Climate Science Glossary

Term Lookup

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

Settings

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

Settings


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

Home Arguments Software Resources Comments The Consensus Project Translations About Donate

Twitter Facebook YouTube Pinterest

RSS Posts RSS Comments Email Subscribe


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



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

Latest Posts

Archives

The 5 characteristics of scientific denialism

Posted on 17 March 2010 by John Cook

A fascinating paper well worth reading is Denialism: what is it and how should scientists respond? (Diethelm & McKee 2009) (H/T to Jeremy Kemp for the heads-up). While the focus is on public health issues, it nevertheless establishes some useful general principles on the phenomenon of scientific denialism. A vivid example is the President of South Africa, Thabo Mbeki, who argued against the scientific consensus that HIV caused AIDS. This led to policies preventing thousands of HIV positive mothers in South Africa from receiving anti-retrovirals. It's estimated these policies led to the loss of more than 330,000 lives (Chigwedere 2008). Clearly the consequences of denying science can be dire, even fatal. 

The authors define denialism as "the employment of rhetorical arguments to give the appearance of legitimate debate where there is none, an approach that has the ultimate goal of rejecting a proposition on which a scientific consensus exists". They go on to identify 5 characteristics common to most forms of denialism, first suggested by Mark and Chris Hoofnagle:

  1. Conspiracy theories
    When the overwhelming body of scientific opinion believes something is true, the denialist won't admit scientists have independently studied the evidence to reach the same conclusion. Instead, they claim scientists are engaged in a complex and secretive conspiracy. The South African government of Thabo Mbeki was heavily influenced by conspiracy theorists claiming that HIV was not the cause of AIDS. When such fringe groups gain the ear of policy makers who cease to base their decisions on science-based evidence, the human impact can be disastrous.
  2. Fake experts
    These are individuals purporting to be experts but whose views are inconsistent with established knowledge. Fake experts have been used extensively by the tobacco industry who developed a strategy to recruit scientists who would counteract the growing evidence on the harmful effects of second-hand smoke. This tactic is often complemented by denigration of established experts, seeking to discredit their work. Tobacco denialists have frequently attacked Stanton Glantz, professor of medicine at the University of California, for his exposure of tobacco industry tactics, labelling his research 'junk science'.
  3. Cherry picking
    This involves selectively drawing on isolated papers that challenge the consensus to the neglect of the broader body of research. An example is a paper describing intestinal abnormalities in 12 children with autism, which suggested a possible link with immunization. This has been used extensively by campaigners against immunization, even though 10 of the paper’s 13 authors subsequently retracted the suggestion of an association.
  4. Impossible expectations of what research can deliver
    The tobacco company Philip Morris tried to promote a new standard for the conduct of epidemiological studies. These stricter guidelines would have invalidated in one sweep a large body of research on the health effects of cigarettes.
  5. Misrepresentation and logical fallacies
    Logical fallacies include the use of straw men, where the opposing argument is misrepresented, making it easier to refute. For example, the US Environmental Protection Agency (EPA) determined in 1992 that environmental tobacco smoke was carcinogenic. This was attacked as nothing less than a 'threat to the very core of democratic values and democratic public policy'.

Why is it important to define the tactics of denialism? Good faith discussion requires consideration of the full body of scientific evidence. This is difficult when confronted with rhetorical techniques which are designed to distort and distract. Identifying and publicly exposing these tactics are the first step in redirecting discussion back to a focus on the science.

This is not to say all global warming skeptic arguments employ denialist tactics. And it's certainly not advocating attacking peoples' motives. On the contrary, in most cases, focus on motives rather than methods is counterproductive. Here are some of the methods using denialist tactics in the climate debate: 

  1. Conspiracy theories
    Conspiracy theories have been growing in strength in recent months as personal attacks on climate scientists have intensified. In particular, there has been accusations of manipulation of temperature data with the result that "the surface temperature record is unreliable" has been the most popular argument over the last month. This is distracting people from the physical realities of global warming manifesting themselves all over the world. Arctic sea-ice loss is accelerating. Antarctic and Greenland ice sheets are losing ice mass at an accelerating rate. Spring is coming earlier each year. Animal breeding and migration are changing in response. Distribution of plants are shifting to higher elevations. Global sea level is rising. When one steps back to take in the full body of evidence, it overwhelmingly points to global warming.
  2. Fake experts
    A number of surveys and petitions have been published online, presenting lengthy numbers of scientists who reject man-made global warming. Close inspection of these lists show very few qualifications in climate science. On the contrary, a survey of climate scientists who actively publish climate research found that over 97% agree that human activity is significantly changing global temperature.
  3. Cherry picking
    This usually involves a focus on a single paper to the neglect of the rest of peer-review research. A recent example is the Lindzen-Choi paper that finds low climate sensitivity (around 0.5°C for doubled CO2). This neglects all the research using independent techniques studying different time periods that find our climate has high sensitivity (around 3°C for doubled CO2). This includes research using a similar approach to Lindzen-Choi but with more global coverage.
  4. Impossible expectations
    The uncertainties of climate models are often used as an excuse to reject any understanding that can come from climate models. Or worse, the uncertainty of climate models are used to reject all evidence of man-made global warming. This neglects the fact that there are multiple lines of empirical evidence that humans are causing global warming .
  5. Logical fallacies
    Strawmen arguments abound in the climate debate. Often have I heard skeptics argue "CO2 is not the only driver of climate" which every climate scientist in the world would wholeheartedly agree with. A consideration of all the evidence tells us there are a number of factors that drive climate but currently, CO2 is the dominant forcing and also the fastest rising. Logical fallacies such as "climate has changed before therefore current climate change must be natural" are the equivalent of arguing that lightning has started bushfires in the past, therefore no modern bushfire is ever started by arsonists.

Update 16 April 2012: Many thanks to Mark Hoofnagle for pointing out that the 5 characteristics of science denial didn't originate in Diethelm and McKee's paper but in an article written by Mark and Chris Hoofnagle. This is an article very worth reading for anyone interested in climate change and public discourse about science. Credit has been updated accordingly.

0 0

Bookmark and Share Printable Version  |  Link to this page

Comments

Prev  1  2  3  4  Next

Comments 101 to 150 out of 158:

  1. tobyjoyce (#99), your dismissal of Geo Guy based on the provably false testimony of Tamino shows that you have not looked at the evidence.

    D'Aleo and Watts make 15 allegations concerning temperature records based on surface stations. Like you I was sceptical about their claims so I decided to check them myself, starting from raw data. As I don't have the time or skills to check all 15 claims I went for D'A&W claim #4:

    "....more than three-quarters of the 6,000 stations that once existed are no longer reporting."

    I downloaded the NOAA/GHCN v.2 raw and adjusted data sets and counted the number of stations versus reporting year. There were over 6,000 in 1972, falling to under 1,000 in 2009. You can easily do the analysis yourself by downloading from the NOAA web site:
    ftp://ftp.ncdc.noaa.gov/pub/data/ghcn/v2/

    When I point out Tamino's economy with truth on this blog, my comments usually get deleted; he seems to be some kind of minor deity around here. This time I have been very careful with my "rhetoric" so maybe it will get through.

    If you don't have time to do the analysis yourself take a look at:
    http://diggingintheclay.blogspot.com/2010/01/station-drop-out-problem.html
    0 0
  2. #100 Ned at 09:33 AM on 19 March, 2010
    "Can you explain what you think that sentence means? Because there are certainly lots of stations north of latitude 65 in the GHCN dataset used by NOAA for their surface temperature record"

    I think I know what he has meant. He was referring to recent Canadian data. In that country only one station north of 65 has data in GHCN for 2009-2010.

    40371917000 EUREKA,N.W.T. 79.98 -85.93

    And in fact only three more north of 60

    40371964000 WITHEHORSE, Y 60.72 -135.07 (1942-2010)
    40371966000 DAWSON,Y.T. 64.05 -139.13 (1897-2010)
    40371915000 CORAL HARBOUR 64.20 -83.37 (1933-2010)

    It's weird, since otherwise GHCN has 119 stations in Canada north of 65, 118 of them are discontinued. And 298 stations north of 60.

    Between 1989-1991 the number of active GHCN stations in Canada dropped from 496 to 44.
    0 0
  3. 101/ Huh?
    Tamino states "Two of the most prominent claims of global warming denialists have proven to be utterly false." No one is denying that fewer stations are reporting. Tamino's analysis is refuting the false claim that the cutoff introduces a warming trend. Doing this analysis, he compares subset before and after the cutoff. Where is the "economy with truth" here? Where is the "provably false testimony".
    0 0
  4. (1) There are more than enough existing stations to compile the monthly global mean temperature data, so the decline in the number of stations is irrelevant.

    (2) The stations that were dropped have no different trends than those that kept reporting. Furthermore, selectively deleting high latitude, high altitude, or rural stations has no impact on the trend. This has been shown repeatedly in separate analyses by Tamino, Ron Broberg, and Zeke Hausfather. Neither Watts, D'Aleo, nor EM Smith bothered to actually test their claims statistically before loudly claiming that the dropped stations affected the trend. They owe NOAA and NASA an apology for their false accusations of fraud (and they owe people like gallopingcamel and Geo Guy an apology for misleading them).

    (3) Even without the statistical analyses, everyone who thought about this knew that Watts, D'Aleo, and Smith were wrong. The satellites show warming, and there's no UHI in space. The oceans show warming, and there's no UHI in the middle of the ocean. In fact, we know from longstanding physical climatology that the warming trend over land is going to be larger than that over the ocean.

    In conclusion, all of the complaints about station dropouts are irrelevant. The observed warming is real and not an artifact of composition of the station lists.
    0 0
  5. scaddenp (#103), thanks for accepting that the station drop off is real. Do you have any explanation for the drop off? If scientists discard ~80% of the data before starting their analysis some kind of explanation should be given.

    In the case of NASA/GISS we are still waiting for Tom Peterson, Gavin Schmidt or James Hansen to explain what is going on. The HADCRUT3 situation and the associated IEA (Russian) station drop off is still under investigation. I am not sure who should be speaking up for NOAA/GHCN.

    The "before and after" Tamino statement you mention is meaningless. What is needed is a comparison of the full data sets with the truncated data sets. Wild claims by Tamino (or anyone else at this moment) are paper tigers unless they have the missing information.

    You can prove or disprove anything using statistics. For a pungent explanation on this point, there is a well known saying that Mark Twain attributed to Benjamin Disraeli that I am not allowed to quote on this blog.

    Several of the D'Aleo & Watts allegations have to do with station quality control which NASA and NOAA have already admitted is poor. Are you defending a position that has already been abandoned?

    Like you, I don't trust D'Aleo & Watts but I am still keeping an open mind while weighing all 15 of their allegations.
    0 0
  6. Ned, (#104),
    Your claim (1)
    Take a look at post (#102). Canada is the largest country in the world. Can you justify eliminating all but 44 of that country's weather stations?

    Your claim (2)
    You can put all the statisticians in the world end to end but they will still not reach agreement. Your claim may be correct but the only way to prove it would require comparing the full data sets with the truncated sets.

    Your claim (3)
    The satellites do not show warming at least in the last 12 years, as noted by D'Aleo & Watts. The ground station data is diverging from the satellite data. Check John Cristy at UAH.
    Sea Surface Temperatures (SSTs) are very high but are currently falling.
    http://vortex.nsstc.uah.edu/data/msu/t2lt/uahncdc.lt
    0 0
  7. gallopingcamel (#105),

    You obviously have not looked into this at all. Older station data in the record comes from multiple historical sources.

    1. It has been gathered and digitized over the years from something like 30+ sources. Most of these sources no longer produce temperature data and the current record is updated 'real time' from just three sources - the most important being the World Meteorological Organization.

    2. There is no secret about this - it is documented in a paper by Petersen (1997 I think).

    3. Hansen also discusses this in a paper (2001 I think) about how the GISS record is constructed, so no, he does not 'need to explain' because if you cared to look you would find it has already been explained - long ago.

    Could we please leave all this nonsense over on Watt's site?
    0 0
  8. I'm a little confused here - on the map which Ned links to in #100 there are many stations in northern Canada, but in #102 BerenyiPeter claims that there is only one. Is the map maybe not up to date?

    And just for the record, gallopingcamel, Russia is the largest country in the world (you knew that, didn't you?)
    0 0
  9. Watts site has a fine example of denialism in action at the moment. It seems that they have dredged up a copy of National Geographic from 1976 which shows a chart of NH temperature from 1880 to 1976 showing a marked drop from 1945. Of course it differs from the CRU record which gets the author quite excited about "hiding the decline". As far as I can tell the National Geographic piece is not peer reviewed, or based on something that is peer reviewed. There is no mention of methodology or data sources. It seems to have very little authority whatsoever.

    Nevertheless it is totted out to sow "doubt and confusion". Desperate times indeed for Watts and fellow travelers.
    0 0
  10. Marcel (#108), thanks for setting me straight. Canada is only #2. It appears that Russia still accounts for 12.5% of land area so the UEA/CRU needs to explain why they discarded the majority of the data they got from the IEA.

    quokka (#107), I think you are referring to Peterson & Vose (1997). That paper does record the station drop off that was occurring at that time. We should thank D'Aleo & watts for bringing us up to date on the continuing drop off saga.

    While you are right to say that the station drop off is no secret we are still waiting for a valid explanation. Do you know of one?
    0 0
  11. gallopingcamel writes:

    Take a look at post (#102). Canada is the largest country in the world. Can you justify eliminating all but 44 of that country's weather stations?

    Canada is not the largest country in the world. Temperature anomalies are spatially correlated over large areas, and 44 stations are enough to cover the country. And neither NASA nor NOAA are "eliminating" weather stations -- they use the data that are provided to GHCN by participating national meteorological programs in other countries, and in some cases those stations are dropped by their home countries or there are delays in reporting. For example, with Canada, there are many more stations with data currently through 2008 which presumably will be providing updated data at some point. If you have a problem with this, complain to Canada, not to NASA or NOAA.

    Continuing, gallopingcamel writes: You can put all the statisticians in the world end to end but they will still not reach agreement. Your claim may be correct but the only way to prove it would require comparing the full data sets with the truncated sets.

    This is exactly what Tamino and others have done.

    More to the point, there are good a priori reasons to believe that dropouts of high-latitude, rural, and high-altitude stations would if anything artificially decrease the warming trend, not increase it. Since these areas are warming faster than the globe as a whole, when they drop out it would tend to decrease the overall trend (if the gridding methods weren't sufficiently robust to handle the dropouts).

    So, it was highly irresponsible of Watts, D'Aleo, etc. to accuse NOAA and NASA of "dropping stations" to fraudulently increase the warming trend. They made those accusations without ever testing them. Now their claims have been tested by different people using different methods, and have been found to be false.

    At this point, if you want to claim that there IS an artificial warming trend introduced by dropping the stations that are warming fastest, you need to prove that. All the evidence is against you.

    Moving on, to the next point: The satellites do not show warming at least in the last 12 years, as noted by D'Aleo & Watts.

    That's a very interesting claim. Let's look at all 10-year, 11-year, 12-year, etc. trends in both the UAH and RSS temperature series.

    In the UAH record, trends running through last month and starting in March of the following years are positive:

    2000, 1999, 1998, 1997, 1996, 1995, 1994, 1993, 1992, 1991, 1990, 1989, 1988, 1987, 1986, 1985, 1984, 1983, 1982, 1981, 1980, 1979

    UAH trends starting in the following years are negative:

    (none)

    Well, let's look at the RSS record. Again, trends starting in March of the following years are positive:

    2000, 1999, 1997, 1996, 1995, 1994, 1993, 1992, 1991, 1990, 1989, 1988, 1987, 1986, 1985, 1984, 1983, 1982, 1981, 1980, 1979

    And RSS trends starting in the following years are negative:

    1998

    It's interesting that your choice of "twelve years" just happens to be the only period where either one of the satellite records shows a negative trend, that it just happens to start with the largest El Nino on record, and that even so only one of the two records shows a negative trend from that year.

    Continuing: The ground station data is diverging from the satellite data. Check John Cristy at UAH.

    Actually there's very little divergence between the two, and in the past disagreements between the satellite and surface records have turned out to be due to errors in the methods used to process the satellite data. Thus, over time, the satellite records have progressively become closer to the surface record.

    Still continuing: Sea Surface Temperatures (SSTs) are very high but are currently falling.
    http://vortex.nsstc.uah.edu/data/msu/t2lt/uahncdc.lt


    Those data aren't SSTs! They're the same MSU lower-troposphere data, just disaggregated into different spatial regions (northern vs southern hemisphere, land vs ocean, tropics vs extratropics vs poles, etc.)

    Actual SST data are available here, or see Kelly O'Day's excellent graphs and scripts here.

    And there's no very meaningful sense in which "sea surface temperatures are falling" is true. You can only get a negative trend if you cherry-pick a length of 5-9 years. Trends of 1, 2, 3, or 4 years, or of any number of years greater than 9, are all positive.

    ==================

    So ... this all leads to the question of why people employ such weak, wrong, or misleading arguments. If there were really good, convincing arguments to support the "skeptic" position, presumably there wouldn't be a need for the kinds of cherry-picking and factually questionable claims we see in this thread, nor for papers like McLean 2009, Lindzen & Choi, or that embarrassingly wrong Chylek one (link, link). Or E.G. Beck's physically impossible claims about chemical CO2 measurements being representative of actual global CO2 concentrations.

    Unfortunately, it takes a lot more time and effort to track down and refute these kinds of incorrect claims than it takes to make them in the first place.
    0 0
  12. gallopingcamel writes: UEA/CRU needs to explain why they discarded the majority of the data they got from the IEA

    "IEA" is an obscure right-wing Russian economics "think tank" associated with the US Cato Institute. They don't operate met stations and they don't provide data to GHCN.

    Can you please stop churning out one incorrect claim after another?
    0 0
  13. gallopingcamel, you stated that "The satellites do not show warming at least in the last 12 years, as noted by D'Aleo & Watts."

    Well, John Christy reckons there has been warming since 1978 :

    'Global climate trend since Nov. 16, 1978: +0.13 C per decade' -
    http://www.atmos.uah.edu/essl/news.html


    Figures here :
    http://www.atmos.uah.edu/essl/news/UAHDataset5.3.pdf

    Who shall we believe ?



    You also stated : "The ground station data is diverging from the satellite data. Check John Cristy at UAH."


    RSS trend is +0.156 C per decade :
    http://www.ssmi.com/msu/msu_data_description.html#msu_decadal_trends

    HADCRUT3 is 0.15 °C per decade
    http://www.metoffice.gov.uk/climatechange/science/monitoring/

    GISS is 0.2°C per decade
    http://www.giss.nasa.gov/research/news/20100121/

    Can't be bothered to look up GHCN, NCDC or JMA, but I'm sure they're all similar.
    So, what exactly does John Christy say ?

    In fact, have a look at this graph to see how close the trends are in all the ones I've given :
    http://en.wikipedia.org/wiki/File:Satellite_Temperatures.png


    You finally stated : "Sea Surface Temperatures (SSTs) are very high but are currently falling."
    http://vortex.nsstc.uah.edu/data/msu/t2lt/uahncdc.lt


    As far as I can see from your link, the global sea temperature anomaly is currently the highest since 1998. What do you mean by 'falling' ? Am I reading the table wrong ?
    0 0
  14. As for gallopingcamel's concern about station drop offs, why not look at what the people actually involved have to say, i.e. like this :

    The reasons why the number of stations in GHCN drop off in recent years are because some of GHCN’s source datasets are retroactive data compilations (e.g., World Weather Records) and other data sources were created or exchanged years ago.
    An Overview of the Global Historical Climatology Network Temperature Database
    0 0
  15. As far as I can see from your link, the global sea temperature anomaly is currently the highest since 1998. What do you mean by 'falling' ? Am I reading the table wrong ?

    You're reading the table right. gallopingcamel is wrong when he says the numbers are "falling", and he's wrong when he says the table is "sea surface temperatures". (See my too-lengthy comment above).
    0 0
  16. Ned, your response is a great illustration of how troublesome it is to counter the usual arguments from denial : gallopingcamel types out several assertions (some with poorly referenced links), and you have to post lots of detailed information (with proper links) in response. However, unbiased readers should be able to see clearly where the facts are - in your posts.

    Thanks, also, for that Kelly O'Day site - very informative indeed.
    0 0
  17. Thanks, also, for that Kelly O'Day site - very informative indeed.

    Yes, it is. The recent posts are mostly R scripts for downloading and processing climate data, but if you look back there are some other interesting things as well.

    If anyone either uses R or is interested in learning R, particularly for working with climate data, I highly recommend Kelly O'Day's site.
    0 0
  18. #108 Marcel Bökstedt at 20:31 PM on 19 March, 2010
    "I'm a little confused here - on the map which Ned links to in #100 there are many stations in northern Canada, but in #102 BerenyiPeter claims that there is only one. Is the map maybe not up to date?"

    The map is not outdated, just has a dot for each station ever used in GHCN.



    Since 1991, station number in Canada dropped to less than one tenth of its previous value. Dropout in the North is even more serious.

    To check this claim you can download the GHCN dataset whenever you want.
    0 0
  19. Nick Stokes (the moyhu site's post from 2010/2/16) has illustrated the irregularity with which temperature station data arrive in the GHCN database, by compiling a list of updates across two example months. No conspiracy is evident, just the messiness of gathering real data in the real world.

    (Sorry if somebody already pointed to this; I thought so but I can't find a previous pointer.)
    0 0
  20. John, I think Diethelm & McKee should consider adding a sixth characteristic to their list: Sheer unyielding repetition. This thread is a case in point:

    (1) Geo Guy introduced the topic of "station dropout" in this comment.

    (2) Two comments later, tobyjoyce pointed out that station dropout had no effect on the results as long as a gridded method was used, citing and linking to Tamino's analysis of GHCN data.

    (3) Geo Guy then replied with a longer comment, repeating his complaint that the number of stations has decreased but not addressing the question of whether this has any measurable impact on the results.

    (4) tobyjoyce reiterated that at least one examination of station dropout shows no effect.

    (5) gallopingcamel replied to tobyjoyce by ... pointing out that a bunch of stations had dropped out. Still no response about whether this has any effect on the trend.

    (6) Berényi Péter then chimed in to agree that yes, lots of stations have dropped out.

    (7) scaddenp replies and points out that gallopingcamel has not actually addressed whether station dropout affects the trend.

    (8) Ned adds some detail, noting that it's not just Tamino's analysis -- similar studies by others have shown the same results. The decrease in the number of stations reporting climate data hasn't significantly affected the trend.

    (9) gallopingcamel replies, again saying that stations have dropped out. His only remark about whether or not this actually matters is to say that we need a comparison of the full and truncated data sets, which happens to be exactly what Tamino and the others have provided.

    (10) quokka points out that the decline in station numbers isn't exactly news -- NOAA was discussing it more than a decade ago.

    (11) gallopingcamel admits that it isn't news, but still wants and explanation (and doesn't address the question of whether it impacts the temperature trend).

    (12) Ned writes an excessively long comment again pointing out that there are good reasons to expect that past station dropouts should not affect the trend, or if they did it should lead to an artificial cooling rather than warming trend. He further notes that multiple people have now confirmed this empirically, using different methods, and that neither D'Aleo, Watts, nor anyone else bothered to check this before making wild accusations.

    (13) JMurphy provides a link to the original Peterson & Vose paper explaining the decline in station numbers over time.

    (14) Berényi Péter says that a lot of stations have dropped out since 1991. Still no comment on all the studies showing this has no effect on the trend.

    And around we go, again and again. Since I expect that any moment another "skeptic" commenter will be popping in to let us all know that some stations have dropped out since 1991, I'd like to pre-emptively provide the following links:

    Here is Tamino's analysis of trends associated with pre- and post-dropout GHCN data: link. It shows no effect.

    Here is Zeke Hausfather's analysis of trends associated with pre- and post-dropout GHCN data: link. It also shows no effect.

    Here is Ron Broberg's analysis of trends associated with pre- and post-dropout GHCN data: link. It also shows no effect.
    0 0
  21. John,
    One aspect of denial is commonly omitted: the preventative projection of one's own behaviour patterns and character traits into the opponent (to distract from oneself and to portray the opponent as a copycat). The German language offers a good phrase for this:

    "von sich (selbst) auf andere schliessen".

    For example:

    Deniers frequently refer to AGW proponents as "believers". This is unfounded as AGW proponents/scientists base their opinion on facts and not on faith. On the other hand, skeptics, like creationists, build much of their argument on faith.

    Another example: AGW proponenents are frequently referred to as "alarmists". Inhowfar is somebody who investigates datasets or identifies scientific facts from the broad body of scientific research panicing? On the other hand, skeptics like Monckton use fear mongering (alarmism) to promote their agenda; in Monckton's case it is the fear of world communism. Fear mongering is used by all religions to keep people at bay and it is an effective tool used by the Republicans, e.g. in their warfare against health insurance. I was in Seattle last summer where some normal looking young people tried to convince me that the Obama administration was going to introduce euthanasia with their healthcare package, aiming to kill old people.
    0 0
  22. Ned,
    Hats off to both the content and style of your comments. I think that at some point, we have to assume that certains commentors are not interested in learning or hearing the truth. I am concerned that part of the repetitive questioning is to "trap" someone into an inaccurate response or goad one into an unprofessional retort. I think that this will continue to be a long battle, so stay rested and avoid carpal tunnel. Thx for your efforts
    0 0
  23. Ned, lovely comment here. Repetition is indeed one characteristic of a poor argument.

    This is also a frequent tactic in psychology - repeating a question over and over, forcing the patient to explain in greater and greater detail, until they have eventually said much more than they intended. In the case of bad arguments, it seems to be used to draw out some overstatements, rather than actually trying to discuss what has been said.
    0 0
  24. Ned (#112), you denigrate the IEA and anyone else with evidence that conflicts with your unshakable faith. If the IEA is wrong, their allegations will evaporate as soon as the UEA/CRU demonstrates they used and published the full data sets from Russia. Until the CRU does that you are "Whistling Dixie".

    Enjoy "Russia Today"
    http://www.youtube.com/watch?v=ElPunkm1zYQ

    In similar fashion you have completely missed the point about Tamino. If he has the full and truncated surface station data sets from NOAA, NASA or UEA he needs to prove it. Only folks who have not heard of Climategate accept the word of climate scientists or their statisticians "Ex Cathedra".

    I collect energetic photons for a living and the idea of discarding 80% of them before starting my analysis is beyond absurd. When "Climate Scientists" at NASA, NOAA, the UEA or the denizens of this blog fail to understand such a simple point, it is hard for me to take them seriously.

    You seem to believe that dropping stations makes no difference to the results of the analysis. You may be right but how you can prove it without comparing the full data to the truncated data?

    OK, so I am repeating myself but you keep refuting allegations I am not making.

    If it will help this discussion get out of a rut, I don't know whether the station drop out causes a warming bias or a cooling bias.

    There is some work (not peer reviewed) suggesting that the records have been biased in both directions:
    http://diggingintheclay.blogspot.com/2009/12/physically-unjustifiable-noaa-ghcn.html

    What I like about the above link is that it is specific enough to allow fact checking. Rather like post (#102) in this thread.
    0 0
  25. JMurphy (#113), this may be a "Failure to Communicate" (cf. Cool Hand Luke). I believe that the world has been warming for the last ~10,000 years and that there has been significant warming since 1850, probably in the range 0.7 and 0.9 degrees Celsius.

    Satellite data has only been available for ~37 years but over that time scale there has been a warming trend. In short, I agree that warming has occurred . My point is that there has been a cooling trend since 1998, contrary to what Hansen et al. claim, based on surface station records.

    Here is the UAH data converted into a graph:
    http://wattsupwiththat.com/2009/07/03/uah-global-temperature-anomaly-for-june-09-zero/

    Here is NASA's view based on surface station records:
    http://data.giss.nasa.gov/gistemp/

    These graphs appear to be diverging over the last 12 years. Am I wrong?
    0 0
  26. Ned (#115), the SSTs were very high into January 2010. I found a link at UAH that showed the February 2010 SST sharply lower but it has disappeared. Can anybody help me out?
    0 0
  27. gallopingcamel> Indeed, there does seem to be some kind of serious break down of communications here. I don't know much about the actual facts behind this, I'm just trying to figure out the various positions on the question. There seem to be several points at stake, but discussing them all at once will probably just increase the confusion, so lets concentrate on the meaning and importance of the missing (Canadian and other) stations.

    There has been a very emotional debate about this on other sites,
    so we have to be very very careful here and try to behave like
    detached scientists. Or if thats not possible, at least try to behave.

    No one seems to deny that there are many stations that reported earlier, which are not reporting today. It's not entirely clear to me if the data from them is has been recorded, but not in digital form, or if the data was never collected. In any case, for our purpose, it does not exist. It's always better to have more data, so from that point of view it is sad that the stations are not there, but we can't do much about that. To make this worse, it seems possible that there is a bias in the sample, that stations in high and cold places dropped out first. On the other hand, our record will never be totally complete, because there will always be a point on Earth where there is no thermometer.

    So how do we handle this. We can't recreate the missing data. We can still compute a global temperature from the data we have, but we don't know if it is exactly the same as what we would have obtained from a more complete record. This is actually a very common situation in science: How do we use the data we have as well as possible, and how sure can we be
    that our incomplete data set (data sets are ALWAYS incomplete) tells us the true story?

    I think that what I have said so far is more or less acceptable to all involved, yes? But
    we are getting into deeper water. The point where it seems to be a total break down of communications is about Tamino's comments.

    The way I understand this (correct me if I'm wrong), he analyses data which I understand to be publically available? He divides the available data into two piles: A is the pile of data from stations that have been available for all time, B is the pile of data from the other stations, those that we lost on the way.

    We would all be happy if we could compute the global average temperature at all times from A+B. We can't do that, since B is only partially available. However, we can compute the average over A at all times, and compare with the average of A+B (as available).Tamino claims that they essentially agree, but that A+B(as available) shows more warming than A. The deatails of this would depend on exactly how we conpute the average etc... I haven't checked on it.

    If he is correct, it strongly suggests that the hypothetically computed average of B would
    show stronger warming than the computed average of A, and thus
    the hypothetically computed average of A+B(complete) would show stronger warming than the computed average of A+B(as available). So the missing stations would underestimate the warming.

    I think that the math is clear - if we believe Taminos result, A+B(as available) is underestimating global warming. This implication should be uncontroversial, yes?

    The controversial point is whether or not we believe Tamino's claim, TC. Since communication is difficult around that nexus, I expect that I have misunderstood,
    so please bear over with me and correct me if I'm using strawman tactics here. I don't intend to.

    I think that gallopingcamel says :

    (1) I only belive TC if you can produce the total data set B
    (2) If Tamino has FULL access to the total data set B, he has to prove it.
    (3) We are discarding 80% of the data BEFORE the analysis.

    My answers would be
    (1) is clearly an impossible condition to fulfil, and gallopingcamel is aware of it.
    (2) Tamino does not claim to have access to B, only to publically available data.
    (3) It seems to me that we are not discarding any data - some data is simply not available to us.

    But this only stands for me, and it's probably not concensus. Maybe gallopingcamel could comment on it anyhow - whether I have misrepresented him, what he really wants to say, and if and why he does not agree with my answers?

    Still, one can have doubts about TC. If it is wrong, we are troubles, because than we know less than we thought about what is coming our way. One could think of weak points in TC.

    For instance, it is not peer reviewed. Incidently, neither are the claims to the contrary.

    It is also not entirely clear to me that the method is valid - what the statistical analysis can show is that the difference between A and B is small in the periods when the data from B is available. That suggests, but does not completely prove that they don't differ after that time. I feel that there is a need for a method discussion here. But we are getting into difficult territory.

    The paper of Peterson and Vose does not seem to approach this particular problem.

    There are others who have been doing the analysis (Ned links to them), and they seem to
    get results similar to Tamino.

    For what it is worth, there is also supporting evidence that the Arctic is warming faster than the rest of the globe, which would suggest that if we are missing data points in the North, we would underestimate global warming, not overestimate it.
    0 0
  28. gallopingcamel writes: Ned (#112), you denigrate the IEA and anyone else with evidence that conflicts with your unshakable faith.

    I didn't "denigrate" the IEA, I simply stated the facts. You said that CRU "gets data from" the IEA. That's wrong. IEA is not a meteorological organization and they don't have any weather stations. They're a right-wing economic "think tank" that issues press releases and position papers.

    In similar fashion you have completely missed the point about Tamino. If he has the full and truncated surface station data sets from NOAA, NASA or UEA he needs to prove it. [...] You seem to believe that dropping stations makes no difference to the results of the analysis. You may be right but how you can prove it without comparing the full data to the truncated data?

    I provided links to not just one but three different people who each did these comparisons. Since you seem to have missed it, I'll provide the links again:

    Here is Tamino's analysis of trends associated with pre- and post-dropout GHCN data: link. It shows no effect.

    Here is Zeke Hausfather's analysis of trends associated with pre- and post-dropout GHCN data: link. It also shows no effect.

    Here is Ron Broberg's analysis of trends associated with pre- and post-dropout GHCN data: link. It also shows no effect.


    Each of those links has a graph showing a comparison of stations that dropped out before some particular date with those that did not drop out.

    I don't know why you say Tamino needs to "prove" anything. All the data he used are publicly available, and he described his methods in more than sufficient detail for them to be replicated. And thus at least two other people have now managed to show the same thing using their own analyses.

    Again, however, when Watts and D'Aleo publish a report claiming that NOAA and NASA have been conspiring to inflate the temperature trend by selectively dropping cooler stations, it's not up to Tamino or someone else to prove them wrong. It's up to them to provide evidence to support their claims. And since those are very serious allegations, they should have had ironclad evidence in hand before publishing. Unfortunately, they have no evidence at all of any actual effect on the trend. They didn't have that evidence before they published, and they still haven't been able to provide any evidence or to refute any of the above-linked studies showing that they're wrong.

    The behavior of Watts and D'Aleo in this instance was the height of irresponsibility. If they had any sense of shame they'd apologize to the people at NOAA and NASA, and to the readers whom they misled.

    I collect energetic photons for a living and the idea of discarding 80% of them before starting my analysis is beyond absurd.

    Did you miss the explanation that NASA and NOAA do not delete stations? As I said in my earlier comment: "And neither NASA nor NOAA are 'eliminating' weather stations -- they use the data that are provided to GHCN by participating national meteorological programs in other countries, and in some cases those stations are dropped by their home countries or there are delays in reporting. For example, with Canada, there are many more stations with data currently through 2008 which presumably will be providing updated data at some point. If you have a problem with this, complain to Canada, not to NASA or NOAA."

    The only thing that's "beyond absurd" here is that you persist in attacking the honest, hardworking scientists of NOAA and NASA for a situation (declining station numbers) that is (a) not their fault, and (b) does not actually bias the temperature trend.

    From my perspective, this whole "station drop-out" argument perfectly illustrates the five characteristics of denialism that John refers to in this post:

    1. Conspiracy Theories: The report by D'Aleo and Watts is actually titled "Policy Driven Deception" and includes allegations like "NOAA and NASA conspired in the manipulation of global temperature records" and "there has been a coordinated effort to manipulate instrumental data. This manipulation has produced an exaggerated warming that is blamed on man’s influence."

    2. Fake Experts: You cited Watts, D'Aleo, and the Moscow-based economic think-tank IEA. Geo Guy cited SPPI and EM Smith. All of these are perfect examples of "fake experts."

    3. Cherry Picking: Geo Guy originally claimed NOAA uses "just one thermometer [for measuring] everything north of latitude 65 degrees." I pointed out that there are in fact many, many stations north of latitude 65. Berényi Péter then said that, well, the claim actually just refers to stations in Canada north of 65 that have contributed data to GHCN since 2008.

    4. Impossible Expectations: You have complained repeatedly about NOAA and NASA "discarding" data when other countries have failed to keep stations running or to update their data in GHCN. This is like blaming the post office for not delivering a letter that the sender left sitting on their own desk.

    5. Misrepresentation and Logical Fallacies: This is perhaps the most important point in this thread. Watts and D'Aleo make the assumption that when high-latitude stations drop out it makes the global trend warmer. That is on its face illogical, since on average those stations have been warming faster than the globe as a whole, so dropping them should introduce a cooling bias (if the gridding process weren't able to compensate for station dropout, which it is). Several people have now empirically demonstrated that Watts and D'Aleo's illogical claims are false, by analysis of actual dropped vs non-dropped stations (and by other methods as well).

    So ... I'd say that this thread hit all the major points of the Diethelm & McKee 2009 paper.

    Back to gallopingcamel: If it will help this discussion get out of a rut, I don't know whether the station drop out causes a warming bias or a cooling bias.

    Or no statistically significant bias at all? In any case, glad to hear that you're finally willing to say this. Perhaps we can move on now, and let NASA and NOAA get back to their work without further harassment.
    0 0
  29. Marcel Bökstedt writes: One could think of weak points in TC.

    For instance, it is not peer reviewed.


    Yes, I agree. Apparently Tamino has submitted or will be submitting a paper based on his analysis for publication. I believe Zeke Hausfather has also speculated about doing that. In the past on this site I've tried to only cite peer-reviewed works, but on this topic we have some rather nasty allegations that people are "conspiring" to increase the temperature trend, and there aren't yet any peer reviewed studies.

    It is also not entirely clear to me that the method is valid - what the statistical analysis can show is that the difference between A and B is small in the periods when the data from B is available. That suggests, but does not completely prove that they don't differ after that time. I feel that there is a need for a method discussion here.

    Well, people have tried different ways of looking at this. For example, Zeke Hausfather and Ron Broberg both looked at different kinds of comparisons (e.g., high latitude vs low latitude stations, high altitude vs low altitude, rural vs non-rural, etc.)

    I think those comparisons do a good job of addressing the argument that the dropout stations and non-dropout stations were drawn from different populations with different trends. There does not in fact seem to be a significant difference in the trend among any of those populations (rural, high lat, high alt, etc.)

    Anyway, Marcel, thanks for your long and thoughtful comments, which I generally agree with.
    0 0
  30. One final point (I've already contributed way too many words to this topic!)

    Watts has a rather problematic history with this kind of thing. He got his start with the "Surface Stations" project, in which he and others made a lot of strong accusations that various problems with the fine-scale siting of met stations were artificially inflating the temperature trend.

    He's posted lots of photographs of stations, and lots of blog posts suggesting that these have "contaminated" the temperature record. But he still hasn't done a single quantitative analysis of the actual impact on the trend of these siting issues.

    There have been at least one early non-peer-reviewed analysis of Watts's data (by John V. with OpenTemp) and one new peer-reviewed paper (Menne 2010) on this topic. As discussed in John's post here (On the reliability of the U.S. Surface Temperature Record), Menne 2010 seems to show that there is no significant impact on the temperature trend from the kinds of site classification issues that Watts had raised.

    Now, Watts has criticized JohnV and Menne et al. for publishing their analyses using incomplete data (not all the Surface Station site classifications were completed). But Watts certainly didn't wait before making the claims that the record was "contaminated"! Ethically speaking, he should have had convincing evidence first before making these claims. Instead, he blames others for trying to clean up the mess he created by his premature assumption that the surface station trend is inflated.

    Watts responded to Menne 2010 by publishing the "Policy Driven Deception" paper about station dropouts. Once again, he was making very strong and potentially libelous accusations against people based on a leap of reasoning (fewer cold stations must mean an artificial warming trend). Once again, he didn't do the quantitative analysis before making the accusations. And once again, when people started looking into this, they found that the evidence seems to contradict Watts's claims.

    So this isn't just a one-time thing. Twice now, Watts has unleashed serious allegations against NOAA and NASA without having the evidence to back up his claims. Both times, it's fallen to other people to do that analysis and prove Watts wrong.
    0 0
  31. Marcel (#127),
    Your remarks seem entirely reasonable.

    In my opinion it is not likely that much of the raw station data is irretrievably lost. In most cases there will be local files that could be used to rebuild the files at NASA, NOAA & CRU. For example you can be sure that Canada's Weather Office keeps copies of all the data it sends to NOAA.

    Rebuilding the raw data files is likely to require a large effort so would it be worthwhile?

    Most of the people on this blog seem to be arguing that the outcome would be the same with 6,000 stations as it is with 950. However, that assumes "facts not in evidence".

    To get back to the main subject of this thread, this discussion has clearly demonstrated all five of the "characteristics of scientific denialism". Unfortunately, the vast majority of the posts demonstrating these traits are from AGW Alarmists.

    The Alarmists display additional failings owing to their unwarranted certainty that has no place in science. Until Climategate they were often inclined to go as far as to claim "the science is settled" while personally attacking anyone who disagreed with them.

    Real scientists disagree over scientific hypotheses and then start looking for ways to test those hypotheses.
    0 0
  32. Perhaps some people are confused by the term "dropped," as in "dropped stations." That term does not always mean a person took the action of dropping a station's data, as in "those stations' data were dropped by Goldmember."

    That term also is used to mean the passive new absence of data, as in "The video stream had dropouts" and "That station was in the first ten years of data but then dropped out."

    That latter sentence does not mean that ten years of that station's data initially were there but then vanished. In the context of a series of data from that station, that sentence means that the station as a data provider dropped out of the series. It does not mean that the data the station already had provided have dropped out. In other words, it means "Data were gotten from that station for every year of the ten years up through 1994, but no data were gotten from that station from 1995 on."

    Perhaps some commenters whose first language is not English are especially susceptible to misunderstanding that term's use here, since the differences are subtle. Maybe some of them could provide explanations of that term in other languages?
    0 0
  33. gallopingcamel, you are incorrect (again) in assuming the raw station data have been "lost" and so are in need of "rebuilding" the files at NASA, NOAA, and CRU.

    As many people have explained to you repeatedly, and as the scientists themselves have explained in their publications repeatedly, temperature data dribble in from weather stations around the world, over the course of years and even decades. Even "dribble in" is a misleading phrase, because it implies a single, central authority to whom all the stations must report.

    In reality, temperature data often languish in notebooks in offices or homes for years until somebody discovers and sends the notebooks to somebody in that nation's or state's meteorological organization, who in turn might have them under a pile of papers in their office for years, until they get around to typing them up in, say, a paper printed report that nobody outside that office sees, until somebody from the outside inquires for the eighth time and finally gets ahold of the printed report, then months later finds time to transcribe the numbers from the report into a database.

    Consequently, the big databases get updated every month not just with new data that are the newest to be gathered at the stations, but also with data that were gathered long ago but are "new" in this database. Those data never were "lost" from the big database. They never were "lost" from anywhere. They simply took a long time to make their way into the database.
    0 0
  34. gallopingcamel writes: In most cases there will be local files that could be used to rebuild the files at NASA, NOAA & CRU. For example you can be sure that Canada's Weather Office keeps copies of all the data it sends to NOAA.

    Rebuilding the raw data files is likely to require a large effort so would it be worthwhile?


    Neither NOAA nor NASA have "lost" any data, and there are no files that need to be "rebuilt." Canada (and other countries) have been slow in updating their own data in the international GHCN database. NOAA and NASA properly use all the data in GHCN. They cannot use data that other countries have not yet provided.
    0 0
  35. A factor that feeds scientific denialism characteristic "1. Conspiracy Theories," is misunderstanding of the day to day practicalities of scientific work. Most people who are not working scientists get their impression of the scientific enterprise from movies and TV, where scientific organizations are monolithic, uniformly high-tech, and incredibly well organized. One case of this problem is the CSI effect.

    The reality is far messier and more mundane, and not just for climate science. Every morning while I'm packing my lunch, I listen to my housemate using his beat-up cell phone at the kitchen table while he is eating his oatmeal, in a conference call to plan the Mars rovers' next few days of activities. My office was a cubicle right next to the chronically malfunctioning restroom with missing ceiling tiles, in a trailer; after three years I got an office with a door, but it's still in the trailer and the restroom is unchanged, right down to the missing ceiling tiles.

    James Hansen is not the boss of everybody at NASA. Goddard runs no temperature stations. It is not true that all the world's temperature stations are obliged to report their temperatures to Goddard. Gentle cajoling and personal contacts often are needed for Goddard to acquire temperature data from people who have stations' data, and even to discover that stations' data exist. Even the national weather services that are supposed to get stations' data, often must resort to cajoling and personal contacts. Only a tiny minority of temperature stations have automatic satellite or phone line delivery of their data to anywhere, let alone Goddard.

    Raw data frequently, even usually, are not useful until they have been processed, sometimes drastically. That includes ground station temperature measurements. Nor do satellites directly measure temperature, any more than real criminologists' lab equipment spits out instant DNA matches from blood samples. Temperature must be inferred by manipulating the satellites' raw data, and also must be inferred by manipulating the ground stations' raw data.

    When the LCROSS spacecraft sent its data about the moon impact on October 9, many members of the public were disappointed because the images were not spectacular, and because the LCROSS team made only vague statements about the presence of water. Not until a month later was the announcement of strong evidence for lots of water. Conspiracy theorists could assume that the raw data indicated lack of water, or that the raw data after CSI-like processing in minutes indicated lack of water. So conspiracy theorists could take the one-month lag in the report of water's discovery, and especially the difference between the low-key, equivocal initial announcements and the subsequent elated, strong announcements, as evidence that the LCROSS scientists spent the month manipulating the original data into supporting their desired conclusion. But the reality is that spectral analysis is very tricky. There are sub-sub-specialties in spectral analysis, so there aren't many people in the world who are capable of doing the particular kind of analysis that was appropriate to LCROSS. Even those people had to do lots of different analyses, each time reflecting on the results before designing the next kind of analysis. And the spectral analyses' results had to be interpreted by specialists in other fields. A month of much longer than 40-hour work weeks by a whole bunch of smart, dedicated, experienced, underpaid people was necessary to finally produce the conclusion that in a movie would have appeared on Clint Howard's computer screen as a red flashing "Water Present!" message after a tense 15 seconds of waiting after the LCROSS impact.

    Genuine working scientists know that's how all fields of science really work. Unfortunately, people who took a few science classes, and even some people who got undergraduate degrees in science (alas, even some people who got Masters degrees) do not know that. That's an example of the Dunning-Kruger effect.
    0 0
  36. Gallopingcamel wrote : Satellite data has only been available for ~37 years but over that time scale there has been a warming trend. In short, I agree that warming has occurred . My point is that there has been a cooling trend since 1998, contrary to what Hansen et al. claim, based on surface station records.


    What 'cooling trend' ? Instead of just stating what you believe, can you link to a graph (or something) that shows this 'cooling trend' ? And you can you show how it relates to what 'Hansen et al. claim' ?



    Gallopingcamel wrote : Here is NASA's view based on surface station records:
    http://data.giss.nasa.gov/gistemp/
    These graphs appear to be diverging over the last 12 years. Am I wrong?



    Hard to tell when it is unclear what you are actually trying to prove from that link. The only graphs I can see there are a global temperature one, and one for Japan with no detail as to axis labels. What do you claim is diverging, and from what ?



    Gallopingcamel wrote : Ned (#115), the SSTs were very high into January 2010. I found a link at UAH that showed the February 2010 SST sharply lower but it has disappeared. Can anybody help me out?


    Forgive me for intruding but this shows that there was no UAH SST reading for the 2nd to the 5th of February this year, and the 6th to the 7th reading were only in 3rd highest place since 2003. Since that date, however, and before, the readings have all been higher :

    UAH SST



    Gallopingcamel wrote : To get back to the main subject of this thread, this discussion has clearly demonstrated all five of the "characteristics of scientific denialism". Unfortunately, the vast majority of the posts demonstrating these traits are from AGW Alarmists.
    The Alarmists display additional failings owing to their unwarranted certainty that has no place in science. Until Climategate they were often inclined to go as far as to claim "the science is settled" while personally attacking anyone who disagreed with them.



    And that is just bizarre, wrong and totally divorced from reality, I'm afraid. Either you can't see that or you don't want to. Shame, either way, but anyone reading all these posts can see where the truth lies.



    Gallopingcamel wrote : Real scientists disagree over scientific hypotheses and then start looking for ways to test those hypotheses.


    Which particular hypothesis did you have in mind ?
    0 0
  37. Tom Dayton at 04:58 AM on 21 March, 2010

    What a terrific post. Thank you.
    0 0
  38. Thanks, Doug. I should have added that despite the trailer office and missing ceiling tiles, I have got my dream job that I never, ever want to leave.
    0 0
  39. J Murphy (#136)
    My apologies for sending the Japan link in my (#125) post. Here is the one I meant to send:
    http://www.giss.nasa.gov/research/news/20100121/

    The second graph (Global Land-Ocean) shows a rising trend over the last decade.

    The UAH one (Lower Troposphere) shows a falling trend over the same period.
    http://wattsupwiththat.com/2009/07/03/uah-global-temperature-anomaly-for-june-09-zero/

    Am I comparing apples and oranges?
    0 0
  40. gallopingcamel wrote : The second graph (Global Land-Ocean) shows a rising trend over the last decade.
    The UAH one (Lower Troposphere) shows a falling trend over the same period.
    http://wattsupwiththat.com/2009/07/03/uah-global-temperature-anomaly-for-june-09-zero/
    Am I comparing apples and oranges?



    You are comparing an insignificant trend to an insignificant trend, because you are only using 10 years, so it is a waste of your and my time. Try it for 30 years, which will be far more significant.

    However, if you insist that you want to use an insignificant trend over the last decade, have a look at this and see how similar those insignificant trends are :

    Insignificant 10 yeart trends
    0 0
  41. PS Both of the trends in the link above (for UAH and GISS) show a rising trend for the last decade, as seen by looking at the data here.

    And I've just noticed, gallopingcamel, that your link to Watts is for July 2009. Why don't you use something more recent ?
    0 0
  42. I thought #17 was about climate change true-believers for a moment.

    Discoverers make their claims directly to the popular media, rather than to fellow scientists.

    The claimed effect appears so weak that observers can hardly distinguish it from noise. No amount of further work increases the signal.

    Anecdotal evidence is used to back up the claim.

    But I digress.

    The "Impossible expectations of what research can deliver." was most interesting to me. Unlike hard science, climate science has no expectation that anything is reproducible. The density of water is the same here as in Russia (there is consensus on this I believe). But the earth undergoes constant change that doesn't repeat. The day is slowing down, meteor strikes, radioactive elements decaying in the crust, the moon receding every year, continents drift, etc.

    So from the get-go, climate science is never going to meet expectations. Especially compared to hard science.

    AGW depends on about a dozen assumptions, some of which are very strong (China and India are not going to play ball), to some that are very weak (warm weather is universally bad).

    That secondhand smoke example? The NCI estimates 3,000 lung deaths per year due to secondhand smoke. It might even be true. Within the limits of what can be estimated, that is as good as it gets. Over 1,000 people die in highway accidents each MONTH in the US. That is not an estimate, actual numbers. In terms of risk mitigation, highway safety probably deserves 10 times the money spent compared to secondhand smoke.

    AGW has some hard numbers, CO2 rising, sunspot activity going back hundreds of years, recent satellite data ... and estimates like tree ring proxies, surface temps from ground stations, all the way to complete WAGs a hundred years out.

    They are not the same and no one in AGW appears to be especially troubled by it.

    And no matter how solid a theory is, there is no guarantee it won't be found lacking.

    Gamma ray bursts are a prime example. Energetic, and just HAD to be within the Milky Way, otherwise basic laws of physics would be violated. The key assumption was that the gamma radiation was uniformly radiated. That assumption turned out to be false.

    Science is replete with stuff like this.

    The key assumption in AGW is that CO2 determines global temperatures. Particularly CO2 emitted by fossil fuels.

    Within limits of climate research, this is really hard to pin down. Maybe impossible. It doesn't appear that CO2 was a driver at times in the past, so folks would like to understand what the differences are today that would make it so.

    Soooooooooooooooooooooo, great article.

    Since your website got hacked, figure you are now in the big leagues.
    0 0
  43. And #2 ... when the AGW evidence gets sorted into the good, the bad, and the ugly ... AGW's credibility would be vastly improved.
    0 0
  44. Unlike hard science, climate science has no expectation that anything is reproducible.

    Actually, lots of research in climate science is reproducible. You can look at the delta O18 records from ice cores in Greenland and in Antarctica, and they match up quite nicely. The Clear Climate Code project has reproduced the results of GISTEMP using their own software operating on the same input data. Events like the PETM or the Younger Dryas show up in all kinds of different paleoclimate studies. Trends in CO2 concentration at stations ranging from the Arctic to the tropics are comparable. There are small differences among the UAH, RSS, GISTEMP, HADCRU, and NOAA global temperature data sets, but they generally match up pretty closely. And one could go on and on...

    I'm not sure where you get the idea that nothing is reproducible. Lots of stuff gets reproduced, in fact almost everything out there is being done by at least two groups.
    0 0
  45. Ned

    I said no expectation of reproducibility. Since the earth's state is constantly evolving, and nothing ever repeats through time, no climate model based on past history can predict the future. Certainly not to within a gnat's eyelash that fearmongers would like.

    This is just an extraordinarily difficult problem, even if climate scientists were completely dispassionate, with no grant riding on support of AGW, and no preconceived expectations to confound their research. And if AGW was discarded tomorrow, no one in the hard sciences would think much of the failure. Just another hard problem people took a crack at. Brownie points for effort.

    The Yellowstone supervolcano is overdue to erupt. That probability of eruption (which is unknown) is not taken into account by climate models. Why is that? The assumption is that the probability is low. But the probability is actually unknown. How about a "nuclear winter" that was fashionable years ago? Now the probability of total thermonuclear war, if one can believe the Union of Concerned Scientists, isn't low. So why is this scenario not taken into account?
    0 0
  46. oracle2world - this is an extraordinarily weird definition of AGW theory that you are working with. It is really a physical theory of climate from which you deduce that if x extra amount forcings (GHG, land change, aerosols) is of human origin, then it will result in extra warmth, and that the observed warming since 1940s is primarily of human doing. AGW does not predict future emissions at all; nor super volcanoes, nor changes in solar. Instead, it is a theory which says that if you continue to increase GHG to this level, you get this climate. It can also happily tell you what the likely effects on climate of another volcanic eruption - or a prolonged solar minima.

    If you want to send a rocket to mars you don't rely on history - you calculate based on known physics. Just so with climate. Do you think greenhouse gas effect isn't hard science?

    Model is doing pretty good so far at predicting climate if you check the model /data comparisons.
    0 0
  47. There is this concept of "hard" science as applied to physics and chemistry. The word was clearly used to separate these from other "softer" fields of study such as psychology or economics, the reason being that in these areas you can not apply the scientific method, (i.e., predicting and reproducing results in a lab).

    There seems to be a clear parallel here with Climate Science in that the Earth's climate does not fit into a lab, nor lends itself to predictable and repeatable testing. It would therefore be helful if climate scientists recognise these limitations, and simply own up to the fact that theirs is also a mix of hard and soft (since it "relies on physics and chemistry), but on the whole, it is a soft science.
    0 0
  48. RSVP wrote :

    "It would therefore be helful if climate scientists recognise these limitations, and simply own up to the fact that theirs is also a mix of hard and soft (since it "relies on physics and chemistry), but on the whole, it is a soft science."


    How is it a 'soft science' ?
    Models have been making predictions for 20 years or so, since Hansen's famous one with the 3 scenarios - have you tested that ?
    How about the prediction of rising temperatures alongside rising CO2 - have you tested that ?
    How about Tropospheric temperatures - tested by you ?
    0 0
  49. RSVP, we experimental psychologists say "There are the hard sciences, and then there are the really difficult sciences."

    Every year, thousands of students in labs for classes dutifully execute psychological lab experiments that always have yielded the same, predicted-by-theory results for the previous tens of thousands of students over the previous many decades.

    Experimental psychology has a long history. Even the brief entry in Wikipedia reveals your statements to be false.

    But your bigger error was equating the scientific method to "predicting and reproducing results in a lab." Most broadly, science is what scientists do.
    0 0
  50. Aha, yet another replication of Tamino's conclusion that the GHCN stations that dropped out before 1991 have a similar trend to those that did not drop out:

    # GHCN Processor v1.0 by Residual Analysis

    # Graph of the results

    It seems to nicely match the previous results from Tamino, Zeke Hausfather, and Ron Broberg.

    More to the point, Joseph seems to have come up with a really nifty and flexible program for doing these kinds of studies.
    0 0
    Response: Thanks for the link, I've added it to the list of links on the 'Dropped stations introduce warming bias' argument (and while I've got everyones' attention, a quick exhortation to submit any useful links you might encounter so we can build a comprehensive resource of useful global warming links).

Prev  1  2  3  4  Next

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



The Consensus Project Website

TEXTBOOK

THE ESCALATOR

(free to republish)

THE DEBUNKING HANDBOOK

BOOK NOW AVAILABLE

The Scientific Guide to
Global Warming Skepticism

Smartphone Apps

iPhone
Android
Nokia

© Copyright 2014 John Cook
Home | Links | Translations | About Us | Contact Us