Senator (and Presidential pre-candidate) Kamala Harris and Rep. Alexandra Ocasio Cortez (not a candidate but lefty star) have published a draft Climate Equity Act. Here it is (pdf). It provides for principles, an Office, reports, consultations, and a platform for “frontline communities” to share their pain with the denizens of the Beltway. It reads like the work of a New Age therapist working in the bureaucracy of the late Austro-Hungarian Empire.
Missing: any proposals for action that would actually do something for unemployed American coal-miners in say Harlan County, Kentucky.
Here’s my idea.
A 100% renewable electricity grid – actually a 90% one – based on cheap wind and solar electricity needs a lot of backup or firming to cover the gaps when there is no solar output (called “the night”) or little wind (week-long lulls mainly created by the procession of anticyclones that drive the weather in middle latitudes). Today, there is enough legacy baseload coal and nuclear power to reduce the problem, and natural gas to deal with what’s left, but they are all going to phase out soon in the GND. Actually the coal will go anyway regardless of the GND from price competition, and nuclear from age, but this plan is for GND supporters.
There is a longish list of technically feasible solutions or part-solutions. None of them are really cheap; but then, a good part of the cost of the electricity you buy today is to cover the rarely used peak generation capacity and the unused reserve. There are no free lunches here.
There is a lively argument in the “100% renewable” expert trade about the best method of firming. Very lively. Mark Jacobson went so far as to sue Christopher Clack for a hostile rebuttal of his first scenario for the USA, relying for firming on a rather peculiar scheme, since dropped, of retrofitting all existing US hydropower dams to run in burst mode, at much higher outputs for much shorter periods. I don’t include this false start.
Some of these technologies are in flux, others mature. It is therefore impossible to predict now the lowest-cost firming mix ten years ahead. The problem is that in a ten-year GND transition, there isn’t time to let things settle down. Some big spending decisions will have to be taken in the next few years, and some of them will turn out to be wrong in the sense of diverging from the optimum – there is not much risk of being stuck with an asset that simply does not work. The priority is as always to ensure a reliable supply, not to assure ratepayers suffering from power cuts that you were prudently trying to save them every last cent on their bills. The compressed timescale also calls for a strong federal policy lead and assumption of risks.
I want to make a case here for off-river pumped hydro storage (PHS).
It may not work out the cheapest in the end, but it’s a mature technology with no technical risk, known and reasonable costs, long working life, modest environmental impact (note off-river), and scaleable to any volume you want. Existing plants (pdf) provide 95% of the current US utility storage capacity. Its problem is that dams take a long time to build: at best five years, though with much less construction risk than nuclear plants. If the USA is going to rely on pumped storage to any significant extent, it will have to start building it out by 2025. There is no technical reason not to start sooner. Storage replaces peak gas immediately as soon as there is a worthwhile volume of wind and solar, which you already have.
At least one expert, Andrew Blakers of the Australian National University, strongly recommends pumped hydro as the basis for firming a wind/solar power supply, along with more HVDC transmission. He has constructed 100% renewable scenarios (pdf) for the Australian NEM (the grid covering the populated East and South) using just these four technologies, with hourly balancing to match the current demand. This balancing costs an additional midpoint US$21 per Mwh on top of the raw wind+solar LCOE of midpoint US$49, a markup of 43%. His paper gives the (narrow) ranges and offers a large number of variants tweaking the assumptions in different ways. His base case calls for 16 GW of storage for 31 hours, making 490 Gwh, balancing a total annual demand of 205 Twh. The capital cost of the storage, based on replicating a standard unit costed by a hydro engineer, is US$600 per kw or US$9.6 bn for the whole package.
To get an order of magnitude for a US programme on the same lines, we will just scale up Blakers without any apology or attempt at adjustment. US consumption of electricity is 4,070 Twh a year, so the model calls for 318 GW of capacity at a cost of $191 bn. (Cross-check: the one-off PHS plant at Bath County, originally 2.1 GW, cost $1.6 bn in 1985, so on that basis 318 GW would have been $242 bn. The order of magnitude is OK, and there has been technical progress since in reversible generators and in tunnelling.)
Since we don’t know whether the alternatives will be cheaper or dearer, it does not make sense to put all the eggs in one basket. However, we can be pretty sure that PHS, as the dominant historical storage technology and still much the cheapest, will play a significant part. Picking with a pin, a 100 GW initial programme looks reasonable. As of 2017, 40 new PHS sites were already under active investigation by utilities and licenses applied for with eight, so we won’t start absolutely from scratch. But if we do, it will cost a ballpark $60 bn. In the context of the multi-trillion overall cost of the GND, this is clearly doable. The plants are long-lived revenue-earning assets: storage has a price, sometimes a high one. I don’t know what the ROI will be, and doubt if it matters very much.
PHS plants are very flexible on size and can adapt to different geographies. The world’s largest PHS plant, at Bath County in Virginia, has a capacity today of 3GW / 24 Gwh. But many working plants are much smaller, down to 100 MW or so. The programme could be met with 33 Bath Counties or 1,000 100 MW plants, or anything in between. The power generated is proportional to the head, and you can get more work from a given size of reservoirs if you can site the upper one higher. This all gives the planners a great deal of flexibility.
Where should the dams go? As a climate justice measure, it has to be Appalachia, since that is where most of the unemployed miners are and will be.
The mountain range is very extensive, seismically inactive, and high enough with typical crests of 900m. You only need 300m or so height difference for a decent PHS scheme. The number of potential sites is so large that the choice can often be made on grounds of economic deprivation. Socially, dam-building is a nearly ideal economic stimulus. The jobs are manly to match an old-fashioned culture, moderately skilled (highly skilled for tunnelling), and last for several years. Contrast suggestions that unemployed Appalachians should be retrained for installing solar in a foggy climate, or wind turbines on the few suitable hilltop sites, clashing with recreation.
How many jobs will be created? At its peak, Bath County had 3,400 workers on site. Applying the same ratio to our 100 GW programme, that would give 113,000 jobs. This is not realistic: smaller dams have different demands to big ones, the employment peaks won’t be synchronised, tunnelling machines are much better, and so on. But it is certainly enough to put a sizeable dent in unemployment across the region, before counting the spending multiplier in local communities. The ambition of the whole programme may even be constrained by the availability of workers. The jobs are only for a decade, but this buys time to develop other opportunities.
How to set up the programme? It is both large and specialised. The obvious solution is to copy Roosevelt’s TVA and set up the Appalachian Storage Authority, under a joint federal/inter-state governance structure, with borrowing and eminent domain powers and so on. It could have a fixed 20-year life, and sell the dams on to states or utilities before winding up. A programme of earmarked federal grants to states would risk sabotage by GOP state governments, which have shown on the Medicaid expansion that they are prepared to sacrifice the welfare of their citizens to ideology. Centralisation and standardization should also work out cheaper in design and project management. There are risks either way.
I don’t know if the scheme can realistically be extended to the Powder River Basin miners in Wyoming. Since their mines are open-cast and highly automated, the miners are far fewer – 5,535 in the state in 2018. The Rockies have even more and better potential sites for PHS than the Appalachians but they are not SFIK anywhere near the mines. I suspect the climate justice warriors will have to think of something else.
Question to Senator Harris and Representative Ocasio-Cortez:
- Do you support this plan or something like it?
- If not, what is your alternative plan that gives former coal miners decently paid jobs where they and their families want to live?
Suppose you both win your political and electoral battles. If you content yourselves with just creating a cool new federal bureaucracy for climate justice, the miners will say: you may be prettier and better spoken than Mitch and Manchin, but in the end you are just another pair of politicians who spin fine words and let us down. They won’t be entirely wrong.
In this post I have ignored the steelworkers and other groups in Appalachia whose situation is often just as bad as that of coal-miners. The issue here is framed by the two representatives as climate justice, implying specific action for those who must lose their jobs to secure the essential energy transition. In Appalachia, that means coal-miners, and they are the measuring-stick for my plan and for any alternative. The plan will of course benefit other groups as well, and these wider benefits should be considered in the planning.
I have no idea what to do for Texan oilfield roustabouts. They are doing all right for now, but that won’t last. Let’s think of something.
The title is, as alert RBC readers will have spotted, a h/t to this famous passage of Keynes:
If the Treasury were to fill old bottles with banknotes, bury them at suitable depths in disused coalmines which are then filled up to the surface with town rubbish, and leave it to private enterprise on well-tried principles of laissez-faire to dig the notes up again (the right to do so being obtained, of course, by tendering for leases of the note-bearing territory), there need be no more unemployment and, with the help of the repercussions, the real income of the community, and its capital wealth also, would probably become a good deal greater than it actually is. It would, indeed, be more sensible to build houses and the like; but if there are political and practical difficulties in the way of this, the above would be better than nothing.
General Theory, Chapter 10, section VI
My dams, being useful, are “houses and the like”.
As in Australia, the national grid is a good way of keeping storage costs down through geographical smoothing. The Australian population, and hence the variability of demand, is crammed into a single vertical time zone. An HVDC line from Sydney to Perth captures useful smoothing of wind and solar supply but not of demand. From New York to San Francisco, it does both. The grid has an even higher payoff in the USA, lowering the storage costs.
If anybody wants to talk to someone who really knows about this stuff, Andrew Blakers is in the phone book: +61 2 612 55905, email@example.com
Blakers points me to a world atlas his team has prepared with 616,000 (not a typo) potential pumped hydro storage sites identified from satellite images. The theoretical collective storage capacity is a hundred times anything we are likely to need. Some of them are in places like Patagonia and Kamchatka that are fairly safe from the bulldozers, but that still leaves innumerable more useful locations. The database lists 33,000 site pairs in the USA, the majority in the Rockies but a good number in the Appalachians – eyeballing, a few thousand. Total US potential storage 1.5 million Gwh. (The huge spreadsheet does not help you find geographical locations, to explore you have to work off the detailed zoomable map, example here, and then copy and paste the coordinates into Google Earth). Some of these sites will be home to protected snail darters or the like, others would drown the governor’s hunting cabin. That still leaves plenty.