As I anticipated, my post on Tesla’s new battery provoked some pretty hostile responses, most notably from pro-nuclear diehards. I’ve written plenty on this (use the search facility), so rather than repeat myself I’ll make an observation drawing on the previous post.

Ten years ago, solar PV was a faintly hopeful technologica prospect, making a minuscule contribution to electricity generation. Today, it’s a reality that is creating massive disruption for electricity utilities around the world. As I said in the previous post, the availability of even moderately cost-effective storage removes the last big obstacle (more on the economics soon)

By contrast, ten years ago, nuclear energy was a mature technology which seemed to be at the beginning of a renaissance. Today it’s further away, in almost every respect, than it was in 2005. Construction times have blown out, costs have turned out to be twice as high or more than expected, the operating record (thanks to Fukushima) is far worse, and the various new technologies (SMRs, Gen IV) have receded even further.

None of this means that the replacement of fossil fuels with renewables+storage is going to happen under current policy settings. But such a replacement is now clearly feasible, much faster, more reliably and at much lower cost, than attempting to reboot the failed nuclear renaissance.

Backing the nuclear horse was a reasonable choice in 2005. But it’s dead, and flogging it won’t revive it.

No one seems to have spelt this point out, but there’s an obvious potential for Powerwall to be used in ways that benefit coal, nuclear and geothermal power, as well as renewables like wind and solar. Advocates of these technologies love to cite the fact that they are “baseload” supplies, but this is a misconception. Because they are costly to turn on and off (or even up and down), these technologies produce too much power at times of little demand (late night and early morning).

If owners of home solar systems, connected to grids with an off-peak excess supply, install battery storage on a large scale, it would make sense to run two cycles per day. The systems (most sensibly oriented west) would charge up from solar panels in the early afternoon, and supply power in the evening. Then they would recharge from the grid in the early morning, and supply power to meet the morning peak associated with getting ready for work, school etc.

What’s the net effect of this. First, obviously, it makes storage a more appealing economic choice for householders. Second, although it reduces costs for any kind of electricity that is not fully dispatchable, the benefits are bigger for renewables for two reasons. First, the variability of these sources is greater. Second, pricing systems, at least those in Australia, are already set up to encourage use of off-peak grid power, whereas current feed-in tariffs discourage solar PV.

From our current starting point, effect of adding more systems with a combination of solar PV and storage will be to reduce total demand for coal-fired power (and, where it exists, nuclear power), and to enable more efficient use of existing capital stock. So, it’s likely to discourage new investment in these sources. However, unless we have a carbon price, or other measures in place, it won’t necessarily accelerate the closure of existing coal-fired plants.

Update A note on the economics: Calculations I’ve seen on the web assumed that lithium batteries have a life of 1000 recharge-discharge cycles, but it appears this number can be improved drastically. These guys are claiming 20 000. More on this soon, I hope.

I expanded my earlier analysis of the Galilee Basin mines in this piece for The Guardian. The really striking number is 483, the number of long-term new jobs the Carmichael mine is estimated to generate in the local (Mackay Isaac Whitsunday) region. That estimate comes from a computable general equilibrium (CGE) modelling exercise by Adani’s own consultants, ACIL Allen. Before the Queensland election, of course, much bigger numbers of 10-20 000 were bandied about. That’s partly a difference of coverage – the bigger numbers envisage, implausibly, that all the proposed mines in the Basin will go ahead, along with rail lines and port expansions.

Also, some of them focus on peak numbers during construction for each project, so that the jobs in question would only last a year or so. But the big difference is that the larger estimates were made using the discredited input-output method, in which each job created directly generates many more indirect jobs. This is an extreme version of the Keynesian multiplier effect, valid during a deep recession. But, as ACIL Allen observes, it makes sense only if you assume that the recession is going to last for the life of the project.