Renewable energy generation is not as robust in the face of enemy attack as it looks. Nuclear power, even though it would probably be concentrated in a few large generating stations, should in fact be a little more dependable in wartime.

This warrants careful consideration, because an enemy may choose incapacitation of electricity supply to pressure Australia into cease fighting, just as Russia has repeatedly attacked Ukraine’s power generation and distribution system.

Yes, solar and wind electricity generation is dispersed into many units of modest output, each relatively uneconomical to attack, as Chris Douglas points out in an article in The Strategist. But supply-firming installations, such as batteries and peak-demand generators, are not as numerous and can be targeted instead. Moreover, renewable generation is just as reliant on a limited number of key transmission lines and substations in the distribution grid as big power stations are.

Also, whether an enemy would take the huge step of attacking nuclear power stations is doubtful.

How easily an enemy could cripple electricity supply can be analysed in terms of the criticality of points in the system, their vulnerability and their recuperability.

Consider a hypothetical but plausible case in which seven nuclear power stations have a combined capacity of 11 gigawatts. Each is critical, because losing its output would cause great economic disruption. But the same applies to supply-firming installations that back up renewables. Australia’s eight largest planned firming installations have a capacity of about 11 gigawatts, so the concentration of supply in critical assets is quite comparable in the nuclear-generation and renewables-generation scenarios.

As for vulnerability, concentration of generation capacity in nuclear stations would simplify protection with air and missile defences. In the all-renewables scenario, defences would instead cover the largest firming assets, but we would have to accept risk with the remainder.

If an enemy overcomes air and missile defences and scores hits, nuclear power stations may actually be less vulnerable than the equivalent firming assets, because they are necessarily built with impressive, passive defensive and protection measures. To achieve a high probability of kill our adversary would likely need to increase the volume or yield of the missile attack.

Now, penetration of the containment building is likely to knock out the station out regardless of damage to the reactor, because cooling systems, heat exchangers or condensers will likely be damaged. But a catastrophic release of radiation is less likely. And in fact the more vulnerable part of the station is the adjacent turbine hall. But it would be no more vulnerable than firming installations—gas or hydro power stations or large lithium batteries.

Then there’s the question of recuperability—how quickly installations can be brought back on line after damage. In the nuclear-generation scenario, reactor damage would be catastrophic and take months to years to recuperate. But, as noted above, it would be improbable, and the more likely problem would be damage to ancillary systems, turbines or the station’s transformers. They could be fixed or replaced in about as much time as the turbines or transformers of firming installations could be, and possibly faster than an array of large lithium batteries. Note that replacing ancillary systems, turbines and transformers would not be far from current Australian manufacturing capabilities.

So, in both nuclear-generation and renewable-generation scenarios there is risk in concentrated sources of electricity, but on balance the nuclear set-up is not appreciably worse and may even be better.

Regardless of the means of generation, the main vulnerability is in the distribution network. While renewable generation assets are numerous, they are also functionally useless in isolation: the system works only if networked together so that when the sun isn’t shining in Queensland, for example, the South Australian wind can take up the slack.

There are about 305 substations across the National Electricity Market, and destruction or disablement of only the four main interconnector transmission lines is likely sufficient to splinter the system. Destruction of only eight substations is likely to isolate the main generating regions from the capital cities, regardless of how much renewable capacity there is.

Substations, being compact and not particularly robust, are also much more vulnerable to attack. Vulnerability will rise in the renewable-generation scenario. Because renewables capacity is often built far from demand, the distribution grid grows, increasing the area to be protected, and relying more on internet enabled control systems. This increases vulnerability to additional attack vectors, including cyber and sabotage. The protection against this may be far more draining on limited national capabilities.

The distribution network would be relatively recuperable, because replacing a transformer is a simpler task than repairing a turbine or pressure vessel. US experience tells us that the job typically takes eight to 35 days. But the sheer ease of re-attack and quantity of targets may exhaust our stock of repair or replacement components quickly. Similarly, trying to protect each node with surface-to-air systems would exhaust our available forces.

Lastly, it should be pointed out that targeting nuclear power stations has been deliberately avoided by both sides in the Ukraine war, and it is beyond dispute that attacks on nuclear facilities in any conflict would be regarded as escalatory. While non-binding common practice should not be used as a mitigation in its own right, it adds a degree of complication to the enemy assessment.

Nuclear power stations may actually be the safest, and most defendable part of the electrical system.