I know, you’ve heard it all before from me and at least a few others: The energy/water nexus is going to bite us hard, and it’s already started. You need water to make electricity (for the most part) and you need energy to provide water. Climate chaos can change the quantity and characteristics (e.g. temperature) of the water available for thermoelectric plant cooing at most locations, meaning that every time we build a thermoelectric plant we’re making a decades-long bet that we’ll have the needed cooling water, and that we can use it without causing other major problems downstream.
Once again, this has gone from “nice theory” to “ugly fact”, as Joe Romm points out in France imports UK electricity as summer heatwave puts a third of its nukes out of action:
To avoid maxxing out on my July quota of irony in the first week of the month, I will simply report this as a straight news story. The UK Times reports:
With temperatures across much of France surging above 30C this week, EDF’s reactors are generating the lowest level of electricity in six years, forcing the state-owned utility to turn to Britain for additional capacity.
Fourteen of France’s 19 nuclear power stations are located inland and use river water rather than seawater for cooling. When water temperatures rise, EDF is forced to shut down the reactors to prevent their casings from exceeding 50C.
…
EDF warned last month that France might need to import up to 8,000MW of electricity from other countries by mid-July – enough to power Paris – because of the combined impact of hot weather, a ten-week strike by power workers and ongoing repairs.
EDF must also observe strict rules governing the heat of the water it discharges into waterways so that wildlife is not harmed. The maximum permitted temperature is 24C. Lower electricity output from riverside reactors during hot weather usually coincides with surging demand as French consumers turn up their air conditioners.
One power industry insider said yesterday that about 20GW (gigawatts) of France’s total nuclear generating capacity of 63GW was out of service.
Much of the shortfall this summer is likely to be met by Britain, which, since 1986, has been linked to the French power grid by a 45km sub-sea power cable that runs from Sellindge in Kent to Les Mandarins.
A statement from EDF played down the heat problems, saying that the French system continued to meet customer demands – but similar heatwaves have caused serious problems in France in the past.
In 2003, the situation grew so severe that the French nuclear safety regulator granted special exemptions to three plants, allowing them temporarily to discharge water into rivers at temperatures as high as 30C. France has five plants located by the sea and EDF tries to avoid carrying out any repairs to them during the summer because they do not suffer from cooling problems.
Aside from the energy/water nexus point, which, try as I might, can’t be stressed too much, this situation also highlights a more general issue: The danger of becoming so reliant on one form of electricity generation. It’s very appealing to say, “All we need to do is standardize nuclear power plants. Come up with one design for the entire plant, make sure it works as desired, and then replicate it.” The problem is that then you have a large portion of your electricity generation, whether based on nuclear or any other technology, that has the same strengths and weaknesses. And in a time of rising temperatures and shifting rainfall patterns, cooling water for river-fed thermoelectric plants is quickly emerging as a much more serious weakness than we thought, as great as fuel supply and CO2 emissions.[1]
The reason for this “surprise” is simple: We’ve become accustomed to a remarkable level of stability in our climate. There are certainly anomalies that arise on yearly and shorter time frames, including drought, heat waves, and severe winter weather. But for a long time we’ve been able to look at places like the continental US or most of Europe and predict with confidence that certain locations will “always” have a good supply of cooling water for an electricity plant. We then build the plant with a 40 to 60 year, or longer, lifespan, and sleep soundly, “knowing” everything will work out.
Thanks to climate chaos, that’s suddenly a much riskier bet, and we’ll increasingly have to build thermoelectric plants that use much less or no cooling water, or site them on gigantic, some would say Great, lakes or near the oceans. At least I’m hoping we’ll have the sense to do that before we suffer a major electricity crisis, much worse than Europe in 2003 or what almost happened in the US SE a couple of years ago. Of course, we’ll still have to deal with the installed base of thermo plants for decades. Consider it yet another example of how our past ignorance, kicked up a notch with at least a pinch of hubris, will force us to live with decisions we wish we could undo.
[1] No, I’m not overlooking the issue of the size of the individual generating plants, in case you were wondering. If you build a 1GW generating plant, you have to deal with the problem of losing a lot of electrons all at once if/when it goes offline, for whatever reason. The more diversified and decentralized our electricity generating infrastructure is (and therefore the less we try to force any technology into being a silver bullet), the less vulnerable it is to individual outages.
