Objectively testing the "higher variability of climate change" theory

[Roger Smith]

One of the more controversial (and difficult to test) aspects of the climate change theory is that of higher variability. We are informed that climate change will mean more severe storms, droughts, heat waves, and even snowstorms and cold spells in some versions that assign those outcomes to displacement of the polar vortex.

The problem with these assertions is that mainly anecdotal evidence is offered as proof (or disproof). 

I thought of a way to test more objectively for such outcomes, and I should add the disclaimer that I don't have much if any idea whether the claims would prove true, false or inconclusive. 

This method could be called the "forwards-backwards test" and it works like this: 

(a) name any climate extreme that you postulate may be exacerbated by climate change

(b) determine how many years back in time you have to go to find its worst known example (in recent times, let's say since 1800 and the onset of widespread atmospheric measurement)

(c) now if the event does not at least duplicate itself within that number of years going forward, the claim is false. If it duplicates in a shorter period, it is true (on this count at least). If there is a tie, then the matter is inconclusive. 

(d) to test the assertion relative to forest fires or wildfires, some concept of exposure to risk must be applied, clearly forest fires are having greater impacts but how much of that is due to larger numbers of human-forest interactions available due to changes in lifestyle? 

Here's one example of the test. If we are going to see "worse heat waves" in the future, then the 1936 heat wave would need to be outdone within 83 years (by 2102). If it isn't, that claim is false. 

If we are going to see "worse hurricanes" then I suppose the primary test would be central pressure, since damage is a rather quirky function of track, and death toll is a further function of track combined with effectiveness of warning. I wouldn't hold the theorists to having Galveston destroyed before 2138, but perhaps the central pressure of Gilbert (1988) or Wilma (2005) equalled or lowered before 2048 (2031).

For a "worse tornado outbreak" I would be looking for numbers of F4 or F5 tornado reports in one outbreak similar to either 1925 or 1974, before (in each case) 2111 or 2062.

For "worse snowstorms" I would be expecting to see heavier amounts than fell in Jan 1996 within 22 years, or Great Lakes blizzards of the calibre of Jan 1978 before 2058. 

Critics are going to argue that these tests are too specific and too difficult to verify precisely. But if we don't hold the theory to this sort of testing, then what sort of testing is appropriate? In its absence, we face the likely prospect of the political arm of the science just claiming verification after each notable weather event, whether it meets this test or not. And while this may alarm many people, it is not actually in the scientific tradition to claim verification from uncompared data. 

And the problem is that we probably don't have the luxury of waiting that long to assess the theory. I have to wonder, though, if there isn't something to be said for a competing theory that variations will become less intense due to climate change. Perhaps the greenhouse gases are also interfering with stability profiles in a way that, overall, induces less extreme weather. If that were the case, would we actually want to "fix" that "problem" or not? 

 

[skierinvermont]

How about frequency of 100F days in Chicago? Or 102F? Whatever the threshold for "extreme" is. You've made the threshold so extreme (one in 20, 50, 100 yr events) you've made it difficult to verify. There are also lots of events that are clearly 1 in 1000 year events that happened only 10 or 20 years ago. To expect them to recur in 10 or 20 years is unrealistic. 

We could test the frequency of 20" or 24" snowstorms in NYC. (Rather than using the record snowfall of ~30" or whatever it might be). We could test the frequency of 3" rainfalls in 24 hour periods in Kansas City.

[Vice-Regent]

17 hours ago, Roger Smith said:

One of the more controversial (and difficult to test) aspects of the climate change theory is that of higher variability. We are informed that climate change will mean more severe storms, droughts, heat waves, and even snowstorms and cold spells in some versions that assign those outcomes to displacement of the polar vortex.

The problem with these assertions is that mainly anecdotal evidence is offered as proof (or disproof). 

I thought of a way to test more objectively for such outcomes, and I should add the disclaimer that I don't have much if any idea whether the claims would prove true, false or inconclusive. 

This method could be called the "forwards-backwards test" and it works like this: 

(a) name any climate extreme that you postulate may be exacerbated by climate change

(b) determine how many years back in time you have to go to find its worst known example (in recent times, let's say since 1800 and the onset of widespread atmospheric measurement)

(c) now if the event does not at least duplicate itself within that number of years going forward, the claim is false. If it duplicates in a shorter period, it is true (on this count at least). If there is a tie, then the matter is inconclusive. 

(d) to test the assertion relative to forest fires or wildfires, some concept of exposure to risk must be applied, clearly forest fires are having greater impacts but how much of that is due to larger numbers of human-forest interactions available due to changes in lifestyle? 

Here's one example of the test. If we are going to see "worse heat waves" in the future, then the 1936 heat wave would need to be outdone within 83 years (by 2102). If it isn't, that claim is false. 

If we are going to see "worse hurricanes" then I suppose the primary test would be central pressure, since damage is a rather quirky function of track, and death toll is a further function of track combined with effectiveness of warning. I wouldn't hold the theorists to having Galveston destroyed before 2138, but perhaps the central pressure of Gilbert (1988) or Wilma (2005) equalled or lowered before 2048 (2031).

For a "worse tornado outbreak" I would be looking for numbers of F4 or F5 tornado reports in one outbreak similar to either 1925 or 1974, before (in each case) 2111 or 2062.

For "worse snowstorms" I would be expecting to see heavier amounts than fell in Jan 1996 within 22 years, or Great Lakes blizzards of the calibre of Jan 1978 before 2058. 

Critics are going to argue that these tests are too specific and too difficult to verify precisely. But if we don't hold the theory to this sort of testing, then what sort of testing is appropriate? In its absence, we face the likely prospect of the political arm of the science just claiming verification after each notable weather event, whether it meets this test or not. And while this may alarm many people, it is not actually in the scientific tradition to claim verification from uncompared data. 

And the problem is that we probably don't have the luxury of waiting that long to assess the theory. I have to wonder, though, if there isn't something to be said for a competing theory that variations will become less intense due to climate change. Perhaps the greenhouse gases are also interfering with stability profiles in a way that, overall, induces less extreme weather. If that were the case, would we actually want to "fix" that "problem" or not? 

 

We could re-fix much of the current climate extremes and rosby wave perturbations by simply progressing deeper into global warming. I also believe that Hansen's theories about polar ice melt driving superstorms is likely to be incorrect because the rate of emissions is so extreme. AMOC superstorms are largely confined to the Eemian period and with good reason. (A period of high solar obliquity and low CO2)

You need specific conditions for their generation and a hyper-significant temperature gradient. There will still be significant sea level rise but we may be able to rest easier because the storm tide range will not change much from now due to a lack of hypothesized superstorms. However there will be a strong uptick in landfalling tropical cyclones and they will also become more intense wherever they form.