Over twenty years of operation, the use phase decides a heat pump versus gas boiler comparison, not the carbon content of the kit: manufacturing a heat pump weighs 1 to 2 tonnes of CO₂ against tens of tonnes released on the gas side, since every kWh net CV of natural gas burned emits 0.203 kg CO₂e in combustion alone according to DESNZ 2024. Two life cycle assessments only compare if they hold the same unit, delivered or primary energy, the same service, heating alone or heating and hot water, and the same end-of-life boundary. The two values to survey before costing are the SCOP actually reachable at the flow temperature of the existing network and the refrigerant leak rate, the only ones that still move the result once the boundary is fixed.
Understanding LCA: what you need to measure before comparing a heat pump and a boiler
LCA, emissions, primary energy: concepts to clarify on site
An LCA (life-cycle assessment) adds up a piece of equipment's impacts over its entire life cycle, most often in kg CO₂e. On site, check that you're comparing the same unit (final or primary kWh) and the same service delivered (heating only, or heating plus hot water). Note the annual heating need, the emitters, the setpoint, and the climate zone. Without these reference points, two studies can reach opposite conclusions for the same building.
Study boundaries: manufacturing, transport, use, end of life
Two LCAs can reach opposite conclusions if the scope changes. Make sure the analysis clearly includes:
- manufacturing of the generator, and for a heat pump, the refrigerant
- transport, installation, accessories (hot water tank, controls)
- consumption in use, backup heating, any leaks
- maintenance, replacements, recycling at end of life
The data that changes the result: the 2026 electricity mix, real-world efficiency, maintenance
In 2026, the carbon content of electricity depends on the generation mix and the time of consumption. The reference factors are the ones DESNZ publishes each year for greenhouse gas reporting, the British counterpart of ADEME's Base Empreinte®. For electricity, two values coexist: generation alone (≈ 207 gCO₂e/kWh in the 2024 set) and generation plus transmission and distribution losses (≈ 229 gCO₂e/kWh), and a heating-season figure runs higher still, because winter leans harder on gas plant. On site, use the actual SCOP (defrost cycles, flow temperatures), the boiler's real-world efficiency, and maintenance. To go further, see real-world heat pump performance.
LCA of a heat pump: where does the environmental impact actually come from?
Manufacturing and metals: the weight of hardware in the LCA
In an LCA, manufacturing weighs in mainly through the extraction and processing of materials. A heat pump packs a compressor, heat exchangers, and electronics. The heaviest contributors are copper, aluminium, and steel, plus logistics and assembly. Available FDES and PEP environmental declarations for residential air/water heat pumps converge around 1 to 2 t CO₂e for manufacturing an 8 to 12 kW unit. This is also where choosing robust, repairable, correctly sized equipment avoids paying for the impact twice.
Use phase: electricity consumption, SCOP, and emission factor
In use, the impact depends on the kWh consumed and the electricity mix. Seasonal SCOP varies with outdoor temperature, heating flow temperature, defrost cycles, and settings. For a typical heating need of 13,500 kWh/year, a heat pump with a SCOP of 2.5 (a conservative retrofit value) consumes around 5,400 kWh/year of electricity, i.e. roughly 22 t CO₂e over 20 years at today's grid factor, and closer to 16 t if the emitters allow a SCOP of 3.5. Both figures fall further as the grid decarbonises: the factor has more than halved since 2013. A well-insulated house, low-temperature emitters, and careful control reduce consumption, and therefore impact, without changing the machine.
Refrigerant: R32, R290 (propane), and best practices
The refrigerant is a sensitive point because some have a high global-warming potential. R32 heat pumps (GWP 675) remain the market standard; the newer R290 (propane, GWP 3) generation reduces the refrigerant's contribution to the LCA to almost zero. Leaks are limited by clean commissioning, careful fittings, recovery at end of life, and compliance with the leak-tightness checks required under the F-Gas regulation. Installation quality directly determines the refrigerant's share of the overall footprint.
LCA of a gas boiler: which factors weigh most on the environment?
Combustion and direct emissions: CO₂, NOx, and combustion settings
In an LCA, the number-one contributor remains the use phase. Every kWh net CV of natural gas burned emits 0.203 kg CO₂e (combustion only, DESNZ 2024). And if the boiler is poorly tuned or dirty, consumption rises. NOx mainly affects air quality. Fine tuning, proper flue evacuation, and a real-world efficiency close to nominal limit excess emissions.
Upstream gas: extraction, transport, and methane leaks in the LCA
Before it reaches the burner, gas already has an energy history. Extraction, processing, compression, sometimes LNG, and transport all weigh in the LCA. The sensitive point is methane leaks. According to the IPCC AR6, fossil methane has a GWP100 of 29.8 (and a GWP20 of 82.5, more relevant for a short decarbonisation horizon). Adding the well-to-tank factor to combustion brings natural gas to ≈ 0.241 kg CO₂e/kWh net CV, i.e. +19% compared with combustion alone.
Lifespan, maintenance, and replacement: the hidden impact over 15 to 20 years
Over 15 to 20 years, the impact isn't limited to the new appliance. Maintenance visits, replaced parts (circulator pump, sensor, expansion vessel), and end of life all add up. Annual servicing limits breakdowns and avoids the excess consumption that drags down the overall impact.
Comparing heat pump and boiler with a useful LCA: a simple method for choosing by dwelling
Assumptions used for the "Key figures" block
The orders of magnitude shown on this page rest on the following assumptions, to be adapted to the real case: heating need of 13,500 kWh/year, 20-year lifespan, DESNZ 2024 conversion factors (electricity, generation plus losses: 229 gCO₂e/kWh; natural gas, net CV including well-to-tank: 241 gCO₂e/kWh), retrofit SCOP of 2.5 (air/water heat pump with high-temperature radiators) to 3.5 (heat pump on underfloor heating in an insulated building), seasonal boiler efficiency from 75 to 95%. On the gas side that gives around 3.8 t CO₂e a year, close to 77 t over the same twenty years. Methodological source, for the French equivalent: Base Empreinte® ADEME.
Partial vs whole-house renovation: when insulation changes everything in the LCA
In a partial renovation, the LCA is driven by the use phase. If the dwelling remains an energy sieve, the heat pump runs longer, is sized higher, and its real SCOP deteriorates. After coherent insulation work, needs drop, the heat pump can be smaller, and the environmental gain becomes clearer. The decision therefore follows a logical order: insulate first, then size for the target building envelope.
High-temperature radiators vs underfloor heating: consequences for consumption
With high-temperature radiators, the heat pump has to work harder, and its SCOP drops (the Electrification of Heat demonstration measured a mean seasonal performance close to 2.8 across its retrofits, and lower where flow temperatures stayed high). The result is more kWh consumed for the same comfort. Underfloor heating, or radiators oversized for low-temperature operation, improve efficiency. On the boiler side, condensation works better with a colder return temperature. Lower temperatures win points on both sides.
Climate, domestic hot water use, controls: the scenarios that flip the verdict
In a cold climate, a heat pump can lose performance, especially at peak demand. A large domestic hot water (DHW) need, a poorly set heating curve, or faulty ventilation can also cancel out the expected advantage. Conversely, good controls, optimised DHW, and a correctly sized generator make the LCA more favourable. Document the real-world scenarios before deciding.
Talking points for tradespeople in 2026: how to explain the LCA to your client
Speak in concrete impacts: kWh, tonnes of CO₂ over time, and orders of magnitude
Explain the LCA as a balance sheet over the whole cycle: manufacturing, transport, installation, use, end of life. Then bring everything back to two reference points: the kWh saved and the tonnes of CO₂ over 15 to 25 years. With sourced orders of magnitude (the DESNZ conversion factors, the products' own EPDs), the client quickly understands what really matters. Those benchmarks come out of the pricing itself when Argile works out the energy savings of the chosen heat pump and shows them to the client.
Supporting documents and traceability: paperwork, product sheets, maintenance, compliance
For a credible LCA, keep simple evidence. Technical data sheets, references, batch numbers, and environmental declarations such as FDES or PEP when they exist. Add the maintenance manual and commissioning report. A well-kept file avoids disputes and protects the job.
2026 funding and obligations: linking the LCA to the Boiler Upgrade Scheme, ECO4 and certification
In 2026, funding looks at performance and compliance: MCS certification for the Boiler Upgrade Scheme, TrustMark and PAS 2030:2023 for ECO4, detailed invoices for both. The LCA isn't always required, but it helps decide between two eligible solutions. It's your guiding thread for making the right choice.
Sources and references
- Base Empreinte® ADEME, the French database; the British factors used here are the DESNZ greenhouse gas reporting set
- F-Gas regulation, leak-tightness and traceability obligations for refrigerants
- IPCC AR6 (2021), GWP100 / GWP20 factors for fossil methane
- FDES and PEP declarations from heat pump manufacturers (from the INIES database)
