Embodied vs. sequestered carbon in a model conventional house
In this post, I try to take a look at the amount of embodied carbon emissions, as well as the captured carbon in lumber etc, for an entire house, as well as a very quick comparison of the operating carbon emission to the embodied carbon emissions.
These are somewhat notional calculations for a simplified house. Partly, this is because as soon as we try to be very accurate, we would have to start to get very specific to a particular site and house design, and set of building codes, and the increased accuracy about that particular house wouldn't generalize. And other-partly, it's because it's too much work for a blog post to try to track down the embodied energy of the door-handles. Finally, embodied carbon emissions in building materials are sufficiently variable and uncertain that there isn't any point in trying to be super-precise about anything else. So here, I just confine myself to the main structure of a simplified house, built in a somewhat generic North American style. Still, I think I've been comprehensive enough to capture the main components of the house.
This post will consider a reference house using fairly conventional construction techniques. Then in future posts I can use the same spreadsheet to consider variations. Commenters should feel free to chime in with corrections.
The assumptions for this particular model house are as follows:
- 45'x29' internal dimensions of the house, with two stories above ground.
- Full basement, with eight foot high, 12" thick poured concrete walls on a 2'x1' footing, and with a 4" thick basement slab
- 2x6 walls, 24" centers, with R-18 fiberglass insulation in the cavity, drywall on the inside, 3/4" OSB sheathing plus 1" of rigid foam insulation (R-5) and 1/2" clapboards on the outside.
- Partition walls of 2x4 plus drywall on either side that run fully across the middle of the house in both directions (essentially cutting each floor, and the basement, into four equal size rectangular rooms - obviously this is particularly simplistic floor plan, but stands as a proxy for all the infinite variations of real floor plans).
- Floors with 2"x10" joists, plus 3/4" OSB subfloors and 1/2" hardwood flooring.
- 2"x10" rafters, with R30 insulation between them.
- Asphalt shingles, on the 8:12 pitch roof gable end roof, with an 2' all round overhang.
Some things that are currently neglected:
- Furnishings and appliances
- Windows are assumed to be similar in embodied emissions to the wall that would be there without them
- Emissions associated with construction professionals traveling to the site (don't fly in an architect from the other side of the country repeatedly!)
My embodied carbon numbers come from ICE, as do the material densities. Captured carbon in lumber components are calculated based on the assumption that the wood has 15% moisture content and the dry portion is of chemical composition CH2O. It should be realized that embodied carbon emissions vary widely according to the specifics of where and when the calculation was done. So these numbers should be viewed as having uncertainties of a few tens of percent.
In particular, the embodied energy in lumber is dominated by transportation energy, so it's tremendously advantageous to source lumber close to the house to reduce its embodied energy.
Also, I break out the sequestered carbon in the lumber separately, since it's controversial whether or not you should take credit for it or not. Traditionally, I think a lot of environmentalists viewed cutting down trees to put the lumber in houses as basically raping the earth and bad per se. I guess I would tend to concur if the lumber in question was coming from the remaining old growth redwoods or virgin tropical forests (complete with hunter-gatherer inhabitants).
However, from the standpoint that climate change is the biggest overarching threat to all ecosystems everywhere, it seems to me that sequestering carbon in tolerably managed secondary forests and then putting it into buildings for an extended period (the longer the better) is basically a good thing. It beats burning the wood and putting the carbon immediately back into the atmosphere anyway.
Given the above, the overall embodied carbon emissions and sequestered carbon for "v1.0" of my model looks as follows:
As you can see, the sequestered carbon in the conventional house offsets a sizeable fraction of the embodied emissions. (In my view, especially in my region where trees are growing like weeds, the right column is basically "good" and can be viewed as offsetting the "bad" in the left column). Indeed, if one were to a) skip the basement, and b) locally source the lumber, you could probably have a conventional house that sequestered more carbon than had been emitted to manufacture the materials.
As several commenters pointed out in earlier posts, these numbers are small compared to what the house will likely use during its life. If we take this graph:
for average direct residential energy use, which comes from this post, we can see that the average household uses 100 million btu per year, which probably comes to carbon emissions of around 2.5 tonnes/year - again with an uncertainty of a few tens of percent depending on the mix of fuels etc. So amortized over the 30-300 year life of building components, the embodied or sequestered emissions will only be a few percent of the total used.
Still, they are probably particularly important right now, as the sooner they go into the atmosphere (or come out of the atmosphere), the longer they will have to influence the climate during the critical twenty-first century.
Finally, for anyone who wants to double-check the details, here they are: dimensions in feet, densities in kg/cubic feet, and emission/sequestration coefficients in kg carbon/kg material, and total emissions in tonnes of C. Click to get a version big enough to read.