Based on feedback on a comment in my last diary, I found out that there's interest here in a project I'm working on: the construction of an earth-sheltered (underground) home in Iceland. We're still in the early stages, but I'd be glad to talk about the concept, what we've learned so far, etc, and how we plan to proceed. :)
For most of the past year, I'd been looking for land to build on... and I must say, the search was frustrating and disappointing. Lots usually fell into one or more of several categories: 1) Too small /too in town; 2) Too large and unwilling to go through a rezoning to break up the lot; 3) Pre-financial-crash prices; 4) Simply too far from Reykjavík (I work in town); 5) Too high up (roads become more dangerous / difficult to get through in the winter with increasing altitude); 6) Uninteresting - just plain flat land. At one point, there was a piece of land I was quite interested in, and the owner was willing to do rezone. But Reykjavík took 5 months just to be able to tell me whether it'd be possible to rezone, and by then the owner had changed his mind on selling. Frustrating.
But then I found Stapagljúfur.
(Above: The entrance to Miðdalur. Stapagljufur is right inside the valley, from the east of the road to the bottom of the river canyon)
Stapagljúfur ("Mesa Canyon") is 8,35 hectares (about 20 acres) near the entrance to Miðdalur (Middle Valley), the first valley in Hvalfjörður (Whale Fjörd), about half an hour from Reykjavík (20 minutes if they build Sundabraut). It directly borders the greater Reykjavík area but is in Kjós instead, a countryside area with a much more relaxed back approach to dealing with construction and land-use changes than the hyper-strict Reykjavík construction administration. It lies on a gentle south-facing slope, which suddenly plunges down into a 20-meter (60 foot) deep canyon with a small, glacial-melt river at the bottom. In some places it's a 35-degree slope; in others, sheer cliffs. The half-kilometer long canyon begins just before our property and ends just after it, so we and the property across the river (Tindstaðir) basically own the whole thing. Little babbling creeks run through the land, creating wetlands, tiny waterfalls, and still pools. One has carved out a 6-meter deep ravine of its own. Sizeable marshes lines the creeks, though most of the property is high and dry grassland. A 4-meter tall hill between the flatland and the canyon creates a ledge on the canyon slope perfect for building a home. Crowberry plants can be seen scattered throughout. A well-maintained lowland road leads back to the main arterials to Reykjavík. Mountains to the south block out Reykjavík's light pollution.
It's not perfect, of course. Direct sunlight doesn't hit the land at all until the start of February, and not during midday until late February, due to the 850-meter high mountains to the south. And having been grazed down by sheep for literally a thousand years, erosion is a problem in some areas, with visible landslide scars in a section of marshy gravel. While better than the western slopes of Kjalarnes where cold winds race down Esja to the sea, it can still get sizeable gale-force, even hurricane-force winds flowing down from the valley to the fjörd during storms.
But that said, I think it's perfect for our needs. Even the neighbors (great people, btw) are supporting our crazy idea.
Wait, what crazy idea? Well, here's our concept drawings of what we want to build there:
Okaaay, so... unusual? Check. But if I was satisfied by traditional houses, I wouldn't have wanted to build in the first place, I'd just have bought. My dream is an earth-sheltered home in the countryside.
Let's go into the design for a minute here. While it looks like a cave, it's not to be bored into a slope. That'd be all too difficult, expensive, and dangerous. Instead, we plan to take the natural shelf on the canyon slope, dig it down even further while levelling it out, build there, then fill in over the house and replant. We were initially planning to use shotcrete, but according to the people our architect consulted, it's apparently over 4x more expensive than regular concrete, so instead we'll be building with traditional slabs and then texturing them, possibly with an acid wash. Sand will be sourced from neighboring properties, and gravel will be sourced from our own land, where an erosion-caused landslide exposed a gravel deposit (afterwards we'll shore up the it up and replant). We want to use basalt fiber rebar instead of steel. Basalt fiber - simply melted and stretched basalt (what could be more Icelandic than that?) held together with a binder - is a composite related to fiberglass (but with better material properties and durabilities). Steel guarantees that concrete will begin to "spall" eventually -30 years, 50 years, 75, 100... sooner or later,the steel will rust, and rust is nearly ten times larger than steel. Even coating the steel in epoxy doesn't seem to stop it. It's a shame to think of how the Romans built concrete structures that lasted millenia while few of ours will even last a century. I seek to avoid that. Steel also means keeping the concrete at a very high pH, which unfortunately means maximizing the prescence of weak calcium hydroxide and minimizing the prescence of stronger carbonate, silicate, and aluminate compounds.
(Above: A 30-year-old bridge section in Florida, ruined beyond repair)
Of course that's not concrete's only problem. Cement has a high CO2 footprint. The calcium carbonate it creates isn't very waterproof, has limited self-healing capability, and tends to leach over time. Cement isn't a local product and all must be shipped in. Its raw materials have to be shipped in in the first place. Wouldn't it be great if one could use, say, trash to replace a lot of it?
Well, it turns out you can. The Romans used something called "pozzolana", which was a type of volcanic ash, in their concrete. Now called "pozzolans", they're a whole family of silica and aluminum minerals that have been exposed to high heat and undergo what's called a "pozzolanic reaction" in concrete. Though volcanic ash is most definitely an option around here, the type of pozzolan I want to use is fly ash, which here in Iceland has become a storage problem from Reykjavík's trash incinerator. Up to 60% of the cement should be replaceable with pozzolan. There is one primary disadvantage to pozzolanic cement - it's slower to harden. After 3 months, it has only 70% of the strength of traditional concrete. After 6 months, 85%. It takes years to reach the full strength. But the reaction keeps going, and consequently its ultimate strength is higher. It's also more waterproof (2x better after 6 months, and only increasing thereafter), has significant self-healing ability (due to the slowly ongoing pozzolanic reaction), does not readily leach, and has been clearly shown from ancient concrete to have extreme long-term durability.
(Above: Various pozzolanic materials available on the market)
Lastly, I additionally want to use microfibers in the concrete. These fibers -usually plastic - help ensure that there's tensile binding everywhere evenly in the concrete, not just near the rebar. They've also been shown to significantly reduce spalling in the case of fire by turning into conduits through which steam pressure can escape.
Where possible, of course, I'd rather not use concrete. Bedrock would be best. But what's down there? Honestly, I don't know. There's cliffs to the east and gravel to the west. I'd love to have a ground penetrating radar figure out but I don't think there's one in the whole country - and any "digging" to try to see what's down there, this time of year, would involve a pickaxe. If we find bedrock, we'd like to simply level it, polish it, and seal it as flooring (as well as use it for part of the wall if possible). If we don't find bedrock, flooring will be smooth volcanic scree held together with mortar.
You probably noticed in the concept drawings a total lack of drywall. This is no accident. Drywall is mold-prone if exposed to moisture, easy to damage, and makes maintenance difficult, as well as being an extra cost, both for building materials and labor. Instead, we plan to go for a steampunk look, with pipes and conduits sought and emphasized for their appearance. We may add some unnecessary meters and valves, perhaps some sections of clear tubing on water pipe, etc, to give the sense of living within a working "system". We want to avoid plastic as much as possible, but where necessary, we would like to either have metal piping on the outside or utilize some techniques we've read about to give it as metallic of a look as possible.
We considered infloor heating but decided instead to go with traditional geothermal radiators to reduce maintenance (we don't want leaks hidden from the eye, we want everything visible and accessible). A geothermal well to feed the valley is being dug right now and will be coming online next year (cold water will need to come from our own property). We're hoping to have the hot water line run to the house underneath the driveway, so that any heat it loses on the way helps melt ice. Any used hot water that doesn't need to be fed back into the system, we'd like to dump into a small outdoor pond to have a chance to cool before it runs into the river, to avoid depleting it of oxygen and to give us and the local birdlife a warm bath.
(Above: There's fish in that river -midsize fish in this section, larger fish (salmon) downstream. The last thing I want is to disrupt the river ecosystem!)
Being big into both cost-savings and recycling, we've already started doing the whole "pickers" thing and gathering up good, used materials in good condition. So far we got dozens of square meters (hundreds of square feet) of parquet flooring (for the master "bedroom" loft), six radiators and fittings, and a bunch of other stuff. Our best find has been large amounts of free, undamaged double-pane glass (it amazes me how willing people are to throw away perfectly good glass when they change out their windows). We plan to step up our picking activities once we have formal blueprints and know exactly what we'll need, including regular trips to the dump and roadtrips to hunt for natural materials (driftwood, interesting rock, etc) and disused buildings that we can salvage from (with permission of and fair payment to the owners, of course). We've purchased a 20-foot shipping container to store building materials in (which we plan to bury), but unfortunately it's not on the land yet and won't be for a month or so, so until then we're packing our apartment full of the stuff ;)
We want the project to be personal, to be something we can point to and say "we did that" - even if that means taking many years to complete the project. My fiance (a professional driver) plans to do the digging and is current taking a course in excavator operation. We're considering buying an excavator rather than renting so that we can take as long as we want, then selling it afterwards to raise money for finishing the interior; our only digging costs would then be maintenance, depreciation, and fuel. My fiance has worked a little with concrete before, but we're looking to have his father (who's worked extensively with it) give us lessons and to spend time volunteering for practice. In accordance with Icelandic building regulations, there will also be a professional, licensed construction manager who supervises the project. We've got a number of friends and family members who've volunteered to help us with the construction this summer, who want to likewise have a part in such a project. That said, we don't plan to do everything ourselves. Some tasks take difficult-to-acquire skills or special licenses, and we do not plan to try to become experts in everything.
So enough about "plans" - what have we actually done so far?
Well, not much. We only closed on the land in January, and the architect has been taking his time. We've meet with him twice in the office and once on our land so he could "get the smell of the place". We've been picking and acquired a shipping container. We're talking with the power company about getting power on-site, and are working to acquire an circuit box to set in the shipping container. We've been meeting with the zoning architect to merge our lots and lay out where to have the driveway, outbuildings, etc. We took measurements two days ago for extending our fence all the way to the river to better keep out sheep, and we'll be making some hardware store trips soon to pick up the 35 meters of wire mesh, barbed wire, and posts needed to complete the project. But all in all, what we can do without formal blueprints is rather limited.
We're working on landscaping. As always in Iceland, erosion is a problem. Iceland was largely forested when settlers arrived,but a thousand years of woodcutting and especially sheep grazing has led to only a fraction of a percent of the forest being retained. In the modern era there's been a lot of work trying to correct this, but in a country larger than Ireland with only 320,000 people, adverse weather and depleted soils, this isn't going to happen fast. One school of thought says to use only native plants, pointing to out-of-control lupine that was introduced to try to stop erosion and restore the soil (which it does, and it ultimately does get outcompeted by native plants after it restores the soil enough in an area, but that takes decades, and in the mean time it pushes plants like bilberry and dwarf birch). The other school of thought argues that any forest is better than no forest, that they're not competing with native trees because in the vast majority of the coutnry there are no trees at all, and that it'd be far easier (even profitable) to cut down unwanted trees and plant natives later once the rampant erosion and soil depletion is under control and you've got existing trees acting as a windbreak. It's quite true that many non-native species are hardier and grow faster than the very limited selection of natives and are thus better suited to quickly recover the amount of forestland. Basically, the argument is, "if it can grow and spread faster than native species into the unforested regions... good for it!"
(Above: Fields of lupine)
I have mixed feelings but in general support this latter school of thought. I can see the erosion on my land with my own eyes, what a thousand years of grazing has done. I also support the concept of food forests, providing habitat for wildlife in the same land that feeds one's self, and I want our kids to grow up thinking about where their food comes from. So we're not limiting ourselves to native species, and instead focusing on trees that are some combination of useful (especially for food), noteworthy, fast-growing, and hardy. The hardiest plants, like Russian pine (hardy to -60C, grows up to 50m tall, yields pine nuts) and sea buckthorn ("only" hardy to -30C, but famously resilient against harsh weather and with very nutricious berries), will go on the edges and most eroded slopes, acting as a windbreak and anchoring the soils. The most water tolerant varieties of plants (for example, swamp cypress, cranberry, etc) are to anchor the marshlands, and the less hardy plants (most fruit trees, for example) in the more sheltered areas in the middle. We're still going to leave a good bit of land as grassland, but it'll be allowed to grow free rather than be constantly eaten down to the ground. Most of our seeds are being sown loose (I don't exactly have time to plant several hundred thousand seeds individually!), but I will be planting some in trays indoors as a backup plan.
We've been learning about home construction, of course. Based on feedback from the architect, for example we know that we have to move the guest room further out to the west side so it can have a window, in accordance with building codes. We need to shorten the "tunnels" between rooms, because they're exterior walls, and exterior walls are expensive. This raises a problem of slopes - if a tunnel runs from the first floor up to the second, the shorter you make it, the steeper it gets. We've proposed a number of solutions that are found in real caves, including "human-added" steps, tunnels entering a room at different heights with a ladder to reach the upper entrances, rooms connected by only one tunnel where previously two were connecting, etc.
This information also led us to research how external wall costs are traditionally minimized in earth homes. The best approach appears to be the "umbrella":
Normally, underground (for example, basement) insulation and waterproofing are directly attached to a house's outer walls. Here, they are over the house, acting like an umbrella that keeps water out and heat in. The gravel/sand underneath remains dry, and dry ground is a far better insulator than wet. A huge pocket of trapped heat lines beneath, extending to the sides and deep below the house, of a scale that a total loss of heating only results in a temperature change of a few degrees per month. About half as much surface area insulative and waterproofing material is required, and the insulative material is half the thickness needed for a regular home. And the design is altogether much simpler: Instead of needing to follow the contours of the home, it's just a big layer over everything.
Thre's more we've learned, and far, far more that we're going to learn. We'll either succeed or go bankrupt trying. ;) But it's a labor of love.
And I hope to live there some day.
As an additional followup to yesterday's diary, the additional protests to the conservative government's breaking of their campaign promise to let people vote on whether or not to join the EU (probably the most important decision of this generation) keep going. Here's a video and some pictures I took at yesterday's protest (I arrived quite late because of work so it was winding down, but oh well!)