- JoNova - https://www.joannenova.com.au -

How to make climate graphs look scary — a reply to XKCD

This week XKCD (a popular Geek comic site) posted an epic cartoon called “A Timeline Of Earth’s Average Temperature”. It was a cutesy long godzilla hockey-stick — “scary” to the unwary.

It’s easy to make a scary historical-looking temperature graph — so easy that the artist probably didn’t even know how. (Thank Shakun, Marcott, Annan, Hadcrut and the IPCC for doing the tricky part.) First,  guesstimate temperatures over last 20,000 years with anything at hand: tree-rings, ice bubbles, coral, fossilized tea leaves, whatever. Blend. Then stop the proxies, tack on thermometer data that was recorded in a different way with different errors and a very different response to faster temperature changes. Finally, launch that line into the future with unvalidated, skillless multivariate models that predict a fingerprint which 28 million weather balloons can’t find. Then take the models that didn’t work for the last twenty years, and run with the errors to the next century… Voila!

I took the 14,000 pixel cartoon and squeezed it to one shot that shows the curve that matters. See the error bars? Me neither.

(But who needs an uncertainty range when you have faith?)

Hockeystick Graph, XKCD, Cartoon, Holocene.

Click to enlarge.

The secret to a good hockey-stick graph is to never use the same type of data from start to end. If things like tree rings and ice bubbles were so good at measuring the temperature circa 5,015 BC, why don’t we use them in 2015?

Could it be that thermometers will measure every hot day, but corals don’t?

Matt Briggs takes on the graph  Stream: xkcd’s Global Warming Time Series Mistakes

Adding together lots of errors and uncertainties will make a nice smooth line. Any noise can be averaged to one note. This gives the illusion that the climate was once stable.

The Medieval Warm Period was recorded in hundreds of studies. Temperatures a thousand years ago were not so different to today. In the XKCD graph, that bump’s gone, blended to nothing. xkcd calls it “regional” but 6,000 boreholes drilled all over the world suggest otherwise, so do warm Indonesian waters, receding glaciers in New Zealand and melting ice in Antarctica. (See NIPCC.)  How many other bumps in the last 20,000 years disappeared like this too? About 20,000 years worth. It’s like a thousand year smoother was run over the graph up until the last 100 years.

What happened to those error-bars?

As the Great Matt Briggs says:

The picture xkcd presents is lacking any indication of uncertainty, which is the major flaw. We should not be looking at lines, which imply perfect certainty, but blurry swaths that indicate uncertainty. Too many people are too certain of too many things, meaning the debate is far from “settled.”

Global temperatures vary less than polar ones. So I took data from Vostok and Greenland and shrunk to half its actual variation and slapped it over the xkcd line. See it below. This is just an indication of the variability missing from the “smoothed” proxies. The invisible error bars on the original XKCD graph would be wide.

Holocene temperatures, Marcott, GISP, Vostok, XKCD, cartoon, Graph.

Click to enlarge.

Apparently the data for the XKCD graph comes from Marcott (ha ha, UPDATE See Climate Audit: The Marcott Filibuster, and others on Marcott. My post on Marcott: “Ponder how researchers can find 5,000 year old Foraminifera deposits, but not ones from 1940?”

Another good quote so relevant to the graph: That’s 300 year smoothing. We should average the climate from 1700 to now. How scary would that look?

“Marcott et al clearly say there is “…essentially no variability preserved at periods shorter than 300 years…” So if there were, say, occurrences of a warming rise exactly like the last century, this graph won’t show them.”.

You want noise? This is noise:

Anyone who has looked at proxies, knows that they don’t make long smooth lines. This is just the last 2,000 years of variability in the Northern Hemisphere. There are a lot of not-hockey-sticks.

Temperatures, Northern Hemisphere, Climate change, Medieval warm period.

Click to enlarge

Things we know for sure: The Earth was warmer for thousands of years than it is now during In The Holocene peak. Corals and polar bears survived. CO2 was not to blame for the heat that didn’t kill corals or bears.

h/t To Todd, Dennis.

Psst:  Matt Briggs explains why BCE is a spelling mistake:

The plot purportedly shows the average global temperature, presumably measured right above the surface, beginning in 20,000 BC and ending in the future at 2100 AD. Mr Munroe misspells “BC” as “BCE” throughout the cartoon, incidentally, and leaves out “AD”.

No, I’m kidding. “BC” means “Before Christ”, which some academics, sensitive creatures that they are, find offensive on behalf of people they haven’t met, and so they change it to “Before the Common Era”. And how do they demarcate the “Common Era”? By the birth of Christ, a.k.a. BC. The same people who gave us “BCE” gave us “safe spaces”. Skip it. — W. M Briggs.

Update:  :- )

It’s no wonder that skeptics deride,
What hockey-stick warmists decide,
When some smooth operator,
Can make warming come later,
To be man-made, extreme and worldwide.

 — Ruairi

7.5 out of 10 based on 109 ratings