Do You Trust the Climate Change Experts?

• Author: Kleit, Andrew N.

Climate Rationality: From Bias to Balance

By Jason Scott Johnston

280 pp.; Cambridge

University Press, 2021

University of Virginia law and economics professor Jason Scott Johnston's recent book Climate Rationality offers an extensive critique of climate change science, built upon his broad research in the area. This is a worthy task, as the climate change literature is increasingly important in directing public policy. In doing this, Johnston attempts to thoroughly evaluate a broad collection of academic articles. The result is a dense tome deeply critical of climate science and its applications, with a reference list of over 550 academic articles. While this work is needed, not all parts of the book are equally compelling.

Some literature and its applications / The academic literature Johnston examines is far too vast to review in depth here. Instead, I focus on his views of three aspects of climate science: how changes in temperature and other relevant factors are measured, how changes in climate are forecasted, and how the economic effects of climate change are estimated.

Climate science relies on climate data. Johnston points out several weaknesses in how these data are measured.

Because the methods of calculating maximum temperatures on land have changed, as have the physical areas around many or most temperature reporting sites, the raw temperature data from land-based weather stations are subject to a great deal of adjustment before being used in analysis. Of course, the more adjustment that occurs, even with the best of intentions, the more error is likely to occur. Johnston claims that this process is subject to "upward" bias, inaccurately increasing temperature recordings. Determining the accuracy of Johnston's claim is very difficult.

He also suggests that using the relatively cool 1950s as base years rather than the relatively hot 1930s serves to bias the estimates of temperature change upward. Essentially, the base year does matter. Today's temperatures may be higher than they were in the 1950s, but perhaps not higher than in the 1930s. This, in turn, may change our viewpoints on the threat of climate change.

Theory predicts that climate change will have larger effects in the troposphere (the lowest part of the earth's atmosphere) and the upper reaches of the ocean than it will on land. The temperature measurements from these places are in large part taken from satellites. These readings have both the advantage of not needing as much adjustment as land-based measurements and the disadvantage of a shorter period of data measurement. Johnston tells us these studies show far less warming than the studies based on temperatures from land-based monitors. The inference is that, in these circumstances, the land-based readings are less trustworthy.

Similarly, reported increases in hurricanes in general and intense hurricanes in particular can be attributed to the rise of satellite reporting, which today observes essentially all hurricanes, rather than pre-space-age ship-based reports. Similar challenges exist for measurements of sea level. Even with satellite-based data on sea level, the magnitude of needed corrections and potential errors in these corrections swamp the magnitude of the measured sea level change.

Global climate models (GCMs) are used to forecast how much the climate in the future will change because of carbon emissions. These models are highly complex and, again, subject to error. An unfortunate feature of GCMs is that, for almost everyone, the models are "black boxes" whose internal characteristics are not well understood. Of course, the quality of any model is unlikely to be better than the quality of the assumptions going into it. The limited amount reported on these models indicates that they are dependent in large part on assumptions about the feedback on climate from both clouds and aerosols. Unfortunately, as Johnston ably explains, the values of these variables are essentially unknown. These models also generally omit the warming effects of black carbon (what the EPA calls "soot"). A further problem with GCM models...