Guest post by Indur M. Goklany
1. Krugman in Gradual Changes and Extreme Events forgets that there is a threshold on the left hand side, below which cold kills. In fact, in the moderate to higher latitudes more people die daily during the cold months than in the rest of the year. See Winter kills: Excess Deaths in the Winter Months.
2. How does Krugman know that the distribution does not become narrower due to warming?
3. Where is the data that shows extremes have become more intense or more frequent, after one corrects for better detection, increased population, and better communications? It certainly doesn’t hold for cyclones, as Ryan Maue’s ACE graph shows. Events more extreme than any we have witnessed over the past 30 years (or whatever) have occurred before and will, no doubt, occur again, even absent any anthropogenic climate change.
4. Empirical data show that even if extremes are more frequent and intense, lives lost have declined. As noted in the A Primer on the Global Death Toll from Extreme Weather Events — Context and Long Term (1900–2008) Trends, long term (1900–2008) data show that average annual deaths and death rates from all such events declined by 93% and 98%, respectively, since cresting in the 1920s (see Figure). These declines occurred despite a vast increase in the populations at risk and more complete coverage of extreme weather events.
Source: Goklany, IM. 2009. Deaths and Death Rates from Extreme Weather Events: 1900-2008. Journal of American Physicians and Surgeons 14 (4): 102-09 (2009). Available at http://www.jpands.org/vol14no4/goklany.pdf.
5. Similarly, empirical data also do not show any significant upward trend for property losses once increases in population and assets-at-risk are accounted for. See (a) Pielke, Jr’s weblog on Normalized Disaster Losses in Australia, (b) Bouwer, L.M. (2010), “Have disaster losses increased due to anthropogenic climate change?” Bulletin of the American Meteorological Society, doi:10.1175/2010BAMS3092.1, (c) Neumayer, E., and Barthel, F. (2010), “Normalizing economic loss from natural disasters:A global analysis, ” Global Environmental Change http://dx.doi.org/10.1016/j.gloenvcha.2010.10.004.
Professor Krugman offers two new courses . . .
ECO314: How to deceive with simple charts
ECO425: How to make millions while destroying your credibility
/sarc off
Weird
If English is not your first language then I apologise but in that case I would suggest that you have missed the nuance in Krugman’s article.
He uses the word “event” which in the context of what he is writing, unless he simply means a temperature above his “threshold” (which makes the whole article totally pointless), infers an actual event:- ie, hurricane, flood, or similar misfortune caused by that elevated temperature and which (again unless the article has no point) in some way detrimental to humanity.
There is no evidence that the rise in temperature has resulted in an increase in these “events”.
People can easily “freeze to death” due to a lack of heat. People cannot easily “boil to death” due to a lack of cool. Before, the 1950’s most of the world survived very well without air conditioning.
Indeed, we are near a tipping point with CO2 at the present 390 ppm. That tipping point is 90 ppm at which photosynthesis ceases and this green world as we know it would come to an end. In some greenhouses CO2 is injected to keep levels at 1,000 ppm which plants consider nirvana.
Does Krugman acknowledge the relationship of CO2 and temperature for the last 5.5E8 years as documented by Berner and Scotese? Is Krugman aware that there is absolutely no relationship between CO2 and temperature for this geological period?
I’m confused about Goklany’s point #4, in which he offers what he seems to think is a criticism of Krugman’s claim.
Krugman claims that as global average temperatures increase, the probability, and hence frequency, of “extreme events” will increase as well.
Krugman defines an “extreme event” as “a case in which the temperature exceeds some threshold.”
Goklany’s point #4 is that the number of deaths resulting from extreme events has sharply decreased since about 90 years ago.
But what does that have to do with Krugman’s claim about the frequency of extreme events increasing? Krugman made no claim about the frequency of deaths. And point #4 seems to have no bearing on the claim that Krugman in fact made.
Is Goklany trying to define an extreme event in terms of “number of lives lost”? If so, then he’s clearly not addressing Krugman’s claim (with point #4). And if not, then I’m not clear on how this is relevant. What have I missed?
Krugman’s graph is faulty.
1) It ignores the threshold for cold events. It shows only the threshold for warm events. There should be a threshold on the left as well as on the right. According to his graph, there are no extreme events associated with low temperature, which is plainly false.
2) Also, the shape of the probability distribution assumes a normal distribution, which is not realistic. Cold kills more surely than heat because of energy requirements. Typically you need a fire to keep warm, and only water to keep cool.
It doesn’t matter a cracker that the posters here in the main excoriate Krugman their views will not get into the MSM. And Brian you comment on “increasing frequency of extreme events” which in relation to cyclones at least, is clearly not correct. Tim Flannery, another in the Krugman mould, has just been apponted Climate Commisioner in Australia. God help us
There are actually two main climates, Summer and winter. While the warmists spend all their time worrying about a 0.6 C rise in temperature in the last hundred years,
temperature rises of 15-20 C occur from winter to summer.
Strong weather systems thrive on temperature contrasts which, of course, happen in winter months, not the result of overall warming.
I suspect those 7000 include deaths from the Indian Ocean tsunami of 2004. About six (??) months ago there was a discussion on WUWT (I think it was here) when some alarmist organization tried to pull that fast one, lumping tsunami deaths under the term “natural disasters” and then using equivocation to shift the pea to blame AGW.
An “extreme event” is a synonym for a catastrophe, a catastrophe implies lost lives and property damage.
Wobble says: “2. If the distribution becomes more narrow, then there will be less extreme events from both warm and cold – thus lives will be saved. If the distribution stays the same (which you are assuming), then the increase of extreme warm events will be offset by the extreme cold events – thus lives will be saved. This is exactly why the CAGW narrative pushes the fact that CO2 will increase the distribution width – hence the name change to climate disruption. They’re doing this because it’s much more difficult to disprove.”
Yes, and remember that all the points on the curve are radiating to space by T^4. Those on the right end are radiating a LOT more than those on the left. That curve is VERY resistant to being shoved to the right, and the farther a point is to the right, the harder it is to move it further. Thus the curve will get narrower as it heats up.
Roger Knights said:
An “extreme event” is a synonym for a catastrophe, a catastrophe implies lost lives and property damage.
I suppose you could (a) stipulate that that’s what you mean by “extreme event,” or (b) argue that that’s what everyone ought to mean by “extreme event.” But (a) is irrelevant to Krugman’s claims. He defines an “extreme event” as “a case in which the temperature exceeds some threshold.” So talking about the number of lives lost doesn’t engage Krugman’s argument (hence my comment a few posts up). As for (b), well, let’s see the argument.
jorgekafkazar says:
February 9, 2011 at 6:01 pm
Wobble says: “2. If the distribution becomes more narrow, then there will be less extreme events from both warm and cold – thus lives will be saved. If the distribution stays the same (which you are assuming), then the increase of extreme warm events will be offset by the extreme cold events – thus lives will be saved. This is exactly why the CAGW narrative pushes the fact that CO2 will increase the distribution width – hence the name change to climate disruption. They’re doing this because it’s much more difficult to disprove.”
Yes, and remember that all the points on the curve are radiating to space by T^4. Those on the right end are radiating a LOT more than those on the left. That curve is VERY resistant to being shoved to the right, and the farther a point is to the right, the harder it is to move it further. Thus the curve will get narrower as it heats up.
I think you are missing the effect on spatial components and oversimplifying the issue. The warming of the Arctic has impacted the Arctic Oscillation, pushing it into negative mode more often. This has created more extreme snow events in parts of the Northern Hemisphere in the winter time, including the US.
http://www.appinsys.com/GlobalWarming/RS_Arctic.htm#aoi
Let’s put the stats aside for a moment and ask a more fundamental question.
What is the scientific basis which supports the ASSUMPTION that a warmer earth will have more extreme weather events? I never see the scientific explanation of why this should be true, and if one stops to consider for a moment, the science says a warmer earth should have LESS extreme weather events.
Consider some simple facts. The first is that as the earth warms, cold regions warm more than hot regions. Break down GISS or HadCrut by latitude and you will see exactly that. The temperatures in the tropics over the entire temperature record are almost exactly the same. Temperatures in temperate zones have increased more and in arctic zones still more. Which is EXACTLY what Stefan-Boltzman says should happen, regardless of what is driving it. This applies to daily and seasonal temps as well. Daily lows increase far more than daily highs, and seasonal lows increase more than seasonal highs. Now onto the second fact.
Where exactly do people thing weather events, extreme or regular, come from? Fact is they come from energy flowing from high concentrations to low ones. Example, what is wind? It is air moving from high pressure (high energy) zones to low pressure (low energy) zones. What makes the wind blow extremely hard instead of softly? THE DIFFERENCE IN PRESSURE (ENERGY) BETWEEN THE ZONES. The greater the energy DIFFERENTIAL the greater the weather “event” that accompanies it will be to redistribute energy more evenly.
Now let us ask, what caused the pressure zones in the first place? Anthony can explain better than I, but the driving factor is the daily heating and cooling of the earth as it spins in space. Heating when facing the sun, cooling when not. So let us now put the science of extreme weather events into perspective and ask what logical conclusion should we draw.
If the arctic zones warm faster than the tropics, the temperature differential between them declines. All air currents, ocean currents and other factors driven by the temperature differential between the poles and the equator must also decline.
As seasonal lows increase faster than seasonal highs, the seasonal differentials must decline, and hence any weather factors driven by the changing of the seasons must also decline.
And as daily lows increase more than daily highs, the daily differentials must decline, and hence any weather events driven by daily heating and cooling must also decline.
The assumption of the warmists, which I rarely see challenged, is that higher temperature should simply be likened to a higher voltage in an electric circuit. That would be absolutely correct if there was only one voltage source in the circuit in which case a higher voltage would in fact mean higher current in the circuit. But that’s a lousy and highly misleading analogy.
Consider a circuit with TWO voltage sources. Let’s say (and I’ll make the numbers lopsided to illustrate the point) we have a 10 volt source and a 100 volt source. The differential is 90 volts. Now let’s increase the voltage of both sources, but with Stefan-Boltzman relationships in mind. The low voltage source goes up to 90 volts and the high voltage source to 120. Now the differential is 30 volts. One cannot argue that the energy potential in the second scenario is MUCH higher than in the first one, but IT DOES NOT MATTER. The current that flows in the circuit is driven 100% by the differential. Up it to 1,000,000 volts and 1,000,030 volts and you will get the EXACT SAME CURRENT.
Back to weather now. Everything we call weather is ultimately driven by warming and cooling which in turn create air and ocean currents, pressure cells and so on. If the world warms, as both Stefan-Boltzman laws of physics require, and as the temperature record itself shows, the earth’s temperature differentials, daily, seasonaly and latitudinaly, decline. The total amount of energy in the system may be larger as a result of warming, but with lower temperature differentials at every level of the equation, the science is clear.
A warmer earth means a more stable earth. A warmer earth means less extremes in temperature differentials, and hence less extremes in all the weather processes derived from them. Which is all of them.
AGWeird (and Brian)
1: He [Krugman] does not forget about cold extreme events. In fact, he mentions this several times in the article.
RESPONSE: I don’t see it on the curve – do you? May be it was just Freudian slip.
2: When the distribution shifts to the right, there are three alternatives of what may happen: the distribution becomes wider, more narrow or no change at all. The most reasonable would undoubtedly be to assume that there is no change in the width of the distribution. As far as I know, there is no reason to believe that the change of the distribution has a higher chance of becoming narrower than becoming wider.
RESPONSE: I have a hypothesis that it is more likely to be narrower for the following reason: a good part of weather is driven by the need to reduce the temperature differentials between the warm equatorial areas and the cooler poles. But if, under global warming the increase in temperatures in the polar regions exceeds the increase in the equatorial areas, the temperature differential should decline and winds should diminish. This is only a hypothesis so don’t take it to the bank, but it offers a mechanism for less intense winds (and therefore a narrower probability distribution for winds).
3: If the number of warm and cold events has not changed, only our ability to detect it, why do you think that we would get a bias toward warm weather?
I believe that if that was the case we would see that the number of observed cold and warm events would increase approximately the same.
RESPONSE: First, I would not disagree that there has been some (unquantified) amount of warming, but do we know that this is anything more than a rebound from the Little Ice Age? A warming over a few decades doesn’t necessarily mean that the distribution has changed permanently in one direction or another.
4 (a): He [Krugman] never mentioned anything about the relationship between lost lives and increased extreme events in this article. In fact, he didn’t mention lost lives at all. You call the article nonsense because of something he hasn’t said.
RESPONSE: Just as the term “climate change” usually is code for anthropogenic climate change due, most likely, to man-made greenhouse gas emissions, or the term “denier” is code for an anti-Semite, if not, a pro-Nazi person, increased probability of extreme events in the context of climate change is code for more floods, droughts, natural disasters and all else that might entail (including more deaths and property loss). You can get a flavor for this if you go to the website of Joe Romm, Krugman’s climate change guru, which has a testimonial in the sidebar from Krugman — “I trust Joe Romm on climate”. [Romm’s website is at climateprogress.com. Once there, click on the link to “Extreme Weather”. Once you wade through this you may get a flavor for what an increase in extreme events implies in warmist circles.]
4 (b): And where is the data where you have taken increased ability to adapt to extreme weather events into consideration?
RESPONSE: I don’t recall saying anything about adaptation in this post, but no matter, since my graph is based on empirical data, it automatically incorporates whatever adaptation measures that populations may have availed themselves of. But because adaptation is automatically accounted for, I do not equate the empirical declines in deaths and death rates from extreme weather events to a decline in their frequency or intensity.
Incidentally, the above figure includes deaths from droughts, floods, wildfires, slides, heat waves, cold waves, cyclones and other storms.
5: The same goes here. He never mentioned anything about property losses.
RESPONSE: See response to 4(a), above.
BTW, what I was attempting to do in points 4 and 5 was to suggest to the reader that regardless of whether extreme events were becoming more frequent or more intense, in the real world human beings were making them less consequential. In other words, the issue of whether the frequencies and/or intensities were increasing (or not) were really a tempest in a shrinking teapot.
AGWeird (and Brian)
1: He [Krugman] does not forget about cold extreme events. In fact, he mentions this several times in the article.
RESPONSE: I don’t see it on the curve – do you? May be it was just Freudian slip.
2: When the distribution shifts to the right, there are three alternatives of what may happen: the distribution becomes wider, more narrow or no change at all. The most reasonable would undoubtedly be to assume that there is no change in the width of the distribution. As far as I know, there is no reason to believe that the change of the distribution has a higher chance of becoming narrower than becoming wider.
RESPONSE: I have a hypothesis that it is more likely to be narrower for the following reason: a good part of weather is driven by the need to reduce the temperature differentials between the warm equatorial areas and the cooler poles. But if, under global warming the increase in temperatures in the polar regions exceeds the increase in the equatorial areas, the temperature differential should decline and winds should diminish. This is only a hypothesis so don’t take it to the bank, but it offers a mechanism for less intense winds (and therefore a narrower probability distribution for winds).
3: If the number of warm and cold events has not changed, only our ability to detect it, why do you think that we would get a bias toward warm weather?
I believe that if that was the case we would see that the number of observed cold and warm events would increase approximately the same.
RESPONSE: First, I would not disagree that there has been some (unquantified) amount of warming, but do we know that this is anything more than a rebound from the Little Ice Age? A warming over a few decades doesn’t necessarily mean that the distribution has changed permanently in one direction or another.
4 (a): He [Krugman] never mentioned anything about the relationship between lost lives and increased extreme events in this article. In fact, he didn’t mention lost lives at all. You call the article nonsense because of something he hasn’t said.
RESPONSE: Just as the term “climate change” usually is code for anthropogenic climate change due, most likely, to man-made greenhouse gas emissions, or the term “denier” is code for an anti-Semite, if not, a pro-Nazi person, increased probability of extreme events in the context of climate change is code for more floods, droughts, natural disasters and all else that might entail (including more deaths and property loss). You can get a flavor for this if you go to the website of Joe Romm, Krugman’s climate change guru, which has a testimonial in the sidebar from Krugman — “I trust Joe Romm on climate”. [Romm’s website is at climateprogress.com. Once there, click on the link to “Extreme Weather”. Once you wade through this you may get a flavor for what an increase in extreme events implies in warmist circles.]
4 (b): And where is the data where you have taken increased ability to adapt to extreme weather events into consideration?
RESPONSE: I don’t recall saying anything about adaptation in this post, but no matter, since my graph is based on empirical data, it automatically incorporates whatever adaptation measures that populations may have availed themselves of. But because adaptation is automatically accounted for, I do not equate the empirical declines in deaths and death rates from extreme weather events to a decline in their frequency or intensity.
Incidentally, the above figure includes deaths from droughts, floods, wildfires, slides, heat waves, cold waves, cyclones and other storms.
5: The same goes here. He never mentioned anything about property losses.
RESPONSE: See response to 4(a), above.
BTW, what I was attempting to do in points 4 and 5 was to suggest to the reader that regardless of whether extreme events were becoming more frequent or more intense, in the real world human beings were making them less consequential. In other words, the issue of whether the frequencies and/or intensities were increasing (or not) were really a tempest in a shrinking teapot.
Indur M. Gloklany
But if, under global warming the increase in temperatures in the polar regions exceeds the increase in the equatorial areas, the temperature differential should decline and winds should diminish. >>>
Bravo! But I beat you to by a single comment and provided the physical basis at the daily, seasonal, and latitude levels as well. This to me is the real argument. The science says extreme weather should diminish in a warmer world. I keep seeing the warmist assumption of the opposite going unchallenged, and I have yet to see a reasoned scientific explanation to support the assumption. And I don’t think there is one.
davidmhoffer
As a one time electrical engineer I liked the voltage analogy.
Indur – thanks very much for the reply.
You said that two of your five points (#4 and #5) were not intended to actually rebut anything Krugman said, but rather to suggest to the reader that:
regardless of whether extreme events were becoming more frequent or more intense, in the real world human beings were making them less consequential. In other words, the issue of whether the frequencies and/or intensities were increasing (or not) were really a tempest in a shrinking teapot.
That helps to clarify what was going on – it was clear that #4 and #5, as they stood, did not really engage Krugman’s point, which was that if average temperatures increase, the frequency of “extreme events” will increase. It seems you didn’t mean to engage that point, and meant instead to argue that even if Krugman is right, it wouldn’t matter much.
No doubt you agree that all else being equal, increased frequency of extreme events will cost lots of money and lots of lives (i.e. compared to the money and lives that would be lost without that increase in extreme events)?
If you grant that (seemingly uncontroversial) ceteris paribus claim, then there seems to be only a Pickwickian – and a bit perverse – sense in which we’re talking about a “shrinking teapot,” don’t you think?
Furthermore, the cogency of your points #4 and #5 depends also on how much you think we should care (granting for the sake of argument, as you do in #4 and #5, that Krugman is right) about this “shrinking teapot.” A quick perusal (perhaps you know some better sources) indicates that typical annual losses due to extreme weather events is in the multiple tens of billions (US dollars), and tens of thousands of lives.
I’m not sure what you would consider “consequential” costs in human life, capital, development, or opportunity, but given that we’re not talking trivial numbers, why characterize such costs as unworthy of the concern which you believe Krugman enjoins us (in “code”) to have?
Um, this was Krugman’s approach. We are merely commenting on his approach. How is jorgekafkazar oversimplifying anything?
davidmhoffer says:
“A warmer earth means a more stable earth. A warmer earth means less extremes in temperature differentials, and hence less extremes in all the weather processes derived from them. Which is all of them… The science says extreme weather should diminish in a warmer world. I keep seeing the warmist assumption of the opposite going unchallenged, and I have yet to see a reasoned scientific explanation to support the assumption. And I don’t think there is one.”
I can understand how your simplified view would work if the temperature increase was universally consistent throughout the system, but it doesn’t happen that way. The temperature increase is initially localized and must diffuse from there.
If the warming is being driven from the atmosphere down, the temperature difference between the ocean depths and the troposphere is initially going to grow and remain larger for a long time (from the perspective of a human life). The heat capacity of the atmosphere is nothing compared to the oceans. You can’t expect a 1 C temp increase in the troposphere to translate into a 1 C temp increase to the ocean depths for decades.
So any weather patterns that depend on deep ocean upwellings of cool water could indeed become more “extreme”. I know of at least one weather pattern that fits that bill. (hint: the name is Spanish)
jorgekafkazar says: Yes, and remember that all the points on the curve are radiating to space by T^4. Those on the right end are radiating a LOT more than those on the left. That curve is VERY resistant to being shoved to the right, and the farther a point is to the right, the harder it is to move it further. Thus the curve will get narrower as it heats up.
eadler says: I think you are missing the effect on spatial components and oversimplifying the issue. The warming of the Arctic has impacted the Arctic Oscillation, pushing it into negative mode more often. This has created more extreme snow events in parts of the Northern Hemisphere in the winter time, including the US.
And I think you are missing the fact (or desperately trying to) that it was Krugman who oversimplified the issue. My statements are correct, and your red herring has neither addressed them nor disproven them. You’re a little late parroting the “extreme snow events” meme. If you wanted to make a valid contribution to climate science, the time to have spoken up was 11 years ago:
http://www.independent.co.uk/environment/snowfalls-are-now-just-a-thing-of-the-past-724017.html
You’re correct, I should have been more careful in phrasing my criticism. I didn’t bother to read Krugman’s actual article and instead relied on what I remembered him saying elsewhere.
On the basis of what he’s said elsewhere, his definition of an extreme event is not as limited as it was in this article. There he says that extreme events equate to catastrophes like droughts. For example, “Kalpa” writes, at http://bigpictureagriculture.blogspot.com/2011/02/debunking-krugman-nyts-soaring-food.html , this article: “Debunking Krugman: NYT’s “Soaring Food Prices – Blame the Weather””, which contains the following:
Incidentally, in place of the opaque abbreviation FSU, and the long-winded term “former Soviet Union,” I suggest the short and self-explanatory “XSSR.”
Indur Goklany wrote:
BTW, what I was attempting to do in points 4 and 5 was to suggest to the reader that regardless of whether extreme events were becoming more frequent or more intense, in the real world human beings were making them less consequential. In other words, the issue of whether the frequencies and/or intensities were increasing (or not) were really a tempest in a shrinking teapot.
In terms of immediate fatalities you are correct, but you are responding to an argument that is not being made, by Krugman or others who are concerned about the impact of global warming/climate change.
The problem now is habitability of areas impacted by drought and floods, combined with population increases. Climate refugees are going to be more numerous, and this will increase misery and instability. Problem areas are evident in Africa and Asia already.
eadler;
Warming will substantially expand habitable areas, lessen extreme events, and even green the deserts (currently the Sahel/Sahara is shrinking fast). So, per usual, none of what you say is even facing the same hemisphere as the truth.
Why would climate refugees be more numerous? The number of climate “events” will be the same.
Why are you so slow to grasp this concept?
Wobble: eadler was replying to Goklany’s points 4 and 5, in which Goklany has made clear that he was assuming, for sake of argument, that Krugman may be right about an increase in the frequency of extreme climate events, but denying that such an (hypothetical) increase would be problematic.
Eadler was suggesting that, given the same premise which Goklany accepts (for sake of argument), an increase in extreme events would be problematic.
So, as you can see, your reply to eadler isn’t at all on point, since you simply deny the hypothetical premise that is not in dispute between Goklany and eadler (at least in this particular context). Make sense?
(Goklany argues: “If X is true, it’s no big deal.” Eadler replies: “If X is true, it is a big deal.” You come along and say: “X is false; so you’re wrong, eadler.” But that doesn’t follow, for eadler was here just talking about the relation between X and its likely consequences, just as Goklany was. It would be like trying to “refute” Goklany’s points 4 and 5 by saying “but Krugman is wrong!” That wouldn’t refute his points at all, don’t you think?)