From: Robert J. Bradbury (bradbury@www.aeiveos.com)
Date: Wed Feb 02 2000 - 18:25:20 MST
On Wed, 2 Feb 2000, Stirling Westrup wrote:
>
> The ultimate problem with solar power is that there isn't enough of it.
> I've seen projections in which the planet's power needs by 2050 will be
> such that a band of 100% efficient solar cells 10 miles wide and wrapping
> the whole equator will not give enough power. So, even if we converted
> all highways and all roofing materials to 98% efficient solar cells,
> we're *still* going to have to go out and find other fuel sources.
Well, my comment would be -- *never* believe everything you read.
As I said at Extro3 -- "Trust but verify"...
Draging up an Excel sheet (I just love Excel...),
Earth's Circumference: 6378 km x 10 mi, gives about 2*10^9 km^2
Assume perhaps 30% of it is getting sunlight at a rate of about
750 w/m^2 that is being harvested at an efficiency of about 20%
(after all if we are enginering for planetary power we can afford
the "good" solar cells...). That gives you ~10^10 W (J/sec).
Assuming maybe 8 hours of light per day gives you an annual
production of ~1000 EJ.
According to "Energy: A Guidebook", in 1974, our annual energy
consumption was 370 EJ (year round constant drain of 12 TW).
So even with some pretty pessimistic assumptions we should be
able to harvest 2-3x the amount of energy we are currently consuming.
So, I think there are some interesting built-in assumptions
in your claim, like:
(a) You can't put power collectors in the ocean...
Yet here in Washington State we have several "floating"
bridges...
(b) That power consumption will continue to grow at past
historic rates... But in the developed countries per
capita power consumption seems to have leveled off...
(c) That there will be no improvements in energy efficiencies
(automobiles, electric lights, etc.) [LEDs for example
are 50%+ efficient, new electrodeless sodium vapor lamps
are 2-3x as efficient as incandescent.]
(d) The solar cells are the really crappy 8% efficient versions
rather than the 30+% efficient versions that we can build...
(e) That we don't use other more efficient technologies such
as heliostats and heat pumps to harvest the solar power...
But we are talking 50 years here. Everyone always assumes the same
thing when discussing these limits -- the population grows, energy
consumption grows and technology stands still. Makes me want to gag.
And 10 miles wide is nothing, we've got agriculture production over
much of the land area of the Northern hemisphere hundreds to thousands
of miles wide! I've previously discussed how terribly poor our
agricultural efficiencies are. If we can convert sunlight into
carbohydrate at a *real* 1% efficiency, we'll so much agricultural
land available that we probably have 1-2 orders of magnitude more
power generation potential than the population would require.
> (Note that I'm assuming we don't get around to huge solar collectors in
> space. THAT way we could get enough power for the planet.)
Agreed. Though the power delivery area requirements remain a concern.
>
> Don't count on the oil running dry anytime soon. If you read "The State
> of Humanity" (which I can't seem to find on my bookshelf, so I can't
> tell you the authors' name...) you'll see that there are estimates that
> we have over 200 years worth of oil, even taking into account our
> projected growth of energy use, and that by THAT time we should have
> more reserves available than we do now, due to increases in extraction
> processes.
I would agree if we limit our oil use to "energy" supply uses.
The Jan/Feb 2000 issue of "Foreign Affair" has a interesting article:
"The Shocks of a World of Cheap Oil" by Amy Myers Jaffe and Robert
A. Manning. They make some interesting counter-arguments to the doom
and gloom scenarios painted by the Club of Rome and people like
geologist Colin Campbell.
However, if we get to the point of building interesting diamondoid
things out of oil, I can see the carbon supplies going rapidly.
But by then it will be a very different economy, so it really
won't matter.
Robert
This archive was generated by hypermail 2.1.5 : Fri Nov 01 2002 - 15:26:35 MST