Why no one is asking questions about Spain’s mysterious missing nukes

May 5, 2025May 5, 2025 by Ketan Joshi In Longform 40 Comments

Large scale blackouts happen all the time, but when they happen on a grid with a decent volume of wind and solar, a disinformation machines spins up within minutes to lay the blame solely, loudly and repeatedly on renewable energy. See: South Australia, 2016. UK, 2019. California, 2020. Texas, 2021. It’s a thing.

It was no different after a major blackout hit Spain last week. A narrative pushed by a few business columnists, nuclear power advocates, centre-right ecomodernists, very-right fossil fuel advocates and climate delay practitioners contributed to a widespread assumption that the blackout occurred thanks solely to Spain’s investments in wind and solar.

There has been no official detail on the sequence of events that led to the blackout, besides a few vague details on disconnected generation. That hasn’t sapped the vibrating excitement the people above have shown in aggressively pre-blaming the blackout on renewables. The debate centres around the inherent characteristics of different types of generation technologies: traditional generators use large spinning rotors that have, in the past, helped secure systems through the inertia of that spinning mass. Newer technologies like wind and solar operate on a different paradigm, and deeper structural changes are needed in systems with high volumes of both to ensure grid reliability. There’s a thick layer of politics and identity on top of that, along with direct financial interests and political corruption going in many different directions.

This post is not about spinning rotors, nor is it about what ’caused’ the Iberian blackout. Until we know more, I want to interrogate a narrative that’s sprung up in the wake of the incident: that nuclear power is the cure.

Spain is a global leader in the deployment of both solar and wind power. It’s #6 on the global country ranking for wind/solar proportion (43%), for 2024. But Spain is also one of 30 countries in the world with substantial nuclear power (20th on that list). It has a 20% share of nuclear in 2024: higher than the US, UK, Russia, Canada or China.

The widespread claim that Spain could’ve prevented the blackout with nuclear power seems to weirdly ignore the fact that Spain already has nuclear power – more than most other countries in the world. So….what happened?

The confused narrative

The government plans (very controversially) to phase nuclear out by 2035 – a decision that seems to have been wavering in the weeks before the blackout and that has also provoked the ire of communities that rely on those plants for jobs and local economic benefits.

That future plan somehow transmogrified into the broad belief nuclear had already phased out. You can find thousands of examples of this, but here’s a particularly badly phrased one:

More absurdly, Bloomberg’s columnist Merryn Somerset Webb called for “speed building” nuclear power in response to the crisis.

“Perhaps if we’d embraced nuclear properly 30 years ago, we wouldn’t need to worry about emissions— or for that matter grid security. Which would be nice. Relaxing even” (Merryn ends the piece with a list of investment suggestions for uranium stocks).

Got some good news for ya, Merryn. Spain did in fact embrace nuclear properly thirty years ago, and that capacity has remained steady ever since, according to the latest update from the World Nuclear Association:

Spain has plenty of nuclear power (in addition to hydro, gas and a small amount of coal, too). Nuclear power generation has barely budged over the past two decades, in terms of how much it generates as a proportion of the country’s total. Whatever Spain’s planning to do in the future, at the time of incident on April 28th, it had as many nuclear plants as it did when I was five years old, and in recent years, those plants have generated at a steady pace.

So what happened in April, 2025?

The shutdown

At the time of incident, Spanish nuclear power was generating 3,384 megawatts of a potential 7,123 megawatts. As a percentage of total possible nuclear output (‘capacity factor’), it was 47.5% (which isn’t all that far from offshore wind’s average capacity factor of about 40%). That is unusual for nuclear power, which almost always operates at a far higher capacity factor.

That is strange. First, I checked some recent data – you can see that the week preceding the blackout was one that saw high (but not unusually high) wind/solar output, along with weirdly low nuclear output:

To give you an idea of how weird nuclear power output was, here’s the country’s daily nuclear generation from the start of 2023 to April 27th, the day before the blackout. You can see the dip down low on the far right of the image:

Or, to demonstrate it another way – you can compare April’s nuclear power generation between 2025 and 2024:

Maybe this happens regularly outside of April, and we’re looking too narrowly? No – it doesn’t. This reduction in nuclear power is highly unusual for Spain. Even if you only look before the blackout, April 2025 was inarguably destined¹ for the lowest nuclear power generation since at least 2015, according to Ember’s data:

This is significant, right? Nuclear power was abnormally low in the weeks preceding and on the day of the blackout. So, I wanted to find out why.

The weird justifications for Spain’s nuclear shutdown

The most direct reference I can find is this, from a Reuters article, referencing the two units of the Almaraz nuclear power station (only 2 gigawatts of the total missing 3.7 at the time of the blackout):

“[A week after a debate on the future on nuclear in early April], Almaraz temporarily shut down the two units citing abundant wind energy supply as making operations uneconomic. One unit was still offline on Monday [the 28th, the day of the blackout]”

Another sources cites this reason:

“….the economic unviability caused by the high taxation of nuclear plants, the low electrical demand in recent days, and the high generation of renewables” 

It is absolutely zero surprise here that the shutdown of nuclear power is being blamed on renewable energy. Nuclear advocates have a pathological decades-long history of blaming every single one of their woes on the dirty greenies, and this seems no different. Bloomberg recently ran this line, but it was vigorously denied by France.

Almaraz 1 and 2 comprise two of Spain’s seven gigawatts of nuclear: both the oldest plants, connected to the grid in 1981 and 1983 respectively. That means they are both first on the chopping block for Spain’s nuclear phase-out: “The planned closure of two nuclear reactors at southwestern Spain’s Almaraz plant, starting in 2027, will increase the risks of blackouts, European power lobby ENTSO-E said in April”.

We can do a simple sense check on both of those claims: first, that wind power output was freakishly high in Spain, and also whether the price of power in April has been abnormally low. Nuclear power has never shut down to this depth in Spain – so these circumstances must be similarly abnormal, to justify the shutdowns, right?

Well – not really. April was not a particularly high month for wind power output in Spain. In fact, the days preceding and the day of the blackout were kinda low for wind power:

This is…..pretty significant. For at least two of the missing gigawatts of capacity, the rationalisation offered, blaming high wind output, seems to be questionable.

Maybe they meant wind and solar together? It is, unsurprisingly, the same story: wind and solar’s combined contribution in April 2025 was high but absolutely within expectations and recent history:

What about the claim of the ‘economics’ of running nuclear in a high VRE grid? It is true that, generally, more wind and solar depresses wholesale power prices. It often results in smaller profits for every power generator (including wind and solar, in a self-destructive way).

While April’s wholesale electricity price was low relative to recent months, the post-energy-crisis earning per megawatt hour have been massive, and 2025’s April prices were on average higher than 2024, 2020 and 2016. Low wholesale prices have happened before in Spain and don’t seemed to have triggered the sudden obliteration of more than half the country’s nuclear capacity.

There could be an element of truth to this, given the prominence of negative prices in April 2025 – but this was a Europe-wide phenomenon, and I haven’t seen reports of nuclear plants in other European countries responding by simply shutting down entirely and de-rating output to a fraction of capacity. It is also worth noting that wholesale prices were driven negative in Spain partly due to hydro power plants generating in the middle of the day rather than just wind and solar output.²

Spanish wholesale power prices were unusually high in the months leading up to nuclear’s shutdown in Spain. They were relatively low in the weeks prior, but not abnormally so. What was different about mid-April 2025, that justified the unprecedented, partly-intentional, partly-mysterious shutdown of half of Spain’s nuclear fleet?

Make this make sense

Looking deep into this data flips the script. The story told in media is that Spain was suffering from ‘too much solar’. Who has been asking why half of Spain’s nuclear fleet was taken offline at the time of the grid instability?

We can say with confidence that on April 28th, Spain was not seeing an abnormal volume of wind or solar generation relative to the past few years. When the country’s nuclear plants were shut down, wholesale prices were low, but again not abnormally or unusually so (and not relative to other similarly nuclear-enabled countries).

The only truly unusual profile that emerges is nuclear power seemingly being downrated to its lowest level of generation for at least ten years (if not much longer), in the weeks preceding the blackout. The only verifiable information for some of this downrating relates to a politically red-hot nuclear facility currently in the midst of an intense debate about its future, which seems to have intentionally shutdown with at least questionable justifications around opaque economics.

The muddled story of nuclear power as a saviour for grids, which we should “speed build” immediately, is questionable. If we accept that as true, the implications are massive: even when nuclear power is installed and present, it can’t be relied upon to supply energy or stability unless the cost of power is above a certain level – erasing all of the benefits of capturing the free fuel of sun and wind. Nuclear shutting down if power is cheap means that as electricity gets cheaper for consumers, nuclear becomes significantly more unreliable.

If we accept the rationalisations offered by nuclear power operators, it screams of a broken incentive system around grid stability that seems to have had extremely significant consequences for Spain and Portugal. In this world, nuclear power is fundamentally incompatible with a world in which we seize on free liquid, atmospheric and solar resources to supply electrical energy. Nuclear is only the hero of the power grid when the price is right³. It raises the simple question: surely there’s a cheaper, more reliable way of integrating wind and solar? One that doesn’t demand a persistently high power price?

If we’re a little more cynical, I think there’s at least a chance politics played into these oddly-rationalised decisions, considering how significant the blackout will be when it comes to decisions around the future of nuclear power in Spain. Strategically witholding capacity as a tool for creating the political, media and social conditions for self-preservation isn’t unprecedented. Without suggesting at all this is what has happened here, it’s worth revisiting the Enron scandal, in which power traders withheld capacity and caused major rolling blackouts for profit. All this comparison means is that some basic questions should be asked here – but they’re not.

Which brings me to something important: why the broader energy media has failed – really, really badly – to interrogate the behaviour of nuclear on Spain’s grid. Here are some questions I think should be asked:

• Can the operators of the Almaraz nuclear power plant provide justifications for not providing electrical energy to the grid in any detail beyond vague claims of ‘economic’ decision making?
• How expensive does electricity need to be for “reliable” generators to actually provide reliability? Do they publish this anywhere?
• Nuclear power plummeted to a minimum of ~2 gigawatts on the 17th of April, meaning many other plants beyond Almaraz were deactivated to a collectively unprecedented level, and Spanish nuclear had a capacity factor of 29%. Which plants? Why? Were these outages, or intentional shutdowns like Almaraz? Were they planned?
• Was this decision process known to Spain’s grid operator? Are they factoring in nuclear power’s declared inability to generate electricity when power is cheap into their models around system strength and system security – and what alternatives exist?
• What is the near-term outlook for Spain’s nuclear power plants? Given they’ve now operated far outside their normal bounds, how much can they be relied upon to provide energy or system strength?

Other points…

It is striking that on the day of the blackout, April 28th, solar power output was high, but not abnormally high⁴. It was at a share of 60% when the blackout occurred – there are hundreds of intervals, mostly in 2024, which saw a higher share of solar power among all generation types in Spain. While June hasn’t come around yet, 2024 still beats 2025 in terms of record-high solar intervals:

It is also worth noting that in the same way wind and solar generation were not abnormally high on April 28th, the Iberian Peninsula is not abnormal when it comes to high shares of variable renewable energy. This September 2024 IEA report lays out, very clearly, that other regions like the UK, Germany, Denmark, Chile, Ireland, Australia, Kyushu (Japan), Texas and California all manage varying levels of wind and solar, with varying degrees of interconnection into other regions – each with their own suite of challenges, too.

The entire report is excellent – and a demonstration that the complex challenges of grid integration were being discussed, honestly and openly, far before anything happened to Spain’s grid.

1. Outside of COVID19, the lowest month of nuclear generation in Spain since 2015 was 3.46 GWh. By April 27th, cumulative nuclear gen in Spain was 2.7 TWh. To beat the next lowest month, it should’ve generated 0.76 GWh over three days (28th to 30th), or 0.25 GWh per day. The most it has ever generated on a single day is 0.17 TWh. So – before the blackout, Spain was destined to have its lowest ever month of nuclear power output ever – for sure. ↩︎
2. They have too much water and need to use it to generate power, in short. Read this article for a good explanation ↩︎
3. I should have mentioned this earlier, but Brett Christopher’s book, ‘The price is wrong‘, is a really good exploration of what cheap power actually means for privatised, profit driven markets, including the negative consequences for renewable energy operated for profit. The deeper issue here is that greedy gouging by corporations works best with large, big thermal generators and isn’t particularly compatible with smaller, distributed energy resources running off passive, non-extractive fuels. ↩︎
4. There were a few articles about solar reaching ‘100%’ of Spain’s power in the weeks preceding: this is share of total demand so it differs materially from the data I’m using, which is share of total generation at the given moment. I don’t have interconnector flows or good demand figures, and ENTSO-E doesn’t seperate rooftop PV, utility scale PV or solar thermal, so it’s tough to do this properly. ↩︎