Another Way to Lie with Statistics
By Howard Hayden, The Energy Advocate, July 2024
Imagine taking a temperature reading every hour for thirty years (one climate data point) and plotting the probability that the reading is (say) 15ºC versus temperature. You’ll get a curve something like the gray curve in Figure 1. The left-to-right width of the curve from the coldest night to the hottest summer day will probably be around 60ºC (ca. 110ºF)
If you continue taking data for another 30-year period, you may find that the temperature has risen, so the new curve is shifted to the right a bit. Let’s have a look at the dishonest representation in Figure 1 from Thomas Karl et al [1].
· Notice the absence of record cold weather.
· The arrow labeled more hot weather refers to the pink area between the two curves, not the implied area under the upper curve. (The drawing shows an abrupt end of the gray Previous climate curve above the Hot label, although the normal curve never reaches the axis, but approaches it asymptotically.)
· Ditto for the more record hot weather curve.
· Given that the temperature range of the Previous climate curve is about 60ºC, the implied temperature increase to the New climate curve is a whopping 12-13ºC.
Figure 1: Highly exaggerated probability graph showing more hot weather and more record hot weather from ref. 1.
In our little exercise of taking data in two successive 30-year periods, the realistic temperature change of late would be less than 0.5ºC. But let us presume that Karl et al meant the temperature shift to be 3ºC, the putative most probable temperature rise caused by doubling the CO2 concentration. Figure 2 shows the probability curve of the previous climate and that due to a 3ºC temperature rise. Note that that there is precious little increase in the probability of hot weather or of record hot weather, as shown by the minuscule shaded area between the two curves.
Figure 2: The shift of the temperature probability curve for a 3ºC temperature rise (with no attention paid to the coldest weather). Bear in mind that even Figure 2 is a bit of an exaggeration, because the 3ºC shift is IPCC’s “most likely” temperature increase due to CO2 doubling, even though it is physically impossible for the radiative forcing from CO2 doubling to block the additional infrared from the surface that must accompany a 3ºC temperature rise.
Also bear in mind that when “climate scientists” endlessly tell us that storms, floods, cyclones, hurricanes, droughts, heat domes, atmospheric rivers, and other such weather phenomena are caused by “climate change,” they do not speak the truth.
[1] Thomas R. Karl, et al, eds., “Weather and Climate Extremes in a Changing Climate,”
https://www.globalchange.gov/reports/sap-33-weather-and-climate-extremes-changing-climate. A similar graph can be found in IPCC’s Third Assessment Report.
https://view.officeapps.live.com/op/view.aspx?src=https%3A%2F%2Fwww.sepp.org%2Fscience_papers%2FAnother%2520Way%2520to%2520Lie%2520with%2520Statistics.docx&wdOrigin=BROWSELINK
