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

Twitter Facebook YouTube Pinterest MeWe

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


What has global warming done since 1998?

What the science says...

Select a level... Basic Intermediate

Every part of the Earth's climate system has continued warming since 1998, with 2015 shattering temperature records.

Climate Myth...

It hasn't warmed since 1998

For the years 1998-2005, temperature did not increase. This period coincides with society's continued pumping of more CO2 into the atmosphere. (Bob Carter)

Even if we ignore long term trends and just look at the record-breakers, 2015, 2014, 2010, and 2005 were hotter than 1998.

The myth of no warming since 1998 was based on the satellite record estimates of the temperature of the atmosphere.  However, as discussed in the video below by Peter Sinclair, even that argument is no longer accurate.  The satellites show warming since 1998 too.

There's also a tendency for some people just to concentrate on atmospheric or surface air temperatures when there are other, more useful, indicators that can give us a better idea how rapidly the world is warming. More than 90% of global warming heat goes into warming the oceans, while less than 3% goes into increasing the atmospheric and surface air temperature.  Records show that the Earth has been warming at a steady rate before and since 1998 and there is no sign of it slowing any time soon (Figure 1). 

Fig 1

Figure 1:  Land, atmosphere, and ice heating (red), 0-700 meter ocean heat content (OHC) increase (light blue), 700-2,000 meter OHC increase (dark blue).  From Nuccitelli et al. (2012).

Even if we focus exclusively on global surface temperatures, Cowtan & Way (2013) shows that when we account for temperatures across the entire globe (including the Arctic, which is the part of the planet warming fastest), the global surface warming trend for 1997–2015 is approximately 0.14°C per decade.

Ultimately, every part of the Earth's climate system is warming, and has continued warming since 1998.

This rebuttal was updated by Kyle Pressler in September 2021 to replace broken links. The updates are a result of our call for help published in May 2021.

Last updated on 29 September 2017 by dana1981. View Archives

Printable Version  |  Offline PDF Version  |  Link to this page

Argument Feedback

Please use this form to let us know about suggested updates to this rebuttal.

Further reading

Tamino further explores the warming trend since 1998 in Garbage is Forever and Wiggles.

I've kept my original treatment of the subject as other websites hotlink to the images. My original treatment uses similar arguments to Fawcett and Jones 2008 although their analysis is much more rigorous (as you'd expect in a peer-reviewed paper).

Further viewing


Prev  1  2  3  4  5  6  7  8  9  Next

Comments 51 to 100 out of 414:

  1. Jeffomatic, well, i wouldn't say that the oceans are not storing heat; look at the response to comment #40. And I would not say that "temperature is not showing it" either; see for example fig. 4 in this post or here if you like statistics more. Although we all may agree with Trenberth that "it's a travesty" that we can not track the details of the heat flow through the climate system, neverthless we can see that more heat is here.
  2. Jeffomatic writes: In fact, it's April 04, 32 F. outside and snowing right now! That kind of anecdote is not particularly useful. On the same date here in northern New England (where there's normally still snow on the ground at this point) it was in the mid-70s F (after having been an astounding 83 F the previous day).
  3. A number of people make comments about the Earth heating up without considering the massive thermal capacity of the Earth. In the following example I have exaggerated in every possible way to see what is required to raise the temperature of the planet by just one degree. I have used a very simple model of the Earth made of uniform material with reasonable conductivity If the Earth absorbed all the Suns radiation that landed on it and absolutely no heat ever escaped. How long would it take for the temperature to rise by 1 degree centigrade. Formula used Pxt =cm(temperature rise) P=1367W/m2x(crosssectional area of Earth) t =time in seconds C = specific heat capacity = 1000 (you can tweek this number if you like) m = Mass of Earth =6×10power24 When calculated it turns out to be 1080 years.
  4. how many degrees has the ocean's surface temperature risen by since 1950?
  5. amymarshall95 asks "how many degrees has the ocean's surface temperature risen by since 1950? " The US agency NOAA has data on sea surface temperatures here. For the most recent 12 months (May 2009-April 2010) they averaged 0.54 C above normal. For the 12 months of 1950, they averaged -0.09 C below normal. So SST has risen by about 0.6 C since 1950. Kelly O'Day has a very nice website with R scripts for analyzing climate data. Here's Kelly's example of a script for plotting historical SST data:
  6. suibhne writes: "A number of people make comments about the Earth heating up without considering the massive thermal capacity of the Earth. [...] I have used a very simple model of the Earth made of uniform material with reasonable conductivity If the Earth absorbed all the Suns radiation that landed on it and absolutely no heat ever escaped. How long would it take for the temperature to rise by 1 degree centigrade." Aha! Perhaps this explains why you've found the concept of AGW so difficult to accept. If you were under the misapprehension that the entire mass of the earth had to change temperature at a uniform rate, then of course it would be impossible for humans to raise the planet's temperature by 2-6 degrees C. Of course, in that model, the glacial/interglacial cycles, and shorter-term temperature fluctuations like the Younger Dryas, MWP, and LIA would never have happened, either. The earth's temperature would have to be effectively constant over time. So that's an indication that you might be misunderstanding something, right? So, here's a question for you, suibhne: where do you think the error is in your assumptions? You might also want to think about what happens when you cook a turkey in your oven. Does it cook at a uniform rate all the way through, or does it cook more rapidly on the outside and more slowly on the inside?
  7. I think John Cook and Peter Sinclair's contribution to communicating climate science to the educated layman has been indispensable. However, there comes a point of diminishing returns where the best mode of communication to the remainder of the public might not be an entirely rational one. Politicians have realised since time immemorial that mindless rhetoric and appeals to the core values and fears of the people are far more effective than hard facts. Deniers understand this better than the 'good guys' and it may be necessary to take the gloves off to reach the multitude of people. So perhaps the best way to communicate with these people is not by explaining climate change in technical terms, but to emphasise the effect of climate of migration patterns, the increased chance of their beach residence being flooded by increased storm intensity and to emphasise our need for energy security with the onset of peak oil.
  8. 304 is the number of consecutive months we have now had with temperatures greater than the mean for the 20th Century. Thats every month in the last 25 years. Every single year from 2001 onwards has been warmer than every year prior to 1998. According to the NOAA, May 2010 was the hottest May on record. Also: June was the hottest on record July was the hottest on record new records this year for: The warmest March and the warmest April The warmest January to April period The warmest January to May period The warmest March to May period The warmest Jan through June period Warmest 13 month.......................... . The NSIDC gives us the latest Arctic sea ice maximum on record (March), and The rate of Arctic sea ice decline through the month of May was the fastest in the satellite record. 2005 had (by far) the lowest Arctic sea ice extent on record at the time. It is now the 4th lowest (and likely to become 5th this year) 1997 was (at the time) the warmest year on record. It is now 12th.
  9. Phil Jones NEVER said that there had been no warming since 1995; He only said that 15 years was too short a time span to have 95% statistical certainty for the warming trendline. It was actually 92%+ statistically significant. If you ask the same question, starting with 1994 instead of 1995, you get 95% statistical certainty. And that number is more or less arbitrary as a benchmark, anyway.
  10. "Of satellites and temperatures Guest post by Ned There are a variety of rumors floating around the "skeptic" blogosphere involving claimed problems with satellite temperature measurements. Unfortunately, there is a great deal of confusion on this point." "Global Warming has stopped" video good clear debunking of this meme Global cooling -"3 levels of cherry picking in a single argument"
  11. Mizimi @ 15 "If that trend continues in a linear fashion, then by 2108 the GMT will be 15.8C assuming all other things remain equal. Not impressed" Why not? Thats well within projections of climate scientists. If the global mean temp is now 14C, and we have already increased temp by 1C in industrial times, and you add your 1.8C difference to that, you get 2.8C change since the industrial revolution. In addition, you are assuming constant rate of CO2 emissions. Without mitigation efforts, CO2 emissions will increase over the coming decades. Current projections for global temp rise average about 3C and within a range of 2-5C . The lower end are conservative estimates. We will likely see 2C even with mitigation efforts. These projections do not take into account any amplification from releasing of methane from melting Tundra or the seabed. We are now experiencing the warming and climate change from emissions of 30 years ago. You haven't seen the effects of our current emissions yet. 2.8C change in GMT will bring catastrophic climate change. At 1C change, the glaciers, ice caps and sea ice are already melting, seasons and weather have changed and extreme weather is increasingly frequent, to name just a few observable signs of global warming that is already upon us. 1000 year heat wave in Russia is a sign of how the climate is changing. No you can't attribute any weather event to climate change, but these extremes are more frequent. The last decade had twice as many record highs as record low temperatures. The record high to record low ratio has been increasing for three decades in a row.
  12. I don't know if Bob Carter would be high on my list of good sources of climate information. "Climate scientists continue to respond to badly flawed, politically driven, papers by those who deny the strong evidence for humans affecting climate in ways that portend major future disruptions". "Such papers have confused the public debate, but increasingly scientists are stepping up to provide strong refutations. Last year, John McLean, Chris de Freitas and Bob Carter, published a paper in the Journal of Geophysical Research using a mathematical procedure that eliminates long term trends to claim that there is no long term trend in global temperatures." "How Low can you Go?" April 3, 2010 McLean, de Freitas and Carter rebutted... by McLean, de Freitas and Carter that is not clearly shown in this graph and only discovered through analysis of the original data is that the mean values of the weather balloon and satellite data during their period of overlap differ by nearly 0.2°C. Splicing them together introduces an artificial 0.2°C temperature drop at the boundary between the two. In other words, they "hide the incline". and pervert.html
  13. In post 61, sailrick projects temperatures based on 2 or 3 times warming feedback of water vapor on top of the increase from CO2 (sensitivity). Sailrick claims a 30 year delay of temperature rise to CO2 increases. Did the oceans expand enough over 30 years to show that kind of heat storage? Finally sailrick acknowledges that extremes are increasing. But that means that the negative feedback is already kicking in (even before it should). Catastrophic warming from water vapor feedback (multiples of 2 or 3) only works if the water vapor is evenly distributed, not concentrated so as to produce record rains.
  14. @Eric (skeptic) #63 Bear in mind that increased water vapour (leading to increased clouds and precipitation) would also be in the (generally) much bigger gaps between any clouds. Some confuse water vapour (invisible and a powerful greenhouse gas) with condensed water vapour (steam) which makes up clouds.
  15. This is a response to fydijkstra in another thread, where he claimed there has not been any warming since 1998. fydijkstra, there has been warming since 1998, as this graph demonstrates: Second, 1998 was an exceptionally warm year. Using it as a starting point to determine a trend is cherry-picking at its best (not to mention the fact that the time frame is too short to indicate statistical significance). In spite of this, the graphs still show modest warming. However, look at what happens if we measure trends right before or after 1998: It's clear that starting in 1998 gives an erroneous idea of the actual temperature trends, which is why contrarians like to do so. That is what we call cherry-picking.
  16. When ENSO-adjusted, why is 2006 the hottest year on record? But NASA GISS shows 2005 as the hottest year on record. Which year is the hottest 2005 or 2006?
  17. The new images here should remove any doubt that those still clinging to the 'it hasn't warmed since 1998' mantra are truly in deep denial. The above is temperature anomaly for the period 2000-2009. The clincher is the companion image, both available full scale here, which represents 1970-1979, when things were a lot bluer. Be sure to check the color scale.
  18. A really nice temp chart is here at NASA Earth Observatory
  19. I recently heard that NASA and the National Oceanic and Atmospheric Administration reported that 2010 tied with 2005 as the warmest year on record. I can't believe there hasn't been more said about this in the media, particularly with so many climate change deniers citing cold snaps in some parts of the world as being evidence of global cooling.
  20. The 'further reading' links at the top of the first page are broken: "Tamino ... Garbage is Forever and Wiggles." If they've been recovered can you update the pointer?
    Response: [Daniel Bailey] Garbage is Forever can be found here while Wiggles can be found here. A compendium of recovered Open Mind posts from Tamino thought lost can be found here.
  21. "2005 was the hottest year globally, and 2009 the second hottest." According to majority 1998 was warmest on record (Hadley, RSS, UAH) and 2005 not even close. 2010 was the second warmest and tied only in UAH. GISS record years are just artifacts due to homogenization and unjustified interpolation. Again, "Skeptical" Science does cherry picking and pickes the only dataset which shows any warming on this ENSO-neutral interval which is 1998-2010.
    Response: [DB] Incorrect. According to the WMO, which uses data from NASA,NOAA/NCDC and MET/UEA, the year 2010 was jointly ranked with 2005 and 1998 as the warmest year on record. Hansen 2011 indicate we have reached the levels of the Holocene Maximum/Altithermal. Removing ENSO and other exogenous factors from the records yields (using monthly averages) this, from Tamino: And now this, using annual averages:
  22. protestant, you need to look at Monckton Myth #2: Temperature records, trends and El Nino to learn more about the different datasets and how global their coverage is. You also need to look at Are surface temperature records reliable?, before backing up your accusation of "unjustified interpolation".
  23. protestant, I notice you still haven't retracted or backed-up your accusation of "unjustified interpolation", so, to help you out, please look at the following comment. You can reply there to retract or back-up your accusation.
  24. GISS record years are just artifacts due to homogenization and unjustified interpolation. Again, "Skeptical" Science does cherry picking and pickes the only dataset which shows any warming on this ENSO-neutral interval which is 1998-2010. As far as I can see, every statement in here is incorrect. If you are going to make a claim like this, then please present data to support it. This appears to be channelling of some pseudo-skeptic site (like other of your posts). Tell me, how many articles on such site would we have to refute before you realized that you were being suckered and stopped reading it? I'm curious.
  25. In my opinion, the evidence is just not there and more data is needed. For example, scientists have always led us to believe that tree rings are a sign of global warming as they represent temperature changes over the years. Yet there is simply no evidence to support this! It is caused by Lunar Cycles and Solar Flares (of which there is to be the largest one on Monday night). For anyone else who is interested in this kind of thing, I would suggest this website: It's an informative read for skeptics and anyone, like me, who questions what we are told about climate change and global warming.
    Response: [Daniel Bailey] Welcome to Skeptical Science! There is an immense amount of reference material discussed here and it can be a bit difficult at first to find an answer to your questions. That's why we recommend that Newcomers, Start Here and then learn The Big Picture. I also recommend watching this video on why CO2 is the biggest climate control knob in Earth's history. Further general questions can usually be be answered by first using the Search function in the upper left of every Skeptical Science page to see if there is already a post on it (given the plethora of posts [I get paid extra for using big words and alliteration :-) ] odds are, there is). Or you can search by Taxonomy. If you still have questions, post them on the most appropriate thread & someone will attempt to help you. Please adhere to the Comments Policy when composing your comments and remember to use the Preview function before submitting. I'm afraid the vast majority of your comment is simply incorrect. The warming of the globe is an accepted fact. That humans are causing a good part of it is accepted at over a 90% scientific certainty level.
  26. The Tamino post in further reading is no longer active. Fortunantly, the wayback machine can help us.
    Response: [Daniel Bailey] Thanks; that one and many more recoverable lost Open Mind posts are linked at the Archive.
  27. I have been struggling with a concept that seems to be conceded by [--snip--] and [--snip--] alike: that the super el ninio of 98-99 raised global average temperature. At what time scale is the global energy budget reconciled? If the answer is anually, to concede this point means that the ocean-atmospheric system had to draw down some "fossil" savings account to boost the temperature.
  28. Decades? Ocean cycling is something close to 1000 years. ESNO amounts to ocean/atmosphere heat transfer. You dont see much sign of "draw down" on the OHC to 2000 meters shown in #40 of this thread, let alone total global energy budget.
  29. My best understanding is that el ninos are thought to be caused by an atmospheric failure of trade winds into the ITC in the easern Pacific, depriving the tropical Americas of the oceanic Ekman transport that fosters upwelling. The reverse would presumably be true for la ninas. If I may be forgiven an observation as a naturalist rather than an expert, I have been watching the SST anomalies in this years la nina and it appears to me that the cold water is emerging in a standpipe plume (a point source)in the western Pacific and splaying eastward against the prevailing winds and currents. If the ENSO and possibly other ocillations have deep enough pockets to skew global average temperature that much, and the periods of the ocillations appear irregular, how do we know anything at a decadal scale?
  30. Its a coupled ocean/atmosphere phenomena. Tricky to assign a "cause" to the atmosphere. Why do the trades fail? However, the point really is that upwelling is heat exchange. Look at total OHC. If the current warming was just an ocean cycle, then why is total OHC increasing? So far decadal prediction eludes us. That's why climate is defined in terms of 30 year averages. What the overprint of global warming tells you is that the temperature of the peaks in ENSO events of the same magnitude is increasing.
  31. scaddenp at 06:57 AM, if total OHC measured 0-2000m is increasing, whilst OHC 0-700 is not, then what mechanisms are allowing this to happen. If it is due to the atmosphere/ocean interaction, then SST's will show that, but they don't, actually trending down the last 8 years or so. For the deeper ocean 700-2000m to be accumulating heat whilst 0-700m is not, it can only mean, assuming measurements are adequate and accurate, that whatever solar radiation being absorbed in the upper layers is either being offset somehow or being speedily transferred to the lower levels. But that amounts to a lot of heat that has to transfer through the surface, and then 700m of water, without leaving a trace, to deposit enough heat in 1300m of water, enough to cause the average OHC of 2000m of water to increase.

    [DB] "If it is due to the atmosphere/ocean interaction, then SST's will show that, but they don't, actually trending down the last 8 years or so."

    Demonstrably false:

  32. Convection? More marked in upper ocean. As SSTs are not trending down. The upper 700 do show ESNO variation. I don't see the problem here.
  33. Response: [DB] "Demonstrably false:" Demonstrably wrong! The image you posted is the anomaly for the entire period Feb 2003-2011 against the average 1951-2002, so it is irrelevant to the very simple observation I had made. This image, even at the scale it is, allows the downtrend to be observed. Perhaps you could instead produce actual data for the last 8 years that can be examined in finer detail if you believe that this image does not support my assertion.

    [DB] Wreeennncccch (sounds of metal protesting as goalpost are moved)!

    As it stands, your initial phrase was false, so I pointed it out. And it still is, as you specified no comparator period. Where your logic falls apart is in your cherry-pick of a short period of time that is statistically insignificant. So let's recompute, using the period shown in your graphic (why 60S-60N when more data is available?):

    Yup, still false.

  34. Ignoring for a moment the futility and folly of using short-term trends (less than 15 to 20 years), consider the linear trend in global SSTs from Hadley for different periods of time. Note what happened between 1979 and 1987. Now can we please stop playing this juvenile and unscientific game of cherry picking "skeptics"?
  35. DB, the reason it still appears false to you, is that you haven't changed anything. It is still the complete period Feb 2003-2011 that you are using as a benchmark instead of examining what is plainly evident within that period. I think that you took your eye of the pea whilst trying to switch thimbles. There was no comparator period necessary, the trend of the period covered by the last 8 years is the subject in question. The challenge still stands, if you believe that there is no downtrend evident over the last 8 years then produce the relevant data that can be analysed. If you are now going to start claiming 8 years is insignificant, then why did you start using a 8 year period to try and make a point. Further to the point of being consistent, if data beyond 60S-60N is that important to you then why did you start posting maps that indicate no data beyond that point. Also it would help tremedously if your replies were made in the same manner as all other participants so that your posts appear at the top of the comments page as they are made. That also allows the order of each response being posted to be followed if replies to different posts, and different posters, are being made.

    [DB] You've made an extraordinary claim, were called on it and have since hastily waved hands while backtracking. The burden of proof is on you to prove your claim with some actual analysis, not for others to disprove it.

  36. johnd #85: "if you believe that there is no downtrend evident over the last 8 years then produce the relevant data that can be analysed." Produced here and here. Just look up this very thread or at a number of other 'global warming stopped in ... ' threads. If you're going to be a serious skeptic, you must try harder.
  37. muoncounter at 11:33 AM, your "here and here" links do not lead to SST data.
  38. DB, on the comments page your responses are labeled as below. Moderator Response: [DB] You've made an extraordinary claim, were called on it and have since hastily waved hands while backtracking. The burden of proof is on you to prove your claim with some actual analysis, not for others to disprove it. ......................... ( -Moderation complaints snipped- ). ( -Moderation complaints snipped- ). I have not backtracked at all. The assertion I made was very, very simple and straight forward. It is you in fact that is backtracking, having first producing irrelevant information in order to challenge my assertion, now having thrown your hands in the air and deciding it's impossible and trying to turn it around. I have nothing that disproves what is evident on the graph I supplied, sorry. End of story I think.

    [DB] Again, you made an (unfounded) assertion simply unsupported by the science and data at hand, you were called on it and have done little since but wave your hands, backtrack and complain about it.

    It is up to the asserted to support claims with published peer-reviewed literature and/or analysis of their own which supports their contentions (this is called science). Which you have not done and complain about having to do. That is your choice and right.

    The readers of, and the participants in, this forum remain skeptical of your claims until then.

  39. johnd #87: No, those links point to global temperature anomalies. Following the data trail, one finds oceans included in those global temperatures. More data are better, yes?
  40. muoncounter at 12:40 PM, only for those unable to distinguish the difference between quality and quantity. You were saying......?

  41. I find this a little irrelevant but I am a little out of depth. To me the "its the ocean's" argument runs like this. "An ocean cycle is causing the warming" Okay, so where is the energy coming from. "Well out of the oceans of course" So if is warming is just cycle of ocean atmosphere energy exchange, then why is total OHC increasing? I thought is was supposed to lose energy to atmosphere. Someone with more knowledge of this can tell me what I am missing.

    [DB] Good question. If it's the ocean (discussed here), then OHC anomaly should be net zero. Since it isn't (it's actually positive, as you point out), than the ocean is retaining energy/heat (because of the Earth's radiative imbalance), despite the continual loss of heat to the atmosphere and to the melting of Arctic ice (Patrick Lockerby just released his March Arctic Ice Update #2).

    See the links I just gave you for relevant discussion; a search on the term "It's the ocean" yields these results. Hope that helps.

  42. RE: #7 response: Response: The cool temperatures of Jan 2008 are due to an unusually strong La Nina effect (the strongest in a decade). Did the models predict this "strongest in a decade" La Nina and if so could you point us to the statement where this was predicted before it actually occurred
  43. Bruce @92, The forecasts for 2010-2011 can be found here for IRI. Also see here for CPC.
  44. Bruce#92: "Did the models predict this "strongest in a decade" La Nina" What an utterly irrelevant question. Climate forecasts deal with long term trends, not transient weather events. To answer the question of this post, see Tamino: No, global warming did not stop in 1998: "Big wheel keeps on turnin'"
  45. I would like to ask if anyone is aware if there have been any refutations of this paper by Stockwell and Cox? I have been following the climate debate for some time - especially on Drum. I really have no truck with Cox as I find his approach to be dishonest and dissembling. But I have to say I find this paper to be pretty sound? I really hate to give Mr Cox any credence or airtime at all if it's not justified. But I also believe in maintaining an open mind. My own approach to AGW was one of a genuine sceptic. When I first encountered suggestions of AGW back in the late 80's as a newly minted PhD in physics I was fairly dismissive - I felt in particular that the heat capacity of the ocean and solar irradiance variations would be much more powerful forces. But in the early "noughties" as the issue gained more attention I did my own secondary research (the wonders of the internet) and one by one my doubts were addressed by the evidence. Actually your site with the neat little thermometer down the side is a fairly apposite reflection of my own doubts that have been addressed. I am now convinced of the realities of AGW and have been since about 2003. Those who deny the science really upset me for their ignorance and blindness bias but I always try and maintain an open mind. I confess I've sometimes been fairly "direct" and forceful in some of my comments posted on the Drum. Over the years I've checked on many so called "refutations" of AGW in the "literature" - always to find that either the publication is highly suspect (such as "Energy & Environment" or anything published by Bentham Science ) or if the journal is good the paper gets comprehensively refuted - such as the case with G. Gerlich, R. D. Tscheuschner: Falsification Of The Atmospheric CO2 Greenhouse Effects Within The Frame Of Physics. International Journal of Modern Physics B, Vol. 23, No. 3 (30 January 2009), 275-364 - I'm AMAZED that even got published - complete piffle but I guess the technical "density" (pun intended) of it proves that "bullshit baffles brains". I'd like to think my "science bullshit" detectors are pretty acute but this paper by Stockwell and Cox actually seems sound to me. The journal is reputable. The Chow test is very valid, and whilst it was developed I believe for econometric data it can really be applied to any time series analysis. Also, looking at the temperature data one can reasobaly conlude that there do appear to be step changes and therefore it's not an "invalid" basis for forecasting. IF they are right it doesn't suggest AGW isn't real - only that the size and scope of the problem is much lower than we feared and most scientists have suggested. I can't see that they've made the usual "error" of cherry picking their data points (a la Carter) Given that this issue - how far/how fast will we warm - is I think the one genuine area where there is room for scientific uncertainty then I must, out of intellectual honesty, pay attention to their findings. Of course Stockwell and Cox don't explore what mechanisms might be at work that may have mitigated warming lately - they are only interested in creating doubt not genuine physics. I might even speculate that the recently reported grand solar minima may be at play, not to mention that aerosol effects may indeed be more than we had thought - but at this point have not been able to find any papers that have really examined that hypothesis - namely that warming may be temporarily in a reprieve due to such effects. I have found on New Scientist articles that suggests the solar minima phenomenon may at most mitigate things by 0.3 degrees - but I'm not a subscriber so can't access the full article Anyway - love this site. But in the genuine spirit of open minded sceptical inquiry I would really like to know if anyone is aware of any rebuttals of the paper above or can suggest any ideas?
  46. The abstract of Stockwell and Cox suggests a rather poor grasp of multiple hypothesis testing and the nature of cherry picking. Finding a statistically significant result does not refute the possibility of cherry picking. If you look at 100 independent events (e.g. trends at different site), at the 95% level you would expect to see five statistically significant events even if all trends were due to random chance. Cherry picking is about searching for events that make your argument, in contradiction to the broader picture; this is still possible if you restrict yourself to "statistically significant" events. Also it is rather unsurprising that there is a change in the datasets around 1997. In 1998 there was an extremely strong El-Nino event. It is difficult to detect the difference in a noisy dataset between a step change and a linear trend with a quasi cyclical variability (ENSO). Physics provides an answer, the linear trend model has a plausible physical explanation, can the same be said of the step change model? I'll comment again when I have read the rest of the paper. This is just my impression from the abstract.
  47. Having skimmed the paper, it looks to me like another case of someone thinking stats trumps physics; they are wrong, it is the other way round (I am a statistician, and I greatly prefer a model with a strong physical explanation than merely statistical support). The Chow test is not reliable in this case, as it is based on sum-of-squares errors, which implicitly assumes a Gaussaan noise process. However the noise process for climate data is non-gaussian; it has a quasi periodic cyclic component, due to ocean circulation such as ENSO. Now if they were to find a step change in the data after it had been adjusted for the effects of ENSO (i.e. look at the resudials of a regression of temperatures on MEI, as has been done frequently), that would be a different matter entirely. I very much doubt that the Chow test would identify a step change in that case, which would mean that surface temperatures are affected by ENSO, something we have known for decades. The reason that Easterling and Wehner find periods of low warming or cooling is exactly that - internal climate variability. Secondly, it is obviously cherry picking to start a trend at a conspicuous maximum. This is true even if you include statistical significance becuase if you optimise the start date to bias the result in a particular direction (e.g. by choosing to start at a maxima) then it invalidates the significance test anyway. Essentially you are performing many simultaneous significance tests, one for each start point. Each test has a probability of a false-positive or a false-negative. If you pick and choose the test to make the argument you want, you vastly increase the overall chance of a false-positive or false-negative, and hence it is cherry picking. In stats this is the "multiple hypothesis testing problem". Lastly, using the start point of the dataset is not cherry picking, it doesn't mean the results are robust to the choice of start point. It is not reasonable to expect a linear trend in temperatures on a centennial scale; climate forcings have changes in many ways over that period - solar forcing explains much of the waming in the first half of the 20th century, aerosol cooling explains a mid-century plateau and CO2 radiative forcing has become dominant from the late sixties. So the Chow test is a bit of a straw man. It is always going to find a break point, simply because the linear trend is known to be wrong. That doesn't mean that it is a step change though, that is just the only alternative offered to the straw man of a linear centennial scale trend. I suspect that if they had started in say 1960 (giving a period where CO2 radiative forcing is asserted to be dominant according to e.g. the IPCC), I suspect the Chow test would no longer identify a step change because a linear trend over that period is more plausible. The paper shows a lack of self-skepticism. Has the paper been published in a peer reviewed journal? If so, I'll give it more than just a skim, but as (essentially) a statistican, I am not impressed by it.
  48. Yes I've come across this Stockwell and Cox paper before. Aside from the dubious affiliation (I mean has anything scientifically worthwhile come out of the Institute of Public Affairs?), it just seemed to me like a bunch of statistical description with a very limited attempt to place the data in the context of a credible physical model.
  49. The paper was submitted to Journal of Forecast but hasn't been published, not there at least. The statistics leave much to be desired, physics is absent; but even if they were sound, they discovered that the climate in Australia is sensitive to no less than the nearby ocean! How could they come to global conclusion and even make projections remains a mistery.
  50. Cheers Riccardo (good summary ;o). If it does get published, it would be good if a peer-reviewed response were submitted. Submitting it to the Journal of Forecast is a neat trick, they wouldn't be able to get it published in a climate journal because of the lack of physics, the reviewers at JoF are likely to be only able to review the stats, but not the significance of the findings to climatology. Obviously that won't stop some drawing firm conclusions about climate from it.

Prev  1  2  3  4  5  6  7  8  9  Next

Post a Comment

Political, off-topic or ad hominem comments will be deleted. Comments Policy...

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

Link to this page

The Consensus Project Website


(free to republish)

© Copyright 2022 John Cook
Home | Links | Translations | About Us | Privacy | Contact Us