Saturday, 30 July 2011

"Mark One Eyeball" - an Effective Weapon Against AGW

About a year ago, I read with some amusement, a blog posting by someone who accused sceptics (of AGW) of "cherry picking" temperature data to prove their claims that climate warming had stopped in the current century. "They are looking at the wrong data" he opined "look at this". "This" was a temperature chart showing global temperature from 1900 to 2010. "It's the long-term trend that's important" he continued. I'll use his tip if my bank account goes into the red in the future. When I get the letter from the bank, I can tell them "You're cherry-picking data - It's the long-term trend that's important".

So what's the "Mark One Eyeball" I refer to in the title?
This term is actual British military slang for eyes or eye sight, derived from British Royal Navy nomenclature for distinguishing sequential variations of a piece of equipment (i.e., "Mark 13 Depth Charge"). Since the human eye has not changed, it is called the Mark I Eyeball. The term is typically used when someone relies too much on their equipment: "Use your Mark One Eyeball!". It is used likewise in the United States military and other predominantly English-speaking countries. 
The point I'm making is that the "Mark One Eyeball" used with our "Mark One Brains" (no better versions available currently) is a valuable tool for spotting trends in graphical data, and to a lesser extent in tabular data. The eye/brain combination seeks to identify shapes and lines in collections of dots, blobs or points on a graph. The MOE is also good at spotting conflicting scientific or statistical claims, sometimes made by the same person. In December 2009 Dr. Vicky Pope, who is head of the climate predictions programme at the Hadley Centre of the UK Met Office, was (apparently) quoted in a BBC report "figures indicate that the years since 2000 - the "noughties" - were on average about 0.18C (0.32F) warmer than years in the 1990s; and that since the 1970s, each decade has seen an increase of about the same scale". MOE was telling anyone who cared to look at ANY temperature dataset that this couldn't possibly be true. This is the Hadley Centre's own data for 1991-2009:

Just a year later in November 2010 she was saying "The long-term warming trend is 0.16C, in the last 10 years the rate decreased to between 0.05 and 0.13". That's rather different from the earlier statement, presumably based on the same data, and the quoted rate has a range of 1:5. sounds rather imprecise - the difference in the range is half the long-term decadal trend. Perhaps they've got too much data to analyse, and use of the MOE would give a better result. In 2005 it was common to see plots showing 1970 to 2005 with a nice smooth "eyeball" trend:

Now we're told that plots like the first above (1991-2009) is "cherry-picking" and we should be looking at:

 .... which of course shrinks the "inconvenient" last decade at top right. Unfortunately, "Mark One Eyeball" more easily interprets that part as a downward trend, which in fact is what it is. I'm not concerned whether that trend is "statistically significant" or not, global temperature is clearly dropping since 2002. After such claims were ignored for years, admissions of a "reduction" in warming are more common. The first research papers detailing possible (but not proven) reasons have been published. Of course those papers don't question any of the basics like climate sensitivity to CO2 or cloud cover. That would be akin to heresy. Never mind that scientists should be endeavouring to disprove their own hypotheses, rather than leave it to others. The question in mainstream climate science is not "am I right" but "can I stop a rebuttal paper being published".

No comments:

Post a Comment