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 Support

Bluesky Facebook LinkedIn Mastodon MeWe

Twitter YouTube 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...

New? Register here
Forgot your password?

Latest Posts


10 Indicators of a Human Fingerprint on Climate Change

Posted on 30 July 2010 by John Cook

The NOAA State of the Climate 2009 report is an excellent summary of the many lines of evidence that global warming is happening. Acknowledging the fact that the planet is warming leads to the all important question - what's causing global warming? To answer this, here is a summary of the empirical evidence that answer this question. Many different observations find a distinct human fingerprint on climate change:

10 Indicators of a Human Fingerprint on Climate Change

To get a closer look, click on the pic above to get a high-rez 1024x768 version (you're all welcome to use this graphic in your Powerpoint presentations). Or to dig even deeper, here's more info on each indicator (including links to the original data or peer-reviewed research):

  1. Humans are currently emitting around 30 billion tonnes of CO2 into the atmosphere every year (CDIAC). Of course, it could be coincidence that CO2 levels are rising so sharply at the same time so let's look at more evidence that we're responsible for the rise in CO2 levels.
  2. When we measure the type of carbon accumulating in the atmosphere, we observe more of the type of carbon that comes from fossil fuels (Manning 2006).
  3. This is corroborated by measurements of oxygen in the atmosphere. Oxygen levels are falling in line with the amount of carbon dioxide rising, just as you'd expect from fossil fuel burning which takes oxygen out of the air to create carbon dioxide (Manning 2006).
  4. Further independent evidence that humans are raising CO2 levels comes from measurements of carbon found in coral records going back several centuries. These find a recent sharp rise in the type of carbon that comes from fossil fuels (Pelejero 2005).
  5. So we know humans are raising CO2 levels. What's the effect? Satellites measure less heat escaping out to space, at the particular wavelengths that CO2 absorbs heat, thus finding "direct experimental evidence for a significant increase in the Earth's greenhouse effect". (Harries 2001, Griggs 2004, Chen 2007).
  6. If less heat is escaping to space, where is it going? Back to the Earth's surface. Surface measurements confirm this, observing more downward infrared radiation (Philipona 2004, Wang 2009). A closer look at the downward radiation finds more heat returning at CO2 wavelengths, leading to the conclusion that "this experimental data should effectively end the argument by skeptics that no experimental evidence exists for the connection between greenhouse gas increases in the atmosphere and global warming." (Evans 2006).
  7. If an increased greenhouse effect is causing global warming, we should see certain patterns in the warming. For example, the planet should warm faster at night than during the day. This is indeed being observed (Braganza 2004, Alexander 2006).
  8. Another distinctive pattern of greenhouse warming is cooling in the upper atmosphere, otherwise known as the stratosphere. This is exactly what's happening (Jones 2003).
  9. With the lower atmosphere (the troposphere) warming and the upper atmosphere (the stratosphere) cooling, another consequence is the boundary between the troposphere and stratosphere, otherwise known as the tropopause, should rise as a consequence of greenhouse warming. This has been observed (Santer 2003).
  10. An even higher layer of the atmosphere, the ionosphere, is expected to cool and contract in response to greenhouse warming. This has been observed by satellites (Laštovi?ka 2006).

Science isn't a house of cards, ready to topple if you remove one line of evidence. Instead, it's like a jigsaw puzzle. As the body of evidence builds, we get a clearer picture of what's driving our climate. We now have many lines of evidence all pointing to a single, consistent answer - the main driver of global warming is rising carbon dioxide levels from our fossil fuel burning.

0 0

Printable Version  |  Link to this page


Prev  1  2  

Comments 51 to 99 out of 99:

  1. Damped the effect, AlanR? What effect are you speaking of? C02 is C02, excepting isotope markers which don't affect P-chem. If you have a system that has been cycling C02 in a rough state of equilibrium and you dump a large additional amount into that system, you'll see a bulge of unprocessed C02, failing the emergence of increased processing power. The IPCC refers to the lack of excess capacity in the existing system. In this case, once enough time has passed we'll see the bulge eventually diminish, ironically in part because of increased temperature in the system. If the system had previously had -excess- capacity we'd have seen a trend of decreasing C02 prior to the C02 glut we've imposed on the system, if the trend were short enough to have allowed us to develop the capacity to make such measurements. Easy enough. The remaining mystery is, was that actually a mystery?
    0 0
  2. Why Doug, the warming effects we noted at #47. ACO2 is CO2. I think so too. The IPCC models treat them differently. Easy yes. Mystery no.
    0 0
  3. "We," AlanR? Superficially speaking "we" have some hints that you are sufficiently motivated as to register here and ask questions in a way that suggests you are familiar with this topic but at the same time insufficiently curious to make an effort to find answers on your own, a seeming paradox. Or perhaps you are not curious, it's hard to say exactly without more data. Now you appear to be departing from the world of facts, presumably imagining "cooling" or the like in the face of overwhelming evidence to the contrary as well as sliding into the fuzzy world of semantics and rhetoric, which helps to confirm that further discussion of science with you is probably pointless. I could be wrong, of course.
    0 0
  4. I didn't expect my motivations to be examined here. One on't cross beams gone owt askew on treadle.
    0 0
  5. Timely arrival of new integration of anthropogenic C02 emissions w/carbon cycle: "What’s new about this research is that we have integrated the carbon cycle into our model to obtain the emissions data," says Erich Roeckner. According to the model, admissible carbon dioxide emissions will increase from approximately seven billion tonnes of carbon in the year 2000 to a maximum value of around ten billion tonnes in 2015. In order to achieve the long-term stabilisation of the atmospheric carbon dioxide concentration, the emissions will then have to be reduced by 56 percent by the year 2050 and approach zero towards the end of this century. Although, based on these calculations, global warming would remain under the two-degree threshold until 2100, further warming may be expected in the long term: "It will take centuries for the global climate system to stabilise," says Erich Roeckner. More: New carbon dioxide emissions model
    0 0
  6. I have just realised that most of the observations in the illustration is strictly not evidence of anthropologically induced global warming, only global warming. Natural warming could induce more water vapour into the atmosphere producing many of these radiative effects. The remaining observations only suggest that we have simply released more carbon into the atmosphere! To show AGW we need to show a) the basic theory of GHG warming in conjunction with evidence that these GHGs are human based (eg. from isotopic analysis), you have done this elsewhere. b) the measured relationships between GHGs and temperature, particularly the sudden rise in global temperatures when large quantities of carbon emissions were first released around 1900. Have any statisticians assessed the probability this rapid change, in conjunction with GHG concentrations might have happened by chance (the null hypothesis). I guess this would be very low. c) the absence of any plausible natural mechanism, once again you have covered this elsewhere.
    0 0
  7. There is some additional progress reported here on identifying a specific human fingerprint to extreme weather events In their biggest success, climate scientists led by Peter Stott of the British Met Office analyzed the 2003 European heat wave, when the mercury rose higher than at any time since the introduction of weather instruments (1851), and probably since at least 1500. After plugging in historical and paleo data, and working out climate patterns in a hypothetical world without a human-caused greenhouse effect, they conclude that our meddling was 75 percent to blame for the heat wave. Put another way, we more than doubled the chance that it would happen, and it’s twice as likely to be human-caused than natural. That’s one beat shy of “Yes, we did it,” but better than “There’s no way to tell.” Scientists are now applying the technique to other extreme weather, especially deluges and droughts. They have reason to be optimistic. One of the signal successes of climate science has been identifying the “fingerprints” of the culprits behind rising temperatures, fierce storms, and other signs that a 10,000-year-old climate regime has been knocked for a loop. Fingerprinting has shown that the rise in global temps follows the pattern you’d expect from the greenhouse effect and not an increase in the sun’s output, for instance. A hotter sun would heat the upper atmosphere more than the lower, but in fact the upper layers have cooled while the lower have warmed, Santer explains. Fingerprinting has also nailed the greenhouse effect for warming the oceans. Natural forces such as El Niño warm some seas and cool others, but every major ocean is hotter than in the 1950s. Similar analyses have been done for today’s extreme rainfall patterns (drought followed by deluge, not precipitation spread out evenly) and the retreat of arctic sea ice. “Natural causes alone can’t explain any of these,” Santer says. “You need a large human contribution.”
    0 0
  8. Typos in item 9: "stratophere" (twice) should be "stratosphere".
    0 0
    Moderator Response: [DB] Fixed; thanks!
  9. Here is a possible 11th indicator: wave height.
    0 0
  10. Oops - posted on wrong thread. The above link should be on ten indicators of a warming world. Sorry.
    0 0
  11. Hi John, Thanks for your useful site. About #7: The decreasing diurnal temperature range (DTR) is listed as a fingerprint of Anthropogenic warming, with references. However, you do not say what the basis is for this claim - why would the DTR be expected to decrease under greenhouse warming rather than solar warming of some other forcing? Initially it seems intuitive because obviously the sun shines in the day, and not at night -so increasing night-time temperatures "should" be due to something else. However, upon more thought I started doubting that this is the case. Indeed, going back over the references you gave, plus further back into for example the Stone and Weaver (2002, 2003) papers, and even the Easterling papers - I don't see anywhere a definitive statement that a reduced DTR is indeed a fingerprint of additional greenhouse forcing. There is some mention of this in the intro's but never anything further. It seems to come down to a matter of clouds and soil moisture in the Stone and Weaver papers, and in Braganza (2004). All of them say it is a good index of recent climate change, while I don't see any of them actually saying it provides evidence of anthropogenic warming.
    0 0
  12. Also, a suggestion: I would suggest that you add in the increase in Ocean Heat Content Anomaly (OHCA). We expect about 90% of the Anthropogenic forcing to go into heating the upper ocean - and this is indeed what is observed, and this close link between TOA radiative imbalance and OHCA increase I believe is some of the strongest evidence of anthropogenic warming. There are good recent updates on ocean heat content: Lyman et al. (2010) And there are other papers connecting the TOA radiation with the oceanic warming: you have some Trenberth ones, but Levitus principally talks about this: Levitus et al. (2005) Levitus et al. (2009) Apologies if this was already obvious and listed elsewhere.
    0 0
  13. neil, night time temperatures increasing faster than day time temperatures is indicative of enhanced greenhouse warming because of the way that greenhouse warming operates. That is, increasing the concentration of greenhouse gases in the atmosphere (which is being caused by human industry) decreases the rate at which the planet cools. Consider a 100 degree day in Miami vs a 100 degree day in the Arizona desert. Once the Sun goes down the temperature starts dropping... but Miami is very humid (lots of atmospheric water vapor, a greenhouse gas) and can actually stay warm all night. The desert on the other hand gets very cold very fast because it has almost no water vapor and thus the daytime heat escapes quickly. Thus, if we increase the level of atmospheric carbon dioxide and other greenhouse gases which have much lower geographic variance than water vapor we are decreasing the rate of night-time cooling for the entire planet... nights stay warmer longer and thus the average night time temperature increases faster than the day time temperature. As to ocean heat content... a strong indication of warming, but not of anthropogenic causes. Any warming forcing would result in most of the energy going into the oceans. For instance, if the observed global warming were being caused by increased solar radiation we would expect to see days warming faster than nights (because there is no sunlight at night) and ocean heat content rising as most of the solar forcing went into heating the upper ocean. Ergo, we'd have the ocean warming either way, but the day vs night warming speed would be different.
    0 0

    [DB] Additionally, what was predicted by models and then subsequently confirmed later by observational studies is summarized here.  For example:

    Nights warm more than days Arrhenius 1896 Dai et al. 1999
    Sherwood et al. 2005
  14. @CBDunkerson Thanks. I understand the concept of greenhouse gas induced radiative forcing. I note that in Arrhenius (1896) there is the prediction of reduced DTR, which has been confirmed experimentally. I submit to you that more recent research (like Stone and Weaver (2002, 2003) ) has shown that water vapour and soil moisture have an overwhelming effect on DTR - its not simply a direct radiative effect of anthropogenic CO2. So, it may be wise to note that (although I accept increased water vapour may be attributed to anthropogenic CO2). As for the ocean heat. Of course increasing ocean heat content itself does not prove anthropogenic warming - so if that's how my line read I apologize. The strength of the argument lies in that there is a measured decrease in outgoing long wave radiation at the Top of the Atmosphere - and a concomitant measured increase in downward longwave radiation at the earths surface - corresponding to the CO2 absorption bands in the spectrum. That is the anthropogenic link - and yields a radiation imbalance or anthropogenic radiative forcing if you will (with some error). The ocean heat content increase IMPLIES a radiative forcing which is in line (though smaller) than the observed radiative forcing, and also in line with that predicted by GCMs doing global warming experiments. Essentially my point is, the Ocean Heat Content Anomaly is a far more robust measure of global warming than the surface temperature record, because the ocean absorbs most of the heat.
    0 0
  15. I too feel that Skeptical Science is overselling DTR changes as a greenhouse signature (it is probably an ANTHROPOGENIC signature - not the same thing). I have never seen a paper about observed changes in DTR that attributes it to an enhanced greenhouse effect. Instead every paper that I have seen (including the ones listed on skeptical science) attribute the DTR changes to a suppression of daytime solar radiation - probably from anthropogenic aerosols. I think that the problem with CBDunkerson's argument above is that nighttime minimum temperature is not independent from daytime maximum temperature. If radiative cooling to space is suppressed, and the nighttime minimum temperature increases, then as soon as the sun comes up the daytime warming starts from this new warmer temperature. If this is the case then the daily max should increase as well. One other thing. In the IPCC AR4 summary for policy makers, it says this: "A decrease in diurnal temperature range (DTR) was reported in the TAR, but the data available then extended only from 1950 to 1993. Updated observations reveal that DTR has not changed from 1979 to 2004 as both day- and night-time temperature have risen at about the same rate. The trends are highly variable from one region to another." If Skeptical Science wants to claim that DTR changes are due to an enhanced greenhouse effect then they are cornering themselves by making it necessary to come up with an explanation for why DTR hasn't changed in recent decades.
    0 0
  16. Please explain how carbon in fossil fuels becomes 'newer' than the fossils it came from by using it. .
    0 0
  17. Not sure what your point is, but the fossil carbon is millions of years of sequestered carbon. We are releasing it over few hundred years. Does this imply CO2 was higher in past? Yes, it was but sun was weaker and the issue is the rate of change not the absolute temperature.
    0 0
  18. i am new to the debate and if my point has been covered elsewhere i would appreciate a link. do we sufficiently understand the driving forces behind the cycle of ice ages and subsequent warming to state categorically that co2 from the burning of fossil fuels is the principal cause of the current warming?
    0 0
  19. New Boy. Concluding that CO2 is "the principal cause of the current warming" is a result of two processes. One. Physicists and chemists identified the properties of CO2 and advised (and then warned) that increasing the proportion of CO2 in the atmosphere would result in warming. Two. Other people have done lots of work on analysing all the various causes and sources of both warming and cooling now and in the past. Having measured, analysed, checked and analysed yet more data, they've concluded that the other known causes of variations in the climate cannot account for the current warming. The properties and the increasing concentrations of CO2 can explain it. One - theory about ghgs predicted warming long before there was any real evidence. Two - measurements of warming coincided with other measurements, greenhouse gases, TSI, volcanoes, declining ice, orbital variations, and found that only greenhouse gas warming can account for the increases we've seen.
    0 0
  20. #68 and #69, can I add ... Three: we observe the increase in downward longwave radiation and decrease in outgoing longwave radiation at CO2-specific wavelengths. Not only is CO2 the only explanation for warming, we actually see it happening directly. That's the "less heat escaping to space" in the graphic above.
    0 0
  21. NBotB, for info on 'ice ages' I'd suggest reading the article on Milankovitch cycles. As the other responses indicate, our understanding of the glaciation cycle is just one of many independent lines of evidence which confirm that fossil fuel CO2 emissions are the primary cause of recent warming. The list of 'skeptic' arguments and the 'search' box at the upper left of the page are good places to find answers to various questions.
    0 0
  22. Suggested thread for peacetracker's comment on anthropogenic/greenhouse warming fingerprints & subsequent responses.
    0 0
  23. Got a question. Can some one point to me a peer reviewed publication/s that gives the exact percentage of atmospheric CO2 is from anthropogenic sources? What I would most like is a paper that spelled it out in percent. I have always understood this to be about 39% from looking at the Keeling Curve but I need a source for the exact number. Thank You.
    0 0
  24. Trent1492, there is some disagreement on whether and how much ancient humans contributed to atmospheric CO2 levels through deforestation and other 'land use' changes. If we set that aside and accept the pre industrial revolution value of about 278 ppm as a starting point it gets much simpler. Given that we know (see below) from multiple lines of evidence that humans are responsible for all of the increase from that point, calculating the 'human percentage' is just a matter of dividing the current level (about 392 ppm) by 278 ppm... which gives about a 41% increase. The 39% figure you cite was probably based on 386 ppm atmospheric levels a couple of years ago. A good overview, with links to source papers, of how we know humans are responsible for the recent increase can be found here.
    0 0
  25. i suggest the inclusion of this link to help explain proportion of CO2 accumulating in the atmosphere from Human/nature.

    The latest increasingly popular ploy used by politicians and industrialists during media interviews to avoid directly addressing the issue of the increasing threat from accumulating CO2 from fossil fuel use is : "... yes we accept that climate change is occurring, but are not sure as to how much of that change is due to human activity".

    It seems to have become the standard  'do nothing' cop out, when confronted with the undeniable physical evidence of increasing cyclone/flood/fire disasters. 

    0 0
  26. Interestingly, some of the alternate energy sources proposed as soltuions will not work.

    Nuclear - speeds the decay of nuclear materials which produces more heat now than the natural decay of these same materials in the earth.

    Solar - If the CO2 traps radiation trying to exit the earth after reflection, than trapping it here at the surface just accomplishes the same thing in a more efficient manner.

    Nuclear and Solar may be better than coal/oil, but will still cause global warming. Wind and Hydro capture energy that is already here and thus do not warm the planet.

    0 0
  27. cynicalanddisgusted - Anthropogenic greenhouse gas trapping of IR adds about 100x the forcing to the climate as all of human energy use; see the waste heat discussions. Any move to a non-carbon based energy utilization will reduce global warming, including nuclear and solar. 

    It's important to actually look at the numbers.

    0 0
  28. Just wondering why in point 7 it states

    "If an increased greenhouse effect is causing global warming, we should see certain patterns in the warming. For example, the planet should warm faster at night than during the day. This is indeed being observed (Braganza 2004, Alexander 2006)."

    Is not observed in the CRUTEM4 DTR data.

     CRUTEM 4 Diurnal Temperature Range



    0 0
  29. Forgot to add that data is Crutem4 sourced via KNMI

    0 0
  30. POJO @78, if the night times warm faster, then the difference between daytime and nighttime temperatures will decrease, as is shown on your graph.  Consequently I am not sure what your point is.

    0 0
  31. Tom Curtis @ 78


    It is a very broad history i have shown there.  Here is the last 35 years.

    Keen to hear an explanation for this apparent diversion.

    1980-2014 Crutem4 DTR

    0 0
  32. POJO @81, first, you have missed the point of my comment.  You are presenting data that, at least broadly, is in agreement with the peer reviewed assessments cited above as contradicting those assessments.  That at least requires some detailed explanation by you as to why you think that data contradicts those assessments.  Absent that explanation, you have not point.  Indeed, absent that explanation you are at least contradicting the spirit of the comments policy were it states:

    "No link or picture only. Any link or picture should be accompanied by text summarizing both the content of the link or picture, and showing how it is relevant to the topic of discussion. Failure to do both of these things will result in the comment being considered off topic."

    (Underlinging mine)

    Indeed, I would go further, you have taken no effort at all to show "how it is relevant ot the topic of discussion".

    When I challenged you on that by asking you, "what was your point" you simply evaded the question.  You do not make a point, preffering to set me homework.  I pressume that is because you do not have a coherent point at all.

    Second, what I make of your second graph is that you have (deliberately) increased the smoothing window, thereby eliminating information and preventing a coherent intepretation of the data.  From the first graph over the same period, and by eye, I would suggest that DTR decreases as a result of volcanic erruptions, and also as a result of El Ninos; and that it increases as a result of La Ninas, and that this explains the recent history.  I am not firmly committed to that interpretation because my alignments are by eye only.  However, if that is the case the recent history of EL Nino's and volcanic erruptions would result in a spurious upward trend in DTR from 1980, despite the fact that the trend is flat.  That in turn requires an explanation as to why that spurious trend has been nullified - with Braganza (2004) providing just such an explanation.

    0 0
  33. #82 Tom,

    Sorry Tom,  I assumed i had made it clear.

    #7 stated that diurnal temperatures should be decreasing.

    The first chart supplied for last century shows the decreasing trend of DTR up until the 1980 ish  It then shows the deceleration of that trend and then subsequent rise (in my view)

    The next chart posted was for your benfit.  To remove the noise I did smooth it out more.  

    I am shocked that you have inferred my intentions were not honest. I refute that.

    All I have done is use a 5 year average that NOAA, NASA and even Skepticalscience use when showing charts.  I am confused why that is acceptable for them to apply to Crutem4 data but it is not when i apply a 5 year average to the CRUTEM4 data.


    So is their an explanation as to why DTR has not decreased.  At worst why has it stalled when it clearly should be accellerating in its shrinking of the DTR??

    0 0
  34. Sorry I should point out that my reference to #7 in the above post was to the opening piece statement

    7. If an increased greenhouse effect is causing global warming, we should see certain patterns in the warming. For example, the planet should warm faster at night than during the day. This is indeed being observed (Braganza 2004, Alexander 2006).

    my bold highlighting

    not poster 7 on this thread

    0 0
  35. POJO @83 and 84:

    1)  Braganza et al show that DTR has a correlation to T mean of -0.24 (ie, as temperature increases the DTR falls), and that DTR has a trend of approximately 0.08 C /decade (0.4 C/50 years).  They specifically show that "This trend is due to larger increases in minimum temperatures (0.9C) than maximum temperatures (0.6C) over the same period" which is inconsistent with the primary driver of the warming being increased insolation or decreased cloud cover.

    Alexander et al show that the frequency* of warm nights has increased faster than the frequency of warm days, and that the frequency of cold nights has decreased faster than the frequency of cold days:

    (* technically the average % of land area experiencing a warm night etc.)

    They use a different, and much more extensive dataset than used by you or Braganza et al, and do not show the large fluctuation in trend evident in your graph @78 between 1960 and 1980. That suggests the large fluctuation in trend may be an artifact of the data set.  Note further that Braganza et al (which used an earlier version of the dataset you use) excluded cells which had less than 40 years of data between 1901-2000.  (This raises the possibility that that odd variation in trend is an artifact of variation in reporting stations over that period.)

    In any event, the graph you show does not contradict either of the findings by Braganza et al, nor the findings by Alexander et al.  Nor do you show any analysis that suggests that it does.  Frankly that should be the end of the story.  Eyeballing of data does not trum peer reviewed analysis.

    2)  Being a stickler for accuracy, what I suggested was that you deliberately altered the smoothing window.  I made no speculation as to why you altered that window, but did note that the move eliminated relevant data (such as the El Nino related trough in 97-98).

    I note that you then drew particular attention to the "apparent diversion" from 1985-1993 which is at least partly an artifact of that smoothing window (due to the aggregating of the effects of the 1983 volcano plus El Nino with the 1988 El Nino due to the five year smoothing window), and whose prominence is owed to the smoothing away of peaks of similar magnitude.  That shows that your deliberate change of smoothing window was either due to dishonesty or utter carelessnes with the data.  Again, I made no speculation as to which, but either is reprehensible.  Particularly given that you are attempting to refute peer reviewed literature by the mere eyeballing of data.

    Finally, as I am being a stickler for accuracy (and as this inflation of language gets my goat), you have not "refuted" the non-existent inference that you are dishonest.  To refute something, you need to mount a successful argument to the contrary.  Nor have you even rebutted it, for which you actually need to mount an argument.  All you have done is reject the claim.

    I will accept that rejection on face value, and from my imputed dichotomy infer that you are as careless with the presentation of data, as you are reticent in actually presenting an argument and/or analysis.

    3)  You ask " their an explanation as to why DTR has not decreased."  I have already provided one @ 82:

    "From the first graph over the same period, and by eye, I would suggest that DTR decreases as a result of volcanic erruptions, and also as a result of El Ninos; and that it increases as a result of La Ninas, and that this explains the recent history. ... However, if that is the case the recent history of EL Nino's and volcanic erruptions would result in a spurious upward trend in DTR from 1980, despite the fact that the trend is flat."

    To double check on my interpretation, I have looked at the 12 month running mean data in 10 year windows centered on major ENSO and/or volcanic events and can confirm the relationship ascribed.  Ergo, the strong volcanic erruptions (1982/1993) coupled with the super El Ninos early in the period (1983, 1997/98) and strong La Ninas late in the period (2008, 20011/12) would definitely result in a spurious positive trend in DTR.  That the trend is instead flat (0.00 C/decade) shows there is some countervailing factor resulting in a negative trend.

    In short, your special window is merely another example of a cherry picked pause such as those in mean temperature extending from 1997.  You are merely able to extend it further because of the lower trend in DTR and the coincidence that volcanos and El Ninos have the same sign in influencing DTR.

    The accelerating decrease in DTR you refer to is an artifact of your own imagination, given the small relative effect on DTR and the lag between forcing and temperature response.

    4) Your case has continued to be based on eyeballing and bald assertion.  Absent actual mathematical analysis of the data in support of your position, I doubt you will have anything new to say on this topic worth responding to (and which I have not already dealt with).  Ergo absent that analysis, do not expect further responses.

    0 0
  36. Tom Curtis,  For some unknown reason I have just spent a1/2 hour responding.  I am peeved that  it did not go through after posting.

    Nonetheless,  here is the shortened version

    Your very first paper cites Brazanga,

    He in the very first paragragh states that he is using decadel data.  He then go on to say in section 3


    3. Simulated and Observed Changes in
    Diurnal Temperature Range

    Paragraph 2 of section 3

    [10] We use the decadal standard deviation as a measure
    of variability in DTR

    I find this very peculiar that you have just finshed berating me for being less than honest and yet their is no problems on your behalf to use a paper that uses decadel averages..  Am I perplexed.

    YES I AM


    Further to that I except that extreme weather events aka EL NINO's do in fact enhance the DTR signature as stated by Brazanga.  that is observed data even in the 60 month average

    0 0
  37. Dear Tom,

    In regards to your second paper  Alexander et al of point numebr one.  I note that his observations are based on decadel trends

    4. Results

    4.1.1. Absolute and Percentile-Based Temperature

    [29] Figure 2 shows the decadal trends in extremes
    between 1951 and 2003 for the percentile-based temperature


    and then 


    4.1.3. Seasonal Results
    [35] Warming is observed in all seasons. Figures 4 and 5
    show decadal trends and time series of global anomalies for
    the seasonal occurrence of cold and warm nights, respectively


    4.2.1. Annual Results

    ......and  Decadal trends in the simple daily
    intensity index (Figure 6d) also agree well with the
    results from Kiktev et al. [2003] although unlike Kiktev
    et al. [2003] the decreasing intensity in the western
    United States is not identified to be statistically significant
    in this study.


    Again decadel trends.  I have used the ever reliale CRUTEM4 data.  A 12 month average shows the increase in DTR,  A 60 month average highlights it. 

    I am at aloss why you do not accept basic monthly data on a 60pt average yet except in your papeers cited the 120 month point averages.

    Note that the finger print as outlined in point number 7clearly states that 

    If an increased greenhouse effect is causing global warming, we should see certain patterns in the warming. For example, the planet should warm faster at night than during the day. This is indeed being observed (Braganza 2004, Alexander 2006).

    I have shown 35 (ish)years of data that contradicts this  Is their something else with a 60 or lower point averagethat can corrorbarate DTR as a fingerprint.  Cloud cover maybe?  

    Ciou for now

    0 0
  38. POJO @86 and 87, evidently you do not know the difference between climate and weather.

    Braganza et al were looking for the effects of change in climate on DTR, not looking for the effects of short term fluctuations on DTR.  Therefore they chose a smoothing value that eliminated such fluctuations, but left the climate effects unaltered.  The effect of that was to weaken the correlation of DTR to Mean temperature from to -0.37 using the montly data, relative to their reported -0.24.  It leaves the trend in DTR unchanged at -0.36 C per 50 years (slightly misreported by me in error in my prior post, for which I am sorry).

    In contrast to Braganza et al, you were looking at the effects of fluctuations over a short interval, particularly highlighting a fluctuation whose total duration in the smoothed data was just 10 years.  That is, you were looking at weather.  Yet you chose a smoothing window that largely obscured the short term fluctations.

    With regard to Alexander et al, they used unsmoothed daily data.  In the timeline graphs, they showed the annual data along with a smooth.  In the mapped graphs they showed the decadal trends.  The difference between a decadal trend and an annual trend is just that the former is the later multiplied by 10 (ignoring rounding issues).  That is, the difference is merely a matter of the unit chosen to report the result - not the value of the result.  In other words, what Alexander et al did was nothing like what you did.

    To summarize, you apparently do not know what was wrong with your smoothing, despite my having pointed it out.  In defense of that, you point to a greater smoothing by Braganza without noticing their very different purpose to yours.  You compound your error by accusing Alexander of making a similar error, when they did nothing of the sort.  (The later strongly suggest incompetence rather than deceit as the reason for your inappropriate smoothing.)

    Finally, inorder to check Braganza's result against the monthly data, I downloaded the monthly Crutem 4 data.  As a result I was able to compare the normalized inverted DTR data to the Mean data and can confirm that the sudden change in slope is a feature of both.  The difference in the two largely arises because the effect of the two large volcanic erruptions (El Chichon in 1982; and Pinatubo in 1991)  have same sign on DTR and Mean temperatures.  That artificially flattens the rapid slope visible from 1960-1980.  The trend from super El Ninos (1983, 19897/98) to large La Ninas (2008, 2011/2012) artificially reduces both the positive trend in Mean temperatures and the negative trend in DTR. 

    0 0
  39. Tom @88

    Yes my apologies for getting the decadel wrong.  That was a bad error.  So I am now using the annual data,


    Note that Alexender etal data was based on data from 1951 to 2003

      So i have now accuired the annual (yearly) data from KNMI / CRUTEM4.  No smothing applied yet

    The story does not change.  DTR saw a rapid decline and has since stabalised and even increased

    Annual DTR


    Now below the decadal smooth is applied.  And it is evident as per ALexander etal that infact the trend remained intact until 1998-2003.

    Since then the decadal DTR has since increased.

    10 year smoothing on annual DTR



    I find it odd that the DTR is now being assessed in terms of cold nights and warm night extremes.

    The science is clear that as GHG's increase so to does DTR decrease.  Extrene events are termed weather by many!!!!!  

    No matter what variation is done to the monthly data it is evident that the change has at the very least increased from 2000 ish.  Which coincides with available data at the time that Alexander etal had.

    As you alluded to my earlier comment cherry picking short events is weather.

    So I have now heard what you have said and now have attached the climate charts as per above.


    Perhaps this change in DTR also can be used to best highlight the current pause.  

    As per Brazanghi final summary that DTR can be used to

    ....Diurnal temperature range appears to be a suitable
    index of climate variability and change,......

    page 4 of 4 SUMMARY

    0 0
  40. POJO @89, briefly:

    1)  Thankyou for acknowledging and apologizing for your error.

    2)  Annual data is already smoothed from the montly data, so the claim of "no smoothing yet" is false.

    3)  Alexandar et al examine frequency of warm or cold, night and days.  That frequency can not be identified from monthly data, let alone annual data.  Further, Alexandar et al examined the percentage of the Earth covered by warm or cold, nights and days.  For that you need the full gridded data, not a time series.  It follows that your analysis above is in no way a replication of Alexander et al.  Rather, it is only another shot at replicated Braganza et al.

    4)  I do not trust a decadel smooth that returns "observational" values for 2018 from data that terminates in December of 2014.  Ergo your second graph is near to useless.  (I am aware that may not be your fault.)

    5)  Regardless of the merits of the smooth, you appear to back at the game (already refuted) of using smoothed data to discuss the causes of trends in a noisy data set obviously significantly effected by a variety of factors.  However, it may be that you are only trying to illustrate the broad trends.  That being the case, the period of the smooth is appropriate, but the method is dubious, both because of the spurious years added to the tail of the smooth, and because that tail hides a downtick at the end visible in the 10 year running mean data.

    6)  There is nothing odd about cross checking an analysis by checking what happens in an obviously correlated variable.  The relative frequency of warm nights and days is not DTR, but is obviously related.

    7)  Extreme events are weather.  So are daily or monthly measurements of mean temperature or DTR.  Climate, however, is the statistics of weather.  Alexander et al test to see if the statistics of warm or cold, nights and days changes over time.  It does, and with a consistent long term trend.  That later represents a change in climate.

    More detailed comment:

    A)  Here, and in comments elsewhere, you talk about refuting a fingerprint.  If you look at the original post, and the accompanying illustration, it talks about the pattern of 10 features being a fingerprint.  By that analogy, an individual item among those ten may be shared with other sources of warming (as individual details may be common between two peoples fingerprints), but the overall pattern of all 10 is unique.  Consequently, as it stands you have not refuted any fingerprint.  You have not even refuted the DTR detail.  At best you have shown this particular detail to have multiple causative factors (as was already known, and is the actual IPCC science, contrary to your claims), and so that its weight in contributing to the overall "fingerprint identification" is small.

    B)  The OLS DTR trend since 1980 is flat and not significant (0.01 +/- 0.12 C per century).  As twice explained above, and twice ignored, that has a prima facie explanation in the two large volcanoes in the late twentieth century, the two super El Nino's in the same period, and the two large La Nina's late in the data period.

    I can now add to that that if you cherry pick your start period to coincide with the 97/98 El Nino, you can just squeeze a positive statistically significant trend out of the data (0.45 +/- 0.32 C per century).  In all other periods less than 30 years I have checked for a trend, the trend has not been significant.  Further, the data is obviously autocorrelated, so the OLS trend calculation will overestimate statistical significance.  Using a more appropriate model, it may well be that even the 1997-2014 trend is not statistically significant.  I have not checked that, however, as the maths is beyond my pay grade.  Further, the most recent trend from 2012 is strongly negative (-3.03 +/- 4.63 C per century).  That is entirely unsurprising considering the visible relationship between DTR and ENSO states.  Of course, so short a trend is just weather, but that relationship does suggest the 2015 data will eliminate any statistically significant trend from 1997.  In other words, trends since 1997 are too short to be representative or changes of climate in regards to mean temperature, and are certainly too short to represent changes in climate in DTR (which typically has a lower trend).

    More later.

    0 0
  41. @Tom #90

    thanks To for the response and the net surfing.  Great work :)

    Regards to OLS.  Must be honest,  that is escaping me at this stage.  Need to know what that is before proceeding

     Thank you Tom

    0 0
  42. Thanks Tom for the response,

    After getting through all of that. I guess the only bit that matters at the end of the day is what the observed data is saying. If someone is telling the chart is going down but I can see that it is going up I can conclude two things. It is at best stable or perhaps marginally going up.

    I say it is going up and hence forth my reasons to support this below.

    • I note that the IPCC have withdrawn their support from the DTR story. As from AR4 in as so far they have distant themselves from the previous AR reports where they conclusively aligned themselves with DTR. Note that I do agree that DTR is a good proxy for cloud cover and surface solar radiation amongst others.

    With all the factors going into establishing DTR you mentioned a number of issues that attempt to explain the effect aerosols had on DTR. I.E The Abrupt decline.
    You inferred that aerosols – Decease DT. And they do.
    El Ninos and La ninas. Well I look at them as have a neutral effect. Warms here but cools down over there. And we are continuously reminded that that the Oscillation effect is NET ZERO.

    So taking on board that globally aerosols have a reduction effect in DTR I find it pointless discussing this. We both agree that DTR declined with absolutely minimal effort from the two biggest movers and shakers of Aerosols last century El Chichon and Pinatubo.


    Re below Mauna Loa radiation transmissions from aeresol thickness

    Mauna Loa Solar radiation transmitted

    Also this pic  Will not upload?


    It is obvious when looking at Crutem Mean Temperature Data that the Mean uptrend was very steep since the mid seventies.

    Whilst I do not discount the findings of Brazanghi And Alexander it is worth noting that their data was only up to 2000 and 2003. As the IPCC felt strongly about the DTR fingerprint in FAR, SAR and TAR I can see why so did Brazanghi et el and Alexander et al. But since then the IPCC have distanced itself from the DTR fingerprint. As for the reasons I have already given. The observations clearly show it.

    Whilst I do not discount the findings of Brazanghi And Alexander it is worth noting that their data was only up to 2000 and 2003. As the IPCC felt strongly about the DTR fingerprint in FAR, SAR and TAR I can see why so did Brazanghi et el and Alexander et al. But since then the IPCC have distanced itself from the DTR fingerprint. As for the reasons I have already given. The observations clearly show it.


    This from IPCC AR4 Climate Change 2007: Working Group I: The Physical Science I am sure you have seen this part before.

    Some aspects of climate have not been observed to change. {3.2, 3.8, 4.4, 5.3}
    • A decrease in diurnal temperature range (DTR) was reported in the TAR, but the data available then extended only from 1950 to 1993. Updated observations reveal that DTR has not changed from 1979 to 2004 as both day- and night-time temperature have risen at about the same rate. The trends are highly variable from one region to another. {3.2}


    Hence I am at a loss to understand why you are claiming Aerosols as a cause for abrupt dtr decline and then going on the limited data that Brazanghi etal and Alexander etal had they claim GHG's are responsible.  The claim to defend the DTR Fingerprint as a result is conflicting from SKS in this case.  
    This logic contradicts itself.

    So if IPCC are stepping back from the DTR fingerprint what reason do I have to believe what is written in this threads OP.

    You mentioned that the decadel data had extra pips on it. And you are right. I checked the data and it is their. I am damned if I know how it does that. I reloaded the sheet again and it is still their. Nonetheless you were able to establish this and consequently discarded those last two pips.

    In summary Tom,
    The observed data of DTR is clear that the decline has been arrested.

    Whether you use monthly, annually or decadal. It is obvious.



    Monthly low pass 12




    And Decadal


    To me it is obvious.  As to do the IPCC!

    0 0
    Moderator Response:

    [BW] Comment edited to make graphics fit the page. Please make sure to restrict width to 500px.

  43. POJO @92.

    I am not sure why you harp back to AR4. AR5 has happened since and surely supersedes AR4. Mind, I'm not so sure you have read AR4 properly. Or even given its pictures a critical look. Do you not think the inconsistencies between your data and that presented in AR4 Figure 3.2 should be telling you something?

    0 0
  44. MA Rodger @93.

    Is it not enough that IPCC have openly stated that have backed away from DTR??  

    In reference to AR5.  Please advise me where IPCC have opted back in??

    Regards to Fig 3.2 of AR4.   After just explaing and providing the evidence that the IPCC have backed away from DTR as a fingerprint you are effectively questioning the IPCC's wisdom!

    It is clear in the the low resolution anomaly chart that the anomaly has flatlined since  1990 ish.

    Also note that the IPCC chart you reference in Fig 3.2 of AR4 uses a DECADAL anomaly.  The decadal chart I posted is on post 92.  It clearly show's what has happened since the 90s.  Ignore the last two ticks. 


    I may add as I have in previous comment that the data that is being used to support the Fingerprint as pointed out by Tom was only upto 2000 and 2003.  It is now 15 years later and you are defending the the Brazanghi analysis.  I applaud you for that.  On reflection I am not disputing his analysis.  He made a analysis with the data he had at the time.   

    Note that Brazanghi etal paper was more about the suitability of using DTR as a fingerprint for CLIMATE VARIABILITY AND CHANGE.  

    I have already stated that I support this assessment.  But people need to move on.  Another 15 years has elapsed from the last data used for his assessment.   There is nothing in his paper that states his findings do not apply if DTR is moving in a opposite direction now is their.


    So in conclusion it is evident to me that DTR is no longer a endorsemnt (Fingerprint) for Co2 effect on greenhouse gas.  DTR is stable and Co2 is pumping along.

    It is a endorsement of correlation between cloud and temperature.  More cloud means a smaller range between min and max temps.  That is all it means.  

    0 0
  45. POJO @92 and @94, seems intent on creating a fiction about the IPCC discussions of DTR.  According to his mythology, the IPCC supported "the DTR story" up to, and including the Third Assessment Report (TAR), but thereafter withdrew that support.  In fact, the IPCC discussion of DTR has been nuanced throughout.  This, from the First Assessment Report (FAR) we have:

    "Because the ocean has a large heat capacity, diurnal temperature variations in the ocean and in the overlying air are considerably muted compared with those over land and, from a climatic point of view, are likely to change little Over land, diurnal variations are much less restricted so the potential for relative variations in maximum and minimum temperature is much larger Such relative changes might result from changes in cloudiness, humidity, atmospheric circulation patterns windincss or even the amount of moisture in the ground Unfortunately, it is not yet possible to assess variations of maximum and minimum temperature on a hemispheric or global scale However in the regions discussed below, multi-decadal tiends of day-time and night-time temperatuies have been studied and do not always appear to be the same"

    Discussion clearly focusses on the multiple possible causes of changes in DTR, and notes that hemispheric and global data is not availabe.  From that it follows that they cannot have been pushing "the story" POJO attributes to them.

    The Third Assessment Report concludes its discussion of DTR by saying:

    "Minimum temperature for both hemispheres increased abruptly in the late 1970s, coincident with an apparent change in the character of the El Niño-Southern Oscillation (ENSO) phenomenon, giving persistently warmer sea temperatures in the tropical central and east Pacific (see Section 2.6.2). Seasonally, the strongest changes in the DTR were in the boreal winter (−0.13°C/decade for rural stations) and the smallest changes were during boreal summer (−0.065°C/decade), indicating some seasonality in the changes. Preliminary extensions of the Easterling et al. (1997) analysis to 1997 show that the declining trends in DTR have continued in much of North America and Asia. 

    Figure 2.3 shows the relationship between cloudiness and the DTR for a number of regions where long-term cloud cover data are available (Dai et al., 1997a). For each region there was an increase in cloud cover over the 20th century and generally a decrease in DTR. In some instances the correlation between annual cloud cover and annual DTR is remarkably strong, suggesting a distinct relationship between cloud cover and DTR. This would be expected since cloud dampens the diurnal cycle of radiation balance at the surface. Anthropogenically-caused increases in tropospheric aerosol loadings have been implicated in some of these cloud cover changes, while the aerosols themselves can cause small changes in DTR without cloud changes (Hansen et al., 1998 and Chapter 6)."

    The strong trend in DTR from the 1970s is attributed to a change in the ENSO regime.  The effect of cloudiness is discussed extensively, and some attention is also given to aerosols.  Again we have the nuanced discussion we expect from the IPCC.

    In short, the only thing that has changed in the IPCC accounts prior to AR4, and after is the addition of more data allowing a better characterization of the trends and uncertainties.

    0 0
  46. POJO @94.

    I'm not entirely sure what you mean by "backed away" and "opt back in." They are not terms I have met in scientific papers. AR5 Section is titled Diurnal Temperature Range so DTR is still a subject under discussion by the IPCC, as ever. Perhaps the question should be asked as to why there are no DTR analyses plotting out global DTR time-series post-Vose et al. (2005)? Is there a difficulty with such analysis? Indeed, why do your graphs present data that is clearly contradictory to that presented by Vose et al. (2005)?

    0 0
  47. MA Rodger


    The IPCC stance as noted on pge 9 of AR% for policy makers.

    "A decrease in diurnal temperature range (DTR) was reported in the TAR, but the data available then extended only from 1950 to 1993. Updated observations reveal that DTR has not changed from 1979 to 2004 as both day- and night-time temperature have risen at about the same rate. The trends are highly variable from one region to another. {3.2}"


    Backed away and OPt back in.  They may not be scientific terms MA.  But they very obvious to everyday meaning.  I am sure you know what they mean.  

    In regards to Vose etal paper.

    Again the obvious stands out.  Back then old data was used.  Since then the global data has had several changes to it via its rigourous homeginisation process.

    Crutem 4 is the LATEST dataset.  Analysis now is at a fingertip.  Tmax - Tmin.

    Voss anomalie chart is the same as the IPCC in TAR.

    And here is Vose's response to the chart is below

    Both maximum and minimum temperature increases from 1979-2004 whereas the DTR is basically trendless. The maximum and minimum temperature trends are nearly identical (0.287 versus 0.295°C dec-1),............


    And lastly the data has moved on since way back then.  Know one needs a paper to look at a chart.  It is what it is.


    If you have a problem with my source for the Charting best you take that up with the home of Data.  It is fresh from The CLIMATE RESEARCH UNIT.  This is where CRUTEM 4 comes from.

    0 0
  48. POJO @97.

    The use of every day terms within science is entirely acceptable if their meaning is scientifically precise. I do not believe the two terms you used @94 in any way provide such precision.

    As for you consideration of Vose et al. (2005), I would agree that "the obvious stands out" but I don't think anything is gained by stating it. You are wrong saying that Vose et al. (2005) "is the same as the IPCC in TAR." I think the reason for that is "obvious."

    As for CRUTEM4 Tmax and Tmin. You suggest these data have undergone "a rigourous homeginisation process." Do you think it is adequate for your purpose? (Note that maybe "one needs a paper" to answer that.) If CRUTEM4 is accurate enough, the dramatic change in DTR estimates between Vose et al. (2005) and that afforded by CRUTEM4 would be worthy of comment, assuming your graphing is also accurate.

    0 0
  49. MA Rodger @98, The DTR data used by POJO is not from CRUTEM4, contrary to his frequent assertions.  Rather it is from the CRU TS 3.23 gridded dataset, from which a timeline is produced by the KNMI Climate Explorer software.  The CRU TS 3.10 dataset (updated to 2009) is described Harris, Jones, Osborne and Lister (2013).  I am not aware of a paper explicitly describing version 3.23, but it is unlikely to differ significantly, except for the end date.

    Harris et al write:

    "The CRU TS dataset is not specifically homogeneous.  Whilst many of the observations will have been homogenized (often by national meteorological agencies) prior to publication and use in the process, this is not a requirement for inclusion. With the use of climatological normals (and synthetic data in the case of secondary parameters) to supplement observations, it would be neither appropriate nor straightforward to assess homogeneity throughout the dataset. This dataset should only be used for climate trend analysis, therefore, if the results are treated cautiously, and we recommend that such analysis should be complemented by comparison with other datasets. For example, in Section 4 we compare longterm changes in CRU TS3.10 with CRUTEM3 and GPCC over world regions, and find good agreement at the chosen spatial scales. We also compare CRU TS3.10 mean temperature with CRUTEM4 at that dataset’s resolution, finding that long-term (∼50 year) and full-term trends are consistent, with only one or two exceptions where the trends are significantly different at the 95% level."

    For what it is worth, here are the comparison between CRU TS3.10 and CRUTEM3.  The three numbers in each category are in order, the TS3.10 trend, the CRUTEM3 trend, and the correlation of annual values in the timeseries.

    Northern Hemisphere  0.10 0.09 0.97
    Southern Hemisphere  0.05 0.08 0.94
    Global 0.07 0.08 0.97

    0 0
  50. It would be inteeresting to hear some comments to this more recent paper related to (9)

    /Wilmer T

    0 0
    Moderator Response:

    [PS] Fixed link. Please learn to do this yourself with the link tool in the comments editor.

Prev  1  2  

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 2024 John Cook
Home | Translations | About Us | Privacy | Contact Us