Currently, air source seems to be attracting more attention than ground source. This could be explained by the fact that the industry is developing, or it could be that the goal posts have been moving.

In the mid noughties, I, like many others, concluded that Ground Source was the best method since it promised the best year-round efficiency, and COPs (or SPF as we should say) of 4, for space heating had been demonstrated, given a well engineered system installed in an appropriate building. However, the installation cost and disruption is considerable, and it is no wonder that interest in air source has grown, since these can be installed very simply, and the entire ground-works is eliminated.

In the depth of winter, there is no doubt that the energy efficiency of an air source system will be lower than that of a good ground source system, and it is quite likely that more back-up heat will be required if an air source unit is fitted. This back-up heat often comes from a direct electric heater. Whilst the total (annual) amount of energy provided by back-up heaters is relatively small, it remains a concern that this load adds to the total capacity on the mains grid at a time when power stations are already working to capacity.

It has become clear that not all ground source installations have achieved their expected performance. This may in part be due to the fact that it’s expensive to install the ‘ideal’ ground loops, and some cost-cutting compromise is often chosen to make GSHP installations more viable, thus SPFs in the mid/lower 3’s seem to have been accepted rather than the hoped-for SPFs of 4. Putting an actual figure for a desirable SPF value for the UK (given its generation mix) seems an almost impossible task due to the multitude of variable factors.

The Efficiency of any heat pump unit (model to model) can vary, and is dependent on many things, namely - Compressor type/efficiency, heat exchanger sizes etc. Air source units require a good refrigerant control mechanism since they operate over a varying temperatures range (maybe -15°C to +25°C). They also require a good defrosting mechanism. Over recent years there have been some improvements in these areas, so the efficiency gap between good air source and ground source may have narrowed (not forgetting that cheaper inefficient units still exist).

Another factor to muddy the waters is domestic hot water (DHW). This was only 20% of the total heat in the past. However, and houses become better insulated, the percentage heat requirements for DHW increases, and this reduces the SPF since most hot water heating functions at a lower COP. Furthermore, the increased concern over legionella has probably increased the use of direct electric heaters. This further reduces the SPF values. One conclusion drawn from the EST report on heat pumps was that DHW function (to tap) was considerably worse that we thought (but this applies to boilers too).

An important factor to consider is the official pollution figures for mains electricity. This rose during 2010 from 0.43kg CO2/ kWh to the current figure of 0.517. This is close to our current actual whereas the 0.43 figure was a predicted future value) This revised value was bad for heat pumps since their CO2 savings over other fuels diminished. However, this seems not to have done what it might have done, and pushed up the COP/SPF requirements.

So, COP or SPF values are likely to lower than we first hoped for. At a time when you might expect us to look for ways to improve matters, it seems that many decided that air source are only slightly worse, so maybe that is a simpler and more viable way to go.

Efficiency and running-costs aside, the RHI initiative has changed the goalposts, and the economic case is no longer simply due to the fuel cost saved, but is dependent on the money income from RHI, of which the difference between air and ground source is relatively small.
In summary; whilst ground source promise to give the best long-term benefits, it can be hard to justify their considerable extra installation cost, and an alternative strategy could be to fit an air source system, and take other measures to minimise your heat demand. Means of reducing the winter peak is another tack, and wood fuelled winter back-up is one suggestion. (maybe boiler back-up for cities)

My personal choice still favours ground source, and real-life tests in Germany confirm that this system may be on average 20% better than air source. Furthermore, a ground source unit will tend to outlive an air source unit, furthering the long-term benefits of ground source.
There are of course other considerations; will severe winters like 2010 become frequent? Whilst the extreme periods are relatively short, it is far from ideal relying on electric heating when air source heat pumps are failing, so finding the best solution is quite a challenge.