What does a house carbon footprint actually mean?
The question about a building carbon footprint usually arrives once the house is already designed as energy efficient. That is understandable and also late. A house carbon footprint has two parts, and each of them is decided at a different moment. Operational carbon accumulates gradually, over years of occupation, and can still be improved later by changing the heat source or adding insulation. Embodied carbon is released during raw material extraction, manufacturing, transport and construction, before anyone spends a first night in the house. Once the building is handed over, nothing can be done about it. It has been paid.
The sum of both parts is the building whole-life carbon, and it is inside that sum that the centre of gravity has moved in recent years.
Why do embodied and operational carbon trade places?
As the envelope improves, heating demand falls. In a well designed new Slovak house it is already low, and at passive house level almost symbolic. Operational carbon falls with it. Embodied carbon does not, and often rises slightly instead: thicker insulation, triple glazing, mechanical ventilation and solar technology are products with footprints of their own.
The result is that in an energy efficient new house embodied carbon makes up a large share of the whole-life total, and all of it is spent up front, before the first day of occupation. Published assessments disagree on the exact share, roughly from a third to well over a half, and that spread is not a measurement error: it depends on the assumed service life, on the assumed electricity mix and on where the system boundary is drawn. The direction is consistent, and it will deepen as Slovak electricity decarbonises. Whether the passive house premium pays back in euros is a different question, covered in a separate article on payback.
How is a building life cycle divided into modules?
So that two solutions can be compared at all, an assessment is split into modules. This is the dullest part of the subject and the part on which most comparisons found online fall apart.
| Module | What it contains | What gets forgotten |
|---|---|---|
| A1-A3 (production) | Raw material extraction, manufacturing, transport within the supply chain | Most product declarations report only these modules and stop at the factory gate |
| A4-A5 (construction) | Transport to site, assembly and running the site | Offcuts, packaging and waste that never made it into the building |
| B (use) | Operational energy, maintenance, repair and replacement | Facade, waterproofing, windows and the heat source get replaced over a building life |
| C (end of life) | Deconstruction, transport and waste processing | This is where biogenic carbon in timber is released, if it is burnt or landfilled |
| D (beyond the boundary) | Benefits from recycling and energy recovery | Reported separately, and must not be subtracted from the result |
The practical consequence: a cradle-to-gate figure (modules A1 to A3) is not a whole-life figure, only the beginning of one. A material with an excellent A1-A3 number and a ten year service life can end up worse than one with a higher starting number that lasts the whole life of the house. Module D is reported separately so that future recycling cannot disguise a present-day emission.
Where does embodied carbon in a house actually sit?
This is where clients are most surprised. The decisions discussed for months, meaning tiles, floors, the kitchen and finishes, matter less to the carbon footprint than the ones taken in a single evening over the site plan: how big the house is, whether it has a basement and how it is founded.
| Element group | Typical weight in embodied carbon | Why |
|---|---|---|
| Structure, floor slabs, stairs | Dominant | The largest volume of material and usually most of the concrete and steel |
| Foundations and substructure | Dominant, decisive on poor ground | A basement, deeper footings or piles add a lot of concrete without adding living space |
| Insulation and envelope | Significant | Large areas and growing thicknesses, with higher production emissions for oil-based and mineral insulants than for bio-based ones |
| Windows and glazed walls | Significant | Glass and frames are energy intensive, and large glazed elevations add up |
| Services (heat source, ventilation, photovoltaics) | Moderate, repeated in module B | Shorter service life than the structure, so it is counted more than once |
| Finishes, claddings, fit-out | Smaller than clients expect | Little material by volume, but short life and frequent replacement |
Cement is a special case: a substantial part of its emission comes not from fuel but from the chemical conversion of limestone during firing, so it cannot be removed by a better kiln, only by a different mix design and by using less concrete. Steel is demanding when made from ore. A large garage under the house or extensive foundations on poor ground therefore outweigh dozens of decisions about finishes. The first question is not which material, but how much material there is at all.
How do you count timber honestly rather than as marketing?
Bio-based materials, meaning timber, wood fibre insulation, cellulose, hemp or straw, usually carry lower production emissions than their mineral alternatives, and they also hold carbon the tree took out of the atmosphere. That storage must not be counted as an automatic bonus.
Two conditions have to hold. The material must not end its life in an incinerator or a landfill, where the carbon is released again, and that release shows up in module C. And the harvested timber has to grow back, meaning the feedstock needs documented regeneration. This is why biogenic carbon is reported separately, and why design for disassembly matters most of all for timber: a bolted joint keeps a second life possible, a fully glued and foamed detail destroys it. A CLT panel acting as both structure and finished surface also saves a layer of materials that would otherwise arrive as plaster and cladding. I compared masonry, timber frame and CLT in a separate article on choosing a structural system. The division is clean: operational demand is decided by the envelope build-up and workmanship, the carbon footprint by the material.
Why does renovation almost always beat demolition?
This is the least intuitive and, for a Slovak reader, the most useful conclusion here. An old house has its embodied carbon already paid. Demolition does not remove it, it writes it off, and adds a whole new building including new foundations on top. Deep renovation or adaptive reuse therefore usually wins on carbon even when the renovated house performs worse than a new build on the same plot would have.
| Criterion | Deep renovation | Demolition and new build |
|---|---|---|
| Embodied carbon up front | Only the new layers: insulation, windows, services, local strengthening | The entire house again, structure and foundations included |
| Structure and foundations | Retained, and with them the single largest item | Manufactured again from scratch |
| Achievable operational demand | Very low, but constrained by geometry, orientation and details | Unconstrained, passive house level is reachable |
| Construction waste | Smaller, mostly from replacing layers | The whole mass of the demolished house to be processed |
| When the emissions occur | Spread out, and can be phased | All at once and up front, before the first day of use |
| Main risk | Hidden defects the survey did not find | That the new building cannot repeat the old position and size |
Exceptions exist: a house with a failed structure, a permanently damp basement or an unusable layout may be better removed. That decision belongs after the survey, not before it. The permitting and ownership side of the comparison is in the article on whether to renovate or demolish.
Where do you get numbers that can actually be compared?
The source is an Environmental Product Declaration, or EPD. It is third-party verified, so it is not a manufacturer leaflet. Its headline indicator is global warming potential in kg CO2e, conventionally over a hundred year horizon.
Two EPDs are comparable only when three things match: the same declared unit (per kg, per m2 of build-up, per m3), the same modules and the same product category rules. A per-kilogram value cannot be compared with a per-m3 value, and an A1-A3 figure cannot be compared with an A1-D figure. Anyone who takes two numbers without that check has not compared materials, they have compared boundaries. In practice you avoid this by comparing a functionally equivalent wall build-up per m2 at the same U-value. Only then can you verify whether the house is heading towards a low-carbon building or merely towards a better written brochure. Whole-life carbon reporting is also being introduced by the revised European directive on the energy performance of buildings, starting with large new buildings.
Why is the refrigerant in a heat pump a hidden item?
A heat pump contains a few kilograms of refrigerant, and that refrigerant is not CO2. Older HFC refrigerants have GWP values in the thousands, so one kilogram equals tonnes of CO2e. Newer HFO refrigerants have GWP values in the single or double digits. If the circuit leaks or the unit is disposed of improperly, this item turns into a real emission, and in a house with near zero operational demand it is no longer negligible. The question about refrigerant type and its GWP belongs in a quotation comparison as much as the question about efficiency.
What could overturn this conclusion?
It is only honest to say where the argument is sensitive. First, the grid: if Slovak electricity stayed carbon intensive, the operational side would keep a larger share and materials would matter less. Second, service life: an assessment over sixty years and one over a hundred give a different ranking, because a longer life spreads embodied carbon and amplifies the replacements in module B. Third, the system boundary: whether furniture, paving, retaining walls and utility connections are counted. A figure has value only with its assumption stated. Without it, it is a number without a unit.
What can a homeowner actually decide, strongest lever first?
- Do not build new if you can renovate. The largest saving is not manufacturing the structure and foundations twice.
- Build less. Every m2 carries its material. A well planned hundred metres beats a badly planned hundred and fifty, in carbon and in money.
- Do not add a basement without a reason, and found the house modestly. A basement on poorly draining ground is the most carbon intensive storage space you can buy.
- Reduce concrete and steel where they are not essential. Shorter spans, fewer cantilevers, less in-situ concrete.
- Choose bio-based materials in the envelope. Insulation and non-loadbearing elements are where substitution is easiest and structurally free.
- Ask for EPDs on the three largest items by volume. Not on everything, nobody will pay for that. The three deciding lines are enough.
- Deal with refrigerant, joints and replacements. A low-GWP refrigerant, demountable joints and components swappable without demolition all reduce modules B and C.
A house carbon footprint will not fall because of the tile you pick. It falls with how much house you build, what you build it from, and whether you build at all instead of keeping what already stands on the plot.