It’s official. After five recent posts criticizing Michael Mann, a half dozen criticizing Skeptical Science, another half dozen criticizing the work of Stephan Lewandowsky, and several others criticizing various people, I’ve gone to war… with Steven Goddard.
Weird, right? I mean, I’ve written one post about him, and even then I said I didn’t want to have to write it. That, plus some comments at his blog hardly seems like the opening salvo to a war. Still, I’m told that’s all the precursor there was to today’s declaration:
Even the friendly user Foxgoose described this as a war:
I don’t get that, but whatevs. Let’s roll with it. Today’s combat began when I saw a message on Twitter which said Sun News had a story about NASA fudging historical temperature data. I was curious, clicked on a link and found Marc Morano had given an interview on television about the issue, using talking points from Steven Goddard.
I scrolled through the post discussing Morano’s interview and almost immediately spotted a problem. The fourth picture was introduced:
NOAA/CRU/NASA erased the post-1940 cooling, just like they said they were going to do in their private E-mails.
The primary quote offered for this said:
if we could reduce the ocean blip by, say, 0.15 degC, then this would be significant for the global mean
The problem I spotted almost immediately is the discussion was about an ocean blip, yet the image offered was this one:
Notice the figure is titled Global Temperature (meteorological stations). If we go the page Goddard’s image originated from, the description of the image says, it’s for an “analysis using only meteorological station.” We do not have meteorological stations scattered throughout the ocean.
Even if NASA adjusted ocean temperature data to fit some preconceived bias, that wouldn’t have any effect on an analysis which didn’t use ocean data. That means Marc Morano got on live television and pointed to a graph which couldn’t possibly show what he claimed it showed.
But it gets worse. They claim a warming blip of the 1940s was removed by NASA altering ocean temperatures. It’s true there was no blip in the analysis which doesn’t use ocean data, but it is in the analysis which includes ocean temperatures:
How does one explain that? I don’t know. I don’t know how Steven Goddard picked a graph which doesn’t use ocean data as showing ocean data was manipulated. I don’t know how Goddard missed the graph on the very same page which included ocean data and showed the blip he was focusing on. I also don’t know how Marc Morano failed to catch such an obvious problem.
What I do know is this is not some exception. Morano also used this image from another one of Goddard’s posts:
That’s a damning image as long as you don’t look closely. If you do though, a problem becomes obvious. Goddard’s red line does not stop where one would expect the original line to stop. Goddard’s red line extends up to and on top of the y-axis. We’d expect the original line to stop before the y-axis. It does.
You can’t tell that from the image because Goddard aligned the two series at the end of the original line. The two lines meet at that point, but you could never tell one ended there. However, you can tell where the endpoint is in the original source. That source also gives us the answer as to what Goddard’s trick was. Part of the image Goddard cut off at the top was the title: “Observed temperature, (5-year running mean).”
If you were plotting a five year average for 1980, you’d need data for 1978, 1979, 1980, 1981 and 1982. Hansen only had data in his series up to 1980. That meant he couldn’t plot a five year running mean past 1978 (which uses data from 1976, 1977, 1978, 1979 and 1980). That’s why Hansen’s graph ends before the y-axis labeled as 1980. Steven Goddard, on the other hand, had data into the 21 century. That meant he could plot as much as he wanted, and he did:
His line shows points up to about 1981, meaning he used data from as late as 1983 (1979, 1980, 1981, 1982 and 1983). That means he’s claiming past temperatures were altered by comparing an 1880-1980 series to an 1880-1983 series.
If there was warming in the four years he added, it’d make the new series look warmer than the old one. Given his line goes up at the end, we know those years were warmer. That means we know his results were caused, at least in part, by him comparing current data after 1980 to old data before 1980.
Steven Goddard claimed an ocean warm blip was removed by citing a series that didn’t use ocean data. The series which did use ocean data showed the blip he claimed was removed. Steven Goddard claimed past temperatures were adjusted to make current warming seem bigger than it is. He made a graph which hid the endpoint of the old series while showing the new series well beyond that endpoint (reminiscent of this).
Despite those problems being relatively obvious, Marc Morano repeated them on live television. I pointed them out, and the responses I got all suggested I should just ignore the issue and focus on other things.
That is not okay. If being a skeptic means anything, ridiculous arguments like these should not be tolerated. They’re every bit as bad as the “tricks” global warming proponents use. If they’re the public face of skepticism, how are skeptics any better than alarmists?
On a related note, the responses I’ve gotten to my criticisms of Steven Goddard’s arguments have been practically indistinguishable from the ones I’ve gotten to my criticisms of Michael Mann. The same is true of when I criticized Christopher Monckton and Cook et al.
Either both “sides” have similar problems and dislike me for pointing them out, or I’m a terrible person who nobody likes.
Judging by my social life, I think I know which it is.
February 17th edit: In response to a user on Twitter, I’ve edited this post to add an image showing where Goddard’s trick is directly relevant.
The point I was unclear about previously is when Goddard aligned the lines at 1978, he made it look as though they met at that point and continued on together. They did not. Only one line continued on. That hid the fact most of the apparent disagreement in his figure was just a matter of what baselines he used.
Had he used a different alignment, such as the one in the image I added, the past disagreement would have been much smaller (note, he doesn’t show the past when he aligns the series at 1940). There also wouldn’t have been much disagreement in the ~1980 period if he stopped his line at the same point as Hansen stopped his. I’ll add an image showing this as soon as I work out the kinks in overlaying the images.
Image Update: These images are rushed and not the best quality, but they should show the point effective. The first image is my attempt at replicating Steven Goddard’s. They’re similar, but you’ll notice his is more exaggerated than mind. I have no explanation for that. I’ve checked both figure’s values against the data files themselves, and his are off.
Regardless, I adjusted the baseline of the current temperature record to show the visual impact of his choice of baseline (this image is not displayed) if you leave the extra data in. I then did the same but removed the extra data so both lines end at the same point. I think the figures speak for themselves: