Continuing University of Guelph professor Ross McKitrick’s look at Steven E. Koonin’s landmark book Unsettled: What Climate Science Tells Us, What it Doesn’t, and Why it Matters.
Theoretical physicist Steven Koonin moved with his wife to Chevy Chase, Maryland in May 2009 to join the Obama Administration as Under Secretary for Science in the Department of Energy. Seven months later the snowiest winter ever recorded hit the Capital area, including a storm dubbed “Snowmageddon”. But in Chapter 7 of his book Unsettled, “Precipitation Perils – From Floods to Fires” Koonin resists the temptation to treat the event as proof (or disproof) of anything related to climate, and instead presents a graph of Washington DC snowfall totals from 1889 to 2018, within which it becomes clear that 2009 was an outlier against the context of a long, slow decline in average snowfall in the Washington area. When talking about precipitation, it takes a lot of data to establish the context, and that gives plenty of openings for the cherry-pickers to engage in trickery.
The declining trend in DC snowfall leads Koonin into the larger topic of precipitation trends, and specifically questions related to trends in snowfall, rainfall, droughts, flooding and wildfires. Here Koonin makes a radical departure from just about every other public commentator on the subject when he says (p. 130) “We’ll turn to the data to answer those questions.” And that approach makes it inevitable that this chapter would skewer yet another batch of alarmist slogans. Anyone who works with precipitation data (as I have done) knows it is extremely variable and trends in one location may run counter to those in nearby locations. And while one can easily cherry-pick data to tell a story, long term big-picture conclusions are extremely elusive.
Koonin begins his survey by explaining the physical basis of climate modelers’ view that global warming will intensify precipitation. But he then shows long term (115 year) graphs of global and US precipitation rates, which show minor net increases but with very large natural variability and extended periods with trend reversals. The view that there is no detectable trend is supported by published literature and past IPCC reports, both of which Koonin quotes.
He also shows data indicating that an increase in heavy precipitation events was observed in the US from 1910 to 2015. But he notes the changes are uneven across regions (and John Christy and I have shown that the apparent trends disappear using longer datasets where available.) Koonin further observes that the IPCC draws only a tepid conclusion regarding whether such increases are observable globally. As for average Northern Hemisphere snow cover, season-specific data show reductions in Spring and Summer since the 1960s, but increases in Winter, with no annual trend after the late 1980s despite the observed warming. Yet, as Koonin notes, the most recent US National Assessment states as one of its key findings, with no explanation or accompanying pesky data, that Northern Hemisphere snow cover snow cover metrics “have all declined.”
Turning to floods and droughts, Koonin again finds that the recurring pattern is the absence of a pattern. US data on flooding indicates a wide variety of changes over time, as does global data, which leads the IPCC to have only “low confidence” even in the sign (positive or negative) of the trend globally. Likewise with droughts: Despite the repeated use of local drought events as journalistic “proof” of climate change, both US and global data show much variability but no long-term trend. If anything, droughts in the 20th century appear to have been shorter and milder than those in past centuries. Yet – as Koonin notes and his readers by now will have anticipated – this information is “entirely absent” (p. 141) from the 2014 US National Climate Assessment. The 2017 report contained a brief mention of the evidence of the past millennium, but then devoted twice as much space to discussing the California drought then ongoing.
It is ironic that one of the false charges leveled against Koonin is that he doesn’t understand the difference between weather and climate. Of course he does, and carefully distinguishes the two throughout his analysis driven by, of all things, data. That charge should instead be leveled against the National Assessment authors, who ignored millennial-scale evidence in favor of a highlighting short-term local drought event (which reversed to wet conditions shortly after the report was published).
After also reviewing the evidence on wildfires (spoiler alert: they’re declining globally) Koonin ends his chapter by examining a 2015 speech by former Central Banker and now full-time UN climate guru Mark Carney, in which the latter grabbed hold of a 2014 forecast by the UK Met Office and used it as a basis to warn his audience that UK winter rainfall would go up by 10 percent over the next 5 years. The data show that it instead fell by almost 40 percent over the forecast interval. Carney, of course, learned nothing from this episode. But Koonin’s readers will by this point have learned that science bureaucracies, and their cheerleaders like Carney, are not to be trusted.