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 Mastodon 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


We’re now breaking global temperature records once every three years

Posted on 23 January 2017 by dana1981

According to Nasa, in 2016 the Earth’s surface temperature shattered the previous record for hottest year by 0.12°C. That record was set in 2015, which broke the previous record by 0.13°C. That record had been set in 2014, beating out 2010, which in turn had broken the previous record set in 2005.

If you think that seems like a lot of record-breaking hot years, you’re right. The streak of three consecutive record hot years is unprecedented since measurements began in 1880. In the 35 years between 1945 and 1979, there were no record-breakers. In the 37 years since 1980, there have been 12. The video below illustrates all of the record-breaking years in the Nasa global surface temperature record since 1880.

Nasa global surface temperature data, with record-breaking years shown in red. Created by Dana Nuccitelli.

These include a spate of five record hot years between 1937 and 1944; however, ongoing research is investigating whether some of those are artificial, due to changes in the way temperatures were measured during World War II. 

Even including World War II, in the first 100 years of the Nasa data, the high temperature record was broken seven times. It’s been broken seven times in just the past 20 years.

This rapid rate of record-breaking heat (once every three years) is consistent with climate scientists’ expectations. A 2011 paper by Stefan Rahmstorf and Dim Coumou found that as global warming continues, we should expect to set new records about once every four years.

Indeed, if we only use the data of the past 30 y, these show an almost linear trend of 0.017°C/y, yielding an expected 2.5 new record hot temperatures in the last decade [1 per 4 years].

Rahmstorf told me that so far, the rate of record-setting temperatures is in line with a consistent human-caused global warming trend:

There is no statistical evidence for recent acceleration, just as there never was statistical evidence for a “slowdown” in global warming before the recent series of records. It’s all still within the noise (which could well be hiding an acceleration, but we cannot tell yet from these data).

Global temperature wasn’t the only record-setter in 2016. Global warming causes climate change, and North America saw its highest number of storms and floodsin over four decades. Globally, we saw over 1.5 times more extreme weather catastrophes in 2016 than the average over the past 30 years. Global sea ice cover plunged to a record low as well. California endured a fifth consecutive year of its worst drought in over a millennium. A drought also savaged the maize harvest in Southern Africa, causing a famine. The list of climate consequences goes on.

Alternative Facts

To paraphrase Taylor Swift, “deniers gonna deny.” As expected, the anti-climate policy advocacy group GWPF claimed that the record-breaking global temperatures somehow prove that global warming has “paused,” because, they argue, the record was only set due to an El Niño event. Not so, say mestatisticians, and climate scientists.

Quite obviously, human-caused global warming is the driver behind these frequent record-breaking hot years. Usually an El Niño event will help push a given year over the top, as happened in 2016 and 2015 (but not 2014). However, today’s El Niño years are hotter than past El Niño years because of global warming.

Nasa global surface temperatures divided in to El Niño, La Niña, and neutral years, with linear trends shown for each.

Relatedly, testifying before the Senate on the day the record temperatures were announced in a bid to become EPA administrator, Scott Pruitt claimed that we don’t know how much humans are contributing to global warming:

This is simply wrong. We wouldn’t be setting a new temperature record every three years if not for global warming, and there’s no question that the warming is predominantly human-caused. The latest IPCC report stated with 95% confidence that humans are the main cause of global warming since 1950, and most likely responsible for 100% of that temperature rise. 

Every study quantifying the various contributions to global warming has found humans are the dominant cause. Our fingerprints are all over climate change – the changes are precisely in line with what we’d expect to see as a result of an increased greenhouse effect from human carbon pollution. That’s why there’s a 97% expert consensus on human-caused global warming.

Climate denial is evolving. All of Trump’s nominees rejected his claims that climate change is a hoax, but all cast doubt on the degree to which humans are contributing, and to the threats it poses. It’s a softer, cuddlier form of climate denial that doesn’t reject all scientific research – just the vast majority – and yields the same end result of obstructing climate solutions.

Click here to read the rest

0 0

Printable Version  |  Link to this page


Comments 1 to 18:

  1. “deniers gonna deny.”

    The 'latest'?

    “Unadjusted and satellite records combine to suggest that there has been no more than 0.6 degrees C warming since the middle of the 19th century.”

    0 0
  2. Postkey @1, Breitbart's source is Tony Heller of Real Climate Science, now renamed "The deplorable climate science blog" (the one accurate thing about it).  Tony Heller is one of the worst of the climate deniers.  His output does not even rise to the level of pseudoscience, being rather a straightforward series of outrageous distortions of the data.  His style can be shown by another recent article on the popularity of Trump, where he accuses "fake news networks" (by which he means the mainstream news) of reporting Trumps approval rating at 32% whereas, he assures us, it is actually 56%.  In fact, the media have been reporting approval ratings around 40%, some based on a Fox Poll showing 42% favourable to 55% unfavourable.  Looking at RealClearPolitics shows what is happening.  Their average of 9 polls shows 41.8% favourable, to 49.9% unfavourable, but with the Rasmussen poll as a clear outlier, and the only one to show a net positive approval rating for Trump.  Heller, of course, reports only the Rasmussen poll.  Thus, by first distorting the news reporting, and then cherry picking his data, he purports to find media bias, wheras he only demonstrates his own.  (I would go much further than that were it not for the comments policy.)

    I another recent example, he indicates he has "Smoking gun of fraud by NASA and Gavin Schmidt".  The "evidence" is a report on Real Climate claiming there was no pause.  This is compared to a scientific paper, "Making Sense of the early 2000s slowdown" which said:

    "It has been claimed that the early-2000s global warming slowdown or hiatus, characterized by a reduced rate of global surface warming, has been overstated, lacks sound scientific basis, or is unsupported by observations. The evidence presented here contradicts these claims."

    Heller claims, based on the Real Climate article that "Now, the same people claim there never was a global warming pause or hiatus, and it was just a “red herring.”  

    Heller's evidence that these "are the same people" is that Michael Mann was both a co-author of the scientific paper and is a contributor to Real Climate.  But a one person overlap hardly counts as "the same people".  Nor is even that overlap relevant given that the Real Climate article was written, not by Mann, but by Rasmus Benestad.  To summarize, Heller accuses Schmidt of fraud because one of his colleagues (Mann) supported one opinion in a scientific paper, and a different one of his colleagues supported a diferent opinion, later, and with more evidence.  Needless to say, as an evidentiary basis this is beyond weak.  As evidentiary support, it ressembles more a magic trick where he uses random facts to distract the audience from the fact that his thesis has no support at all, but is rather simply an unjustified slander.

    Given that background, it will be no surprise that his evidence, rellied on by Breitbart, consists only of appending the running five year mean of the 2000 version of GISTEMP up to (approximately) 1995 to the running five year mean of RSS TLT up to 2014.  That is, he has cherry picked the satellite data with the lowest trend knowing full well it does not measure surface temperatures and appended it to a surface temperature record whose endpoint is reduced to early 1990 levels (ie, Pinatubo volcano levels) by the inclusion of that volcano but not the 1998 El Nino in the averaging frame.

    In short, it is just more prestidigitation.

    No surprise then, that it should be echoed on a fake news site like Breitbart. 

    2 0
  3. Thanks for that.

    2 0
  4. Tony Heller, aka Steven Goddard. The guy who claimed it snows CO2 atop the Antarctic ice dome. The guy who claimed the reason Lake Superior is cold is it "remembers" the last glaciation.

    'Nuff said.

    0 0
  5. How much hotter was 2016 ? I read at the beginning of the the piece that 2016 broke the 2015 record by 0.12 degrees celsius according to NASA. Yet I have before me a news item taken from the ABC web site January 19th this year where the temperature rise is cited at 0.04 degrees celsius. The news item does not clarify which temperature data set that figure is taken from. I go to the Climate Council web site for more information and yes 2016 was the hottest year again by 0.04 celsius. Reading on I discover that their figure is based on NOAA temperature data. Throw in the UK Met Office and lower troposphere satellite data and it all becomes a bit of a dogs breakfast. Would it ever be possible when it comes to global surface temperatures for the NASA's, NOAA'S ect to collaborate come up with one figure ? Then there is the satellite data much loved by climate change deniers, but on this forum often accused of being problematic and unreliable. Did their 2016 data better that 1998 spike ?  In the end it all comes down to clear communication. 

    0 0
  6. Richard McGuire @5, the four major surface temperature records show the following differences between 2015 and 2016:


    Berkeley Earth Surface Temperature LOTI 0.078 C

    NOAA LOTI: 0.04 C

    HadCRUT4: 0.013 C 

    Of these, NOAA and HadCRUT4 do not significantly cover polar regions, with HadCRUT4 also missing significant parts of Africa, the Middle East, and Australia.  Apparently the Arctic was unusually hot, so that would account for the low values of those two relative to GISSTEMP and BEST.  Further, BEST uses more stations, and probably has a better statistical technique than does GISSTEMP.  NOAA and GISSTEMP use almost the same stations, with just a few extra for GISSTEMP, with GISSTEMP using the better technique; and HadCRUT4 has significantly less stations than the other three, and probably the worst technique.

    Finally, each dataset has its own error margin.  Further, at this stage each is liable to further, small adjustment as data from late reporting stations comes in.  For both reasons, the figures should be considered as indicative rather than set in stone.  

    1 0
  7. Thank you for the information Tom Curtis. However given how politically charged the Climate Change issue is, and with the Trump administration poised to set any global progress in mitigation back decades I don't think it is good enough. If the GISSTEMP's, NOAA's and HadCRUT's can't collaborate on their temperature data collection, maybe they could co-operate when it comes to releasing such important information to the media. As things stand it is confusing. Contradictory information gets out there, GISTEMP's 0.12 c verses HadCRUT's 0.013 c. The average person out there has no idea who has the most Arctic weather stations, nor should they need to know. Must somehow do better.   

    1 0
  8. Richard McGuire @7, I disagree.  The most important thing in science is replicability.  The different temperature products represent four different replications of the same measured quantity.  That four different methods, using different (if significantly overlapping) data produce essentially the same result shows the result to be robust.  Where they to cooperate by using precisely the same data and precisely the same method, the different "products" might represent audits of each other, but they would not be replications.  In scientific terms, they would be of no greater interest than if their were just one team, with no replication.

    As to joint press releases, ie, cooperating when they release the data, GISTEMP and NOAA already do hold joint releases, but the other teams are not US government agencies and should not be expected to do so.  Further, how far would this cooperation need to extend.  Would we need the (very infrequently cited) Japanese version?  What about the 5 or 6 odd versions put out by private individuals?

    If you actually look at the media statements by the respective groups, they tend already to indicate why there are major divergences between the groups (if there is).  They also tend to indicate the uncertainty.  What we need is an expectation that the press report those relevant details, rather than that the scientists should self censor.

    2 0
  9. Richard McGuire @7 and Tom Curtis @8, it seems to me that the spirit of what RM is suggesting is not necessarily to scrap replication, but for someone to distill the conclusions in a way more digestible by the average person.  

    In that regard, would it be feasible (and meaningful) for the groups' results to be synthesized in some way, to produce a quantitative "best estimate" of average surface temperature derived from the four (or however many) primary estimates, taking into account differences in their respective methodologies when determining how best to combine them?  Each of the four independent "products" could still be reported, in addition to an aggregate/synthesized "product".  

    As I (by no means an expert on climate science, physical measurement, or statistics) write and edit and re-edit this question, feeling the need to pack in more and more parentheticals and qualifiers as I go, I become more and more convinced the answer is surely no.  I suspect the task of combining the products in a meaningful way would be monumental, and at the end it would produce a result with no coherent methodology behind it, sort of divorcing the number from the physical quantity it represents.  It seems better to keep each curve separate, perhaps presenting them on a single graph and explaining how scientists interpret any apparent differences, as is essentially being done by honest and diligent climate news sources.

    Another thing, whatever group took the four temperature curves and synthesized them into one sounds an awful lot like the IPCC, and as such we already know would be looked upon by deniers with great suspicion as part of a massive global conspiracy.

    0 0
  10. "The latest IPCC report stated with 95% confidence that humans are the main cause of global warming since 1950, and most likely responsible for 100% of that temperature rise."  

    I would add here (Dana correct me if I am wrong) that the precise meaning of the statement is that humans are most likely responsible for at least 100% of that temperature rise.  As phrased in the article, the average person might interpret 100% as the physical maximum humans could have contributed.  However, I understand that not to be the case, as the human contribution would be more than 100%  of the temperature rise if the aggregate impact of natural forcings (TSI, ENSO, volcanic activity etc.) would have been a temperature decline, but human activity more than made up for those factors.  And if we parse out GHG emissions from other human contributions, we could add aerosols to the list of forcings with a cooling tendency, which were overcome by human GHG emissions and anything else on the warming side.  

    0 0
  11. My comment @10 is of course not for Dana's benefit, who knows this a lot better than I do, but for visitors to the site.

    0 0
  12. And now the rich and super rich are trying to buy up bolt holes in New Zealand to escape the chaos they have caused.  It would be interesting to research, from the America side, which of these paragons of virtue are trying to prepare a place where they can escape the mess they have caused.  They have caused the ship to sink and now they want preference in the life boats.

    0 0
  13. jdixon1980 @9, perhaps what you are looking for is something like the World Meteorological Organization's press release on 2016 temperatures.  They state:

    "The globally averaged temperature in 2016 was about 1.1°C higher than the pre-industrial period. It was approximately 0.83° Celsius above the long term average (14°C) of the WMO 1961-1990 reference period, and about 0.07°C warmer than the previous record set in 2015.

    WMO uses data from the US National Oceanic and Atmospheric Administration, NASA’s Goddard Institute for Space Studies and the UK’s Met Office Hadley Centre and the University of East Anglia’s Climatic Research Unit. WMO also draws on reanalysis data from the European Centre for Medium Range Weather Forecasts and the Copernicus Climate Change Service, which use a weather forecasting system to combine many sources of data to provide a more complete picture of global temperatures, including in Polar regions."

    And later:

    "Notes to Editors

    Global temperature anomalies are computed using three global datasets: HadCRUT4, jointly produced by the Met Office Hadley Centre and the Climatic Research Unit at the University of East Anglia, United Kingdom; the GISTEMP analysis (2016 version), produced by the National Aeronautics and Space Administration Goddard Institute for Space Studies (NASA GISS); and the NOAA Merged Land Ocean Global Surface Temperature Analysis Dataset (version 4.0), produced by NCEI. WMO also uses European Centre for Medium Range Weather Forecasting ERA-Interim reanalysis data. Copernicus ECMWF press release available here. Differences between the datasets arise largely from the way that the data-sparse polar regions are handled. There is a margin of uncertainty of +- 0.09 °C in the anomaly value.

    A number of definitions exist for the pre-industrial period, the most commonly used being 1850-99 and 1880-99. The value of 1.1 °C is valid (to the nearest 0.1 °C) whichever of these periods is chosen."

    For completeness, the ERA (ie, the European Centre for Medium Range Weather Forecasts and the Copernicus Climate Change Service) data is a reanalysis product.  That is, it is produced by running a climate values with cell values slaved to observed values were known, which allows a prediction of unobserved values.  It showed global temperatures to be nearly 0.2 C greater than 2015.

    I am not sure how they derive the 0.07 C figure.  It is too high to be the simple mean of the three observationatal values (which rounds up to 0.06 C), and too low to be the simple mean of the three observational plus ERA interim records (0.09 C).  For the record, the simple mean of the four observational records including BEST rounds down to 0.06 C. 

    2 0
  14. Tom Curtis @8 "That four different methods using different (if significantly overlapping) data produce the same results shows the result to be robust" With due respect Tom the only same result produced was that 2016 was warmer than 2015, therefore the warmest year on record. By how much is open to debate. Does this matter ? In some cases, such as in 2014 it does. This article that appeared on the ABC's Drum in 2015 may help explain my point better 

    0 0
  15. Richard McGuire @14, you are being inconsistent.  You point to an article which claims that uncertainties are important (they are), and that consequently 2014 was not the hottest year todate as of the end of 2014, but a statistical tie with two other years for that title.  But then you insist that the four observed 2016 values from my post @6 are different, even though they are within uncertainty of each other, and will within uncertainty of the mean value.  If Walker's article on 2014 has merit, as you suggest, then the four records are effective replications of each other.

    This is more obvious when the full overlapping record is shown:



    Indeed, quite independent of the merit's or Walker's article, replication cannot achieve an accuracy greater than measurement error.  It may happen by coincidence that two values from different observations are identical, but that is no more significant than if they were further appart, but within error.  That is because we do not know the observed value any more accurately than the limits of measurement error, so exact coincidence my be pure coincidence.  But two values falling within measurement error of each other counts as a replication because, with the information to hand the real, underlying value may well coincide, with the difference between observations being down to error.  Indeed, as examples from other sciences show, particular obserations may lie outside measurement error, and still count as a replication provided they are one of a set of observations that progressively (but not monotonically) close on a particular value.  That is because we can also be in error about the magnitude of the measurement error. 

    So, with that in mind, I will note that while Walker's more general point about the need for more scientifically accurate reporting is correct, it is not correct to claim that 2014 did not hold the temperature record as of Jan 2015.  That is because records, in all fields of endeavour, are based on nominal values, without concern about uncertainties.  Put simply, 2014 was the warmest year on the record, but the record is known to be slightly inaccurate due to uncertainty, so that the warmest year on record may not be the actual warmest year.

    1 0
  16. Richard McGuire @7: in terms of what the average person needs to know, consider the analogy of investing for retirement. Some investment advisors tell their clients not to check the value of their portfolios every day, but rather to trust in the market fundamentals that guide an advisor's long-term investment strategy for the client. Stock prices fluctuate randomly from day to day and year to year, but all that really matters for the buy-and-hold investor is the long-term performance.

    Our present response to the future threat of climate change does not depend on knowing precisely how much hotter the global average surface temperature was in 2016 over 2015. In the year 2050 nobody will care about that. Since we only have one available planet to inhabit, we are all long-term investors in its future. The average person needs to focus on the fundamentals, not the noise.

    The challenge is to persuade people to give up something that gives them real, tangible value now, to avoid causing abstract harm to other people and other species in the future. For example, if someone takes a holiday flight to Cancun or Tahiti, s/he experiences undeniable and immediate hedonic rewards. The several tonnes of carbon dioxide equivalent s/he dumps into the atmosphere to get those rewards will go on incrementally heating the climate system for centuries.

    It's similar to smoking cigarettes for short-term hedonic reward, except that with greenhouse gas pollution, it's like smokers giving cancer to someone else they've never met.

    There's no way to mask this moral quandary with any combination of policies or rhetoric. To avoid wrecking the future, people today must change their values drastically, to get the average per capita carbon footprint below the globally equitable emission allowance. The rise of Trump reflects the average person's refusal to do this. Trump represents the interests of everyone who wants to keep flying to Cancun - or to a scientific conference, for that matter (instead of figuring out how to virtualize the conference).

    Even in the liberal enclaves of California, Oregon, etc. nobody is shutting down the highways and airports. Given that a single long return flight causes a whole year's worth of allowed emissions for one person (leaving nothing for other activities such as eating), we can't have any flying if we want to stabilize the climate. We need many other equally drastic adjustments, but I focus on flying because it is one of the most egregious, least necessary, and least equal ways in which humans assault the climate. Flying is a good test case of our seriousness about mitigating climate change. It will be much harder to cut emissions from things everybody needs, such as agriculture.

    The level of response we need isn't going to be motivated by figuring out whether last year was 0.12°C or 0.04°C hotter than the year before. By analogy, imagine trying to abolish lynchings or stonings by determining whether the victim dies in 345.4 seconds or 345.7 seconds. The only way to durably eliminate lynchings and stonings is to persuade people that participating in them is morally unacceptable. As Trump is showing, merely passing laws accomplishes nothing that the next election cycle can't destroy.

    2 0
  17. I have no scientific qualifications, but have keenly followed the debate over many years, from the hockey stick to the hiatus debate. Have watched and engaged climate change deniers who have cherry picked temperature data, especially UHA and RSS satellite data. The climate change debate remains highly politically charged. Which is why temperature data that is comprehensible to the media and public at large is so critical. So 2016 was warmer than 2015. The next obvious question is by how much ? Debates about margins of error and which of half a dozen temperature data sets is closer to the mark is unlikely to cut through in the wider world.

    0 0
  18. I think that is really the wrong question. Year to year it is just weather. What matters is trend, and trend over a 20-30 year period if you are talking about surface temperatures. That is what the climate models really predict. If you want to talk about shorter time periods, then need to look at something less noisy that surface temperature. Ocean Heat Content for instance.

    1 0

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 2023 John Cook
Home | Translations | About Us | Privacy | Contact Us