In earlier posts, Climate Change I, Climate Change II, and Climate Change III, (and later ones, Climate Change V and VI that I’ve already posted) I’ve discussed aspects of the science of Climate Change as I see it, and as I presented it to my Physics 125 students at the University of Tampa.
In the previous essay, Climate Change III, I told how the author of Physics 125 textbook, Dr. Richard Muller, a UC Berkeley physicist, created the Berkeley Earth Surface Temperature project. He proclaimed that he would review the temperature record raw data to check for himself whether the climate professionals had properly done their work. He let it be known that he was skeptical, and, therefore, he drew funding from various climate denier groups. In the end, however, Dr. Muller’s analysis reproduced that of decades of work by many climate researchers. He admitted as much in a New York Times op-ed essay.
In this essay, I will describe another interesting aspect of the climate scientists’ work measuring the Earth’s temperature. This illustrates the thoroughness with which they are studying this matter and the methods followed by researchers to resolve discrepancies in the data and differences of opinion between scientists. In this matter, I’m writing as an informed amateur, as I am not a climate scientist. I’m also not an historian of science. I’m just going to tell you about some of what I know, but further details are readily available on many climate
Measuring the Earth’s temperature is a tricky and complicated matter, with plenty of details necessary to consider and confounding factors to clear up. Thermometers were not invented until the 18th century, and did not become widespread until the 19th century. Even so, there weren’t many people recording the temperature, and they weren’t spread around in a regular pattern. Of course, 2/3rds of the Earth’s surface is ocean, and data only came in from passing ships. Without going into details, I’ll just say that properly interpolating and extrapolating and averaging is a heroic task involving immense amounts of data gathered from varied sources of varying continuity and reliability.
Of course, the researchers are seeking a single quantity to represent a variable quantity: day to night, winter to summer, equator to pole, mountain peak to Death Valley. They do a lot of interpolation and averaging.
Given these problems, a key test that researchers use is to examine if different temperature records match up. The main data for the past 150 years is from direct temperature measurements. Scientists call other estimates of the temperature proxies. One scientist reports her estimates of temperature from California’s bristle cone pine trees. These go back several thousand years, but are from one location. Another from his Greenland or Antarctic ice cores. Some of these go back several hundred thousand years, and are from several locations. Do these connect and overlap with European thermometer data from England or Germany or North Atlantic shipping lanes? This problem arises anytime a new method or source appears.
If these and other temperature measurements do not overlap, then one or more of them may have an unknown systematic error. All measurements have random fluctuations, which scientists suppose (based on experience) vary around the unknown correct result and average to zero, relative to the unknown (alas) correct result. But measurements may have systematic errors. If a scientist knows that her methods have a systematic error, say that a thermometer in a city’s “heat island” reads high relative to nearby rural thermometers, she will calibrate her urban instrument and correct for this error.
Scientists like to compare observed or experimental results for a quantity, in this case, global average surface temperature, that arise from different methods because unexpected differences in the results suggest the presence of an unknown systematic error.
An important new data source appeared at the end of the 1970s. This was satellite measurement of the lower atmosphere’s temperature. These satellite instruments measured microwave radiation sent upward from the atmosphere, and from these data researchers could estimate the temperature in the lower atmosphere. By the 1990s, however, a major puzzle had emerged, which was that the data published by a University of Alabama, Huntsville, group differed from other satellite records and from ground-based thermometer measurements. The UAH data showed a slight cooling of the lower troposphere when, basically, everyone else showed warming!
The principle investigators for the UAH system are Dr. John Christy and Dr. Roy Spencer. As lead investigators on a major experiment, tenured academics, and NASA researchers they weren’t unthinking uninformed climate change deniers. They published their results in important, refereed journals. They had confidence in their instruments and analysis. As far as I know, they were known and well-thought of in the climate science community. They often spoke out in public forums with the proposition that the harms of climate change were exaggerated, that the science was not well understood, and, indeed, it was possible that the climate was cooling, not warming. This tended to make them less popular with their colleagues, I suppose. I have read that these researchers were evangelical Christians, but I can’t say that their beliefs leaked into their scientific work. Consider Spencer’s three books (which I copied from the Wikipedia article about Spencer):
Spencer, Roy W. (2010). The Bad Science and Bad Policy of Obama’s Global Warming Agenda. Encounter Books. ISBN 1-59403-482-6.
Indeed, Dr. Spenser still advocates these views, as you can read about here: ___ This is a report about a climate change court case in which Dr. Spencer testified for the “Nothing to see here. Just move along now” side. Evidently Dr. Christie also holds and advocates similar views in scientific papers and public forums such as the Wall Street Journal.
During the 1990s there was a major problem in the temperature data. Most datasets and analyses showed that the Earth’s surface temperature was rising, but the UAH analysis showed that the temperature of the lower troposphere was falling. This satellite data also differed from other measurements of the lower troposphere, such as radiosondes, a fancy term for weather balloons.
Scientists like to make measurements of the same quantity from different directions using different techniques, different algorithms, and different graduate students. As scientists believe that there is but one world that they are observing, different observations of the same quantity suggests that one or more of them suffer unknown systematic errors. But which one: the data intensive, highly interpolated, highly averaged compilation of point records, or the modern satellite record that measures, not surface temperature, but atmospheric microwave emissions, from which the atmospheric temperature must be unfolded?
To shorten this to a few sentences, by the late 1998 there were more than one research group with similar satellite observations, and they made their raw data and algorithms available to each other. It turned out that Drs. Christy, Spencer, and their group had a key error in their analysis. They had not properly accounted for the gradual decay of their satellite’s orbit, which gradually changed the volume of the atmosphere their instrument observed. Correcting this error flipped the sign of their temperature record from cooling to warming, and put it close to the other groups’ data. Subsequently, additional, but smaller errors and corrections emerged, all of which placed the UAH temperature data into closer agreement with that of other groups.
Dr. Christy and Dr. Spencer (who I believe is now retired, at least from this group) acknowledged these errors, and they have applied the corrections to their data, working both back in time and to newer measurements. These days, as far as I have heard, their data now, basically, shows what all of the other groups’ data shows.
Remarkably, however, this has not changed their overall views on the subjects of whether the climate changes are natural or man-made, or whether humans need to do anything about this, or about whether the climate models are useful, and so on. Indeed, researchers consider the so-called climate sensitivity to be a key quantity for any assessment of the effects of greenhouse gases on future climate. The climate sensitivity measures how much the Earth’s averaged surface temperature will change in response to a doubling of the CO2 concentration. A high sensitivity means bigger changes than a low sensitivity. The one number contains within it many effects. Let’s say that warming of the air and oceans produces more evaporation of water and more clouds. Clouds reflect sunlight back into space, so perhaps the clouds will counteract the increased absorption from increased greenhouse gases. There are many similar possible positive and negative feedback effects. The modelers try to get them all, and with the correct magnitudes.
It’s difficult and tricky, but the climate change community has, it thinks, a pretty good idea about this after decades of research. Would you be surprised if I were to tell you that Dr. Christy and Dr. Spencer are among those (few) researchers who believe that the climate sensitivity is notably lower than most of their colleagues around the world. Indeed, while some of their modeling work, originally based upon their cooling temperatures, minimized the climate sensitivity. And, it seemed, the farther their publications got from the refereed professional literature, the more they asserted that climate sensitivity was small, insignificant, not well known, and fluctuated.
In recent publications, in the Wall Street Journal, Dr. Christy has complained that his colleagues have for the most part reached agreement on temperature trends, on climate sensitivity, and on climate change in general. Supposedly this is 1984-style group think and he says that the world’s climate scientists are like Nazis trying to impose a single view on everyone. Really, like Nazis!
This post took me too long to write, and I’ve probably written too much. I want to turn to other topics, such as further thoughts on Confederate Monuments, and I want to respond to your posts.