Jungle Construction at its Best with State-of-the-Art SOYP* Engineering

The closest Home Depot is 3300 km (~2000 miles) away in Miami. The nearest cement truck is on the other side of the Andes Mountain range. It requires two boats and a three-wheeled, motorcycle-driven cart to move materials from the nearest town. And when you arrive, there is no road to the construction site. There is also no lumber-yard, no back-hoe, no crane, no cell signal and no electrical grid. But there are also no building plans, no permits and no inspections.


These are just some of the challenges (and advantages!) of building a field station in the heart of the rainforest. When a violent windstorm took out some of our buildings at the Santa Cruz Forest Reserve in November 2020, I decided to document the rebuilding process, so as to share how remote field stations are actually constructed. Mind you, the Santa Cruz Forest Reserve is by no means truly "remote", and does not face some of the nearly unimaginable logistic, cultural and political challenges of other field sites (check out Searching for Pekpek, by Andrew Mack, a grad school colleague of mine who worked in Papua New Guinea).


Back to the Amazon however! We needed to build a completely new kitchen and caretaker housing at Santa Cruz. Simply repairing the damage wasn't an option as the photos below illustrate.

Top: left to right: The remains of one set of bathrooms & showers; the kitchen freezer narrowly missed a direct hit, but was still trashed; another view of the bathrooms.

Bottom: left to right: A flattened trail shelter; the water tank at the kitchen (the polypropylene tank amazingly popped back into shape when the tree was removed); the mangled remains of a bunk bed in the caretaker quarters.


I purchased my tickets to Peru, excited about the opportunity to travel again, despite the challenges of flying to Peru during the on-going pandemic. In preparation for my arrival, our manager, Fernando Rios, sent out a small work crew to clear away the building and tree-fall debris and prepare the site for the new construction. Opening up the trail leading from the river to the field site was, of course, the very first step.


I arrived in Iquitos early on 24 November, and within hours, most of the money I had brought was gone. Well, not truly gone, but transmogrified into cement, rebar, roofing, nails, files, chainsaw chains, gasoline, hammers, machetes, food, medication, TP, PVC tubing and a few dozen other miscellaneous and sundry items. The building materials and supplies were loaded on one of our boats in Iquitos that same day, and at 7 AM on 25 November, I and the rest of the work crew were ready to head to Santa Cruz. It was the first time I had seen our multi-talented crew since mid-March and the start of Peru's pandemic shutdown, so there were fist- and elbow-bumps all around. After months of enforced inactivity, the crew were as eager to get underway as I was. Putting some additional money into their pockets before the Christmas holiday season was an added reason for pushing ahead with the reconstruction as soon as possible.


Day 1 (25 November) of construction


With supplies already loaded, all it took was my arrival at the Bellavista port in Iquitos for the team to be underway. I'm usually the last to arrive for such departures. Not because I'm chronically irresponsible, but mostly because I'm frantically trying to finish up as many emails as possible before being out of cell phone range for an extended period of time. Bandwidth is seriously limited in Iquitos, and with many students trying to take their classes on-line, it slows to snails pace by 8 AM, and is invariably very slow late into the night. Generally I'm up at 4 AM to take advantage of a few hours of "reasonable" internet speed. Forget about attaching videos or full size photos though!


The first leg of our journey was a 35 km (~22 mile) boat ride from Iquitos to Indiana. At least we didn't need to go to New York, San Francisco, Canada, France or Ireland, all of which are much further downriver on the Amazon (sometimes you just have to haul out the old atlas to assign names to communities!). Upon arrival in Indiana, all the materials needed to be carried from the port area up to a 3 meter (9') wide road, loaded onto carts and transported some 30 minutes to the town of Mazan.


This is a land where humans battle (and profoundly respect) the implacable whims of the mighty Amazon River. Why hasn't anyone devised a motorized pulley system to haul heavy loads up from the river? The answer is easy - the river could carry it all away overnight. River levels change by up to 15 meters (48') over the course of the year, and the "port area" could be a few steps distant from the road, or several hundred meters (or yards) away. Sand-bars come and go, islands appear and disappear over the course of a year or two, and large sections of riverbank cave in on a regular basis. The Amazon River simply has "a thing" against permanent structures and it goes where ever it wants to go. As such, we're stuck with that most resilient and reliable of technologies - human muscle.

Workers carry sacks of river sand and bags of cement up from the edge of the Amazon river. The sand is for a school construction project in Indiana, the cement is for our field station. Each bag of cement weighs 42.5 kg (93.7 lb). The guy hauling 2 bags of cement is carrying 187.4 pounds. Uphill. In sandals. In 90% humidity and 30C (85+ F) temperatures. The sacks of sand are probably 70-80 kg (154-176 lb). I was sweating just taking the video!


While materials were being loaded onto "trucks", I headed on to Mazan with everyone's personal gear, and to meet up with Emerson Torres, the supervisor of the Santa Cruz Forest Reserve. We made some last minute purchases at small shops in Mazan, and waited for the building materials and rest of the crew to arrive. At our port on the Napo River in Mazan, our boat was already waiting, but it was the tricked-out yacht that pulled in that really caught my eye. The yacht had it all - entertainment system, organic gardening area, laundromat, fuel depot, depth finder, twin propulsion and more.

Top - from left: Our transportation from Indiana to Santa Cruz, the wooden boat at right; An Amazon river (pamacari-style) yacht pulls into port with the crewman at the prow manning the depth-finder.

Bottom - from left: Yes, that really IS a satellite TV dish on the back of the yacht - just above the chicken coop and generator, and behind the herb garden and fuel storage area. Twin peke-peke motors provide reliable but ear-splitting propulsion. Right - another view of the port area with some of our supplies on the dock waiting to be loaded.


Eventually the rest of the crew and the building materials arrived, we loaded up as much material as would fit without sinking the boat, and the first trip was underway. It took two trips to carry everything since 40 bags of cement, 45 sheets of roofing, 30 rebar and a wide assortment of other materials, crew and food weigh a lot! I stayed behind for the first trip to keep an eye on things. The crew returned some two hours later, and after the last items were loaded and we were underway, it was definitely 5 PM somewhere.

Our crew loading cement onto our transport boat in Mazan.


From left: The "truck" with our roofing materials arrives at port; a view of the houses overlooking the port (in all fairness, the closest house is abandoned, the next one over is available for rent - "river view"); we start the second run with construction materials at "5 PM" out to the Santa Cruz Forest Reserve up the Mazan River.


Arriving at the field station port area, it was still a 1.8 km (1.1 mile) hike inland to the construction site. Every bag of cement, every sack of sand, every re-bar, every roofing panel and every nail, bolt, screw and washer made that trip on someone's back or shoulder. Usually not mine, however. After sitting in front of a computer for most of the past eight months, I was surprised at how out of shape I actually was. I elected to stay at our facilities down at the river edge, so I had a couple of extra trips to make each day. If I wanted coffee in the morning, it was a 6 AM hike up to the construction site. At the end of the day it was another hike back to the river, and sometimes a couple of round trips during the day as well. I quickly started to get back in form.


Arriving at the field station, the remains of the old kitchen and bathrooms had been cleared away. Young fellows from the nearby community were already starting to carry bags of cement and re-bar up from the river, and the work crew started to get immediately to work.

Top from left: The cleared kitchen site, ready to start work; the rubble of the old kitchen & caretaker housing - some of this material was repurposed during the rebuilding process.

Bottom from left: Salvaged boards from the old structure. The wood is andiroba (Carapa guianensis), a very durable and insect-resistant wood. A little trimming on the ends and these boards will serve for decades; sacks of sand already on site, hauled up from the river prior to my arrival.


After consulting our SOYP* software, we laid out the orientation and dimensions of the new structure. Our mason (and expert snake finder) Edwin started assembling reinforcing bar supports for the concrete columns of the building, and others began the work of excavating the footers for the columns. Not an easy task when you are working in clay soils that stick to everything and you have to continually scrape the clay off your shovel. The rest of the crew headed out to select and prepare downed trees for lumbering - clearing away the branches and vines, and clearing away storm debris on the trail to the selected trees.

Top from left: Edwin starts assembling the re-bar columns (1 & 2); Antonio, the local caretaker, works on digging a column footing.

Bottom from left: Juan Luis (foreground) and Walker (back) work on additional footings; By the end of the afternoon, nine column footings, each about 80 cm x 80 cm x 80 cm (~30" x 30" x 30"), were excavated.


Day 2 (26 November) of construction


The heavy lifting really started on this day as we settled into a regular routine. Coffee at 6 AM (or earlier), followed by a couple hours of work, then a large Peruvian Amazon breakfast, more work, a large Peruvian lunch in the early afternoon followed by a digestive pause, then more work till about 5 PM. After that, there was time to clean up before the sun went down at about 6 PM. After a bit of reading, I was in bed most nights by 8 PM, but also up at 5 AM when the bird chorus started to overtake the frog chorus.


Birding-from-bed in the Amazon is always an interesting experience - various owls, nightjars, potoos, ibises, falcons and other species chime in, and there are usually some shrieks, squeals and raucous maniacal laughs that can't be precisely identified, but which add to the many Amazonian legends of ghosts and spirits, both human and animal. As the day shift takes over, chachalacas, nunbirds, wrens, orioles, antbirds, parrots and many other bird species start their daily chatter.


But what, you ask, is a Peruvian Amazon breakfast? This is clearly the most important thing to know! For starters, it isn't something that is likely to be on the menu at IHOP or Denny's any time soon. It is almost invariably a soup or stew, and definitely hearty and tasty! For the record, I do NOT post pictures of my meals on Facebook, and rarely take pictures of them to start with. This is a rare exception. The first grouping are all piping hot breakfast dishes, and there was always a selection of spicy Amazon peppers to heat things up even more.

Top from left: Mealtime with the work crew; a beef, noodle, carrot and plantain stew; chicken soup with potatoes, carrots, pasta shells, noodles and cilantro.

Middle from left: Beef soup with potatoes, carrots, noodles and plantain; a variant of beef stew with noodles, potatoes, cilantro and plantain; majas stew (a large rainforest rodent - Cuniculus [Agouti] paca) with noodles, potatoes, ginger, carrots and plantain.

Bottom from left: Mazamorra of shovelnose catfish (Pseudoplatystoma sp.) with fariña (toasted cassava), plantain and cilantro; fish soup (Semaprochilodus taeniatus) with fariña and plantain; Chicken soup with potatoes, carrots, noodles and cilantro.


A couple of lunch meals (invariably with rice!) and a decidedly non-Amazonian breakfast at Le Bateau Doré, a French-Belgian restaurant in Iquitos...

From left: Roasted chicken with spicy Peruvian cream sauce, split peas, rice, egg, and beet and heart-of-palm salad; Catfish with beans, rice and vegetable salsa; the "countryman's breakfast" at Le Bateau Doré.


The serious chainsaw work got underway on this day. We needed a large quantity of timbers of various dimensions for the construction, as well as boards for concrete column forms. Our master carpenters - Emerson Torres and Segundo Rios, pulled out the SOYP* software, ran the calculations, and came up with a list of how many beams of each dimension we would need. The previous day, they had selected a large fallen caimitillo (kai-mee-tee-oh) tree (Pouteria sp., Sapotaceae) as "the tree". It was close to the construction side, and had fallen across a trail, so cutting it up served two useful purposes.

Clockwise from top left: Emerson (orange shirt), Segundo (cap) and Rider (blue shirt) inspect the fallen trunk; the trunk sectioned into 5 meter (~16') lengths; another view of a section of trunk with Segundo and Armando (light blue shirt); rot at the base of the tree, a contributing factor to its fall; white latex, characteristic of plants in the Sapotaceae family, oozing from a cut section of trunk.


The largest section of trunk closest to the base of the tree was a bit of a challenge. Even with a 36" chainsaw blade, we couldn't cut all the way through the log. It required the use of makeshift wedges to spread the the two halves of the trunk apart enough to be able to insert the chainsaw body (as opposed to just the blade) far enough to make the deeper cuts that would split the log in two. This was tough work. In the next video, Segundo starts to cut the first section of trunk in half.


Preparing a tree trunk often takes longer than the eventual cutting of beams and boards. Splitting a log is the first step - you need to have a flat surface on the wood to work from.


After the initial cut with the chainsaw to the depth that the blade would reach, Armando used wedges of wood to start to spread apart the two sides of the partially split trunk.


Then it was more cutting with the chainsaw, and larger wedges to spit the two halves even more.

Rider (with chainsaw) and Armando (sweat flying!) work at getting that first trunk section split.


As the trunk halves were split apart to a greater degree, it became possible to get the body of the chainsaw itself between the two parts, giving the blade sufficient reach to make the final cuts.

Of course, right when I hit "pause" on the video, Rider casually reached over and gave one half of the trunk a shove, and over it toppled. I missed the "dramatic" moment! In reality, the following videos where the other half was tipped over (after some trimming) was much more dramatic.

Segundo trims off an offending bulge on the 2nd half of the trunk.

And, with a little help from his friends, over it goes!


This is very heavy and hard wood that we were working with, and chainsaw teeth get dulled with use and require frequent sharpening. Fortunately we weren't working with bloodwood, which has a very high silica content, and is like using your chainsaw on sandpaper. As is often the case, there is more than one way to clean your teeth. The traditional method using regular chainsaw files as illustrated by Rider;

and the lazy man's way, using an angle grinder with a metal cutting blade (running off of solar power), as demonstrated by Segundo.

The above videos were not intended to serve in product placement, but for what it is worth, Stihl chainsaws are hands-down the best brand available in the Amazon, and the only brand that we use. It's hard to beat German engineering and manufacturing!


Back at the ranch, the rest of the crew were clearly slacking off (NOT!) - the rebar column sections had multiplied considerably (tribbles?), and a horde of Labidus sp.