Epic Fail : Hansen’s 2010’s Forecast From 1988

Yes, Hansen actually made these forecasts (degrees C) in the year 1988. We should be baking already, and Manhattan should be drowning.

The standard response from our friends is that “well … ummm … we know a lot more now than we did 30 years ago.”

I agree, which is why we should be clever enough to realize that they will be saying the same thing five years from now about about 2010 forecasts.

So instead of making clueless forecasts about Arctic storm frequency in the year 2100, perhaps we should be having an honest discussion about the fundamental lack of skill in climate forecasting – and an honest discussion about the fact that all these climate forecasts being passed on to the press have no legitimate scientific basis.

Below are the “actual” (GISS adjusted) trends since 1988, with colour scales shifted to approximately match the forecast.

About stevengoddard

Just having fun
This entry was posted in Uncategorized. Bookmark the permalink.

84 Responses to Epic Fail : Hansen’s 2010’s Forecast From 1988

  1. R. de Haan says:

    You can rest your case.

  2. M Carpenter says:

    I totally agree Steven! You’ve hit the nail on the head.

  3. MikeTheDenier says:

    My forecast for the climate 100 years from now. It will by somewhat different from today. In between today and 100 years in the future? Some periods MAY be cooler, some MAY be warmer.

  4. ChrisD says:

    Steve, it is your considered belief that the decadal means for the 2010s should have been achieved before the end of the first year of the decade?

    And, Steve, what were these changes relative to? Were these relative to the actual 1988 temperatures, or to something else?

    • Ya … I think that after La Nina hits with full force, temperatures will shoot up 300% more in the next 24 months than they did in the last 120 years combined.

      Because that is what would have to happen.

    • I also expect Manhattan to drown before 2008.

    • ChrisD says:

      What are those increases relative to, Steve?

    • ChrisD says:

      And, by the way, there are nine years left in the decade. So why do you say “24 months”? Does the decade end in October, 2012?

      • So when he said 2010’s he actually meant 2020. I get it.

      • Dikran Marsupial says:

        No, he meant an average for the decade. As his projections had temperatures increasing with time you might expect 2015 to match the mean for the 2010’s.

      • ChrisD says:

        It’s a decadal average Steve. Do you know what that means?

        And I’m still waiting for you to tell us what those numbers are relative to. When it says that South Mudflap will be 2C warmer, what does it mean? 2C warmer than what?

      • On average, Manhattan will be underwater this decade.

        2001-2010 had no warming, so we can reasonably expect 2011-2020 to warm up about 2-3C.

      • Dikran Marsupial says:

        The point remains, if someone makes a decadal prediction, you need to wait until the end of the decade to demonstrate that it was wrong, not the first year.

      • Paul H says:

        So are you still holding to Hansen’s forecast?

      • Paul H says:


        This is like saying a horse could still win a race when it has broken its leg halfway round the course and the rest of the field are in the last furlong.

      • Dikran Marsupial says:

        The point remains, it is an epic fail only if it is an epic fail at the end of the decade, not at the start, whether you or I personally think it will be a fail is irrelevant. Waiting for the observations to become available is pretty standard in model-data comparisons!

        BTW, it is possible that the prediction in question is for the most extreme emission scenario. If actual emissions don’t match that scenario, then it isn’t an epic fail if the projection is wrong. It would be if the projection for the best matching scenario were wrong. Steve, what emissions scenario does that projection assume?

      • ChrisD says:

        Paul, it is just a bit too early to say the prediction for a decade is an epic fail when more than 92% of the decade is still in the future. Sorry.

      • ChrisD says:

        Dikram, it would’ve been nice if Steve had labeled it, but just by eyeball it appears to be Scenario B (page 9348, left column, bottom).

      • ChrisD says:

        Typo: Dikram s/b Dikran, sorry.

      • Dikran Marsupial says:

        No worries, it isn’t my real name anyway! ;o)

    • ChrisD says:

      I’m still waiting for one of you to answer “2C warmer than what?”

      • ChrisD says:

        And I might add, this is quite significant because the correct answer means that Steve’s “actual” map is very, very wrong.

      • Dikran Marsupial says:

        The caption of the figure clearly states that the differences are relative to the control run, which was a 100 year simulation with GHG levels fixed at 1958 levels (performed to estimate the unforced variability of the climate). So far from 1988 being the correct point for comparison, it should be 1958!

        Perhaps not so much of an epic fail (other than in Steve’s ability to understand the caption of the figure). To Steves credit though, he did pick the most appropriate scenario, even if this wasn’t specified in the article.

      • Plate 2 in the Hansen article shows 1980s as essentially zero deviation, and my comparison starts in 1988.

        Sorry to disappoint you, but the epic fail is real.

      • Dikran Marsupial says:

        The colours on the diagram are so heavily quantised that you can only tell that the white areas are within about 0.5 degrees of zero. That is not essentially zero when the effect in the prediction for most of the globe is only about 1 degree (yellow).

        Again, the 80’s is a decadal mean, so if you use 1988 as a baseline that will be warmer than the decadal mean.

        This means you have biased your comparison against Hansen in two different ways. That sort of thing isn’t remotely acceptable scientific practice. If someone makes a prediction relative to a baseline, it must be evaluated against that same baseline.

      • As usual, you can’t see the forest for the tree.


        It’s Hansen’s map – not mine.

      • ChrisD says:

        Plate 2 in the Hansen article shows 1980s as essentially zero deviation, and my comparison starts in 1988.

        Bzzzzt. Sorry, but thanks for playing.

        The Hansen temps are relative to a 100-year control run using GHGs fixed at the 1958 levels. The map does not show “Temps will be this much hotter than they were on X date”, no matter what X you pick. They are not relative to any historic date.

        It’s pretty obvious that you didn’t know what the map shows when you wrote this post.

      • ChrisD says:

        It’s Hansen’s map – not mine.

        The second one certainly is yours; it’s wrong; and it’s the entire basis for your argument in this post. Without the second map, there’s nothing in the post that shows anything at all.

      • It is pretty obvious that you have no interest in an honest evaluation of Hansen’s forecasts, and that you haven’t read my other articles about his 1988 paper.


      • ChrisD says:

        It is pretty obvious that you have no interest in an honest evaluation of Hansen’s forecast

        Actually, that’s what I would say about you.

        You aren’t going to admit that your map showing deltas from 1988 has no relevance even though Hansen’s map isn’t relative to 1988, are you? That the two maps show completely different things?

        Honest evaluations don’t compare apples and oranges and proclaim an “epic fail” because the orange doesn’t look much like the apple.

      • Brendon says:

        ChrisD, this is the part where Steve doesn’t return to answer your question.

        Seems to be a habit of his looking at quite a few of the other threads.

        Self-preservation kicking in I guess.

      • You got that right. You guys will waste all my time if I let you.

      • Dikran Marsupial says:

        Steve wrote: “It’s Hansen’s map – not mine.”

        Absolutely, which is why any analysis should be based on Hansen’s description of what is shown, not yours.

        The projections are relative to a control run based on 1958 conditions. Using a baseline of 1988 is (rather obviously) not representative of the conditions in the control run. If the differences were random, that would be one thing, but they are not, the difference biases the evaluation against Hansen’s projections, possibly by half the effect size. If done deliberately, that sort of thing would be regarded as scientific fraud; I will assume that it was done out of ignorance instead.

        As I have said before, your scientific claims will be strengthened if you pay attention to the criticisms made and attmpt to address them. If you just bluster and refuse to accept there is any problem, you will continue to be a laughing stock in the “warmist” blogs and deservedly so while you continue to provide them with material like this to work with.

      • Brendon says:

        Steve writes “You got that right. You guys will waste all my time if I let you.”.

        Yes, those little details like getting the facts right can be so bothersome!

  5. Douglas DC says:

    Simply put, Hansen is wrong.
    Period. they have no idea what’s going on..
    With La Nina coming on this next year is going to
    be-Interesting times…

  6. Scott says:

    Steve, do you have an image for to-date 2010 temp anomalies (or anomalies for a recent year like 2008 or 2009)? Obviously Hansen was off by a large magnitude, but I wonder if he was also off with location. IIRC, the US has stayed constant or showed slight cooling in recent years, definitely contradicting the image here.


    • I tacked the GISS reported trends on to the bottom. These are adjusted upwards and I don’t particularly trust them.

      • Scott says:

        Was that image there before and I just missed it? Guess I was too interested in reading the comments. 😉

        Interesting that even the “adjusted” values are way off for Antarctica, Siberia, and NW North America. Heck, the whole SH seems to be off by a good measure. About the only areas that seem somewhat correct are northern India and the Arctic (according to GISS anyway…DMI doesn’t show that level of trend to my knowledge).


      • ChrisD says:

        The problem with all this is that Hansen’s predictions are not relative to 1988 temps, so Steve’s map is completely meaningless as a comparison to Hansen’s map.

        Still waiting for any evidence that Steve (or any of you “skeptics”, for that matter) knows what Hansen’s anomalies are relative to.

      • Matt says:

        Chris – from the 1988 paper, it appears Hansen’s anomalies are relative to the 1951-1980 climatology. Based on than, I’d guess his numbers might not be too far off, considering large portions are shown to be between 1-2 °C higher than the 1951-1980 climatology, and the same areas are shown to be 0.5-1°C in the 2nd map relative to 1988 only.

        In any case, that paper is still more scientific than anything I’ve seen Steve ‘The Oracle’ Goddard do.

      • Scott says:

        Eyeballing it from the linked graph below, “global temperature” has only increased by ~0.6 C from the 1951-1980 average. So to reach 1.5 C increase, we’ll need to see ~1.5x the increase we’ve seen in the last 30 years during the next 10 years (thus a rate increase of 4.5x) just to get up to Hansen’s predictions by 2020. And because we’re discussing the decadal average, the rate would have to be faster than that. This is ignoring things like UHI contamination.

        I’m on the fence wrt AGW, but over the top predictions like Hansen’s turn me off and give skeptics very good ammo to use.

        And who called Steve Goddard “The Oracle”?


        Temperature plot link:

      • ChrisD says:

        No, Scott, these anomalies aren’t relative to any historical date. They’re relative to a 100-year control run of the same model with GHGs fixed at 1958 levels. The map does not show what Steve thinks it shows, and his comparison map is of no value.

      • Scott says:

        Ah, but it does…it just needs one added piece of information. And that information would be the top left map in Figure 4, which shows most of the world in the deltaT=0 region for the simulation, thus establishing that decade as essentially equal to the base simulation anomaly. Given that, maybe Steve’s image would be better if it were for the 1980’s average, but it definitely has some value.

        If Steve would just add said image to the post, it would be very relevant.


  7. R. de Haan says:

    Libs know nothing about branding, that’s the real problem (Lol)

  8. James Sexton says:

    Recently, I had a discussion with an alarmist. I pointed to all the missed dire predictions such as the underwater Manhattan, and lamented to silly laws and subsequent increase in the cost of power. The response to the missed predictions? ———- “See, our efforts are already working!”—————- 😐

    • Dikran Marsupial says:

      That is indeed pretty amusing. However, a better approach is to ask him to find justification for those predictions in the IPCC report, and if he can’t, suggest he reflect on what that implies.

      • James Sexton says:

        Yes, well, anyone that believes CO2 reduction efforts have been effective…………well, I’m not sure they’d be able to spell IPCC much less search for a justification. I just picked up my glass and moved down the bar and prayed he wasn’t representative of the general public.

      • Dikran Marsupial says:

        “Yes, well, anyone that believes CO2 reduction efforts have been effective”

        didn’t realise we had given it a serious try!

      • Even if the US cut CO2 emissions by 20%, it wouldn’t make much change to global emissions.

      • James Sexton says:

        No, we haven’t, and no it wouldn’t make any difference.

        Well, assuming our purchase of goods would remain the same, it is likely global CO2 emissions would increase if the U.S. decreased their emissions.

  9. chris y says:

    Now I understand why Hansen and GISS are so intent on extrapolating surface temperatures beyond +70N latitude. The models forecast lots of warming in regions that have poor or non-existent measurement coverage, creating an opportunity to apply climate models to fabricate *measurements* to validate…climate models.

    To borrow a turn of phrase from Michael Crichton, it is a perfect refuge for scoundrels.

  10. It's always Marcia, Marcia says:

    Some are asserting in this thread that manmade co2 level is going down? It has not. It continues on the same linear rise it has for decades.

    Some here commenting should just admit James Hansen is wrong.

    • ChrisD says:

      Some are asserting in this thread that manmade co2 level is going down?

      Not that I’ve noticed.

      Some here commenting should just admit James Hansen is wrong.

      He may or may not be–there’s nothing in this post that offers any evidence either way.

    • It shows that Hansen predicted much more warming than has occurred. Some areas have actually cooled.

      • ChrisD says:

        Hansen’s map is not a prediction of 2010 temps vs. 1988 temps. Your map is. What is the justification for comparing two maps that don’t depict the same thing, and how is this supposed to illuminate anything?

      • harry says:

        This thread sums up exactly why I called you a cretinous (is that a word? If not I’m submitting it to Oxford with a picture of ChrisD, a pic of a baboons ass will do) muppet. ChrisD you are bordering on mental retardation such is your desperation to cling to the shire you’re pedalling. Its embarrassing, I’m close to pitying you, but that means I give a shit, and clearly I don’t.

      • harry says:

        That’s shite by the way, I know you Yanks struggle with Anglo terms!

      • ChrisD says:

        Wow, Harry, that would be a whole lot more convincing if you offered up a single shred of anything that’s actually wrong with my comments. But you didn’t. Oh, well.

  11. Amino says:

    I know, let’s play a game. It’s called Let’s Make Pretend James Hansen is Not Actually Creepy but is Really Smart and His Predictions are Right, if He Even Did Make Predictions—Cause That’s a Make Pretend Part Too. Oh wait…… I see some are already playing.

    • ChrisD says:

      Nobody did any of that, Amino. We’re simply pointing out that Hansen’s map doesn’t show what Steve apparently thought it did, and that the “comparison” map doesn’t show the same thing as Hansen’s map.

      But you’ll defend anything, won’t you? It seriously does not matter to you that the two maps show different things, yet are used as a comparison to show “epic fail”, does it? It seriously does not matter to you that Hansen’s map is not a map of projected warming since 1988, yet is used that way, does it?

      And you guys say that we are the ones with religion.

      You all want to debate the science, fine. Debate it. Show what you think is wrong with it (like Steve does in his latest Hansen post). Nobody has a problem with that. But this is bogus crap. It’s wrong, and you all simply could not care less. That’s just dishonest. If your case were stronger, you wouldn’t have to do this.

  12. NS says:

    I’ve been reading the Hansen paper and it seems like a reasonable study so far.

    “There is no obviousley significant warming trend in either the models or observations for the period 1958-1985.” (Hansen P.9346 Model forecasts….)

    Question for AGW experts:
    1. Why is there no warming during this rapidly industrialising period?
    2. Why 1958 baseline?

    Maybe more to follow….

    • Dikran Marsupial says:

      I suspect the 1958 baseline is due to accurate measurements of background atmospheric CO2 start with Keeling’s Mauna Loa dataset in 1958.

      Not that I would claim to be an “AGW expert”.

    • Dikran Marsupial says:

      I suspect changes in sulphate aerosols are the mainstream explanation for the lack of warming in the 1958-1985 period. The chapter on understanding and attributing climate change in the most recent IPCC WG1 report mentions that “Aerosol forcing appears to have grown rapidly during the period from 1945 to 1980, while greenhouse gas forcing grew more slowly (ref). Global sulphur emissions (and thus sulphate aerosol forcing) appear to have decreased after 1980 (ref)” (starting at the bottom of the second column on page 672). Later on (halfway down the first column of page 674) “Sulphate aerosol forcing results in cooling throughout most of the globe, with greater cooling in the NH due to its higher aerosol loading (fig ref), thereby partially offsetting the greater NH greenhouse-gas induced warming”. The effects of forcings dues to atmospheric constituents are also discusssed in chapter 2, I expect there is a lot more information there. HTH

  13. NS says:

    Got another already!

    Model B states a “1 degree C level of warming” within 25 years of the paper (A is in 20, C is 0.5 in 15 years).
    The map shown above (Hansen’s 2010 prediction over 58-85 b/l) seems to indicate by the use of color a rise in excess of 2 deg C.

    Is this due to feedback effects or just an “unfortunate” use of color?

    • Dikran Marsupial says:

      The colour for most of the globe in the 2010’s for scenario by is white/yellow (less than 1 degree C) and orange (1-2 degrees C). An area weighted average may well be about 1 degree within 25 years.

    • ChrisD says:

      NS, Hansen’s 1988 B forecast shows the 2010 anomaly in global average temperature at about +1.0. The GISTEMP observation is about +0.7. That’s really not too bad.

      Part of the error is that, while Scenario B is the closest of the three to what actually happened, it does overestimate emissions by about 10%. The rest is that he overestimated climate sensitivity. (Of course, he had much less data to work with than we do now.)

      • You cherry-picked the peak monthly anomaly of a strong El Nino?

        Very Hansenesque of you.

      • Dikran Marsupial says:

        He is looking at the same year you were using to declare Hansen’s prediction an “epic fail”, so it was your pick. Ironically enough, this seems to be one of the few articles you have posted recently where the error is NOT cherry picking.

      • ChrisD says:

        You cherry-picked the peak monthly anomaly of a strong El Nino?

        Say what? I’m looking at Hansen’s 1988 forecast for 2010. Same as you.

      • GISS does not provide a way to generate a trend map vs. a baseline period. You are making a straw man argument based on unavailable data for the sole purpose of generating noise.

      • Dikran Marsupial says:

        GISS provide the data on their FTP site, there is nothing to stop you from using some package such as R or MATLAB etc. to generate whatever plot you need to make your argument.

      • ChrisD says:

        The GISTEMP data are readily available. The current 12-month mean anomaly is just a bit under +0.7. That is what I said. I don’t know what your problem with this is.

    • Dikran Marsupial says:

      Actually, checking again, I can see your plot is a difference between 1998 and 2009, so appologies, it wasn’t Steve’s pick. However, picking this year for evaluating a decadal prediction for the 2010s as there is only one year to choose from in that decade, and even that isn’t complete yet.

      • Anyone with the slightest bit of objectivity can see that Hansen’s 1988 forecasts were much too high. Everyone in the climate science community recognizes this. Why are you being so obtuse?

      • Dikran Marsupial says:

        They are a bit high, as has already been pointed out, partly because the best matching emissions scenario overestimated actual emissions and the climate sensitivity of the models was too high. Also the internal variability of the climate means there is a limited accuracy with which predictions can be made. These days this is dealt with by running an ensemble, the actual climate is only expected to lie within the spread of the ensemble. Hansen couldn’t do that because he didn’t have the computeing power we do now. So my view is that Hansens predictions we about o.k., not bad for the state of knowledge at the time, but nothing more than that.

        Nowhere have I said that Hansen’s predictions are not too high. What I have said is that your analysis is biased against Hansen, becuase you chose to use a different (warmer) baseline than he did. Hansens predictions are STILL too warm even if you use the correct baseline, so you don’t even need to bias the comparison to show that Hansens predictions are too high. You would have to bias the comparison though to make them look like an “epic fail”.

        As I have frequently said, if you want your science to be taken seriously, you need to eliminate the bias and the cherry picking and the misrepresentations, but it is up to you.

  14. NS says:

    “A warming of 0.5 deg C per decade implies…….” (doom and disaster).
    Where does he get this figure from?

    When predicting extreme hot summers (same page) he baselines at 1950-1979, and makes comparison & extrapolation on model based values (58-85). Is this good scientifc method?

    “If it (land temps) rises and remains for a few years above an appropriate significance level, which we have argued is about 0.4 deg C for 99% confidence…it will constitute compelling evidence….”
    (summary p.9359)

    This is not compelling evidence in a system with proven 11, 30, etc. year cycles. The terminology used is also un-scientific.

    I read through this paper in an hour and could post all day long on the innaccuracies, wild guesses and statistically invalid extrapolations.

    D- must try harder.

    • Dikran Marsupial says:

      “A warming of 0.5 deg C per decade implies…….” (doom and disaster).

      The full sentence was “A warming of 0.5 deg C per decade implies typically a poleward shift of the isotherms by 50 to 75 km per decade”. He then goes on to say this means there will need to be some adaption in agricultural practices, especially where species are near their maximum temperature tolerance limits. He predicts there may be some coral bleaching. He also points out that there will be some mitigation of this because of the “fertilisation” effect of CO2 (it is apparently plant food).

      The fact you present that as “doom and disaster” is an obvious misrepresentation and the only thing it demonstrates is your bias. I wish I hadn’t bothered looking up the answers to help you with your other questions; I suspect you weren’t that interested in the answers to begin with.

      • NS says:

        “a poleward shift of the isotherms by 50 to 75 km per decade”
        = doom & disaster

        To say that Hansen’s predictions do not equate to doom and disaster is in fact a complete mis-representation.

        And furthermore if I were completely misguided / unable to understand English, and you’re right (that he does not say it represents doom and disaster) then the case for massive mitigation efforts is completely undermined.

      • Dikran Marsupial says:

        “a poleward shift of the isotherms by 50 to 75 km per decade”
        = doom & disaster”

        Rubbish, it just means that it is on average as warm at a particular lattitude at the end of the decade 50-75km poleward of where it was at the start. Big deal, it just means that where there are crops at there temperature thresholds you have to start farming them 50-75 km further northwards OR plant different crops that are not at their temperature threshold.

        It also means that land in the north that was previously too cold for some crops can now support them.

        So a bit of adaption in agricultiural practices will be required. If you think that is “doom and disaster” then it is you that is the “alarmist” rather than Hansen.

  15. Jose Suro says:

    Although Steve is correct, the discussion that followed is kind of pointless, focusing on tenths of degrees differences, rather than on Hansen’s conclusions. Hansen’s main point was not that temperatures would go up but that the pseudo-predicted consequences “should”, “could” be catastrophic. Either way, the catastrophes have not materialized.

    I read both the paper published in 1988 and his presentation to Congress in the same year. IMHO, the most damaging part in Hansen’s documents is the graph presented in both his paper and presentation to Congress. The pages marked 9347 and 48 respectively therein. The presentation to Congress is more telling because it was addressed to policy makers, not scientists. The PDF of the presentation to Congress is here (Figure 3, Page 48) and has a more complete footnote:

    Click to access ClimateChangeHearing1988.pdf

    The graph includes three scenarios: a,b and c. You can look at the year 2010 and figure out the rest yourselves. And yes, that is the year 2010, not the decade.

  16. Pingback: Global Warming Hoax Weekly Round-Up, Oct. 21st 2010 « The Daily Bayonet

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s