How the approximate CO₂ estimate works

Model 0.4.0 · researched 14 September 2026 · article revision 6.

Purpose: an educational lifetime estimate to explore choices, not an audited personal inventory. Tonnes mean physical CO₂. No methane, nitrous oxide or aviation non-CO₂ multiplier is included.

Reference factors

InputCentral coefficientOrigin / boundary
Petrol car0.16084 kg CO₂/vehicle-kmDESNZ 2026 average car, operating emissions
Diesel car0.17097 kg CO₂/vehicle-kmSame
Hybrid car0.12844 kg CO₂/vehicle-kmSame
Local bus0.10077 kg CO₂/passenger-kmDESNZ 2026 average local bus
Train0.03056 kg CO₂/passenger-kmDESNZ 2026 national rail, UK proxy
Short economy flight0.07344 kg CO₂/passenger-kmWithout RF, distance uplift already included
Long economy flight0.06826 kg CO₂/passenger-kmSame
Natural gas0.18194 kg CO₂/kWh fuelGross calorific value
Heating oil0.24557 kg CO₂/kWh fuelBurning oil, gross calorific value
Wood logs0.35 kg CO₂/kWh fuelBiogenic CO₂ reported outside scopes; included here without regrowth credit

Source S004: DESNZ 2026, corrected July workbook. Exact selected rows and their stable factor IDs are saved in selected-co2-factors.json; complete source workbook is desnz-2026-flat.xlsx. Its “CO2e of CO2” field contains only the CO₂ component (GWP 1), not the total-GHG field. Flight rows explicitly exclude RF. Do not add a second distance uplift. These UK references are transferred across countries/decades as a deliberate approximation. Direct fuel/car/flight factors exclude fuel production and vehicle manufacture; those are only approximately represented in the residual allowance.

Food (S005): Scarborough et al., Table 2. Central kg CO₂/person/day: plant 2.16, vegetarian 3.33, mixed (medium meat) 5.34, meat-heavy 7.28. Multiply by 365/1000. Adults standardised to 2,000 kcal; no child or country adjustment. Low/high values below come from the study but their lifetime application is a scenario.

Deliberate simplifications

A002 home: 9,000 kWh useful heat at 80 m²; gas efficiency 90% gives 10,000 kWh fuel. Electricity reference 2,500 kWh/year. These reference demands are anchored to England/Wales 2024 medians in NEED 2026, Table 2.1 (S008). The reference area is an authored choice, not the median home's measured area. Heat scales with area; electricity scales with 0.5 + 0.5 × area/80. Oil efficiency 85%, wood 80%, heat pump COP 3 are scenario assumptions. Fuel energy bases follow the selected factors. Divide home emissions by household members.

A001 country grid proxies are retained in assumptions.json. They are not sourced national measurements or historical series. HDI is only menu organisation. Country choice affects home energy and electric cars, not every category. Ember's methodology (S009) was checked and its all-GHG lifecycle intensities were not adopted as CO₂-only factors.

A004 electric cars use an assumed 0.18 kWh/vehicle-km times the grid proxy, divided by occupancy. No separate invented manufacturing coefficient is added.

A005 flights assume total return great-circle distance of 3,000 or 12,000 km, economy class. Distances are authored examples. Values are 0.22032 and 0.81912 t CO₂ per return trip. Substantially different routes/classes need different inputs.

A010 other goods/services 1.2 t and public services/infrastructure 0.5 t per person-year remain editable scenario allowances, not researched national statistics. OECD (S007) supports the allocation concept, not these quantities. Residual goods/services exclude the food supply chain already counted, but approximately cover unitemised purchases, vehicle manufacturing and energy supply chains. Allocation and overlap cannot be audited from a lump sum.

There is no automatic age or historical technology multiplier. Users enter personal chapter averages, including childhood. This is a substantive limitation, deliberately retained to avoid a larger application or unsupported age adjustments.

Lower and higher cases

CategoryLower / higher relative to central
Home0.64 / 1.64
Driving0.70 / 1.30
Flying0.70 / 1.50
Bus0.65 / 1.30
Train0.50 / 2.00
Other goods/services0.50 / 2.00
Public services/infrastructure0.50 / 2.00

Home bounds are inspired by electricity quartiles 1,600/2,500 and 4,100/2,500; using them for the entire home category is an authored stress test, not a household CO₂ confidence interval. Transport and allowance bounds are editorial scenario choices to expose transfer/recall uncertainty, not fitted distributions. Food low/high kg CO₂/day: plant 1.81/2.94, vegetarian 2.57/4.46, mixed 3.86/7.33, meat-heavy 4.90/12.23.

All categories move down/up together. The displayed total is rounded outward to tens of tonnes. It does not claim 95% coverage; real values may lie outside it. A zero activity or allowance stays zero in all cases, so the range does not fill omitted activities.

Funding and reproducibility

A007: full monthly payment first funds current CO₂ credits, then historical removal. Both default to €50/t, inclusive of all costs, with full quantities recorded at month end. These are user-supplied scenario terms, not evidence of measured delivery. The two ledgers remain separate.

A008: whole-year chapters to January 2026; monthly funding to age 100; selected annual decline compounds once a year. No previous contributions assumed.

Build: node tool/build.cjs. Verify: node checks/revision6.test.cjs. Inputs: research/assumptions.json. Model: tool/model.js. Raw workbook extraction: checks/extract-factors-v6.py (Python standard library). No additional dataset imports or full historical calibration were attempted after the scoped evidence check.