Pico Bolivar – Colombia Highpoint

Pico Bolivar (5720.42m)

Highest mountain in Colombia

Eric Gilbertson, Ginge Fullen, Jonathan Suarez

Dec 12 – Depart Seattle
Dec 13 – Arrive Santa Marta 0m
Dec 14 – Buffer day 0m
Dec 15 – Drive 4×4 to San Pedro 1400m
Dec 16 – 4×4 to Yerbabuena village, meet indigenous community, hike to 2350m
Dec 17 – Las Lagunas 3700m
Dec 18 – Laguna Corazon 4030m
Dec 19 – Laguna Herradura 4200m
Dec 20 – Laguna Carcuina 4400m
Dec 21 – Hike over 4900m pass, Basecamp 4600m
Dec 22 – Carry climbing gear to base of glacier 5050m, return to basecamp
Dec 23 – Summit Bolivar 5720m
Dec 24 – Barrios camp 3900m
Dec 25 – Laguna Corazon 4030m
Dec 26 – Las Lagunas 3700m
Dec 27 – hike to Yerbabuena, 4×4 to Santa Marta 0m
Dec 28 – Buffer day
Dec 29 – Flight out 
Dec 30 – Arrive Seattle

The highest peaks in Colombia are Pico Cristobol Colon and Pico Bolivar, located in the Santa Marta Mountains in the north of the country. Pico Bolivar was first climbed on Feb 2, 1939 by Krause, Praolini, and Pichler approaching from the south and climbing the east ridge. Krause surveyed the summit elevation using a hypsometer and measured 5520m [1]. This measurement was later corrected by the Augustin Codazzi geographic institute to 5794m, though the error bounds on the measurement were not given [1].

Pico Colon was first climbed a month later on March 16, 1939, by Praolini, Bakewell and Wood. This climb was part of the larger American Cabot expedition, which conducted a trigonometric survey of the area considered the most accurate ever done. That expedition measured Pico Colon 5775m and Pico Bolivar 5773m [2]. For that expedition surveyors took an Abney level to the summit of Colon and measured that Bolivar was 2m shorter. Thus, since 1939 Pico Colon has been recognized as the highest peak in Colombia. 

The next measurement was conducted in 1989, when Cristóbal Von Rotkirch and Juan Pablo Ruiz used an altimeter brought to both summits to measure Colon 5775m and Bolivar 5790m [1]. In general altimeters have much higher errors than an abney level or trigonometric survey, so this merely showed the peaks were of similar elevation. Colon was still recognized as the country highpoint.  

More recent satellite-based measurements from the 2001 SRTM mission [4] show Colon higher than Bolivar, though errors can be up to 16m for sampled points and measurements were only taken every 30m horizontal spacing. So those measurements are not definitive. The most accurate measurements to date have been the 1939 trigonometric survey. 

Since the late 1990s this mountainous area has been largely off limits to climbing from a combination of local religious beliefs, political instability, and drug trafficking. The area around the summits is controlled by four indigenous groups: Kogi, Arhuaco, Cancuamos, and Wigua. In their culture the Santa Marta Mountains are the heart of world, and the high peaks area is called Chundua. It has historically been their belief that it is forbidden to touch the glaciers of the high peaks for religious reasons. 

The last ascent of the peaks before this time of closure was in 1998 when Juan Pablo Luis climbed Pico Bolivar.  

In 2015 Petter Bjorstad and John Biggar secured permission from the Kogi to climb both peaks. They climbed Colon but not Bolivar. A handheld GPS on the summit of Colon recorded an elevation of 5730m (WGS84 EGM96 geoid). This is approximately 5725m in Colombia 2004 geoid (see trip report here: https://www.ii.uib.no/~petter/mountains/5000mtn/Cristobal/Trip-report.html). Handheld GPS units generally have vertical accuracy around +/-10m. 

In 2017 Bjorstad and Biggar received permission to attempt to climb Bolivar, but the trip ended up not being successful (see trip report here: https://www.ii.uib.no/~petter/mountains/5000mtn/Bolivar/Trip-report.html).

In 2018 Mountain Madness secured permission to lead guided trips to both peaks. However, before the trips started a group of independent climbers attempted to sneak in to illegally climb the peaks. They were turned around and the indigenous communities revoked permission from Mountain Madness. 

I had been in communication with Mountain Madness in 2018 and was ready to climb Bolivar and Colon, but was then told in October about the independent group trying to sneak in. Mountain Madness told me they still had permission to climb Simmonds peak, the third tallest in the country, but not Colon or Bolivar. I told them I would wait since my goal was to climb the country highpoint, not a shorter mountain. 

In 2022 the Arhuaco community began giving permission to several Colombian climbing groups. Cristian Alarcon led six trips to Pico Colon between 2022 and early 2024. Bolivar still remained unclimbed since 1998. Cristian measured an elevation of 5721m on the summit of Colon in 2022 with a garmin inreach (converted to Colombia 2004 geoid, error bounds unknown). 

I heard about the successful Colon trips in early 2024 and started trying to get on one. I was originally in communication with Andean Adventures, another group that had secured permission. They originally said they had permission to climb Colon but not Bolivar. I think this was because Colon was thought to be the country highpoint and is a significantly easier climb than Bolivar. 

I suspected it was unclear which peak was the true country highpoint, given potential melting of the summits since 1939, so I tried to do some research in advance to make sure I would climb the correct mountain. I found one of Cristian’s pictures on Facebook (Fig 1.) taken from the summit of Colon looking at Bolivar in January 2024. The sky was clear and many smaller peaks were visible in the background.

From my surveying background in WA I’ve learned to use a special piece of photo analysis software written by Edward Earl called Geopix [5]. This software allows you to find the relative elevation between the camera and a peak of interest in a photograph. You have to enter the coordinates, elevations, and pixel locations of background peaks and the peak of interest. Then it outputs the relative height of the peak of interest above or below the camera. It accounts for image distortion and atmospheric affects. 

I’ve validated the accuracy of the software on over a dozen peaks in WA using dGPS, Lidar, and theodolite measurements and it is very accurate.

I consulted Johnathan de Ferranti for help identifying background peaks and elevations and Nick R helped apply image distortion corrections in advance. I entered 14 background peaks and found Bolivar was 7m +/-2m taller than Colon (95% confidence interval). With this data I definitely would not join a trip that only climbed Colon. Only a ground survey would definitively determine which peak was higher, but I was very confident Bolivar was the true country highpoint. The peaks must have melted down at different rates since 1939, with Colon melting down faster. 

I mentioned this to the WhatsApp group of interested climbers, and the guide eventually agreed to climb both peaks and let me measure them. I said if we go in December I could join and measure the peaks, but I had to teach in January. 

However, the group then ended up scheduling their trip for January during my winter teaching quarter. So I couldn’t join. They would still climb both peaks based on my measurements though. I was pretty disappointed. 

But another friend got in touch with Cristian and we were able to organize a trip during my Christmas break in December when I was free. Cristian had permission to climb both peaks and was on board with me bringing professional surveying equipment up to determine the true country highpoint. 

Another peakbagger, Brian, emailed me asking about joining my trip, but I wanted to keep the group small for the technical climb up Bolivar. I mentioned to him I suspected Bolivar was taller, and I planned to measure both peaks to be certain. He then wanted to climb both peaks also. 

He ended up joining another scheduled climb from Andean Adventures that would start a few days before my scheduled trip. Because of my analysis, that group also planned to climb both peaks. 

So it appeared what had been several years of climbers just climbing Colon was suddenly changing into all climbers planning to climb both peaks, all based on my Geopix analysis. The main problem, though, was that nobody knew the route up Bolivar except that it looked technical. 

I planned to bring a good ice and rock rack, technical crampons, and two technical tools to be ready to lead it if necessary. From pictures I’d seen I felt comfortable leading a way up Bolivar. 

I also planned to bring enough survey equipment to get many different measurements from different places. I’ve learned from surveying high altitude technical peaks over the past six years that the full survey plan never completely succeeds. There are always equipment failures or unforeseen difficulties. So I usually plan lots of redundant measurements. 

This time I would bring two differential GPS units capable of 2cm absolute vertical accuracy measurements (Trimble DA2 and Trimble Promark 220). Note: these units are what professional surveyors use and are orders of magnitude more accurate than consumer grade handheld units. I have access to them through Seattle University, where I am a professor, and from Trimble. I would also bring two abney levels with 10 arc minute accuracy. This is the same instrument used in the 1939 survey to determine Colon was 2m taller than Bolivar. I would have two phones capable of logging data from the DA2 and three spare batteries for the Promark. I’d also have inclinometers on two phones and a Garmin 62S handheld GPS. (The handheld unit would have high errors, but I figured I might as well bring it along since it could potentially provide corroborating data for which peak was higher if I brought it to both summits). 

My primary goal was to get 30 min or longer dGPS measurements on each summit with both units. I would take angular measurements between the summits with the abney levels. 

Additionally, at basecamp I would set up both dGPS units and take angle measurements up to each peak. This alone could allow me to determine absolute elevations. 

I would similarly set up the dGPS units at at least one location on the approach hike and take angular measurements with the abney levels and inclinometers. This would also allow for absolute height measurements. 

If I could climb both peaks I could additionally measure heights with the Garmin 62S. This would have higher errors but could help corroborate the relative heights. 

I knew probably not all of the dozen or so planned measurements would work out. But if even half of them worked that ought to be sufficient to prove which peak was higher. In general I will only publish a surveying result if at least three measurements are consistent. I really really don’t want to publish anything that’s incorrect. 

Ginge Fullen, Serge, and Thais would be joining. Ginge has climbed the most country highpoints of anyone in the world, at 174 before the trip. I was at 143 and Serge at 114. 

On December 12 I finished grading my last final exam at SU, then took a taxi to the airport for an evening flight out. I arrived in Santa Marta with Serge on Dec 13 in the afternoon, and we met up with Ginge. We’d last all met a few years earlier in Chad on a failed attempt for that highpoint. 

Dec 14

We met up with our guides Cristian and Jonathan to discuss logistics, and I described my analysis of the peaks. They were on board with targeting Bolivar as the primary objective and Colon after if there was time.

That afternoon I got a short walk up Cerro de la Virgen Millagrosa, which had a great view of town. That evening I took a practice measurement with each dGPS unit on the beach and verified everything worked fine. Sometimes I’ve found that these units take a while to acquire satellites when turned on halfway across the world. I wanted to ensure that didn’t cause problems up in the mountains. 

The units looks sort of suspicious to someone who has never seen them before with the external antennas mounted on tripods, and I’ve been kicked off mountains by military before. So that’s why I did my test measurements at night. 

Dec 15

In the afternoon we loaded up a big land cruiser, drove to the airport to pick up Thais, then continued to San Pedro. San Pedro is a mountain town at 1400m and definitely required a hard core 4×4 vehicle to access. It was lightly raining in the afternoon when we arrived and we stopped for the night at a nice guest house with sweeping views of the coast and jungle below to the west. 

Dec 16

In the morning we drove farther up into the mountains on rough and steep dirt roads. One switchback was too tight for the vehicle, so the driver actually drove backwards up the road, then on the next switchback continued forwards. We stopped at the national park boundary at a good turnaround spot marked by a large national park sign. The sign had the highest altitude of the park labeled as 5775m (referring to the 1939 elevation of Colon). 

The road behind the sign was too rough to proceed, and mules would carry our gear from there. 

We hiked up with day packs a mile to Yerbabuena village. There we meet members of the Arhuaco community who would be our porters once we got high enough that mules could no longer proceed. After lunch we hiked up the the road for 10 minutes, then proceeded on a trail through the jungle. After a few hours we reached a small house where we would spend the night at 2350m. 

The mules brought our gear to the house and the family cooked us dinner of chicken, rice, and plantains. Ginge refreshed our memories on playing backgammon and we got in a few rounds before dark.

Ginge keeps a log book of every climber he’s played backgammon with around the world. He was on Everest in 1996 and famous climbers like Rob Hall are in the book. I felt honored to get my name written down there. 

Dec 17

The next morning we hiked up trails through cleared land and jungle above the house. The locals have cleared steep slopes for cattle to graze. After a few hours the trees began getting smaller and eventually we reached treeline around 3000m. 

The slopes turned to grass and it appeared to be excellent grazing terrain. We made it to Las Lagunas, a small set of lakes, at 3700m in the late afternoon and put camp there. In general we would hike from 830am to 4pm most days. 

That set of lakes was as high as the mules could make it before the trail got too treacherous. That evening the porters hiked up to meet us. 

For dinner the porters would generally collect fire wood and make a fire to boil a huge pot of water. Then Cristian or Jonathan would add some dehydrated food and we’d all get a bowl full.

If these mountains ever get popular I expect cooking will need to be done over gas stoves instead of every group collecting firewood. But for a small number of groups it might be sustainable. 

That night the lead Arhuaco member Abel conducted a religious ceremony to allow us to touch the glaciers and climb the peaks. 

Dec 18

Above Las Lagunas the route became steep, with no real trail but some cairns marking the way. Our progress slowed to around one mile per hour and the porters helped carry all our extra gear. I still had a full backpack because I wanted to personally carry all the survey equipment so I could be extra careful with it. 

We climbed up to a 4200m pass and there got our first good view of the peaks in the distance. Colon, Bolivar, and Simmonds were all three clustered together as a big mass of white in the middle of the dark rocky surroundings. 

It was amazing to see such snowy peaks after hiking so far through the jungle. We took a break and I was able to take my first measurements there.

I took a dGPS reading of the elevation and used the inclinometer on my phone with 30x zoom to measure the declination from the summit of Bolivar down to the summit of Colon to the nearest 0.5 degree. 

By using the coordinates of each peak and my known dGPS elevation and coordinates along with the declination angle, I was able to measure Bolivar as 5-10m taller than Colon.

That was consistent with the Geopix analysis, which was good news. I was gaining confidence that Bolivar was indeed higher. 

We descended to a small stream below at 4000m for a lunch of tortillas, cheese, and dried meat. Then we climbed up and over a 4300m pass before dropping down to Lago Corazon, our camp for the night at 4030m. 

The high passes during the day and lower camps at night were great for acclimation. This followed the mountaineers mantra of climb high sleep low. 

There was an excellent sunset from the location as the valley drained due west down to the ocean. 

Dec 19

From camp we climbed back up to 4200m and started the longest day of the approach. This involved going up and over multiple passes and scrambling along the sides of peaks in between. 

A few of the scramble sections were steep and tricky, and unfortunately Thais slipped once and twisted her ankle. 

We all continued slowly and cautiously from there, and ended up pitching camp a bit before our intended destination of Bohios. We found a flat spot with water at 4200m just below Lago Herradura.

Dec 20

In the morning Thais’s ankle had still not improved, and she decided to turn around. Cristian, Serge, and several porters joined her. 

The remaining team would be me, Ginge, Johnathan, Abel, and five porters. One of the porters, Antonio, had brought his nine year old daughter Carolina along and they would also continue with us. 

We started up the pass and soon encountered another group hiking out. Brian recognized me and we spent a few minutes talking about the peaks. 

He had successfully climbed both peaks and said a phone GPS had shown Bolivar 10m taller than Colon. That by itself wasn’t definitive, since handheld GPS units can have high error, but it was another measurement in favor of Bolivar being higher. 

Brian gave me valuable beta about climbing on the south side of the rock steps of the East Ridge, and to be careful of very loose rock. 

In his group two guides with two clients had each climbed both peaks, and another set of clients and guides had just climbed Colon. 

We soon continued up to the pass and Herradura lake, then descended down the other side. By now we were following more of a trail, and we reached the Bohios camp in a few hours for lunch.

From there we merged with the Mamarongo route that Petter Bjorstad had taken in 2015, and the trail got better. We hiked back up to a set of lakes, and pitched camp at Laguna Carcuina at 4400m. 

Dec 21

We continued up the valley past more alpine lakes to the base of Simmonds peak. Simmonds Basecamp at 4800m had some nice stone wall campsites next to a big lake. I’ve heard mules can make it up to there from Mamarongo. If that route ever opens again it might be one of the easiest approaches. 

Past the camp we crossed a 4900m pass, then dropped steeply down to a small tarn at 4700m on the other side. 

We waited there for the porters to catch up, and it got pretty cold. Abel decided to make a fire, but there weren’t any sticks anywhere. There were plenty of dried cow pies, though. It took a lot of effort but we eventually got some flames going with dried grass and the patties. 

A few other porters caught up and we then hiked the short distance down to a large grassy flat area at 4600m. That would be our basecamp and room for plenty of tents.

Up until then every day had been dry, with clear skies in the morning and clouds building in the afternoon. However, that day it started snowing at 3pm and didn’t let up for the next five hours. We even had brief bouts of thunder snow. 

A few cm of snow accumulated and I was worried this might make the route up Bolivar trickier. We planned to take a rest day the next day, though, to help with acclimation, and we hoped that would also help the snow melt off the route. 

Dec 22

In the morning while there skies were clear I decided to take another measurement. I set up both dGPS units on a hill behind camp with a view of Colon and Bolivar. 

I took a one hour measurement with each unit, and measured angles up to each peak. I measured that Bolivar was 5m +/- 3m taller than Colon. Yet again, another measurement showing Bolivar the taller peak. 

We decide it would increase chance of success if we could ferry a load of gear up to the base of the glacier that day. That would let us save time and energy on summit day.

We sorted out gear into a few packs and the porters offered to bring gear up. Jonathan and Ginge rested in camp but I decided to join the porters. This way I could record a GPS track for us to follow and I could be certain I knew exactly where the gear was stashed. This turned out to be critical to our success. 

Abel led the way up a route that didn’t have any cairns and would not have been obvious. He’s been there before, though, so knew the route that worked.  

We reached the base of one lobe of the terminus of the Colon Glacier at 5050m and stashed the gear under a boulder. The porters decided to have a little fun before heading down, though. They used the ice axes to chop steps up a steep snow/ice slope and marched up in their rubber boots to a good viewpoint. Even nine year old Carolina joined in. 

I snapped some pictures before they returned. We made good time back to camp and I think tagging that altitude helped me with acclimation a bit more. By then all the snow had melted away from basecamp. 

We had an early dinner and snuck in a few hours of sleep. We were then up and moving by 11pm for our summit push. The goal was to get to the Colon Bolivar col at sunrise so we could have light to climb the route and time to get back down before the likely afternoon clouds and precipitation came. It was important to be able to see the route since it was basically unknown to us and route finding would likely be tricky. 

Dec 23

We made it to the gear stash by 1am and I was very happy to have joined the porters the day before. Jonathan had remembered an alternate route up the glacier from the last time he had been there, but the gear was stashed in a different location. That could have easily proved a show stopper if we couldn’t find our gear. 

I located the gear under a boulder and we were roped up and moving by 145am. 

We followed Abel’s chopped steps up a steep slope, then continued up the glacier. We made quick progress with Jonathan in the lead, soon joining up with his old route. 

We wove around many crevasses, and a few bridges were quite thin. I bet the route will soon get difficult if those bridges collapse. Luckily we were there early season (the dry season is December through February). The recent snow covered up evidence of the previous group, but we made good progress through the crevasse maze. 

The slope soon steepened and we eventually reached the ice and snow slope just below the Colon-Bolivar col. A bergschrund spanned most of the face, guarding access to the north face of Bolivar. But there was a narrow bridge just below the col. 

Above the bridge was a short stretch of WI2 ice followed by steep snow. We each got out two technical tools and I belayed Jonathan up. After a full 60m pitch he reached the top. Our plan was to all three climb on one 70m rope to increase efficiency, and the two followers would tie in ~8m apart on the end.

This meant the third person needed to be extra careful not to fall and pull off the second person. But in general the route was easy enough that this was not a problem. 

We quickly climbed up to the col by 515am and caught the first hint of alpenglow to the east. We approached the east ridge of Bolivar, and it was indeed non trivial. Two rock steps guarded the bottom of the ridge, with an easier snow slope above. If we could get past those rock steps the summit would be ours.

The steps were each around 30m tall, and the lower one looked around M4 to me, with the upper slightly harder. They looked doable, but Brian had told us the rock was lower angle on the south face. I was more interested in the highest chance of success in getting elevation measurements than getting a first ascent of a new route, so I decided to try to find his route (which was led by Ricardo Rubio). 

The south face at the col level looked like a tricky exposed traverse, but there was a steep snow slope that dropped down lower. Jonathan set up a picket anchor and belayed me as I down climbed the slope with two tools about 30m down. By then there was enough light to get a good view of the route. Indeed, the rock looked much easier down low.

 It looked like a few traversing and ascending pitches would get us around the lower rock step. I couldn’t see beyond that but it looked promising. I yelled up that this was the route, and Jonathan and Ginge rapped down to join me. 

We had brought a 60m and 70m rope, since we knew it was a 60m rap to get off the col. We left the 60m rope on the snow slope to jug up on the return. We then continued climbing with the 70m rope. 

Jonathan took the first pitch, ascending a little then traversing on ledges to a belay stance about 30m away. We climbed in crampons since there was a mix of snow, ice, and rock on the route. It was low enough angle, through, we could stash our tools and climb with our hands. 

The plan was for me and Jonathan to swing leads to increase efficiency. I took over the next pitch, where I continued ascending and traversing until I reached a point just below the gap between the two rock steps.

The upper step looked difficult so I decided to stay on the south face and try to get around it like the first step. I continued down a ledge and around a corner and built a solid anchor at the end of the 60m pitch. It was important on the route to be careful with gear placements to reduce rope drag but still protect the followers on the traverses. 

Jonathan took the next lead that was purely ascending up a face of snow, ice, and loose rock. He dislodged one basketball-sized block, but I was belaying safely out of the way. Ginge and I then followed.

I ended up taking the next two pitches. I was very concerned about rope drag, so I first led a short pitch up to a horn where we found a rap anchor from Rubio’s team. Next I led a pitch straight up to the ridge crest at the top of the upper rock step. This was the crux of the route, I’d rate it 5.6 M2 and 35m. It was all rock with some vertical sections and some very loose blocks to be careful of. But it was most efficient to just climb everything in crampons. 

I saw below there turned out to be low angle snow ledges that continued traversing and ascending, and these would have been much easier. I later learned Rubio’s group had followed this easier route. 

So it appears our route ended up being a variation of Rubio’s route. I would rate our route 5.6 M2 WI2 and their route 5.4 M1 WI2. So, for future reference, that is likely the easiest way to the summit from the Colon-Bolivar col. 

On the top of the ridge I slung a boulder, which I planned to be our rap anchor. I always plan my descent as I climb, and rapping from an anchor here would just barely reach the good horn below.  

Jonathan took the last pitch, climbing up the mix of rock and snow on the ridge until it gave way to a continuous snow slope to the summit. 

When we all regrouped I made a proposal. I knew I needed at least a 30min dGPS measurement on the summit to get good accuracy, but Ginge doesn’t do as well at altitude and would likely not want to stay up there for 30min. I proposed I would solo ahead quickly as Ginge and Jonathan took their time. Then I could get a head start on the measurements and hopefully get the full 30min.

We all agreed, so I took out the two tools and made my way up. I was wary of a potential cornice on the north face, so kept my distance. Luckily I noticed the exact summit was cornice free, though still a sharp snow summit. 

I leveled out a patch on the summit and quickly mounted the DA2 on the mini tripod so the bottom of the antenna was level with the highest point of snow. I started logging data at 11:15am. 

I had originally planned to bring both dGPS units, but given the technical nature of the route I’d opted to just bring the lighter weight DA2. Indeed, the summit was sharp enough that it would have been tricky to deal with the big pelican case and extra equipment of the Promark up there.

While logging data I pulled out the abney level and pointed it towards Colon. I measured a 40 arc minute declination down to the summit. Bolivar was definitely the country highpoint. Every single measurement was consistent about that. Using known coordinates from each summit I would later calculate a distance of 655m between summits. Using a bit of trigonometry this meant Colon was 7.63m shorter than Bolivar. 

The dGPS quickly converged to a preliminary altitude of 5720m. I knew I’d later need to post process results to get the final value, but this would be pretty close. It was way lower than the 1939 height of 5773m!

I noticed the Colon summit appeared to have a rock sticking out just a few meters below the top. This was not visible in any pictures from Petter’s 2015 trip. There were also rocks not far below the Bolivar summit. It seemed like the peaks must have been melting down a lot in recent years. 

After 15 minutes Ginge and Jonathan reached the top and we took a bunch of pictures. I realized that the south ridge of Bolivar looked like a mellow non-technical snow slope. That would definitely be the easiest ascent route, though it may be difficult to get permission for that route. 

As Ginge was taking pictures his phone dropped down the south face and I was sure it was gone forever. But amazingly it came to rest on a mellow slope about 50m down. Jonathan built a picket anchor and rapped down to retrieve it. 

For better or worse, this retrieval bought me enough time to get a full 45min measurement on the summit, which was great for increased accuracy.

With the phone retrieved we packed up and headed down. I led to the rock horn, which I threw a few slings over and clipped a locking beaner to. I rapped down all the way to the horn below. It was nice to have the 70m rope, since I don’t think a 60m would have reached. 

Luckily there were no issues with rockfall or snagged ropes. We reused the other team’s anchor at the horn, and rapped all the way down to a good ledge below. We made sure to stay in the snow and ice to avoid bringing more rocks down. 

From there the descent was a bit tricky since it was mostly a traverse route. I slung a horn for a short 10m rap into a snow gulley, then led across a ledge system around a corner. From there Jonathan built an anchor up high and we were able to diagonally rappel exactly back to our starting point at the base of the longer snow gully. 

Our stashed rope was still there, so I put a prussik on it then climbed up with one tool using the prussik as backup. 

We topped out at the col around 330pm, with two hours of daylight left. I wanted to then climb Colon to get a final measurement there. Unfortunately there was not similar excitement among the group of tagging on another peak after Bolivar. I had already definitively measured Bolivar higher with many corroborating measurements, and climbing Colon would only stand to get slightly tighter error bounds on its elevation. And it was now pretty much just a sub peak of Bolivar anyways. 

It would be too dangerous for me to climb solo up the glacier, so we stuck together and all headed down. We rapped off a dead man picket to get off the col, then followed our tracks down the glacier. 

We exited the glacier exactly at sunset, then loaded up our packs for the hike out. I was very happy to have my gps track to follow on my watch since fog rolled in and even with headlamps we were essentially navigating in a whiteout to get down.

We stuck together and eventually made it to camp around 8pm for a 21 hour day. The porters had hiked down the valley a bit that day to find fire wood, and they cooked up some tasty rice and meat. 

Dec 24

The next morning we slept in a bit and felt recovered enough to start the hike out. We hiked over the 4900m pass, then all the way down to Bohios camp to end the day. It rained all afternoon and we realized we had been very fortunate with the weather on our summit push. It had been sunny pretty much all day, which is unusual compared to all the other days of the trip.

Firewood was much more plentiful down at 4000m. Carolina did an excellent job climbing up a boulder with a machete to hack off some dead tree limbs for wood. I don’t know too many nine year olds in the US who could do that (or would be allowed to by their parents). 

Dec 25

Christmas morning the weather cleared and we hiked back up to Laguna Herradura. A group of three condors were curious and flew near us, allowing for some excellent pictures with my 30x zoom camera. 

We then descended down past an old plane wreck, and continued to Laguna Corazon. Luckily the tricky slabs had dried out enough to be safe. I would not want to hike that section in the rain or snow. 

I managed to get some good timelapse videos of sunset, and we hung out around the fire late into the night.

Dec 26

Some of the porters left early that morning and pushed all the way home to Yerbabuena, dropping their packs off at Las Lagunas for the mules to pick up. We took our time and called it good at Las Lagunas after a 5 hour hike. 

In the evening a group of porters from Andean Adventures hiked up and joined us for dinner. They had brought a lot of gear up with mules and were planning to stash it at higher camps for an upcoming trip. I think this is the trip I had originally planned to join before the dates got switched. 

Most of the porters ended up hiking home to Yerbabuena after leaving their packs at camp. Only Abel, Jonathan, Ginge and I remained. It rained again that afternoon but cleared up by sunset. 

Dec 27

In the morning Diego arrived with the mules and we helped him load up. We then hiked out all the way to Yerbabuena, and took a 4×4 back to Santa Marta that evening. 

Results 

I was able to process my summit measurements with TimbleRTX and CSRS-PPP, which gave consistent results. All measurements were consistent and my final reported results will be those with the lowest errors. This is the TrimbleRTX elevation for Bolivar. For Colon this is the elevation from the abney level relative measurement from Bolivar to Colon plus the Bolivar absolute measurement. I will report results as orthometric height using the Colombia 2004 geoid, the standard geoid used currently in Colombia. Error bounds are 95% confidence intervals. 

Bolivar 5720.42m +/- 0.08m
Colon 5712.79m +/- 0.87m

The error bounds for Bolivar are given by TrimbleRTX. The error bounds for Colon assume +/- 5 arc minute measurement error from the Abney level. 

Corroborating measurements, all consistent, of height Bolivar above Colon are as follow (see box-and-whisker plot of results in Fig. 2):
Geopix: 7m +/- 2m (95% confidence intervals)
Abney level from base camp relative to Promark 220 and DA2 dGPS measurements: 5m +/- 3m
Handheld GPS from Dec 18 taken to both summits: 10m (unknown error bounds, assumed +/-10m)
Inclinometer from hike in, relative to dGPS DA2 measurement: 5m-10m.

This means Pico Bolivar is the country highpoint of Colombia and the fifth most prominent peak on earth. Pico Colon is the second highest peak in Colombia with a prominence around 130m.

This also means that Bolivar has melted down 53m and Colon has melted down 62m since 1939. See Fig.3 for elevations of both peaks over time. Note that estimated error bounds are high on the 1989 measurement, so it is likely that Bolivar and Colon maintained height between 1939 and 1989, but started rapidly losing elevation in the 1990s or early 2000s. This is consistent with when historical modeled temperatures began significantly warming and precipitation began significantly decreasing on Pico Colon and Bolivar, based on ERA5 dataset (Fig. 4) [3].  

The highest visible rock on Colon was approximately 2m below the summit and the highest visible rock on Bolivar was approximately 3m below the summit. Thus, if the melting continues, Bolivar will remain the country highpoint.

Note: I am currently working with Colombian glaciologists at IDEAM to publish these results in a formal document.

Link to technical document detailing measurement methodologies and results: Pico Colon and Pico Bolivar Elevation Survey 2024

References

1. Ruiz, J. “The Roof of Colombia,” Alpine Journal, 1990, p 150-156.

2. Echevarria, E. “The Santa Marta Range, Colombia,” Alpine Jouranl, 1963, p 209-215.

3. Hersbach, H., and Coauthors, 2020: The ERA5 global reanalysis. Q.J.R. Meteorol. Soc., 146, 1999–2049, doi: 10.1002/qj.3803. Accessed at https://www.meteoblue.com/en/climate-change/pico-crist%c3%b3bal-col%c3%b3n_colombia_3685669 

4. Farr, T. G. et al., 2007, The Shuttle Radar Topography Mission, Rev. Geophys., 45, RG2004, doi:10.1029/2005RG000183. (Also available online at http://www2.jpl.nasa.gov/srtm/SRTM_paper.pdf)

5. Earl, E., 2017, “GeoPix Photo Analysis,” available for download at https://github.com/ericgilbertson1/PhotoAnalysis/tree/main

© 2025, egilbert@alum.mit.edu. All rights reserved.