse rope was anchored to our side of the pit with one bolt and continued out of sight around a corner where it appeared that the crawlway continued on the other side of the pit. There were also several short ropes stashed here by the previous team, which was convenient for us because we were low on rope at that time. We had no qualms taking the rope, because the last team that had been here left them here 14 years ago and had no plans to come back. We backed up the 1-bolt anchor with a new bolt we placed, and I went first across the traverse rope. Because I didn't know what it was anchored to on the other side, I put an ascender on the traverse line, facing the 2-bolt anchor on our side, so that if the anchor on the other side failed, my ascender should catch me.

The rope was anchored to a huge boulder on the other side: totally bomber. I told Christian and Reilly that the anchor on the other side was good and they didn't need to use an ascender on the traverse line for safety. We all got across and continued along the crawlway on the other side. Notably, the crawlway lost almost all of the airflow at the pit; it was very faint on the other side. The air must be going down the pit or down one of the other passages connected to the top of the pit that aren't the 2 passages on both ends of the traverse rope. There was a good-looking horizontal passage at the same height of the traverse rope on the other side, making the top of the pit a 3-way intersection. Later we looked at the map and saw that they did get to that passage, and that it went for about 100 ft before ending at a spot labelled "crack too tight, takes air".

Despite being slightly disappointed at losing the air, we continued down the crawlway on the other side of the pit going towards our lead. The next obstacle was a small downclimb and slope labelled "slick mud, rope needed". When I got there, I went down it then back up, and found it doable without a rope but certainly agreed that a rope there would be helpful. We used one of the short ropes we found stashed at the pit traverse. Because it was an old stiff 11 mm rope, I used a somewhat creative and funny way to attach the rope to a bolt: rope straight through the bolt hanger and a barrel knot on the other side of the hanger. This is very efficient as it uses very little rope and no quicklink. It's totally safe because we were using Pingo Bonier hangers with very round edges that let you tie the rope straight to the hanger, and because we were using old swollen 11 mm rope that would never pull through the bolt hanger hole. It's just a very unconventional way to rig.

Just beyond this chimney downclimb then mud slope was our lead: a wide open pit with a large, comfy, walking-height room where we would get on rope. The final station from the last push trip 14 years ago, E48, was clearly labelled on a wall right where you would want to get on rope. Throwing rocks down the pit and timing them was consistent with a depth of about 70 ft, and we could hear water below, just like the lead was labelled on the map: "unexplored pit about 70 ft deep with murmuring water sounds below". Time to start some new exploration, woohoo! We As Christian and Reilly stopped to unpack and switch gears to survey mode (I was to continue rigging), I was dismayed to find out that I only had 2 bolts left. It was only 6 pm, and we were in no rush to get out early, so I had to be stingy with the bolts. A change from how I had been using them the past 2 days: rigging everything as perfectly alpine as I could, eliminating any and all rope rub. Oh well, at least the next day we would return with many more bolts and fix whatever MacGyver-rigging I was about to do.

We all looked around for natural anchor options, hoping to save the bolts for whatever came next. We didn't find anything bomber; one small horn I considered shattered when I hit it with a hammer, so that was definitely out as a natural anchor option. I decided to place the last of the 2 bolts we had right at E48 and just roll with it, hoping that we would make it to the bottom of this pit without excessive rope rub. Indeed the rope rub was not bad: the lip of the pit was mud, which meant that chunks of mud did fall down on you as you were on rope below the lip which was annoying, but I wasn't concerned about excessive rope wear from just one party going down then back up the rope.

Christian went first down the rope, because then I had to poop (thankfully there was a large muddy area away from where we walk there, so I could bury it and not pack it out). After we all got down the pit (which was in fact about 70 ft), we started surveying down a crawlway at the bottom. This crawlway was muddy and had no air and no water, so it didn't seem super promising, but after only a few shots we got to a standup room where there was another pit! We were out of bolts at this point, but thankfully there was a large boulder we could tie the rope around. I went down this pit first. The IRT rigging around the boulder was not ideal—it rubbed badly over multiple edges, and it put you right in the path of the waterfall (the passage picked up water right before the pit). We would have to fix this when we returned the next day. But, it was good enough for me and Christian to get down about 40 ft to a large comfy ledge where we could get off rope. Reilly didn't come down then because the rigging sucked and we were gonna come back the next day anyways and do it with better rigging. We got the survey down to that ledge, beyond which the pit continued down as 5 ft wide, very tall canyon, with no floor visible below. We would need bolts to go any further. A great lead to return to the next day!

One more thing we had to do before gunning for the surface was tie in the bottom of the first new pit from that day with E48, as we started the survey at the bottom of that pit. We did that in a few shots, left the drill and remnants of our bolting supplies (8 quicklinks, 3 bolt hangers, no bolts) at E48, and bolted for the surface. I started exiting from E48 around 8:45 pm and made it to the surface just before 10 pm. While ascending out of the cave I noticed myself coughing a bit—was I getting whatever sickness, possibly COVID, that Michael had? I wasn't too concerned at that moment because my sickness seemed pretty mild. It didn't hold me back while caving hard in the cold all day.

When I got to the surface it was cool out and lightly raining, so I hiked back in my cave suit, which saved some trouble over changing out of my cave clothes as the conditions mandated I do the previous day. I was in bed just before midnight.

Wednesday, July 10

We woke up this morning eager to return to where we ran out of rope the day before: at a wide open pit dropping down into a tall canyon that meandered around a corner out of sight. All night that canyon was calling our names, beckoning us to follow it around the next corner and see where it would go. Reilly had a job interview (via Zoom) that morning, so the plan was for me and Christian to head into the cave in the morning without Reilly, and she would join us when her interview was done. Thankfully she could get cell service by scrambling up the hill to a high point above camp and above all the caves. Quite the place to conduct a job interview—atop a rocky hill overlooking the ocean in the backcountry of an Alaskan island.

Reilly left camp early to get to her high point while me and Christian took our time, only getting out of camp at 11 am and into the cave around noon. Part of slowed us down was that is started raining that morning, so we were naturally reluctant to leave our comfy tarp city that was the camp kitchen and get out into the rain. We all hoped that Reilly was handling her interview OK while sitting out in the rain (I later learned that she got the job, so she did in fact handle it OK!). We stopped before the lead where we turned around yesterday to fix the substandard rigging we did the day before after running out of bolts. This took longer than I thought due to some awkward, muddy, precarious positions I had to drill from, but after a few hours I made the rigging on all the new drops nice and freehanging. After making it to our deep point from the previous day, I forged ahead with the drill and rigging supplies while Christian and Reilly (who had made it into the cave and caught up with us) continued the survey behind me.

One more drop from the ledge where we turned around the day before brought us to a nice walking canyon, 4-5 ft wide and taller than you could see. The floor was a mix of bedrock and small breakdown that made for nice pleasant walking. One nuisance drop in this canyon and a short bit of walking brought me to the next substantial drop, at which the cave changed significantly and made the rigging more complicated and tricky.

At this drop (F12), the cave walls picked up a thick mud layer. I was able to scrape some mud off one of the walls right before the pit and get in 2 good bolts for a lead in line. But all the rock after the lip of the pit was terrible—I would hit it and it would either reverberate with that bongo-drum sound that tells you to stay away, or crumble under the force of the hammer. The lip was smooth enough that we probably could have IRT'd this drop without damaging the rope; but there was quite a substantial stream at this point that made a waterfall down the pit, meaning you definitely could not just throw the rope and rappel down without getting yourself dangerously soaked in 38 degree water. So I absolutely needed to find good rock for a rebelay to keep the rope (and caver) out of the waterfall.

After testing all of the rock remotely near the lip of the pit, I found a patch of good rock awkwardly off to the side that I could get 2 good bolts in. This rebelay was far enough away from the water, but it would not make the rope freehanging. Thankfully we still had some beefy 11 mm rope we could use for this non-freehanging drop. It was quite awkward and strenuous drilling way out to the side of the pit entry, and required my putting in a traverse line to access the anchor from back before the lip, but I got it done and it allowed us to get down this drop without getting soaked.

Probably this explanation of the strange geometry of this pit makes no sense without pictures, but we were all either too busy (me) or cold due to waiting around (Christian and Reilly) to bother to get good pictures of this pit. Here's one we got of Reilly trying not to freeze while I worked on this tricky drop.

After that awkward section you land on a breakdown ledge where you need another rebelay to make it down to flat ground where you can get off rope. Thankfully here there was an obvious spot with good rock for a rebelay that would make the rope freehanging and keep you out of the water (although the waterfall was lightly spraying me at this point, which was totally unavoidable). But before I could get in that rebelay, I had to trundle all the large and clearly unstable breakdown on that ledge. These blocks could easily be dislodged by someone swinging around on rope into one of them, or dragging a tethered pack across them, so they definitely had to go before anyone went any lower. This took a long time as I had to lift and move some pretty heavy rocks all by myself, sometimes while awkwardly hanging on rope. Christian and Reilly continued to shiver while I worked alone on this ledge. While I was trundling, something slightly scary happened I was standing on part of the ledge, hauling around various medium-sized rocks and chucking them off the ledge, while I was attached to the rope with some slack via a tied-off descender. All of a sudden the entire floor of the section of the ledge I was standing on (an area about 3ft by 3ft) collapsed and fell beneath me. I fell onto my tied-off descender, not far at all because I only had a few feet of slack, but it was still spooky. As I fell I swung around on the rope and got under the waterfall for a bit, getting pretty soaked. I stopped trundling a bit to calm down and collect my thoughts, as the unexpected (but very short and totally safe) fall had startled me. After my heart rate went back down, I continued trundling until the ledge was clear of dangerous blocks.

One more rebelay at an overhanging face just past the end of the edge brought me down to flat ground where I could get off rope and call for Christian and Michael to come back down. They had been waiting around for quite a long time while I trundled and rigged that tricky pit, and they were very cold. By this point all of us were coughing up a storm; we had clearly caught whatever sickness Michael brought. Being moderately sick certainly didn't make sitting around getting cold and wet any more fun.

The bottom of this pit was a decent-sized room, about 30 ft diameter, without an obvious best way on. The water poured into breakdown in the floor that seemed impassable. The floor sloped up into walls with interesting-looking holes in them. I scrambled up into one of them that didn't go (plugged with mud). I started scrambling up to another good-looking hole in the wall, but didn't like the slick mud slope guarding it, so I retreated. It seemed like just an infeeder anyways. Eventually I realized that the best way on was a small hole in the mud floor of the room, where the floor and walls met. There was a small gap between the mud floor and bedrock wall at one spot, about 1 ft tall and 2 ft wide, so too small for anyone to fit through, but with a larger void space visible beyond. I tossed some rocks down the hole to see what they would do—although they didn't go far or make any nice echoes, I could tell that the void space beyond the constriction was promising and worth digging into. It would be an easy dig, since I would just have to trench out the mud floor that formed the bottom half of the constriction.

I took all the rigging gear off my and removed my harness to get setup for digging at that hole. I also ate some food to warm up and fuel up for digging effort I was about to put in. By the time I was ready to start digging, Christian and Reilly had gotten the survey down to this room and were looking around, assessing all the nooks and crannies in the walls and floor to see if I missed anything. They too concluded that the best way on was this easy-looking dig at the hole in the mud.

We had no digging implements on us, but there were plenty of flat rocks around to use as shovels. I got to work shoving the sharp point of one of those rocks into the hard but malleable enough mud, scraping away a few inches at a time until I could fit down the hole. Eventually I got to the point where I wanted to be on rope to continue digging, as I was stuffing my legs down the sloping mud hole without seeing what was below me beyond a dark void space. Reilly set 2 bolts in a boulder right before the dig, and I got on rope to trench out the last part of the mud chute before declaring it big enough. I then rappelled down the mud chute, which was now about 2 ft high and 3 ft wide—certainly big enough for all of us to fit through, but still awkward, especially when carrying bulky things like the drill and a cave bag full of rope.

The mud chute went down into a little standing room about 4 ft in diameter. At the bottom of this room was another small hole in the mud with the sound of water audible below. This must be the water that disappeared into the breakdown before this mud dig. I scraped mud away from the sides of the hole, still on rope at this point, until I hit bedrock on all sides of the hole. There was still no way I would fit. Without specialized digging gear (which we had no desire to haul down here, let alone use for the hours it would take to widen this constriction), there would be no getting through to the stream below us. It seemed that this was not the way on, and that we would have to push other holes in the walls in the large room before the dig.

Ascending the rope back up the tight mud chute was difficult, as the concave nature of the lower surface of the chute meant the rope gut sucked into the mud which made it difficult to slide ascenders up the rope and make upward progress. After a little bit of wiggling and squirming, I made it out of the mud chute and told Christian and Reilly the news. Christian went in to solo survey to the end of the cave. After he came back out, he announced that he had gotten beyond the tight spot! Apparently I missed an opening that went horizontally across the top of that small standing room with the too-tight hole in the mud floor. Christian solo surveyed into that opening, which immediately opened up into a pit that he shot partway down, and got to just over 800 ft of depth for the cave. That made Snowhole Cave officially the deepest cave in Alaska, surpassing Vive Silva Cave at 797 ft deep. Woohoo! We now had at least one big success of the expedition.

We would be back tomorrow to rig down that pit and continue exploration in whatever was waiting for us at the bottom. But now, we were all cold and tired (and coughing up a storm from probably-COVID) and ready to leave. We ditched all the gear in the room before the dig and exited with near-empty packs. I got out in a little under 2 hours from near 800 ft deep. At The Stripper and the other tight spot just above it, I noticed that there was way more water than there was when we entered this morning. It must have continued raining on the surface. I left all my warm layers—a fleece layer and balaclava—on while exiting, as I was soaked and never truly warmed up even while ascending rope. Back at camp, I tried to dry off my cave clothes with my body heat by continuing to wear them, but this just made me even colder without drying them out much. It was raining and the air was fully humid. I shivered under the kitchen tarp while I ate my dinner then retreated to bed at 12:40 am. I slept in my wet cave clothes (fleece layers minus cave suit), hoping that a whole night would be enough to dry them out.

Thursday, July 11

The next morning, all that water from my cave clothes had migrated to my down puffy and quilt. At least the clothes next to my skin were dry, so I was reasonably warm and comfortable during the night. Unfortunately, my down layers would be much harder to dry out—not exactly ideal gear to bring on a long backcountry trip in a cold Alaskan rainforest, but it was what I had on me, so I would have to make it work. When packing for Alaska, I also had to pack everything I needed for my subsequent caving trip to Tears of the Turtle Cave in the Bob Marshall Wilderness in Montana, as I would be flying straight from Ketchikan AK to Kalispell MT after this. The Bob Marshall Wilderness is very dry and warm this time of year, unlike Prince of Wales Island. I didn't want to bring heavier synthetic insulation into the Bob, and I certainly didn't want to pack 2 sets of insulative gear in my suitcase for Alaska and Tears of the Turtle, so I had to pick one set to bring, and I chose down. Oh well, I would just have to deal with the consequences of my choice to bring down, which lacks any insulative power if it gets wet and is difficult to dry out.

Thankfully we had some intermittent bouts of sunlight that morning, so I (and others who had much wet gear to deal with) spent some time in the morning hanging gear up where it would get sun and drying it out. Christian, Reilly and I were all feeling quite beat up, from our mystery illness and the many hours spent being wet and cold just as much from the caving itself, and we contemplated the idea of taking a rest day and returning with a vengeance on Friday. While we were imagining how nice it would be to spend a while day drying off our feet, sipping hot drinks under the kitchen tarp, and sunbathing for the occasional moment when the sun peeked out from behind the clouds, we got some info that would nix those plans. Anna, one of the US Forest Service employees, got word on the radio that a major storm was heading our way, and that it would be wise to be off the mountain by Friday night. The forest service folks on the radio kept using the term "atmospheric river" to describe the unusually severe rain event they were predicting. Because we had to pack up camp on Friday, we wouldn't be able to cave on Friday beyond possibly a few hours in the morning. So Thursday, today, would be our last full day of caving. We decided to hit the cave hard one last day today and give it a full day of exploration. We would have Friday morning to derig if needed, although we would try to derig today if we could so we would have all day Friday to pack up camp and hike down the mountain.

The others at camped hemmed and hawed for a while about whether they wanted to continue caving+ridgewalking that day, or hike out today, one day before the atmospheric river. None of the other cave leads had gone so well as our Snowhole lead, so the others had spent much time ridgewalking out in the rain all week. They were tired of being out in the rain all day every day and were ready to hike down. Even I was feeling a little beat up from the constant dampness—I was getting some classic feet issues, specifically some small open wounds on my feet that would not heal due to being constantly wet and in fact were getting worse—but I was certainly not about to hike down early given that we had a wide open lead at the bottom of what was now the deepest cave in Alaska calling our name. Amelia considered coming with us that day, but ultimately decided to hike down instead, alongside Anna, Alex Lyles, and Jared. We made sure to get a group picture with all of us before some of us started hiking down

We sure did enjoy our morning on the surface, slowly packing for the day without any particular rush, as it was just before 3 pm when we finally got in to the cave. We got down to the bottom of the cave uneventfully, and I went to work rigging the drop beyond the dig from the previous day. I was quite surprised that I missed the obvious hole horizontally across the top of the standing room with the too-tight bedrock constriction in the floor. A lesson I've learned too many times: don't get tunnel vision only trying to go down deeper in a cave! Looking up and can be fruitful.

Beyond that hole was a tube in the top of a keyhole-shaped canyon coated in mud. I stemmed out across the tube until the crack going down from the tube got to a comfortable width, and set a traverse line out to that point. Getting in the bolts for the traverse line was quite strenuous and time-consuming, as I had to power-stem into the slick slime coating all the walls, reach high up near the ceiling of the keyhole passage to find good rock not coated in slime, all while merely on rope with tons of slack to give myself room to move around and work. Eventually I got in some bolts that anchored the rope for a traverse line out to where it made sense to start rappelling down into the keyhole, and down I went. It was a low-angle fissure pit with a thick layer of mud coating both walls. As I rappelled down, I couldn't help but dislodge chunks of mud with my feet and dangling cave pack. I was worried about chunks of mud hitting me when Christian and Reilly made their way down. I did get one rebelay in partway down the pit that moved the rope a little to the side, but it could only do so much as the bottom of the pit was a small room without much space to get out of the way of falling mud. Oh well.

The pit was about 90 ft deep, and the bottom was a 4 ft by 10 ft room where the stream, that we lost as it went into breakdown at the bottom of the previous put, and that we could hear but not access beyond the mud dig, finally reunited with us. It poured out of the wall shortly before the bottom of the pit, where the rock changed from muddy to clean-washed, and followed the bedrock floor around a corner down a small passage that I reluctantly squirmed into. The passage went only a few body lengths before getting to a too-tight bedrock constriction, with bedrock walls, floor, and ceiling, and the stream continuing in that bedrock floor past the constriction out of sight. There was no air movement at this constriction. This was a hard ending to this passage, unless we wanted to come back with digging implements (we certainly didn't).

While Reilly and Christian were surveying towards me, I scrambled up a breakdown slope on another wall of the pit than where I came down, aiming for an interesting-looking hole in the wall to inspect. I got most of the way there but didn't like the steep loose breakdown guarding the last part of the climb into the hole, so I backed down and retreated to the floor at the bottom of the pit. I was definitely caving in "DNFU mode" at that time—I had heavily internalized that we were very far from any source of help beyond the 9 other cavers on this trip. Even getting their help would take several precious hours, as one would have to exit the cave (~2 hours), possibly wait several hours for people on the other teams to get done with their trips and back to camp, make them pack up supplies for more caving, then bring them back down to the bottom (another hour or 2). All while whomever was hurt was possibly succumbing to hypothermia or complications from whatever injury they got. If anyone needed more help than the handful of cavers on this trip, it would be at least a whole day (possibly 2) for cavers from the continental US and/or Canada to fly to Ketchikan, get to Prince of Wales Island, and hike up to camp then to the cave. There were no other cavers on Prince of Wales island, and certainly the local search and rescue squads would have no idea how to perform a cave rescue. As always in caving, but especially in these extremely remote circumstances, we took extreme care to be safe and self-reliant, and operated under the assumption that help might not arrive in time if anyone got hurt.

So I retreated from the climb up into the hole in the wall on the other side of the pit and yelled at Christian and Reilly, above me on rope, to look at that hole while rappelling down and swing over into it if they could. I also yelled at them to avoid dropping rocks if they could help it. Since there was nothing left for me to do, I started packing up my things to take out of the cave, then did squats to stay warm until they got down to me. Christian confirmed that the hole in the wall of the pit did not look like a good lead at all, then they both made it down to the bottom of the pit. I confirmed that it was now time to leave the cave, as we had no good leads going down, although we certainly had some good infeeder leads going sideways/up. I packed up as much as Christian and Reilly could give me, and bolted to the surface while the others finished up surveying the short horizontal passage at the very bottom of the cave. I had gotten quite cold waiting there, and was glad to be moving and generating lots of body heat caving with 2 packs.

I carried 2 packs out of the cave because Christian agreed to go last and stage-derig the cave as we left. Stage-derigging is partially derigging the ropes from the drops so they get (mostly) protected from damage during the high waters of the wet season, but not fully removing them from the cave. What you do is remove the ropes from all intermediate bolts in all the pits except the top bolts of all the pits, and coil the rope and leave them attached at the top of every pit. That way, they stay out of the way of raging waterfalls that go down the pits and can damage ropes when water levels are especially high, but you don't have to carry them all out of the cave. Even when you're not carrying them out of the cave, stage derigging is still quite a bit of work, and I volunteered to take 2 packs so Christian wouldn't have much to carry while he stage derigged. We also agreed that he should fully derig and carry out the rope from the landing at -328 ft upwards, as the entrance drop gets tons of debris falling into it from the surface and ropes left rigged there for the whole year are especially likely to get damaged.

Ascending up from that last pit was quite the chore, as the low angle of the wall meant the rope got sucked into it and jammed into the slimy mud. Both my ascenders quickly clogged with mud, and I had to manually actuate their cams with every movement up the rope. And push out from the wall before every movement to make space to push my ascenders up the rope without them being pinned against the wall. This significantly slows down the process of ascending rope and makes it take much more effort, but the bright side was it meant I very quickly warmed up and ceased shivering, as I had been while waiting around at the bottom of the cave. I figured that doing all this with 2 packs would be good training for Tears of the Turtle Cave in Montana, where I would go right after Alaska, as that cave is full of horrible mud that makes one's ascenders never work right, and there are always many heavy things to carry, although we usually stuff them in 1 big bag, not 2 medium ones.

I got out of the cave a little after 10 pm. Reilly was right behind me, although we had to wait maybe half an hour before Christian his way out of the cave, after which we declared victory for the week, with everyone out of the cave and a new depth record for Alaska achieved. The depth of Snowhole Cave with the new discoveries is now 911 ft. I carried 450 ft of 11 mm rope (what we used to rig to the landing at -328 ft) down to camp, awkwardly while the loose coil snagged on every tree branch, but feeling too victorious to be bothered by such trivialities. The mystery illness was hitting all 3 of us especially hard around this time, but again we were in too good of a mood to care. Everyone was asleep at camp when we got there, so we would have to wait til the next morning to tell everyone how our day went.

That night, I slept under the kitchen area tarp, instead of under my small tarp shelter. My reasoning was that it wouldn't get in anyone's way, because the next morning we were going to pack up camp and leave, so there would be no disruption from having to leave my camp setup in the kitchen. Plus I thought the larger, taller kitchen tarp would provide more convenience when getting in/out of my sleeping setup, as my personal tarp was small and thus low enough to the ground so that I couldn't stand up and change/pack up under my tarp. It was raining hard now, and I really wanted to be able to stand up around my sleeping setup without getting soaked by the rain. Well, it turned out that the kitchen tarp was totally inadequate to keep my dry while sleeping, as the heavy winds meant the rain would move sideways into the kitchen area and soak my even when I was completely under the tarp. Also, the ground there was waterlogged and muddy, and brown water from the kitchen area floor slowly worked its way onto my ground sheet over the course of the night. My down puffy and quilt were totally soaked at that point and provided very little insulation. I was quite cold that night, which was especially discouraging as I hadn't really warmed up since exiting the cave the night before.

Friday, July 12

The next day it was time to pack up camp and hike down. The "atmospheric river" was supposed to hit some time that evening. A forest service helicopter had hauled up most of the gear we had up at camp, and would haul most of it down, but we had to package everything up and store it properly so it would last the several weeks until the forest service could schedule helicopter time to take everything down. That meant packaging up all the food in dry bags and Darren Drums to seal in the scent away from bears, hanging it up in the best bear hang we could muster (which wasn't great—it would probably stop a bear that had other easy sources of food, but it certainly wouldn't stop a hungry bear), taking down the kitchen structure we had, and deciding what gear we wanted to take with us and hike down the mountain. That ended up being a lot more stuff than we thought. We made the mistake of not leaving any caving gear back down at the road, instead helicoptering everything up to camp, which was a silly mistake because we had to use some of that gear for the caving we were going to do the next week! In addition to all of our wet and muddy cave clothes and backpacking gear, we also hiked down a bunch of the group food, bolt hangers, quicklinks, and drill and batteries that had been helicoptered up there for us. Note for next time: if you're gonna cave somewhere else after the 1st week, don't helicopter all the caving supplies you'll need for the 2nd week up to your 1st week spot, else you'll have to hike them back down!

It was shortly before 2 pm by the time we were all packed up and ready to hike down. My pack was overflowing with heavy things that made the steep, loose, off-trail hike down especially unpleasant. The sickness was hitting me hard at this point, and I felt uncomfortably hot due to what I assumed was some combination of a fever, the exertion of carrying a way-too-heavy bag down uneven terrain, and the intermittent sunlight we were getting while I was dressed to be cold and wet all day. I kept thinking—isn't the atmospheric river supposed to hit us today? Why are we getting intermittent sun? When we got down to the road, we eventually heard that this atmospheric river forecast was for a totally different part of southeast Alaska, and wasn't even supposed to hit Prince of Wales Island at all. Anna got bad information from the forest service folks who simply told everyone southeast Alaska to go inside. So we lost a day of caving. Oh well, by that point we were all very beat up and sick and ready to be done caving for a day or several.

Another thing we determined when we got down to the road was how heavy my bag was. Hannah had a luggage scale with her. My bag turned out to be…wait for it…too heavy for her luggage scale! We had to use a bathroom scale that she keeps in her converted ambulance that is her house. My bag was exactly 90.0 pounds. Way too heavy! Reminds me of my previous post where I pull-tested my home-sewn caving harness and the strength was too high for the testing setup—it broke the steel carabiner connecting the harness to the break-test machine.

After a little bit of socializing at the El Cap Cave cabin with the cave guides, we departed to Christian's house. On the ride back to his house in Thorne Bay the sickness finally hit me real hard, and I got the chills real bad. As soon as I got to his house, I collapsed into the bed in his guest room and didn't feel like getting up for the rest of the day. Reilly took my temperature and it only read 99F, so I probably didn't have a proper fever, but was just feeling especially fatigued after working so hard many days in a row with a minor-to-moderate illness making it a bit worse. I then proceeded to sleep for 14 hours, from 6:30 pm to 8:30 am the next day.

Saturday, July 13

I awoke from my slumber feeling like a new man. Those 14 hours expunged the illness from me real good. Meanwhile, Christian and Reilly started feeling noticeably worse. John had a COVID rapid test on him, and said I could use it if I was curious what I had. I took it, and it turned out that yes I did have COVID, and so did Christian and Reilly too (and Michael, who seemed to have it earlier than all of us and was thus recovered by now). The others who were with us at Christian's house at this point, John and Amelia, didn't bother isolating from us, as we had already spent so much time together at close quarters.

Amelia and John were leaving early the next morning to take the ferry back to Ketchikan then fly home, so they both had to get packing. Reilly, Michael and I were staying another week, so we were in no rush for anything and were in relax, refuel, and COVID-recovery mode. One fun thing we did that day was take Christian's boat out on the water to set crab pots out in the water just outside Thorne Bay. Christian had eagerly told us about the boat he acquired then spent much time and effort fixing up to be water worthy, so we were all excited to see the boat and take it for a spin out on the ocean. We drove down to the Thorne Bay public boat launch, and once we were all in the boat ready to get out onto the water, the engine wouldn't start and kept stalling. Classic old fixer-upper boat problem! After some fiddling with the engine, Christian realized that the issue was a faulty connection on one of the fuel lines, and he got the connection working well and off we went into the bay. We enjoyed the scenic views of the bay surrounded by a heavily forested, uninhabited ridge on one side, and the small, quaint town of Thorne Bay on the other as he brought us out into the middle of the bay where he assured us he had caught crabs before. We loaded the pots with some fish heads then boated back to town.

We got back from boating and Christian made us a nice dinner of split pea and ham soup. It was our first proper meal that wasn't dinner mix since hiking up to basecamp on Monday morning. After dinner, Gooseberry stopped by to bring us some fish he had just caught. He hiked down a day earlier than the rest of us and spent the whole day fishing. He caught a ling cod (unrelated to standard cod) and gave us a nice fillet. After that, we went to nearby Sandy Beach to make a fire and hang out with everyone one last time before Amelia and John had to leave the next morning. When we were there we ran into several of Christian's friends from work at the Forest Service office in Thorne Bay. We chatted with them by the fire for a while. It was quite nice to casually get to know some of the Thorne Bay locals. Thorne Bay is a very different place than Albuquerque, and hanging out with some of the locals gave us a nice little glimpse into what day-to-day life is like in Thorne Bay.

Sunday, July 14

Amelia and John left early that morning for the ferry back to civilization, and it was now just Christian, Reilly, Michael, and me who were sticking around for another week of caving. Because Christian and Reilly felt even worse from their COVID (we assumed that everyone who got sick had COVID, as I had the same symptoms as everyone and tested positive for it), we decided to not get back to caving just yet, and instead spend the day ocean fishing on Christian's boat. Not exactly what I signed up for and was expecting when coming to Alaska, but a fun and welcome activity nonetheless. My feet were feeling pretty bad at that point, as several small open wounds I had on my feet, acquired either before the trip or during, were not healing and in fact had gotten much worse due to being cold and wet all week. Hiking everywhere in my wellies certainly didn't help that problem—next year, I'll definitely bring shoes for hiking around when not in the cave. Given the state of my feet, I was glad to not have to wear wellies nor be on my feet much for a few days while we fished instead of caved.

Us out-of-staters bought single day Alaska fishing licenses, then the 4 of us got in Christian's boat and went out into the ocean again. The boat started up without issue today, and we went to where we left the crab pots to check on them. There was only one small crab in one of the pots, too small to eat, so we let it go and put some more bait in the pots. We then boated our of the bay that Thorne Bay is in and into the wide open ocean. The 4 mile boat ride from Thorne Bay out into the open ocean between Prince of Wales Island and the Cleveland Peninsula on the mainland was beautiful. After we were a little ways out from Thorne Bay, all sign of civilization disappeared, and we were surrounded by steep ridges rising out of the water on both sides of the bay, that were covered in a thick, lush blanket of dark tall pine forests. A few tiny islands dotted the bay and provided minor obstacles to navigate around. Christian even let me drive the boat at one point. One of the tiny islands in the bay, Tolstoi island, had a group of seals hanging out on the rocky shore.

We boated around and tried fishing at several shallow spots near the mouth of the bay. The first thing anyone caught was a hunk of coral that I dragged up from the sea floor. After that, we all started catching tons of rockfish. Rockfish are apparently slightly endangered in this area, and only Alaska residents are allowed to keep them, and only 1 per day at that. We were all somewhat surprised that rockfish are threatened, because we all caught many of them. Maybe they're dumb and easy to catch and that's why they're threatened yet you always seem to catch plenty of them.

After switching fishing spots a few times, we finally found a spot where we caught plenty of fish that we could keep. Once we made it to this spot, Reilly, Michael and I started catching Pacific cod every few minutes. We caught 6 cod in the remainder of the afternoon and kept all of them as they were all a pretty good size. My total for the day was 3 cod, 3 rockfish (none of which I could keep), and one pretty hunk of coral. One of the cod I caught freed itself from the hook at the last second just as I was about to reel it into the boat, but I was able to grab it and snatch it out of the water before it could swim away. Probably it was in too much shock to swim away after being hooked and dragged up from the sea floor. I was particularly proud of that catch and definitely wanted to keep and eat that one.

One of the coolest sights we were treated to while catching cod at this spot was several bald eagles circling us, clearly looking to capture any of the fish we were bringing to the surface that they could probably smell and see. Once when we caught a rockfish and threw it back, the fish was stunned (or dead?) and hung out on the surface for a bit. A bald eagle saw it and swooped in trying to capture it. But it missed! We were surprised it missed given that the rockfish was sitting in the water immobile.The whole thing was a classic Alaska sight. Michael even got the whole thing on video. After the eagle left, the fish continued drifting out of sight on the surface of the water. I hope it came to and made it.

Once we got home with our haul of 6 cod, it was time to gut and fillet them. Christian knew how to do this of course, but figured it would be a fun learning experience to leave us to our own devices to figure it out on our own. I had done this once a few years ago with trout, but cod are anatomically different, and I didn't even remember how to gut trout anyways. Reilly and I pulled up a YouTube video and got to work while Michael watched—he could tell that we were both very eager to learn and practice this new skill, while he was pretty neutral about it, so he let us handle it. We watched the YouTube video a few times and figured it out without a hitch. Honestly, the hardest part was touching our phone to rewind the video while our hands were covered in fish guts. After we got all 6 cod gutted and filleted, we froze some of them and used the rest to cook a delicious dinner of cod and rice. Now we were finally eating like kings and queens! No more dinner mix for now.

Monday, July 15

The next day Christian and Reilly still felt pretty bad with COVID, so it was not yet time to go caving. More fishing! Instead of going out on the ocean and fishing for cod (with the chance to catch Halibut), today was a day to fish at a river and get some Salmon. Reilly's COVID was worse than the day before, so she decided to stay home and rest. Christian, Michael and I went to hatchery creek, which is set aside by the state for subsistence-only fishing. Christian has a rural subsistence fishing license, so he was allowed to fish there, and he was allowed to fish via dip netting, which is using a large handled net to scoop fish out of the water. Dip netting is much more effective than fishing with a rod in terrain like Hatchery Creek, where salmon running upstream are funneled into a few narrow areas where you can clearly see them and scoop them out of the water. The sockeye salmon were running at this time, and went we got to the creek I could immediately tell that this would be a good location and time to catch salmon as I saw multiple jumping out of the water.

This specific spot at Hatchery Creek has several short waterfalls that the salmon jump up as they swim upstream looking for a spot to mate. I didn't know that salmon could jump so high out of the water! It was quite impressive seeing them launch 6+ ft into the air, stick the landing into the top of the waterfall, and continue swimming upstream as soon as they were back in the water. Such gymnastic feats were nice to watch, but also useful as they provided a dip netting fisherman a signal that a salmon is coming their way. Just upstream of the little waterfalls here were narrow channels that the salmon get funneled into, so right when you see one jump you know to keep your eye on the channel and scoop with your net when you see a salmon swim through. Christian caught 6 sockeye salmon in about 2 hours using this method.

Michael and I were not allowed to fish via dip netting, as that method is for subsistence use only in Alaska, so we entertained ourselves by watching the salmon jump, watching Christian scoop them out of the water, and wandering around picking blueberries and salmonberries to eat. Both blueberries and salmonberries (like blackberries, but with a rich yellow or red color) were abundant here and now. The abundant berries and salmon made me think this is perfect bear habitat, and I'm sure that bears would perpetually hang out at this spot if humans didn't frequent it. In fact, we ran into 2 locals who were hunting bear there, and chatted with them for a while. They live in Coffman Cove, an even smaller, more isolated community than Thorne Bay on Prince of Wales Island.

Once we got back to Christian's house with another haul of 6 gorgeous fish, it was again time to gut and fillet them. Salmon are slightly anatomically different than cod, but the skills transferred and I gutted and filleted 5 salmon after Christian showed me how to do it using 1 of the 6. At Reilly's request, I also saved the roe (eggs) from the female salmon, as apparently it is a delicacy. Like the day before, we froze a bunch of the fish and kept some out to cook a delicious dinner.

Reilly prepared the roe with a half hour salt cure. I tried some when it was done curing and was a little put off by the intensely salty brine that gets released into your mouth when you chew and pop the eggs. The eggs have a fairly mild fish flavor, but the salt was a bit much for me. I figured they would be excellent to eat with rice to dilute the flavor a bit, rather than straight up. I cooked up 2 fillets from 1 fish for everyone, and we ate it with rice and roe. The roe was indeed excellent eaten with rice; it seasoned the rice and gave it a wonderful yet mild fishy flavor and aroma. I'm glad she made me save it from the salmon, as I wouldn't have thought to eat it, but now I know how delicious it is. Now we were eating salmon fillets and caviar, way more luxurious than dinner mix!

After dinner, I started a repairs to address some gear damage from the previous week. I needed my gear to last through some caving this week, but also the next 2 weeks at Tears of the Turtle Cave in Montana, a harsh, difficult cave notorious for destroying gear, among other things. I used some Shoo Goo to glue together the foam at my helmet where it cracked, and I started extending the homemade pocket for kneepad foam in my cave suit to make it better cover my knees. Again I was glad that I brought my speedy stitcher. I cut up some fabric form my rain pants, which were full of holes and not worth keeping at this point, to extend the internal pockets. I also extended the foam itself my cutting up my flip flops and attaching their foam sole to my kneepad foam with the speedy stitcher and some Gorilla Tape. We were planning on finally caving again the next day, so I needed to have functioning cave gear by the morning.

Tuesday, July 16

Today it was finally time to cave again (not that I minded the previous 2 days of fishing and, more importantly, eating lots of fish). I finished up my kneepad modifications in the morning before everyone else was up, and started packing for a day of caving. Reilly was still having pretty bad COVID symptoms so she decided to stay home again. Christian, Michael and I would be a team of 3. We would be recreating the Fast and Heavy team, as we called ourselves, from the previous year.

We had a lead in mind in Zina Cave in the Staney Creek area, which is a flatter, low-elevation karst area with more horizontal than vertical passage about an hour from Thorne Bay. Local caving legends Kevin and Carlene Allred explored Zina in the 2000s before moving away. Zina is a little over a mile of mostly horizontal passage, with some short vertical drops in a mazy area near the entrance that guards the large, walking stream passage that forms the main trunk of the cave. There was one lead left on the map marked "continues large and horizontal", that seemed like an infeeder to the large and horizontal trunk passage. It sure sounded promising!

We were on a fairly late lazy schedule and didn't get into the cave until 3 pm. The entrance is a pit in a large sinkhole that leads into an interesting and confusing mazy area full of walking and crawling tubes going in various different directions. Christian had been to the trunk passage before, so he navigated us through the confusing tubes to the drop into the trunk passage. This drop is a small hole in the floor of a walking tube that one could easily overlook if there weren't a rope bolted to the wall going down into the hole. This drop had several tiers with ledges between them, and there was bad rope rub on some of them that Christian wanted to fix, so he went first and took the drill to put rebelays in the drop and make it freehanging. After he announced that the drop was rigged properly for all of us, Michael and I followed him down to the water. I was eager to get moving at that point, as I had gotten pretty cold waiting for Christian to rig. Now I was getting a taste of my own medicine, waiting around in the cold and shivering while someone else went ahead and took a ton of time rigging.

At the bottom of the drop, we scrambled down to the floor of a tall walking canyon with a stream in it. The rock was polished and hard, and it seemed marble-like, although this could have been just due to fast-flowing water polishing the rock rather than metamorphosis of limestone into marble. We walked down this canyon as it slowly transitioned from narrow sideways-walking to spacious borehole. There was one tricky section where we had to find a way to downclimb a ~15 ft waterfall drop. And we did expend some effort tiptoeing around the spots with the deepest water, and stemming high above the water, to avoid flooding our wellies and getting our feet wet. Other than that, the travel to the lead was straightforward once we were in the trunk passage. It appeared as an obvious large passage on the right, above the stream level in one of the largest, most impressive sections of the trunk passage.

We went a bit down the passage to see what it did before getting into survey mode, and found that…the passage immediately died. Bummer. It turned into a too-tight wet crawl. We surveyed a few shots from the nearest station to the end of the passage, just to get something on the map, and got 60 ft of survey. Can't win 'em all.

We turned around and spent the trip back poking around in some of the areas that seemed a bit confusing on the map, that seemed like they had potential for leads. No luck. Once we were back in the entrance area tubes, I replaced a handline anchor that was one old rusty bolt with a new 2-bolt anchor. This handline wasn't exactly necessary, but if it failed while someone was holding onto it, they could easily fall backwards and bash their head, so if it was gonna be there it might as well be on a bomber anchor. Because we had no extra rope, and because the rope that was already there had no extra length to tie more knots, I had to be very efficient in my use of rope attaching it to a new 2-bolt anchor. I ended up doing more "meme rigging" as I called it, which meant feeding the rope directly through 2 vertically offset bolt hangers and using barrel knot as stopper knots to keep the rope in place. It makes more sense to explain with a picture:

We were out of the cave at the reasonable hour of 7:30 pm. Not the most successful day caving, but we still had fun as we got to see a pleasant, fun cave. There is very little large horizontal passage in caves on Prince of Wales Island like there is in Zina, so that was nice to see.

Wednesday, July 17

Today was supposed to rain hard all day, so we decided not to go caving. Most of the leads in the area that we knew about could become dangerous flood risks with heavy rain. This is in contrast to the caves up on El Cap Peak that we were at earlier in the week, which are more consistently vertical and therefore don't present a flood risk. I was somewhat thankful to have a day without caving to let my feet heal, as getting my feet wet and cold and leaving them that way for a whole day seemed to undo the progress of a day of healing them on the surface. We all slept in and had a luxurious breakfast of sockeye salmon before figuring out what to do with the day.

We decided that fishing would be a much more suitable activity for today. In the morning we walked around Slide Creek, a few minutes north of Thorne Bay on the way to Sandy Beach, to see if any salmon were running there. The Slide Creek area was lush and beautiful and pleasant to walk around, but we didn't see any fish jumping around, so we decided to fish at the more dependable Thorne River, even closer to town.

It was just Michael and I fishing today, as Reilly still felt pretty bad with COVID, and Christian had better areas to fish that were reserved for only in-state residents with subsistence licenses, so he stayed home and worked. There were visible fish jumping all over the river, although we had no luck catching any. I did get a few bites, but nothing I could hook and reel in. Both of us managed to get the lure stuck at least once each, and we had to cut some. When we were there we ran into another party of locals who were fishing, and they also had no luck. So probably it wasn't just us and our complete lack of fishing skill—the salmon weren't running much right now. It seemed that in the few days since we went to Hatchery Creek, the sockeye run died down a bit, and the coho run had yet to start. We only had one rod, so we switched off, and when I wasn't fishing I wandered around enjoying the lush green vegetation that I don't get to see living in the desert southwest, and I picked a bunch of berries. The salmon weren't exactly in right now, but the salmon berries were abundant and delicious. We fished for a few hours in the rain, enjoying the rain and the scenery, before heading back to Christian's house.

Thursday, July 18

Today was Michael and I's last day on Prince of Wales. on Friday I would travel to Kalispell, Montana to go on my next caving adventure: Tears of the Turtle Cave. We wanted to do a big day for our last day on the island. The rain had subsided, which meant we could go out on the ocean as there wouldn't be huge waves rocking Christian's tiny boat. So we decided to go to Kosciusko Island ("Kos"), a small island adjacent to Prince of Wales Island with even fewer people than Prince of Wales and just as good karst. We all went to Kos last year and explored many good leads there. On the last day of last year's expedition, I turned around at a great lead in Laser Eyes Cave, as Ellen Whittle named it that year. I had just dug open a pit entrance sucking tons of air in the floor of some horizontal passage at the bottom of the entrance pit of Laser Eyes. I rigged this pit and rappelled down until I ran out of rope. There I was dangling in space at the end of the rope, getting dripped on by a small waterfall, looking down to a ledge after which the pit continued out of sight.

That excellent, wide-open lead was one of the leads we had in mind on Kos, but we also had another lead that was less of a hike from the shore than Laser Eyes that Christian had found earlier. This lead was a surface pit that took a large waterfall and seemed to be heading in the direction of Cascade Cave, a resurgence cave near the shore of Kosciusko Island that ended in a sump. The plan was for us to push this lead, and if it ended (or if the water was too high making the waterfall pit dangerous), we would move on to Laser Eyes pit. This surface pit was under 500 ft elevation, so it wouldn't go super deep, but since we only had one day, we figured it would be good to focus on this pit since we could possibly connect it to Cascade Cave in the one day we had, which would be a cool find if we made it happen. Reilly's COVID still made her not feel like caving, so her and Christian would fish from the boat that day while Michael and I caved on Kos. With any luck, we would find some cave and Christian and Reilly would have some delicious fish waiting for us when we got back.

We drove back to El Cap, where we started the trip nearly 2 weeks ago, as the boat launch there was just a half mile from the spot on the shore of Kos where we would begin hiking. We got to the boat launch at 12:30. When we boated across the straight to Kos, we were surprised to see a group of people on the shore near Cascade Cave, the resurgence cave that we suspected our lead might connect to. Hanna and Ian, who were with us on El Cap Peak for the first week of this trip, were there with several of their cave diver friends from Anchorage. We knew that this week, they were running a dive trip to dive a sump in El Cap Cave, just on the other side of the straight on Prince of Wales Island. Apparently the heavy rain the day before meant no more diving there, as that cave can get fast moving water and would be dangerous to dive (and dangerous to do normal caving in, to get diving gear to the sump) during/after heavy rain. So instead they were scoping out the sump in Cascade Cave to see if it would be worth a dive. What a crazy coincidence to run into them here! We told them what we were up to and that it would be legendary if we ended up running into them on the other side of a sump that their divers got to from the other side. They told us that couldn't quite happen today, as today they were only scoping out the sump for dive potential (apparently it was promising), not diving it.

Michael and I got off the boat around 1 pm and started the hike up to the lead. We had a GPS track to the lead, and once we were there, at a pair of large (40 ft diameter) sinkholes, we spent quite a while wandering around the pair of sinkholes looking around for a surface pit. Everything seemed to be plugged with vegetation and breakdown. There was a stream flowing into the upper sinkhole that disappeared into a crack in the bottom of the sinkhole. While this crack seemed marginally large enough for a human to pass through, it certainly wouldn't be safe to do so with the large amount of water currently flowing down the crack. Even if you didn't drown, it would be difficult to get back up the crack with the raging water pushing you back down. We hoped that wasn't the lead that Christian described, as we certainly wouldn't be going in it in these conditions if so.

We spent maybe half an hour wandering around both sinkholes, checking every nook and cranny and fighting with the dense vegetation, and were considering continuing on to Laser Eyes Cave when I finally found a way in to a cave there. I crawled under some large boulders at the bottom of the lower sinkhole into a small space that didn't look like a cave entrance at all from above but quickly opened up into walking passage. After walking down it a bit until I could hear the sound of a raging waterfall that was likely the same water that went down the crack in the upper sinkholes, I left and found Michael (who was still looking around in the upper sinkhole) to tell him the good news. We wouldn't have to hike the extra mile with 400 ft of elevation gain to Laser Eyes; we could start surveying at this alternate entrance instead. When I brought Michael back to the lower sinkhole, I had trouble finding the entrance again, as the hole between breakdown blocks was quite obscure. Once I found it, we we started surveying, eager to see what we would find when we met up with that loud waterfall I heard.

It was 2:30 pm when we got in the cave and started surveying; we told Christian and Reilly we would meet them at the shore at 6 pm so this did not leave us tons of time for surveying. In a few shots we got to a black dike that sloped down and made a slick wet slope. I placed 2 bolts for a handline that wasn't exactly necessary, but useful to make the slope casual. Below this slope was a very interesting breakdown floor: the breakdown from the jet black dike rock was coated in white calcite. It seemed that the floor got sprayed on sometimes from a faint waterfall in the ceiling (not the louder waterfall that I could hear in the distance and that we were getting tantalizingly close to), and when the spray dried, it precipitated out a white calcite coating. It was beautiful and unique; I had seen white calcite coatings like this before (see for example the Snowy River passage in Fort Stanton Cave, or Silent River Cave in the Grand Canyon for examples with tons of pretty pictures), but never on a jagged breakdown floor like this.

Beyond this room was a small downclimb into a crawling passage that led into a pit where we finally gained the raging waterfall we could hear from the entrance. The frothing white jet of water coming down looked like the stream from a power washer blown up 100 x. It roared like a jet engine, so loud that I had trouble communicating with Michael even when he was right next to me. The crawling passage that intercepted it was just a window in the wall of the pit, as the waterfall came down from way above and continued down below us. Once we got the survey to the lip, I went back into rigging mode and started thinking about how I could get the rope rigged in a way that would keep it as far away from the water as possible. The waterfall was on the left side, so I looked to the far right wall of the pit. Once I got below the lip, the rock on the right wall and below my feet turned into that same jet-black, non-carbonate rock that we saw at the dike earlier in the cave. I had to spend quite a bit of time trundling at the lip, as the black rock was fairly loose. The only good spot for a rebelay ended up being in that strange black rock layer. I was a little nervous about setting bolts in that unfamiliar rock, but it was clearly solidly attached to the cave walls and made a good sound, so I set a 2 bolt rebelay in it, right under a roof that provided a little bit of shelter from the spray of the waterfall, and went down about 50 ft to the bottom of the pit.

The bottom of the pit was a 30 ft by 50 ft room with a tall ceiling. Once I got down there I quickly got off rope, called for Michael to come down, and started looking around. It was just about our turnaround time now, so we wouldn't get the survey down to here, but rather we would just look around and make sure the cave kept going before telling Christian and Reilly they should come back. All the water from the waterfall seemed to end up in a pool in the floor and sink there; could this mean the cave ended in a sump here? Thankfully after poking around the walls I found a way on: I scrambled up a breakdown pile at the far end of the room and found a pit at the top of the pile, continuing about 40 ft down into more large passage below. A great lead! But not one Michael and I would be returning to, as this was our last day here. Christian and Reilly would have to return as soon as their COVID symptoms allowed.

As soon as Michael touched down I started heading out of the cave. We were going to be close to the 6 pm agreed upon time that was the latest we should be back at the shore. We derigged the ropes and made it out of the cave at 5:15 pm. We packed our bags as fast as we could, and speedily hiked down to the shore. I made sure to enjoy the pleasant lush green rainforest one last time on this hike down, as I wouldn't be immersed in this kind of terrain again for a while. We made it to shore right before 6, just on time. At the shore we were rewarded with pleasant views of the clouds partially enveloping the rolling green hills of Prince of Wales Island across the sea from us, and with the sight and sound of a sea otter swimming past us near the shore. We sent an inReach message to Christian telling him we were ready for pickup, and before long we heard the boat in the distance then saw it round a corner from the west.

When we met up with Christian and Reilly we told them about all about our finds, and that they should definitely return to this place as soon as they could. They told us about thee day they spent fishing and boating around the various straights and channels in the nearby islands, without much fishing luck but with amazing views. They only caught some rockfish, but Christian kept one so we would be able to try one. On the half-mile ride back to the El Cap boat launch, the boat ran out of gas! Thankfully we had oars in the boat that we could use to get us the rest of the way there. No problem; if anything it meant a little more time to hang out on the ocean and enjoy the scenery. I savored the foggy, lush green hills and calm ocean around me as we gently oared ourselves back to shore.

We named the cave Hi-Contrast Cave, after the jet black rocks coated in white calcite that gave the cave such an interesting color scheme in that spot. Reilly and Christian ended up returning to it that weekend, as Reilly's COVID symptoms had finally abated by then. The cave went for several more drops and had some nice large horizontal canyon passage before ending in a sump. It's not yet at the same elevation as the sump in Cascade Cave. So if Hannah and Ian and their diving friends ever end up diving around here, they should dive this upper sump, as it will be easier to follow whatever passage lies beyond the sump downwards to the Cascade Cave sump, rather than follow some passage upwards from Cascade Cave.

We ate the rockfish for dinner that night. It has a delicious and distinctive flavor and meaty texture; I'm a fan. Apparently rockfish have not that much meat on them and it's difficult to remove it from the skeleton, so people rarely keep and eat them. But I'm glad that Christian kept one and gutted+filleted it for us so we could try it, as it was wonderful. That night Michael and I packed our bags, as we would have to leave very early the next morning to catch the ferry to Ketchikan.

Friday, July 19

We woke up at 5:30 am in order to make it to the ferry terminal in Hollis by 7 am. Michael and I boarded the ferry to Ketchikan; Reilly stayed with Christian on Prince of Wales for another week. Michael was flying back home to Nashville, while I was flying to Kalispell, Montana to participate in the Tears of the Turtle Cave expedition. To hear about that you'll have to read my next post when it comes out!

I use a homemade harness for caving, rather than any commercially available one. This tends to alarm some people, as one's harness is a critical, single-point-of-failure piece of life support equipment. It tends to alarm more people from the rock climbing world than from the caving world, as caving has much more of a DIY/homemade gear culture due to its niche nature, with fewer large gear manufacturers mass-producing any and all gear that a caver could ever want.

The other week, back in May, I had the chance to pull my harness to failure and determine its breaking strength. AKS Tower Supply, a local business selling rope access/work-at-height equipment in Albuquerque, has one of these break testing machines, and opened it up to the public for one night for members of the New Mexico Mountain Club to come and break test all of their old, worn out, or otherwise questionable gear. I was super eager to test a bunch of my old and worn out caving gear, especially the homemade gear, altho there were a lot of people with stuff to break so I only had one chance to break something. I chose one of my homemade harnesses, which was 1 year old but heavily used.

It broke at...drumroll please...actually, it didn't break! At 30.52 kiloNewtons, or 6861 pounds, the steel carabiner connecting to to the break-test machine broke instead of the harness! That was a very pleasant surprise.

The line scale after the break test, connected to the semi-broken harness.

As the steel carabiner snapped, it cut both the leg loops of the harness, and damaged the waist loop without fully severing it. So the harness was not even fully broken (altho I certainly won't be using it ever again!)

The harness after the pull test. The leg loops got severed by the sharp carabiner as it snapped, and the waist loop was damaged but not fully severed.

For context, a force of 7 kN in when falls start to become dangerous. Much higher than that and there's a real chance of breaking your pelvis or other bones as the harness crushes you. So 30.5 kN is way overkill. As it should be—harnesses should be stronger than the human bodies that they protect, because harnesses can and do wear out over time and become weaker. (So do humans, I guess...). The UIAA standard for harnesses specifies that harnesses should hold 15 kN. The standard for connectors like slings and quickdraws specifies 22 kN—higher because the rope running from the belayer to the gear then down to the fallen climber creates a 2:1 pulley system, altho with friction between rope and the carabiner it tends to be more like 1.5:1 in practice. So my harness tested well above both of those standards.

The broken steel carabiner from the test.

I break tested it by pulling, on one side, both harness attachment points, and on the other side, the waist loop and both leg loops, approximately equalized. This is somewhat similar to how a harness is actually loaded in a fall—the attachment points are connected to the rope, and your bodyweight is approximately equally distributed among the waist and leg loops. The steel carabiner actually broke below its strength rating of 50 kN, probably for 2 reasons: (1) the 1.5" webbing didn't sit super well in the bottom of the carabiner, and was loading it somewaht close to the nose, a configuration where carabiners are weaker; and (2) this carabiner had been used for many break tests and may have experienced some metal fatigue.

I've made a few of these harnesses for some friends of mine, and they were all super psyched to hear the break test numbers, as they were trusting their lives to my homemade harnesses without them having any official strength ratings besides my assurances that they were "super good enough". I use a homemade harness, rather than a commercial one, not primarily for cost reasons, but because I like my homemade harnesses better. My harnesses are a fixed size custom fit to myself, with no adjustment buckles. This means less snaggy things sticking out of the harness to get stuck, and thus more durability, as the webbing that sticks out where the buckle is tends to wear out first due to it constantly rubbing against the cave walls. No metal buckles nor extra webbing for adjusting the fit also means the harness is lighter and packs down smaller. I also put extra gear loops on my harness where I want them. No commercially available harness has these features (and it would be hard for a manufacturer to pull them off, because the lack of adjustment buckles means the harness has to be custom-fit), so instead I just make my own.

I use a standard doestic sewing machine (Brother LX2500) that I found at a thrift store. It sews thru 3 layers of seatbelt webbing with a #18 needle with no problem. The webbing I use is 1.5" polyester webbing from Strapworks, with a strength rating of 4800 pounds (the harness has multiple strends sharing the load, so it makes sense that it broke way higher than that). I use size 69/Tex 70 bonded nylon thread with breaking strength 11 lb. Each harness attachment point (there are 2 of them) gets approximately 300 stitches. Multiplyingh 11*300*1.5 gives 4950 pounds per attachment point (the factor of 1.5 comes from the fact that each stitch has 2 strands, but they tend to not be full-strength because the thread has a sharp bend at the stitch, and 1.5 is a common fudge factor). So 9900 pounds, or 44 kN, total strength for pulling at both attachment points, theoretically. I don't think I'll ever break test a harness with a machine that can go that high, so that number will forever remain a theoretical prediction. And lately I've been using size #92/Tex 90 nylon thread instead, with a 15 pound breaking strength, as my sewing machine has no trouble with that thick thread. So the strength of the harnesses as I make them now may be even higher.

Close-up of the harness attachment points stitching after the test. The stitching looks unaffected.

Huge thanks to AKS Tower Supply and the New Mexico Mountain Club for putting on this event! It was tons of fun and very informative. We also broke a bunch of old, sun-damaged webbing collected from various rappel routes in the Sandias that night, and it was alarming to see how weak some of it was. Also—AKS tower supply stocks some caving gear, such as Petzl Simples, Basics, and other essentials. It's pretty rare for any in-person store in the US to stock caving gear. So if you ever need to buy some caving equipment in Albuquerque, check them out!

Many people are surprised to hear that the Grand Canyon has caves, since people associate the towering red cliffs of the Grand Canyon with sandstone, not cave-forming limestone. But in fact one of the major cliff-forming rock layers of the Grand Canyon is the Redwall Limestone, which is naturally grey but is stained red by layers of red shale and sandstone above it. The Redwall Limestone is a great cave-forming rock—it is from the Mississipian era, the same geologic era as other cave-forming limestones in the western US such as the Madison Limestone and Pahasapa Limestone.

I met up with Stan Allison in Farmington, 3 hours northwest of Albuquerque, the night before the trip, and we drove together to the Grand Canyon the next morning. We met up with everyone around midday, spent an hour or so packing our bags and picking out group gear to carry, and set off for our campsite above Silent River Cave early that afternoon. The hike in to our campsite above the cave was typical of Grand Canyon caving approaches: beautiful, wild, off-trail, and thus somewhat rough with plenty of steep/loose terrain and bushwhacking. We started descending from the north rim area around 2pm and were quickly rewarded with wonderful views of the canyon when we got to the rim itself, which was a short distance from the road.

Jason and me at our first view of the Grand Canyon on our approach to the cave, at the rim.

The view from the rim. The large red cliffs are Redwall Limestone, which is the great cave-forming rock layer in the Grand Canyon. From here we descended into that thick brush.

It was only 1 mile from the rim to camp, but it took over 3 hours (!!!) since there was not trail and we had many obstacles along the way: bushwhacking through dense vegetation with plenty of thornbushes, carefully negotiating some steep gullies where missteps could send loose rocks tumbling down the gully onto people below, and tricky climbs down small cliff bands where we lowered our packs to make the climbing easier. We got to camp in the late afternoon with plenty of daylight left to set up camp and collect water from the cave. The campsite is not the best: it's a dry wash in smooth, water-polished bedrock, scarce on flat space for sleeping and cooking/hanging out. The drainage floods if it rains a little, so we made sure to check the weather right before leaving and verify that there was no chance of rain before setting up camp there. Our campsite made up for its less-than-ideal topography with its amazing views and surreal location within the depths of the Grand Canyon right above a huge Redwall Limestone cliff, and right above the cave. When hanging out at camp we could see towering red cliffs all around and above us. And the dry wash, despite presenting a flood risk and not having a ton of flat ground, did have the perk that the smooth, water-polished bedrock of the wash was very pleasant to walk around on barefoot.

There were several small patches of snow in shady nooks on the approach. These provided a nice cold respite from the heat.

Most of us set up our campsites in the wash, set back from the lip a bit. Sam was the boldest and set up her campsite right at the edge of the cliff! Certainly her boldness rewarded her with the best setting and views when going to sleep and waking up in the morning. For my camping setup I brought (at Jason's suggestion) my single-point-suspension portaledge hammock. This portaledge-hammock hybrid allows you to rig a hammock-like sleeping setup from a single point such as one tree (unlike hammock which require 2 rigging points with just the right geometry). The inflatable sleeping pad you embed in the bottom results in a much flatter sleeping surface than a hammock, although it's not quite as flat as a portaledge. Still, it's an order of magnitude lighter and smaller than a traditional rigid-framed portaledge, which means you can feasibly carry it with you when backpacking, even when carrying a bunch of other heavy things like caving gear. The first night I setup my portaledge-hammock in a tree just outside the wash.

Sam's camp setup right at the edge of the Redwall Limestone cliff. The rock in the dry wash where we setup camp is grey, not red like most of the other Redwall Limestone, because occasional floods wash away the red stains that come from other rock layers above and color the rest of the Redwall.

From our camp in the dry wash, you access the cave by going into the forest outside the wash a bit, to a large cluster of bushes where we rigged our rope. From there you rappel thru the steep forested slope to the top of the large Redwall Limestone cliff right below our camp. The cave entrance is a huge hole in the cliff shortly below its top that we rappel into. The first night we went into the cave to collect water for that night and the next morning. There are some drips in the ceiling a few minutes into the cave that are the only water source around, as the surface is totally dry. By the time we had gotten our camp setup and collected water from the cave, it was about 7pm, time to make dinner (which for me for every meal was my homemade Dinner Mix loosely based off the Cheve Dinner Mix) then start winding down for the night.

Jason rappelling into the entrance of Silent River Cave.

Our plan was to do long trips to leads near the back of the cave (about 3 hours travel time) on our 1st and 3rd caving days, and a short trip to some leads closer to the entrance on the 2nd day of caving. The 1st caving day, Saturday the 25th, Stan Allison took Hailey Galit and me out to a lead he had found and started surveying on his previous trip to Silent River, 2 years ago. It took 3.5 hours to get to the lead, although that was with plenty of stops for routefinding, regrouping, seeing some nice formations in side passages, etc. Later we would make it out to that lead and back much faster. The travel through Silent River Cave was sporty and challenging, in a fun way. There was lots of climbing, stemming, and a little bit of crawling, with plenty of mud and loose rock around to keep you on your toes. Very unlike typical Grand Canyon caving, which is stereotyped as lots of large easy borehole although with heinous off-trail approach hikes. One way that Silent River Cave is like other Grand Canyon caves is that it is very decorated. On the way in to the lead we passed fantastic gypsum needles, and a gypsum rope formation, which I didn't even know was a thing before seeing it here.

A gypsum rope formation. That white rope-like material not a manmade textile, but gypsum rock.

More wild gypsum formations.

Some nice aragonite and popcorn decorations in the main passage.

One annoying part of the travel out to the lead is that it required us to get our feet wet when travelling thru a wide canyon with ankle-deep pools. I did not think I was going to get my feet wet on this trip, and so I only brought cotton socks—oops! Caving with wet feet all day was slightly demoralizing, although I did manage to dry my socks out the night after that trip by sleeping with them.

An important intersection where all 8 of us stopped and split up into separate teams on the first day.

Our lead was at the current known end of the E survey, a passage which Stan and his team started exploring on their previous trip to Silent River. After travelling thru E1-33, we stopped at the limit of exploration E33 to get setup for surveying. E1-33 was small walking passage, with minor obstacles such as short crawls, some stemming, and narrow sideways-walking. The passage continued beyond E33 as larger walking passage, although it quickly shrunk to breakdown crawling. After a few stations we popped through the breakdown into a very different character passage and I was delighted to lay my eyes upon what lie beyond the breakdown. Not only did the passage change character to a larger, stand-up walking canyon, but this canyon also had a calcified crystal formation on the floor following the flat streambed, not unlike the Snowy River formation in Fort Stanton Cave that I just blogged about. This Snowy-like formation had a light brown color, not quite the creamy white of Snowy River proper, but it was unmistakably morphologically the same as it had the same cauliflower texture and flat flooded streambed origin as Snowy River.

The Snowy-River-like formation where we first found it, where it was light brown.

Something interesting happened at a (dry) eddy pool in the canyon. In the eddy area, the Snowy formation abruptly changed to pearly white, in contrast to the light brown color in the main portion of the dry streambed. I think this is because the light brown color is caused by dirt, and dirt particles are too heavy/have too much momentum to get sucked into the eddy pool when this passage flows with water.

Now that we were at the floor of the canyon passage that was the E survey, not higher up above breakdown, the travel and survey became much more pleasant. The passage was mostly upright walking, 2-4 ft wide. We eagerly surveyed ahead, following the Snowy River passage in the floor the whole time. To our excitement, the formation got thicker, wider, and more pearly white in color as we continued down our E survey. Several times the path of least resistance in the passage brought us to an upper level above Snowy, but it was never far from us, and the travel stayed relatively easy the whole time.

Part of the passage where Snowy goes under a low roof and the path of least resistance goes above Snowy (but not far). Note the series of small rimstone dams in the passage.

Here you can see the cauliflower texture of the Snowy formation, very similar to the texture of the actual Snowy River in Fort Stanton.

When it was getting time to turn around to head back to the entrance and to camp, we went a few stations further so we doubled the number of stations in this passage (we went from E33 to E66). This worked out well because E66 was at a pleasant, large room, about 15 ft in diameter, and would be a good spot to unpack and restart the survey when we returned in 2 days. At this room the passage and the Snowy formation visibly continued off into the unknown. We surveyed 583 ft that day in a trip that was a little under 13 hours.

We got back to camp a little after 9pm and told everyone about our exciting find and good amount of survey footage. They were all happy to hear that our lead was going quite well, as the other teams didn't have any huge finds. Rachel and JJ were still in the middle of their aid climb, and Jason and Sam's pit leads at the back of the cave didn't go very far.

Rachel in the middle of the aid climb that she and JJ worked on that weekend.

That night, Sam had the excellent idea of using my portaledge hammock to bivvy on the rim of the cliff right below camp, instead of in an unassuming tree set back from the edge in the forest. How epic would it be to fall asleep in a portaledge hammock rigged to our rope going over the Redwall Limestone, and to wake up hanging on a sheer cliff with nothing but the walls of the Grand Canyon around me in all directions? It would be a bit of an ordeal getting set up to bivvy at the rim, especially at this late hour of the night, as the portaledge hammock is pretty awkward to get into without a good ledge to stand on and use to move yourself into position. Even when it's rigged right above flat ground, it's still a bit awkward to get into the hammock and into your sleeping bag. Still, I knew that a few minutes of awkward struggling getting into the hammock while hanging from a rope on a sheer vertical cliff would be worth it for such a legendary bivvy. Once I got hammock rigged to the right spot on the rope just below the rim, I used my ascenders to carefully and awkwardly down-ascend into the hammock. Once I was comfortably sitting in it, I switched from being attached to the rope via my ascenders to being attached to a knot in the rope via my cowstails, which gave me more slack and room to toss and turn. With some difficulty I got myself wrapped up in my sleeping bag and went off to bed. I slept with my vertical gear (harness and such) on, and didn't even bother taking my shoes off (I just loosened the laces) because I didn't feel like doing the awkward free-hanging situp in the hammock to fully grab them and take them off.

When I got into my hammock and fell asleep, it was pitch black out, which made waking up the next morning all the more dramatic and surreal. I was dangling in a little fabric cocoon by a few thin straps that held the hammock to the rope, with a sheer vertical wall of the Grand Canyon pressed up against one side of me, a magnificent view of more of the canyon meandering off into the distance on the other side of me, and a long way to the ground beneath me. I basked in the glory of my bivvy for as long as I could, watching the sun-shade line creep down lower on the walls above me as the sun rose higher in the sky, until I couldn't wait any longer to get up because I had to pee. Getting out of the hammock onto the rope was much easier than the other direction the night before. I carefully packed up my hammock while dangling on the rope, making sure not to drop my sleeping bag which wasn't really attached to anything as I got out of it, then climbed up over the rim and walked over to surface camp, eager to tell everyone how marvelous my night on the rim was.

My bivvy at the top of the Redwall Limestone. The large entrance to Silent River Cave is below.

The next day at camp I realized that while hiking my hammock setup thru the brush to the rope we had rigged off a tree, I had popped my Thermarest inflatable sleeping pad. I thought I had the sleeping pad wrapped up and well protected by the hammock when I dragged it thru the brush, but I guess not well protected enough. Thankfully I brought some Gorilla tape with me in case this happened, and I was able to patch it. Tape, especially sturdy impermeable tape like duct tape or Gorilla tape, is a great emergency repair material to bring with you when backpacking, and I always do.

That next day was slated to be an easier, shorter day of caving, in between our planned longer days of caving at leads near the back of the cave. I personally was eager to return to our E survey, but also didn't mind a shorter day as the previous day did have quite a bit of commuting. I went with Jason, Hailey, and Robin to check out some high leads within half an hour or so of the entrance. The first objective was to ascend the rope on an aid climb that Tommy Cleckner did on a previous trip, check if it goes, and pull the rope down if not. Jason was pretty sure it didn't go but wanted a second look before pulling the rope down. When I got up there, I did get a little further than Tommy did by trenching out the dirt floor of a too-tight squeeze, but the small amount of passage ended in breakdown shortly after, so I decided to pull the rope. I set up a pull-down rappel, came down, and started pulling the rope. It pulled down partway only to get stuck after the other end was out of reach! Both me and Jason bounced on it with our combined bodyweight and we could not get it to move any further. I could see what was stuck above us: it seems that the other end of the rope got twisted and tangled around the end we were pulling. After some time spend struggling with the rope to no avail, we declared that retrieving the rope was a lost cause and decided that someone would have to come back with aid climbing gear to climb partway up to the tangle in order to untangle it and retrieve the rope. Not a great start to the day so far: we were down one rope.

After that, the 4 of us went to some other high leads around the cave that looked free-climbable. We were looking to see where the air in the cave went. At the entrance the cave sucks air and it's very strong, although the air disappears shortly into the cave. We wanted to check out a few possible leads at a complex junction area just past where the air disappears, in hopes of finding the air and thus hopefully much more passage. Me and Jason chimneyed up into the ceiling of the passage in a few interesting-looking spots, but could not find any free-climbable way into new passage. I also checked a tight hole in the breakdown in the floor of the passage at one spot, but this didn't go very far. We then exited the cave, feeling mildly defeated after losing one rope and failing to get anywhere while poking around in the ceiling and floor.

That night everyone got out of the cave fairly early (a little after 4pm for my team), with plenty of daylight left (except Rachel and JJ, who had a longer day working on their aid climb). When we got out, Sam surprised us all with some fun little luxuries she packed in for us to enjoy during our one shorter day with liesure time on the surface. She brought in fun colored ties for everyone to wear, and nice chocolates, and a little boxed wine. It was time for a fancy party! We all donned a tie, looking somewhat silly with our one formal article of clothing on top of our dirty cave clothes or slightly-less-dirty camp clothes. I actually brought a collared shirt for my camp shirt (because I like linen in the heat) which went well with the tie.

Us hanging out at camp with our fancy ties.

We spent the rest of our daylight on that liesurely evening hanging out at camp and getting up to various shenanigans such as acro yoga. Sam already knew some, and Stan and Jason tried it for the first time. It was fun to do that in such a unique location, although it was a little bit of a challenge being stable in the wash which wasn't perfectly flat nor even.

I taught Stan this basic acro yoga pose.

Sam and I in the throne pose.

That night, Sam wanted to try out sleeping in the portaledge hammock in a crazy setting just like I did the night before. She took it even further by bivvying not on the edge of the cliff below camp, but dangling from the freehanging rope that drops you into the entrance! After Rachel and JJ got back to camp (so she wouldn't be hogging the rope), Sam set up the hammock portaledge on that freehanging rope. It must have been even more awkward getting into the hammock on a freehanging rope, instead of on a rope with a cliff face on your side like I did the previous night, but she said it was super worth it for the epic bivvy. She woke up to not only the towering red cliffs of the Grand Canyon surrounding her, but also the large impressive cave entrance.

Sam in her bivvy in the cave entrance the next morning.

Although I don't think the portaledge-hammock was really necessary to have enough flat(-ish) sleeping spots for everyone, I'm glad I brought it because Sam and I were both really psyched to use it to setup some epic bivvies.

The third and final day of caving, after our relaxing second day, we were all eager for a longer and harder trip. Stan and I were particularly psyched to return to the E survey where we turned around at a good-sized room with the Snowy River formation continuing off into the unknown. Stan, Sam, and I returned to E66, the nice room where we turned around day 1. This time we managed to keep our feet dry by doing some wide, somewhat exposed stemming high above the pools in the bottom of the main canyon passage of the cave. This was a nice morale boost: no need to wear wet socks the whole day. We also made it out there noticeably faster than before. I think our travel time to the lead was 1.5-2 hours, instead of the 3 hours it took each way day 1.

We got set up to survey at this nice room and were delighted to find the passage continuing as even nicer, larger canyon passage, with the Snowy formation still following us the whole way. Just like day 1, we did occasionally have to take high routes above Snowy, but never too far from it, and this time the high bypasses were generally larger, easier srambling above large breakdown instead of crawling in smaller breakdown spaces. The passage had some nice colorful speleothems on the walls, and the Snowy formation got even thicker, wider, and more white. Eventually we ran into pools of water on Snowy, and spots where it got very deep, up to 10 ft of white calcite coating below the paleo flood line. The pools of water were an especially nice find because we didn't pack much water for our day of surveying; I only had half a liter on me and I think the others packed the same.

One of the deeper parts of the Snowy formation. There's a pool of water at the bottom of the depression there, not quite visible.

With the Snowy formation getting more prominent and our shots getting longer in the larger, straighter passage (at one spot we had a 60 ft and 70 ft shot back-to-back), it seemed that this passage was never gonna end! We even passed a few (small, not super promising) side leads on our way. However, our first inclination that something was going to change about this passage came when I saw a bug in the passage lying on a formation. Bugs shouldn't be able to make it this far into the cave. I thought maybe this one got lost and died, as it wasn't moving. Then I saw another bug flying around and I new we had to be getting close to an entrance. This hypothetical entrance would have to be a new entrance, as we knew we weren't heading towards the main entrance. Eventually I started seeing packrat midden (debris they bring in such as leaves, sticks, and bones) on the floor and on ledges in the passage and I knew that this new entrance had to be close. Stan even heard 2 scenic tour helicopters flying around, thru the cave walls, another sign that we had to be close to the surface. He later decided to name this passage "Whirly Bird Passage" after these helicopters.

The passage abruptly changed from large walking passage to a small crawl at E112. There it also got super warm and there were spider webs around. We were all eagerly awaiting popping out of a hole in the Redwall and looking out at the Grand Canyon! But unfortunately, as I went thru a squeeze then turned a corner in the now-low-crawling passage, it ended in dirt fill. It was 3 ft wide, 2 ft high passage that very abruptly ended in a dirt mound (possibly moved around by packrats? They often fill cave entrances shut with debris they bring in). There was not even a too-tight continuation to the passage, and no air. I turned off my light to see if there was any faint daylight poking through, but there was none. After surveying the last few shots to the end (E115), I quickly backed out of the crawl as I was getting quite hot in there. Slightly defeated, we all agreed this was a good time to turn around. The side leads we found would have to wait for another trip.

The 3 of us travelled back to the entrance very efficiently. We made it back from E115 in under 2 hours of travel time, not counting a long break at the intersection near Rachel and JJ's aid climb where we waited for them to be done with their aid climb so we could pick up some aid gear from them to haul out. The 49 stations/1076 ft of new survey that day took only 15 minutes or so to travel due to its large and comfortable nature. And by this point we were well-practiced with traveling thru this passage so the rest of the travel went quickly.

We surveyed 1076 ft that day in a trip that was a little under 13 hours—same length of time as our first day where we only surveyed 583 ft. The nicer passage with easier travel, longer shots, as well as our faster travel time to and from the lead all contributed to us getting much more survey footage this day in the same length trip. The new survey in the Whirly Bird Passage this weekend totalled 1659 ft. There are still a few side leads in the Whirly Bird Passage, so we will surely be back. After Jason entered in the survey data, we saw that the survey put the potential new entrance somewhat close to a cliff lead in the Redwall we could see across camp, although not exactly on it. It could be that this potential new entrance is in fact the cliff lead we could see across camp and the survey is slightly off (much of the data is old, from the compass-and-tape survey era), or the potential new entrance could be somewhere else we can't see from camp. We'll be exploring the area to find it for sure. The dirt plug we found at E115 could probably be dug out, and it would probably be easier from the other side where we would presumably have more space and wouldn't be filling up the low crawling passage with dirt. It would be nice to have a faster way to get to the leads in the E survey, although if accessing this other entrance requires lots of bushwhacking far from camp then it might not save much effort.

That night I just slept on the ground in the wash instead of in my hammock, to save time the next morning when packing up camp to hike out, and because there was enough (barely, but enough) flat space for everyone in the wash. That next morning, Tuesday, we got up at 5am with the aim to start the hike out at 6am, in order to get it done before it got too hot and to give everyone time to drive home that afternoon. Jason, for example, had a long drive back to San Diego and wanted to get it done that Tuesday night. I took a while packing up my camp and didn't get going until 6:30am, but I caught up with everyone ahead of me before we got back to the cars. In fact I was still the first back at the cars (although that certainly would have been Jason had he not gone back to help out someone who was struggling with the hike by taking their pack). The hike out was actually faster than the hike in for me, at slightly under 2 hours rather than 3, despite the fact that it was steeply uphill rather than downhill. This is because steep, loose, vegetated terrain provides more of a challenge going down than going up, and that extra difficulty was greater than the extra difficutly associated with going uphill on the way out.

Rachel (front) and Hailey on the hike out. Classic Grand Canyon off-trail hiking: steep, loose, and densely vegetated.

I got quite scraped up from all the thorny New Mexico Locust on the way out. Because I wasn't with a group at all times during the hike and was largely just going my own pace, taking my own path between the obvious features, I did not always take the most efficient path. When I wasn't sure the best way to get from point A to point B I generally just went in a straight line, which was often thru dense clusters of thornbushes. At one point, a small bush I was using as a handhold on a steep slope ripped out, and I fell backwards into a bunch of thornbushes. That wasn't too fun. By the time I was back at the cars, my arms and back were aggressively scraped up. That week at UNM people kept asking me what I was up to the past few days when they saw my arms. Of course I was eager to tell them all about the Grand Canyon when they asked.

The state of my arms and back after the bushwhacky hike out.

There was a small lake where we parked the cars. Apparently, every other time Jason had been here, the lake had been dry, but now it had water. At first I was a little hesitant to swim in it due to the muddy, opaque water, and due to the bugs flying all around it, but Stan went in first and told me the water was nice so I jumped in. And the water was indeed nice. It was pleasantly cold and very refreshing. I was able to get most of the dirt and sweat off me although the water in the lake definitely deposited some more mud on me in places. Still, it resulted in me being cleaner and less smelly for the long drive home which was nice.

All 8 of us got lunch at a restaurant in Jacob Lake just outside the park, then went our separate ways to drive home to places as far away as San Diego and Wyoming. After stopping in Farmington at Stan and Gosia's house for dinner, I drove the rest of the way home by myself and made it to Albuquerque just before midnight that night.

When you're all the way back there, you are extremely isolated from anything and anyone on the surface of the Earth. You have no communication with the outside world, and no tools or other amenities of modern civilization other than what you and your team carried in there. If someone gets hurt or something goes wrong in any way, expect a a full day before anyone on the surface knows about it (via some of your team members taking off for the surface to tell them in person), and 2 days before any supplies or people they send to help arrive at you. This sense of extreme remoteness and disconnect from the outside world, altho sometimes scary, is one of the major appeals of caving for me.

The Snowy River Passage in Fort Stanton Cave is notable not only for going extremely far from the entrance (it goes from ~1 mile from the entrance all the way to ~12 miles from the entrance), but also for having a beautiful white calcite crystal formation covering the floor uninterrupted for the entire 11+ mile length of the passage. This formation is extremely unique, with nothing like it of the same scale anywhere else in the world. The Snowy River formation is the world's longest single cave formation, and the furthest reach of the cave out at the current known limit of exploration of Snowy River is by far the most distant one can get from a cave entrance underground, in terms of both in-cave travel distance (13 miles at a side passage off of Snowy River) and straight-line distance.

Besides the challenge provided by the extremely long and remote nature of the far Snowy River trips, there is also a challenge associated with conservation of Snowy River. Snowy is a clean, beautiful white cave formation surrounded by muddy and otherwise dirty passages. Carelessly traveling on Snowy and adjacent dirty passages would track dirt all over it and ruin its pristine, pearly white appearance. Anyone who travels on Snowy River has to take great care to avoid getting Snowy dirty. This involves changing between clean and dirty clothes multiple times per trip when switching to traveling on Snowy to traveling on dirty passage for extended periods of time, and putting on clean or dirty shoe covers when briefly switching between Snowy and dirty passage. We also only travel on Snowy River when it is dry, without any pooled or flowing water in areas where we have to walk or crawl, because Snowy River gets soft and malleable when wet and could get permanent footprints if we traveled on it in that state. Snowy River floods irregularly and takes months to dry up after it floods, so conditions when it is dry all the way out to its current known furthest point are cherished opportunities to travel to the frontier. This spring there has been an unexpected and welcome period of dry conditions on Snowy, and I am thankful to have been invited on a far Snowy River trip during these conditions.

The furthest known reaches of the cave are still full of unexplored passages that keep going off into the unknown. We were going to travel to those far reaches at the limit of exploration. We had 2 goals on this trip:

1. Collect some samples for scientific analysis. My friend Riley Drake, a biology PhD student at Emory University, studies cave life (among other things). She was interested in studying the water in various pools in the far end of the cave, where very few humans have ever been. In fact, one of the pools we were to sample water from, had only ever been visited by the original team that explored the section of passage the pool is in and found the pool; that's 4 humans ever. We would be the second ever group of humans to visit this place and would collect water from it while it was still highly unimpacted by humans.

   Riley also wanted to collect samples of small snail shells that previous explorers had found in a dry streambed in a passage very close to Midnight Junction. She wanted to determine when these snail shells were washed in to the cave, and what species they are (if they even are any extant species).

2. Exploration of new passage. We had a long list of leads (unexplored passages in the cave) near Midnight Junction Camp. After we collected all the samples we wanted to, we were to pick whichever leads looked interesting to us and start surveying them.

I got to the Fort Stanton field house where Riley, John and Ramon were at around 8:30pm on Friday May 17th, the night before we were supposed to enter the cave early Saturday morning. The others had arrived there less than an hour before. We quickly packed all of our bags, double checking to make sure we had everything given how rushed our packing was. I was done with packing and in my sleeping bag by midnight. We all got up at 7am, wishing we had a little more sleep but still excited to get in the cave. I choked down my massive breakfast that was a whole box of mac and cheese, and we all got in the cave at 8:30am.

There is about a mile of travel through the "historic" section of the cave (known for hundreds of years, including to some Native Americans who explored parts of the cave near the entrance using cane torches) before one gets to Snowy River. Much of this is huge walking borehole, and it takes about 45 minutes normally. However, a section of this huge walking borehole called Conrad's Branch was flooded with calf-deep water, which slowed us significantly. The water was murky with low visibility and the floor under the water was sloppy mud. Falling in the water would mean that all of our gear, including our clean gear for Snowy River, would get soaked with brown muddy water and be way too dirty to clean up in a reasonable amount of time, which would mean we would quite literally have to turn around and can the trip. To make sure that there was no chance we would fall in the water, we all brought a ski pole to use as an extra point of contact when traveling through the deep water. We also brought a separate pair of dirty-mode pants, rain boots, and socks for travel through the flooded area, in addition to our standard dirty-mode clothes, as these clothes got way too wet and muddy to bring with us on Snowy River, even packed away in bag after we changed into clean clothes. The water was just deep enough to flood our rain boots (slightly below knee-deep), so we all did this section of cave with wet feet.

We got on Snowy River at Turtle Junction, after a little over an hour of travel from the entrance. Much of Snowy River is large, pleasant, easy walking passage. However, there are still plenty obstacles that slow you down. Occasionally the ceiling in Snowy River passage does get low and you have to crawl on it. Whenever you travel on Snowy River, but especially when you crawl on it with the ceiling/walls very close to you, you have to avoid touching the ceiling and walls with you or your pack. This is because the ceiling and walls are often covered in dirt and other sediment that will fall off and get Snowy River dirty if you so much as slightly brush against them. Whenever you travel on Snowy River you carry a bag with damp clean rags with you, so you can clean up any messes you do make. You may have to clean up Snowy River itself or your clothes/skin/pack so you don't transfer dirt from yourself to Snowy River. Minor accidents that spill a little bit of dirt are not a big deal as long as you immediately notice when they happen so you can stop and clean them up right then and there. But stopping to clean up yourself and/or Snowy does slow your entire team down, so everyone does try very hard to avoid touching anything other than Snowy. It's the reverse of that childhood game "the floor is lava"—in this case, the floor is safe, and everything else (walls, ceiling, breakdown blocks in the passage) is lava.

There used to be a section of Snowy River that was a long section of crawling called the Crawl from Hell. But it was bypassed via a passage called the Black Rock Bypass. The Black Rock Bypass is a dirty-mode passage that takes about 45 minutes and is not exactly easy (it's full of stoopwalking, crawling, and various other obstacles) but still saves a ton of effort over the worse crawling section, that includes low belly crawling, on Snowy River at that section.

Sometimes the ceiling directly above the Snowy River formation can be really low, crawling height or even impassable, but the passage itself is taller off to the side of Snowy. Then it makes sense to walk off to the side of Snowy on the dirt banks. In order to not track dirt from the banks onto Snowy, we put plastic tarps over these dirty sections so you can stay in clean mode while stepping off of Snowy onto the dirt-floored parts of the passage with taller ceilings. We call these plastic tarps "magic carpets". The magic carpets can save a lot of time and effort, but sometimes they get annoying when not placed well or when they get dirty. My first and only other Snowy River trip, last October, was a magic carpet maintenance trip, where we improved the route by working on some of the worse tarps. This included staking down tarps that were on slopes and thus liable to shifting around when people walk on them, moving tarps further away from obstacles so you don't have do tiptoe around obstacles when walking on the narrow tarps, adding more tarps where they were useful but hadn't been placed on previous exploration trips, and cleaning tarps that had gotten dirty. That trip was very satisfying because the trip out of the cave was noticeably easier after we improved the tarp situation on the way in.

Sometimes you have to get off Snowy River onto dirty surfaces where you can't use tarps for various reasons, such as the dirty surface being uneven and rocky, or too long to carry that length tarp out there. Then you use shoe covers to keep your clean shoes off the dirt. Putting on and taking off shoe covers is an intricate dance in and of itself: you have to balance on one foot with your heavy pack, often on an uneven surface, stick out your foot over a dirty surface so you don't drop dirt from the dirty shoe covers onto Snowy, put the shoe cover over your shoe at the end of your outstretched leg, then step onto a dirty surface, all without touching anything dirty like a wall for balance. This dance gets quite tedious when you do it hundreds of times during a single trip, and it can be somewhat high-consequence as a slip and fall onto a dirty area could spill quite a bit of dirt onto Snowy which you would have to clean up.

One of the more extensive shoe-cover obstacles is Mount Airy. Mount Airy is a breakdown pile atop Snowy River about a third of the way (in terms of travel time) to Midnight Junction. You gain maybe 100 ft of elevation by scrambling over breakdown for a few hundred ft. The standard procedure is to put shoe covers on and take your gloves off so you can touch the dirty breakdown with your hands without getting your clean gloves dirty. Scrambling around breakdown in shoe covers, while making sure not to touch anything with your knees elbows, or other body parts, is certainly a unique challenge that I haven't experienced in any other adventures.

After Mount Airy is a section of Snowy called the Eggshell Trail, because here Snowy gets very thin and easily cracked. We travel very slowly thru this section, all walking single file in the same spot so we contain any impact we make to one spot. By carefully stepping on the spots that look thickest, you can generally avoid cracking Snowy, but it requires soft steps and careful attention to where you place your feet.

Our first chance to collect water for drinking in the cave was at Finger Lakes, which is a series of pools on Snowy River about 2/3 of the way to Midnight Junction by travel time. Riley wanted to collect some water samples for analysis before we collected drinking water which would possibly contaminate the pools. This was the first scientific sample collection stop of many on the trip.

Immediately after Finger Lakes is Rough Country, a section of breakdown-filled borehole not too unlike the Mount Airy breakdown pile, except much longer (~1 hour of travel) and with some fairly exposed, high-consequence climbing above tall dirt/rock cliffs in the breakdown. Rough Country is also far too long and difficult to do in shoe covers to keep your shoes clean, unlike Mount Airy. Instead our clean/dirty strategy was to switch our bottom halves (shoes, pants, kneepads) to dirty clothes, while keeping our shirts and our packs in clean mode. We couldn't switch our packs to dirty mode as we left our dirty over-packs at the Black Rock Bypass, the last extended section of dirty-mode travel before camp. We didn't get any pictures of this section, as we were all focused on not falling off the numerous cliffs in the breakdown and on not getting our upper halves dirty.

Rough Country, despite being full of obstacles and a lot of work, is a very complex, interesting area with many leads. In several spots you can see the Snowy River formation at the bottom of the passage off to the side, although the main travel route stays high and off Snowy for the entire length of the breakdown passage. The passage is huge, even bigger than most of the big parts of Snowy River, and the uneven breakdown floors often make it hard to see the entire passage. It can be hard to tell what is a real lead going off in another direction versus what is just another way thru the breakdown in the same passage. On this trip in general, but especially in this passage, I got in the habit of pestering John (who knows the cave very well) every time I saw another passage "is that a lead?", "does it go?", "does it still go?". Quite often the answer to all those questions was "yes".

After about an hour we were through the Rough Country breakdown and back on Snowy River. From here to Midnight Junction is fairly straightforward Snowy River travel, although there are very few magic carpets out this far as exploration teams stopped wanting to bring them out this far on their long survey trips. This meant tons of putting on and taking off shoe covers to cross dirty sections in the Snowy River passage. I lost count of how many times we had to swap into and out of shoe covers between Rough Country and Midnight Junction—probably around 50? I was pleasantly surprised that no one on the trip ever lost their balance and fell during the hundreds to one-legged balancing acts that are putting on and taking off shoe covers while standing on uneven ground and wearing a heavy pack.

We finally made it to Midnight Junction at at 9:30pm, 13 hours after we entered the cave. This was a bit longer than the 11 hours or so I had been told to expect to get to Midnight Junction. At Midnight Junction, the passage splits. The right passage is the continuation of Snowy River, and the camp (which has much gear including sleeping bags and a stove permanently stashed there) is in this passage just past the junction itself, on a flat mud bank just above Snowy. The left passage takes you to Harmony Hall after a few minutes, which is an intersection where you gain Midnight Creek, a stream passage that goes both ways and is the water source for camp.

Although Midnight Junction is supposed to be a nice camp, I actually wasn't a huge fan of it. It is flat, with a soft dirt floor, and warm, which is nice. But my objection about the camp is that you are somewhat in tarp jail when you're there—the tarps are clean and you have to be in clean mode when you're on them, but the dirt bank around you is dirty. To leave your tarp you have to put on shoe covers, and even then you can't sit down on the dirt; you have to be in full dirty mode to do that. The tarps are not huge at about 5 by 10 ft, and all of your stuff that you unpack onto the tarp takes up a good amount of space. I felt a little trapped on my small tarp, unable to get off it and hang out around camp without changing into dirty mode. Still, it worked well enough for camping and I slept well there on the soft flat dirt. At least it wasn't a hammock camp.

We quickly unpacked our gear on the tarps, made some dinner (which for me was my version of dinner mix, loosely modeled after the Cueva Cheve dinner mix they use for long underground camp trips), and switched into dirty caving mode. It was time to collect water then begin our second science sample collection mission: collecting snail shells that had been washed into the cave near camp. From Midnight Junction we took the left passage, which quickly brings you to the Harmony Hall intersection where Midnight Creek is. Before we collected drinking water there, Riley collected a water sample from Midnight Creek. We then went right (downstream) into a passage called La Culebra. where the snails were.

We went slowly down La Culebra, scanning the passage for snail shells. We found several spots where several snail shells were lying on the floor, often right in a pile of charcoal too. Probably some event that washed in some charcoal also washed in these snail shells. I wonder how related the charcoal and snails are—maybe some large fire event that made a bunch of charcoal on the surface also killed a bunch of snails, all of which were washed in together? Identifying and dating the shells should help answer this question. Another interesting observation we made in this passage is that in multiple spots, the footpath thru the dirt had been visibly washed away where the footpath crossed a low dry water channel in the dirt floor of La Culebra. So La Culebra must have had flowing water some time in the past few years. That's a nice data point to have, as it's not obvious that this passage takes water anymore. Midnight Creek drains away from the passage right after the Harmony Hall intersection, and there was previously no indication that the passage ever takes water from anywhere else.

After collecting several samples, we turned around and made it back to camp at 1:20am. When I cave camp, I like to get on what I call "cave time", which is when you do long days of exploration that result in you getting back to camp very late, then sleeping for 9 or more hours after that no matter what time that results in you getting up the next morning (or afternoon). This means that your "days" become longer than 24 hours, often 30 or more. I find that irregular schedules like this work really well for me when cave camping, because really long days get me nice and tired which helps me fall asleep easily, and because the lack of daylight and irregular noises from surface weather means I sleep really well even when I'm off my usual 24 hour schedule. However, the rest of the team said they do not work with that schedule, and that they would be up early the next morning according to their internal 24 hour clocks no matter what time we all went to bed. Oh well, no "cave time" for me.

We woke up at 8:30am the next day (thankfully not too early) and set off for Laguna Lechosa, a large pool at the current known furthest reach of the cave. We were out of camp at around 9:50am, in dirty mode at this point. From Midnight Junction, the route to the end of Snowy River actually follows a different passage than Snowy for a while before getting back on it at 4:30am junction (yes, named for the time of day when it was first discovered). The route goes to Midnight Creek, the water source just outside of Midnight Junction camp, then turns right and follows La Culebra (which transitions from merely pleasant walking passage into huge borehole) for a while. After following the impressive La Culebra borehole for a while, there is a mildly confusing area with a few turns. The route first brings you thru an amazing section of velvet flowstone ("Red Velvet Passage") that requires clean shoe covers to keep it clean. After another intersection, the passage abruptly changes to an annoying low belly crawl with grabby cobbles on the floor and many stalactites sticking down from the ceiling that you must avoid breaking ("Catch 23"). The belly crawl is high-effort but short, and quickly open up into walking passage that brings you to 4:30am junction where we finally regained Snowy River. The original route to this area followed Snowy River all the way from Midnight Junction, but apparently Snowy gets very thin and fragile there, and the team that explored that route recommended that it not be used again.

Along the way there was one handline downclimb, perhaps 15 ft, over a dirt cliff. The handline was anchored to 2 rocks which were embedded in a not very secure space in between the dirt floor and a bedrock protrusion in the ceiling. As the first person went down the handline I noticed that the anchor for the handline shifted a bit, and I immediately jumped onto it to secure it by pushing my shoulder into it. After the first person went down safely, we tried to move the rocks into a more secure position, but we didn't have much to work with. We decided that everyone else would climb down using the handline, with me backing it up by lieing right next to it, ready to push my shoulder into the rocks to hold them in place in case they shifted more. But I wouldn't actively push myself into the rocks otherwise, so they could be tested so the last person down (myself) could go down without a backup, having tested the anchor.

Once we had regained Snowy River at 4:30am junction, we encountered pools in Snowy River with increasing frequency. Riley stopped at a few of them to collect samples. At this point we were caving in dirty mode, mostly on dirt banks to the side of Snowy, and frequently crossing it with shoe covers. It seems that more tarps would be very useful here if people traveled this section in clean mode, but given how often you are off Snowy in this section I'm not sure that bringing tarps and doing it in clean mode would increase efficiency at all. We crossed Snowy River at SRS772 and found it flowing with water—not just with a standing pool, but actual flowing water. This was the first time we had seen flowing water on Snowy this trip. Thankfully we were still able to cross it by hopping over some rocks which made a near-bridge over Snowy.

Shortly after crossing Snowy with flowing water, we arrived at a large breakdown room around SRS779. This breakdown room is confusing, with Snowy nowhere in sight, but many leads heading off in different directions at different levels. We needed to find Laguna Lechosa, a large pool that was the main target of Riley's science sample collection mission. We also wanted to find the route that regains Snowy River from this room, to check on it and see if it was flooded (as it was during the one and only previous trip to this place). After poking around in several different passages for an hour or so, we finally found the Laguna. Laguna Lechosa had only been visited by one other party of humans—the original exploration party that found and surveyed this area in 2022. The area is extremely isolated from, and unimpacted by, humans, and it is extremely interesting to ask what kind of life is in the water here. I look forward to hearing about her results and will share them here in the future.

Update 2025-07-09: Riley told me that she found primarily Polaromonas glacialis bacteria. That is a species primarily found in glaciers and ice caps, and occasionally in seasonal snow. Are they coming from snowmelt in the Sierra Blanca mountains above? Or are they an ancient remnant from a time when glaciers covered this part of New Mexico? Riley will figure it out! Stay tuned.

While Riley was collecting her samples from Laguna Lechosa, I went down the passage from this room that regains Snowy River, to see if Snowy was wet. The passage went as a belly crawl for maybe 20 ft before dropping into a nice sized walking passage with Snowy River right there in the center of it. And Snowy was indeed flowing with water, as we expected. That doesn't bode well for the next trip—with the amount of water flowing (I estimate about 0.5 cubic feet per second). I can't imagine it will dry up in time for the scheduled exploration trip at the far end of the cave this June. Oh well, they'll have plenty of other leads in that area to explore instead.

After Riley finished collecting her samples and I was done checking on Snowy River, we headed back to camp to unpack the science gear and make dinner before switching to exploration/survey mode. At the first shoe cover crossing, Riley couldn't find her shoe covers, and she figured she had left them somewhere in that complex junction area near the current known furthest point in the cave. Riley and I went back to that area to look for them, which was thankfully only 5-10 minutes away at that point. While we were looking for the shoe covers (which we found at the Yellowstone River where we had last used them), John and Ramon surveyed a small side lead at SRS769 (marked on the survey as "2x6 gets low bad") that went as a crawl for a few stations before ending in a sump. After that we made it back to camp uneventfully, stopping in La Culebra several times to collect a few more snail samples that we didn't have time for the previous night.

We were back at the Midnight Junction camp at 6pm. At camp, we dumped the samples and science gear, made a quick dinner, and switched to exploration+survey mode. We were out of camp by 7pm on our way to survey some leads in the La Culebra Arriba passage (MJ121+) and MA passage. The route to these passages involved turning left at Midnight Creek (the stream where we get water a few minutes from Midnight Junction) instead of turning right towards La Culebra. We travaled for maybe 20 minutes upstream in Midnight Creek, which involved one short crawl where we had to take care to stay dry and clean(-ish, as we were in dirty mode at that point, but still didn't want to get our dirty clothes covered in slimy mud). Scrambling up through a large hole in the wall/ceiling brings one to the intersection of La Culebra Arriba, a huge borehole that goes left from the intersection, and the MA survey, another huge borehole that goes right and is somehow unnamed despite being an extremely impressive borehole. This intersection was extremely impressive with ~60 ft tall boreholes going off in either direction.

The first lead we went to was in La Culebra Arriba (turning left at that intersection), at MJ133, marked as "10x10". On our way there we passed another lead at MJ124, marked "5x10", which we thought looked pretty good and would be a good option if our 10x10 lead didn't go very far. The lead was exactly as described, nice 10x10 walking passage. We tied in at MJ131 because we couldn't find MJ132 nor MJ133, and started surveying. John and Ramon sketched, while Riley set points and I was front sight.

The passage continued as nice walking passage with some beautiful velvet flowstone for 200 ft before reconnecting with La Culebra very close to midnight junction at MK4. Although we would have preferred that the passage continue forever with its large, decorated, pleasant character, it was useful to close a loop and provide a shortcut back to camp. This new route back to camp would bypass the wet crawl in Midnight Creek and result in perhaps 10 minutes less travel time back to Midnight Junction. This was MJ131A-E.

Another side lead in this passage also connected to La Culebra after a few stations and had more beautiful velvet flowstone. This other side passage connected at a 14 ft overhanging pit that we couldn't downclimb. This was MJ131H-J.

After connecting to La Culebra in 2 different spots, we went back to the 5x10 lead further back in La Culebra Arriba, which started out as stoopwalking passage with some nice gypsum needles in the floor. This went for an even shorter distance before connecting to the intersection between the Midnight Creek Passage, La Culebra Arriba, and the MA passage, via a tortuous route through the breakdown in the floor of that intersection room. This was stations MJ124A-G.

After killing those 2 leads there was some talk of going back to camp, but we decided we would rather continue surveying as there were more leads not too far down the MA passage and we all had plenty of energy. I'm glad we did, because the 3rd lead we surveyed ended up being quite interesting, and I got to see the MA passage which is more huge impressive borehole like La Culebra Arriba.

From the intersection, we now went right, down the MA passage, which was similarly impressive borehole that got 80 ft tall. The first leads were a pair of side passages on both sides of the borehole at MA3 (only 3 stations from the intersection). The lead on the left wall of the borehole was marked 3x3, and the lead on the right was marked 6x3. Naturally we chose the lead on the right and started surveying.

This lead started out as fairly nice walking passage, 3 ft wide and just a body length tall, and quickly got even nicer. It started averaging about 4 ft wide, and the ceiling slowly gained height as the passage went on. The passage was fairly flat with little inclination, as was the floor which was mostly soft flat dirt. Eventually the passage gained a classic keyhole cross-section as the ceiling got tall, around 25 ft. The passage took many right turns which made our shots short (often ~20 ft) and the survey somewhat slow, although we certainly didn't complain as we were surveying nice, pleasant, virgin walking passage that didn't seem to want to end and was heading towards a large blank area on the map. And with 2 fast sketchers (John and Ramon), the sketch was able to keep up with Riley and I setting points and taking shots. There were no side leads in this passage, just one obvious walking passage heading off the map.

We decided to name this passage the Porcupine Passage, after a cluster of thick gypsum needles embedded in a clump of mud at the entrance to the passage that I thought resembled a porcupine. All of us New England cavers love porcupines and encounter them often when caving in the winter, as they love to shelter in the warm caves in winter in New England.

The Porcupine Passage was full of nice gypsum formations like needles, but one formation in particular stood out. We found a thin strand of gypsum wound in a perfect helix shape, like a spring. I had never seen a formation like that before. We flagged it so others will see it and avoid accidentally brushing up against the wall it's on and destroying it.

It was at this point that I realized that Fort Stanton Cave is much more than just the Snowy River Passage. The cave is over 44 miles long, and Snowy River is only 13 of those miles. Side passages off of Snowy River make up much of the rest of that mileage, and many of those "side" passages are extremely impressive and interesting passages in their own right, such as the huge boreholes of La Culebra Arriba and the MA survey. And those side passages are full of great leads. There is lots of exploration to do in Fort Stanton, and I am excited to be a part of it during the next several years I'll be in New Mexico.

Once we used up a whole alphabet of stations MA3A-Z, we decided that would be a good time to turn around and head back to camp. At that point we had just over 1000 ft of survey for the trip. We scooped 30 ft of passage from there to make sure it didn't immediately end, and saw the passage continue around the next corner into the unknown. We used our new shortcut that we surveyed earlier that evening to get back to camp nice and quick. MA3 was probably 25 minutes of leisurely travel from camp using the shortcut. We were back in camp around 2am that night. I snacked a bit then quickly went to bed, knowing that we had to get up early tomorrow for our long day of travel back to the entrance.

We got up bright and early (really, dark and early because it's always dark in the cave!) at 7:30am the next morning for our exit day. Packing our bags and organizing camp took a while, as it always does, and we set off from Midnight Junction towards the entrance at 9:15am. I was glad to be out of tarp jail and up and moving fast. I estimate that my pack was 40-50 pounds, even heavier than the way in. Usually when caving or backpacking, one's pack is lighter on the way out, but the Fort Stanton rule that cavers must carry out all pee and poop from Fort Stanton means that one's pack gets heavier as the trip goes on. Water you collect from the cave to drink gets turned into pee which must be carried out. Lightweight dehydrated food with no water weight gets turned into poop which has water in it. I carried out 5 liters of pee from my 3 day camp trip, which is 10 pounds. The Fort Stanton camp packs have hip belts on them, which I was glad to have. This is the only time I've ever wanted a hip belt on a cave pack due to the weight.

The trip out was uneventful. Once I got past the last stoopwalking obstacle in the historic section of the cave, I took my shirt off and did the final hour of caving shirtless. At 55F, the cave might not seem super hot, but the high humidity means you do get quite hot and sweaty when working hard, especially when wearing long sleeves and pants in dirty mode to keep your skin clean. It was nice to be no longer concerned about keeping my skin clean for Snowy River, and to cave with minimal clothing to heat me up.

The historic section of the cave felt like it took forever, with the calf-deep water (usually not present) slowing things down significantly. This section is usually very straightforward walking or even running in huge borehole, but the murky, low-visibility water with sticky mud floors meant we had to go slow and use our ski poles for balance. We made it to the entrance at 9:50pm. It took us 12.5 hours to get to the entrance from Midnight Junction, a little faster than our 13 hours on the way in. Not bad considering our packs were heavier than on the way in.

After unpacking my bag and dumping all my gear at the field house for the others to deal with in the morning (thank you Riley John and Ramon for dealing with that!), I started my drive back to Albuquerque. It was shortly before midnight when I started the 3 hour drive. I considered spending the night and leaving early the next morning, but I did want to be back at UNM early for the NQI Quantum Algorithms Workshop that was happening starting 8am the next morning. In fact I originally declined to go on this trip because I wanted to be at the workshop on Monday, its first day, but John talked me into coming (thanks John!). So I figured I would start the drive that night while I still had some energy, and pull over and sleep in my car as soon as I got tired. As I started driving, I was still full of energy and excitement from the amazing trip I had, but the sleepiness started to kick in around the halfway mark, just outside of Socorro. There I pulled over onto a BLM road and napped in my car until 5am. Then I drove the rest of the way to UNM, not even stopping at home, to save time and because the physics building has showers where I could clean up and change for the conference. Thankfully the conference had plenty of free coffee to get me through that long day without much sleep.

Thanks to Riley, John, and Ramon for many of the photos, which are also courtesy of the Fort Stanton Cave Study Project, BLM, and USFS.

First, some background. Quantum computers work by taking advantage of the ways that the laws of physics at the small scales of atoms and molecules (that is, quantum physics) work differently than the the laws of physics at larger scales that we are all familiar with (that is, classical physics). We know we can use these different laws of physics to store and process information differently than how we do so on regular, "classical" computers. In classical computers, the basic unit of information storage is the bit, which can take the value 0 or 1, and we do computation by flipping bits between 0 and 1 based on the values of other bits. Quantum mechanics allows us to store information in "quantum bits", or qubits. A qubit can be any physical system following the laws of quantum physics that has 2 basic states: one analogous to a "0" and one analogous to a "1". Quantum physics tells us that the allowed states of such a system are not just the 0 state or the 1 state, but also any combination of them that contains some about of 0 and some amount of 1. This is distinct from allowing the state to be any number between 0 and 1, like 0.5 or 0.25. Quantum mechanics actually gives us more power than that.

The fact that quantum states are continuous, rather than discrete like the classical bits that can be 0 or 1, means quantum computing gives us more computational power but also comes with some problems. Continuous states means that small errors can add up over time, over the course of a long computation, and ruin the computation. This is not a problem in classical computation for 2 reasons:

1. The discrete, 0 or 1 state of classical bits means that if we get small errors, we can remove them by simply rounding to the nearest 0 or 1, and the error goes away. For example, we could store the state of a classical bit by the voltage in a capacitor: 0 volts or 1 volts, for example. We could always just continuously measure the voltage in the capacitor and round down to 0 volts or round up to 1 volt, and we would never have any problems with sufficiently small errors. We can't do this with quantum states due to their continuous nature.

2. Also, us humans have become so incredibly good at manufacturing extremely accurate, error free components for classical computers. One talking point that quantum error correction people like to mention is that the dominant source of errors in computers at the electronics level (I mean, caused by the hardware acting up, not by humans using the device incorrectly) is stray cosmic rays from outer space striking the computer. That's how good we've gotten at manufacturing computers! Still, even if we got that good at manufacturing quantum computers (and we are very far away from that), we would still have to deal with errors due to point number 1 above.

Thankfully, we know how to deal with this problem of errors occurring during computation in quantum computers. We know how to do quantum error correction. This involves encoding the information in our qubits using many more qubits for redundancy in a way that allows us to detect and correct errors. This is analogous to error correction with classical bits, where you encode some information in classical bits with many more bits for redundancy. The simplest classical error correcting code is called the "repetition code": if you want to send a bit over some noisy channel where the bit could be flipped between 0 or 1 accidentally, just send many copies of the bit: send 0000... or 1111..., however many times you want, instead of 0 or 1. Then whomever receives the bits at the other end can take majority vote of the bits they received to recover the original bit. Of course, this is a very inefficient way to do error correction. There are many ways to do error correction in a way that uses fewer bits but retains the same amount of redundancy, but this is still an instructive example. We can do a somewhat similar song and dance with qubits, encoding some qubits with a larger number of qubits to protect against errors that occur during computation.

The quantum states we want to protect with quantum error correcting codes are kind of like the probability distributions of classical bits that I just talked about. Quantum states can be combinations of different states, with different weights that determine the probabilities of those different states. Except, unlike with classical probability distributions where the "weights", or probabilities, have to be nonnegative real numbers, the weights in quantum states, or "amplitudes" as we call them, can be negative numbers, and even complex numbers! In quantum error correction, we don't only need to protect the values of the individual states that make up the whole quantum state (this would be called protecting against "bit flip noise"), we also need to protect the values of the complex number weights of all the individual states that make up the whole quantum state ("phase flip noise" is one such type of noise that affects the complex number weights of the individual states that make up the quantum state but doesn't affect the values of those individual states). So quantum error correction is really a more general, complicated, and interesting task than classical error correction.

One of the most promising quantum error correcting codes that we think we will use to build a quantum computer is called the surface code. The surface code has several properties that make it practically useful:

1. All of the error corrections are geometrically local in 2D. This means that we can build our quantum computer on a grid on a 2D surface, where only neighboring qubits can talk to each other, and we can still run the surface code on such a quantum computer. This turns out to be exactly how we build some types of quantum computers, in particular superconducting quantum computers, which are what Google and IBM are investing in.

2. We know not only how to encode information in the surface code to protect it against errors, but we also know how to do computation on the encoded information. It's one thing to encode some qubits using an error correcting code to protect them against errors, but that's not all we need: we also need to do our computations that change the state of the qubits, all while the qubits are encoded in this error correcting code.
   

3. It has a very high tolerance for errors, moreso than many other quantum codes.
   

With all these nice properties, the surface code has been a favorite of the quantum error correction community for many years. An active area of research about the surface code has been decoding algorithms for it. Decoding is the task of determining what error happened and how to correct it, given some measurement results extracted from the code that give some indirect information about the error that happened. Decoding is a task for regular old classical computers: an essential part of any large, useful quantum computer will be classical computers churning alongside them, processing information about what errors happened on the quantum computer and figuring out how to correct them.

A diagram of the surface code. It can be defined on any sized grid, not just the 5x5 grid shown here. There is a qubit at every point/circle on the grid. Every colored square or semicircle on the grid corresponds to a certain type of measurement done on the qubits to determine some information about the error that happened. Blue squares correspond to a "Z measurement" (like measuring the 0/1 bit value) on the 4 or 2 adjacent qubits. Green squares correspond to a "X measurement" (measuring the +1/-1 sign of the weight of part of the quantum state) on the 4 or 2 adjacent qubits. 

OK, with all of that context out of the way, now for my results and how they fit into this picture. I have shown that 2 closely related but different decoding problems for the surface code are "hard" computational problems in a formal sense. This means that we can't have decoding algorithms that are both efficient and accurate, in a certain technical sense of both of those words. There are 2 different decoding problems associated with quantum error correction: what we call Maximum Probability Error (MPE) decoding, and Maximum Likelihood (ML) decoding. I will explain the difference between those 2 decoding problems in a bit, but first let me get to the results. I have shown that MPE decoding is NP-hard and ML decoding is #P-hard. First I'll explain what it means to be NP-hard or #P-hard, then I'll explain the difference between those 2 decoding problems.

One of the most surprising, interesting results in computer science is that a large class of computational problems that we call NP are all equivalent in the sense that if even one problem in that class has an efficient algorithm that solves it, then every problem in that class automatically does as well. This is surprising because many of these problems in NP seem not at all related on the surface: they range from planning the shortest route to many locations (the travelling salesman problem), to coloring regions on a map so none of the bordering regions have the same color (the graph coloring problem), to optimally packing bins into boxes (the aptly-named bin packing problem), to solving classic games like minesweeper and many Nintendo games. Computer scientists have shown this incredible phenomenon by constructing algorithms to convert between any of these problems. So an algorithm that solves one of them can solve all of them. Computer scientists have searched for decades for efficient algorithms for all these problems, many of which are very practically important problems, yet no such efficient algorithms have turned up. They view this as overwhelming evidence that all of these problems must not be efficiently solvable, for if there were efficient algorithms for all these problems, then surely we must have found one of them by now! Formally proving this empirical observation that all NP-hard problems are not efficiently solvable is the famous P versus NP problem that has a million dollar prize attached to it by the Clay mathematics institute.

A problem is NP-hard if it is one of those special problems that can be used to efficiently solve all problems in NP. A problem is #P-hard if it is a problem that not only allows one to efficiently solve an NP-hard problem, but also count the number of solutions to any NP-hard problem. Intuitively, a problem being #P-hard is much stronger than merely being NP-hard, because counting solutions (as for #P) should be harder than merely determining whether a problem has 0 or ≥1 solutions (as for NP).

OK, now for the explanation of what those 2 different decoding problems are. In classical error correction, you send a message that is a bunch of bits, encoded using a larger number of bits that has some redundancy so you can detect and correct errors. You detect the error by figuring out which bits in the message were flipped, and you correct the error by flipping those same bits so they go back to their original value. Given any message that is a noisy, corrupted message from an error correcting code, there may be several possible original messages consistent with that corrupted code word. Generally, you want to find the message that requires flipping the smallest number of bits to get from the corrupted message to that original message. The idea is that the message that requires flipping the smallest number of bits to get between that message and the corrupted message is the one that is the most likely to have been the original message.

You can do a similar strategy with quantum error correcting codes. You have some measurement results that you've done on error correcting code that give you some indirect information about what error may have happened. These measurement results are called syndromes. In general there may be many possible errors that are consistent with the syndromes, that would have caused those syndromes to occur. The task is to find an error consistent with the syndromes that has maximum probability of occurring, among all the errors consistent with the syndromes. This is called Maximum Probability Error (MPE) decoding.

There is another decoding strategy associated with quantum codes, however. This decoding strategy takes into account to a special property of quantum codes that different errors can be corrected by the same correction operation. This special property of quantum codes is called degeneracy, and it really doesn't have a counterpart in classical error correction. I'll give a simple example to explain this property. Say you've done your measurements on your quantum code and determined that there are 4 possible errors consistent with the measurement results: E₁, E₂, E₃, and E₄. Due to that special property of quantum codes we call degeneracy, it could be the case that E₁ and E₂ are corrected by the same correction operation, and E₃ and E₄ are also corrected by the same correction operation (different than the one for E₁ and E₂). We would say that E₁ and E₂, as with E₃ and E₄, are logically equivalent. So, if the actual error was E₁, but your decoding algorithm returns E₂ instead, there's actually no problem at all: you will still do the "correct" correction operation. So to determine what error correction operation to apply, you want to consider the sum of the probabilities of E₁ and E₂, versus the sum of the probabilities of E₃ and E₄. Whichever probability is higher tells you which error correction operation you should apply.

This is what the Maximum Likelihood (ML) decoding task: you consider all the possible errors consistent with the syndromes, partition them into groups (or "cosets") of errors that are logically equivalent (which means they share the same correction operation), and look at the sum of error probabilities for those groups. We call this sum of error probabilities for a group of logically equivalent errors the coset probability. You output an error correction operation corresponding to a group of errors with maximum coset probability.

ML decoding doesn't maximize the probability that you found the original error, but it does maximize the probability that you found the right correction operation, which is what really matters at the end of the day. So ML decoding is really "better" than MPE decoding in that sense. Although, ML decoding can be more computationally intensive than MPE decoding, so MPE decoding is still a viable strategy to use for quantum computing. Intuitively, it seems like ML decoding may be much harder than MPE decoding, because you have to consider sums of probabilities of many errors, instead of probabilities of individual errors. My results confirm this intuition.

OK, now that I've explained what the results say, I'll say just a little about how I proved them. The classic NP-hard problem, the one that started this whole field of computer science about NP-hard problems, is the SAT problem. SAT, short for "satisfiability", is the problem of determining whether any boolean formula has a TRUE/FALSE assignment of its variables that makes the formula output TRUE. OK, that's a lot of jargon, let's break that down. A "boolean formula" is an expression involving some variables (which can take the values TRUE or FALSE) and AND, OR, and NOT expressions. We want to know if we can make TRUE/FALSE assignments to the variables in a way that makes the whole boolean formula output TRUE. For example, a boolean formula could be (x AND NOT y). In that boolean formula, x and y are the input variables. We would say this formula is satisfiable because we could set x = TRUE, then the formula would output TRUE. The boolean formula x AND y AND NOT x is not satisfiable, for example.

Likewise, the problem #SAT is the problem of, given a boolean formula, determining how many different assignments to the input variables make the formula output true. This is a harder problem than determining whether there is zero or more than zero such assignments, as in the SAT problem.

The technical parts of my paper are about showing how to convert a SAT problem into a surface code MPE decoding problem, such that the answer to the decoding problem tells you the answer to the SAT problem. I take a boolean formula, which is something that can be laid down in a flat 2D plane, and carefully embed it in the flat 2D planar structure of the surface code. That establishes MPE decoding as NP-hard. Once we have that result, extending it to show that ML decoding is #P-hard is a much shorter exercise. Recall that ML decoding is related to summing probabilities of many different errors, which should already seem related to counting solutions of an NP hard problem (specifically, the NP hard problem that is MPE decoding! But also, SAT, since I showed that SAT and MPE decoding are in a certain sense equivalent problems).

One interesting part of my results is that we formulate the decoding problem slightly differently, slightly more generally in fact, than is standard. Specifically, we consider the input to the decoding problem to be not only the measurement results from the code, but also a description of the noise process that occurs on a per-qubit level. This is a very reasonable setting for the decoding problem, since quantum computers (in particular superconducting quantum computers, where the qubits are large-ish [on the order of a millimeter] and subject to manufacturing irregularities) will not have the exact same error processes occurring on each qubit, and we know we can do error correction more effectively if we take into account this information on the specific noise present when we decode. Although people often think about the performance of error correcting codes and decoding algorithms with only very simple noise that is the same for all qubits, we know that this is not actually how error correction will go in practice with real quantum computers.

A heatmap of error rates on one of Google's superconducting quantum computers with 23 qubits. Each square is a qubit. Each qubit has different error rates, and we certainly want to use that information when we decode! That will help us do error correction more effectively. Thanks to Andrew Zhao who made this figure using data from Google's cirq software library and let me use it.

Is this bad news for quantum computing? Does this mean we will have to abandon the surface code? No and no! While these results are certainly theoretically interesting, they don't actually spell bad news for using the surface code to build quantum computers. This is because these results are all about worst-case hardness: they say that these decoding problems are too hard to have an efficient algorithm that always gets the right answer for every possible input. But it turns out we don't need such strict standards to use the surface code in quantum computing. It turns out that all we need to use the surface code in practice are decoding algorithms that are efficient and correct in the average case: they need to get the right answer for all errors that are likely to occur in practice. My results show that there are always edge cases that defeat all efficient decoding algorithms, although these edge cases are unlikely enough to occur in practice that they won't interfere with using the surface code in practice.

So, what is the practical takeaway of these results? The impact of these results is that they will inform further research into decoding algorithms. Whenever you have some formal hardness results for practical problems of interest, you have several options:

• Give up ☹️. No way, we're not doing that! ❌

• Reformulate the problem. Are we even thinking about the problem the right way? Are we translating the real world problem we want to solve into a formal definition of a problem we can feed into a computer the right way? For surface decoding, yes we are thinking about the problem the right way—reformulating the problem isn't an option. ❌

• Try for approximation algorithms—algorithms that always return something close to the right answer, even it's not the exact right answer. Well, I also proved some hardness of approximation results, that it's hard to output an error or correction operation that's at all close to the right answer. So this is not an option. ❌

• Look for special cases (restrictions of the problem) that you can solve exactly. Yes, this is an active area of research, and there are several interesting special cases that we can solve exactly. ✅

• Think about heuristic algorithms—algorithms that don't always work, but work often enough to be useful. Yes, much of the research on surface code decoding algorithms focuses on these types of algorithms. ✅

So the practical takeaway of these results is to inform decoding algorithms researchers that you can't possibly hope for an efficient algorithm that is guaranteed to solve the decoding problem in full generality—you have to search for algorithms in the last 2 categories of the above list.

In the northern reaches of Grand Teton National Park, between Red Mountain and Elk Mountain, lies over 2000 vertical feet of Madison Limestone and an impressive array of karst features visible on the LIDAR. Ever since I became aware of the high quality LIDAR available for the whole park and the impressive potential it suggested for that area several years ago, I had been looking to ridgewalk (caver parlance for ‘walk around looking for caves’) there and see what the karst looks like up close. This August, I finally made it happen. I was in the western Wyoming/southern Idaho area for some other caving, and I took 3 days to do a big solo ridgewalking/backpacking trip to check out the area.