A Lot of Hot Air?

Yesterday, I wrote a long post on "my energy plan." It wasn't so much a plan as a set of semi-baked ideas that sit well with me in terms of addressing our energy needs. While I've been sort of dancing in this direction for a while, I haven't really sat down and written out my thoughts- and the organization that takes place while writing is important. But in focusing on primarily traditional forms of energy, I neglected some of the alternatives in my post. I also forgot to embed the shorter Pickens video I mentioned:

In short, Pickens proposes installing windmills across the Midwest and Great Plains states in large enough numbers that we can shut down or retool the 22% of the nation's electricity generated by natural gas. That natural gas could then power about 38% of the country's vehicle fleet, and reduce our imports of petroleum by nearly a half- in today's prices, about $300 billion a year.

Off the top of my head, I have two reactions: first, Great! Let's go! Second is the more realistic realization that we're not really talking energy independence here; we would still need to import about 32%, a bit less than a third, of our oil. However, strategically, that means that we would be much less dependent on OPEC. OPEC will do just fine without us, but putting another major energy stream into competition with them would lower their profits somewhat. Furthermore, OPEC is truly holding us hostage right now; if they chose, they could cut us off. If you think $4.00 per gallon gasoline is bad, try to picture $15.00 on that pump (with 9/10 of a cent at the end, of course). Much of the funding for Muslim terrorist groups comes out of oil profits. Severely reducing our consumption of Middle East oil is unlikely to affect that funding substantively (again, there are lots of customers waiting in line to buy OPEC oil), but I for one consider it asinine to purchase a weapon for your enemy, regardless of the fact he can easily find one elsewhere. (This should not be taken to imply I consider the Middle East, Muslims, or OPEC to be enemies- just the terrorists, and the groups that funnel money to them.)

There are lots of questions. First, it should be pointed out from the start that Pickens has invested heavily in gas; I have also read (though I want to verify it a little more thoroughly) that he has invested in wind turbine manufacturing. So no matter the ultimate impact, following his recommendations is going to net him a great deal of money. This should raise a red flag for anyone in our culture. If someone is recommending you spend a bunch of money that the person is going to profit from, you should definitely check and double check the claims and reasoning behind those recommendations. That's not to say you should simply dismiss them, but you should check them out.

So do we have enough dependable domestic supply of natural gas (henceforth in this post ng.)? First it should be pointed out that he's not proposing that the supply per year should increase, but that our current supply feeding electricity generation be diverted to transportation. For those who use ng for heat, hot water, cooking, etc. this is good news. If he was proposing adding to ng demand, price would inevitably rise. From what I hear, ng is already pretty spendy during the heating season for those with cold winters. Furthermore, he seems to be suggesting that ng as vehicle fuel is sort of a stopgap, an interim solution. He suggests hydrogen or electricity as closer to the desirable endpoint than ng. The question remains, do we have enough to make this a reasonable proposal? Well, as always it depends on what numbers you want to pay attention to. NaturalGas.org provides estimates from three sources ranging from 1191 trillion cubic feet (Tcf) to 1779 Tcf. Averaging the three estimates gives me about 1430 Tcf. I had to look around for some conversion factors to find the energy equivalent of a barrel of oil to natural gas, and found a good site here. Divide BTU's per barrel oil by BTU's per cubic foot ng and you get 5653 cf ng equals one barrel of oil (There are likely some efficiency differences, but this is a good ballpark estimate). The previous page with the link to the conversion factors provides the figure of roughly 21 billion barrels in proved oil reserves. Dividing 1430 Tcf ng by 5653 cf per barrel yields an estimated oil equivalent of 253 billion barrels- about 12 times the amount of oil we have.

However, there's a caveat: in the example above I was comparing estimated ng to proved oil. Looking at a further EIA page, we find the proved reserves are only 211 Tcf (as of Dec. 2006), less than 15% of the estimated total, or only one and a half times the estimated oil equivalent. Oooh, what a letdown! But wait! There's more!

I have to admit, I'm having fun with this ride. Let's call it a wonkercoaster. My point is that it's really good idea to pay attention to the fine print.

If you look at this still further pdf (5.4 Mb), and read the first column of the first page, you'll see that our proved reserves went up 3% in 2006, and that we "proved" 36% more than we produced. In other words, we're still finding ng at a greater rate than we're using it. Crucially, the opposite is true for oil; if you look at this pdf on page 4, you'll see that during the period of 2005-06, total proved reserves fell by 785 million barrels- which takes into account new discoveries and "proofs" over that period. This left a total of just under 21 billion proved barrels. If we were to continue the loss of proved reserves at that rate, we would run out of oil in about 26.7 years from Dec. 2006, or about August 2033. In the meantime gas is being discovered faster than we produce it.

In short, if you want to object to Pickens' suggestion on the basis that natural gas is in short supply, the numbers don't back you up well. This was quite surprising to me; as I've mentioned before, the energy crunch of the late '70's made a lasting impression on me. Ng was stretched, and for a while, if I remember correctly, there was a moratorium on hooking up new customers. I had not realized that the estimates for reserves were that high.

The next issue is the capitol investment involved in building enough wind turbines to cover 22% of our electricity production. Our annual national electricity production is 4.1 billion megawatthours (Mwh), according to the EIA. 22% of that is 902 million megawatt hours, or 902 billion kilowatthours. The California Energy Commission provides a rough estimate that "Large-scale wind farms can be installed for about $1,000/kW." If we assume that these turbines generate at a constant nominal rate (this is not a valid assumption, but it eases off-the-cuff calculations, and helps adjust for the fact that Pickens is specifically targeting the most productive wind areas), then we multiply 902*10^9Kwh/year * 1 year/365 days * 1 day/24 hours * $1000/Kw. Cancel units, we're left with $, good. The arithmetic reduces to 902* 10^9*$1000/365 * 24. Fire up the calculator and we get $103 billion. This sounds like an awful lot, yes? But keep in mind that's roughly 10 months of what we spend in Iraq. And keep in mind the estimated savings by displacing imports of foreign oil are $300 billion per year. In other words, savings from reduced oil imports would cover the capitol investment costs in 4 months.

Now that last sentence is a little ingenuous. There are a number of other costs I haven't accounted for, including, off the top of my head:

• Infrastructure for distributing the electricity


• Infrastructure and stations for redistributing ng to vehicles


• Maintenance of turbines


• Refitting the vehicle fleet to run on ng (Rawley pointed out to me that refitting fuel injection engines would be more difficult/expensive than engines with carburators; I'll take his word for it; my knowledge of cars sputters out after muffler bearings and that little propeller thingie up front)


• Land purchase or leases

These are not minor costs, though I have been assuming that the cost of purchasing and erecting the turbines would be more expensive than any other single cost. But let's pretend that's a very bad assumption and that these additional costs bring the total to ten times the installation cost. We would still cover the capitol investment through reduced imports in less than three and a half years!

And let's look at the flip side: reasons that costs might be lower than the estimate above:

• With a serious and prolonged push to manufacture wind turbines, economies of scale would kick in, and technical development would speed up, reducing up front costs, increasing efficiency and dependability, and perhaps reducing maintenance costs.


• Much needed manufacturing, construction and maintenance jobs would be created in large numbers. This would increase incomes in some areas of the country that need additional income badly, and increase tax revenues.


• Manufacturing infrastructure would provide export opportunities. Technological advance could make US products highly desirable elsewhere. At the present time, from what I've heard, most of our wind turbines are imported.


• There are many other places where wind power would be cost-efficient. They may not be as widespread over regions as the Midwest/Great Plains are, but I can tell you that the west coast tends to be pretty windy, as do many western mountain passes. Utilizing such areas could provide still more positive feedback to decreasing wind generation costs, and at the very least a continuing market following the completion of the Pickens plan.

This list is speculative, and I don't have the expertise to quantify or estimate how much these factors might reduce costs. But I would be surprised if some of these didn't kick in.

The final issue I want to address is the footprint of such a project. Looking at GE's PDF on their 2.5 Mw turbine, the rotor diameter is 100 meters. This means the spacing between individual towers must be a minimum of 100 meters in all directions. Rather then calculating the area for hexagonal packing I'll just treat the packing as square (again, it makes for easy math). This means each tower would require a hectare, about 2.5 acres. Basically repeating the calculations from above, we get 902*10^9Kwh/year * 1 year/365 days * 1 day/24 hours * 2.5 acres/2500 Kw. Running through the calculations, I get 257 thousand acres. 640 acres is a square mile, so that is equivalent to just over 400 square miles, or as one massive block, a square about 20 miles on a side. Quite doable, considering the size of the area under consideration. This land would not be limited to only wind generation: the footprint of the towers themselves is small compared to the diameter of the rotors. Farmers could lease out their lands and still raise crops or livestock on most of the area; I can't imagine they'd forgo 10% of it. My guesstimate is closer to 5% (I couldn't find figures on the tower footprint).

The one environmental consideration I'm aware of is that the turbines are known to kill birds indiscriminately. I don't know facts and figures on this, but it is an issue that should not be ignored, especially along major migratory flyways. Given the amount of area in the wind maps Pickens uses in the video, those areas should be avoidable. Nevertheless, we should try to find ways to reduce this risk.

My impression is that this is a plan that's doable. It has limitations: what doesn't? It's a small piece of the overall energy puzzle, and a promising step in a promising direction. We're caught between a rock and a hard place: declining petroleum on one hand, and deep concern over climate change on the other. The easy answer would be to turn to coal, but until we develop some confidence in carbon sequestration, we shouldn't simply turn to the easiest answer. As I noted in yesterday's post, I don't think there's going to be one or two magic bullets that solve our energy problems. I expect whatever solution we come to will be made of many little pieces all linked together to create a cleaner, safer, more sustainable picture.