REF analysis of trends in the deployment of solar photovoltaic technology in the UK has been used in a press story today in the Sunday Telegraph (23.06.13) with an editorial comment. Since the development trend towards very large solar revealed by the analysis seems to be surprising to many, we are publishing the following details.
1. In preparing data on operational solar projects we used our own enhanced versions of Ofgem’s Renewables Obligation (RO) data, and the Feed-in Tariff (FiT) register, both of which are available at no cost on the REF site.
2. To analyse future trends in solar development we downloaded DECC's own renewables planning database (RESTATs), which also includes information on newly operational solar projects that have not yet become visible in the RO or FiT databases.
The Renewables Obligation and FiT Databases
3. There is very little Solar PV listed in the RO database at present (about 30 MW), and most of this has in fact transferred to the FiT, leaving only 6 MW active in the RO (we expect RO listings to increase rapidly very soon). Since most of the active capacity is at present in the FiTs database, we began by concentrating our attention on that.
4. The FiT database on the REF website is updated quarterly when Ofgem releases the latest figures for Feed in Tariff supported installations. The latest available set of data is for the period up to 31 March 2013. From this we can see there is 1,585 MWs of capacity of solar PV at 369,912 sites. Of this, the majority is of domestic scale: 1,181 MWs at 358,783 sites. Of the remainder, 342 MWs is commercial (8,978 sites), 18 MW is community (1,619 sites), and 44 MWs is classed as industrial (532 sites).
5. Only 71 out of the 369,912 sites are larger than or equal to 500 kW, with a total capacity of 203 MWs.
6. It is clear that the UK's current 1.6 GWs solar capacity is predominantly of small scale, with the average size of the projects listed in the FiTs database being about 4 kW (0.004 MW).
7. However, that is now changing, with another 1,700 MWs of capacity in the wings, as can be seen from DECC’s planning (RESTATs) data.
DECC’S Planning (RESTATS) Database
8. The government's planning database gives information for solar PV sites of 10kW or greater. It lists 604 sites with a total capacity of 2,359 MWs falling into the categories of Built (632 MW), Under or Awaiting Construction (968 MW) or Submitted for planning permission (759 MW).
9. The vast majority of this capacity is large scale, with 2,336 MW being greater than or equal to 500 kW.
10. Over half of the capacity listed in RESTATS is very large – greater than 5 MW – and comprises 1,522 MWs at 222 sites.
11. Solar farms require about 5 acres of land per MW, so even the smallest of these 222 sites will take 25 acres (a typical suburban garden might be about 1/10 of an acre).
12. Realistically, the land use of any solar installation over 1 MW will seem large to a passerby or neighbour, and there are 359 sites of that capacity or larger, with a total capacity of 2,272 MWs, or 95% of all capacity in DECC’S planning database.
13. Of those 359 sites of over 1 MW in capacity, 143 (615 MWs) have been built, 133 (952 MWs) are either under construction or awaiting construction, and 83 (753 MWs) have been submitted but are still awaiting a planning decision.
14. On the basis of this analysis it is clear that there is a significant development focus on large scale solar farms, with the average size of those currently in planning being 9 MWs.
Mr Barker’s Ambitions for Solar
15. In a speech given in Germany last week, Mr Greg Barker, MP, Minister of State for Climate Change, revealed his ambition for 20,000 MWs (20 GWs) of solar photovoltaic in 2020.
16. As we have seen, there is a significant volume of solar being built or trying to get built, and with the subsidies still very high, it seems that investors might want to build as much as Mr Barker suggests, though we are a long way from his target.
17. While the exact composition of the 20 GWs, should it be built, is uncertain, it seems likely that it would predominantly be larger scale solar, probably registering under the RO or the new Contracts for Difference, rather than the FiT.
18. A conservative assumption for the break-down of subsidy mechanism for solar in 2020 would be 50:50, FiT:RO (or Contracts for Difference, which will replace RO in 2017). In other words, we could estimate that half would be under, and half would be over, 5 MWs in capacity.
19. We know from large scale planning applications for ground based solar that about 5 acres is needed per MW of installed capacity, so on that basis the land take for 10 GWs of very big solar would be about 50,000 acres.
20. Realistically, any project over 1 MW is likely to be ground mounted, and it seems probable, on the basis of the current dash for large solar, that the vast majority of the 20 GWs referred to in Mr Barker's speech would be of that scale (i.e. greater than 1 MW). If three quarters of it, 15 GWs, were of that scale and ground mounted this would require about 75,000 acres of land.
21. Estimating the subsidy support needed for Mr Barker’s 20 GWs of solar is not straightforward, since the banding for solar is very complex and we cannot be certain about the breakdown of size and type and subsidy level in 2020.
22. However, bearing in mind the probability that most of this capacity would be in large scale solar farms, which earn less subsidy than smaller scale projects, our feeling is that a conservative estimate would be to assume a low figure of about £60/MWh, which is the lowest level of subsidy under the RO at present (1.2 ROCs), and apply this to all 20 GWs. It is in fact likely that there will be higher levels of subsidy paid to a large part of the 20 GWs, but for the purposes of this estimate we have ignored this potential.
23. Thus, assuming a 10% load factor, 20 GWs of solar would generate about 17,500,000 MWhs, and receive subsidy of about £1 billion a year for approximately 20 years
Solar PV and the Operation of the GB Electricity Grid
24. The cost of managing 20 GWs of solar in the UK system is not well understood, but National Grid has published some remarks in a briefing note for DECC, where they assess the impact of 22 GWs.
25. The briefing note suggests very strongly that successful management of 20 GWs of solar is practically infeasible without major changes. National Grid writes:
22GW of Solar PV could theoretically be accommodated on the system assuming the interconnectors were in service, capacity to export could be secured and there was a purchaser on the continent. In reality, such an arrangement would not be robust against interconnector problems or losses of demand from the system and would not be acceptable
26. However, the measures required to facilitate the introduction of this much solar generation, including special system management procedures, electricity storage, new grid and distribution network reinforcement, and constraint payments to solar (and presumably to other technologies; see Diagram 4 in National Grid’s document) would all add significantly to consumer costs, and when combined with the subsidy costs sketched above, would probably not be tolerable.