At first glance, the green credentials of ground source heat pumps (GSHPs) look unquestionable: because you’re harvesting free heat from the ground, you can get up to four times more energy out of the system than you put into it. Sure, it runs on electricity, which is more carbon intensive than gas, but because of this favourable ratio of output-to-input (called the COP for coefficient of performance) the system should still emit less carbon than a gas boiler – in theory.
But the claimed benefits are reliant on incorrect assumptions. A new house will emit about the same carbon using a ground source heat pump as with a new gas boiler. Here’s why:
The key factor is the carbon intensity of electricity from the grid. This is commonly stated as 0.422 or 0.43 kgCO2/kWh. But these figures were proposed several years ago, when everyone thought the steady decline in grid intensity would continue. And it didn’t. Instead it levelled off at around 0.53 kgCO2/kWh as shown in the graph below. (Data from BRE available here – pdf). For a realistic carbon calculation, we need to use a realistic grid intensity figure.
In addition, to maintain a high COP, GSHPs need to keep their output temperature low: around 40°C. This is why they work so well with underfloor heating. But domestic hot water should be stored nearer to 60°C, so a backup system is required to heat up the water the rest of the way. This is usually an electric immersion coil, similar in principle to an electric kettle, which requires more grid electricity – but this time without the COP advantage of a ground source heat pump.
For the purposes of comparison here’s a baseline house, using a condensing gas boiler and grid electricity.
| Primary energy | Assumed efficiency | Useful energy | CO2 kg/kWh | kgCO2/yr | |
| Space heating | 3580 | 88% | 3150 | 0.19 | 680.1 |
| Water heating | 3181.8 | 88% | 2800 | 0.19 | 604.5 |
| Electricity | 2800 | 100% | 2800 | 0.53 | 1484 |
| Total kgCO2 | 2768.7 | ||||
Now here’s our house with a ground source heat pump and immersion coil backup for domestic hot water. I’m assuming the GSHP meets 60% of the DHW load.
| GSHP | Backup systems | ||||||||
| Useful energy demand (kWh) | Demand met (kWh) | Primary energy | CO2 kg/kWh | kgCO2/yr | Demand met (kWh) | Primary energy | CO2 kg/kWh | kgCO2/yr | |
| Space heating | 3150 | 3150 | 787.5 | 0.53 | 417.4 | 0 | 0 | 0.53 | 0 |
| Water heating | 2800 | 1680 | 420 | 0.53 | 222.6 | 1120 | 1120 | 0.53 | 593.6 |
| Electricity | 2800 | 0 | 0 | 0.53 | 0 | 2800 | 2800 | 0.53 | 1484 |
| Subtotal CHP CO2 | 640 | Subtotal backup CO2 | 2077.6 | ||||||
| Total kgCO2 | 2717.6 | ||||||||
| Saving | 1.8% | ||||||||
So our GSHP house saves less than 2% of carbon relative to the baseline. That hurts, especially when you consider that installed costs for a GSHP are around £1000/kW, six to ten times as much as a gas boiler. In addition, the above figures assume a COP of 4, but in practice the performance isn’t always as good as this. For example, Barratts recently found that the GSHPs at their scheme in Chorley Lancashire are averaging a COP of about 2.6.
So here’s an illustration of how the assumed carbon intensity of the grid affects the carbon savings of GSHP versus the baseline gas boiler system:
From the above graph it looks like the scheme at Chorley would have emitted much less carbon if they’d stuck with condensing gas boilers.
Having said that, the environmental performance of GSHPs does look better if compared against a system based on a higher carbon fuel, such as heating oil. Here’s the same table, this time compared against a baseline using heating oil at a carbon intensity of 0.265 kgCO2/kWh:
So the GSHPs look better here. At today’s grid intensity, a GSHP with a COP of 2.6 (like the ones at Chorley) save 6.5% carbon vs. a new system based on heating oil. At a COP of 4, they save 17% (though a gas system would save 15.5% anyway since gas is so much less carbon intensive than heating oil).
Where does this leave GSHPs? The use of domestic GSHPs where gas is available is very questionable. If gas is unavailable and you’re replacing a high carbon fuel such as heating oil (and the GSHP performance is able to match manufacturers’ claims) then they do make environmental sense. But you’d need to look carefully at whether you’d get more carbon benefit by spending your money elsewhere (e.g. biomass).
And of course you could always power a GSHP from a turbine or PVs but again, you may want to look carefully at alternative heating systems that would give you more carbon savings for your money.
carbon limited 
Interesting. I took issue with your breakdown of the ratio of space heating to DHW, but if I use an 80/20 split ( the extreme other end), the figures come out even worse, with the same house now emitting 2,961 kg
Sorry- ignore that last entry. If you use an 80/20 split, the emissions figure is 2627 kg/co2/yr, which is still not much better.
GSHP never do well when compared for purely heating, but the story changes when cooling is added. Now, I am most definitely not advocating cooling in the UK (yet) for the domestic market, but for those clients who demand cooling (think high end of the market, especially in city apartments where opening windows is less than desirable in the summer due to grime), then GSHP becomes more efficient than standard splits (or did last time I did the calcs).
Great post as always – ever think of pulling these altogether as pdf’s?
How does the embodied energy and embodied CO2 of conventional gas fired appliances compare with GSHP. All that Augering, excavation and pipework must attract a premium unless energy piles are used. Has anyone done a whole life payback exercise for this ?
Carbon intensity of the grid leveled off because the eco-lobby continues to villify nuclear energy. It is vitally important to build out lots of nuclear energy sources in order to fill the gap between fossil fuels and sustainable/renewables.
[…] Other Resources Grants may be available for UK installations see http://www.lowcarbonbuildings.org.uk/micro/ Proceed with caution where gas is available according to carbonlimited.org […]
Interesting post!
but unless I have misunderstood your figures Jason is right about your split of hot water energy vs space heating energy. The government suggests that this is generally around 20/80 as Jason suggests, though as buildings become better insulated this figure pushes towards 25/75
The answer is to heat domestic hot water using solar thermal and back up using an immersion heater on a timer to come on in the evening, a strategy advocated by Kensa heat pumps for a long time (no I don’t work for them, I install solar systems!). Heat pumps should not be used to heat DHW
There appear to be many heat pump manufacturers suggesting performance will be much better than is realistic, and I worry about a general lack of understanding of the technology by architects and specifiers. I think ultimately the people selling heat pumps are to blame as you can find many of them suggesting heating DHW is perfectly acceptable.
I have heard that one manufacturer claim that it sells a 14kW heat pump which is in fact an 8kW output heat pump with a small cylinder inside the casing and 2x 3kW immersions, so in fact not a true 14kW heat pump at all.
Another intersting (!) point is that SAPS calcs don’t take into account the energy taken to deliver fuel to a biomass boiler, something that often appears to be forgotten when it comes to these issues
assuming 15-20 year lifespan of a biomass boiler thats a lot of lorry journeys carrying 1-25tons of fuel depending on the scale of the installation
lifespan of a good heat pump should be similar or higher than this
Increasing the use of electricity will have no effect on the level of emissions within the EU Emissions Trading system – the overall level of emissions is capped across the industries covered for the whole of Europe. Increasing the use of electricity by installing a GSHP will marginally increase the cost of allowances, this may increase emissions in the UK, but over the whole system the market for allowances, and their marginal increase in price will incentivise compensating abatement elsewhere in the scheme. Switching from fossil fuels to an electric heating will therefore reduce the global level of greenhouse gas emissions by the reduction in domestic gas/oil use. The savings are the total emissions associated with the alternative gas or oil boiler! Consideration of the bigger picture is essential. DECC have adopted this approach in all their analysis.
[…] January 13, 2010 by cbuster “At first glance, the green credentials of ground source heat pumps (GSHPs) look unquestionable: because you’re harvesting free heat from the ground, you can get up to four times more energy out of the system than you put into it. Sure, it runs on electricity, which is more carbon intensive than gas, but because of this favourable ratio of output-to-input (called the COP for coefficient of performance) the system should still emit less carbon than a gas boiler – in theory.” Casey Cole […]
B Whittle,
It’s an old wifes tale to include biomass transport like this as all fuels have indirect emissions not just biomass and the Grid electricity figure used excluded these emissions. Here’s the latest figures;
Wood Pellets 0.03895 kg/CO2/kWh indirect emissions (DEFRA 2010, Annex 9, Table 9C)
Natural gas 0.1968 kg/CO2/kWh indirect emissions (DEFRA 2010, Annex 1, Table 1b)
Electricity indirect 0.07185 kg/CO2/kWh indirect emissions (DEFRA 2010, Annex 3, Table 3C)
For example Natural Gas has 5 times more indirect emissions than Wood pellets while electricity has double.
Policywonk,
proper wonky stuff. the idea of advocating an inaffective solution due to some capitalist money exchange is very frustrating.
Shane
ohhh, just noticed the figure for Natural Gas is 0.01799 kg/CO2/kWh indirect emissions (DEFRA 2010, Annex 1, Table 1c rather than 1b as above)