2026 Europe heat-pump efficiency by emitter and climate zone

The 2026 map: what EPREL says about heat-pump performance by type

The broad EPREL picture for 2026 is clear: heat-pump type changes average efficiency, but the distribution of listed products is dominated by one category. Across the hydronic categories with SCOP data, water-water units have the highest average SCOP at 6.15 with an average capacity of 35.65 kW and average outdoor noise of 42.0 dB (type_efficiency / EPREL Public API · type aggregation). That is ahead of ground-water at 4.77 SCOP, 18.45 kW, and 58.8 dB (type_efficiency / EPREL Public API · type aggregation), and air-water at 4.54 SCOP, 11.83 kW, and 59.8 dB (type_efficiency / EPREL Public API · type aggregation).

The type ranking matters, but so does scale. EPREL lists 30,452 air-water models, making air-to-water by far the most common category (type_efficiency / EPREL Public API · type aggregation). That compares with 21,065 air-air models, 9,228 heat-pump water heaters, 213 ground-water models, and just 31 water-water models (type_efficiency / EPREL Public API · type aggregation). For buyers and installers, that means the most efficient average category overall is also one of the smallest in market presence: water-water represents 31 models versus 30,452 for air-water (type_efficiency / EPREL Public API · type aggregation).

A compact ranking is below.

Readers wanting the product-level spread rather than averages should check the live EPREL catalog, the top SCOP leaderboard, and the type-specific air-to-water SCOP table. The methodological definitions behind SCOP and aggregation are set out in the site methodology and linked back to the underlying EPREL database.

Emitter matters: where underfloor heating beats radiators, and by how much

This is the point where the corpus becomes restrictive. The article brief asks for quantified SCOP differences by emitter pairing and for the gap between underfloor-heating and radiator-heavy systems. The supplied corpus does not contain any emitter-level EPREL aggregation or radiator versus underfloor SCOP splits. The registry extracts here are grouped by heat-pump type and by country conditions, not by emitter temperature class or installed emitter system.

So the numerical answers to these two questions cannot be given from this corpus:

• which emitter pairing has the highest average SCOP in 2026;
• how large the average SCOP gap is between underfloor and radiator-heavy setups.

What the available data does support is the editorial direction: operating conditions vary enough across countries that emitter temperature will materially affect the value of a given SCOP. In other words, the system-design question is more important than brand shopping alone, even if this corpus does not provide the emitter split directly.

That is also why the most useful next step for a project is usually not a brand page but a design workflow: the sizing calculator, the climate-fit tool, and the HVAC glossary for flow temperatures and seasonal metrics. For product browsing, the ground-source leaderboard and the quietest models table help narrow the shortlist after the emitter question is settled.

Climate matters: which European regions make low-temperature systems shine

Climate varies sharply across the 32-country comparison, and that alone changes the design case. On annual heating degree days at 18°C, Norway is highest at 5,039.96 HDD, Iceland is close at 5,028.63 HDD, and Liechtenstein reaches 5,023.68 HDD (country_compare / Eurostat · NASA POWER · EEA · Househeating Pulse subsidy register). At the warm end, Malta sits at 492.27 HDD, Cyprus at 819.26 HDD, and Portugal at 851.63 HDD (country_compare / Eurostat · NASA POWER · EEA · Househeating Pulse subsidy register).

The colder group therefore has roughly ten times the annual heating demand indicator of the warmest markets: 5,039.96 HDD in Norway versus 492.27 HDD in Malta (country_compare / Eurostat · NASA POWER · EEA · Househeating Pulse subsidy register). That is a large enough spread to make low-temperature emitter design far more consequential in northern markets than in Mediterranean ones.

The climate-zone labels in the country comparison reinforce that pattern. Countries marked colder include Sweden, Finland, Norway, Iceland, Estonia, Lithuania, Latvia, and Denmark in this dataset (country_compare / Eurostat · NASA POWER · EEA · Househeating Pulse subsidy register). Among those colder countries, Sweden combines 4,242.38 HDD with a very low grid intensity of 14 gCO2/kWh (country_compare / Eurostat · NASA POWER · EEA · Househeating Pulse subsidy register), while Norway combines 5,039.96 HDD with 18 gCO2/kWh, and Iceland combines 5,028.63 HDD with 11 gCO2/kWh (country_compare / Eurostat · NASA POWER · EEA · Househeating Pulse subsidy register). Those are some of the strongest conditions for electrically driven heating on carbon terms.

Subsidy support varies just as much. The largest maximum support in the comparison is Poland at €31,000, followed by Austria at €23,000 and Germany at €21,000 (country_compare / Eurostat · NASA POWER · EEA · Househeating Pulse subsidy register). Among colder countries, however, the registry records no active subsidy entries for Sweden, Finland, Norway, Iceland, Estonia, Lithuania, or Latvia in this snapshot (country_compare / Eurostat · NASA POWER · EEA · Househeating Pulse subsidy register). So the countries combining the coldest climates with the most favorable subsidies are limited in this corpus. The strongest cold-climate tariff-and-carbon case appears in Scandinavia, while the strongest subsidy case appears in central and eastern Europe, especially Poland and Germany (country_compare / Eurostat · NASA POWER · EEA · Househeating Pulse subsidy register).

For a wider comparative view, the 32-country dashboard, the country index, and the subsidy tracker are the most direct companion pages.

The running-cost test: electricity, gas, and the SCOP 4 break-even line

For a simple running-cost screen, a heat pump with SCOP 4 breaks even with a gas boiler at an electricity-to-gas price ratio of 4.0. The price-ratio dataset lets that line be tested directly because it reports household electricity and gas tariffs and the resulting ratio by country (price_ratio / Eurostat household band DC (electricity) / D2 (gas), latest semester).

Countries below the 4.0 break-even ratio for a SCOP 4 system include:

• Sweden 1.3 (price_ratio / Eurostat household band DC (electricity) / D2 (gas), latest semester)
• Netherlands 1.49 (price_ratio / Eurostat household band DC (electricity) / D2 (gas), latest semester)
• Portugal 1.73 (price_ratio / Eurostat household band DC (electricity) / D2 (gas), latest semester)
• France 1.78 (price_ratio / Eurostat household band DC (electricity) / D2 (gas), latest semester)
• Italy 2.0 (price_ratio / Eurostat household band DC (electricity) / D2 (gas), latest semester)
• Bulgaria 2.09 (price_ratio / Eurostat household band DC (electricity) / D2 (gas), latest semester)
• Liechtenstein 2.37 (price_ratio / Eurostat household band DC (electricity) / D2 (gas), latest semester)
• Slovenia 2.44 (price_ratio / Eurostat household band DC (electricity) / D2 (gas), latest semester)
• Greece 2.59 (price_ratio / Eurostat household band DC (electricity) / D2 (gas), latest semester)
• Denmark 2.63 (price_ratio / Eurostat household band DC (electricity) / D2 (gas), latest semester)
• Austria 2.68 (price_ratio / Eurostat household band DC (electricity) / D2 (gas), latest semester)
• Spain 2.79 (price_ratio / Eurostat household band DC (electricity) / D2 (gas), latest semester)
• Lithuania 2.86 (price_ratio / Eurostat household band DC (electricity) / D2 (gas), latest semester)
• Latvia 2.97 (price_ratio / Eurostat household band DC (electricity) / D2 (gas), latest semester)
• Luxembourg 2.99 (price_ratio / Eurostat household band DC (electricity) / D2 (gas), latest semester)
• Estonia 3.03 (price_ratio / Eurostat household band DC (electricity) / D2 (gas), latest semester)
• Croatia 3.05 (price_ratio / Eurostat household band DC (electricity) / D2 (gas), latest semester)
• Slovakia 3.05 (price_ratio / Eurostat household band DC (electricity) / D2 (gas), latest semester)
• Ireland 3.11 (price_ratio / Eurostat household band DC (electricity) / D2 (gas), latest semester)
• Germany 3.16 (price_ratio / Eurostat household band DC (electricity) / D2 (gas), latest semester)
• Hungary 3.23 (price_ratio / Eurostat household band DC (electricity) / D2 (gas), latest semester)
• Czechia 3.35 (price_ratio / Eurostat household band DC (electricity) / D2 (gas), latest semester)
• Poland 3.71 (price_ratio / Eurostat household band DC (electricity) / D2 (gas), latest semester)

Only Belgium at 3.9 is still below but very close to the line (price_ratio / Eurostat household band DC (electricity) / D2 (gas), latest semester). Countries above the line are the United Kingdom at 4.63 and Romania at 5.11 (price_ratio / Eurostat household band DC (electricity) / D2 (gas), latest semester). Romania is therefore farthest above the SCOP 4 break-even threshold in this dataset, by 1.11 ratio points (price_ratio / Eurostat household band DC (electricity) / D2 (gas), latest semester).

That result is where emitter choice becomes commercial rather than abstract. In countries well below the 4.0 line, a lower-flow-temperature system has more room to beat fossil alternatives on running cost. In countries near or above the line, preserving seasonal efficiency through emitter matching matters even more. The site payback calculator and subsidy calculator are built for exactly that screen.

Three country snapshots: Germany, France, and Sweden as different operating conditions

Germany is the classic design-sensitive market in this corpus. It has 3,308.21 HDD, January mean temperature of -0.73°C, electricity at €0.3869/kWh, gas at €0.1223/kWh, and a power-to-gas ratio of 3.16 (country_profile / Eurostat tariffs (band DC/D2 latest); NASA POWER 30y normal; EEA grid CO₂; subsidies captured manually from official programme pages; price_ratio / Eurostat household band DC (electricity) / D2 (gas), latest semester). Grid intensity is 366 gCO2/kWh and the main subsidy reaches €21,000, capped at 70% of eligible cost, with a €30,000 eligible cost cap (country_profile / Eurostat tariffs (band DC/D2 latest); NASA POWER 30y normal; EEA grid CO₂; subsidies captured manually from official programme pages). The data therefore supports a decent running-cost case, a strong subsidy case, and a meaningful carbon case, but not a forgiving one for poor high-temperature design.

France is milder and much cleaner electrically. It records 2,759.65 HDD, January mean temperature of 3.0°C, electricity at €0.2561/kWh, gas at €0.1436/kWh, and an electricity-to-gas ratio of 1.78 (country_profile / Eurostat tariffs (band DC/D2 latest); NASA POWER 30y normal; EEA grid CO₂; subsidies captured manually from official programme pages; price_ratio / Eurostat household band DC (electricity) / D2 (gas), latest semester). Grid intensity is just 56 gCO2/kWh (country_profile / Eurostat tariffs (band DC/D2 latest); NASA POWER 30y normal; EEA grid CO₂; subsidies captured manually from official programme pages). Its main subsidy reaches €11,000, with programme rules showing €5,000 support for air-water and €11,000 for geothermal in the top income band tier listed here, while air-air is excluded (country_profile / Eurostat tariffs (band DC/D2 latest); NASA POWER 30y normal; EEA grid CO₂; subsidies captured manually from official programme pages). France therefore gives one of the strongest all-round cases for hydronic heat pumps in this corpus.

Sweden is colder but unusually favorable on operating economics against gas. It has 4,242.38 HDD, January mean temperature of -3.08°C, electricity at €0.2711/kWh, gas at €0.2092/kWh, and the lowest electricity-to-gas ratio in the dataset at 1.3 (country_profile / Eurostat tariffs (band DC/D2 latest); NASA POWER 30y normal; EEA grid CO₂; subsidies captured manually from official programme pages; price_ratio / Eurostat household band DC (electricity) / D2 (gas), latest semester). Grid intensity is just 14 gCO2/kWh and the subsidy register shows no listed subsidy in this country profile snapshot (country_profile / Eurostat tariffs (band DC/D2 latest); NASA POWER 30y normal; EEA grid CO₂; subsidies captured manually from official programme pages). Even without subsidy support, Sweden’s tariff ratio and ultra-low grid carbon make it one of the strongest real-world environments for heat pumps in the corpus.

What buyers and installers should prioritize: system design before brand choice

The data supports a narrow claim, not a sweeping one. Brand-level rankings are not included in this corpus, so the registry here cannot quantify brand spread. But it does show that the largest efficiency differences buyers can actually act on are often structural.

First, type matters: average SCOP is 6.15 for water-water, 4.77 for ground-water, and 4.54 for air-water (type_efficiency / EPREL Public API · type aggregation). Second, climate matters: annual HDD ranges from 492.27 in Malta to 5,039.96 in Norway (country_compare / Eurostat · NASA POWER · EEA · Househeating Pulse subsidy register). Third, the tariff backdrop matters: the electricity-to-gas ratio runs from 1.3 in Sweden to 5.11 in Romania (price_ratio / Eurostat household band DC (electricity) / D2 (gas), latest semester). Fourth, subsidy policy matters: maximum listed support spans from none recorded in many markets to €31,000 in Poland, €23,000 in Austria, and €21,000 in Germany (country_compare / Eurostat · NASA POWER · EEA · Househeating Pulse subsidy register).

That combination means the strongest real-world conditions for heat pumps, once climate and energy prices are considered together, are found in colder countries with low electricity-to-gas ratios and clean grids, especially Sweden with 4,242.38 HDD, a 1.3 price ratio, and 14 gCO2/kWh grid intensity (country_compare / Eurostat · NASA POWER · EEA · Househeating Pulse subsidy register; price_ratio / Eurostat household band DC (electricity) / D2 (gas), latest semester). The weakest combined conditions in this corpus are markets with high electricity-to-gas ratios and less favorable grid intensity, especially Romania at 5.11 ratio and 240 gCO2/kWh, or the United Kingdom at 4.63 ratio and 207 gCO2/kWh (country_compare / Eurostat · NASA POWER · EEA · Househeating Pulse subsidy register; price_ratio / Eurostat household band DC (electricity) / D2 (gas), latest semester).

So the practical order of operations remains straightforward: check climate, tariff ratio, and subsidy; then design for the lowest workable flow temperature; then shortlist products in the full heat-pump catalog or browse the manufacturer directory. The data here does not support the emitter-specific SCOP deltas requested in the brief, but it does support the larger point: in 2026, the commercial outcome depends more on fitting the system to the building and the country context than on brand choice alone.

Sources

• EPREL Public API · type aggregation — snapshot 2026-07-12
• Eurostat · NASA POWER · EEA · Househeating Pulse subsidy register — snapshot 2026-07-12
• Eurostat household band DC (electricity) / D2 (gas), latest semester — snapshot 2026-07-12
• Eurostat tariffs (band DC/D2 latest); NASA POWER 30y normal; EEA grid CO₂; subsidies captured manually from official programme pages — country profiles for DE, FR, SE, snapshot 2026-07-12

Continue reading

• Heat pump sizing guide — Start with heat loss and emitter temperature before comparing model brochures.
• Heat pump payback guide — Use tariff ratios and subsidies to test whether a high-SCOP design pays back faster.
• Radiators vs underfloor heating guide — A practical framework for deciding when existing emitters are good enough.
• European heat-pump subsidies guide — Compare how grants and bonuses change the upfront case by country.