Global warming predictions spot-on

Climate scientists have been getting a lot of flak. Something I repeatedly read is that their prediction of temperature rises have been wildly inaccurate and so therefore we shouldn’t believe anything they say. This is simply not true. One forecast made in 1999 by Oxford physicist, Myles Allen, has proved to be impressively spot-on. WottsUpWithThatBlog has already made a good post about this at Tests of a climate forecast, so I won’t go into too much detail here other than to show the forecast.


The climate forecast is the dashed black line. The red line represents actual temperatures.


A paper published this year, Test of a decadal climate forecast, which tests the accuracy of this forecast comes to the same conclusion:

 Early climate forecasts are often claimed to have overestimated recent warming. However, their evaluation is challenging for two reasons. First, only a small number of independent forecasts have been made. And second, an independent test of a forecast of the decadal response to external climate forcing requires observations taken over at least one and a half decades from the last observations used to make the forecast, because internally generated climate fluctuations can persist for several years. Here we assess one of the first probabilistic climate forecasts with a full uncertainty assessment that was based on climate models and data up to 1996. Using observations of global temperature over the ensuing 16 years, we find that the original forecast is performing significantly better than a hypothetical alternative based on the assumption that decade-to-decade temperature fluctuations consist of a random walk, that is, a sequence of random fluctuations with no externally driven warming trend. The original climate forecast also outperforms a very simple interpretation of the climate models used for the latest Assessment of the Intergovernmental Panel on Climate Change (IPCC), supporting the conclusions of previous assessments that the spread of such an ensemble is not, on its own, an adequate measure of forecast uncertainty.

85 thoughts on “Global warming predictions spot-on

  1. Thanks for this Rachael. We’ve heard much from the denier community about the so called ‘end of warming.’ Too much and too devoid of facts for any reasonable person to agree with.

    1. Indeed, but science isn’t about everyone being right. It’s about a process in which we gain more and more understanding of how things work. That Myles Allen’s & Co. model does indeed appear to match well the global temperatures for the last decades does mean that the claim that climate models are all wrong is incorrect.

    2. Yes, I agree with Wotts. The blogs written by people who do not endorse AGW constantly tell us that *all* the climate models are wrong and all the predictions are wrong. This example clearly shows us that it is their claim that is wrong. I think it also illustrates the bias on those blogs.

  2. Rachel,
    Go to and read 2 posts,” EPIC FAIL:73 Climate Models vs. Observations for Tropical Tropospheric Temperature “, and ” STILL Epic Fail: 73 Climate Models vs. Measurements running 5 Year Means “.
    As you and Watts appear to understand, the 73 CMIP5 climate models currently in prospect for AR5 ,and used by the UN IPCC, are completely unarguably and utterly wrong in their predictions.
    To quote Dr. Roy Spencer,” ….the day of reckoning has arrived.The modellers and the IPCC have willingly ignored the evidence for low climate sensitivity for many years …..The discrepancy between models and observation is not a new issue ……just one that is becoming more glaring over time.”
    ” It will be interesting to see how all of this plays out in the coming years. I frankly don’t see how the IPCC can keep claiming that the models are “not inconsistent with ” the observations. Any sane person can see otherwise.”
    The 1990 FAR predictions of the UNIPCC were wrong .They predicted that if human emissions stayed the same,temperatures would rise by 0.3 C per decade.Even by the most generous interpretation of the data the highest rate they can find is 0.18 C per decade ( probably overestimated), and in any case below the lowest estimate of 0.2C per decade.( highest 0.5 C per decade).

    1. Let’s see. I have read those posts by Roy Spencer. Firstly he shows the mean of 73 models using the most extreme possible forcing scenario RCP 8.5. Also, the satellite observations that he shows are the averages of two different datasets (UAH and RSS) which differ quite substantially. In fact, the RSS data has a trend that is 3 times greater than the UAH trend. So basically he has chosen the most extreme forcing for the models and averaged two datasets that differ, to try and show that the models have failed. If he had chosen a different forcing (which may be more appropriate) and compared it with the RSS dataset only, many models would have compared well. Roy Spencer’s post hasn’t shown that the models have failed. His post has shown that you can make it appear as though the models have failed if you persevere and try really hard. If at first you don’t succeed, try, try again.

  3. Rachel ,
    While you are on Roy Spencer’s site you might notice the Global Microwave SST update for May 2013, at minus 0.01 C .Also the UAH Global Temperature update for May 2013. The temperature anomaly is +0.07C.Not much global warming there.

  4. Wotts,
    No.Not from a single month but over time the temperature anomalies will show a trend ,or no trend .
    As with Roy Spencer showing the temperature deviation by way of an anomaly from a baseline of 1981-2010, there are a number of such temperature the various temperature records.There seems considerable difficulty in giving an accurate lower troposphere average surface temperature ,without some margin of error.Let’s mention one suggested in the David Rose article in the Daily Mail last October-
    January 1997 – 14.37C GAST
    August 2012- 14.45C. GAST
    Warming – 0.08C -statistically insignificant if the “error margin ” is say 0.16C.
    (You may want to say I am cherry picking the start date etc.)
    A good example of the issue is the Global Warming Challenge on
    In 2007, University of Pennsylvania Professor of Marketing J.Scott Armstrong sought to place a bet with Al Gore .He claimed that having spent decades studying the science of forecasting Armstrong was unable to find a single scientific forecast to support the claim that the Earth was becoming dangerously warmer or colder.He claimed some scientists were using improper forecasting methods to forecast.
    He contacted Gore and suggested they co-operate in a validation test of dangerous global warming forecasts.He suggested a ten year bet for which he would forecast no long term trend at all in climate while Mr. Gore could chose forecasts from any climate model.Gore declined but Armstrong pursued the” bet.”
    As shown on Professor Armstrong’s forecasts have been more accurate and the latest global temperature anomaly is little different from the beginning of the “bet,”in 2007.I don’t know if any statistically significant warming or cooling can be discerned from the forrest of figures.
    Incidentally what is the GAST currently with margin of error ?

    1. Firstly, I think simply looking at the anomalies for two months separated by some time interval is meaningless unless you know something about the trend. You could easily be comparing a warm month in the past with a cool month now. The time period considered by David Rose is 15 years. The difference is 0.08 degrees celsius. Not inconsiderable. We’re concerned about rises of between 0.1 and 0.2 degrees C per decade. Plus, if you consider the various datasets for the period 1997 to 2012 the trends are between 0.06 and 0.09 degrees C per decade with errors around 0.15 degrees C per decade. So maybe there’s be no rise in surface temperature but maybe it’s been considerable. We really don’t know yet.

      The other issue is this issue of statistical significance. People have interpreted this as suggesting that because the error in the trend suggests that the trend could be negative that we can conclude that there has been no warming. This is simply incorrect. Let me see if I can explain why I think this.

      Imagine we consider some medical trial in which we’re testing a new drug. We already know what would happen is nothing was done to a group of patients with a particular condition (this could be the null hypothesis). We run a test with this new drug and discover that the results are statistically consistent with the null hypothesis. Hence we reject the hypothesis that this drug has some positive effect on the patients in favour of the null (i.e., it does nothing). This could just be a sample size problem and maybe a bigger sample would give a different result. We won’t know until we do the bigger trial. In the case of global surface temperature trends, if you want to argue that the trend is flat, you need some reason why (scientifically) a flat trend is the correct null hypothesis. There isn’t one as far as I know. In fact, given that solar insolation is dropping, the correct null might be that the trend is negative (which is not really consistent with observations). Furthermore, we have models that suggest that the trend should be about 0.1 degrees C per decade which is consistent with the observations, so why shouldn’t we just accept that as the correct trend. Just because the error is large, doesn’t immediately mean that the trend is flat. Statistically a trend of 0.2 degrees C per decade (for the period 1997 to 2012) is as likely as a flat trend.

      As far as the bet is concerned, I don’t know much about that. One thing I will say is that using surface temperatures alone to determine if global warming is happening is also not correct. If you only consider global surface temperatures only then, strictly speaking, you’re only really determining if global surface temperatures are rising or not. You’re not really determining if global warming is happening or not. To do that you need to consider all the possible datasets that measure changes in energy in the climate system.

  5. CO2 increase from 1800 to 2001 was 89.5 ppmv (parts per million by volume). The atmospheric carbon dioxide level has now increased since 2001 by 25.46 ppmv (an amount equal to 28.4% of the increase that took place from 1800 to 2001) (1800, 281.6 ppmv; 2001, 371.13 ppmv; May, 2013, 396.59 ppmv).

    The average global temperature trend since 2001 is flat.

    Rationalize that the temperature increase to 2001 was caused by a CO2 increase of 89.5 ppmv but that 25.46 ppmv additional CO2 increase had no effect on the average global temperature trend after 2001.

    A simple equation at calculates average global temperatures since they have been accurately measured world wide (about 1895) with an accuracy of 90%, irrespective of whether the influence of CO2 is included or not. The equation uses a single external forcing, a proxy that is the time-integral of sunspot numbers. A graph is included which shows the calculated temperature anomaly trajectory overlaid on measurements.

    1. I’ll be honest and say that I haven’t looked through you “model” in any detail (I shall try and do so) but it does appear to simply be a fit to the temperature anomaly data using an equation with a number of tunable factors. Without the physics explaining why these number of reasonable, this has very little actual value (IMO at least). I will say one other thing about this. Surface temperatures have increased by about 1 degree celsius since about 1800. This means that the energy flux from the surface has increased by about 5.5 Wm^-2. We still have a measured energy excess at the top of the atmosphere of 0.5 Wm^-2. If we were in equilibrium sometime around 1800, then that requires that the total forcings plus feedback must be about 6 Wm^-2. The change in solar flux at the top of the atmosphere since about 1800 is estimated to be 0.25 Wm^2. If you want to invoke solar effects, then this requires a feedback factor of about 20 which is extremely difficult to explain and also extremely difficult to explain how such a large feedback factor has allowed our climate to be so stable over long timescales (i.e., if small changes to solar forcing can lead to feedback factors of 20 or so, paleo records should show much greater variability.).

      As far as your CO^2 statement is concerned. CO2 forcing is estimated to be the following – Delta F = 5.35 ln(C/Co). If you consider 1800 – 2001 then this gives something like Delta F = 5.35 ln(370/280) = 1.49. The Transient climate response is estimate to be about 1.8 degree C for a change in forcing of 3.7 Wm^-2, therefore the change in forcing from 1800 – 2001 should have produced a temperature change of about 0.7 degrees C – which is about right. If we consider the period 2001 – 2012 then Delta F = 5.35 ln(395/370) = 0.34 Wm^-2. Converting to Delta T gives about 0.17 degrees C. The trend for the period 2001 – 2012 may be flat, but the errors are large, so not inconsistent with this estimate. You have to bear in mind that the temperature change due to changes in CO2 depends on the ratio of the CO2 concentrations, not the absolute change.

      One other thing to bear in mind is that fundamentally the increase in CO2 produces a top-of-the atmosphere energy imbalance which is the Delta F I’ve been computing above. The measured energy imbalance is consistent with what we expect from CO2 forcing and not consistent with what we’d expect from ocean cycles (i.e., none) or from changes to solar flux (much smaller than currently estimated).

      1. Wotts – It is not a model in the usual meaning of the term. It is an equation that was set up using conservation of energy and the hypothesis that the first derivative of entering energy is proportional to sunspot numbers. All that is left for anyone to do is arithmetic, or just look at the graph that the equation produces. The high accuracy (R2=0.9) demonstrates that the hypothesis is valid.

        As to ‘tunable’ factors, it depends on what is meant by tunable. They are described in some detail in the paper. The factor C can be set to zero with no significant effect on accuracy and D is equivalent to using a different reference temperature for the anomalies. B is a proxy factor for the sunspot numbers and A accounts for what is left; the natural ocean oscillations of which the PDO is a major contributor. Thus it boils down to only one independent ‘tunable’ factor, B.

        Most people can’t seem to get past thinking that the metric for solar influence is TSI. Calculations that show the high sensitivity of average global temperature to low-altitude cloud area change as shown at might help. A theory described there connects sunspot number to low-altitude cloud area.

        The predictive ability of the equation can be tested right now. For example, I used the temperature anomalies and CO2 measurements to 1995 to set the coefficients and then used the equation, as calibrated by measurements to 1995, to predict the anomaly trend in 2012. It predicted the 2012 anomaly trend within 0.03K.

        The use of trends instead of yearly average temperatures is mandated by the random uncertainty in the reported average annual anomaly values which have equivalent standard deviation of approximately ±0.1K. This is described further in

        I demonstrated that noncondensing greenhouse gases had no significant influence on average global temperature in which was made public in March, 2008.

      2. Dan, a quick question. How does your formalism/equation conserve energy. As far as I can tell, it probably doesn’t (or at least doesn’t really deal explicitly with energy) but maybe you can correct me.

    2. Thanks, Dan, for an easy to follow interpretation of scientific data that isn’t overloaded with mathematical symbols designed to confuse the average reader.

  6. Rachel,
    Professor Freeman Dyson was the successor to Einstein at Princeton University,and here is his take on climate models-
    “My first heresy says that all the fuss about Global Warming is grossly exaggerated.Here I am opposing the wholly brotherhood of climate model experts and the crowd of deluded citizens who believe the numbers predicted by the computer models.Of course they say I have no degree in meteorology and I am therefore not qualified to speak.But I have studied the climate models and I know what they can do.The models solve the equations of fluid dynamics,and they do a very good job of describing the fluid motions of the atmosphere and the oceans .They do a very poor job of describing the clouds ,the dust , the chemistry and the biology of fields and farms and forests.They do not begin to describe the real world that we live in.The real world is muddy and messy and full of things that we do not yet understand.It is much easier for a scientist to sit in an air conditioned building and run computer models than to put on winter clothes and measure what is really happening outside in the swamps and the clouds.That is why the climate model experts end up believing their own models .
    There is no doubt that parts of the world are getting warmer,but the warming is not global.I am not saying that the warming does not cause problems.Obviously, it does.Obviously we should be trying to understand it better.I am saying that the problems are grossly exaggerated.They take away money and attention from other problems that are more urgent and more important ,such as poverty and infectious diseases and public education and public health,and the preservation of living creatures on land and in the oceans, not to mention easy problems such as the timely construction of adequate dykes around the city of New Orleans.”
    From “A Many Coloured Glass : Reflections on the place of Life in the Universe (Page-Barbour Lectures).

    1. There’s a good interview with Dyson at Yale Environment 360 –

      His basic argument seems to be that since CO2 is good for plants and so a warmer planet is a good thing. I’ve already written about the claim that CO2 is good for plants in another post –

      There’s one point he raises that is certainly worth discussing and that is that the people of Greenland like their new warmth. I can’t verify this so can’t comment, but I do think it’s a question worth asking. I have a Finnish friend who tells me that all her friends and family back in Finland have been enjoying the unprecedented warmth this summer. The impact of global warming will be different for different places and some of it may be good. But it’s my guess that many more people will be adversely affected than will benefit simply because there are more people living in the tropics and subtropics than in Greenland and Finland combined. The Sami people who were interviewed in the Thin Ice movie were not happy with the changes though (

  7. Rachel,
    I know you and Wottsupwiththatblog dislike WUWT but there are 2 new posts on the ” failure ” of models.See “Model – data Comparison : Australian Land Surface Temperatures & Anomalies – not so ” angry ” after all, ” And ” Policy Implications of Models on the verge of failure”.
    The picture emerging is that models repeatedly err on the upside of temperature increases, often by significant factors.

    1. Douglas, I haven’t really had a chance to read the Australian Land Surface post, although I believe Sou at hotWhopper has addressed some of the claims here. The basic claim of the WUWT post seems to be that how can it have been an angry summer if no single site broke a record. This appears to be because in the past individual sites may have broken records, but at no single time did almost all sites have temperatures close to the record. This appears to have been what happened last year. Even though no single site broke a record, the average for the country as a whole was higher than it has ever been.

      As far as the Policy post is concerned, it is a rather poor piece of work. They compare the measured mean trend over time periods up to 15 years with model trends. Firstly, they ignore errors in the measured trends. If you do so, the comparison is pretty good. Secondly, the time interval chosen is heavily influenced by the large ENSO event in the late 90s that has made the trend shallower than it has been in the past. Thirdly, if you simply consider a 20 year interval, the measured trend is very similar to the model trends. And finally, even if you ignore all the above you still can’t reject the models statistically because they are still consistent with the measured trends at the 10% level.

    2. Doug, there’s a good post at OpenMind – – which directly addresses the stuff at WUWT regarding Australia’s angry summer.

      But you can also go to the Bureau of Meteorology yourself and check out the data for last summer, which was a scorcher and the hottest on record –

      As for the post about policy implications, may I suggest you read for a good analysis of this.

  8. Wotts – Conservation of energy can be written:

    Change in stored energy = energy ‘in’ minus energy ‘out’

    Average global temperature change is proportional to energy change through the effective thermal capacitance.

    Energy ‘in’ was hypothesized to be proportional to the sunspot time-integral. The hypothesis is verified by the accuracy of the final equation.

    Energy ‘out’ is proportional to the time-integral of the fourth power of the average global absolute temperature (SB equation).

    Whether atmospheric carbon dioxide change increases or decreases average global temperature depends on what the sign of C turns out to be. The influence of atmospheric carbon dioxide was found to be somewhere between zero and insignificantly positive. Since any influence of CO2 would be on energy rate, it also needs to be divided by the effective thermal capacitance.

    1. You see, I think that’s where your model is incorrect. You’re fitting to the temperature anomaly data and assuming that because your equation fits the temperature anomaly data well that you’ve conserved energy. I suspect you’re going to disagree with me (because that is what typically happens in these discussions) but, as far as I can see, it is simply a curve fitting exercise. It has no physics in it. One reason I can be fairly confident that you’re not conserving energy is that the amount of excess energy entering the climate system per year has remained reasonably constant for the past few decades (about 8 x 10^21 J). The fraction going into the oceans, however, appears to change with time and so one reason that the surface temperatures have not risen for the last decade or so is that most of the energy has been going into the oceans and very little has been heating the land and atmosphere (I guess we don’t have a particularly good idea why this has happened, but we do have measurements that suggest that this is the case). You model completely ignores the energy going into the oceans, so I really can’t see how it is actually conserving energy.

      I guess we’ll know fairly soon if your model is correct as you’re predicting a downturn in global surface temperatures and climate models are predicting a continued rise.

      1. Wotts – I repeat, it’s not a model, it’s an equation.

        The equation was formed using physics as described in my previous post. It uses the conservation of energy equation so it absolutely conserves energy. It implicitly accounts for ALL of the energy ‘in’ and ‘out’ of the planet. The results are calculations. You just plug in the numbers for AGT and sunspot numbers that have been reported. Perhaps you are unfamiliar with the use of the coefficient of determination when applied to the general case where the line is not straight. Wikipedia has a pretty good write up. The formula that I use to compare the equation with the measurements comes from there. It’s

        R2 = 1 – SSres/SStot

        You assert that the equation is incorrect. Certainly you accept the conservation of energy and the SB parts. That leaves only the time-integral of sunspot numbers. I suspect that, instead of explaining the measurements, possibly because of preconceived perceptions, you think that the very high correlation is just a coincidence. Partly because of corroborating studies, I think that coincidence has extremely low probability.

        One corroborating study is described at . This shows a trajectory based on the sunspot time-integral beginning in 1610. The decline of the LIA and rise since approximately 1941 are evident. The close match of the calculations to measurements results when the ocean net oscillation is added.

        It’s during the last decade (actually about 12 years) that the average global temperature (AGT) trend has been flat and, since energy change shows up as AGT change, there hasn’t been any significant energy change for about 12 years. Between about 1973 and 2005 the large AGT increase is consistent with a large energy gain.

        The main influence that the sunspot time-integral has is the effect that it has on the quantity of low-altitude clouds (according to Svensmark). AGT is very sensitive to changes to low-altitude cloud area as shown at (which contains a link to the Svensmark paper). Thus insolation could be constant but AGT could change from increasing before 2001 to constant after 2001 because of increase of low-altitude clouds which increases both albedo and the rate that energy is radiated from the planet combined with temperature decline from net ocean temperature oscillation.

        Most of the rational reports that I have seen show about 93% of the energy going in to the ocean (one place in Wikipedia shows 93.4%). That passes the sanity test for me. One reference (Schwartz) found 14/17 = 82.3% but with a lot of uncertainty. We do know why this is so. Oceans cover 71% of the planet and they have an effective mixing layer depth of about 110 meters.

        I see no reason for the fraction of energy going in to the ocean to change significantly. I see no reason for the energy to suddenly start showing up below 700 meters. How would the energy get past the surface? I put this issue on my watch list. Reported AGTs include the surface or near surface temperatures of both land and ocean.

        Maybe you missed this “The predictive ability of the equation can be tested right now. For example, I used the temperature anomalies and CO2 measurements to 1995 to set the coefficients and then used the equation, as calibrated by measurements to 1995, to predict the anomaly trend in 2012. It predicted the 2012 anomaly trend within 0.03 K.”

        I ran a couple more cases: When calibrated to either 2001 or 2006 it predicted the 2012 trend temperature within 0.04 K. Thus the equation has already predicted the trend 6 years, 11 years and 17 years into the future with an accuracy of about 0.04 K. For the graph, the equation was calibrated to 2012 and projects to 2020 and then to 2037 depending on what the sun does. How many more years is “fairly soon”? My prediction for the trend is the graph.

      2. Hi Dan,
        I’ve heard about Svensmark’s ideas about cloud cover and cosmic rays and while they are interesting, from my understanding they have been investigated by climate scientists and the correlation breaks down after 1991/1992. There’s a good paper about this in the journal of Astronomy and Geophysics called The influence of cosmic rays on terrestrial clouds and global warming. Here’s a graph from the paper which shows the divergence between cloud cover and cosmic-ray flux from about 1991:

        It is my understanding that a climate model is a quantitative simulation of the interactions between various elements in the climate system like atmosphere, ocean, land and ice. Even a very simple equation which models incoming radiation from the sun and out-going infra-red is a climate model.

      3. Dan, I realise that it’s an equation, not a model. Correct me if I’m wrong, but you’re getting your coefficients by fitting to the known temperature anomaly data. So, it is simply a fit to the data? You say that “it implicitly accounts for ALL the energy in and out of the planet”. No it doesn’t. It simply accounts for the energy associated with surface temperatures. You’re not considering ocean heat content, for example. I’m not suggesting that your equation doesn’t fit the data. I’m simply pointing out that it is simply a fit to the data. It may have some predictive power. Judging by your figure, we’ll know by 2020 if your equation has some predictive value.

  9. Rachel – Whether all cosmic rays (they come in a wide range of energy levels) have been adequately assessed or are even involved is arguable. The connection between the sunspot time-integral and average global temperature has been demonstrated with arithmetic. Two sites that do this are linked above.

    Wott – You’re wrong. And apparently you are not technologically knowledgeable enough to recognize why you are wrong.

    Adding to my June 28 post, the CO2 level in 2020 will be about 405. So you believe that a further increase of 8 ppmv will cause the uptrend to restart. I monitor all five reporting agencies, so if that happens, I will know about it. However, as I said in a previous post, I discovered over five years ago that noncondensing ghg have no significant influence on AGT so I don’t expect it. The current measured temperature trend from 2001 is down.

    1. Dan,

      While I often don’t agree with Doug’s comments, there is one thing he said a little while ago which I do agree with. It was something along the lines of: the first person to call the other stupid (or in your words, “not technologically knowledgable enough) loses the argument. That’s you. So before you call me stupid as well, I will say it first. Yes, I’m stupid, so perhaps you won’t mind clarifying something for me.

      You are saying that the current divergence between total cloud cover and cosmic flux is because our observations are wrong or inadequate? How then, do you explain the good correlation prior to 1991? Surely our measurements since then have improved?

      As far as your prediction goes, according to Oxford physicist Myles Allen, you will need to wait at least 15 years from the last observation used to make the forecast to know how good it is. Here’s what he says:

      “And second, an independent test of a forecast of the decadal response to external climate forcing requires observations taken over at least one and a half decades from the last observations used to make the forecast, because internally generated climate fluctuations can persist for several years.”

      1. Rachel – There is a huge difference between ‘not technologically knowledgeable enough’ and ‘stupid’. Not everyone has been a student of technology or may not have a talent for it. They may be very knowledgeable in other areas or they may simply not have gone very far down the path of learning yet.

        If the talent is there, ‘not technologically knowledgeable enough’ can be fixed with learning. Barring a miracle, a person is stuck with ‘stupid’ which is usually described as difficulty in learning. But even this can be in a limited area, i.e. dyscalculia.

        As to cosmic rays, I am not knowledgeable enough on cosmic rays to be confident that they have, or ever had, any significant influence.

        Because of the approach that I used and other corroborating evidence, I have high confidence that the sunspot-time integral is the sole significant external forcing on average global temperature (AGT) and change to the level of atmospheric carbon dioxide has no significant influence. I know that AGT is sensitive to changes in low-altitude cloud area .

        Svensmark showed a correlation between sunspots and low-altitude clouds but I have heard rumblings that some have had trouble verifying his findings. In my perception, the sunspot related solar magnetic shielding of cosmic rays influencing the quantity of low-altitude clouds is likely but not yet well understood and not proven. But knowledge of cosmic rays is not required to discover the connection between the sunspot time-integral and AGT.

        Maybe you missed this response to Motts “The predictive ability of the equation can be tested right now. For example, I used the temperature anomalies and CO2 measurements to 1995 to set the coefficients and then used the equation, as calibrated by measurements to 1995, to predict the anomaly trend in 2012. It predicted the 2012 anomaly trend within 0.03 K.”

        I ran a couple more cases: When calibrated to either 2001 or 2006 it predicted the 2012 trend temperature within 0.04 K. Thus the equation has already predicted the trend 6 years, 11 years and 17 years into the future with an accuracy of about 0.04 K. For the graph, the equation was calibrated to 2012 and projects to 2020 and then to 2037 depending on what the sun does

        So the equation has predicted the average global temperature trend 17 years into the future. This is actually a bit misleading because it used actual sunspot numbers. The projection to 2020 used NASA’s expected sunspot numbers so any change that the actual sunspot numbers will cause must be added.

      2. Hi Dan,

        I had a search through the literature to see what I could find about low altitude clouds and sunspots. It seems like there’s been a fair amount of research in this area already. I found a paper from 2010 which looks at the impact of sunspots on low altitude clouds. It’s called, Clouds, solar irradiance and mean surface temperature over the last century. This paper does find correlations between sunspot numbers and cloud cover. I’ve taken this image from the paper for you to see:

        But their main conclusion is that from 1956 onwards, the solar impact on global temperature is less than 14%. Here’s what they say:

        “Particular attention is devoted to answering the question, ‘what fraction of the observed increase in mean Global temperature () can be attributed to solar, as distinct from man-made, effects?’

        We conclude that a best estimate is ‘essentially’ all from 1900 to 1956 and <14% from 1956 to the present."

        The other question I might ask is how does your idea explain how Earth thawed from its pre-Cambrian snowball state? And how also can it explain the hot-house conditions of the Eocene when there was no ice at either pole and crocodiles were living in the Arctic yet solar intensity was a lot lower than today?

        I am inclined to think that yes, clouds do have an impact on global temperatures and are possibly influenced by sunspots and cosmic rays themselves, but there's a little more to the story than just this. Perhaps as the next step you could take your equation back in time to see how it stands up with global temperatures from past climates.

      3. I have a couple of quick questions. While your equation looks way too complicated for me to grasp, I did notice one thing and that is that you seem to be using the effective sea surface temperature anomaly. Why is that?

        My other query is how come most of the scientific literature finds a divergence between global temperature and solar activity from about 1980 onwards?

    2. Dan, maybe you can explain why I’m wrong. I’m always happy to be corrected but simply telling me I’m wrong and suggesting that I’m insufficiently knowledgeable to know that I’m wrong isn’t really going to do it. Why not at least try and put some effort into convincing me (or others that read this) that I’m wrong. Not doing so does rather suggest that maybe you’re unable to do so.

      1. Wotts – I looked through my posts and observed that, up to now, I have been trying to show what I did and what resulted when that was done (high correlation of calculations using an equation based on physics with measured average global temperature anomalies).

        I reviewed your comments and have an idea of why you are wrong. You have become mired in minutia. You are considering many factors (knobs to twiddle) that, on average, have very little influence on AGT. The simple approach that I used explains 90% of the AGT measurements using only one external forcing. Thus everything else must find room in the unexplained 10%. A big chunk of that 10% is used up in the random variation of the reported measurements (equivalent s.d. ≈ ±0.1 K).

      2. Dan, I’m not mired in minutia. If anything, I was maybe going into too much detail. All I was suggesting is that you have produced your parameters by fitting to some portion of the temperature anomaly data. Am I wrong about this? I also suggested that you’re only considering surface temperatures. Hence you’re not really modelling global warming (i.e., all excess energy) you’re simply looking at one aspect of global warming (surface temperatures). Am I wrong about this?

        When I mentioned we would know if your prediction was right, I was referring to your model apparently predicting a fall in surface temperatures in the next decade or so. Others predict opposite, hence it will be clear if your prediction was better than others. You say, above, that you predicted the last 17 years correctly. What did you assume about the sunspot numbers when you made that prediction?

  10. It always good to read someone who doesn’t merely re-iterate a load of data, but instead gives an interpretation in simple English. Just look back through your blogs, Rachel, and you will see what I mean. Great to see more balance coming into your blog as well.

  11. Rachel,
    WUWT has a new post on another failed model, ” Models Fail:Scandinavian Land Surface Temperature Anomalies.” In spite of Wotts’ denial , the conclusion of this post is that the Climate models being used by the IPCC for the upcoming AR5 show little skill at being able to simulate a number of variables .According to the models the Scandinavian Land Surface Air Temperatures should have warmed at twice the observed temperatures.In the view of the author,Bob Tisdale this is joining a growing list of model discrepancies.

    1. Douglas, not quite sure what I denied. I commented on some issues with Roy Spencer’s post, that you then didn’t really respond to. I don’t know a lot about these models but, as far as I’m aware, they’re primarily global circulation models, so I don’t actually know how you extract small regions (scandinavia, alaska) and what this means. As far as the global modelling is concerned, one still can’t actually claim that the models have failed since they’re consistent at the 5% level (i.e., the models predicted that there would be a 1 in 20 chance that the temperatures we observe would have happened). Another issue is that it is claimed that models fails to predict global warming, when really all that is being considered is surface temperatures. Only a few percent of the excess energy is associated with heating the atmosphere and so by focusing only on surface temperatures, we’re ignoring an awful of other evidence related to global warming.

    2. Doug, Anthony Watts is just trying to motivate his troops. The Earth is heating up, the ice is melting. In short, we have a greenhouse. And now it is beginning to look like finally, the USA is going to do something about it. Anthony Watts will always try to find an error or mistake while at the same time ignoring all the evidence. I really have no idea whether the Scandinavian models are correct or not and it really doesn’t matter.

  12. Rachel – Thanks for the graphs. I will study them more later.

    I think that the mistake that nearly everyone has made is that they only looked at TSI change, which is tiny, and decided that therefore sunspots had no significant effect. Using the sunspot time-integral makes a huge difference. A graph that goes back to 1610 is shown at . I expect that the ocean oscillations, which have been so pronounced since before 1900, will fade in and out over the centuries.

    A neat graph at shows AGT and atmospheric CO2 for the last 550 million or so years. About 440 million years ago the planet plunged into the Andean/Saharan ice age when the CO2 level was about 10 times the present. I see this as one of the things that corroborate that atmospheric CO2 has no significant effect on AGT.

    Having ruled out CO2 as a factor, I suspect that snowball earth, etc. were caused by variations in the sun that haven’t been figured out yet. As to crocodiles at the poles, was tectonic plate drift taken in to account?

    The equation works very well since accurate temperature measurements have been made world wide and fairly well since sunspot numbers have been regularly recorded. It works in the range of a degree or two if sunspots are known. Useless before that.

    I gotta get some sack time . . .

    1. Dan,

      The Eocene was only about 50 million years ago when the tectonic plates were in much the same place as today, thereby excluding theories of continental drift.

      There is an error with the graph of CO2 and global temperatures which you link to at geocraft. They have not taken into account the solar influence on global temperature. But climate scientists do. Climate scientists know that both the sun and CO2 influence global temperature and so they account for the influence of the sun. It is the people who do not endorse AGW who wave that graph about and completely ignore the sun’s influence. When the sun’s impact is accounted for, there is a remarkable correlation between CO2 and global temperature over the last 500 million years.

      If you have a look at the paper I referred to, although it finds a correlation between clouds and sunspots, it does not support your idea. They found that the solar influence on global temperature since 1956 is less than 14% which is vastly different to your 90%. I am not necessarily saying that they are right and you are wrong, because I really am not qualified to judge, but if scientists with years of research behind them are saying something quite different to you, then maybe it’s worth having a look at what they’re saying.

      1. Dan, Rachel has pretty much said all I was going to say. What Rachel highlights is why I would give what you’ve proposed some more thought. It may fit the temperature anomaly data, but that does not mean that it has any long-term predictive power nor that it really helps us to understand the processes involved in global warming.

  13. Wotts,
    As I am a trusting chap, I will accept what you say that the models cannot yet be said to have failed,and that they are not inconsistent at the 5% level.But on the second point,the models appear to be under pressure not only in the regions such as Alaska and Scandinavia,but in the global modelling.
    Specifically, on the growing list of Model discrepancies,see those at the end of Mr.Tisdale’ s post , including the April 20 post ,”Model-Data Comparison with Trend Maps: CMIP5 ( IPCC AR5) Models vs New GISS Land -Ocean Temperature Index.”
    The opening paragraph shows the records that the models can’t simulate.Note that It is the GISS Land -Ocean record which is assailed.You are correct that it is the oceans not surface temperature alone that are critical.
    If I may quote from an article in The Australian today by Maurice Newman,a former Chairman of the Australian Securities Exchange and the ABC, and a non-scientist ,
    ” Despite this ( dire warnings of extreme weather events without evidence ) , the voices of alarm and authority have been unable to hide the reality that statistically there has been no increase in global temperatures since 1997, despite an 8.3% increase in atmospheric CO2. For those who want to cite warming in some records,all data sets agree there has been none since 2000.In fact since 2002,a slight cooling has been observed.Who knew? Well not the warmist scientists.”

    1. A couple of quick comments. In Bob Tisdale’s post he really is cherry-picking somewhat. I do some quite complex modelling of my own and am impressed that using global circulation climate models one can produce surface temperature evolution that so closely matches what is observed (e.g., the first figure in Bob’s post). If you want to extract short periods, yes you can show differences, but the long-term trend is impressive. Also, all that Bob is doing here is comparing surface temperatures (land + ocean), so still isn’t considering the ocean heat content (90% of incoming energy).

      As far as the last decade is concerned, the influence of CO2 on temperature is determined through Delta F = 5.25 ln(C/Co) and then through Delta F = 3.7 Wm^-2 is expected to produce a transient climate response of about 1.8 degrees. Therefore, the influence of a certain increase in CO2 depends on CO. The amount added in the last decade should have produced (using this algorithm) a change of about 0.15 degrees C. This is quite small and is still consistent with the range. However, it is also small enough that it can easily be masked by natural variability on decade long timescales. Just because temperatures have been apparently flat for a decade is not a very good argument against CO2 being an important greenhouse gas (especially as the OHC has continued to rise).

    1. And that is another very biased article with little scientific basis to it but probably what one can expect from The Spectator. The true alarmists are not those who understand basic physics but rather those who oddly think that by shifting to a low carbon economy we will end up destroying jobs, our health, businesses and the economy. This is where the real alarmist nonsense lies.

      1. That I chose The Spectator review is neither here not there. There was a plethora of review articles I could have chosen instead. I am astonished that you so readily discard The Spectator, one of the world’s premier periodicals. Does it still arrive in your household each week? I am sorry to learn it was such a poor choice of reading matter by us for Ben.

      2. Not at all. He very much enjoyed reading it. But every single article related to global warming in the Spectator takes the position of denial. Every single one! So it is typical in that sense. I think I saw one article in the whole two years which accepted the scientific position, but that was it.

  14. Rachel,
    Here is the Author of “The Age of Global Warming : A History”,Rupert Darwall putting to the sword your conclusion on ” shifting to a low carbon economy”-
    ” Based on their collective record, natural scientists would be among the very last people to involve in government policy .During the First Environmental Wave of the late 1960’s and early 1970’s ,they had predicted the imminent collapse of civilisation and called for the abandonment of economic policies designed to satisfy humanity’s material needs and generate rising prosperity.Their dire predictions turned out to be completely wrong and their recommendations disastrous-had any government been foolish enough to have acted upon them.
    And so it proved in the era of global warming.Across every dimension, global warming policy has been a costly fiasco.Unsustainable commitments to solar and wind energy in Germany and Spain; the morally abhorrent diversion by rich countries of resources from growing food into making biofuels; the collapse of the EU’s carbon market;the transformation of the UK’s liberalised energy market producing some of the cheapest electricity to become Europe’s most expensive electricity producer; the scandals associated with the Clean Development Mechanism; the destruction of tropical rainforests to make way for palm oil plantations- all provide material for students of policy failure.
    ……The implications of Copenhagen on the efficacy of global warming policies are nothing short of disastrous.Whatever the predictive merits of the science ,the absence of a regime capping global greenhouse gas emissions rendered the West’s global warming policies completely pointless.The results of global warming’s political experiment already provide a definitive verdict : global warming policies have made the world unambiguously worse off , a conclusion which holds irrespective of the outcome of the geophysical experiment.”
    Do you understand what you are advocating when you speak of “alarmist nonsense” on the part of sceptics?
    I will leave you with the last word.

  15. Already in Australia, the effect of green policies on mining equities is patently obvious with the share price of many good companies plummeting more than 80% in value. Jobs are being lost and investors will see little to no return in the immediate future. All this, the result of alarmist policies based on dubious research.

    1. No, jobs are not going to be lost overall. There will be a net gain in employment by investing in clean energy. Sure the fossil fuel industry will suffer and I wouldn’t be surprised if those shares were already falling. You can probably thank Bill McKibben for that. I would be getting out now if you own shares in these companies. But the money realised through divestment is going to be invested elsewhere and the new industries that spring up as a result of this investment will create jobs.

  16. That sounds delusional to me. You’re living in a world of academia. There are no green jobs. When my shares go down 90%, I’ll have green policies to blame.

  17. Rachel ,
    Look at the sub-title of the paper you have linked at the University of Massachusetts”The Economic Benefits of investing in Clean Energy”. It is “How the Stimulus Program and new legislation can boost the Economy and create jobs”. I have read it.
    The Paper is dated June 2009.The investment figures are predicated on 2 pieces of legislation ,the ARRA Act and the American Clean Energy and Security Act 2009. The second is the Waxman-Markey Bill 2009 which failed to pass the Senate! See the Wikipedia post.
    I think the authors of the Paper might want to reconsider their conclusions.
    The usual alarmist “benchmark ” is the widely discredited Stern Review,on which see Peter Lilley’s GWPF report ,”What is wrong with the Stern Report?”

      1. Well, it’s ruining mine already, and I imagine there must be plenty of others like us seeing their retirement savings similarly decimated. It is all very well to espouse your views from the security of an academic job. At least you can have the satisfaction that your movement is producing “results”.

  18. It’s incredible that the Chairmen and CEOs of the well managed mining services companies I’ve invested so heavily in haven’t thought of this already! But then they are Australian companies not listed on the American stock market.
    Tim Flannery’s Geodynamics,, has similarly seen its share price plummet but at least it’s green. Perhaps, I should transfer the now very small sum remaining to that local company and have the satisfaction that we are at least saving the planet.

    1. Yes, things may be different in Australia. Here’s something recent from about clean energy investment (dated 1/7/2013):

      “For years sceptics have derided the potential of generating electricity from clean energy sources like solar and wind. The key argument has been clean energy is not cost competitive without government subsidies. Solar and wind farms were considered too expensive to compete with King Coal.

      In Australia that rationale has been turned upside down in the case of wind and it appears solar will soon follow the path of declining cost. Earlier in the year Bloomberg New Energy Finance released the following graph, demonstrating that the cost of generating power from wind is already lower than new coal in Australia. Here is the chart:” Go to for the chart and the rest of the story.

      NB: I am not a financial advisor and I am not responsible for possible losses incurred should anyone decide to put their money into the stocks listed at

  19. You mean you wouldn’t put your own money into those stocks? Too late for us, I’m afraid – nothing left to invest now!

  20. But I could, by ear marking some shares specifically for you and investing in companies which would meet your desired criteria. So in an indirect way, we can put your ideas into practice now.

    1. Go for it. I believe we are entering the age of low carbon energy, so that includes things like nuclear and carbon capture technologies as well as wind and solar.

      What do you think of Silex Systems recommended by I notice they have uranium enrichment as well as solar.

  21. Well, I had a look at them. Thanks. No dividends for at least ten years and the share price has gone from a high of $13.50 in 2007 to $2.60 when I looked this morning. So no return there and no capital gains either! In fact, the return on equity is -14.69%. That’s not going to put food on my table, keep us clothed or allow any travel.
    Can you find something with a decent price to earnings ratio and ROE please?

  22. TOX is a bit better as for the past four years it has been giving a dividend. The dividend yield, 1.3%. is still too low to live on and the share price is at an all time high. The company does have some good contracts going forward but the investor return is a turn off. No.

  23. Rachel ,
    I would appreciate a confirmation from you and Wottsupwiththatblog that there has been a hiatus in world surface temperatures over the last 16 years.
    See Knox and Douglass 2010 ,as contradicted by Nucitelli et al 2012 , commented on by Dr.Judith Curry at “Pause discussion thread : Part II”. Note in particular the series of comments by mainstream scientists at “Candid comments by climate scientists”, on Climate Etc.
    Secondly, why do you maintain that the OHC continues to show warming ? This is clearly in doubt if Trenberth is wrong in his dismissal of the Argo buoys.
    Lastly what empirical evidence is there that the Earth’s Heat content continues to rise other than Computer models?
    See Roger Pielke Snr.”My Comment on Robert S. Knox and David H .Douglass-Kevin Trenberth on the Climate Etc.Post on “Missing Heat”. ”
    Recently,the WMO produced a Report that ” the decadal rate of increase (of world temperature ) between 1991-2000 and 2000-2010 was unprecedented.” I understand that Professor Ed Hawkins of Reading University now points out this is no longer true if you compared 1993-2002 and 2003-2012, I.e if you took the most up to date records.
    Is it correct that in that case the latest decade showed a smaller increase over the preceding decade than either of the preceding decades did? In other words ,that the temperature standstill of the past 16 years has begun to show up in the decade by decade data. Correct or not?

    1. Firstly, I’m not really sure why you think I should be confirming anything for you. I don’t think I’ve asked the same of you.

      To give some kind of answer though. If you consider the last 16 years of global surface temperature anomaly data, the trend is positive (between 0.05 and 0.08 degrees C per decade depending on the dataset) but the errors are large. This means we can’t rule out that it hasn’t been flat or possibly even falling. However, that the trend is positive means that it is more likely that the surface temperatures have been rising than not. Therefore, I guess I’m not confirming what you want me to confirm (you have heard of the term confirmation bias I presume).

      As far as the ARGO floats are concerned, I don’t know enough about them to make any kind of definitive statement. Maybe there are issues. However, the current data analysis suggests that the ocean heat content has continued to rise. Furthermore, there is direct satellite measurements of the energy imbalance that is consistent with this continued rise in ocean heat content (although I will acknowledge that the errors in the satellite measurements are quite large). Therefore, there is plenty of empirical suggesting that the Earth’s heat content continues to rise. That models are consistent with this simply adds to the evidence and is not, by any stretch of the imagination, the only evidence.

      As far as the decade by decade comparison goes, I don’t know if Ed Hawkins is correct but I suspect he may be. One has to be careful though. There was a large ENSO event in the late 90s that increased surface temperatures substantially. After this event, the surface temperatures remained high (rather than dropping as they would be expected to do if there was no global warming) so if one considers the period 1991 – 2000 compared to 2001 – 2010, the latter is dominated by the post ENSO high temperatures, while the former is not. If you move the interval to 1993-2002 and 2003-2012, then the difference will be less marked. However, another large ENSO event (if it occurs and presumably it will) will shift the balance the other way. I’m not convinced that this means anything particularly significant at this stage.

      1. Wotts ,
        You have been very kind in answering so clearly and honestly the matters that I put to you.You are right,of course that there was no need for you to confirm anything for me as you had not in turn sought any admissions on my part.
        Frankly I sought your confirmation on the hiatus and the OHC issues because there is so much controversy on both.Some mainstream commentators angrily reject any notion of a temperature standstill and further confidently assert with peer reviewed papers in support that the oceans including the deep oceans are accumulating heat.Sceptics in turn point to peer reviewed papers or critiques of the mainstream papers in rebuttal.
        I now understand your opinion on both.
        As Rachel has now put me in “the witness box “, and in turn asked my confirmation on a number of central matters , I will withdraw for the moment to consider my answers, and give her a comprehensive response.

      2. No, Doug, I do not request your confirmation on those issues. Please don’t answer them. I don’t think it is worth our discussing them at all as I think you are very biased and immovable and I’m sure you think the same of me.

    2. Great response, Wotts. Thanks.

      Doug, to quote MikeM from a different thread, “nit-picking arguments about climate change are apt to distract from the main game”.

      We have discussed recent surface temperatures a number of times before and I see no reason why my explanation will satisfy you now since it hasn’t succeeded previously. This obsession with the past decade or so is a distraction from everything else.

      Perhaps you will kindly confirm for me that the world has warmed by about 0.8 degrees since 1880; that the arctic ice sheet has been steadily melting since satellite records began; that ice records show a very strong correlation between CO2 and temperature over the past 500 million years; that human carbon emissions are rising and so is atmospheric CO2; that the ten warmest years on record have all occurred since 1998; that sea level is rising as expected; that CO2 has very high absorption in the infra-red and so is a good greenhouse gas; should I go on?

      I don’t know about Trenberth and the buoys but if for some reason you doubt his findings, then go to the library and see what other scientists have to say. I’ve just done a quick search and have found two papers which confirm this. Here’s one of them –

      World ocean heat content and thermosteric sea level change (0-2000m) 1955-2010“. Here’s a graph from their paper which seems to confirm what Ternberth and others are saying:

      And from the discussions:

      One important result presented here is that each major ocean basin has warmed at nearly all latitudes. A net warming has occurred despite interannual to decadal variability of the ocean associated with phenomenon such as the El Niño-Southern Oscillation, the Pacific Decadal Oscillation, and the North Atlantic Oscillation as well as other such phenomenon

      1. Rachel ,
        I will honour your request not to answer your questions seriatim.It will free up my time ,and save my typing finger.
        I will say that there is no nit- picking on these issues.I do not know what Mike M thinks is the “main game.”
        There is ample material from Roger Pielke Sr. and others exposing the inadequacies of Levitus et al.2011, the paper to which you referred me.
        The recent peer reviewed material by Professors David Douglass and Robert Knox of Rochester ,New York ,has established that, contrary to various climate extremist assertions , there has been no net accumulation of “missing energy” in the form of heat in the oceans worldwide in the 6 years since ocean heat content was first reliably measured by the 3000 automated ARGO bathythermographs in 2003 .Their finding implies that the amount of warming we can expect from even quite a large increase in CO2 concentration is far less than the IPCC and other fellow travellers maintain.
        I don’t think referencing a contrary peer reviewed paper is “cherry picking . ” I trust that the IPCC does so in 2014 in AR5.

      2. Douglas, you’re certainly entitled to quote those papers. Personally, I think Douglass & Knox is a very poor piece of work. One simply reason, they only use data down to 700m. We have data going down to 2000m now. They also seem to equate the change in OHC with a top-of-atmosphere flux plus geothermal flux. If I understand what they’ve done, they’ve concluded that there are periods when the rise in OHC is slow and hence the TOA flux is zero (no global warming) and the rise is entirely due to geothermal flux. One immediate problem is that the geothermal flux has existed for billions of years and decreases with time. If this energy can be trapped in the oceans, why haven’t they boiled away. One could argue that it is trapped for some period of time and then released. But, there’s no evidence for this. It may be possible but why would we invoke this process when we have plenty of evidence in support of GHG trapping long-wavelength radiation and hence adding energy to the climate system. Also, we have satellite measurements that are consistent with this picture that Douglass & Knox appear to completely ignore.

    1. Hmm. I’m not sure what you think I’ve cherry-picked.

      “Cherry picking, suppressing evidence, or the fallacy of incomplete evidence is the act of pointing to individual cases or data that seem to confirm a particular position, while ignoring a significant portion of related cases or data that may contradict that position.”

      I actually think this is what Doug is guilty of. I have simply gone to the library catalogue and searched peer-reviewed literature for ocean heat content or whatever I happen to be interested in, and from this I get a feel for the overall position of the field.

Leave a Reply

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

You are commenting using your 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 )

Google+ photo

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

Connecting to %s