Climate Science Glossary

Term Lookup

Enter a term in the search box to find its definition.


Use the controls in the far right panel to increase or decrease the number of terms automatically displayed (or to completely turn that feature off).

Term Lookup


All IPCC definitions taken from Climate Change 2007: The Physical Science Basis. Working Group I Contribution to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change, Annex I, Glossary, pp. 941-954. Cambridge University Press.

Home Arguments Software Resources Comments The Consensus Project Translations About Donate

Twitter Facebook YouTube Pinterest

RSS Posts RSS Comments Email Subscribe

Climate's changed before
It's the sun
It's not bad
There is no consensus
It's cooling
Models are unreliable
Temp record is unreliable
Animals and plants can adapt
It hasn't warmed since 1998
Antarctica is gaining ice
View All Arguments...

Keep me logged in
New? Register here
Forgot your password?

Latest Posts


Climate Hustle

Linking Extreme Weather and Global Warming

Posted on 18 June 2011 by Anne-Marie Blackburn

Whenever there is an extreme weather event, such as a flood or drought, people ask whether that event was caused by global warming. Unfortunately, there is no straightforward answer to this question. Weather is highly variable and extreme weather events have always happened. Detecting trends takes time, particularly when observational records are rare or even missing in certain regions. An increase in extreme weather is expected with global warming because rising temperatures affect weather parameters in several ways. Changes in the frequency of extreme events coinciding with global warming have already been observed, and there is increasing evidence that some of these changes are caused by the impacts of human activities on the climate.

How global warming affects weather parameters

Rising temperatures can have several effects on the factors involved in weather. For example:

  • They increase the rate of evapotranspiration, which is the total evaporation of water from soil, plants and water bodies. This can have a direct effect on the fequency and intensity of droughts.
  • A warmer atmosphere can hold more water vapour. The atmosphere now holds 4% more water vapour than it did 40 years ago as a result of increasing temperatures. This increases the risk of extreme rainfall events.
  • Changes in sea-surface temperatures (SSTs) also have an effect by bringing about associated changes in atmospheric circulation and precipitation. This has been implicated in some droughts, particularly in the tropics.

These changes don't automatically generate extreme weather events but they change the odds that such events will take place. It is equivalent to the loading of dice, leading to one side being heavier, so that a certain outcome becomes more likely. In the context of global warming, this means that rising temperatures increase the odds of extreme events occurring.

Changes in extreme weather events are already being observed

In the US, the Global Changes Research Program published a report in 2009 entitled Global Climate Change Impacts in the US. The National Climate Change chapter reports the following findings for recent decades:

  • Heavy rainfall events have increased both in frequency and in intensity by 20%, and are the main cause behind the increase in overall precipitation in the US. The Northeast and Midwest have seen the greatest increase in such events.
  • The frequency of drought has increased in areas such as the Southeast and the West, and decreased in other areas. Rising temperatures make droughts more severe and/or widespread, and also lead to the earlier melting of snowpacks, which can exacerbate problems in vulnerable areas.
  • Atlantic hurricanes have increased both in power and frequency, coinciding with warming oceans that provide energy to these storms. In the Eastern Pacific, there have been fewer but stronger hurricanes recently. More research is needed to better understand the extent to which other factors, such as atmospheric stability and circulation, affect hurricane development.

Similarly, Australia has seen the odds of both heavy rainfalls and droughts increase, and similar patterns are being observed worldwide, coinciding with rising temperatures over the past 50 years.

In conclusion, although it isn't possible to state that global warming is causing a particular extreme event, it is wrong to say that global warming has no effect on the weather. Rising air and sea temperatures have a number of effects on the water cycle, and this increases the odds for more extreme weather events.

NOTE: this is the Basic rebuttal to "Extreme weather isn't caused by global warming"

0 0

Bookmark and Share Printable Version  |  Link to this page


Prev  1  2  3  4  5  Next

Comments 51 to 100 out of 236:

  1. What are the latest findings on how the ENSO is being impacted by climate change?

    What's the most recent SkS article on this issue?
    0 0
  2. Badgersouth @51, a search on google scholar for the terms ENSO & ("global warming" v "climate change") 2009 or later produces the following articles of interest:


    Latif and Keenlyside, "El Niño/Southern Oscillation response to global warming", PNAS 2008

    Collins et al "The impact of global warming on the tropical Pacific Ocean and El Niño" Nature Geoscience 2011

    Both indicate that models disagree about the effects of global warming on the ENSO.


    Gergis and Fowler "A history of ENSO events since A.D. 1525: implications for future climate change" Climatic Change 2008

    Gergis and Fowler indicate a significant anthropogenic effect on El Nino intensities:

    "Although extreme ENSO events are seen throughout the 478-year ENSO reconstruction, approximately 43% of extreme and 28% of all protracted ENSO events (i.e. both El Niño and La Niña phase) occur in the 20th century. The post-1940 period alone accounts for 30% of extreme ENSO years observed since A.D. 1525. These results suggest that ENSO may operate differently under natural (pre-industrial) and anthropogenic background states."

    Doran et al "Analyzing ENSO Period Changes in a Proxy Record Spanning the Last Millennium" Geophysical Research Letters 2009

    Doran et al indicate a change in the frequency of ENSO shifts:

    "The time series model analysis of the Palmyra corals shows ENSO period changes
    that are similar to the broad patterns observed in the wavelet analysis (Figure 2). ENSO
    periods are generally higher during the pre-1850 interval (μ = 6.4 years, σ = 1.9 years, N
    = 9), while periods post-1850 are lower (μ= 4.0 years, σ = 0.60 years, N = 4). This 2.4-
    year decrease in ENSO period from pre-1850 to post-1850 is statistically significant (p <
    0.1). In particular, three of the four estimated ENSO periods in the post-1850 interval are
    at or below the lower limit of periods observed pre-1850. Additionally, there is a shift in
    ENSO period at 1700 that is more pronounced than the shift observed at 1850. This
    earlier shift may reflect changes in forcing, such as increased volcanic activity..."

    Li et al "Interdecadal modulation of El Niño amplitude during the past millennium" Nature Climate Change 2011

    Li et al conclude that warmer conditions are correlated with enhanced ENSO variability, a finding in agreement with Doran et al.

    I am making no claim to expertise, so I can't say that these studies are representative, and for most of them I have read only the abstract so I can't comment on their validity. They are, however, consistent with the evidence as I have seen it summarized.
    0 0
  3. Further to my 52, here is a bar chart from Gergis and Fowler showing the number and intensities of El Nino and La Nina events for each century over the indicated period. The lack of weak events in the 1500's is probably a function of reduced proxy number and quality in that period, but through out the rest of the record, weak events vary in approximate proportion with strong events, and strong events increase in frequency in the 20th century, the opposite of what we would expect if the change in frequency was an artifact of the proxy record.

    What the data shows is that with warmer conditions, both El Nino conditions and La Nina conditions become more frequent, and that the stronger events become more frequent relative to the weaker events. Also, the increase of frequency of El Nino events is slightly greater than that for La Nina events.

    0 0
  4. Sphaerica,
    More data showing the relationship between winter temperature and snowfall. This is mainly due to more days above freezing.

    Does anyone have any data showing an increase in snowfall with an increase in winter temperatures?
    0 0
  5. Tom (#53), that might be suggestive as the authors state, but post 1970's when AGW was a clearer factor the El Nino & La Nina sample size is very small (can't rule out coincidence).
    0 0
  6. Does anyone have any data showing an increase in snowfall with an increase in winter temperatures?

    You're looking too narrowly. The big nor'easters that hit New England are sorta hurricane wannabees that start in the Atlantic off the SE US coast or further south.

    They then roar north and dump snow when they hit arctic air sitting over NE (or further south).

    It doesn't matter if winter temps increase, as long as they're low enough for snow to fall.
    0 0
  7. Eric (skeptic) @53, looking at Gergis and Fowler, and regardless of when the anthropogenic signal was strongest, the entire twentieth century has been unusually warm relative the the preceding 375 years, and the period since 1940 has been unprecedented in the last 485 years for the tropical Pacific, which was quite cool during the MWP. Consequently the fact is that 43% of extreme events occur in the warmest 21% of the time reconstructed, and that 30% of extreme events occurred in the very warmest 12.8% of the time.

    Not only that, but the coldest period, also experienced unusually low numbers of extreme events, particularly El Nino's relative to other periods:

    "Our results are consistent with the conclusions of D’Arrigo et al. (2005) concerning solar/temperature related modification of ENSO
    behaviour. Periods of inactivity were identified throughout the record; most notably, during the 1600s ENSO appears to have weakened, coincident with the height of
    the commonly defined Little Ice Age (∼A.D. 1550–1850) and Maunder Minimum (∼A.D. 1645–1715) epochs. Hendy et al.’s (2002) coral Sr/Ca SST reconstruction shows northern Australian SSTs 0.2◦ to 0.3◦C cooler than the long-term average between 1565 and 1700. This corresponds to enhanced La Niña activity in our reconstruction (1520s to 1660s) and a period of relative El Niño quiescence from the 1600s to 1780s."

    Consequently it is grossly misleading to suggest that this evidence is based on a correlation over just 30 years; and while, as is true of all empirical evidence, it could be just the result of an accidental correlation, it is likely (>66%) that the correlation between increased tropical SST and increased ENSO activity and intensity is not coincidental. In fact, where it not for the fact that we clearly do not fully understand the mechanisms involved, I would say it is very likely (>90%). If you wish to claim that is not a fair range on our certainty, you need to present some substantial evidence.
    0 0
  8. Tom Curtis @ 52 - I'm going to cover a couple of those papers in the next chapter on Amazon drought. The key point, as you mention, is that ENSO variability depends on the mean state of the tropical Pacific.

    When the tropical Pacific is cool (as in the MWP) the frequency/intensity of ENSO declines (although there still can be some pretty wild swings). When the tropical Pacific is warm then ENSO ramps up. That's what the proxies indicate.
    0 0
  9. With regard to cold Winters, I haven't noticed anything to do with abnormally cold temperatures in the Southern Hemisphere so far his year. I remember that the last couple of years had various local or regional unusual cold periods down there (especially in South America) which were highlighted by some posters as an indication of a coming 'ice-age' but I am more interested in whether there is some sort of link between cold spells in the Northern and Southern Hemispheres, because the UK had two cold spells over the last two Winters too.
    If there is a link, I look forward to a nice mild Winter again. Or, at least (because it was only really December that was cold here), a milder run-up to Christmas.
    If there is no link, who knows what the Winter will be like !
    0 0
  10. JMurphy -
    "I haven't noticed anything to do with abnormally cold temperatures in the Southern Hemisphere"

    Brisbane, Australia has been cold by all accounts. Just ask John Cook!

    In contrast, just across the ditch here in northern New Zealand it was 21 °C yesterday and nudging 20°C today. In the thick of winter! Looks like they're getting the "normal" cold further south in NZ though.
    0 0
  11. Thanks for the info, Rob. Seems to be isolated areas, rather than whole regions or countries, so far ? Any idea what it was like this time last year, or the previous one - if this is not too much off-topic.
    0 0
  12. Generally, La Nina (now finished I guess) gives NZ wetter and warmer weather, and El Nino; dry and cool. Last winter was slightly above average, temperature wise. You can find the data at NIWA, but they don't compare the months to the long-term records, which is a bit useless.
    0 0
  13. 54, Eric,

    Again, your latest study is simply looking at short time frames (1949–2005) in one locality (U.S.) in a regime that really predates the bulk of climate change (since we haven't even gotten there yet!). No matter how many papers you trot out demonstrating a snowfall pattern in the past 30-60 years, they're not going to be representative of snowfall changes over longer periods and more dramatic climate swings.

    Did you read up on Hadley Cells (to understand one important mechanism that underlies the problem)?

    Here's another thing to consider... a NY Times article on a scientist's proposition that climate change could cause colder North American winters (and with them, more snow further south than has been usual).

    My main point is that the problem is more complex than any simple kitchen logic will resolve, and there will be different answers for different regions because wind, circulation, and precipitation patterns may all change in unexpected ways, leading to more or less snowfall in various areas.
    0 0
  14. Norman @ 25 and fellow 'skeptics' and contrarians,

    You continue to cherry pick and ignore the overwhelming and collective body of evidence out there. And one paper for one region of the planet does not make your case concerning a global phenomenon, that is why we look at the body of evidence across many disciplines (a.k.a consilience).

    I, and others, have repeatedly directed you to papers which have objectively and quantitatively found an increasing trend in heavy rainfall, droughts and heat waves around the globe. Indeed there is already research, (see here) suggesting that the hydrological cycle may be accelerating/intensifying, as work by suggests.

    This thread has become yet another example of "skeptics" demonstrating that they are not at all interested in the facts and truth, but deliberately seek out those data and interpretations of data that suite their preconceived bias. That is not true skepticism, and it is not good science.

    And talking of extreme events, I would be surprised if "skeptics' were to try and argue that the recent dramatic and accelerating loss of Arctic sea ice doe snot constitute an extreme event in its own right. In fact, that assertion is supported by the findings of Polyak et al. (2010).
    0 0
  15. Albatross:
    The link you posted to river trends is really about worthless.

    I can only suggest that you research the long term Mississippi River basin discharges. You will find that there is a cycle on a long term basis.

    The same applies to the Nile River.

    People are suggesting extreme events are a somewhat recent occurance? How about the floods of the Mississippi River in the 1920's? The recent floods did not break those records. Or the heat waves and fires in Russia when Tchaivosky was creating beautiful music.

    There have been extreme weather events always. That is the nature of a chaotic climate, which we have.

    I have not read reliable data from any source that we are experiencing anything that is abnormal.

    I am very interested in facts and truths. I am not interested in inuendo.

    As far as Arctic is a long term perspective. A Truely climatic perspective.

    The Nares straight opened yesterday, and that will allow ice to freely leave the Arctic. My prediction for Arctic Ice for 2011 summer low still stands at 3.50. Is this an extreme? Yes, on a short term basis it is, but on a climatic basis it appears that it isn't.
    0 0

    [DB] Your linked source is missing p. 114.  Also, most ice export from the Arctic Ocean is through the Fram, not the Nares.  And if you base your closing sentence upon your linked source, then you are reading into it more than what it says.  Considering that this is about the 6th or 7th time you've posted it, you may want to make sure your understanding of it is up to par.

  16. Tom, my evidence is the text in bold in #57 which do not lend strong support for the "extreme El Nino from AGW " hypothesis. The statistics you quote in bold are completely insufficient to reject the null hypothesis: that extreme ENSO is natural. In the paper you quoted "during the 1600s ENSO appears to have weakened, coincident with the height of
    the commonly defined Little Ice Age" (my bold) there is no discussion of cause and effect in the cold period: some of the cold was due to lack of El Nino and that lack was due to longer term dynamic trends.

    For those trends, see this paper where they state "The Holocene is characterized by increasing ENSO event frequency towards the present." That is a natural trend and there in no suggestion in that paper that warming causes El Nino although they don't cover extreme El Nino. Also I don't see a quantity expressed for percentage of extreme El Nino in Gergis and Fowler. If extreme El Nino is 10%, then that is much weaker support (for the AGW - extreme El Nino hypothesis) than 1% or 0.1%
    0 0
  17. DB:
    Yes, I know the Fram is more important than the Nares. But the Nares is open this year much earlier than normal, so it is allowing one more exit path for the ice.
    0 0

    [DB] THe Nares Strait ice bridge is now gone but the Strait itself is still blocked.  At this time in 2010 the Strait was indeed open but not at this time this year.

  18. Camburn
    "People are suggesting extreme events are a somewhat recent occurance?"

    Nope, that is you implying others are.

    "How about the floods of the Mississippi River in the 1920's? The recent floods did not break those records."

    A single event in the past is not a signature of global warming. If the frequency of past events was higher then it would be. Hence your example(s) is fiction.

    "I have not read reliable data from any source that we are experiencing anything that is abnormal."

    A warming climate is abnormal and no sensible person is denying that the climate is warming. Glacial melt, changing seasons and numerous other changes are a very clear indicator.
    0 0
  19. Tom Curtis @57

    Here are some thoughts to consider with your post on the ENSO topic where you are demonstrating an increase in intensity of events that seems to correlate with Global Warming.

    Does Correlation equal Causation?.

    Some expamples that seem to Correlate.

    My main point questioning the linking of more extreme weather patterns to Global warming is that just coming up with a correlation is not good enough especially in something as complex as weather where many factors come into play in determining a weather event.

    This tactic is creating the huge denier syndrome that so many on this site complain about. It is not making people concerned and want to take action, it is generating the opposite effect. It closely resembles the reasoning of some on the Conspiracy sites. Every terrible natural event is caused by H.A.R.R.P. (all weather...floods, hurricanes, tornadoes, blizzards, droughts and even earthquakes). As a scientific minded individual I do not blantanly say this is wrong thinking, maybe it is HAARP, the possiblitly exists. I ask these people for a link, a mechanism a way HARRP can cause all these events. Anger such questions is the response.

    If you want to prove Global warming is causing more extreme ENSO events then you need to find a link or mechanism that directly couples these events.

    It is like AGW theory. The correlation is that CO2 is rising at the same time Global Temperatures are showing a rise. Without a possible linking mechanism it will never be an acceptable theory. Science does have such a mechanism and link. The link is that it can be demonstrated that CO2 absorbs certain bands of IR energy. And in the global arena sattelites can see reduction in IR energy at these bands and ground based sensors can pick up downwelling longwave radiation at CO2 band frequencies and measure the watt/meter of this radiation. So the observed correlation proves to have empirical links and can be considered a legitimate conclusion. With the extreme weather events only some nebulous and questionable correlations are presented without corresponding linking mechanisms.

    Many exterme weather events are caused by blocking patterns (will form heat waves, droughts, and flooding as storm system follow the same tracts as long as the block persists). Can you find or demonstrate a link that would explain how a warming planet will create more intense, greater numbers and longer lasting blocking patterns?
    0 0
  20. Albatross @64

    I checked out the link back to a Skeptical Science article that attempts to prove that preciptiation and river flows are increasing.

    In this same link did you bother to read Charlie A's responses to the author's claims? Post 41,42 and 44?

    In his 41 post, Charlie A actually post a graph which shows the opposite effect and would give a different conclusion.
    0 0
  21. It's also looking as if the US Spring weather is going to be the most extreme to date.
    The UK is looking similar in many respects.
    0 0
  22. Norman:
    "With the extreme weather events only some nebulous and questionable correlations are presented without corresponding linking mechanisms."

    Extreme weather (harsher droughts and heavier rain fall) would be an indicator of climate change, not evidence. If the science of warming is established, then extreme weather purely adds to the list of indicators.

    There are enough changes that relate to weather, a far more prominent one is the change in timing of seasons, eg. starting earlier or later.
    0 0
  23. Paul D @71

    When you say that US Spring weather is going to be the most extreme to date what are you defining? Is it extreme tornado events, floods, droughts, snow (still snowing in some mountain locations). What other springs are you comparing it to?
    0 0
  24. Paul D,

    History of Texas droughts.

    I am seriously questioning the conclusion that extreme weather events are on the increase. While I am limited in time to research the globe I choose one place that is in the news and look at the historical information I can find to see if the news items are taken with a historical view in mind.

    I am "questioning" but not denying the possibility that global warming is leading to more extreme weather events. If linking mechanisms to can be found for given correlations than the credibilty of these claims improves. Predictability would be nice.
    0 0
  25. Paul D @66

    "I have not read reliable data from any source that we are experiencing anything that is abnormal."

    A warming climate is abnormal and no sensible person is denying that the climate is warming. Glacial melt, changing seasons and numerous other changes are a very clear indicator."

    Paul D what happened to the Greenland ice sheet 3200 years ago. Greenland was much warmer than as compared with now...Almost 3 C warmer than today! In fact it was warmer than today for about 3000 years. If our current warming phase is so abnormal and dangerous how did all that ice last through the heat for 3000 years?

    Greenland temps the last 10000 years.

    Why is it that looking at history of weather and climate you cannot find any examples that we are entering uncharted territory and things will get much worse.

    I guess it is because of that nasty weather manipulating and earthquake causing HAARP station in Alaska.
    0 0
  26. Paul D.

    Look at the Greenland temp graph link from the previous post. If you look at 8100 years ago to 8000 years ago you see a drastic temperature rise of 3 C in 100 years. I don't think our last 100 years is even close to this.

    Anyone out there, for looking at temperature extremes and drastic changes, can anyone find a link as to why Greenland's temps are so unstable?
    0 0

    [DB] There are many issues with the way disinformationists use GISP records.  See here and here for two of many posts addressing this.  If you're implying regionally variability in the past trumps current warming known to be global, then your understanding of the science is lacking.  The GISP data do not cover the current warming era.

  27. Eric @66, Gergis and Fowler state:

    "From Table 9, a total of 92 (82) El Niño (La Niña) events were reconstructed since A.D. 1525. Of these, 37 very-strong to extreme El Niño were observed. Nine El Niño events were classified as extreme, including five well-known events of the 20th century (2002, 1982–1983, 1941–1942, 1926, 1905). The 18th century contained a
    further three events (1737, 1723, 1718) while only one extreme El Niño event of A.D. 1650 was recorded during the 16th and 17th centuries. The 20th century represents
    the peak of El Niño activity, when twelve events were classed as either very strong or extreme. A total of 46 very strong to extreme La Niña events were reconstructed,
    twelve of these classified as extreme. Four extreme La Niña events (1998, 1974, 1953, 1950) were reconstructed during the 20th century. The pre-instrumental period indicates relatively more La Niña activity with 6 (24) extreme (very strong) events compared to 4 (18) extreme (very strong) events from the El Niño reconstruction. Considerable La Niña activity is indicated during the 16th tomid 17th centuries when five extreme events are reconstructed. A trend towards increased La Niña activity over the 20th century is also evident from Table 9."

    Thus over the entire 478 year period (1525-2002)of the reconstruction, the relative frequency of extreme events ENSO events is 0.044 over the entire period. Of these, seven occurred post 1939, giving a frequency for that period of 7/63. Given a relative probability of 0.044, the probability of achieving seven positive results in 63 trials is approximately 0.0142, which you are correct does not achieve the standard of statistical significance. It is also just a small part of the statistically relevant information from this study. Once you factor in the increased frequency of very strong events over the instrumental period, the increased frequency in the relatively warm 1991-1939 period, and the reduced frequency of very strong and extreme during the peak of the LIA, and particularly during the Maunder Minimum, I think your claim that the results are not statistically significant are very dubious.

    What is more, Doran et al (as quoted in 52 above do explicitly find that the change in frequency of ENSO events is statistically significant.

    With regard to Moy et al, I note that their conclusion of reduced frequency of El Nino events over the last 1200 years contradicts Gergis and Fowler, Doran et al, and Li et al above, who all find an increased frequency of ENSO events over the last thousand (Doran et al, Li et al) of 500 years (Gergis and Fowler). In view of the fact that Moy et al use a single proxy, their conclusions have to be called into question given that they contradict more recent multi-proxy studies (Li et al, and especially Gergis and Fowler). Is suspect more weight should also be given to Doran et al, even they also only use a single proxy, in that their proxy is a direct proxy of sea surface temperatures in the Central Pacific, ie, the location of ENSO events.
    0 0
  28. Norman, the robust prediction from climate theory on weather parameters is given in the article above (transpiration rate, more water vapour, higher SST). Do you doubt these parameter changes> Going from changes in these how this plays out on a regional weather is far more uncertain science. But do you seriously expect that you could change those fundamental parameters without effecting weather?
    0 0
  29. DB

    The Greenland temps was only a side argument for Paul D. My major question on this thread is to establish concrete links for what seems correlated weather data showing rising extremes.

    But I linked to Crux of the Core part 1 and looked at the Holocene temp graph. For 4000 years the temperature was warmer than it is today. And Greenland's temp is significant for the "melting away of its ice" concept. Greenland was 3 C warmer than today. I am reading that significant ice has been lost since 1979. If a 30 year period of warming can cause such a devestation to the Greenland ice sheet how could it withstand a much higher temp for much longer periods...there should be no Crux of the Core as the glacier should not have existed during those hot phases.

    So during that 4000 year warm period were the floods more devastating? More tornadoes, droughts, heat waves? How did man survive the weather extremes? Why didn't this 4000 year warm cycle melt the permafrost and release all the stored methane gas locked in methane hydrates?
    0 0

    [DB] You may want to study GISP more deeply.  IIRC, the chart shows no data more current than 1895 or so.  Thus the lack of representation of the modern warming era.

  30. Norman and Camburn,

    I'm trying to enjoy Father's day, but your posts are detracting from that ;)

    I'll respond tomorrow.
    0 0
  31. Norman @69, I quite agree with you that mere correlation is not that strong as evidence. In fact, I alluded to exactly that point when I wrote:

    "... it is likely (>66%) that the correlation between increased tropical SST and increased ENSO activity and intensity is not coincidental. In fact, where it not for the fact that we clearly do not fully understand the mechanisms involved, I would say it is very likely (>90%)."

    However, it is not true that climate scientists have merely come up with a correlation.

    The ENSO is a consequence of the walker circulation, a atmospheric circulation that creates an east-west wind at the surface parallel to the equator. That circulation is in turn cause by the temperature gradient between the warm shallow water in the western tropical pacific and the colder water in the eastern tropical pacific which is cooled by the Humboldt current. The walker circulation carries the cold water from the east across the surface of the central pacific, thus cooling the central pacific. If the walker circulation is particularly strong you will get a La Nina, if particularly weak you get an El Nino.

    The oscillation is introduce because once the cold water is carried east, it both reduces the outgoing surface radiation and evaporation from the surface, thus reducing cooling; and exposed a greater volume of cool water to the sun, thus enhancing warming. The consequence is that the cool water carried across the surface is warmed, weakening the walker circulation.

    In these circumstances, prima facie, warmer tropical temperatures will result in the cold tongue of water from ENSO warming more rapidly as it mixes in, thus increasing the rate at which the ENSO switches from La Nina to El Nino conditions. Enhanced tropical warming will also increase the temperature gradient between eastern and western tropical pacific, increasing the rapidity of a switch from El Nino to La Nina. The increased rapidity of switching is itself likely to enhance the intensity of some events in what is essentially a forced oscillator.

    So, this is not a case of an inference from pure correlation, it is an example of an intuitive mechanism which cannot yet be shown to be supported by theory due to complicating factors, along with a clear correlation predicted by the intuitive mechanism.

    Having shown your objection to be ill founded, I also reject in the most strenuous terms your insinuation that

    "This [supposed] tactic [of using correlation without mechanism] is creating the huge denier syndrome that so many on this site complain about. It is not making people concerned and want to take action, it is generating the opposite effect."

    First, I object because it is not a note worthy feature of climate science that simple correlation without some mechanism is used to "prove" anything. Even such ill understood phenomena as blocking patterns are not simply attributed to global warming analysed to determine if there are causal mechanisms involved (note the date on the press story - it is before the Russian heatwave of 2010). Because of the nature of the science, some of the mechanisms are heavily disputed within the scientific community, but those who dispute the mechanisms also dispute the attribution of (changes in frequency or intensity of) particular types of events to global warming.

    Second, I object because reliance on pure correlation is a note worthy tactic of deniers. And people who are so ready to use such a tactic can hardly have been put of by the supposed use of that tactic by others. That is particularly the case in that, as noted, the tactic is often attributed to, but not in fact significantly used by climate scientists.
    0 0
  32. Sphaerica,
    You avoided my question. Do you have any data showing an increase in snowfall with rising temperatures.

    Also, that is the first that I have ever read that glocal warm will cause colder winters. Most have winter temperatures rising the most.
    0 0
  33. Tom thanks for the elucidation, you are right about the single proxy, and Moy et al talk about the potential problems with it, but mostly for 7k or more years ago. Their long term (7k) trend presents a powerful argument for a natural increase in ENSO event although not necessarily extreme events. That is notwithstanding the lower frequency in the last 1200 years.

    The other statistics in Gergis and Fowler are interesting but IMO are not direct evidence against the null hypothesis which is (to restate a bit) that the recent extreme El Ninos in 82-83 and 97-98 were not due to AGW warming kicking in earnestly after the 70's. I would argue that the reduced ENSO extremes in the LIA have potential confounding factors (solar -> weather patterns -> winds and/or clouds) that are not relevant to the post 70's warming.

    An alternative to the hypothesis testing from empirical data above is to describe a physical link from the AGW to the ENSO extremes. ENSO is the response of an oscillatory system from effects that are secondary to warming, e.g. winds. But it is conceivable that the warming changes the system's response such that the peak of the oscillation is increased over normal.
    0 0
  34. 82, Eric the Red,

    Sorry, Eric, I didn't realize you'd asked a question.

    But my answer would be that the search for an increase or decrease in current observations is irrelevant, because temperatures haven't been elevated long enough, and the increase has not yet been great enough, to say what the effects will be for any particular region merely through observation.

    What is needed, instead, is an understanding of what mechanisms will be affected, and what the implications are going to be.

    But then, that in a nutshell is most of climate science. Figure out what's gong to happen, by fitting all of the pieces together.

    I'm afraid I don't agree with your method, which appears to be to basically to wait until something unequivocal actually does happen.

    But, with that said, the trend line for northern hemisphere winter precipitation is slightly positive (and if you look at it from 1990 on, it seems more strongly positive), and spring is strongly negative.

    Rutgers Northern Hemisphere Winter Snow Extent

    Rutgers Northern Hemisphere Spring Snow Extent

    On your comment about colder winters... you're incorrectly comparing a specific, local phenomenon with a global mean prediction. Yes, overall, on average, winters will get warmer. Winter temperatures in total will be warmer. The reference I made was to a specific case where North American winters will still be shorter, but may paradoxically be colder because the much colder Arctic air (which is warmer than normal, but still very cold by North American standards) may frequently get pushed south over the continent.

    This is been my main point all along... that the devil is in the details. You can never, ever take a very simplistic approach (more humidity must mean more precipitation, or warmer winters must mean less snow) and apply it across the board. You simply can't.
    0 0
  35. Eric (skeptic) @83, with respect, "...the recent extreme El Ninos in 82-83 and 97-98 were not due to AGW warming kicking in earnestly after the 70's warming" cannot be the null hypothesis in this case. This is for two reasons.

    First, the simple denial of a hypothesis cannot be the null hypothesis for any statistical test. A test of statistical significance tests against the probability that the test statistic being the result of the null hypothesis, and to do that you must have a defined value for the probability that an event will occur given that the null hypothesis is true. To determine the probability of the denial of a given hypothesis, however, you must know the probabilities of the events under all logically possible contrary hypotheses. As the number of possible contrary hypothesis is infinite, you cannot even enumerate them, let alone sum their conjoint probability. Hence the denial of a hypothesis can never be a null hypothesis because you can never define the probability of the event given that the hypothesis is false.

    Second, the null hypothesis must always be a purely statistical hypothesis. This is because you cannot observe causation. You can only observe correlations. If you use a causal null hypothesis, you could have a situation of perfect correlation over tens of thousands complexly varying observations and still not find statistical significance because there may be some as yet unknown alternative cause.

    Taking this into mind, a sensible null hypothesis is that:

    The probability of a high intensity ENSO event is independent of mean tropical pacific heat content.

    Using that null hypothesis, the probability of the observations of extreme ENSO events observed post 1940 is around 15%. It is highly likely that the probability of the total set of observations in Gergis and Fowler given that null hypothesis is less than 5%, and hence that there is a statistically significant correlation between mean tropical pacific heat content and the intensity of ENSO events. Of course, "global warming" causally predicts increased tropical ocean heat content, so the test is not arbitrary with respect to global warming - but what is tested statistically is statistical, not causal correlation.

    Finally, the feature of Moy et al that is challenged by the other three studies is the correlation between their chosen proxy and ENSO states. If the other three studies are correctly reconstructing ENSO variability, then the ENSO reconstruction of Moy et al over the last thousand years has no skill. Therefore we have no reason to trust it for the seven thousand years before that. But the other three studies either us multiple and geographically diverse proxies (the entire US for Li et al, up to 14 proxies with global coverage and with ten of those being in the Pacific or Pacific rim for Gergis and Fowler), or use proxies directly related to the ENSO by geographical location rather than climactic correlation (Doran et al). Consequently it would be very dubious to accept Moy et al's reconstruction as being more reliable than any of the other three, and certainly not more reliable than the conjoint signal of the other three.

    It is possible that closer examination will show the discrepancy between Moi et al and the other three to be more apparent than real, but until that closer examination is made, relying on Moy et al to contradict the other three is believing against the evidence rather than with it.
    0 0
  36. Sphaerica,
    You seem to be backpedalling rather rapidly. I have not been able to determine if you are making any correlation between temperature and snowfall at all.

    You claim that temperatures have not risen long enough to determine a correlation, but do not want to wait until they do to determine one either.

    Lastly, my references to European snowfall and North American snowfall, while regional in each case, amount to most of the global mid-latitude snowfall. Yet, your refer to this as regional, then make the claim that your statement about colder winters was a specific case in North America.

    You seem to want to extend yout theories across the board without proof, yet deny someone's else even though all the evidence presented supports it. (-Snip-)
    0 0
    Moderator Response: (DB) Someone's profession has no bearing on their ability to formulate a cogent and coherent comment on this forum.
  37. h/t to Norman for this post which Eric and Sphaerica may be interested in:

    "Be that as it may, rather than take a party-line on snowfall accumulation, I rather just see what the science research and basic physics tells us, and ignore the models.

    In general, colder periods on earth were dry, not wet, and snowfall accumulation was lower not higher. The coldest place on earth, Antarctica is also one of the most dry in terms of precipitation. 100,000+ years of ice core data shows that when the planet began to warm, accumulation increases and when it begins to cool (as in going into a a glacial period, accumulation decreases. The thing about glacial periods is not that it is snowing much more, but rather, the spring and summers are much cooler so the snow doesn’t melt, and then when the next winter comes, it snows on older snow, and guess what, you get glacial growth. That’s the way it happens.

    Big snows in the winter don’t necessarily mean anything other than we’ve had a warm enough atmosphere to transport and hold all that moisture. It takes a lot of energy to move all that moisture and warm atmosphere to hold it while it is being moved. You don’t get big snowfalls EVER in the middle of Antarctica– basic physics tells you that.

    Now, as far as “predictions” by the AGW community as to whether there will be snowier or less snowier winters– yes, there will be. But in a system like the climate, that exhibits spatio-temporal chaos, it is a crap shoot to say when and where it will be snowier or less snowier and there is not a consensus opinion among the experts about any of this, but for some reason, AGW skeptics want to paint it that way."

    (From R Gates, a little ray of sunshine in a pit of darkness)
    0 0
  38. Norman, I assume your post was deleted for being off topic. I have therefore responded here where I believe it to be on topic. If you think I have mischaracterized your point, please respond there.
    0 0
  39. Norman:
    "When you say that US Spring weather is going to be the most extreme to date what are you defining?"

    It's not me that is defining it:
    0 0
  40. Norman:
    "I am "questioning" but not denying the possibility that global warming is leading to more extreme weather events."

    The issue isn't that extreme weather events are being observed and that they are caused by global warming. The issue is that global warming will cause extreme weather and the frequency will increase, if the science of warming is established and it is accepted that warming is taking place, then extreme weather is likely to be attributed to those changes because that is what we would expect.
    0 0
  41. Norman:
    "Why is it that looking at history of weather and climate you cannot find any examples that we are entering uncharted territory and things will get much worse."

    Because when you look to the past, you find individual events from history that were caused by regional inputs rather than global inputs.

    It's easy to pick events that are hundreds or thousands of years apart and say that they were not a result of rising greenhouse gases. It's more difficult to find events that occurred within a few years or months across the globe and attribute them to local and/or unconnected inputs.
    0 0
  42. Couple of interesting letters in Nature from this year:
    Anthropogenic greenhouse gas contribution to flood risk in England and Wales in autumn 2000
    Human contribution to more-intense precipitation extremes
    0 0
  43. A new paper to add to the SkS database?

    Reconstruction of false spring occurrences over the southeastern United States, 1901–2007: an increasing risk of spring freeze damage?

    Garrett P Marino1, Dale P Kaiser, Lianhong Gu and Daniel M Ricciuto
    0 0
  44. Tom,

    Nice post. I have also read were places like Mexico and the Sahara were driest during the coldest periods also.

    With regards to snowfall, the general rule is that the closer to freezing, the more snowfall. This is true for both places that are above and below freezing. In places like Antarctica, the temperatures are always below freezing, so any rise will bring the temperature closer to freezing, and subsequently more snowfall (still very little). The areas of Greenland which average slightly below freezing will also see more snowfall as the temperatures rise. Those regions where temperatures are above freezing for most of the year, will experience mroe snowfall as temperatures fall towards freezing.

    Single winter snowfall is indeed a crap shoot, depending on the temperature, jet stream, oceans, etc. Unfortunately, some of these single-year events have been used by many people (believers and non-believers like) to attempt to prove their position. One (or two) years does not a trend make. However, over a period of years or decades, subtle changes can be observed, like Antarctica.

    0 0
  45. Paul,

    Why would we expect an increase in extreme events under global warming?

    Granted, a warmer atmosphere can hold more water and lead to overall increases in rainfall. Conversely, global warming is expected to lead to lower temperature gradients (this has been observed during the last century). Can we say with any degree of certainty that the increasing water content will outweigh the decreasing temperature differences?
    0 0
  46. Tom, your null hypothesis is very reasonable but I have a quibble. How are we going to estimate mean tropical Pacific OHC for the past (i.e. before the 50's) while accounting for ENSO itself? Do you know of studies that do that (my quick search didn't turn up any)?
    0 0
  47. 86, Eric the Red,

    You seem to have a hard time following what I'm saying, so let me try to break it down, and keep it simple.

    My most basic position, all along, is that the logic that you are trying to apply to the problem is over-simplified, and as such grossly flawed. You must look at the details (as I said, "the devil is in the details") and understand things at a deeper level. If you don't do this, then you will perceive self-contradictions in my position, because you are only understanding what you think, instead of what I am saying.

    Main Point [And I've stated this clearly several times now]:

    The impacts of climate change are complex and varied. You cannot use oversimplified logic to draw vast, sweeping, winner-take-all conclusions.


    a) Just because global warming will increase the mean global temperature of the planet does not mean that all areas of the globe will experience warming at all times of the year.

    The specific example that I gave you from the NY Times article is a prime example of this, where a weakening of the polar vortex may allow the frequent intrusion of frigid air into the North American continent.

    For deniers, they must realize that evidence of cooling, even consistent cooling, or uneven warming, is not evidence against climate change.

    b)> Just because global warming will increase the overall moisture content of the atmosphere, this does not mean that every part of the globe must experience increased precipitation, year round.

    The specific example that I gave you of how Hadley Cells function to distribute moisture, and how they will be altered due to climate change, is a prime example of this, where an expansion of the Hadley Cells will widen arid areas, and move areas of major precipitation to new locales.

    Denier claims that major changes in precipitation patterns must mean more droughts or more storms, everywhere, year round, is clearly flawed logic and a simplistic distraction from the truth of the matter.

    c) Your presumption that increased temperatures must amount to less snowfall is similarly flawed. The temperatures will shorten the winter season, delaying the onset of meaningful snowfall, and quickening the onset of spring (and snow loss). This does not, however, preclude the possibility that there can actually be greater snow accumulations during the winter months, due to the increase in overall moisture content in the atmosphere.

    The specific example I gave you of the total snow extent in the northern hemisphere, from the Rutgers Global Snow Lab, is a prime example of this, where spring snow extent has decreased dramatically, while winter snow extent has shown a slight positive trend.

    d) Your attempts to prove that global warming will not affect winter snow accumulations is flawed in three respects.

    i) The first is that the time periods involved are too short, and cover to small a degree of warming to have any statistical significance relative to the idea of major climate change.

    ii) The second is that your studies all involve very specific locations. Trying to cobble together enough of these is much like the effort to define the MWP, but you're trying to do it with one here and one there. Unless you are ready to find hundreds of studies, and to "homogenize" them into something coherent, then it's a pointless effort.

    iii) The third and most important point is that we are talking about the effects on precipitation of a 3+˚C global mean temperature change. We have not yet seen enough of this change in climate to prove, one way or the other, what the total, global effects will be. No accumulation of papers that you can compile, covering the period from 1900-2000, is going to solidify your position on this (especially when many of those papers only run from about 1950-2000).

    Is this clear enough for you?

    Please try to read and comprehend this. There are no contradictions in my position. There is no backpedaling. The system is quite simply more complex than you seem willing to consider, and when you think that you see contradictions in my position, it is because your view of things is too simple. Dig deeper.
    0 0
  48. Eric (skeptic) @96, the simplest method would be by using a decadal mean of proxies for sea surface temperature outside the central pacific region most strongly effected by ENSO.
    0 0
  49. Sphaerica,

    It has always been clear. I am not having a hard time undersatanding you. We simply disagree.

    I stated that snowfall in individual winters cannot be predicted based on global warming, but that several other factors are involved. I do not know why you cannot get past that.

    I understand that you are relying more on models in your analyses, whereas I am using data. Granted, we do not have any data at temperatures of 2-3C higher than today, so a comparison cannot be made between the two. Models are a tool used to comprehend complex systems. However, that model must be able to reasonably correspond to the system data, otherwise it is as useful as a left-handed oil filter remover.
    0 0
  50. 99, Eric the Red,
    It has always been clear. I am not having a hard time undersatanding you.

    Hmmm. I'm not sure why you previously said "You seem to be backpedalling rather rapidly. I have not been able to determine if you are making any correlation between temperature and snowfall at all."

    Oh, well. I apparently misunderstood you.
    I stated that snowfall in individual winters cannot be predicted based on global warming, but that several other factors are involved.
    Excellent. We agree on that point 100%. Many factors are involved, some of which will be greatly influenced by global warming, and so it is an important problem that must be studied, and deserves both attention and deep concern.
    I understand that you are relying more on models in your analyses, whereas I am using data.
    This is not the case at all, and I have no idea where you got this idea (unless by "models" you mean "an understanding of the mechanics that define the system"). I never once used the word "models," or referenced them in any way. I merely feel that the scatter shot way in which you are trying to use the data combined with an overly simplistic "model" of the system is incorrect.
    Granted, we do not have any data at temperatures of 2-3C higher than today, so a comparison cannot be made between the two.
    Excellent. Then you will agree that an evaluation of what will happen based merely on what has happened to date is not feasible.

    Which leaves us to instead try to understand the system in detail, and to avoid trying to draw inferences from an overly simplistic approach to the problem.

    So, in the end, you agree with all of my points, that observations to date are no security against precipitation changes and extreme weather events, that increased snowfall in winter is not of itself a contradiction in climate theory, and that an increase in both droughts and extreme precipitation in different areas, or even the same area, at different times of year (be it spring, summer, fall or winter) are consistent with climate change.

    Now we can move on.
    0 0

Prev  1  2  3  4  5  Next

You need to be logged in to post a comment. Login via the left margin or if you're new, register here.

The Consensus Project Website


(free to republish)

© Copyright 2019 John Cook
Home | Links | Translations | About Us | Privacy | Contact Us