The PAST is Not What it Used to Be (GW Tiger Tale)

Guest Post by Ira Glickstein.

Time machines are a staple of sci-fi. Someone travels back to the past and changes some momentous historical event, expecting his or her heroic action will improve the present and future, usually with disastrous results! Well, NASA GISS has a different type of time machine that does not actually go back to the past, but simply changes the historical temperature data to make the present Global Warming situation appear worse than it really is, and, by implication, lend credence to their CAGW (Catastrophic Anthropogenic Warming) theories.

This is the second of my Tale of the Global Warming Tiger series where I allocate the supposed 0.8ºC warming since 1880 to: (1) Data Bias, the subject of this posting, (2) Natural Cycles, and (3) AGW, which will be the subjects of subsequent postings. Click Tiger’s Tale (and Tail :^) to see my allocation and read the original story.

DATA BIAS
This posting is about how the official climate Team has (mis)adjusted past temperature data to exaggerate warming, and how the low quality of measurement stations and their encroachment by urban heat island (UHI) developments have distorted the historical record.

The above blink graphic alternates between two base charts of historical US Annual Mean Temperatures, both publicly posted by NASA GISS, the older one in 1999, and the most recent downloaded from NASA GISS this month (January 2011). The 1999 image is from a blink graphic comparing NASA GISS 1999 and 2008 data originated by a Netherlands website (zapruder.nl). I first discusssed that graphic in 2009.

Please note that both charts are to the same scale and that my annotations are fixed in place so viewers can see how the data has been changed. I have added a handy scale indicating that the large boxes on the NASA GISS charts are 0.5ºC high, along with a ladder showing 0.1ºC increments. The see-saw (with James Hansen juggling the Earth’s temperature data and our economic future :^) indicates the change between a peak in the early 1930’s and a trough in the mid 1990’s. Note how the slope changes between the 1999 version and that for 2011. In the 2011 version, the 1930’s get COOLER and the 1990’s get WARMER. If you add the changes together, you get somewhat more than the 0.3ºC I have allocated for Data Bias, so I am being quite conservative here.

I have used US data for my example because those sources are more under NASA GISS observation and control than most international data, which may be of poorer quaity. In an earlier post on WUWT I included a graphic with a copy of a NASA GISS email released pursuant to a FOIA request that indicates they felt a need to modify historical data seven times over a period of nearly a decade, until they got it right.

That means the previous six times they admit they got it wrong! Keep in mind that their mid-1990’s data has been in hand for over a decade and their mid-1930’s data is old enough to collect Social Security :^), yet they have made that old data work until they got it right, which, in this case, means more in line with their global warming models. CO2 is going up, therefore, temperatures MUST go up, OR ELSE. (Or else they will wiggle and wriggle and jiggle and juggle the data until it does what must be correct according to their theories, which, in turn, must be correct because real climate scientists thought them up and they are -or were- sincerely convinced they are -or were- saving the whole world.)

NASA GISS has been quite blatant in modifying the data even though they are aware that all the older versions exist in electronic archives. They have got away with it because no one in the major media or Congress seems interested in calling them on it. In my free online novel, set several decades in the future when virtually all data is in electronic storage, officials who control the worldwide data servers create what they call a máquina del tiempo (time machine in Inglañol, the then-prevalent version of US English peppered with Spanish words and phrases) that alters historical documents to further their plan for space travel. In the case of weather data, to cover their tracks, they would also have to alter the original hard-copy documents. This isn’t likely to happen since the NCDC keeps these paper records from COOP weather observers secure in a climate controlled vault in Asheville.

MEASUREMENT STATION QUALITY
The Surfacestations.org project has done a good job of surveying official US temperature measurement stations. I discussed some examples and showed some of their more interesting photos here.

NASA/NOAA specifies measurement sites in five classes, with the best at least 100 m (over 300 ft) from any source of artificial heating or land development and the worst located right on an occupied building (see my graphic). According to a 2009 survey, as of that year, only about 3% of official sites in the US were at Class 1. About 8% were in Class 2, at least 30 m from a source of artificial heat. About 20% were in Class 3, between 10 and 30 m. The remaining stations were closer than 10 m to an artificial heat source (58%) or right on a heat source (11%).

Thus, only about 3% + 8% = 11% were in the best two classes, reasonably distant from artificial sources of heat, while 58% + 11% = 69% were in the worst two classes, easily affected by nearby heat sources. Thus, over 2/3rds of the official reporting stations in the US were close enough to artificial heating sources to be affected. I do not know if the situation has improved much, or at all, over the past couple of years nor if the situation is better for foreign stations, but it may be even worse!

Of course, the Warmists will remind us, Global Warming has to do with changes in temperature. Thus, if a station has been at the same location for decades, any delta in reported temperature should be consistent with actual trends in that area, right?

WRONG!

Stations in urban areas, even if they have been in the exact same place, have been affected by development and lifestyle changes. This includes installation of air conditioning in buildings that had none fifty years ago, more auto and truck traffic, and construction of nearby buildings. But, many stations have been moved from time to time and thus have not been in the same place all this time, and most have been affected and encroached by civilization and changes in land use.

Why are the stations so close to artificial heat sources? Well, fifty or more years ago, all the readings were taken manually by volunteer observers once a day. Some volunteers were not about to walk the length of a football field to do so. Even as automatic reporting stations were introduced, the stations had to be close to buildings so the data cable could be run to the display. Even though the originally specified maximum cable distance was 1/4 mile, most automated COOP observer MMTS sensors ended up within 10 meters (33 feet) of the building, mostly due to the inability of the NWS to trench under driveways and sidewalks which acted as barriers to putting the temperature sensor in open spaces.

NASA GISS adjusts the data when they know that stations have been affected by local development or if they have been moved. However, the Metadata for this is often incomplete or simply missing. Those corrections are, of course, essential to maintaining the quality and integrity of the temperature data network so comparisons are meaningful over the period from 1880 to the present. No one knows if NASA GISS and their international equivalents have been doing that job as honest brokers or if they are using the wiggle room in their analysis to bias the data in the direction their managers would prefer. What do you think?

CONCLUSIONS
It seems to me that my estimate of 0.3ºC for Data Bias and Station Quality is fully justified, but I am open to hearing the opinions of WUWT readers who may think I have over- (or under-) estimated this component of the supposed 0.8ºC rise in global temperatures since 1880.

In my earlier posting in this Tale of the Global Warming Tiger series, I asked for comments on my allocations: to: (1) Data Bias 0.3ºC, (2) Natural Cycles 0.4ºC, and (3) AGW 0.1ºC. Several readers were kind enough to comment, either expressing general agreement or offering their own estimates. A few claim that AGW is ZERO (in other words, rising CO2 and land use changes due to human activities have no effect on temperatures or climate, due to the negative feedback from cloud albedo or other natural processes). I agree clouds have a net negative feedback (most official models assue a net positive feedback) but I do not believe this cancels out all the effects of CO2 on the Earth’s surface absorption of Solar radiation nor of albedo changes due to land use.

What do you think? I have been keeping a spreadsheet record of WUWT reader’s opinions, which I value, along with their screen names, and I plan to report the results later in this series.

This is what you may look forward to:

Normal Seasons of the Sun – How natural processes beyond human control, including Solar Cycles and Ocean Oscillations, are the actual cause of most climate change.

Some People Claim There’s a Human to Blame – Yes, human actions, mainly burning of fossil fuels and changes in land use, are responsible for some small amount of Global Warming.

Is the Global Warming Tiger a Pussy Cat? – If, as many of us expect, natural processes lead to stabilization of global temperatures over the coming decades, and perhaps a bit of cooling, we will realize the whole Global Warming uproar was like the boy who saw a pussy cat and cried tiger.

[UPDATED ~9PM 16 Jan 2011: Some readers don’t like the blinking graphs. We aim to please, Here are non-blinkers.]

87 thoughts on “The PAST is Not What it Used to Be (GW Tiger Tale)

  1. After looking at the data for the Arctic and Antarctic (see my articles here and at P Gossellin’s NoTricksZone), I think the bias is higher, perhaps as high as 0.5 degrees, and the ocean cycles lower. AGW is small; I haven’t decided how small or the sign. The bias at the poles is very high, especially at any location with a population over 10 people in local winter.

  2. George Orwell warned about this in 1984.

    Remember the role of Winston Smith in the Ministry Of Truth: his job was to rewrite historical documents so that the Party was always right. That meant altering photographs and other records if someone in the party fell out of favour.

  3. I love the graph with the james Hansen picture.
    Great article that debunks any warmist comment about activist climate scientists cooking the books.

    They are cooking the books and with science this has nothing to do.

    Besides that Mother Nature has eliminated all the remaining Global Warming since the global average temperatures made a drop of more than 0.6 degree Celsius since August of the last year.

    This fact alone defeats the entire theory that stated that our Co2 emissions were responsible for an unprecedented and unstoppable rise of global temperatures causing the ice caps to melt and oceans to rise.

    AGW AKA climate change as written down in IPCC AR 4 is DEAD

    http://www.accuweather.com/video/748914366001/global-temps-have-dropped-below-running-means-in-jan.asp?channel=vbbastaj

    All we’re left with is a huge pack of political spin, a failed doctrine and a big hole in the tax budget caused by all the money they stole.

  4. How many more times are you going to post the same false information? Each time you take some change, spend zero time investigating why something might have changed, imply something sneaky is going on and encourage your readers to get all hot and bothered about it. This is despite the fact that a) the hansen et al 2001 paper shows very clearly that the TOBS adjustments intoduced into USHCN2 are the biggest component of the change and b) all the Gistemp code and data is online and replicated.

    When does omission of important details become equivalent to simply lying?

  5. What do you think?
    ================================
    When the conversation is all about .5C

    I don’t think there’s anyone of the face of this earth that can estimate UHI….

    …but I do think UHI and changes in land use can account for all of it

  6. Actually, it makes all that talk about a “tipping point” make sense.

    It looks that “tipping point” has already been reached.

    And, added to that is GISS’s steadfast refusal to change their averaging period from 51-80 (knowing that the use of the older period keeps current anomalies higher), and their ESTIMATES of the Arctic temps (from stations 1200km away, also keeping the current anomalies highr), allows them to paint the most alarming picture ever.

  7. Good article, thank you. Put me in the humans have an affect on weather but through land use and particulate pollution rather than CO2.

  8. What amazes me is the lack of metrology control of the equipment. You could have something as simple as the$50-60 thermometer/data-logger offered by Anthony, make 10 days measurements next to the UHI affected gauge, move the thermometer/data-logger off 100-200 yards to a Class 1 area, 10 more days of data, done. You compare the data and calculate the slope/intercept. Do this twice a year (Summer/Winter) for range checks and I would be order of magnitude more confident in the measurements. If you were a purest and a budget like NOAA/NASA for climate monitoring, you setup a “same system” with a data-logger and increase the precision. Now you have just the gauge measurement error.

    Considering the conditions:

    Data bias: +/- 1.5deg (range: 3.0deg)
    Natural Cycles: +/-2.24deg (range 4.5deg)
    AGW: Below detection limits. Maybe detectable with gauge control.

    The adjustment off of “lighting” is beyond retarded. It is a procedural decision you make when fighting a forest fire or disarming a bomb, which is clearly not the case.

  9. “NASA GISS has been quite blatant in modifying the data even though they are aware that all the older versions exist in electronic archives. They have got away with it because no one in the major media or Congress seems interested in calling them on it.”

    That is by far the most scary part of the whole article. In other words, let’s change the data to match the theory and the media and congress look the other way.

    It would be interesting to create a “Fudge Factor” graph showing the difference between the old readings and the “new readings”, just to see how far they twisted things.

  10. The new Congress needs to address this kind of issue.
    Subpoena the folks responsible and get the data.

    Prosecute anyone that can be shown to have intentionally mislead us.

  11. Personally, I do not like the blink graphic since it makes detailed comparison all but impossible. I can see that it is visually effective (much like a sound bite) but I like to scrutinize and this one cannot do.

    Is it possible to post (possibly in addition) both the 1999 and 2011 graph side by side for those who wish to more than quickly glance at the graph?

    [Thanks for your interest. In response to your request, and that of another reader, I just added a non-blinking version at the end of the posting. Enjoy! Ira]

  12. Ira

    Whilst you put forward a case that temperature stations in the US are generally poorly sited and you put forward reasons as to why this may affect the accuracy of temperatures ascertained and even temperature anamolies extrapolated therefrom, no quantative evidence is submitted supporting your view that this can account for 0.3 deg C of the observed 0.8 deg C of observed warming.

    Ftom my reading of your paper, the 0.3 deg figure is ascertained from the assumption that the observed 0.8 deg C of warming is the result of 3 factors, namely bias caused by UHI and/or adjusted temperatures, natural variation and CO2. You assume that the latter 2 factors contribute a total of 0.5 deg C of the observed 0.8 deg C of observed warming. From this you get the 0.3 deg C (ie., 0.8 – 0.5) for temperature/adjustment bias. In fairness, I seem to recall from your earlier post that you did explain how you derived the 0.4 deg C component for natural variation.

    I personally suspect that there is UHI bias in the data set and I suspect that the adjustments made to the data set have not properly eliminated this factor and have, together with station drop outs, very probably exacerbated it. However, I have yet to see good quality data quantifying this very important aspect.

    It is of utmost importance since if temperature/adjustment bias accounts for 0.4 deg C of the observed warming and if you are correct that natural variation accounts also for 0.4 deg C of the observed warming, then the effect of CO2 would be ZERO.

  13. I agree with Richard. Those blink things drive me crazy. I start to swear every time a blinking add pops up on my computer screens. An web sits that blink get booted in about 2 seconds.

  14. Much like the mid evil warming period, they need to make 1998 “disappear”. Revisionist history, just like George Orwell warned as about.

  15. Bob Diaz re: Fudge Factor graph

    Go to climateaudit.org and prowl around the topical threads. Steve McIntyre regularly calculated and posted charts showing the difference between historical data and Hansen-adjusted data. They typically show stair-step patterns where increments of “adjustment” were added/subtracted from the historical records to “make things right”.

    Many commenters regularly use the term “gob-smacked” to describe their reactions.

    Breathtakingly fraudulent.

  16. GISS has a very simple procedure about what to do on station moves.

    http://www.warwickhughes.com/blog/?p=753

    The step (towards cool) that is introduced when a station is moved outwards from a developing city is removed by adjustment. This way, they can get the full benefit of UHI multiple times into a spliced series.

  17. We believe the “data” has been altered and can at least show that data posted in different years was altered to meet some agenda. I’m not sure that I believe the human portion of actual readings has any real effect on the world’s energy budget. Since bias can occur at many times, I know it should be higher. I would have to go 40% bias, 60% natural.

  18. fredT asked:

    “When does omission of important details become equivalent to simply lying?”

    And since when does omission of important details add to the science?

    That would be like your doctor not mentioning the fact that the operation he’s about to perform always causes the death of the patient.

    But hey, at least he didn’t lie to you…

  19. fredT says:
    January 16, 2011 at 3:12 pm

    “How many more times are you going to post the same false information?…..
    …..This is despite the fact that a) the hansen et al 2001 paper shows very clearly that the TOBS adjustments intoduced into USHCN2 are the biggest component of the change and b) all the Gistemp code and data is online and replicated.”

    Lets work through this together Fred.

    1st adjustment: OK, adjustments may be needed as new information comes in.

    2nd adjustment: Still OK, further new info may have come in.

    3rd adjustment: Hmmmm, getting a little suspicious but OK we’ll accept it.

    4th adjustment: C’mon guys, do you know what you’re doing?

    5th adjustment: Is this incompetence or something else?

    6th adjustment: In the private sector, these guys would be fired

    7th adjustment: Fred is confident that these new numbers are totally correct and reliable despite the fact that GISS effectively admits their previous 6 attempts were wrong. The rest of us think there is something fishy going on, but that’s because we are natural sceptics and not gullible lemmings.

    Do I have that about right Fred?

  20. The hochey stick disappeared in the 99 but it seems to have made a come back again this year.I guess when the ship is sinking you try anything.

  21. fredT says:
    January 16, 2011 at 3:12 pm
    “How many more times are you going to post the same false information? Each time you take some change, spend zero time investigating why something might have changed, imply something sneaky is going on and encourage your readers to get all hot and bothered about it. ”

    I think my above comment answers your questions; the linked page explains GISS’ OWN procedure, which must lead to ever stronger adjustments of the past towards cool and the present towards warm. It is their standard operating procedure, and they make no fuzz about it.

    Warmists probably think that that is science.

  22. fredT says:
    January 16, 2011 at 3:12 pm
    “the hansen et al 2001 paper shows very clearly that the TOBS adjustments introduced into USHCN2 are the biggest component of the change”

    Maybe I could add some details, although you may not consider them important. Thomas Karl was an author of what many point to as the relevant paper on TOB adjustment. The paper proposed the use of a computer program to adjust for TOB based on assumptions about changed data collection times rather than actual station records. Interestingly that 1985 paper’s conclusions mention “climate change” and indicate “The main advantage of this model is that it eliminates the cumbersome task of obtaining data at first order stations, and then calculating and interpolating the TOB to the location of interest”

    A further detail of interest may be Reto Ruedy’s email comment from the NASA emails obtained under FOI , “I still think, Steve […] mixes us up with Tom Karl’s group – they “fix” station data, we don’t.”

    While there are valid reasons for adjusting for TOB, the methods used and the resulting adjustments require a lot more scrutiny.

  23. Seems thorough Ira. Can’t remember who it was, but they claimed they had examined the codes used by CRU in adjusting data. The codes had an inbuilt bias of around 0.1 C. For the records from late 1800’s and early 1900’s there was virtually no difference but as the code proceeded towards the end of the 19th century, the divergence was growing noticeably larger.
    Certainly, with your work combined with that type of programming knowledge, certainly shows the variations in temperature are not unusual.

  24. ” Keep in mind that their mid-1990′s data has been in hand for over a decade and their mid-1930′s data is old enough to collect Social Security :^), yet they have made that old data work ”

    Without increasing retirement age that there is no saving Social Security. So your complaintis that they are being responsible! :-)

  25. fredT
    “When does omission of important details become equivalent to simply lying?”

    When you repeatedly try to correct the “omissions” to get the story nicelylined up. Reality has tohew to models or ideology. ANIMAL FARM commandments anyone!

    Oh! BTW I am not at all neither hot or bothered, just ROTFLMAO

  26. Ed Caryl says:
    January 16, 2011 at 3:00 pm
    … I think the bias is higher, perhaps as high as 0.5 degrees, and the ocean cycles lower. AGW is small; I haven’t decided how small or the sign. …

    Congrats Ed Caryl for being the first to comment in this new thread. Thanks for your estimates of allocation for the supposed 0.8ºC of warming since 1880. I put you down in the spreadsheet as 0.5ºC for Data Bias, 0.3ºC for Natural Cycles, and 0.0ºC for AGW.

  27. So the only thing presented as a counter-posit, so far here is effectively this: “I will continue to believe manipulated data, I don’t care how many times, nor how much that data has been manipulated. Now shut up because I am unanimous in my decision.”

  28. fredT says:
    January 16, 2011 at 3:12 pm
    How many more times are you going to post the same false information? Each time you take some change, spend zero time investigating why something might have changed, imply something sneaky is going on and encourage your readers to get all hot and bothered about it. This is despite the fact that a) the hansen et al 2001 paper shows very clearly that the TOBS adjustments intoduced into USHCN2 are the biggest component of the change and b) all the Gistemp code and data is online and replicated.

    When does omission of important details become equivalent to simply lying?

    Thanks fredT for showing up here and for your interest in WUWT. Only time will tell if we need people like you here to straighten us out, or if it is the other way around :^)

    Please re-read my Topic. Be specific about what I may have said that was false information. We have the email from Dr. Makiko Sato to Dr. James Hansen detailing the seven updates, over nearly a decade, plus another update that occured after the email. Some differ by over o.5ºC between 1934 and 1998. If any one of them are right, all the others are wrong, some by 5/8 of the total warming they are trying to measure.

    You and others put it all down to TOBS (Time of Observation) for the old data. OK, but, if the old data included actual TOBS, why recalculate it seven times? If it did not include TOBS, how could they ever account for it? See TOBS Climate Audit debunking.

    If a corporation, between 1999 and 2007, changed their profit reports for their 1934 and 1998 seven times over the course of a decade of recalculating, and all but the last two indicated profits in 1934 were higher, would you suspect they were cooking their books to show false profit increase? What if their supposed profits were $800M and the delta in their calculations was $500M. I’d sell my stock immediately, if anyone would buy it!

    But, you may wish to hold on to your investment in Hansen, Inc. Good luck.

  29. Ira, Steve posted this on his blog, also 1999 Hansen, just 11 years ago:

    “Empirical evidence does not lend much support to the notion that climate is headed precipitately toward more extreme heat and drought. The drought of 1999 covered a smaller area than the 1988 drought, when the Mississippi almost dried up. And 1988 was a temporary inconvenience as compared with repeated droughts during the 1930s “Dust Bowl” that caused an exodus from the prairies, as chronicled in Steinbeck’s Grapes of Wrath.”

    http://stevengoddard.wordpress.com/2011/01/17/hansen-1999-empirical-evidence-does-not-lend-much-support-to-the-notion-that-climate-is-headed-precipitately-toward-more-extreme-heat-and-drought/#comments

  30. A scientific paper on TOBS is not going to tell you anything about what was actually done to the data.

    To do that, you needed to peak over the shoulder of the programmer who was actually making the changes in the database. Then you need to be sure that none of those changes magically appeared in the raw data database as well. I don’t think we can be confident that bias was not introduced when Tom Karl and Tom Peterson and James Hansen were peaking over the shoulder of the programmer.

    The difference in the trends between the satellite measurements and GISS, Hadcrut3 and NCDC indicate the data bias is 0.3C.

  31. I think it should read “warmer” rather than “worse”. I do not think it is appropriate to put a positive or negative connotation on warming. Personnaly, I believe the warming we will experience will be a good thing.

  32. Ira: Interesting post. I don’t have an issue with the way you parcel out the causes of our “temperature” readings. The satellites – which I accord some legitimacy – show some warming over a period of time, but a lot less than the terrestrial analogs. So let’s give 0.5 dC to nature. I’ll even grant 0.1 dC to CO2, since a little apparently gives a lot of result.

    Now, the remainder – the never-ending, unidirectional “adjustments” – are either bad science or cheating. Let’s really air this issue out and determine which. If the former, let us fire the perpetrators (without pensions). If the latter, let’s fit these guys in stripes (or pink, per Sheriff Arpaio’s teachings) for a 20-year gig in modeling (clothing, not climate).

  33. There is so much wrong with this I dont know where to start.

    data source. the graphic here is unsourced. worse than that it does not match Hansen99. A few of us fought long and hard to get the GISS source code so that we could do serious work. I’m sad to see this kind of thing posted. The data sources between H99 and subsequent papers ( and graphs) have changed, but you have to actually read the papers and learn the code to understand how these changes work. in the US for example after H99 Hansen added some addition QA to incoming data streams. In the US that led to a few stations being removed from the record. The data removed was concentrated on the early portion of the record. On of the reasons I requested the code was to see what was being done with these stations. In the end, Hansens choices held up. They were bad stations that needed to be removed. ( yes these were in the US and I checked them all back in 2007 to see if hansens decision was justified )In addition as more data comes in the estimate of the past will change. This is one of the benefits of the Rference stations method. next, in H99 Hansen used population to determine urbanity. Post H2001 he used nightlights. In H99 he used a two legged adjustment procedure for UHI. The hinge point was pegged at 1950. This was improved post H99 and the hinge point is not predetermined.

    The post is so full of speculation and conspiracy type thinking. Sad.

  34. WRT 3C of warming being Bias

    Do not forget that the land is 30% of the total. Do not forget that for the most part the land temps are largely in line with SST. they are slightly warmer and more noisy.

    no data supports a speculation of .3C. The BEST skeptical work suggests .3C in the LAND RECORD… or ~.1C in the combined land/ocean.

    Hansen suggests ( in places) perhaps .1C in the land, jones suggests .06C

    Look at UHA and RSS.. the bias could range from 0 to .1C globally.

  35. Aaron says:
    January 16, 2011 at 5:47 pm
    I think it should read “warmer” rather than “worse”. I do not think it is appropriate to put a positive or negative connotation on warming. Personnaly, I believe the warming we will experience will be a good thing.

    Right Aaron, from a non-CAGW point of view, a relatively small change in mean temperatures is no big deal, and moderate warming is probably better for general human society than cooling.

    But, the sentence you refer to a the top of my posting refers to what the CAGW crowd may be doing to the data to further their catastrophe-based theory that we are headed for a disastrous, runaway, “tipping point” situation where temperatures will increase rapidly by as much as 10ºC by 2100. From their point of view, more warming is worse for life on Earth in that it adds credence to their catastrophic predictions.

    When you are Chicken Little, selling the theory that the sky is falling the worse you can make the situation appear (i.e., by claiming more people got hit on the head), the better it is for your theory, so that makes what is worse for society better for you. Which is why Phil Jones wrote in a Climategate email: “If anything, I would like to see the climate change happen so the science could be proved right, regardless of the consequences.” On the other hand, we skeptics will be shown to be correct if CO2 continues to rise but, contrary to CAGW, we get some cooling over the coming few decades, which, if warmer is better, means what is worse for society (cooling) is better for our theory. Up is down and down is up!

  36. Dr. G has asked for our estimates of the CO2 effect. Well, OK, I will present mine, and the reasoning behind it. I am not a “climate pscientist”; indeed my only scientific achievement was the 11th Grade Trigonometry Prize half a century ago.
    We know that the depth of IR penetration into water is on the order of microns, not even millimeters.
    We know that UV penetration, in contrast, is in the hundreds of meters.
    We know that one of the effects of an “active sun” is an increase in UV, some of which escapes absorption by stratospheric ozone and scattering by the troposphere.
    We know that the top few millimeters of ocean is actually cooler than the mixing layer beneath it, due to evaporation.
    We know that nobody, not even Dr. Curry, has been able to produce a coherent explanation of how increased IR in the atmosphere can heat the oceans. Wind and cloudiness have definite effects; IR apparently does not.
    We know that around 70% of the Earth’s surface is ocean.
    We know that due to the huge difference in heat capacity — the top two fathoms of ocean containing more heat capacity than the entire atmosphere — surface air temperatures cannot measurably affect ocean temperature; the influence is the other way around.
    We know that anything that influences the hydrological cycle (IR, cosmic rays, UV, winds at various levels in the atmosphere) will have an effect on ocean temperature — and anything that influences ocean temperature will have an effect on the hydrological cycle.
    We know that on the one hand, without a greenhouse effect the Earth’s surface would be around 30 deg C cooler. We also know that with a full calculated greenhouse effect, the Earth’s surface would be from 15 to 45 deg C warmer (I’ve seen both figures; the lower is the one I was taught in 8th Grade Earth Science). This by itself implies that negative feedbacks reduce the overall greenhouse effect by between half and three-quarters.

    So my private, layman’s opinion is that the actual effect on global temperature from a doubling of CO2 concentration is somewhere between half a degree C and unmeasurable, with the probabilities favoring the latter.

  37. fredT says:
    January 16, 2011 at 3:12 pm
    …”Each time you take some change, spend zero time investigating why something might have changed, imply something sneaky is going on and encourage your readers to get all hot and bothered about it.”
    ===========
    But…, wait…, I’m not “hot and bothered”, am I lacking encouragement or what?

    Then, you say: “When does omission of important details become equivalent to simply lying?”

    Do you really want to examine the cost/benefit ratios of “green” technology??

  38. I figure the actual temp rise since 1880 is 0.5C after bias is removed and tend to believe that’s the real number for anthropogenic forcings from everything. CO2 is about half that with black carbon (soot), methane, and land use changes accounting for the other half. The warming would be more without aerosols to take some of it away. It was probably a mistake in the 1970’s to limit aerosol emissions by warrant of a bogus acid rain scare and that accounts for the slightly higher rate of warming since then. The physics seem to work pretty well for those numbers. What doesn’t work is the water vapor amplification that turns some beneficial moderate warming into something that might be less than beneficial. There’s no positive feedback from water vapor in response to AGW. Absent that feedback the whole thing becomes a non-issue and we’re left with the only reasonable motivation for cutting back fossil fuel consumption is to conserve a finite resource until such time as a cheaper source of energy is available. The key is the alternative needs to be cheaper because that’s the only path that will result in economic growth and higher standards of living for everyone. Other limited resources, fresh water and phosphorous just to mention two, are just as much a problem as fossil fuels and those are getting shortchanged by the unnecessary alarmism over CO2 emissions. CO2 is plant food, it isn’t a pollutant, and is in and of itself beneficial in rising atmospheric concentration.

  39. “Is the Global Warming Tiger a Pussy Cat?”
    I think the more rational question is if the extra CO2 isn’t actually something beneficial. There is no reason to assume that it is necessarily bad in any way and it is obviously very beneficial in that greenhouses have extra CO2 for a reason. A moderate amount of warming, more than today, was called the Climatic optimum before activists made it fashionable to assume that warmth was kin to evil.

  40. steven mosher says:
    January 16, 2011 at 6:10 pm
    WRT 3C of warming being Bias

    Do not forget that the land is 30% of the total. Do not forget that for the most part the land temps are largely in line with SST. they are slightly warmer and more noisy.

    no data supports a speculation of .3C. The BEST skeptical work suggests .3C in the LAND RECORD… or ~.1C in the combined land/ocean.

    Hansen suggests ( in places) perhaps .1C in the land, jones suggests .06C

    Look at UHA and RSS.. the bias could range from 0 to .1C globally.

    Thanks for your comment, Steven. Although my blink graphs are for US data my allocation of the supposed warming since 1880 is intended to be global, including both land and sea. You wrote “UHA and RSS” but I’m sure you mean University of Alabama in Huntsville (UAH) and Remote Sensing Systems (RSS), the groups that process satellite data. The satellite sensors (according to Wikipedia) “measure radiances in various wavelength bands, which must then be mathematically inverted to obtain indirect inferences of temperature…” Is it this processed data that you and Jones and Hansen say may have a bias of from 0.0ºC to 0.06ºC to 0.1ºC ? What is the bias with respect to? Measuring stations on land?

    If the bias is below 0.1ºC, how can the adjustments in the blink graphic be three times as large? (I know my blink greaphics are US-only and UAH/RSS are global, but, over decades of data I would expect similar bias. Where am I going wrong?)

    Also, satellite data is has been available only since the 1960’s and the supposed warming of the 0.8ºC is from the 1880s. How would you resolve the issue and allocate that 0.8ºC of supposed warming? If Data Bias is 0.1ºC or less, you have 0.7ºC or more to split between natural processes and AGW. The key point of this exercise is to show whether the human contribution (AGW) is a minor issue (or not). I’d appreciate your opinions and reasoning on how to make that allocation.

  41. Its prisontime for the NOAA/Hockeyteam!

    Can someone call the police please? This fraud just have to stop!!

  42. “measure radiances in various wavelength bands, which must then be mathematically inverted to obtain indirect inferences of temperature…” Is it this processed data that you and Jones and Hansen say may have a bias of from 0.0ºC to 0.06ºC to 0.1ºC ?

    ##############

    No. let me be clearer for you.

    The UHI bias in the land record can be estimated in several ways.

    1. Compare Urban sites to Rural sites. These comparisons yield answers from
    0C to .06C to .1C.
    2. Reggression studies. ( McKittrick) puts the value for UHI at about .3C.

    3. Longitudinal studies ( there are only a couple that look at a couple big big cities)
    around .3C per century.
    4. Comparing the temps given by RSS and UAH to the surface. Simply.
    from 1979 to today RSS shows .16 C decade warming in Troposphere
    ( no UHI) this trends matches the surface and means no UHI bias in the surface.
    UAH runs at about .13C and bounds the bias.

    “OK, but satellite data is has only been available since the 1960′s and the supposed warming of the 0.8ºC is from the 1880s. How would you resolve the issue and allocate that 0.8ºC of supposed warming? If Data Bias is 0.1ºC or less, you have 0.7ºC or more to split between natural processes and AGW. The key point of this exercise is to show whether the human contribution (AGW) is a minor issue (or not). I’d appreciate your opinions and reasoning on how to make that allocation.”

    Well, You can’t even begin without actually going out and looking at the data. The other issue is the problem of the definition of natural process and how one goes about understanding aerosols. basically, if you want to attribute a portion of the observed warming, you have to run a GCM or do something similar to what Bob Tisdale and Tamino have done.. subtract the natural oscilations and you are left with a trend.
    Finally the bigger issue is how much additional warming can we expect ON TOP OF the natural cycles. That may be small or may be large.

  43. Bill Illis says:
    January 16, 2011 at 5:38 pm

    A scientific paper on TOBS is not going to tell you anything about what was actually done to the data.

    To do that, you needed to peak over the shoulder of the programmer who was actually making the changes in the database. Then you need to be sure that none of those changes magically appeared in the raw data database as well. I don’t think we can be confident that bias was not introduced when Tom Karl and Tom Peterson and James Hansen were peaking over the shoulder of the programmer.

    ##########

    . You can as some of us have

    1. Read the TOBS paper and implement the algorithm.
    2. Read the follow on papers verifying the approach.
    3. Read the work of notable skeptics on the topic ( back before 2007) and go over their work ( we did this on CA, the data is still around if you want it ) and see that Karl was right.

    and then you will realize that the issue with TOBS is not the adjustment but the uncertainty in the adjustment. the error bars need to be adjusted.

    Further, you could request the code from Karl ( he provides it I believe) and you could realize that his algorithm is only applied in the USA. Other countries do their own adjustments. They have to because the adjustment is based on an empiricial model that needs to be developed for each geographical region.

    Finally, the difference between GISS/NCDC/CRU/RSS is not .3C
    they are roughly identical. ~.16C per decade from 1979 to 2010.
    UAH.. is roughly .13 per decade. So, if you want to say the those 4 are biased high
    by .03C per decade during the period 1979to 2010 you could. That’s a fragile basis to argue for any bias prior to that period, until you identify the exact source of the bias. and since UAH is the odd man out, you got some explaining to do to justify it at the source of truth.

  44. >> steven mosher says:
    January 16, 2011 at 6:10 pm
    WRT 3C of warming being Bias

    Do not forget that the land is 30% of the total. Do not forget that for the most part the land temps are largely in line with SST. they are slightly warmer and more noisy. <<

    Are the SST before satellite data anything more than garbage?

    And what is the effect of measuring air temperatures over land and then combining them with water temperatures? Aren't they adding apples and oranges? Far from large land masses I'd guess that the air temperature is the same as the ocean temperature, but there's a lot of sea area where that wouldn't be true.

  45. Ira – Thank you for your interesting paper.

    I cannot comment on global temperature, but I can report on what I have found about the impact of UHI at two locations that I have studied in Australia, namely the state capital cities of Sydney and Adelaide.

    As An introduction, I advise that I have found that annual rainfall is a good proxy for maximum annual average temperature in many parts of Australia. That helps me to indentify the incidence of UHI, its initial date and the amont it has contributed to temperature each year.

    Sydney: Rainfall very variable but the long term (1866 to 2010) is flat – no trend.
    Temperature – ditto 1866 to 1957. In 1958 there ws a substantial (well documented) change in the built environment. From then on annual temperature rose at an annual linear trend rate of 0.0192 degrees celsius per year (approx 2 degrees per century).
    Long term monthly average temperatures were also trendless up to 1957, but from 1958 onwards began to rise as well. However the trend in the warmer months was very significantly less than in the remaining months, because of a seasonal change in the prevailing wind patterns. This has all been very well documented.

    Adelaide: Ditto to Sydney, including ditto comparison with rainfall, with the change occurring in 1978, with a change in location of the thermometer from the fringe to the centre of the CBD. Until then, temperature and rain moved roughly together. After that, temperature sored up and away.
    Fortunately there is another temperature gauge outside city limits with 20 years history before the move, which does not veer upwards afterwoods, but continues to keep partnership with the rain.

    I have read many reports that temperature in many scattered locations has not changed for over 100 years.
    My conclusion is that the global temperature has not changed since recovery from the little ice age, some time in the middle of the 19th century. All the measured increase since then is likely to be due to UHI, plus possible “after the effect adjustments” which I personally have not seen.

    Anthony has my emil address.
    I can send you much more detail and charts if you are interested.

  46. Ira

    Thank you for updating your post to include the two graphics side by side. At least for me, this made comparisons much easier. I have found your two posts to be interesting and apologise for this lengthy comment.

    You say that you are keeping some sort of record of responses. Of course, most on this site are sceptical and therefore one would expect the majority of respondents to share the premise of data bias. I personally consider that there is a very strong case that the data is significantly adulterated such that it is dangerous to draw too much from it. My personal view is that there has been less warming than the adjusted record shows. i.e., whilst I accept that the US has warmed since 1880, I do not think that it is by as much as 0.8 deg C. If I has to stick my neck out, I would ‘guess’ that the temperature has risen by 0.6 or may be by only 0.5 deg C. This is only a ‘guess’ since the data is too corrupted to reliably extrapolate and I do not accept the validity of seeking an average temperature over a such a large area with so little station data. In my opinion, one would need millions of uninterrupted station data sets to obtain a reliable average,

    At best, the sheer number of adjustments made suggest that the ‘scientists’ have been extremely sloppy in the compilation of the data set, at worst, it smells of incompetence or even deliberate manipulation to support a held belief. Whatever be the explanation, one is left with little faith in a data set that has been so extensively adjusted/massaged.

    I suspect that all this hysteria (and this is an understatement) came about when some fool (and I consider this to be the correct word) decided that they would draw a best fit straight line through the record from 1880 to date. Any sensible person looking at the record would have realised that that fails to describe the data which consists of periods of warming and cooling. A sensible person would either have drawn a series of best fit lines for the warming and cooling periods such that there would be a series of up lines, down lines, up lines, down lines etc or would have drawn a wave curve as you have. Had this been done, it would have been immediately obvious that temperature changes are cyclical in nature.

    How much of the recorded temperature change is natural variation? It is impossible to say since by definition we do not know what those variations consist of and whether they are more or less pronounced than they were in the early part of the last century. Having said that the temperature change from about 1908 to 1945 appears to be 0.6 deg C. During this time manmade actions (burning fossil fuels, deforestation) was relatively modest such that one could argue that all the temperature increase during that period was due to natural variation. Indeed, Phil Jones accepted that the temperature rise through to the 40s was due to natural variation.

    Between 1945 and 1965, natural variation reduced the temperature by 0.4 deg C. This was of course during a period when manmade actions (burning fossil fuels, deforestation) had significantly increased and could (theoretically) have become a factor, If manmade actions were having any effect on temperatures (ie., pushing them upwards), it means that the natural variation during this period was more than 0.4 deg C so that the negative effect of natural variation and the positive effect of manmade factors resulted in an overall cooling during the period of 0.4 deg C.

    Of course, the interesting period is that between 1965 and 2005 where the temperature change is a little over 0.8 deg C. Since there was about 0.6 deg C of natural variation between about 1908 and 1945, there is no reason to presume that natural variation should be any less between 1965 and 2005. Indeed, since we do not know the processes involved or the extent thereof, natural variation could of course account for the entire temperature increase. I consider it probable that it does, but in reaching this view, I suspect that the recent temperature anomalies are not as large as suggested in the data set and that the real temperature rise is somewhat lower at about 0.6 deg C or may be even as low as 0.5 deg C.

    Thus my view (which is a ‘guestimate’ due to the poor quality of the data) is that temperature adjustments/bias accounts for 0.2 or 0.3 deg C and that natural variation accounts for 0.5 or 0.6 deg of the warming that has occurred since 1880.

    I suspect that CO2 has had no measurable effect on the real warming that has taken place (possibly because changes in cloud albedo and the negative feedback that they give rise to). I can see that deforestation could have an effect on climate/climate patterns but through luck whilst deforestation is a growing issue, the increase in CO2 in the atmosphere has resulted in more biomass which has off-set the effects of deforestation. I can also see that increasing urbanisation and all the heat energy that we create annually could have some effect. However, in terms of global area, urbanisation is small and the most significant driver (leaving aside the sun) of climate are oceans and the latent heat that they store and the circulation patterns and currents that they produce. I do not consider that CO2 in the atmosphere can effectively heat the oceans and any change in ocean temperature is due to natural variation very probably changes in cloud albedo

    To summarise: I would not forcefully join issue with your assessment but consider it a matter of some conjecture. My personal suspicions are data adjustment/bias 0.2 to 0.3 deg C, natural variation 0.5 to 0.6 deg C, and CO2 Nil

  47. (1) Data Bias 0.65ºC (including UHI, data-corrections and mislocated stations)
    (2) Natural Cycles 0.15ºC (since pace of natural warming has been slowing down and we are experiencing cooling already + temps in times of cooling drop faster than that they go up in times of warming)
    (3) AGW 0.0ºC (man has no effect on climate, CO2 is plantfood, H2O is the main GHG)

  48. Ira, thanks for the post. Suffice it to say, some of us have been down this road many times already. But each time adds a nuance. I agree with your post. But because I’ve been done this road, I’ll respond to comments, some of them that haven’t been posted yet.

    The tricky thing about our temp data, is that it changes hands. One of the memes you may hear in this thread is that GISS isn’t adjusting the data, but some other agency is prior to delivering it to GISS. (Such as NCDC or NOAA, or whomever. You will also see, as Fred points out, that their formulas and codes are available. These two statements are both true and both tell lies. It is entirely disingenuous to assert GISS doesn’t know what is happening to the data prior to receiving it. If they were, the numbers they produce would be of no value. I assert, but cannot prove, adjustments are made in full knowledge and at the behest of the entire custody chain, but is done by different agencies under different pretexts in order to convolve and obfuscate. So, too, are the codes and formulas. How many times is it ok to adjust data that is 70 years written in the history books? They can publish what ever code they choose and it wouldn’t change the fact that the properties of mercury in a tube were well known even in the 30s. And the TOBS didn’t change 7 times in the 30s. The desired results changed 7 times recently.

    Ira there is a couple of things about your post that I would change. First, you said, “I have used US data for my example because those sources are more under NASA GISS observation and control than most international data, which may be of poorer quaity[sic].” I guess quality in this case is a subjective word. As I pointed out earlier. The data changes hands more times than a basketball at a Globetrotter event. And with each hand, changes are made to the data, either intentionally or though normal errors that occur in transport and interpretation of data. I should note, that often the words “raw” and “un-adjusted”, when discussing temperature data, often take on Orwellian characteristics and mean opposite the connotation these words conjure. My point is, by the time GISS gets done doing whatever it is that they do to the data, its been so homogenized, normalized, and any other “ized” you can think of, I’ve little reason to believe they accurately reflect reality. But, if you’re looking for a temperature equivalent to fun house mirror, it is of good quality, the best taxpayer money can buy!

    Oh, and the other thing I’d change. Typically, when discussing GISS data and adjustments, the text should ooze with loathing sarcasm.

  49. Further to my last post. As regards the period 1945 to 1965, I should perhaps have mentioned aerosols as being partly responsible for the observed cooling. This in my view is theoretically a possibility but upon which there is no quantative evidence and is a mere matter of conjecture.

    One problem is that the satellite data does not go back long enough and if past temperatures pre 1950 are being adjusted downwards, this exagerates the warming that has occurred in more recent times. I consider that land surface data pre satellite observation is now so unreliable that it could fairly be described as garbage.

    As regards TOBs, I do not know how this is done. However, I do know that whilst one can usually point to the hour when temperatures will be warmest, at times one experiences prolonged spells where cloud disipates late afternoon such that temps are at their peak later than usual and also one sees spells where it is fine in the morning but clouds over at midday depressing midday and early afternoon temperatures. Further, one can get warm fronts coming in (possibly with rain) such that the temperature rises as the warm front passes. The point is that to make an accurate TOB adjustment one would need to know the weather for that day. The TOB adjustment could be distorting the temperature as much as making no adjustment.

    In fact in coastal areas and mountainous areas, it may be that peak temperatures are generally experienced at a different time of the day when compared to the norm for central low lying plains. Is there any data on this and if so does TOB take this into account?

  50. If they had tacked on the warming of the 1850’s to 1870’s, we’d see the natural rhythm of climate as a roller coaster ride.
    Instead, they have this ridiculous Global Warming causes Global Cooling that has everyone in stitches.
    It really does hurt to laugh when it’s cold, unfortunately.
    Next time I want to heat my kitchen, I’ll open the freezer door, and next summer I plan to run the kitchen stove to cool off.

  51. Has there been any warming at all? I wonder because I recall reading, possibly at Musing from the Cheifio some time last year, that the international convention for reading daily high and low temperatures was changed in the 1960s to avoid fractional readings with the following convention, temperatures above 0 degrees C rounded up to next whole number, temperatures below 0C also rounded up to nearest whole number. I may be mistaken but to me it sourced the approx 0.5C warming the panic seems to be all about. Coupled with the missing M for minus from the airport data it it certainly had me laughing. Now if I understood E.M.s musing ,the ineptitude of our agencies knows no bounds. This curious insight was soon followed by Environment Canada’s response as to the state of Canadian weather data and ever more interesting relevations. Every time I find myself staggered by the incompetence revealed.Why in this climatology field is it always worse than a cynic could imagine?The phrase, you could not make this s**t up, keeps running thro my mind.

  52. Tom_R says:
    January 16, 2011 at 8:00 pm

    >> steven mosher says:
    January 16, 2011 at 6:10 pm
    WRT 3C of warming being Bias

    Do not forget that the land is 30% of the total. Do not forget that for the most part the land temps are largely in line with SST. they are slightly warmer and more noisy. <<

    Are the SST before satellite data anything more than garbage?

    ###############
    anything more than garbage? Well, If one had the truth to compare the SSt measures to, then one could say with confidence whether they are garbage with conviction. But thn one would have access to the truth and the SST measures would be unnecessary.

    So, they best we can do is make careful note of of the various factors that may bias measures. When we make careful note of all these caveats, then we can begin to understand if the biases are symetrical about zero or not. And we can gain some insight into the overall accuracy. we can also, calibrate. So, only if we we have access to the truth OR if we careless can we conclude that they are "garbage". In short, calling them garbage is a hypothesis, test that.

    "And what is the effect of measuring air temperatures over land and then combining them with water temperatures? Aren't they adding apples and oranges? Far from large land masses I'd guess that the air temperature is the same as the ocean temperature, but there's a lot of sea area where that wouldn't be true"

    If yu were trying to measure "the temperature", then combining air temps and SST temps would be problematic. But the global series are INDEXES. that is, they are "proxies" for the "average" temperature. What matters is that you have a consistent method over time. The primary measure folks are interesting in is the trend.

    If you are really interesting in a physical quantity that makes sense, OHC would be the measure to look at.

  53. Ira,

    Your 1999 graphic is still wrong. I have no idea where you got it but it is nothing like what Hansen Published in 1999. I think you might want to check your sources.

    Ideally, what people should do is request the data for H99. I say this because the stations used in 1999 DIFFER from the stations used after 1999. So, its quite unfair to make this comparison without reconciling the data first. As I noted after 1999 Hansen removed at least 5 US stations. This is noted in his text. These 5 stations all had severe quality issues. As I noted I got interested in getting the code partly to address this issue.

    Next, you have to realize that the algorithms for identifying Rural stations have improved. They still need work, but the 1999 approach was very crude. Also, the urban adjust algorithm changed. Again, the change was for the better.

  54. Can anyone possibly believe there is no conflict of interest with Hansen keeping and manipulating the data while being an absolute 200% Warmist? Talk about re-writing history.

    One question please, if the earth is getting warmer, and UHI is increasing why are the recent temperatures adjusted up instead of down?

  55. If I can raise a minor quibble about the blink-comparison graphs, it would be that they are subtly misleading. For obvious reasons, one graph has ten years more data than the other, but the inclusion of that data on the second graph in the same style gives a visual impression of a much larger change between the two than is actually the case. If the additional section were de-emphasised, it would be better, particularly in the interests of avoiding accusations of deliberate deception or some such.

  56. I am not going to say that there might or might not be issues with Dr. Hansen’s data. At the least he is guilty of observer bias and is in the wrong position as a scientist. Whether he can keep that straight or not is not the issue, this bias happens regardless and this is why scientists should not have positions on what is happening. The bias leads to incorrect assumptions, and I am just saying that this tends to tell people that “Dr Hansen is involved in a conpsiracy.” He might be, but the most likely explanation is that he is not looking at things clearly due to the aforementioned observer bias.

    I think the best way to look at this is to figure out whether the adjustments were done incorrectly or not. Find the errors in the procedure, not in changes in the data. You might be onto something here, and I won’t stop you in that regard…but GISS is now mostly in the public arena with very little code and adjustments not being documented.

    From what I can tell, there are explanations in every adjustment that I might question, but from a scientific point of view its not incorrect persae. The points to question I think come from assumptions he makes that are weak. This goes into the actual models (GCM’s) and not really into the actual data.

    The data could possibly be off by .3 C as you suggest. I would hazard to guess that its lower then that despite serious issues with UHI effects which to put lightly are not done very well…which makes the error fairly high. I just caution that the data itself might have issues in the adjustments, but that overall its going to be a far lesser issue then the real issues which goes into the GCM’s which make assumption about natural variability….

    Real issue is still what part of the warming is natural, and what is from man.

    But to cut this comment somewhat short: I will say this as my guess’s:

    +/- .1 C from data bias/error. I would guess that its more then likely in the positive direction due to the aforementioned observer bias….but that is an assumption to on my part.
    +.6 Natural variation.
    +/- .1 From human effects. I don’t think we understand climate enough to find such a signal in the data from our impact. Someday we might, but today I believe it is somewhere in this range. The negative is from negative feedbacks just to be clear from CO2/land-use changes.

    As a note, I would guess a lot less then .1 for human effects…just that I do think its possible we have effected the climate up to .1 C

  57. John Kerr

    “It isn’t much, but it is significant. There is no visibility as to why the additional corrections were made to temperature data that is 100 years old.”

    you have comparison charts for every station. start looking

  58. It is not only a matter of words that data bias, as used here, should be called data manipulation. There is a more impressive kind of bias in ground temperature measurements, resulting from non-random inclusion and drop out of stations. On an annual basis hundreds or even one thousand stations may appear or disappear. Compute the mean of the included or dropped stations and compare it with the population mean and variance of that moment. You will find 10, 30 or 60 sigma effects. In the GHCN data base you will find effects of 10 sigma or more in the years 1878, 1891, 1907, 1931, 1936, 1941, 1949, 1950, 1951, 1952, 1957, 1960, 1961, 1970, 1971, 1975, 1980, 1981, 1986, 1987, 1989, 1990, 1992, 1994, 2004, 2006. According to my estimates the station drift is good for an increase of 1.6 degrees Celsius per century. Don’t take this figure too seriously but it suggests a bias which is not under control and may completely overrule the manipulations.

  59. most automated COOP observer MMTS sensors ended up within 10 meters (33 feet) of the building, mostly due to the inability of the NWS to trench under driveways and sidewalks which acted as barriers to putting the temperature sensor in open spaces.

    Somehow I knew the automation of temperature stations was just too coincidental to the rise from 1970-2000 (the time most automation occurred in all areas of life) for this not to have affected the reading. At first I thought it might have been that manual readings were done at a specified time …. and people being what they are, they didn’t do them exactly on the time tending to the colder periods.

    But what a simple explanation trenching is! It’s so bl**dy obvious! You’ve had a station positioned well away from buildings, the order to automate comes in, no one wants their prestine lawn, flowerbeds and paths torn up, so the order goes to move the station closer to the building to avoid hassle.

    Individually, the effect is negligible — less than a degree, and as they are all averaged out that individual move won’t be noticed. But when all the stations are under pressure to move closer to heat sources.

    Add to that the clean air acts which e.g. totally removed the London Smogs, the move away from open air fires to more efficient/clean power stations. The modern fad with putting out every single natural blaze which used to burn, putting smoke into the atmosphere with its cooling effect.

  60. In my earlier posting in this Tale of the Global Warming Tiger series, I asked for comments on my allocations: to: (1) Data Bias 0.3ºC, (2) Natural Cycles 0.4ºC, and (3) AGW 0.1ºC

    Starting with 2) The size of the natural cycle is consistent with the signal we have. That is to say, the signal is “normal” and so 0.8C could be due to natural cycles. From recollection around 10-20% of simulations based on the noise produced something that the hysterics would have thought was manmade warming. As I’m not familiar with that statistics of 1/f noise, I’m going to have to guess that 1/3 to 1/2 of the 0.8C would make us fairly confident that in the majority scenarios that change was natural. So, I think I’d estimate natural around the 0.3-0.4 with a range of perhaps -0.3 – +1C

    1) Data bias. For all that has been said, the local position of stations probably doesn’t amount to a lot of bias: a couple of degrees the day the wind blows from the wrong direction once a week. Obviously there are really bad stations, and a lot of really good ones. This bias could easily by 0.1C, I’d doubt it could be 0.5C because even the kind of idiots we saw in Climategate would spot that! So, probably I’d go with a smaller figure (everyone estimates high) for the local effects. I think I’d maybe add on something for regional effects (the change of land use). On top of that I’d add on something for the intrinsic bias of the “scientists” which I’d guess is at least 0.1 but could easily be 0.3C.

    3) Manmade warming. Take away all the fictional multipliers and just use the raw based on CO2 IR warming (although I still maintain CO2 is also a cooling gas, but I’d have to get off my arse to prove it before it would be appropriate to use it)

    4) In addition I would add a contribution for the reduction in global dimming since the 1970s. There’s no doubt in my mind (and it used to be in Wikipedia before the team got their hands on it) that there has been significant change in global dimming with a reduction in particulates leading to measurable increase in temperature (just look at the graph … joke that’s how the post-modernist climategate team doit!)

    5) Solar affects. Still a bit of an “elephant in the room” as no one knows just how big this component is.

    So, with the proviso, someone has to pay me if they want serious numbers, if I were making a bet I’d go with:

    (1a) Local Instrument placing bias 0.1ºC,
    (1b) Upjustments by “scientists” 0.1C (cherry picking of the warmist stations, etc.)
    (1c) Larger scale urban heating bias 0.1C
    (2) Natural Cycles 0.3ºC,
    (3) AGW 0.1ºC
    (4) reduction in global dimming 0.1C

  61. You ask for impressions of data from contributors for your files.

    1. The doubling of CO2 will have in my calculations the catastrophic effect on the world of at least 0.001C
    2. The effect of man on the temperature of the planet has possibly given us a .003C increase or decrease depending on the prevailing industrial processes at the time.
    3. The measured increases and decreases of temperature are natural.
    4. The thermometer measurements of recent times show an anomaly that is a figment of imagination of some.
    5. Unadjusted,unmoved and long term rural thermometers show no warming?
    6. The warming anomaly thus shown by the gurus is fictional, as the record would show if all fudges were removed.
    7. Warming and cooling of the earth are processes not understood and the science is still in stone age.
    8. When the gurus of climate can show falsifiable reasons for ice ages and interglacials they shall gain respect.
    9. When they can show falsifiable reasons for fluctuations of warm and cold periods in the interglacials. their understanding of climate may give them enough cudos to pontificate.
    10. Until that time arrives they are soothsayers and carpetbaggers.

  62. fredT says: “the hansen et al 2001 paper shows very clearly that the TOBS adjustments intoduced into USHCN2 are the biggest component of the change and b) all the Gistemp code and data is online and replicated.

    When does omission of important details become equivalent to simply lying?”

    FredT, first I think there are far too few warmists who post here so thanks because we believe open and honest debate is the only way forward … so I look forward to more from you in the future. But, please put forward the evidence. “Hansen et al” isn’t evidence here, it is a swear word!

    Oh … and when is simple omission equivalent to lying? When you have a Wikipedia article that talks about present warming, when we all know there has been no warming in the last decade.

  63. Mindert Eiting says:
    January 17, 2011 at 2:04 am

    You will find 10, 30 or 60 sigma effects. In the GHCN data base you will find effects of 10 sigma or more in the years 1878, 1891, 1907, 1931, 1936, 1941, 1949, 1950, 1951, 1952, 1957, 1960, 1961, 1970, 1971, 1975, 1980, 1981, 1986, 1987, 1989, 1990, 1992, 1994, 2004, 2006. According to my estimates the station drift is good for an increase of 1.6 degrees Celsius per century. Don’t take this figure too seriously but it suggests a bias which is not under control and may completely overrule the manipulations.

    10 sigma is only significant in climate science if the result produces warming…

    Just kidding, but this sounds very interesting. If you’ve gone that far, please produce a guest post and put it up!

  64. steven mosher says:
    January 17, 2011 at 1:39 am
    John Kerr

    “It isn’t much, but it is significant. There is no visibility as to why the additional corrections were made to temperature data that is 100 years old.”

    “you have comparison charts for every station. start looking”

    Why the h e double hockey stick should someone, anyone have to LOOK. The changes should be clearly marked and explained each and every time they happen. This is not a magicians game.

  65. steven mosher says:
    January 16, 2011 at 9:55 pm
    Ira,

    Your 1999 graphic is still wrong. I have no idea where you got it but it is nothing like what Hansen Published in 1999. I think you might want to check your sources. …

    Steven, my source was http://wattsupwiththat.com/2008/11/14/the-evolution-of-the-giss-temperature-product/, “Questions on the evolution of the GISS temperature product, Posted on November 14, 2008 by Anthony Watts”, where credit for the 1999/2008 blink comparator is given to zapruder.nl.

    I’d appreciate it if you have an online link that shows different US Annual and five-year mean temps from 1880 to 1999 published by NASA GISS in 1999. Since my source is a contemporary (2008) WUWT posting, and there are over 500 comments to that posting, and Anthony did not update his original posting based on any challenge to the base 1999 chart from GISS, I have no reason to doubt the data source.

  66. Data are. Each datum is a reading taken from an instrument, which measures a specific condition at a specific location at a specific time. Period.

    Good data are taken from properly selected, sited, calibrated instruments, read timely. Data which do not meet the criteria above are either bad data or missing data. Period.

    Adjusting bad data does not make it good data; it makes it adjusted bad (“fudged”) “undata”. Missing data is just that – missing. Data cannot be “infilled”. If a specific datum is important, properly site an instrument to measure it; if not, acknowledge that it is unimportant. If a sensor fails, replace it. That is neither rocket surgery nor brain science.

    Data are immutable. They are not the “stuff” of blink comparators.

    A mixture of data, adjusted “data”, infilled “data”, homogenized “data”, etc. is not data. It should not be treated as such.

    If knowing what the temperature is and how it is changing is important, we certainly have the skills and the equipment to to measure it accurately. One wonders why we have not chosen to do so.

  67. Dave says:
    January 17, 2011 at 1:16 am
    If I can raise a minor quibble about the blink-comparison graphs, it would be that they are subtly misleading. For obvious reasons, one graph has ten years more data than the other, but the inclusion of that data on the second graph in the same style gives a visual impression of a much larger change between the two than is actually the case. If the additional section were de-emphasised, it would be better, particularly in the interests of avoiding accusations of deliberate deception or some such.

    Thanks for your comment, but I think it is best to reproduce the base charts from NASA GISS exactly as they were published, to avoid charges of deliberate deception. Please note that my annotations point to peaks and troughs in the 1930’s and mid 1990’s that are common to both the 1999 and 2011 charts, and further than 5 years from the ends where the 5-year mean might be distorted. You might cover the years after 1999 with a piece of paper to avoid the distraction of the additional data in the 2011 chart.

  68. Mike Haseler says:
    January 17, 2011 at 2:46 am

    …So, with the proviso, someone has to pay me if they want serious numbers, if I were making a bet I’d go with:

    (1a) Local Instrument placing bias 0.1ºC,
    (1b) Upjustments by “scientists” 0.1C (cherry picking of the warmist stations, etc.)
    (1c) Larger scale urban heating bias 0.1C
    (2) Natural Cycles 0.3ºC,
    (3) AGW 0.1ºC
    (4) reduction in global dimming 0.1C

    Thanks Mike for your thoughtful and detailed comment on your estimates. Since my spreadsheet for this exercise only has three categories, I’m in a quandry about what to do with your item (4) . There are two interpretations I can think of: (a) the Sun is becoming less bright as it makes its way along the main sequence that stars follow on their way to becoming Red Giants, or (b) reduction of airborne particles as a result of cleaner energy production and so on has made the atmosphere more transparent and therefore the Sun is brighter. (a), being natural, would subtract 0.1ºC from (2) and (b), caused by human actions, would subtract 0.1ºC from (3). In either case, there would be a deficit of 0.1ºC in accounting for the supposed 0.8ºC temperature rise since 1880. I am pretty sure you intended (b), so I am putting you down for (1)= 0.2ºC [your 1a + 1b +1c – 4] , (2)=0.3ºC, (3)=0.1ºC. That all adds up to 0.6ºC, which is shy of the claimed 0.8ºC net warming since 1880.

    Please comment again if I misinterpreted your intentions.

  69. David says:
    January 17, 2011 at 4:52 am (Edit)
    steven mosher says:
    January 16, 2011 at 9:55 pm
    Ira,

    Your 1999 graphic is still wrong. I have no idea where you got it but it is nothing like what Hansen Published in 1999. I think you might want to check your sources.

    Mr Mosher, it looks very similar to this one.

    http://climateaudit.org/2007/08/11/lights-out-upstairs/

    THANKS David. I traced your link back to GISS where I found the original from the horse’s mouth. http://www.giss.nasa.gov/research/briefs/hansen_07/fig1x.gif. Looks like identical data except for some minor relocation of annotation.

    Note that the version I found at GISS today is apparently from a Hansen 2007 paper. It seems they replotted the 1999 data using the newer annotation on the graph axes. Therefore, I believe that the image I used, from zapruder.nl, is a more contemporary version of what GISS actually published in 1999.

  70. steven mosher says:
    January 16, 2011 at 7:32 pm

    “measure radiances in various wavelength bands, which must then be mathematically inverted to obtain indirect inferences of temperature…” Is it this processed data that you and Jones and Hansen say may have a bias of from 0.0ºC to 0.06ºC to 0.1ºC ?

    ##############

    No. let me be clearer for you.

    The UHI bias in the land record can be estimated in several ways.

    1. Compare Urban sites to Rural sites. These comparisons yield answers from
    0C to .06C to .1C.

    So if you choose ten rural urban pairs, seperate them into rural and urban, obtain the trend for each group, then compare the two groups, you say the trends will be almost identicle, 0.05 to 0.1.

    I think this is utter rubbish.

    http://scialert.net/fulltext/?doi=rjes.2011.1.21&org=10

    http://scialert.net/fulltext/?doi=rjes.2011.1.21&org=10

  71. In a Monty Python voice as in, Nobody mentioned the Spanish Inquisition:

    “Nobody mentioned the 1998 temperatures!
    We have two data sets! One that confuses you , one that amazes you and one that…
    We have three data sets….”

    Must be a certain mantra to some folk…

  72. When it comes to warming biases, you also have to consider the “march of the thermometers”, the use of airport instruments that round up to the nearest whole degree, and the substitution for missing data/stations from stations up to 1500 km away and generally at lower altitudes/latitudes. If you accumulate Chiefio’s warming biases they are enough to account for more than the total observed warming. My guess is that the warming from ca 1910-1944 was mostly real, and thet the recent real warming gets us back close to but not above the 1938-1944 peak. If we take 1910-1944 as plus 0.45 degrees, 1944 – 1976 as minus 0.25 degrees, and 1976-2006 as plus 0.6 degrees per the surface instrument estimates, Iwould guess that the cooling to 1976 is understated somewhat and the warming since overstated considerably. Total warming, 1910 to 2006 (valley to peak) is probably not more than 0.45 degrees C, of which at least 0.4 degrees is natural. The long term CET trend is 0.3 degrees C/century if memory serves, so the above SWAG would be fairly consistent. We are still on the upside of the long ca 1000 year cycle, so probably have a couple of more peaks to go, without any AGW effect. However, in the short run, severe cooling to ca 1935 is the most likely direction.

  73. Roger says:
    January 17, 2011 at 9:56 am
    Can anyone explain why the SATELLITE data also shows it is the warmest in 2010?
    Are you saying Roy Spencer and John Christy are also lying.

    UAH

    here is the data base:

    http://vortex.nsstc.uah.edu/data/msu/t2lt/uahncdc.lt

    ______________________________________________
    Roger is doesn’t, 1998 is warmer.

    1998 0.428c
    2010 0.414c

  74. There are almost no faults/errors in modern weather stations (including the environment where based) that cause temperatures to be cooler and need to be adjusted up. Virtually all faults/errors in weather stations cause a warm bias, yet amazingly the high majority adjustments to the modern stations go up. If there is no errors to the raw data then leave it alone, not increase the damn temperature all the time when in fact any adjustments needed to modern instruments should be down.

  75. For those who think the changes are not ongoing. Just before Copenhagen, the US temps changed to bring 2006 onto play from about 5th. The format of the figD.txt means it is not archived on wayback & similar. 1934 also dropped back at the same time.

    DaveE.

  76. Bill Illis says:
    January 16, 2011 at 5:38 pm

    “The difference in the trends between the satellite measurements and GISS, Hadcrut3 and NCDC indicate the data bias is zero.”

    Inasmuch as the satellite measurements go back only to 1979, that’s a leap of faith that a careful analyst would not make. The problem is that actual measuremets made by thermometers in preceding decades have been wantonly altered into manufactured data with a bogus trend over the ENTIRE station record. Defenders of such “adjustments” seem not understand that nothing in a station’s history can rigorously justify staircase–as opposed to step–changes in the data . Nor are they aware that hourly readings, on which empirical TOBS adjustments are based, tell us nothing reliable about the extremes, which are due to high-frequency components that are aliased by the hoursly series. The notion that 1998 was warmer in the USA than 1934 is groundless.

    Glickstein does make the mistake of thinking that GISS s responsible. The adjusted “raw” data used by GISS comes from NOAA.

    Sorry folks, but I can’t take time this week to debate this scandalous issue.

  77. I would gladly bet my house with any AGW proponent as to whether there will be any significance to any other their doomsday scenarios within 15 years.

  78. Because an old browser cut off Bill Illis’s statement that “the data bias is 0.3C” at the decimal point, I misquoted him in my post a couple of days ago. My apologies for the mistake! The main thrust of my comment about the CENTENNIAL trend being most affected by UHCN staircase adjustments nevertheless stands.

  79. I think that changing the distance for which the radiation is fully 100% absorbed from 10 feet off the ground to 5 feet off the ground (doubling of CO2) has a net effect of 0.0 degrees F +/- .03 degrees F.

Comments are closed.