In this article Roger Pielke Jr. notes that it may take a long series of trials to detect shifts in the mean of a variable if those shifts are small compared to its random fluctuations. The example he uses is based on numbers of severe hurricanes observed in the years 1995 to 2022 (which fluctuate wildly between 0 and 7 with an average of about 2.6 and a standard deviation of about 3) whose frequency distribution he then uses to assign random values to the years 2023 to 2050. First unmodified, then with increases (ie shifts of the frequency distribution to the right) of one and two. The ostensible point of this exercise is to show the reader why there are many possible consequences of climate change from which the IPCC does not expect to see convincing evidence in the next few decades. But it is presented with a subtle(?) subtext that appeals to those who deny the need to limit our CO2 emissions by claiming that the media are citing extreme events as evidence for climate change when they might just be the result of random fluctuations(*). Furthermore, by looking just at the overall shape of the resulting time series plots, he overlooks the fact that it is often just those extremes that are truly relevant.
The historic data never goes above 7, and so neither does the randomly generated “no change” projection (even in the longer projection shown later in the article). But the one step up projection reaches 8 (though by chance only once in the run that Pielke chose to show us) and the two step up reaches 8 or more twice.
This might not be so much of an issue with the number of hurricanes per year but might be more so if one considers instead the greatest magnitude occurring in any given year. So let’s look at those randomly generated pseudo-projections as if the number for each year was the highest storm surge that year at some specific location. From the historical data it seems safe to have dykes capable of protecting us from anything up to a 7ft surge, but after the “climate change” that will not suffice and almost every run of the pseudo-projection now yields a disaster within the next 25 years. So, according to that toy model, despite not being able to “see” the effect in a first look at the trend lines of those graphs, if we don’t invest now in higher dykes we are almost certainly screwed and if we don’t stop driving the change we’ll have to repeat that investment again and again.
(*) The article opens with a complaint that “ABC News wrote an accurate story about how climate was not a major or even significant factor in the Lahaina, Maui fire and disaster” but “After being mobbed by the enforcers, the story was changed to emphasize the role of climate.” But although there were certainly other factors (such as a change of local vegetation as a result of reduced market for sugar cane), it would be just as wrong to assert with confidence that climate was not a significant factor as to claim for sure that it was. And actually what the substance of the article discusses would be more in support of a claim that the fire was not significant evidence for climate change than that climate change was not a significant factor in causing the fire – which is a completely different question (hint: both the words and their order are different).
Source: Signal and Noise – by Roger Pielke Jr. – The Honest Broker