So how much wind energy can we realistically use? A paper
(Earth System Dynamics,
DOI: 10.5194/esd-2-1-2011)
by Miller,Gans and Kleidon (MGK10)
has generated a lot of controversy courtesy of a New Scientist
review
(See
Climate Progress for a
good rebuttal). They argue that the effective limit we can safely extract without (bad) climatic consequences
is around 18--68 TW (we currently use 0.2 TW of wind power).
This paper is a follow-on to Kleidons work on the thermodynamics of the climate, where he
points out that the
free energy is the important measure of what energy can be generated by renewables.
The free energy of the atmosphere is on the scale of hundreds of TW, he argues, much less than the heat energy of ~10^5 TW that comes from
the sun. He argues that not enough attention is paid to the free energy budget, and argues that from
MEP theory that free energy flows through different geophysical processes are more
important than people currently think: such energy and momentum flows then lead to faster depletion of gradients
than people currently use in their models. Frequently we use "diffusive-like" approximations to stuff we can't explicitly
handle,such as convection and eddies below the resolution of our computer models.
That the free energy budget, and small-scale processes need more detailed study (such as the transfer of momentum and
energy from wind to water), I agree. But paradoxically, MGK10's estimates based on diffusive processes are what gets
it into trouble.
The critiques (e.g.
Jacobsen and Archer)
seem to concentrate on the estimates for diffusive and other energy losses (eg. wake fields) and climatic consequences
are very loose, and out by factors of 50-100. They also ignore the fact that using wind as a substitute for more
destructive energy sources produces a net gain, climate-wise.
Based on these numbers, the real limits on wind power are more
limited by where we can realistically place turbines, and the use and availability of rare earth metals
in turbine magnets (MW sized turbines use over a ton of neodynium each in magnets!)
One of the focuses of research at
ICHEC and
Met Eireann is wind for wind power: this becomes increasingly important
for Ireland as we have more energy production by wind. I've been recently involved in a
European project
Weather Intelligence for Renewable Energies on investigating
weather and climate forecasting needs of wind energy.
In our weather models, we are getting to finer and finer resolutions
(down to 2-5 km operationally, but working on 1 km in research), but as yet do not take account of wind energy changes.
Modelling the wake turbulence, etc. and other effects of turbines is a 'fluid dynamics' problem done by the wind turbine
manufacturers, mostly, in order to lay out turbines in a farm efficiently; this is done on the timescale of metres and seconds,
while on the other end of the spectrum work is done in climate research on the changes in potential winds to see
if wind energy will change in a few decades. But little is done on 'weather' scales of kilometres, minutes and hours: it turns out
there are wind patterns that we don't predict in our models (vortices, etc.; especially for offshore turbines) that affect wind energy production,
but would be very useful for wind farm and electrical grid operators to know about. So we should think about
adapting our weather models for these.
Either way, it looks like more detailed work on the interactions between atmospheric dynamics and wind farms would be a good thing.
Incidentally, whatever about the alarmist nature of the article, kudos to New Scientist for their linking to papers: they quoted
the names of the journals (Earth System Dynamics,Philosophical Transactions of the Royal Society),
but where the article was not yet published (or behind a paywall),
linked to the Arxiv.org version. More could do this.
Tags climate, energy, thermodynamics, science, wind, renewables
Peter Cox on model complexity
An EOF analysis of annual Irish rainfall (first principal component). We see increased rainfall, with a shift
point around 1978.