What a household footprint includes, and what it leaves out
Every carbon footprint is a sum of activity multiplied by emission factor. The hard part is not the arithmetic — it is deciding where the boundary sits, and being honest about it.
This calculator draws the boundary around direct household energy and travel: the electricity you buy, the fuels you burn at home, the gasoline your cars burn, and your share of the fuel burned by aircraft you fly on. Those are the categories where you have meter readings, receipts and control. In corporate language they are Scope 1 (fuels you burn) and Scope 2 (electricity you purchase), plus one slice of Scope 3.
It excludes everything embodied in what you buy. Food — particularly beef and dairy — clothing, electronics, construction materials, shipping, and the share of public services attributable to you all carry real emissions that this tool does not count. Those consumption-based categories often add as much again as the direct ones, and they are far harder to estimate because they depend on supply chains you cannot observe.
So read the result as what it is: an accurate account of your direct energy and travel emissions, not a complete life-cycle footprint. That narrower boundary is a feature, not a defect. It is measurable, reproducible from bills, and it covers exactly the categories where a household decision changes the number.
Where each emission factor comes from
Fuels are chemistry. Burning natural gas releases about 5.30 kg of CO2-equivalent per therm, propane about 5.72 kg per gallon, #2 fuel oil about 10.21 kg per gallon, and gasoline 8.887 kg per gallon. These come from the carbon content of the fuel and the fact that each carbon atom leaves as a CO2 molecule 3.67 times heavier than itself. They are published in EPA's Emission Factors Hub and are stable from year to year, because chemistry does not change.
Electricity is politics and geography. There is no universal factor, because a kilowatt-hour from a hydroelectric dam and one from a lignite plant differ by more than a factor of twenty. EPA's eGRID database divides the United States into subregions and publishes an emission rate for each in pounds of CO2e per megawatt-hour. The US average total output rate is close to 800 lb/MWh, which is 0.363 kg per kWh, and that is the default here. Find your own subregion's number. It is the single input that most changes the result, and using the national average in the Pacific Northwest or in coal-heavy regions produces an answer that is wrong by a large factor. The electricity CO2 emissions calculator goes further into transmission losses and market-based accounting.
Flying is an allocation. The 0.2414 kg per passenger-mile used here corresponds to about 0.15 kg per passenger-kilometre, a mid-range economy figure across haul lengths. It does not distinguish a short domestic hop, which is worse per mile, from a long-haul cruise, which is better, and it assumes economy seating. For a specific itinerary, cabin class or a radiative forcing uplift, use the flight carbon footprint calculator and enter that result in place of the estimate.
Driving is fuel, not miles. The calculator converts miles to gallons through your fuel economy and then applies the gasoline factor, which is why entering a real-world MPG rather than the window sticker matters. Electric vehicles do not belong on the driving line at all — their charging shows up in your electricity meter, and adding them twice is the most common double count in a household inventory. The driving CO2 calculator handles diesel, E85 and the EV comparison in detail.
Worked example: a family of three in a gas-heated house
Three people, 10,000 kWh of electricity on a grid at 800 lb CO2e/MWh, 600 therms of natural gas, 12,000 vehicle miles at 25 MPG, and no flights this year.
- Convert the grid factor. 800 lb/MWh ÷ 1,000 = 0.8 lb per kWh, and 0.8 × 0.45359237 = 0.362874 kg per kWh.
- Electricity. 10,000 × 0.362874 = 3,628.74 kg.
- Natural gas. 600 × 5.30 = 3,180.00 kg.
- Driving. 12,000 ÷ 25 = 480 gallons, and 480 × 8.887 = 4,265.76 kg.
- Household total. 3,628.74 + 3,180.00 + 4,265.76 = 11,074.50 kg, or 11.07 tonnes CO2e.
- Per person. 11.0745 ÷ 3 = 3.69 tonnes each.
- Breakdown. Driving 38.5%, electricity 32.8%, gas 28.7%. No single category dominates, which is typical of a household with moderate use across the board.
Now add two transatlantic economy return flights for two people — roughly 22,000 passenger-miles. At 0.2414 kg per passenger-mile that is 5,311 kg, which lands the household at 16.39 tonnes and makes flying the largest category at 32.4 percent. Two trips reordered the entire inventory, which is the most important thing this calculator can show you.
Emission factors used, and what one tonne of CO₂e looks like
| Activity | Emission factor | Activity that produces 1 tonne CO₂e |
|---|---|---|
| Natural gas | 5.30 kg CO₂e / therm | 189 therms |
| Propane | 5.72 kg CO₂e / gallon | 175 gallons |
| #2 heating oil | 10.21 kg CO₂e / gallon | 98 gallons |
| Gasoline | 8.887 kg CO₂ / gallon | 113 gallons (2,813 mi at 25 MPG) |
| Electricity at 800 lb/MWh | 0.3629 kg CO₂e / kWh | 2,756 kWh |
| Electricity at 1,400 lb/MWh | 0.6350 kg CO₂e / kWh | 1,575 kWh |
| Electricity at 200 lb/MWh | 0.0907 kg CO₂e / kWh | 11,023 kWh |
| Flying (economy average) | 0.2414 kg CO₂e / passenger-mile | 4,143 passenger-miles |
Every right-hand figure is 1,000 divided by the factor. The three electricity rows show why looking up your eGRID subregion matters more than any other refinement you can make.
How to read your number
Read the breakdown before the total. The total tells you very little on its own; the ranking tells you where to look. A household whose result is dominated by electricity should investigate its grid factor and its largest loads. One dominated by heating fuel should look at the envelope — the insulation payback calculator quantifies that directly — and at whether a heat pump makes sense. One dominated by flying already knows what the lever is.
Per person is the fairer comparison. Home energy is largely fixed by the building, so a four-person household almost always looks better per capita than a one-person household in the same house. When comparing yourself to anyone else, compare per-person figures and confirm that both used the same boundary.
For scale, not as a target: the EIA reports total US energy-related CO2 emissions in the region of 4.8 billion metric tons a year across the entire economy, which divided among roughly 335 million people is about 14 tonnes each. Most of that is industry, commerce, freight and agriculture rather than household activity, so your direct household figure per person will be well below it and should be. Do not treat the difference as a gap you have closed.
Recompute it, do not re-estimate it. The value of this calculation is entirely in the quality of the inputs. Twelve months of actual bills beats a guess by a wide margin, and a year-over-year comparison using the same method is far more informative than one absolute number, because the method's systematic errors cancel when you take a difference.
Assumptions and common errors
- Double-counting an electric vehicle. EV charging is already in your kWh. Enter zero vehicle miles for it, or you will count the same energy twice.
- Using the national grid factor everywhere. Subregional rates vary by more than a factor of ten. This is the largest single source of error in most household footprints.
- Counting flight miles per trip instead of per passenger. Four people on the same flight generate four times the passenger-miles. Count each person on each leg.
- Forgetting the return leg. A round trip is two legs. Half of all flight footprints reported informally are exactly half what they should be.
- Treating this as a total footprint. Food, goods, services and construction are excluded. Consumption-based studies typically find those categories are comparable in size to the direct ones.
- Comparing figures from different calculators. Different boundaries, different grid factors, and different treatment of aviation's non-CO2 effects make cross-tool comparisons meaningless. Compare the same tool to itself over time.
Key terms
- CO₂e
- Carbon dioxide equivalent. Non-CO₂ gases such as methane and nitrous oxide are converted to the mass of CO₂ with the same warming effect over a chosen horizon, usually 100 years.
- Emission factor
- Kilograms of CO₂e released per unit of activity. Chemistry sets the fuel factors; generation mix sets the electricity factor.
- Location-based accounting
- Valuing your electricity at the average carbon intensity of the grid you are connected to. The default approach in this calculator.
- Market-based accounting
- Valuing your electricity at the intensity of the contracts and certificates you actually bought, which can be lower or higher than the grid average.
- eGRID subregion
- EPA's division of the US power system into areas with distinct generation mixes, each with its own published emission rate.
What to do with the result
Rank your categories, then attack the largest one with the cheapest available measure. That ordering matters because the categories differ enormously in how much a dollar buys. Sealing and insulating an attic can remove a tonne a year for a few thousand dollars and keep doing it for thirty years. Replacing a working car with a new one removes nothing until the manufacturing debt of the new vehicle is repaid. Skipping one long-haul return flight removes one to two tonnes immediately at zero cost.
Two structural moves change the shape of the whole inventory rather than trimming a line. Electrifying a heating or hot-water load moves emissions from the heating-fuel line into the electricity line, where the factor falls every year as the grid decarbonises — a gas furnace's 5.30 kg per therm is fixed forever. And changing your electricity supply, through a genuine green tariff with retired certificates or through on-site generation, scales the entire electricity line at once rather than reducing individual loads.
Finally, recalculate annually with the same method. A footprint is only useful as a series. The absolute value carries real uncertainty from boundary choices and average factors; the year-over-year change carries far less, because the same assumptions sit on both sides of the subtraction.