Climate Conversation Group

The Economist carried an online article on 29 June ridiculing the North Carolina legislature for wanting to ban climate-change projections in coastal planning and allow only historical data. (h/t Barry Brill)

It is odd that the always-conservative Economist seems to distrust the temperate, time-tested use of observations to predict likely bounds of future tides. Never mind the success enjoyed by generations of engineers in building our coastal and riverine assets, the Economist now prefers unproven computer models deliberately distorted by the theory that we’re controlling the climate with tiny quantities of an inoffensive gas.

Computers can’t do more than about a fortnight of the weather before accuracy turns to chaos, but the Economist apparently trusts an impressive 90 years of forecast climate.

Am I the lone dullard who doesn’t believe them? More precisely, does anyone else doubt the theory that levels of carbon dioxide (specifically, our tiny emissions of it) dominate the temperature of the oceans, and therefore of the lower atmosphere?

The NC lawmakers were responding to fallout from a 2010 report from a “science panel” selected by the Coastal Resource Commission, who clearly believe that CO2 raises the temperature of the ocean, thus raising the sea level. They wanted to rebuild the coastal infrastructure.

John Droz, jr, physicist & environmental advocate, from Morehead City, North Carolina, has been in the forefront of the fight against the unreasonable predictions of sea-level rise (SLR) forced by the theory of dangerous anthropogenic global warming (DAGW). He describes the report and his response:

In their 2010 report the panel concluded that NC should expect a 39 inch [1 metre] SLR by 2100. Their case was built around a 2007 paper by Stefan Rahmstorf, and was not encumbered by a single reference to a perspective different from Rahmstorf’s. Shortly after the report was released, state agencies started making the rounds of NC coastal communities, putting them on notice that they would need to make BIG changes (elevating roads and bridges, rezoning property, changing flood maps for insurance purposes, etc.).

As an independent scientist, I was solicited by my coastal county to provide a scientific perspective on this report. Even though I wasn’t a SLR expert, I could clearly see that this document was a classic case of confirmation bias, as it violated several scientific standards. But to get into the technical specifics I solicited the inputs of about 40 international SLR experts (oceanographers, etc.).

I compiled and edited their responses to the CRC panel’s report into what I called a critique. This 33-page document discussed how real science works, and then went through the 16-page CRC document, essentially line by line. In doing so numerous specious claims, unsupported assumptions and questionable models were pointed out. It wasn’t pretty.

There’s been an interesting comment on the Economist article from Tom Moriarty, saying the report that kicked off all the agitation was based on sea level predictions put forward by Stefan Rahmstorf and casting strong doubt on his conclusions. Moriarty says:

The state commission’s scientific findings were not as “scientific” as you might think. They were based, in large part, on the various works of Stefan Rahmstorf. In particular, his 2007 Science article “A Semi-Empirical Approach to Projecting Future Sea-Level Rise” and his 2009 PNAS paper “Global sea level linked to global temperature.”

Both of these papers were highly flawed. In some respects they were laughably flawed.

In essence, Rahmstorf tried to build models that take historical sea level data and historical temperature data as inputs to find fit parameters for their equations. Once those fit parameters were found, various hypothetical temperature scenarios for the 21st century could be inserted into the equations to make sea level rise projections for the 21st century. The form of the initial models were highly suspect to start with, and the resulting fit parameters yielded bizarre results.

The 2009 version of his model said that the sea level rise rate was the result of the combination of two things: the temperature above some equilibrium temperature, and the rate at which the temperature was changing. Sounds plausible enough, doesn’t it? The higher the temperature, the higher the sea level rise rate, and the faster the temperature is rising, the higher the sea level rise rate.

But they were surprised to find that their model yielded a counter intuitive result: the faster the temperature was rising, the lower the sea level rise rate. In the parlance of Rahmstorf’s model, the parameter that related the rate at which the temperature was rising to the rate of sea level rise was “b.” Here is what Rahmstorf’s coauthor on the 2009 paper said about “b” in an online forum…

“I downloaded Stefan’s [Rahmstorf's] script, modified it, did the first computations with the same real tide gauge and temperature data Stefan had used — surprise: negative b. Hmmm, strange. That was for real data from the real Earth.”

So, your intuition was the same as theirs, you thought that as the temperature rise rate increased the sea level rise rate would increase. You would think the opposite of that is “strange” and so did they. What is a good scientist to do? Well, they made up a “just so” story about why “b” is negative.

Now, their model is in the form of a differential equation. And it also turns out that that differential equation predicts some very unlikely behavior if “b” is negative. For example, you are able to build various hypothetical temperature scenarios where faster-rising temperatures yield lower sea level rise rates. If you pick your scenario the right way, you can make the temperature sky-high, and make the sea levels unchanging or even drop.

Another incredibly serious problem with the 2009 model was that Rahmstorf used outdated sea level data for the input. They used sea level data from Church and White (GRL, 2006), even though Church and White had made serious recalculations in that data. When the corrected Church and White data (PSMSL, 2009) is inserted into Rahmstorf’s model, his sea level rise projections for the 21st century are drastically cut down. I know this to be true because I personally ran the model with the updated data.

There are always those who will try to end the conversation by implying the conclusions made in peer-reviewed literature cannot be questioned. If so, then please chew on this: six months AFTER Rahmstorf’s 2007 model was published in the journal Science, I had several cordial emails with him discussing his results. He offered to send me the code used to make his projections of “future sea-level rise” and told me “you are the first outside person to test this code.” That is, it had only been reviewed by people “inside” his group. “Pal review” vs. “peer review” is a serious problem in a field that lately has had such a large impact on public policy.

I have much to say on this topic. You can see dozens of postings concerning Rahmstorf’s models, which were so foundational to the NC coastal commission’s findings.

I am a working scientist with over fifteen years of experience in renewable energy. You can see my posts on Rahmstorf’s models at climatesanity.wordpress.com. See the right sidebar for links to indexes of posts about several of his recent papers.

Best Regards,
Tom Moriarty