Tuesday, March 4, 2014

Hurricane Control?

Reader Bayesian Bouffant raised the question of whether we might be able to take some of the strength out of hurricanes, if not to really control them -- by way of an article Massive offshore turbine arrays would help us harness hurricanes.  The article is more positive about chances for weather modification than I, but let's take a look at some details and issues.

By way of background, hurricanes are heat engines.  The heat source is partly warm oceans, but mostly the latent heat released when water vapor condenses in the atmosphere.  The source of that water vapor is, again, warm oceans -- but by way of evaporating water vapor from the ocean surface in to the atmosphere.  When the water condenses, it releases heat, which drives the circulation, and makes for clouds.  The rising air draws more air in, again along the ocean surface. 

Over the years, there have been many suggestions for trying to weaken hurricanes.  One of the least reasonable is to throw a nuclear bomb at the hurricane.  Somehow this is supposed to 'disrupt' the hurricane.  But, hurricanes are heat engines, and nuclear bombs supply heat.  They also don't really have much energy compared to a hurricane!  In one day, a hurricane releases about 52*10^18 Joules (see Christopher Landsea's estimation of the wattage of a hurricane).  The Hiroshima bomb released about 67*10^12 Joules.  Roundly speaking, 1 day of a hurricane is a million Hiroshima bombs.

Since frontal assault is pretty well doomed by the fact that hurricanes are vastly, overwhelmingly, more energetic than anything humans have to wield, any attempt to control, or affect in any meaningful way, has relied on more indirect means.
That brings us to the article, and some related methods.  The idea here is to look back at the 'heat engine' aspect, and try to throw some sand in the gears.  It won't stop the engine entirely, but we can hope it will slow it down.

One of the more direct ideas is to throw oil or other chemicals on the ocean to prevent water from evaporating in the first place.  This runs in to the challenge of throwing enough chemical on the ocean to cover several tens of thousands of square km (or miles), and to keep it there even though the hurricane is kicking up major waves.

The idea in this article is a little different.  Namely, to slow down that breeze which is flowing in towards the eye.  The slower that inflow, the less evaporation to fuel the heat engine.  So goes the idea.

That's the crux.  Is this really manageable?  Or, at least does it pass some Fermi Estimates for plausibility?

While I prefer to work with energies, it is power -- the rate of energy release, that is more typical for looking at wind turbines.  5 MW looks to be a fair Fermi Estimate number.  Landsea's estimate for the power generation by hurricanes (from the latent heat release) is 600 million megawatts.  There would be thousands of turbines, and let's be aggressive and make it 100,000 turbines.  This entire array (let's also assume that all the turbines survive the onslaught of a hurricane) would suck out 1 part in 1200 of the hurricane's total energy.  This goes up a lot if we ignore the evaporative energy and go to the energy of motion (kinetic energy) -- about a factor of 400 (Landsea's note again).  If the turbines are perfect, this could be a substantial fraction of the storm's kinetic energy (about 1/3rd).  Perfect seldom happens in reality.

On the other hand, by the time the entire hurricane is over the entire array, it's already a massive system, which conservatism suggests is too late to really do much with.

But let's look now at the aspect I was foreshadowing earlier -- slowing down the inflow to the hurricane's center.  The tremendous energy of a hurricane is mostly tied up in evaporation and condensation.  Most of the rest is tied up with the energy of the winds swirling in near-circles around the center of the hurricane (the eye).  The portion of the wind flow that's edging towards the eye, rather than swirling around it, is much smaller.  Winds are really in a spiral rather than a circle.  The angle of that spiral is the key.  My round Fermi Estimate value being that the winds going around are something like 8 times faster than the winds flowing inwards.

In this type of geometry, with this enormous array of turbines, and assuming that turbines didn't break when hit by hurricanes, this makes the turbine array able to suck out about as much energy as the inward flow of the hurricane represents.  So it passes this plausibility test.

But ... well, how well can we model hurricanes and, especially, their intensity?  The research mentions running hurricane simulations to support the conclusion of the turbines being able to really extract enough of the inflow to weaken the hurricane.

It's here that I think we hit the biggest problem.  Namely, models for hurricane intensity are not particularly good.  In any case of trying to control the weather, or climate, we need to be extremely confident about the quality of our models.  My student in her doctoral work examined one of the standard models and found that even doubling or halving the rate of evaporation could improve the quality of intensity forecasts.  If you're still at the point where a factor of two in your representations can make improvements in results, you don't have a high quality model.

So I'll count myself skeptical that it can really work.  But this means that evidence can persuade me, even though I lean towards it being wrong.

1 comment:

Anonymous said...

There's also the problem of engineering the turbines to withstand hurricane-strength winds.

Whilst I don't have the information to make a calculation on that, some factors would be the tensile strength of the rotors and associated material, servos that can deal with the loads, and generators that can deal with a wide range of power outputs (and the waste heat that goes with them).

I'm skeptical that such wind turbines can be efficient and economic to operate over such a wide range of wind speeds.

I also wonder if the variation in wind direction from hurricanes could lead to problems keeping the turbines reliably facing into the wind.