A recent project by Tony Heller, whose work in the archives has long commanded our admiration, has been to examine the ways in which U.S. government scientists “adjust” historical temperature readings. And one way is always down in the past, and the farther back you go, the farther down go the readings. Whenever they check them, from Colombia to the Central African Republic to Crawfordsville, Indiana, they find that the people reading thermometers in the past were prone to overstate the temperature so the graph has to be rotated counter clockwise. But then another “way” question is: what possible way could they have of knowing what the temperature “really” was in those places in, say, 1957, since they can’t go back and resample? Which brings us to the third meaning of “the ways”. In what possible way could such a systematic bias decades ago in widely scattered places possibly have crept in to begin with so as to need correcting? And as a bonus if you can think of anything that isn’t laugh-out-loud implausible, what way do they have of determining that it did? So lots of ways. But no way.

It’s certainly possible that temperature readings in the past were inaccurate, because thermometers were imprecise, because thermometers were biased, or because the people doing the readings were incompetent. But if the instruments were inaccurate or the monitors were inept, the mistakes would be random, and then any correction in any place and year would be as likely to raise as lower the figure. Instead with NOAA it’s always down down down as you go back back back. For instance, at the Cali Calipuerto station in Colombia (COXLT957271) we are asked to believe that in the last 20 years the readings were essentially accurate, but in the 1970s they were consistently about 1.5°F too high, in the 1960s 2 to 3 degrees too high and in the 1950s an astounding 5 degrees too high.

Now you may well ask, and we do, how anyone could know such a thing. What possible means exists to compare what the living breathing humans who took the readings back then claim to have seen on their thermometers with the allegedly “actual” temperature at the time? The only obvious answer, though possibly the massive U.S. government agency NOAA that does this weird stuff with data has a better one it just isn’t sharing right now, is that a computer deliberately programmed to say it’s hotter now than in the past duly generated temperature estimates for the 1950s that were 5 degrees lower than today, and then these were compared with the historical readings and the latter were then “fixed”. Which sounds a lot like fiddling the data. But it gets worse.

Suppose we accept for purposes of argument that such a thing might have happened. The question then becomes what it is about the process of reading temperatures that, in the past when “global warming” wasn’t a hot public policy question so nobody had any discernible reason to distort data, could possibly have led people to make systematic rather than random errors of this sort. And we’d love to hear from NOAA on that one.

We mentioned above the possibility that past readings were wrong because thermometers were imprecise. Which they were, especially the alcohol ones that were replaced by mercury before we went digital in this century. But of course if that were the explanation they’d be as likely to read too low as too high. And people who held that view would be most unlikely to speak of knowing how much it has warmed to two decimal places unless, crucially, they insisted that the errors did cancel each other out. Which is the exact opposite of what they claim.

Now let’s examine the hypothesis that older thermometers are biased. Again, it’s remotely conceivable that all alcohol or mercury thermometers were calibrated wrongly, or that the ones in Cali Calipuerto were. But if it were the case, they’d have been wrong by the same amount consistently until the mistake was fixed, and then accurate. To maintain this explanation, you’d have to take the view that someone noticed a problem in the late 1950s and adjusted it, but only by about two degrees, and then a couple of decades later noticed again and adjusted it by another degree or so before finally, around 2001, getting it right. Which is pretty far-fetched especially since from the invention of the “modern” bulb-and-tube thermometer around 1600 they have been calibrated, at least, to the freezing and boiling points of water and scientists were naturally concerned with the reliability of their instruments.

Finally, what about human error? It’s a huge factor in every field where people are active. So yes, the people doing the readings could have been inept for any reason from inadequate training to adequate sloth to excessive fondness for adult beverages. But again, such factors ought to create random errors unless and until they take a night course, start getting up earlier or stop coming to the office drunk. There’s no possible way such an explanation could account for errors that reliably diminish decade by decade. So what are we left with? A thermometer on a pole that is very gradually rising up from the ground so in the 1950s they were looking down at the actual temperature, but progressively less over the years? Or hiring shorter and shorter staff? Bosh.

Of course that suggestion is absurd. But until NOAA offers something more convincing, their dogmatism about the adjustments and silence about the reasons for them is at least as ridiculous. Especially since, Heller demonstrates, it’s not just this one station. In fact over in Bouar in the Central African Republic (CTM00064601) the adjustment back in the 1950s was, gosh, looky here, right about 5°F. Same as in Colombia.

Now you are not going to convince us that the same idiot or idiots were being flown from one to the other and back to take the same readings. Or that the two stations bought the same defective thermometers and then did the same staged adjustments. So the only reasonable conclusion is that NOAA changed the numbers to the same degree in both places because both were actually showing no temperature increase and NOAA was convinced that actually the planet had warmed by roughly that amount. Which is also odd since supposedly the global increase since 1850 has been about 2°F and the poles are meant to be warming fastest leaving the “Central African Republic”, which is apparently in Central Africa or something, to warm more slowly.