Saturday, August 16, 2008

Laugh laugh laugh

Richard Simmons guests on Whose Line is it Anyway?

Monday, August 11, 2008

Sunday, August 10, 2008

Correlation coefficients

Here is an interesting summary of how CO2 concentrations, solar activity and the big oceanic oscillations correlate to recent temperatures. D'Aleo's original paper is linked here.

Seems to me that there are a lot of factors contributing to temperatures. Why single out the one with the lowest r-squared in order to save the planet?

Saturday, August 09, 2008

Confirmation Bias

To a three year old with a hammer, everything looks like a nail. The scientific term for this is "confirmation bias" which Wikipedia defines as
a tendency to search for or interpret new information in a way that confirms one's preconceptions and avoids information and interpretations which contradict prior beliefs. It is a type of cognitive bias and represents an error of inductive inference, or as a form of selection bias toward confirmation of the hypothesis under study or disconfirmation of an alternative hypothesis. Confirmation bias is of interest in the teaching of critical thinking, as the skill is misused if rigorous critical scrutiny is applied only to evidence challenging a preconceived idea but not to evidence supporting it.
The Students on Ice Expedition aims to take 65 international students and 30 scientists on a 17 day journey around the Arctic.

Coordinator Zoe Caron writes "never before has an expedition been so altered by climate change" and proceeds to describe how weather has affected their itinerary and logistics. Fog developed when temperatures dropped. Ice blocked the ship's route. The daily itinerary has changed and changed again. It's because of Climate Change.

Ms Caron is a co-author of Global Warming for Dummies. For some reason I find this funny. And profoundly sad.

Friday, August 01, 2008

The Boogie-man will get us all

Climate change poses big threat to infrastructure, engineers warn

This week the papers were full of a study which concludes that climate change poses a threat to our health and safety because our roads, water supplies, sewer systems and government buildings will become more vulnerable to multibillion-dollar failures.

I like the headline. What if it told the truth?
"Engineering organization reviews design standards based on new data, expresses concerns that old standards were inadequate"

You'd find the story at the back of the paper somewhere, not on page 1.

A common theme amongst the fear-mongers is that extreme weather events will become more common. Have you paused to ask what records we based our expectations about weather on? Why, on the evidence collected and the records kept by those who came before us. Streamflow varies each year, and records of streamflow (or rainfall, or windspeed, or whatever the topic of interest is) can be analyzed statistically to identify the "hundred year flood", a shorthand way of describing how often a given flood level ought to be expected.

Around here we started systematically recording stream flows and rainfall amounts about a hundred years ago. So the biggest storm in a hundred years establishes what the hundred year storm should look like. If you don't have a hundred years of data, look at the actual distributions and use some statistics to project what extreme events might fill in the tails of the curves.

Much of our infrastructure was designed and built fifty years ago, using civil engineering standards based on weather records that were at the time less than 40 years old. Engineers are no fools, so they put safety margins into their calculations. In my experience civil engineers tend to do this more than the other engineering disciplines because their failures are visible to everyone and (not coincidentally) civil engineers get sued more than the other disciplines.

When someone says that climate (by which they mean weather) is becoming more variable, the public should recognize that the records of weather are encompassing more and more years of data, and the shape of the distribution curve is being altered by new actual data, replacing the projections of the past.

The "Public Infrastructure Vulnerability Committee" of Engineers Canada (you'd have to ask yourself whether a committee with a name like that knew their conclusions before they started their study) states that infrastructure was built with "now-outdated assumptions" in the design standards. So while their conclusions are no doubt defensible and some design will prove to not be up to the task it was designed for, the Boogie-man is climate change. Not engineering standards that were based on all we knew or could guess at the time, but climate change.

Handy for everyone, and might even work to get a guy off the hook in a lawsuit. "Not my fault, I used accepted standards. It was climate change, your Honour".

The Boogie-man will get us all.