Running on renewables: How sure can we be about the future?

energy
Credit: CC0 Public Domain

A variety of models predict the role renewables will play in 2050, but some may be over-optimistic, and should be used with caution, say researchers.

The proportion of UK energy supplied by renewable energies is increasing every year; in 2017 wind, solar, biomass and hydroelectricity produced as much energy as was needed to the whole of Britain in 1958.

However, how much the proportion will rise by 2050 is an area of great debate. Now, researchers at Imperial College London have urged caution when basing future energy decisions on over-optimistic models that predict that the entire system could be run on renewables by the middle of this century.

Mathematical models are used to provide future estimates by taking into account factors such as the development and adoption of new technologies to predict how much of our energy demand can be met by certain energy mixes in 2050.

These models can then be used to produce 'pathways' that should ensure these targets are met - such as through identifying policies that support certain types of technologies.

However the models are only as good as the data and underlying physics they are based on, and some might not always reflect 'real-world' challenges. For example, some models do not consider power transmission, energy storage, or system operability requirements.

Now, in a paper published today in the journal Joule, Imperial researchers have shown that studies that predict whole systems can run on near-100% by 2050 may be flawed as they do not sufficiently account for reliability of the supply.

Using data for the UK, the team tested a for 100% power generation using only wind, water and solar (WWS) power by 2050. They found that the lack of firm and dispatchable 'backup' energy systems - such as nuclear or power plants equipped with carbon capture systems - means the power supply would fail often enough that the system would be deemed inoperable.

The team found that even if they added a small amount of backup nuclear and biomass energy, creating a 77% WWS system, around 9% of the annual UK demand could remain unmet, leading to considerable power outages and economic damage.

Lead author Clara Heuberger, from the Centre for Environmental Policy at Imperial, said: "Mathematical models that neglect operability issues can mislead decision makers and the public, potentially delaying the actual transition to a low carbon economy. Research that proposes 'optimal' pathways for renewables must be upfront about their limitations if policymakers are to make truly informed decisions."

Co-author Dr Niall Mac Dowell, from the Centre for Environmental Policy at Imperial, said: "A speedy transition to a decarbonised system is vital if the ambitions of the 2015 Paris Agreement are to be realised.

"However, the focus should be on maximising the rate of decarbonisation, rather than the deployment of a particular technology, or focusing exclusively on renewable power. Nuclear, sustainable bioenergy, low-carbon hydrogen, and carbon capture and storage are vital elements of a portfolio of technologies that can deliver this low carbon future in an economically viable and reliable manner.

"Finally, these system transitions must be socially viable. If a specific scenario relies on a combination of hypothetical and potentially socially challenging adaptation measures, in addition to disruptive technology breakthroughs, this begins to feel like wishful thinking."


Explore further

Renewable electricity by the numbers gets thumbs-up in new study

More information: Clara Franziska Heuberger et al, Real-World Challenges with a Rapid Transition to 100% Renewable Power Systems, Joule (2018). DOI: 10.1016/j.joule.2018.02.002
Provided by Imperial College London
Citation: Running on renewables: How sure can we be about the future? (2018, March 6) retrieved 10 December 2018 from https://techxplore.com/news/2018-03-renewables-future.html
This document is subject to copyright. Apart from any fair dealing for the purpose of private study or research, no part may be reproduced without the written permission. The content is provided for information purposes only.
67 shares

Feedback to editors

User comments

Mar 06, 2018
By the year 2050 we will know (a very accurate estimate) over a billion times technologically what we know now.
How can these people predict what life will be like then?
This study by a well known scientist has been verified over and over by other scientists that were skeptical and in disbelief and wanted to prove him wrong. But they all came out with approximately the same result.
http://theemergin...ment.htm

Mar 06, 2018
It is nice to see a frank and objective assessment. Better decisions can be made with such a mindset. The troll patrol should read this paper and maybe there could be a little more tolerance of differing opinions.

An excellent example of sorting out political non-sense and real world challenges is be at https://www.manha...565.html click on the Download PDF.

Thorium and fusion reactors should be on the table of common sense.

Mar 06, 2018
Interesting study - and as we move forward - we will have more and more real world systems to check the data against. South Australia could be at 80 renewables in less than a decade. http://renewecono...1-21201/

(this is just one example). Interesting times!!!

Mar 06, 2018
How much of the energy can be from wind or solar depends on the cost. If solar is at today's cost, it is only competitive (not counting subsidies) in high-sun regions and if storage losses and costs are excluded.

However if the cost of solar continues to fall, at half the current cost it becomes the lowest cost option in every major market when the sun is shining, and competes against peaking power with enough storage for a daily cycle. At about 1/4 the current cost one can overbuild enough to cover seasonal variation, and to rarely need more than a few days of storage. And if 1/10 the current cost could be achieved then you could overbuild enough to have enough even with a month of cloudy English days. This study only modeled overbuilding 1.8 times as much as today's British grid.

1/10 the cost may be hard for solar to achieve, but 1/4 the cost is fairly likely in less than 20 years, and 1/2 the cost is highly probable within a decade.

Mar 07, 2018
"However if the cost of solar continues to fall, at half the current cost it becomes the lowest cost option in every major market when the sun is shining, and competes against peaking power with enough storage for a daily cycle. At about 1/4 the current cost one can overbuild enough to cover seasonal variation"


That's all neglecting the cost and loss of the storage. Seasonal storage is ludicurously expensive as you're talking about TeraWatt-hours of energy.

Furthermore, the production and upkeep costs of solar goes up as the production infrastructure is de-carbonized. You forget that the same energy has to go back into making the stuff in the first place, and if the cost of energy goes up due to storage requirements then the cost of production goes up.

Mar 07, 2018
For example, consider the EROEI of solar PV panels is around 10:1 but when you close the loop from solar energy back to new solar panels, it drops closer to 1:1

That's because you -also- have to maintain all the batteries, chemicals and mineral refineries, mining and recycling activities, maintenance and cleaning etc. to keep the system up.

EROEI is another way of saying how much of our total social and economic activity needs to be spent on simply producing the energy we need, and an EROEI of 1:1 means we have no energy left over to do anything else - like eat.

3:1 means social collapse. An EROEI of 6:1 is considered energy poverty, as in, you're spending disproportionate amounts of your effort simply to keep the lights on, and solar energy as is - without considering the other factors - is already perilously close. That puts a limit on where solar energy is viable, as moving to higher latitudes the EROEI naturally drops with the available sunlight.


Mar 07, 2018
Currently all that activity that supports the solar PV systems is coming from fossil fuels, which in general operate around 20:1 EROEI if my memory serves right.

Wind turbines on the other hand are cheap, with EROEI from 20:1 to 50:1 before closing the loop. The big unkown is how much energy is lost in closing the production loop and making it self-sustaining.

Currently a loss of 80% is likely, and with advancing technology 50% is within the realms of possibility. 50% off of solar energy puts us into energy poverty, though with a drop in price to 1/4 would put us back to 10:1 EROEI, but that's still pretty bad because it means doubling of energy prices from today.

I'm equating price with energy because money = economic activity = energy.

Mar 07, 2018
This is also an important concept:

https://en.wikipe...nibalism

"Energy cannibalism refers to an effect where rapid growth of a specific energy producing industry creates a need for energy that uses (or cannibalizes) the energy of existing power plants. Thus during rapid growth the industry as a whole produces no new energy because it is used to fuel the embodied energy of future power plants.

For the environmental impact of solar power, the energy payback time of a power generating system is the time required to generate as much energy as was consumed during production of the system. In 2000 the energy payback time of PV systems was estimated as 8 to 11 years[3] and in 2006 this was estimated to be 1.5 to 3.5 years for crystalline silicon PV systems"


So, if you're doubling your solar PV arsenal every couple years, you're not actually generating any net energy while doing so. This puts a limit on how fast it is sensible to grow.

Mar 07, 2018
Numbers like EROI are almost impossible to nail down. Of course solar panels on the equator, are going to have a totally different EROI that those placed far north, or south. Just making numbers up is not helpful Eika - there is no way you can justify saying that EROI on panels that 'close the loop' is 1. That is just nuts. EROI for newer panels is improving - as their efficiency increases. A recent meta study put the number more like 14:1. https://www.scien...16306906

fuels such as tar sands and oil shale (Lambert et al., 2012) deliver a lower EROI, having a mean EROI of 4:1 (n of 4 from 4 publications) and 7:1
From - https://www.scien...13003856

Despite this low EROI - we are still producing plenty of oil this way - right?

Looks to me like wind and solar fair pretty well when put up against tar sands and shale oil - but we are still producing those....

Mar 08, 2018
"Just making numbers up is not helpful Eika - there is no way you can justify saying that EROI on panels that 'close the loop' is 1. That is just nuts."


Of course I'm not saying that categorically - exactly for the reason you stated: solar panels at different locations have different energy payback times. That should be obvious.

"Numbers like EROI are almost impossible to nail down."


There's plenty of studies that place the EROEI of solar panels -on average- around 10:1 or 14:1 since they've apparently improved. That tells you something important - that solar panels around New York or Germany are probably closer to 1:1 than say 20:1, while solar panels in Dubai are the opposite.

"Despite this low EROI - we are still producing plenty of oil this way - right?"


Yeah, because oil/petroleum is used for other purposes than energy. It's more valuable as a chemical feedstock than just burning it all. Think of blacktop roads for example.

Mar 08, 2018
Think about the process of using solar PV to capture CO2 and make Hydrogen, in order to synthesize bitumen in a chemical reactor to make the binder for the road that the maintenance crew uses to clean the dust off of the solar panels.

That process has an EROEI below 1 because it's just consuming energy, yet this cost is not counted against the EROEI of solar panels, for one part because it would be very difficult to measure, and for the second part because it's simply not done.

Currently, that bitumen is a byproduct of petroleum processing and avoids the cost of substitutes, which is what makes it economically viable to utilize rather poor resources like tar sands - because the other sources are worse.

So until we invent the hovering utility truck, solar power needs petroleum at least in this manner, and in many more ways.

Mar 08, 2018
https://en.wikipe...il_sands

"Naturally occurring crude bitumen impregnated in sedimentary rock is the prime feed stock for petroleum production from "oil sands", currently under development in Alberta, Canada. Canada has most of the world's supply of natural bitumen, covering 140,000 square kilometres[14] (an area larger than England), giving it the second-largest proven oil reserves in the world. The Athabasca oil sands are the largest bitumen deposit in Canada and the only one accessible to surface mining, although recent technological breakthroughs have resulted in deeper deposits becoming producible by in situ methods. Because of oil price increases after 2003, producing bitumen became highly profitable, but as a result of the decline after 2014 it became uneconomic to build new plants again."


70% of bitumen goes to road construction, and most of the rest goes to roofing tiles and weatherproof paints. It's got little to do with the EROI.

Mar 08, 2018
>> That's all neglecting the cost and loss of the storage.
No it isn't – I said "at half the current cost it ... competes against peaking power with enough storage for a daily cycle."
Yes, even daily storage is expensive, but peaking power comes from inefficient plants and is very expensive compared to baseload power. (I hope that it is clear that the storage cost is assigned to the solar, not the peaker plant).

>> Seasonal storage is ludicurously expensive as you're talking about TeraWatt-hours of energy.
I agree, which is why I said "At about 1/4 the current cost one can overbuild enough to cover seasonal variation, and to rarely need more than a few days of storage. "

>> ... if the cost of energy goes up due to storage requirements then the cost of production goes up.
That would apply to IF I had discussed fully decarbonizing at today's solar cost; however I discussed exactly the cases where the cost of energy does NOT go up.

Mar 08, 2018
Neglecting wind, hydro and CSP reducing the need to store PV:
PV's nominal EROI has improved from 6.8 in 2010 (Hall) to 11 in 2015 (NREL) so Greenonions' 14 today sounds about right.
A 2x overbuild to not need season storage would cut this in half, and storage of half the electricity at a 50% round-trip efficiency and a battery energy cost equal to the solar would reduce it another 3X.
Decarbonizing TODAY would thus bring the EROI to ~~14/6 or ~2.3, which would indeed be a disaster.

But with energy a major cost, a 4x cost reduction will have a roughly 4X EROI improvement, so when solar is 1/4 today's cost, the EROI including storage will be somewhere around 4*2.3 = ~~9.
In contrast oil's EROI has fallen from ~100 in the 1920s to 18 today (all producing wells) to 8 for new oil discoveries.

While the exact numbers are arguable, the trend is clear: the EROI of solar is improving, while the EROI of oil is declining (natural gas is also about 10).

Mar 08, 2018
Eikka
Yeah, because oil/petroleum is used for other purposes than energy
Wow - how to miss the point. Shale oil is one of the biggest sources of oil - and is of course being turned into fuel. Or are you trying to say that about 50% of u.s. crude oil production goes into making ashpalt? https://www.eia.g...&t=6

Oil has a comparable eroi to both solar and wind - so your machinations don't make any sense.

Mar 08, 2018
I guess this is why Elon Musk, Ray Kurzweil, Steven Hawkings, ABC (Google) etc are so much smarter than us. I pointed out to all of you to a study done by our smartest people(see first post above) and you dismissed it out of hand because you know better than our smartest people. :-)

Mar 08, 2018
>> I pointed out to all of you to a study done by our smartest people... and you dismissed it out of hand ...

Why do you think that any of us dismissed it? I've been saying similar things for decades - in 1985 someone ask when we would reach what is now called 'the singularity', and my guesstimate was 70 years, or 2055, we'd have 10 billion people roughly as interconnected as the neurons of our brain are, so 'humanity' would be roughly as far beyond a person as a brain is beyond a neuron.

But if we all just sat around waiting for that to happen, it wouldn't happen -
we have problems to solve to get there, including energy.

And as much as Eikka and Greeonions argue with each other, they both are doing good. Eikka warns against unrealistic expectations that are counterproductive, and Greenonions pushed to make sure that we work toward a bright future that we would like living in.

Mar 09, 2018
Why do you think that any of us dismissed it? I've been saying similar things for decades


Did you bother to see how the study was done?
Some went back 2000 years to the time of CHRIST. So divide that 10 year projection by /2000 year sampling rate to see the error rate for ten years.
That is a .005% error rate for 10 years. That is 512 times what we know today. A 50 year projection gives us 50/2000 or a .025 error rate with a projection of over 1,000,000,000,000,000 times what we know today. Now that study saw vary little change in acceleration during world wars or plague etc. I don't much want to believe it ether but I am a big believer in facts, whether I like them or not.

Mar 09, 2018
Realscience "And as much as Eikka and Greeonions argue with each other, they both are doing good" I agree with you to a point Realscience. The problem for me is the issue of truth. I feel that we have a huge problem in the U.S. right now (can't really speak for other places) regarding the issue of truth. In science - of course truth matters. Facts matter. Yet we seem besieged by folks who don't care about facts. It is such an upside down world right now. Eikka seems to have no hesitation in making up facts - and then flooding the site with opinions - based on made up rubbish. It is fine to be skeptical - but you don't get to make up your own facts. That is not helpful.

Mar 09, 2018
>>> Eikka seems to have no hesitation in making up facts

Eikka does not make up many facts. Eikka is usually very accurate on where the technology is today (or often, where it was a few years ago). Eikka does err on the pessimistic side (in this case the only recent study I know that shows PV EROI of ~1 used decade-old data pessimistically). But given how many over-hyped technologies die when they meet reality, Eikka is very often right.

What Eikka underestimates is how REALLY GOOD humans collectively are at working around flaws... Just look at Moore's law - 40 years ago it was clear to pessimists that the limit of optical lithography was around 1 micron, and that there were only 3 generations left. Yet I have watched generation after generation, and we now make chips with features 100 times finer than a micron, still with optical lithography!

So you keep pushing for a bright future, while Eikka identifies problems that need to be overcome to get there.

Mar 09, 2018
>> Did you bother to see how the study was done? Some went back 2000 years to the time of CHRIST. So divide that 10 year projection by /2000 year sampling rate to see the error rate for ten years.

I've seen the same basic study taken back to the evolution of DNA a few billion years ago, through eukaryotes with much bigger genomes, through the growth of brain size, through cites and writing, to the printing press, and into the modern world, and yes, it is quite an amazing trend. The doubling time has actually gotten shorter over the eons!

Please don't assume that because people don't comment they either missed your comment, ignored it, or disagreed. Some may, but others are to busy to comment, already know it, or concentrate on places where they can add value, etc.

Also, 1000000000 times the information might be 10000000 times the knowledge, 100000 times the understanding, and only 1000 times the wisdom. (But it will be awesome in any case!)

Mar 10, 2018
I really think two things are going on first the studies projections, no mater how accurate, are just plane unbelievable! Second I think that kind of growth is unfathomable to humans as we are now.
You touch on wisdom. Can you define wisdom? Are we wiser than an ant? If you think so then wisdom is probably dependent on intelligence (barring chemical imbalance and other brain abnormalities). With that kind of technological growth we will soon be enhancing our intelligence by genetics, nootropics etc by a hundred fold or more. Will we be wiser? If wisdom is based on intelligence(ant versus human) then yes much wiser.

Mar 10, 2018
Realscience "Eikka does not make up many facts" And that is not OK with me. Facts do matter. I work in a high school in Oklahoma. The teachers are besides themselves. The kids can't write a paper. Science teacher was recently telling me that cheating is off the charts. Home work come back - and many of the kids have the exact same wrong answer. In other words - they just copied from each other. Same thing happens in tests - cuz the kids text each other answers across the room. I said 'ban cell phones during tests' - but that would be too much trouble. So how do we progress as a species - if making up facts is OK - and we don't value the hard work of learning? Shit - we are studying 2nd World War in history. The teacher said "don't you guys think this is fascinating?' Not one kid agreed.

Mar 10, 2018
>> Eikka does not make up many facts

I should have capitalized the word NOT. I do not know of any case where Eikka has 'made up' a fact (but I cannot say that s/he never does since I have no way of knowing that).

I have seen Eikka selectively quote articles that contain 'facts' that agree with him/her, but then essentially everyone tends to believe things agree with them - I work very hard to avoid confirmation bias, and still often catch myself.

From what I have seen, Eikka fights for the truth as s/he sees it. There is far too much unrealistic hype, and Eikka is very good at seeing through it. But Eikka has gotten so used to cutting through hype that s/he overlooks that the few nuggets that aren't hype reinforce each other, and result in dramatic progress being made over time in spite of the vast bulk being hype.

I am a collector of those nuggets, so I share Eikka's distaste for hype even when I disagree with Eikka's conclusions.

Mar 10, 2018
@rderkis - wisdom depends on intelligence, but it is not a simple relationship; for example, kids are generally as smart as their parents, but are usually nowhere near as wise (although there are exceptions both ways).

My point is that a dramatic change in one aspect can result in a moderate change in another. For example, my computer account 50 years ago had 1 KB of disk space and a time-share of 16 KB of core. I now have on my desk a billion times the disk and full use of a million times the RAM operating 30,000 times faster. While it is certainly a huge improvement, it is more like 100 times as useful, or possibly 1000 times as useful, and not a billion times as useful.

Yes, unless humanity seriously screws up, the future will be amazing.
(But the present is already seriously cool!).

Mar 10, 2018
kids are generally as smart as their parents, but are usually nowhere near as wise (although there are exceptions both ways).


So you feel wisdom is based on intelligence and age(experience)?
Once again with the kind of technological growth we will be seeing in the near future death will be optional. So wisdom will be increasing with age. Since the average age now is around 80 years, that means in 160 years people will be considerably wiser. Plus with that kind of technology memories(experience) will be able to be shared across the whole human race.

Mar 10, 2018
Realscience "I should have capitalized the word NOT" I understood what you said. I don't think it is OK to make up any facts. Saying that EROI on solar panels is 1 - is just an example. Another example is the time EIKKA claimed that commercial electricity in Europe could be bought for around 2 cents a Kwh. I pushed multiple times for a source - and finally got one that actually talked about thermal energy costing 2 cents - not electricity. We were comparing the cost of power for wind turbines - so it was very significant. You are correct - we can all fall prey to confirmation bias. I am not bothered at all by over hype. I read so many articles here - and think "let's just wait and see if that ever really pans out." Many times it does not seem to. Arrogance and negativism are not the same as confirmation bias. Neither is making shit up - just to feel superior.

Mar 11, 2018
>> Saying that EROI on solar panels is 1 - is just an example.

I don't think that Eikka made that up. There was a study a year or so ago that pegged the all-in EROI at 0.85, and it was written authoritatively enough to convince quite a few people who are not in the field. It got some publicity, and I'm betting that that is where Eikka got that data from.

(But the study analyzed the energy used based on a 350-micron wafer and a 300 micron kerf (I might have those reversed), which haven't been used for many years, and modeled decade-old efficiencies, too, and then double-counted energy losses from degradation, and modeled output in cloudy Switzerland, etc.)

Mixing up thermal and electrical energy costs is closer to making things up than merely believing a biased study - I must have missed that exchange. And from what I've seen it would be an exception - Eikka often underestimates the future, but s/he is typically accurate about today.

Mar 11, 2018
I don't think that Eikka made that up
Well - given that current data pegs EROI on solar at somewhere between 8 and 15 - and I linked to a meta study that covered all of those possibilities - it at least makes EIKKA dishonest in citing information that is so obviously out of date. Facts really do matter.

Mar 11, 2018
Let's look at Eikka's initial EROI comment:
>> For example, consider the EROEI of solar PV panels is around 10:1 but when you close the loop from solar energy back to new solar panels, it drops closer to 1:1

>> That's because you -also- have to maintain all the batteries, chemicals and mineral refineries, mining and recycling activities, maintenance and cleaning etc. to keep the system up.

So Eikka did NOT say that PV for immediate use has an EROI of 1. He gave a round number of 10, which is very close to NREL's number from just a few years ago.

Aluminum smelters and glass plants run 24/7 with at most ~20% variation, so if you were to run them from solar you would need storage, and storage both adds energy cost and reduces net output.

If you take into account that most such plants are not in great solar locations, Eikka's comment was actually accurate.

- continued -

Mar 11, 2018
However while Eikka is fairly accurate in pointing out that going to pure solar today doesn't work, it is not all that relevant to the FUTURE. Solar cost (largely energy cost) is dropping rapidly, and efficiency is climbing. Battery cost is also now dropping rapidly, and once solar becomes cheap, one can tolerate less-efficient (even cheaper) batteries.

For immediate use as electricity PV's EROI is ALREADY better than new oil or gas wells (the EROI of the fossil fuel would be hit with the conversion penalty to electricity).

WE are about even in EROI if the oil is kept as fuel.

Solar is still way behind, but catching up rapidly, if few-day storage is needed.
That crossover happens before 2030 at today's trends.

The final barriers are seasonal fluctuation and portable fuels - that crossover is at ~1/10 today's cost (before 2050).

But that's PV/battery on its own - throw in CSP, wind, hydro, etc. and it can happen quite a bit faster!

Mar 12, 2018
"If you take into account that most such plants are not in great solar locations, Eikka's comment was actually accurate." Why don't either of you provide some support for that assertion??? I understood exactly what Eikka was saying which is why I responded "there is no way you can justify saying that EROI on panels that 'close the loop' is 1" And of course - crickets chirping when asked for any support for such conjecture - without any validation.

Mar 12, 2018
@greenonions:
Please understand that I am defending the technical accuracy of a comment whose implication I strongly disagree with.
Eikka's comment SOUNDS LIKE "a solar panel's EROI is only 1", but that is NOT what it says.

First, Eikka did not say that it is 1. Eikka said that it drops from "around 10:1" to "closer to 1:1".
In a geometric ratio anything below sqrt(10):1 is technically closer, to the boundary is 3:1 - anything 3:1 or below, and Eikka's comment, regardless of its relevance, is technically correct.

Second: Eikka said "when you close the loop from solar energy back to new solar panels".
The energy to produce a silicon PV panel is in the glass, steel, aluminum, and refined silicon, all of which are produced by processes run 24/7 to avoid having to cool the equipment down, so these require energy 24/7.

To 'close the loop' one therefore has to store the solar energy and provide it 24/7.

- continued -

Mar 12, 2018
That means that Eikka gets to count all of the cost of providing PV energy 24/7.

That includes having to overbuild to have enough solar in the winter and cloudy weather. And aluminum and silicon purification (the most energy intensive) are in particular placed in places where hydro-electricity is cheap (Pacific Northwest, Norway, Quebec, Iceland), and these places have very little winter sun. Many of these places have 1/4 as much sun n the winter as in the summer (I live in one of them), and aren't all that great in the summer, so Eikka can cut your 14:1 down to 3.5:1 or lower right here.

Now we still have the round-trip efficiency of the battery and the energy embodied in making the battery. I commented earlier that "storage of half the electricity at a 50% round-trip efficiency and a battery energy cost equal to the solar would reduce it another 3X", and Eikka gets to use those hydro places with short winter days, and that's down WELL below 3:1.

- continued -

Mar 12, 2018
So Eikka is technically correct that if you tried to close the loop on solar today it would be a giant fail. And Eikka didn't even count having to convert solar to oils for the transportation industry!

But the BIG PICTURE is that Eikka's point is of little relevance.

Even just considering today, most energy is NOT used to make solar panels, so most solar panels will be in much better locations. And we have wind for diversity, and no one is proposing to close all hydro dams, so the seasonal and climate penalties are both greatly reduced from the pure PV case

But it is the future that really makes a difference: the energy required to produce PV panels has dropped ~4X in the past 15 years, and the panels produce more energy as well.

The trend has been steady for decades so Eikka would have the burden of proof if s/he challenged the continued decline.

And a similar trend applies to the energy required to make batteries.

- continued -


Mar 12, 2018
No one can know the future, but I have done enough work in solar to know what the trends are, and I see no real challenges to solar hitting 1/4 of the current cost/energy.

A ~10X reduction from today is needed to run the world entirely on PV/baatteries, but after ~4x it gets trickier to reduce the cost (although 10x might be possible by 2050).

But 10x may not be necessary - many 'small' contributions such as wind, hydro, adaptive loads (there are already trials of varying the work of smelters +/- 20% on a daily cycle) can probably bridge the gap, and sun-to-fuels would solve the whole storage issue.

Now I have to get back to working on that next four-to-one improvement in PV's EROI (and I say that seriously), so please accept that the problem with Eikka's comment was not that it was wrong in what it actually SAID but that it was not very relevant.

Mar 12, 2018
Quote RealScience "No one can know the future,"

That's a crazy statement. I KNOW with 98% certainty that I am going to put a period at the end of this sentence.
It's just a mater of probability and how far in the future you trying to forecast.
And the technique you are using to forecast it with. If your 10 years old and just guessing, the probability of being correct is much less then in your case where your using your own experience. But the best way SCIENTIFICALLY is to do a study across hundreds or even thousands of years.
And I showed you the studies done by our greatest research/scientists, which you chose to ignore, in favor of your few years experience.

And then you call your self RealScience. :-)

Mar 12, 2018
@rderkis - Scientist are funny - 98% certainty is not knowing', but just a high-confidence prediction. For example, a new particle isn't even considered discovered until the odds of being wrong are below 1 / 3.5 million!

It should be clear from the context of the that you quoted that I am not against predictions and forecasts, such as being pretty confident of a 4X reduction in the cost of solar.

However I cannot KNOW that it will happen - I cannot even KNOW that if I click the next key on my keyboard, the next letter will appear on my screen (I actually have a buggy keyboard connection through USB, so I can assure you that intending to put a period at the end is not the same as knowing that there will be one).

Even with a good keyboard, an asteroid could hit, the phone could ring and you could forget to complete the sentence, your internet connection could dropped, you could hit the comma rather than the period, etc.

-continued-

Mar 12, 2018
As for:
>> And I showed you the studies done by our greatest research/scientists, which you chose to ignore, in favor of your few years experience.

I already told you that I did not 'ignore' your comment. I already explained to you that I had come to the same conclusion myself decades ago, and even that date that I had predicted and the reasoning behind my prediction. I described research that I had read research on the theme that had traced the curve back to the advent of life billions of years ago. So how come you ignored what I said? :-)

Mar 12, 2018
@rderkis - Scientist are funny - 98% certainty is not knowing', but just a high-confidence prediction. For example, a new particle isn't even considered discovered until the odds of being wrong are below 1 / 3.5 million!

It should be clear from the context of the that you quoted that I am not against predictions and forecasts, such as being pretty confident of a 4X reduction in the cost of solar.

However I cannot KNOW that it will happen - I cannot even KNOW that if I click the next key on my keyboard, the next letter will appear on my screen (I actually have a buggy keyboard connection through USB, so I can assure you that intending to put a period at the end is not the same as knowing that there will be one).

Even with a good keyboard, an asteroid could hit, the phone could ring and you could forget to complete the sentence, your internet connection could dropped, you could hit the comma rather than the period, etc

which is why I said 98% certainty not 99%.

Mar 12, 2018
I already explained to you that I had come to the same conclusion myself decades ago, and even that date that I had predicted and the reasoning behind my prediction.

So RealScience you feel reasoning trumps both experience and facts?
RealScience, take a wild guess at how many times or current level of technology (not time frame) needs to improve before our use of solar would be as good as it can get (by today's standards) on earth. 5X 10X or perhaps 100X give me a guess.

Mar 13, 2018
Rderkis - You make too many assumptions - my reasoning was based on my own experience in computing (over two decades using them and almost a decade of designing them already by that point), the trend of Moore's law, and the 'facts' of how the brain was thought to work.

As for solar, the bulk of the improvement can be made without ANY further technology, but just properly applying that which we already have, and probably 90% of the potential gain can be made with the right 10X our current information, as long as it is the right 10X.

However progress is not equal - biotechnology is current advancing at TWICE the pace that that web site you linked has for technology, while computing hardware has fallen to half the pace, and the technical progress in solar even than that (although the economic progress has been great, most of that recently has been from new applications of older technologies).

-continued-

Mar 13, 2018
As for time frames, I think of progress in terms of time frames because it is currently more accurate: the correlations between capability and elapsed time are better understood than the the correlations between quantity of information and capability.

Please reread my comment:
"I've seen the same basic study taken back to the evolution of DNA a few billion years ago, through eukaryotes with much bigger genomes, through the growth of brain size, through cites and writing, to the printing press, and into the modern world, and yes, it is quite an amazing trend. The doubling time has actually gotten shorter over the eons!"

What part of my agreeing with you is it that you do not understand?
It's just not NEW to me anymore, so I no longer rave to people about it unless they ask...

Mar 13, 2018
RealScience Quote - The doubling time has actually gotten shorter over the eons!"


This is the part I dont think is really sinking in for all of us. It is not just doubling it is a exponential growth of technology. Which is what you kind of just said. We are at what is called the knee of the exponential curve of technological growth. Behind us well kind of seem linear but ahead of us will seem almost straight up. Everything we have experienced thus far will not prepare us for the vary near future, That is why many people still think they can forcast 30 years into the future based on their last 30 to 50 years.

Mar 13, 2018
Exactly.

We see evidence of many exponential improvements over the eons.
We don't yet see back to the improvement rate of the first self-replicating molecules, but the dramatic increases in quantity plus major increase in regulatory complexity to eurkaryotic cells, and the dramatic increases to large organisms with complex body plans, each took on the rough order of a billion years.
The dramatic improvements from stone flakes to late stone age tools took millions of years, while the metal age led to things as complex as the anitkythera mechanism in thousands of years, and the industrial revolution (inc. the printing press) led to dramatic improvements in hundreds of years. With electronics we're seeing similar leaps in tens of years, and we can see that as we enhance intelligence itself that will drop to years. Does it then go to months, days, hours, and then "AHA!" spreading lightspeed? My estimate is still 2050 to 2060 for progress to become incomprehensible.

Mar 13, 2018
Quote - we can see that as we enhance intelligence itself that will drop to years. --------- 2050 to 2060 for progress to become incomprehensible. End Quote


The incomprensibleability(spelling :-) of it will depend on our newest level of intelligence.
There are many meanings to the word singularity. I myself think that happens when a intelligence greater than our present intelligence emerges on earth. Whether that intelligence is AI, Aliens, our ourselfs makes no difference to the meaning of singularity.
As far as enhancing our own intelligence, I think it will come soon. We are right on the edge of creating a universal quantum computer and when that happens we will be able to use simulations that will show us how to increase our intelligence. Plus solve the energy problems here on earth. Whether that be solar, fusion, etc

Mar 13, 2018
While a joke now, perhaps a quantum computer simulation will even show us how to create a cold fusion generator. Obviously not how we are trying to do it now.

Mar 13, 2018
>> The incomprensibleability(spelling :-) of it will depend on our newest level of intelligence.
Yes, I was referring to incomprehensible to us having this discussion.
My guess is that it will find its intelligence more comprehensible than we currently find our intelligence.

>> I myself think that happens when a intelligence greater than our present intelligence emerges on earth.
By many measures we are collectively already more intelligent (and certainly have more information) than humanity used to, so to me it has to be a much bigger difference than that - for example, as far more intelligent than a human as human is than a bee.

>> Whether that intelligence is AI, Aliens, our ourselfs makes no difference to the meaning...

Agreed. My guess (and hope) is a hybrid of enhancing what we are today and AI (like the way our brain still contains a core of older brain structures), with the electronic layer also enabling rich connections among billions of individuals.

Mar 13, 2018
>> As far as enhancing our own intelligence, I think it will come soon. We are right on the edge of creating a universal quantum computer and when that happens we will be able to use simulations that will show us how to increase our intelligence. Plus solve the energy problems here on earth. Whether that be solar, fusion, etc

It is already starting to happen. But our brains are more wonderful than usually given credit for, so it will take a little longer than most singularity enthusiasts expect. Not because the speed of change is wrong, but because the goal is farther.

But at the rate of growth, even a lot farther than people expect only adds another decade. Only a colossal disaster in the very near future can delay it more than a few decades.

Mar 13, 2018
RealScience why are you assuming the hybrid if it happens will be based on electrons? Perhaps we will be well beyond electrons and that tech will be obsolete(to slow, to unstable and to easily hacked).

Mar 14, 2018
>> RealScience why are you assuming the hybrid if it happens will be based on electrons?

You are correct - that was careless phrasing on my part. I started with "the technology layer" and then realized that that could include our brain cells if those are enhanced, so I hastily edited it within the 3-minute window.

I do not expect the long distance communications part to be electronic - we're already using photons for that.

As for the enhanced processing layer, there are many choices: electrons, photons, ions, molecular binding and 'other'. Our brains use ions, our computers use electrons, our genes use molecular binding and our communications networks use mostly photons, so an argument could be made for any, but not in 1000 characters or the time I can take now. Let's continue this in a couple of months when I should have more time.

Mar 14, 2018
>> RealScience why are you assuming the hybrid if it happens will be based on electrons?

You are correct - that was careless phrasing on my part. I started with "the technology layer" and then realized that that could include our brain cells if those are enhanced, so I hastily edited it within the 3-minute window.

I do not expect the long distance communications part to be electronic - we're already using photons for that.

As for the enhanced processing layer, there are many choices: electrons, photons, ions, molecular binding and 'other'. Our brains use ions, our computers use electrons, our genes use molecular binding and our communications networks use mostly photons, so an argument could be made for any, but not in 1000 characters or the time I can take now. Let's continue this in a couple of months when I should have more time.

Stephen Hawking died! :-(

Please sign in to add a comment. Registration is free, and takes less than a minute. Read more