XKCDetails

I enjoy the comic strip XKCD. It is often science oriented, frequently clever, and heads off in the same peculiar directions that I find myself pursuing from time to time. But this time, I have a complaint:

XKCD produced this comic, poking fun at global warming skeptics:

XKCD 1321

Two things struck me about this right off the bat. First, this is exactly the mindset used by catastrophists who are alleging that “droughts” and “hurricanes” and “heat wavers” and the “polar vortex” and countless other phenomena are new/different/extreme/much worse than ever before. None of it is true, they simply have short memories, and in some cases are adjusting the data.

The second point was the choice of starting point for the graph above: the very cold 1970s. (Then I saw the data source — the dishonest “Skeptical Science” website.) So let’s look at that data, and go back a little further:

StLouis_cold

The dark blue line along the bottom counts the days below 0°F each year. This matches the XKCD chart, more or less, except that “haven’t had a day like that since the 90s” is not quite correct; there have been several in recent years. As cities urbanize, the lowest temperatures are more difficult to reach. I used NOAA’s data here.

But look at the different story in the upper, pale blue line — the days below 32°F. There, instead of a “never seen before” effect, it looks just like the pattern of the early 20th century. Starting the data in the 1970 timeframe makes the prior pattern invisible — a practice that is less than honest, in my opinion.

The website WUWT (Watts Up With That) had some fun with this comic in a different direction.

WUWT spoof of XKCD

None of this includes the other effect — the fact that US temperature authorities (NASA, GISS) continually adjust older temperatures to tell  a better global warming story, and match each other. (The rest of the world gets similar treatment.) In general, temperatures prior to 1960 are cooled, and after that year are warmed. This is not a one-time adjustment; sometimes several times per year they adjust history over again. It took six years of adjustments before they finally got 1998 to be warmer than 1934, for example, and the difference between those two years is now even larger than it was in 2007. In a few months, it will be larger yet.

And as Steve Goddard points out, the dramatic global warming from the 1920s to 1940s, and subsequent dramatic cooling to the 1970s, has been subsequently erased.

A paper1 recently published (and discussed at WUWT) touches on the issue of those adjustments:

Urban contributions to the overall trends of annual mean Tmin and Tmax reach 100% and 28.8% respectively for the adjusted data. Our analysis shows that data homogenization for the stations moved from downtowns to suburbs can lead to a significant overestimate of rising trends of surface air temperature, and this necessitates a careful evaluation and adjustment for urban biases before the data are applied in analyses of local and regional climate change.

Note that the Tmin, the minimum temperatures, are the most affected by adjustments to city temperatures. This is implicated, perhaps, in St Louis’ reduced days below zero — and the urbanization itself can greatly affect this.

But St. Louis is unusual in one respect: Since the mid-20th century, the population of the city itself has actually been declining. There’s a discussion of the population decline in an article here:

City of St. Louis Population Loss: Yet no city which achieved the scale of a half million residents has lost a larger percentage of its population in peacetime than St. Louis. To some extent, this is a very old problem for a city that was once the largest in the Midwest but was passed in 1880 by Chicago. In 1950 the city population peak at 857,000 people and ranked 8th among the nation’s municipalities. By 2009, the latest estimates, the population was 357,000 (ranked 48th in the nation), a decline of nearly 60 percent from the peak.

===|==============/ Keith DeHavelle

1. Zhang, Lei, et al. “Effect of data homogenization on estimate of temperature trend: a case of Huairou station in Beijing Municipality.” Theoretical and Applied Climatology (2013): 1-9.

Categories