Monday, October 03, 2016

Chart of the Year.

Joe Romm was reading the 2016 update of US DOE's Revolution . . .  Now: The Future Arrives for Five Clean Energy Technologies  when he exclaimed: Chart of the Year


The must-read report reveals the game-changing progress core clean energy technologies have made over the last several years — specifically, solar, wind, LED lights, batteries, and electric cars. Accelerated deployment driven by smart government policies, both domestically and around the world, have created economies of scale and brought technologies down the learning curve faster than almost anyone expected.

The next time an opponent of climate action questions the cost-effectiveness and scalability of climate solutions — or the value of government clean energy policies — the top chart and indeed the whole report should be front and center in your response.

This was picked up all across the tweet zone, Of course, beauty is in the eye of the beholder and some, well some (actually all) of the luckwarmers have no rhythm,  Our good news is their embarrassment, and up stepped Andy Revkin to tweet

Eli, Eli OTOH is a strange cross between a chemist, a chemical physicist, a spectroscopist, a kineticist and, oh well a blogger, so what he saw is something that was better plotted elsewise.

Taking electrons in hand the bunny wrote to the drawer of the graph and person most responsible for the report, Paul Donohoo-Vallett at DOE HQ down the block for the data and Dr. Donohoo Vallett replied post haste

At a minimum the LED curve suggested a first order (exponential) fall off in cost, and to be honest the others looked sort of the same to Eli so he replotted it all as a semi-log plot and fit the decreases to straight lines

Not perfect but a lot better than force fitting to a Tolian (Tolian's being functions that go exactly where Richard Tol wants them to) with a lot less effort.  A bunny can fuss about this brute forcing, but not Eli.  The most interesting part of this is converting the slopes to half lives for each of the technologies.
Wind Utility Scale PV Residential PV  LEDs EV Batteries
Half Life/Yrs 7.68 4.50 5.68 1.70 3.21
r^2 0.94 0.98 0.98 0.98 0.97











Not too shabby and LEDs are beating the current version of Moore's law. 

33 comments:

Fernando Leanme said...

We the same curve in the performance of children's football teams, etc. The problem arises when the curve starts to flatten out and the team isn't competitive. In this case renewables stay uncompetitive due to intermittency, which really degrades performance.

EliRabett said...

Intermittency is the last refuge of the fossil fuel fans.

Arthur said...

Wow, that is a real display of ignorance from Revkin there. "Flattening" indeed. I guess another of those cases where what communicates clearly to scientists is obscure to the nonscientist. But really. Does Revkin expect LED costs to go below zero or something?? When you plot the inverse curve - quantity that can be deployed for a given price - there's clearly no "flattening"...

Phil said...

Note that there are 5 technologies graphed.

LEDs don't have a problem with intermittency.


EV batteries can be charged when there is an excess of power, so reduce the problem of intermittency.

Anonymous said...

Can Andy Revkin really be so innumerate that he doesn't recognize exponential decay?

Jan Galkowski said...

Excellent post and summary, Eli, which I will use, with your permissions and appropriate citations in a brief talk I'm giving on the solar/zero Carbon energy future in Westwood, MA, in November along with some from Stanford's Mark Jacobson and Tony Seba, as well as Bloomberg New Energy Finance.

And, agreed, the appeal to the Intermittency Demon shows a decided lack of imagination. In addition to smart spatial positioning, and great high precision and local wind and sun forecasts, when zero Carbon energy is cheaper to erect at scale than fossil fuel delivery and generation systems cost to merely operate, overbuilding capacity for distributed generation and local consumption is an option which seldom occurs to businesspeople whose solvencies depend upon it being not possible.

-- Jan, http://667-per-cm.net

Florifulgurator said...

Which reminds me of this gem: https://youtu.be/EyPs5ou0DcE
"Is it that hard to introduce the public to the notion of a graph?" -- Murray Gell-Mann to Andy Revkin.

Mike Dombroski said...

Those Energy.gov PDFs are full of hype about new "capacity". When the capacity factors are something like 20%, you can't even call them half truths! They have to be offered with a fifth of something.

Jan Galkowski said...

@Canman, so if Capacity Factor is 20% just build 5x as much as needed. This works because unit costs amortized over 20+ years are near zero, especially for solar PV and land wind turbines. Utility wonks used to fossil fuels have a hard time thinking this way. Not new. IBM execs had a hard time thinking of a world where marginal costs of computing and storage were essentially zero, back in the age of the mainframe, partly because it meant their business model was nearing the end of its useful life.

-- Jan, http://667-per-cm.net

Fernando Leanme said...

So we build 5000 turbines instead of 1000 turbines. And what do we do when the wind stops blowing and 5000 turbines stop generating electricity?

Jan Galkowski said...

@Fernando Leanme, if the 5000 turbines are built over 1000 km, then the meteorological conditions at one part of the array are so statistically decorrelated from one another that this will never happen for the entire array. 1000 km is an upper bound, and in some regions (e.g. and especially offshore New England) it is 100 km.

See http://thesolutionsproject.org/#tsp-section-map

and peer reviewed pubs cited there describing this.


-- Jan, http://667-per-cm.net

JamieB said...

"LEDs don't have a problem with intermittency."

Indeed. In cold climates such as northern Europe, where the highest load on the grid is seen on cold winter evenings, they represent far and away the most reliable resource we have available to our grids. In the UK they will reliably shave somewhere between 5% and 10% off our peak demand when fully deployed.

EliRabett said...

Jan, feel free. - Eli

THE CLIMATE WARS said...

Canman, a couple of fifths of lithium , which costs less by volume than Talisker, are enough to build a Tesla battery big enough to run an air conditioner for a week .

In the long run technology prices tend to converge on a small multiple of materials cost , and when they do lot of fossil fuel aps will fossilize - it's what one soi-disant Scientist for Trump long ago dubbed crative destruction.

Unknown said...

I can tell you are a physical chemist. Ya gotta get a straight line!

Mike Dombroski said...

"In the long run technology prices tend to converge on a small multiple of materials cost "

That may tun out to be true for batteries (or supercapacitors), but wind and solar are known for using huge amounts of materials. That convergence should also apply to nuclear. It looks to me like wind/solar are in conflict with nuclear. Michael Shellenberger has a new TED talk:

https://www.youtube.com/watch?v=TU7kLruNZVU

David B. Benson said...

Despite this, no more wind farms are being built in the Pacific Northwest. These require balancing agents, i.e, backup, and the hydropower resources to do so are at the maximum possible for wind in the so-called Columbia Basin region. Over in Montana there is better wind but all the transmission is otherwise dedicated.

In general, there has to be enough transmission, the price and permitting is maybe becoming exponentially more expensive.

Romm's and Eli's graphs are far too simplistic to describe electricity grid evolution.

David B. Benson said...

Except for LEDs. The effect of that is already noticeable in my electricity rates; the cost is going up by about 6% for the next year or so, until the following rate setting exercise. The consumption of electricity in the utility service area is slightly dropping and the expenses of generating, transmitting and distributing are slightly increasing. Voila!

But as about half of the generation produces carbon dioxide emissions I am not complaining.

THE CLIMATE WARS said...

Sorry Canman, reality no Candu, because reactors require armor and shielding , which PV's and windmills do not.

You are of course at liberty to diss windmill reliability- sailboats seldom last a decade in a gale,
OTOH, I can easily lift ten kilowatts worth of sliced silicon PV stock.

John ONeiil said...

'... reactors require armor and shielding , which PV's and windmills do not.'
Windmills require concrete and steel, as do nuclear reactors.
AP 1000 reactor ( four being built in US, four in China ) takes 93,000 tons of concrete and 58,000 tons of steel. 151,000 divided by 1,117 megawatts output gives 135 tons per megawatt.
Vestas V 164 wind turbine ( used in the Horns Rev offshore farm in Denmark )weighs 1,300 tons, plus 4,000 tons concrete for the base, and makes 8 megawatts - 662 tons per megawatt.
US reactors are averaging over 90 % capacity factors, so an AP 1000 should produce about 8,760 ( hours per year ) x 0.9 x 1,117 = about 8,800,000 Megawatt/hours per year - 17 tons per Terawatt hour.
The very best locations for large wind turbines - parts of the US high plains states, or good offshore sites - might manage 50 % capacity. 8 x 8760 x 0.5 = 35,000 Megawatt/hours. 151 tons per Terawatt hour, about nine times worse than nuclear. Then factor a twenty to thirty year expected life for the wind turbine, versus the reactor being designed for 60 years,
Finally, when the reactor shuts down to refuel - once every eighteen months, scheduled for periods of low demand - the one next to it is still making power. When the wind turbine shuts down for lack of 'fuel ', you can guarantee the one next to it is stationary too, and most likely the one 100 km away as well. 1,000 km ? Do you feel lucky, punk ? Any time you're not, you're burning gas. They can run a car off a battery, but a continent is a whole other story.
As for solar, its capacity factor makes wind look good. You might luck out and find a functioning wind farm within a thousand miles, but when the sun sets on you, it pretty soon sets on everyone else speaking your language. ( Sorry guys, I'm a subject of the Queen.)If you set up a row of solar panels stretching right around the Earth, it's average output would about equal one AP1000. The logistics I leave to you.

Mike Dombroski said...

Running the entire world on sliced silicon PV stock. What a bunch of baloney!

Jan Galkowski said...

@David B Benson: As the Solutions Project and Jacobson's team shows, wind resources cannot be exploited unless there are major upgrades to power transmission lines and controls. This is because, like hydropower, you can't just build wind turbines anywhere. Indeed, that's an advantage to residential solar PV, even if its cost per KWh is high, and, added up across the society, it seems that investment is relatively poor. First, the capital doesn't typically come "from society", and certainly not from ratepayers. Second, the permitting and ancillary requirements are relatively simple.

Also, it is not a question of fossil fuels OR zero Carbon energy using wind. Fields of wind turbines have both some advantages and some disadvantages of utility scale power, including the need to produce concrete which has its own sizeable share of CO2 emissions. The contest is, rather, as suggested by "Romm's and Eli's graphs", that zero Carbon energy, meaning solar PV, or wind, where it is accessible, will plummet in price per KWh. If the local "grid" is incapable of harnessing that, then these generators will be isolated and produce really really cheap power for a locale, probably backed up, as in Minster, OH, by storage. (See a blog post I did on this subject.) That locale will thereby have a huge economic edge over its neighbors. While this process is nowhere near fast enough to convert at large scale to forestall disrupting climate, because it is market based, in the end the decentralized islands will dominate. Also, turbines are mechanical and, so, wear out. PV installations diminish in efficiency over their accounting lifetimes, which are typically 20 years, but good ones are more than 80% as efficient as first installation after 30 years.

In any case, the destination for a region having an ISO which is unenlightened about residential and commercial solar PV is that its utilities will face defection and eventually collapse, as the relatively wealthy members of the grid defect, and all commercial members defect. Companies like IKEA and Apple will help this along, both of which are seeking to sell spare power from their panels, and are looking to team with large storage. This could well be facilitated by schemes like SonnenCommunity.


-- Jan, http://667-per-cm.net

Jan Galkowski said...


As far as @Canman's complaint regarding "Running the entire world on sliced silicon PV stock", every credible resource (several listed below) suggests there are no constraints on materials or manufacturing capacity. Providing build-out of PV electricity is primarily a policy and grid scale problem. As I indicated, that's only a "problem" if the future is imagined to be anything like today. I have also written about that.

References on future build-out of solar PV capacity:
* IEA
* Frauenhofer ISE's study
* Photovoltaic materials abundance limitations literature review
* Materials Availability Expands the Opportunity for Large-Scale Photovoltaics Deployment
* Considerations of resource availability in technology development strategies: The case study of photovoltaics

-- Jan, http://667-per-cm.net

Jan Galkowski said...

@John ONeiil,

Similarly, it's not nuclear versus wind turbines. As I wrote just above, turbines have some of the disadvantages of other utility scale power. I do agree that retaining online many of the present nuclear power installations until zero Carbon energies build out sufficiently is a win. (Then they can be decommissioned.) I don't agree that building new nuclear facilities is a good idea, less from the traditional arguments of long term fuel storage risks, which are appreciable, but from the fact that the nuclear power industry, at least in the United States, has shown no evidence of a business learning curve. This traditionally means that as additional units are constructed, the cost per unit decreases, because of learning how to do it better. U.S. nuclear builds have been progressively more expensive and taken longer and longer, even controlling for regulation.

Then the idea that nuclear power plants only have outages when they refuel is just silly. (Note the 2011-2015 range in background brown in the figure. Attempt to reproduce below.) They often go offline for technical or safety reasons, giving the rest of their local grid little advance notice when they are withdrawing their massive generation capacity, and necessitating the construction of dirty peaking natural gas plants along with them to back up for the eventuality. All generation of electricity is intermittent, and needs to be backed up in some way. In fact, an attractive feature of turbines and distributed solar PV is that units are spatially dispersed, and, so, when they "go down", this withdrawal of power is far less of an event to be managed than the withdrawal of a large generator, like a nuclear plant or even a combined cycle gas plant.

https://drive.google.com/open?id=0B3Nnyie7hrIuTGcyM09CdWY3cU0

https://drive.google.com/open?id=0B3Nnyie7hrIuU3RFV0J1UGRoX0k

THE CLIMATE WARS said...

John, all the panels on your equatorial solar belt would fit into a four mile square, an area comperable to the security perimeter of a reactor and its cooling pond.

Given the modest capex of PV theses days , I expect the shareholders cwould be happy to see the wasted real estate inside the fence covered with PV's ,as auxillary power is nice to have when bad things happen to good reactors.

John ONeiil said...

Jan Galkowski -
Your graph of reactor outages supports my argument, not yours. If you look at daily outages for the peak demand months, July and August, over the last two years they were below six percent, and mostly below four percent. Over the five year period they only briefly spiked once to eleven percent.
For wind, the monthly - not daily - capacity factor over the same months, for the whole country, averaged below thirty percent , and mostly below twenty five percent.
Saying all types of power source are intermittent is just as disingenuous as saying they all cause CO2 emissions. Wind and nuclear both cause about a hundred times less CO2 emissions than coal,per kw/h, but nuclear is about fifteen times more likely to be there when you need it, and that's using an apples to oranges comparison that greatly favours wind.
Distributed solar adds reliability ? When the sun sets in California, it's off for the whole continent, unless you want to power America with Alaskan sunshine. ( Solar thermal in January drops below a five percent capacity factor, and that's in the deserts of the south west.)
www.eia.gov/todayinenergy/detail.php?id=14611

Jan Galkowski said...

@John ONeiil,

If you want to compare capacity factors of nuclear with other sources, clearly they need to be weighted by the amount of power generated for the grid to which they are connected, not by abstract counts of installations.

Facts are that a big chunk of U.S. nuclear installations failed unexpectedly, and sometimes were down for months. Pilgrim Nuclear in Massachusetts is so unreliable, it's owners want to close it. And, on the other side, grid-wide capacity factors for, say, wind ought not to reflect voluntary curtailment when the ISO doesn't want the unit's power. Finally measurements of capacity factors are poorly constrained, because of voluntary reporting and the like. As Delucchi and Jacobson argue, capacity factors as defined simply raise the effective cost per KWh of generation. However, if that is low to begin with, as I've said here before, it machts nicht. Also, grids, like ISO NE, with poorly performing nuclear plants, absolutely require peaking gas plants which, by the way, have the lowest capacity factors of all gas generation (and are dirtier than coal, to boot).

To the intermittency question, with solar there is a backup device using storage, but there is no reason why microgrids relying can't use primarily on solar can't use complementary wind resources as well, resources which do not need to be transmitted long distances. Along many coasts, wind is also anti-correlated with solar generation. And, of course, there are things like the Minster, OH solution I referenced elsewhere. The idea is not to rely upon a distant centralized source for power: Do it locally for your village or town or city. Who cares if, as a whole, this is less efficient by conventional measures. No one is trying to power the entire continent. (We don't even do that today. Most power generation/consumption is limited to ISOs/RTOs. I'm suggesting much smaller units of control.) Those are no longer appropriate for this world. Heck, if a conventional technology is going to rely upon peakers, arguing they are only needed a small part of the time, you can also install natural gas fuel cells at a microgrid, too, and DER.

I can see big generation requirements for cities, but even those are susceptible.

Jan Galkowski said...

And, energy storage begins life in Massachusetts. Whether or not Mrs Hypergeometric and I choose to hoard electrons rather than sharing will depend upon the Infinite Wisdom of the Massachusetts legislature's House, more specifically, it's leadership. The Senate is okay, and the House rank-and-file have their heads and hearts in the right places, but DeLeo, Golden, and Downing give Caesar, Pompeius, and Crassus historical competition.

Hank Roberts said...

"... a four mile square, an area comperable to the security perimeter of a reactor and its cooling pond.

"Given the modest capex of PV theses days , I expect the shareholders cwould be happy to see the wasted real estate inside the fence covered with PV's ,as auxillary power is nice to have when bad things happen to good reactors."

Amen, amen, amen.

A layer of PV, shading the parking lots -- and a field of battery storage sufficient to run the reactor pumps through to cold cooldown in the worst case when the grid goes away (meaning the local system is protected from a big solar flare).

Meanwhile, the surplus electricity gets sold to the grid and keeps all the company's and employees' electric vehicles charged.

Please, Russell, you know how the world works, convince someone who can convince people to lay this out in a way clear enough that shareholders can propose building this and make it happen.

Unknown said...

@David B. Benson

On the West Coast, California is slated to add over 400 megawatts and Oregon, over 250 megawatts
http://ieefa.org/ieefa-update-wind-wires-way/

Oregon is in the PNW

John ONeiil said...

Russell Seitz -
All US reactors already have diesels with enough fuel to power them down to cold shutdown, plus batteries. ( So did Fukushima, but they weren't waterproof.) Since in most cases they will never be used except for testing, their carbon footprint is minimal, and they're a lot easier to secure against disaster than a few acres of PV, which would be useless anyway if the reactor mishap chanced to occur during a cloudy week.
The land around a reactor is often a valuable wildlife refuge, since places most humans are excluded from are rare. In Florida crocodiles and manatees take advantage of the warm coolant water.

Jan Galkowski said...

@John ONeiil, @Russell Seitz,

According to those who've looked at the question carefully, the United States oughtn't be smug about its nuclear power facilities in comparison with Japan. The U.S., after all, still uses pools instead of dry casks. For those with a subscription to Science, this is the full URL of the article. The investigating committee was chaired by Joseph Shepherd, a mechanical engineer at the California Institute of Technology.

A follow-up letter to the editor of Science reported, citing an article in Nuclear News from October 2008, page 83:
Two petitions for Rulemaking Regarding Spent Fuel environment impacts were denied by the Nuclear Regulatory Commission. An August 8 Federal Register notice explained that the petitions filed by the attorneys general of Massachusetts and California (Attorney Moonbeam) presented nearly identical issues and requests for rulemaking concerning the environmental impacts of high-density storage in spent pools. The petitioners asserted that "new and significant information" shows the NRC incorrectly characterized the environmental impacts of high density spent fuel storage as "insignificant" in its generic environmental impact statement (EIS) for renewal of nuclear power plant licenses. Specifically, the petitioners asserted that spent fuel stored in pools is more vulnerable to a zirconium fire than the NRC concluded in its EIS. The NRC said that it denied the petitions because no new or significant information was presented that would invalidate the agency's existing findings on pool storage.

Like I said, not a lot of reason to be smug.

And while manatees might like warm outlet water, no doubt that same water is disruptive to the remainder of the ecosystem which thereby has access to less dissolved oxygen because of the higher temperatures.

John ONeiil said...

I don't have a subscription to Science, but I have looked at the issue of zirconium flammability. The powder burns, but not bulk metal. At temperatures over 700 C in the presence of water, you get an exothermic self-propogating reaction, but those conditions are only possible in an intact pressure vessel - at 700 C, water doesn't usually hang around long. Here's a video of zircalloy cladding getting heated to 2,000 C under a blowtorch.
https://m.youtube.com/watch?v=x__2yWx9zGY
Reactor heating of inlets is a local phenomenon, acidification and greenhouse heating affect the whole ocean. In Florida, there is often a specially constructed circuit of canals for cooling, which the heat-loving critters muscle in on.
Here's an article on what was at the time the largest solar power plant in the US, opened with some hooplah by President Obama and the State governor. The field of solar panels is impressive, but the output is dwarfed by the six natural gas generators tucked in beside them.
https://bravenewclimate.com/2011/05/15/solar-power-in-florida/