Thursday, January 31, 2008

Will we be drowned?

In "An Inconvenient Truth", Al Gore – Nobel laureate, Oscar winner, and erstwhile Vice President – made a very scary prediction: the world-wide sea level would rise a full 20 feet by the end of this century due to the effects of anthropogenic global warming in melting the polar ice caps. I don't believe this for a minute. There have been several published debunkings of this notion, and a ruling by a judge in England that required "government guidance notes" to accompany the showing of the film in British secondary schools (the Town Crank blog post here has the details).

But I'm always on the lookout for more ammunition to fire at this preposterous sort of scare talk. A fellow by the name of Jerome Schmitt made the effort to determine just how much heat would be required to melt enough ice at the poles to cause such a sea level rise. His calculations can be found at The American Thinker.

He determined:
  • How much ice would have to be melted (I won't give the figure here because of what follows)
  • How much air there is in the atmosphere: approx. 5 x 1018 kilograms
  • How much heat from the sun would have to be trapped by global warming to heat that mass of air 5° Celsius – the amount predicted by many global warming scenarios: approx. 2.5 x 1019 kiloJoules
  • How much heat would be needed to melt enough ice to raise the sea level by 20 feet: approx. 7.4 x 1021 kiloJoules
All of the numbers and calculations can be found in the American Thinker article. Mr. Schmitt concludes, using simple physics and nothing more complicated than multiplication and division, that the amount of heat that will raise the temperature of the atmosphere an average of another 5° Celsius in the coming century would only be 1/300th the amount necessary to melt enough ice to cause the catastrophic sea level rise that Mr. Gore has promised us.

That is, 7.4 x 1021 kiloJoules is almost 300 times greater than 2.5 x 1019 kiloJoules. Which means that the amount of heat needed to raise the temperature of the entire earth's atmosphere an average of another 5° Celsius is woefully inadequate to melt enough ice to cause the biggest flood since Noah.

If you read the article, however, you may be struck – as I was – by two statements at the end. First, the author hedges his bets:
Even if I am wrong by an order of magnitude, there is still an enormous difference.
Also there's an asterisked note to that 300 times figure:
*Editor's note: a transposed decimal point led to an incorrect multiple used here when this article was first published. The energy required is nevertheless hundreds of times greater than evidently assumed by Al Gore.
When I read the article the day it was published (January 22nd) it had already been hastily edited because of some faulty math. Apparently, judging by the language in the note ("nevertheless hundreds of times greater"), the figure was very likely 3,000 in the original article. Therefore, as is my wont, I cast a jaundiced eye over all the figures. I've found far too many math errors in publications of every type over the years to trust a long set of calculations, even if they lead to a conclusion with which I'm entirely in agreement.

I ran into trouble right in the middle of the article. Mr. Schmitt calculates that the surface area of the earth covered by water is about 360 trillion (3.6 x 1014) square meters and that a 20-foot rise in sea level is about the same as 6 meters. So far, so good. Therefore the "Volume of water necessary to raise sea level 20 feet" would simply be 6 meters times that 360 trillion figure, or about 2.2 quadrillion cubic meters. However, this is what was in the article I read on the morning of the 22nd of January:
Volume of water necessary to raise sea-level 20-feet: approx. 6 x 1024 cubic meters
or 6 trillion trillion (that is, 6 septillion) cubic meters – 2.7 billion times too much. 6 septillion cubic meters is about 5500 times the volume of our entire planet! About 7½ times the volume of the planet Saturn.

He followed this with the "Volume of ice that needs to melt to raise the sea level 20 feet", and that number is the nearly correct figure of 22 quadrillion cubic meters. I say "nearly correct" since an extra digit got added to the correct figure of 2.2 quadrillion. In other words, that figure is off by a factor of 10 times. Not nearly as bad as being off by 2.7 billion times, but it ain't chump change.

I wrote to the editor of The American Thinker detailing my findings. About an hour-and-a-half later, the editor e-mailed me back asking whether I was referring to the correction that had already been made in the article. I wrote back saying, no, I read the version of the article that already contained the asterisked comment and correction. Therefore two additional errors remained after the first hum-dinger was taken care of.

I visited The American Thinker site a little later and saw that the 6 septillion error had, indeed, been altered ... but it still wasn't correct. The number had been changed to 22 quadrillion, the same number as the existing incorrect figure that was off by a factor of 10. Thus there were two numbers cheek-by-jowl that were both off by that same factor.

So I wrote another friendly missive (my third) to the AT editor telling him that everything was almost copacetic. Those two 22 quadrillion cubic meter figures just needed to be divided by 10 and they'd be set. The editor wrote to say that he'd run it by the author, Mr. Schmitt first.

A day later the two numbers hadn't yet been corrected. So I decided to go over the numbers once again and check the sources for the numbers. I then wrote a fourth e-mail to the AT editor:
Dear Sir,

I decided to look up all the values used in Mr. Schmitt's article, and then follow the calculations all the way to the end. The only remaining error is the repeated error of the volume of water/ice necessary to raise the seal level by 20 feet. The number should be 2.2 quadrillion cubic meters, not 22 quadrillion as in the article.

What this means is that Mr. Schmitt is prophetic at the end of his article where he says:
There is a difference of 300 between these two figures. Even if I am wrong by an order of magnitude, there is still an enormous difference.

I say "prophetic" due to his final comparison of the "difference" between the heat needed to raise the temperature of the atmosphere 5 degrees Celsius (~25 quintillion kiloJoules) and the heat necessary to melt ice to achieve 20-foot seal level rise (~7.4 sextillion kiloJoules). Rather than "difference", he should have said "ratio". The ratio between the two is, indeed, about 300-to-1. However, because of the mistake in the volume of ice needed to raise the sea level by 20 feet, the ratio should be 30-to-1.

Usually an error of a full order of magnitude would be, shall we say, disquieting. The error remains.

As I said, that was my fourth message to the American Thinker editor. I then found the e-mail address of the author who happens to be the principal of NanoEngineering Corporation. I recapped the errors I'd found, a total of four errors. I gently admonished that if one is trying to debunk the absurd claims of others using numbers to back up one's debunking, it's best to make sure they debunk accurately. I couldn't imagine what the American Thinker would say if they had to change the final big important number a second time, taking it down from 3,000 to 300 to 30.

Mr Schmitt courteously replied that he was going to check the numbers one more time and would get back to me. I also heard one final time from the editor saying that they were double-checking, too. That was four days ago.

Today I heard from Mr. Schmitt in a very engaging e-mail:
Dear Mr. Erbach:

You are right. Here's what happened.

In a fit of pique over listening to a Global Warming alarmist announcing on PBS that Miami-Dade county will be under water in 20 years unless the right Democrat is elected president, I dashed off the calculation and article. I asked that it be proofread. The Editor was very happy with the essay and assumed I had completed the calculation correctly. He persuaded me to publish it immediately. To my horror and embarrassment, we were then alerted to a major mistake I had made in transposing data from a website. I probably would have caught this myself if I had waited a day and proofread it myself. This required a very hurried edit the day of publication, complicated by the Editor's lack of internet access that day. Hurried phone calls and recalculations resulted in the text as it is, still containing the error you identified.

For this reason, I am reluctant to ask for another re-edit. Will you indulge me in this, particularly since the furor of the first mistake has died down? As it is, I think I have already torpedoed my credibility with the Editor, although he acknowledges some responsibility for pushing for immediate posting.

Thanks for your consideration.


-Jerry Schmitt

I can't say as either Mr. Schmitt or American Thinker came off looking too well out of all this. Having to make a second correction to the final figure would, apparently, wake up the sleeping dogs rather than letting them lie.

I haven't asked for Mr. Schmitt's permission to publish his e-mail, so I guess I'm letting myself in for recriminations of some sort. But I thought that laying out the situation to the American Thinker people as he did to me would demonstrate a decent respect for the truth. What's the worst that could happen? AT would pull the article completely. That's it. Perhaps publish a small notice saying that there were too many problems in the article. Mr. Schmitt would have to be embarrassed for a while, but it might cause him to not dash off in a fit of pique the next time.

As far as my own exposure in this affair, I rather doubt that Mr. Schmitt or the editor of The American Thinker will ever see this blog posting. You're reading the world's most active least-read blog. I think I'm safe. But, hey! I like the numbers to jibe, OK?