This is a joint project with Peter Dudfield.
This post is acccompanied by the UK Wind Curtailment Monitor App. The monitor is updated every hour to show current and historic levels of wasted wind power.
Last year, the UK generated ~30% of its energy from renewables, of which windpower (~23% total generation) was by far the biggest contributor.
But on the windiest days, we deliberately capped the amount of power our turbines were producing, reducing the total amount generated by 6%. In fact, it’s worse than that: not only did we turn off our turbines, but we paid the owners of windfarms to turn them off. This is called curtailment.
In 2022, a year characterized by extraordinary hikes in energy prices for consumers, we spent £215m on turning windfarms off, and then another £717m turning on gas power plants to replace the lost wind power. In the process, we emitted an extra 1.5 million tonnes of CO2.
Why are we wasting so much wind power?
To understand why this happens, let’s look at the distribution of large windfarms across the UK:
Most windfarms are either in Scotland, or in the sea. That’s because:
- England banned 1 the construction of windfarms onshore in 2015 (a move which looks likely to be reversed)
- Scotland, and the sea, are very windy
- There are relatively few people living in Scotland (or the sea), which makes it easier to get planning permissions without getting snarled up in NIMBYism
However, as you might expect, most of the UK’s electricity consumption is not in Scotland (or the sea). It’s concentrated in the South East of England, where most of the people are.
As a result, we often have to move electricity from North to South. The map below, from the National Grid, shows areas of surplus energy generation in blue and those with a deficit in red, and the resulting transfers needed to balance the grid.
This poses a problem, because moving electricity long distances is expensive. You need big cables, which are serious bits of kit – the last large one we put in cost £1.2 bn. At times, we just have more windpower than we have cables to transmit it. The particular hotspot for this problem is the B6 boundary: the bottleneck for electricity from Scotland to flow to England.
When we’re generating more windpower than we can transmit, the National Grid pays the windfarms to turn off, and pays a (typically gas powered) alternative generator, closer to the demand, to turn on. Consumers end up effectively paying three times for the power they’re getting: the original payment to the windfarm for the electricity, the payment to turn off, and then the payment to the alternative generator.
In the past, it has made financial sense to avoid the expense of building extra cables and instead pay a bit more to replace the lost wind power with gas generation in the South. However, with gas prices surging, this doesn’t look like such a good trade off for consumers, not to mention the planet.
How big a problem was this in 2022?
We’ve built an interactive dashboard for exploring curtailment in 2022. You can explore it here.
At times the UK was wasting as much wind power as we were using.
On Christmas day, we spent £9.2m on curtailment costs, curtailing a total of 76.18 GWh. That’s enough electricity to power ~11’000 households for a year.
What should we do about this?
This problem isn’t going away. As of September 2022, 78% of the new onshore wind projects in the UK were located in Scotland. This represents another 29GW of generation – more than tripling the total wind power North of the border. National Grid’s own projections (Future Energy Scenarios, Exec Summary, Page 11) suggest 15 TWh of curtailment by 2030 in all Net Zero aligned scenarios. Although the cost per MWh is likely to decrease with the expiry of Feed In Tariff and Renewable Obligation Certificate schemes 2, the total TWh curtailed will rise steeply over the next few years.
Build more cables
The simplest way to fix the problem is to build bigger cables, allowing all the wind energy we generate to be routed to the places that need it. The bottleneck in the system is the so called B6 boundary, which is the interface between the Scottish transmission network and the English one.
The most promising plan to increase the bandwidth at this point is the system is to add a High Voltage Direct Current (HVDC) link off the East coast, a twin to the Western HVDC which snakes between Hunterston and North Wales in the picture above. The Eastern HVDC (detailed proposal) will comprise a pair of connections, one between Peterhead and Drax and the other between Torness and Hawthorn Point.
The Eastern HVDC is estimated to cost ~£3.4bn, and will add 4GW of capacity. Maximum curtailment in 2022 was 5GW (on November 11th at 5AM), so this wouldn’t quite have covered that, but that was an unusual day. 99.7% of the curtailment we saw in 2022 would have been solved by an additional 4GW of transmission.
However, laying high voltage cables is slow – much slower than building new wind turbines. In the time it takes for this transmission to come online (2GW by 2027 & 4GW by 2029), we will have added far more new wind capacity North of the B6 boundary.
Simply put: we can’t lay cables fast enough to solve this problem.
Add energy storage at bottlenecks
Battery storage has grown from a few MW to 1.7 GW in the last few years (source). Batteries can soak up cheap wind energy when it’s abundant, and discharge it when the wind has stopped blowing and congestion has eased. Batteries are certainly easier to deploy than cables, so its more plausible that we could add battery capacity fast enough to keep up with new turbines. National Grid project that we’ll have 35GW of battery storage across the whole of the UK (National Grid’s Future Energy Scenarios) by 2050. But filling a battery up and leaving it charged for days on end is not attractive – most battery operators make money by cycling (charging and discharging) at least once a day. Of all the interesting things to do with batteries, it’s not clear that solving curtailment will be the most lucrative.
Pumped hydro – basically pumping water uphill when you have lots of electricity, and letting it run downhill through turbines to turn it back into electricity – is a great option for longer duration storage. This is particularly useful for storing and releasing electricity generated by longer spells of high wind. Drax, who are best known for running a biofuel generator in North Yorkshire, are aiming to expand their Pumped Hydro storage at Cruachan Dam, whilst SSE’s Cloire Glas project will hit 1.5GW when complete sometime after 2025. As with cable laying, it seems unlikely that such large and complex projects – which frequently involve hollowing out mountains – can match the frenetic pace of on and offshore wind deployment in Scotland.
Hydrogen, rarely out of the spotlight, provides another wildcard. Its proponents emphasize its complementarity to sporadic renewables: when there’s too much wind, electrolyzers could generate hydrogen from water, benefitting from free electricity. Its detractors point out that electrolyzers are so expensive that running them only when electricity is cheap doesn’t make sense: buying expensive machinery and leaving it idle for long periods of time is bad business. Moreover, once you’ve generated the hydrogen, you need infrastructure to get the energy back out in a useful form, either by turning it back into electricity (a wasteful round trip process in which you lose about 70% of the energy), or burning it in homes or new power plants. The UK government has been consistently bullish on Hydrogen, and it plays an important role in 2 out of 3 of the National Grid’s Net Zero by 2050 scenarios (source). Given that we lack any examples of large scale electrolyzer deployment paired with intermittent renewable energy sources, it’s very unclear how much of a role hydrogen has to play as a form of storage for Scottish wind power.
Factor location into electricity prices
Electricity prices – which are the main way that incentives are communicated to market participants – are completely location independent in the UK. This means that a generator with a wind farm on the Outer Hebrides can sell a MWh to an electricity supplier with customers in Surrey, and neither of them has to worry about how the energy gets from point A to point B 3. Conversely, a wind turbine in Surrey sees no price benefit from selling its MWh to a customer next door, despite the obvious efficiency benefits.
If generators and consumers aren’t worrying about where energy is flowing from and to, who is? The messy, geographically nuanced reality of ensuring enough electrons are flowing to each consumer becomes the problem of the National Grid Energy System Operator (NGESO), who have the final responsibility for keeping the lights on.
They do this via a system called the Balancing Mechanism (BM). As the name suggests, the BM allows NGESO to fix all sorts of problems that might leave the grid out of whack, ensuring that the amount of electricity generated and consumed are the same across all locations in the network. This is increasingly critical to preventing blackouts. What was meant to be a light-touch process to tweak the equilibrium already established by the free market, has skyrocketted in importance over the last decade. Due to the rise of renewables (and interconnectors to other countries) and the location-agnosticism of the market, NGESO now routinely has to intervene in more than 50% of generation (source).
Unfortunately, this rather frenzied balancing process – which occurs in the hour before power is going to be used – is a very costly way to optimize the system, resulting in the perverse solutions we’ve described in this post. A much more elegant solution would be to inject that locational information into the market, placing more responsibility on generators and consumers to worry about where the energy is coming from and where it’s going. This idea is known as locational pricing, and it’s already used in New Zealand, Canada, and several US States.
Effectively, this would mean breaking the UK energy market into lots of small chunks (formally: nodes or zones), and allowing prices to develop independently in each chunk. This would mean much lower power prices in places with high generation and low demand (the North) and much higher prices in places with lots of demand and not much generation (the South). This would provide an excellent incentive for generators, who might find it profitable enough to brave the teeth-gnashing NIMBYIsm of the South East and build more generation. Conversely, energy-intensive industries would find it appealing to move North, closer to existing generation – conveniently aligned with the Levelling Up agenda.
NGESO is in favour of this plan, and reckons it could deliver on this quite dramatic restructuring within 5 years (Net Zero Market Reform Phase 3 Conclusions), with Energy Systems Catapult & Octopus energy estimating the savings at £30bn by 2035. Other market participants are reticent, remembering the pain of implementing the last large market reform in 2001 4, and wary of destabilizing a system that looks increasingly fragile. The UK Energy Research Centre observe that although the current arrangements are increasingly expensive for consumers, they provide a high degree of certainty for generators, because they get paid regardless of the constraints in the network. This favourable arrangement has underpinned the UK’s impressively rapid growth of wind power, an undeniable success story in the race to Net Zero. There’s understandable reluctance to sour the investment environment in a country which is already suffering from other self-inflicted economic wounds 5.
Locational pricing remains a vigorous topic of debate as part of the government’s Reform of Energy Market Arrangements (REMA). NGESO, an influential participant, has clearly stated their case for a locational pricing scheme. BEIS (Department for Business, Energy, and Innovation) will combine this with feedback from other stakeholders across the industry and perhaps a dash of politics and we’ll see where we end up 6
Ultimately, like many discussions around solutions to climate changes, this is a “yes, and” rather than an “or” choice. We can, and should, build more transmission capacity and storage, whilst reforming the market such that the UK’s phenomenal success in deploying wind power can be finessed to more precisely match our energy needs.
Although the headline figures are dramatic – £900 million 7 is not to be sniffed at – we shouldn’t let them overshadow the fact that the UK’s adoption of wind power is a huge success story. Whilst it’s unfortunate that transmission hasn’t kept pace with generation, it’s partly explainable by the fact that generation has been deployed so quickly. 2022 saw the highest curtailment yet, but also had the highest production of energy ever, at 74 TWh (Bloomberg Green). We’re laying the right pieces of track for a zero carbon energy system by 2035. We just need to be careful to lay them in the right places.
Thanks to Mike Ryan for the initial inspiration and (many months later) for feedback on this post.
- Technically a moratorium based upon impossibly stringent local planning approval processes
- Many wind turbines were built under the Feed In Tarrif (FIT) or Renewable Obligation Certificates (ROC) scheme. If the wind turbine doesn’t generate, they cannot be reimbursed via these schemes – so the National Grid ESO has to pay them the amount instead. As more of the schemes expire, the cost to turn down the wind turbines should decrease to closer to the actual operating cost of rotating the wind turbine blades out of the wind – not very much
- This is a slight simplification, because at the point of the construction, the wind farm pays a connection fee which represents the difficulty of access to the network. However, that fee is one-off; it doesn’t cost the wind turbine more to use the connection when the rest of the grid is congested, and so it doesn’t help with congestion. In fact this connection fee guarantees access to the market – a so called “firm” generation guarantee
- New Electricity Trading Arrangements, or NETA
- The UKERC point out another issue: locational pricing might kill the incentive to install more wind in Scotland, without addressing the roadblocks to installing wind power elsewhere. This concern may be allayed by Sunak’s signals that the moratorium in onshore wind is going to be overturned. However, evidence from Texas suggests that windfarms do not end up near population centres even in markets with locational pricing and liberal planning laws
- . Locational pricing is not the only topic on the table as part of REMA. More exotic schemes, such as splitting the wholesale markets into two separate renewables and gas markets, are also being floated. Most people I’ve read seem to think this is a bad idea – here’s Guy Newey from Energy Systems Catapult’s take.
An alternative is to devise additional incentives for folk North of the border to use power when the grid is curtailed, soaking up constrained generation. This is the idea behind National Grid’s recent deal with Piclo Flex, who provide an interesting platform whereby the people operating the grid can directly incentivize energy users (by paying them!), cutting through the intervening layers of noise and conflicting incentives in the market.
- This seems like a very big number, but total necessary spending on Net Zero is estimated at about £50bn, so 2% on curtailment might be considered a small price for a carbon neutral power system