View Single Post
Old 05-23-2019, 06:32 PM
Sam Stone is offline
Charter Member
Join Date: Jun 1999
Posts: 28,554
Originally Posted by Dr. Strangelove View Post
It's over 3% of total capacity. At that rate, 30 years until total replacement. Just think how far along we'd be if we'd gotten serious 20 years ago.
You of all participants should understand the fallacy of extrapolating a curve like this into the future. Early adoption curves almost always look exponential. The reason is because there is low-hanging fruit, and it gets exploited first. There is no evidence that the adoption of renewables will accelerate, and a lot of good logic that says the opposite will happen as we have to implement it in increasingly dubious regions.

For example, solar power first started being implemented in the south. This makes total sense as the south gets the most solar insolation and has much higher capacity factors because the difference between seasons is smaller. But now we are trying to implement it in Alberta, which gets half the solar insolation as the equator and has a factor of three difference between the amount of available solar power in winter vs summer - a much more expensive proposition.

The same is true with wind. The economics of wind are highly dependent on location. You need winds that are high, but not too high, and as steady as possible. There simply aren't enough of those prime places located near population centers in easily buildable areas. We're already putting them in the best places, but the rate of new power generation will slow as we have to start using more marginal locations.

Likewise, there is no evidence that the rate of production that is sustained at current levels can be easily scaled to produce say, 10X as much without creating major shortages and price spikes of raw materials which will drive up the price of renewables, slowing demand.

Another reason why we can expect adoption to slow as its gains a large percentage is that most renewable power today is heavily subsidized through tax credits, manufacturer subsidies, feed-in tariffs, mandates and the like. This may be affordable when you are only subsidizing 5% of your power, but those subsidies will decline over time and eventually go away as you can't afford to subsidize everyone. Even if renewables are a grid parity, losing the subsidies will lower demand.

Then when you get to about 30% of grid power you run into capacity factor problems with solar, unless you build enough solar to provide maybe twice as much energy as needed in summer. You also run into grid stability problems, which has been plaguing Australia, Germany, and Ontario after they included a significant percentage of intermittent power. Germany has resorted to paying factories to close because they couldn't provide enough power to keep them running.

And so far, the rate of adoption of renewables has not even been fast enough to keep up with world energy growth, let alone replace current energy sources. Even at the growth rates of renewables we've seen so far, global GHG emissions are still increasing. I see no evidence that this pattern will change.

Of course, eliminating the last couple tens of percent of fossil generation is going to be hard. But there's also no reason to believe that at the current renewable penetration, the expansion rate can't be further increased. As I've said before, a huge advantage to renewables is the low barrier to entry, particularly with capital costs, but more generally as well.
At the current renewable penetration, fossil fuels will continue to be burned in about the same amount they are now, although the amount of renewables as a percentage of energy remains about the same.

The rate of change tells you where things are going. This entire discussion is about the future. The rate of change--and the change in rate of change--tells you much more about where we'll be than the current state of affairs.
In a complex adaptive system, measurements of instantaneous slopes of trends mean very little. Malthus, the Club of Rome, Paul Erlich and environmentalists in general tried to terrify us with stories of mass starvation by 2000 due to linearly extrapolating population growth curves. Their predictions look ridiculous today, because being able to read a curve does not turn you into a fortune teller who can see the future. The demographic transition came along, and no one saw it coming. Complex systems are full of noise and chaos, and inside noise and chaos can be found all sorts of patterns - mostly meaningless in terms of predicting what will happen in the future.

And you're simply wrong that renewables can't replace natural gas. Not in every situation today at a given price, but price and technology changes. And in a way that's much more favorable to renewables than natural gas.
They can't replace natural gas for load-following or for baseload power. With wind and solar you get what you get in that amount, and tough luck. So on sunny days Germany generates so much power that the spot price of electricity goes to zero, meaning they can't sell that excess profitably. Then when it's dark or in winter they have to import natural gas to make up the shortfall. This is one reason why Germans pay the highest electricity prices of any large country in Europe. And the only reason they can sell any at all is because their neighbors have not followed suit. If all of Europe relied on wind and solar, they would produce about 3X what they need in summer, and would have nowhere to sell it. And in winter they'd have shortages, and no convenient nuclear powers next door to buy it from.

Maybe a technology will come along that makes all this better. But we don't have it today, and if we have to urgently get started on changing the grid, we can't just hand-wave that problem away as something we will fix in the undefined future.

Yes, of course. We're in this whole mess because people forgot that a few decades isn't very long at all.
No, it isn't. In terms of engineering, if we wanted to get to 100% renewable in two decades we'd already have to have the whole design and be well on the way towards breaking ground for every facility we'd need - many thousands of them. We are still at the point where we are debating what to do, and we've now mired the debate in partisan politics thanks to people like OCS tying climate change to 'fundamental transformations of society'. That will ensure that nothing substantial will EVER be done. We'll continue to fiddle away on the margins and bicker until it's too late to do anything.

I want to be clear about one thing: I generally oppose closing down nuclear plants. This is not a symmetrical situation at all; an operational nuclear plant has already paid most of the cost (not just dollars, but in NIMBY capital, regulatory hurdles, etc.), and so keeping it running as long as possible is a very good idea. Opening new plants--or spending $9B to fail to open a new plant--is not a good idea.
Would you accept a plan to accelerate nuclear utilization through modernizing existing plants or building a new plant on the same site, therefore avoiding a number of regulatory or NIMBY hurdles? That seems like at least a reasonable place to start. We could begin construction of new plants on the sites where existing plants are, and if they come online while the existing plant is still working we can feed both into the grid. Then when the old plant needs to be decommissioned, the new one can just take its place.

It's not disingenuous at all. We are interested in net expansion. The news is good about coal: massive plant retirement and zero expansion. For natural gas it's mixed; they do see significant expansion, but some of that is countered by retirement. For renewables, it's all good news: massive expansion that isn't negated by retirement.
That's quite an unfair comparison as most renewable power is quite new and you wouldn't expect to retire it, while most nuclear and coal plants are getting long in the tooth. Let's see how those numbers look in 20 years. Wind power stations are designed for lifespans of 20-30 years. But they may not show up in decommissioning numbers because the turbines can be replaced piecemeal as they fail. So a station might not be decommissioned for 50 years, but by then it may have turned over its entire array of turbines two or three times. The cost is there, it just doesn't show up in the statistic you are using.

But entire wind stations ARE being decommissioned. Here's one: UK Blyth Power Station Decommissioning

You can go on and on about "that's not a solution that works in the real world", and yet that's what's happening. We do need pumped hydro, big batteries, and other storage systems. These are already known to work at scale, but need to be deployed more widely.
All of which add more cost and more inefficiency. Pumped hydro loses energy at the pump, through friction in the pipes, evaporation in the reservoir, friction coming down the pipes, and generation losses. All in, you're probably losing 20-40% of your energy to storage losses. The same goes for batteries. What does this technology due to the grid parity of wind and solar? It makes it much harder to achieve.

We do need a beefier grid to transport energy across the country. That stuff is easy, relatively speaking--we've been doing it for decades, and just need to spend more capital on it. We need more efficiency. We need a smart grid where we can shut down heavy but optional use (like aluminum smelting) if the load is too high. We need all these things and more.
Things we've really never done at scale, with unknown effects on the economy. Clean, undisturbed, constant power is a vital input to our manufacturing economy. It is dangerous to screw with it. One of the reasons America's manufacturing is so efficient is because of access to cheap reliable power. If we take that for granted like Germany has, we could pay a big price.

And finally, we will probably still have some natural gas use, but it will be infrequent and very expensive because it will only be needed when all the other systems fail, and so there will be strong motivation to reduce the usage as much as possible (as long as we don't let energy monopolies form).
Why would it be expensive? The majority cost of a natural gas plant is fuel cost, and if the demand for natural gas declines the cost of it will decline until there's a new equilibrium. Under your scenario natural gas would decline in price to whatever the cheapest source is, so long as it is sufficient to fill demand.