Steve Chu: “Climate More Sensitive Than We Thought”

The quote in the title comes from Obama’s nominee to become energy secretary, Steven Chu.  Specifically,

Chu’s views on climate change would be among the most forceful ever held by a cabinet member. In an interview with The Post last year, he said that the cost of electricity was “anomalously low” in the United States, that a cap-and-trade approach to limiting greenhouse gases “is an absolutely non-partisan issue,” and that scientists had come to “realize that the climate is much more sensitive than we thought.”

I will leave aside of why hard scientists typically make bad government officials (short answer:  they have a tendency towards hubris in their belief in a technocrats ability to optimize complex systems.  If one thinks they can assign a 95% probability that a specific hurricane is due to man-made CO2, against the backdrop of the unimaginable chaos of the Earth’s climate, then they will often have similar overconfidence in regulating the economy and/or individual behavior).

However, I want to briefly touch on his “more sensitive” comment.

Using assumptions from the last IPCC report, we can disaggregate climate forecasts into two components:  the amount of warming from CO2 alone, and the multiplication of this warming by feedbacks in the climate.  As I have pointed out before, even by the IPCC’s assumptions, most of the warming comes not from CO2 alone, but from assumed quite large positive feedbacks.


This is based on the formula used by the IPCC (which may or may not be exaggerated)

T = F(C2) – F(C1) Where F(c) = Ln(1+1.2c+0.005c2+0.0000014c3)

Plotting this formula, we get the blue no-feedback line above (which leads to about a degree of warming over the next century).  We then apply the standard feedback formula of Multiplier = 1/(1-feedback%)  to get the other lines with feedback.  It requires a very high 60% positive feedback number to get a 3C per century rise, close to the IPCC base forecast, and nutty 87% feedback to get temperature rises as high as 10C, which have been quoted breathlessly in the press.  It is amazing to me that any natural scientist can blithely accept such feedback numbers as making any sense at all, particularly since every other long-term stable natural process is dominated by negative rather than positive feedback.

By saying that climate is “more sensitive than we thought” means essentially that Mr. Chu and others are assuming higher and higher levels of positive feedback.  But even the lower feedback numbers are almost impossible to justify given past experience.  If we project these sensitivity numbers backwards, we see:


The higher forecasts for the future imply that we should have seen 2-4C of warming over the last century, which we clearly have not.  Even if all the past warming of the last century is attributable to man’s CO2  (a highly unlikely assumption) past history only really justifies the zero feedback case  (yes, I know about damping and time delays and masking and all that — but these adjustments don’t come close to closing the gap).

In fact, there is good evidencethat at most, man’s CO2 is responsible for about half the past warming, or 0.3-0.4C.  But if that is the case, as the Reference Frame put it:

The authors looked at 750 years worth of the local ice core, especially the oxygen isotope. They claim to have found a very strong correlation between the concentration of this isotope (i.e. temperature) on one side and the known solar activity in the epoch 1250-1850. Their data seem to be precise enough to determine the lag, about 10-30 years. It takes some time for the climate to respond to the solar changes.

It seems that they also have data to claim that the correlation gets less precise after 1850. They attribute the deviation to CO2 and by comparing the magnitude of the forcings, they conclude that “Our results are in agreement with studies based on NH temperature reconstructions [Scafetta et al., 2007] revealing that only up to approximately 50% of the observed global warming in the last 100 years can be explained by the Sun.”…

Note that if 0.3 °C or 0.4 °C of warming in the 20th century was due to the increasing CO2 levels, the climate sensitivity is decisively smaller than 1 °C. At any rate, the expected 21st century warming due to CO2 would be another 0.3-0.4 °C (the effect of newer CO2 molecules is slowing down for higher concentrations), and this time, if the solar activity contributes with the opposite sign, these two effects could cancel.

Not surprisingly, then, given enough time to measure against them, alarmist climate forecasts, such as James Hansen’s below, tend over-estimate actual warming.  Which is probably why the IPCC throws out their forecasts and redoes them every 5 years, so no one can call them on their failures (click to enlarge chart below)


Because, at the end of the day, for whatever reason, warming has slowed or stopped over the last 10 years, even as CO2 concentrations have increased faster than ever in the modern period.  So it is hard to say what physical evidence one can have that tenperature sensitivity to CO2 is increasing.


Update: First, to answer a couple of questions, the data above is from the UAH, not Hansen’s GISS.  To be fair to Hansen, it has been adjusted to be re-centered on his data for the period before 1988 (since all of the major data sets use different zero centers for their anomalies, they have to be re-centered to be compared, a step many often forget to take).

I have a number of issues with the quality and reliability of surface temperature data, and the GISS data in particular, so I think the satellite data is a better source (just as we abandoned observations by passing ships in favor of satellite measurement of sea ice extent, it is probably time to do the same for surface temperature measurement).    Second, if I understand one of the comments correctly, there is some implication that I am being nefarious is cutting off the data in August of 2008.   Hardly.  Unlike those who work at this full time, I do this as a hobby between crises in my day job,  so I tend to reuse charts for a few months until I have time to create new ones.  Certainly there is nothing in the Sep-Nov UAH temperature data, though, that magically validates Hansen’s forecast.  I think November was a few tenths higher than August (making it just about even with June 1988) but well short of Hansen’s forecast.

One thing I didn’t mention, but Hansen and his enablers are often dishonest in trying to explain away the above forecast.  They will say, well, the A case was extreme and not meant to conform to reality.  But the only differences between the forecasts was in their CO2 output assumptions, and in fact Hansen A actually understated CO2 production and growth since 1988.  If anything, it was conservative!

15 thoughts on “Steve Chu: “Climate More Sensitive Than We Thought””

  1. Chu is the latest in a long line of the exact same type of consensus driven, well educated technocrats who has wrecked the economy.

  2. To add to the craziness, Obama’s new “energy czar” will be Carol Browner – Clinton’s wacky EPA head. She may have more power than Chu, and far less sense or scientific background.

    They might as well put Al Gore in charge of our economy.

    So much for economic recovery or energy independence.

    NOTE: I was unable to post this comment using Firefox. On IE, I couldn’t do it without signing in to TypePad

  3. Are you people crazy??
    “So much for economic recovery or energy independence.”

    What kind of crack are you smoking Moore?! Steve Chu has been pushing LBNL to be a leader in developing alternative energy sources. In case you can’t connect the dots, alternative energy sources will make us more energy independent.

    Either way, having Chu as DOE chief will be good for the US and science. We need someone who understands science deeply to turn the corner from the disastrous Bush administration. Bodman, current DOE head, used to be a scientist 30 years ago. Now he’s a puppet. It’s good to get someone with brains in the office.

  4. Steve Chu is a bright ambitious guy who definitely knows his physics.

    He would not be signed onto the global-warming issue unless he were very convinced it is real, and he would not jump into the DoE role unless he thought he could make a substantial contribution.

    Just another example of how mainstream science has a non-ideological bent – towards reflecting reality rather than “market-oriented” interpretations.

  5. Chu will screw this up and move on in less than 2 years.
    He is clearly a great physics person, but he zip experience with running large programs, and his comment about climate sensitivity shows he is just repeating what people he has mistakenly trust have told him to think on climate.
    If Obama had gotten an experienced engineer with executive level leadership, he would have a better chance of success.

  6. Afer reading some of Chu’s other stuff, I take it back. He might last four years, but he will be a disaster. He is, to be diplomatic, very eccentric.

  7. Neal J. King: “Just another example of how mainstream science has a non-ideological bent – towards reflecting reality rather than ‘market-oriented’ interpretations.”

    dunno … i think academia is much like a business driven by money / funding / connections / ego / personal biases / that pursuit of knowledge truthiness thing / ego / memes / hatred of superman because of personal hair loss / rivalries / cliques / the dark lord sauron / intellectual ponzi schemes / EGO / etc. academia just sells a different product instead of shoes and ships and sealing wax, cabbages and kings. goo goo g’joob.

  8. Hi Warren,

    Can you clarify the source of the actual temperature data in the penultimate graph? I’d like to repost that picture, but I worry that I won’t be able to explain where those actual temperature numbers come from, if they’re consistent, etc.

    In other words, by putting this picture up, I expect the data to be attacked, and I’d like to have some defenses in place.

  9. Crackpot science will dominate the Obama administration. Just look at science advisor John “population bomb” Holden as the template.

    Computer models are not science.

  10. Oh wow. That graph with Hansen’s scenarios is incredible. How on earth can anyone be such an ass that he/she removes the other scenarios from the graph and only shows the one which has nothing to do with what happened???

    Hansen’s predictions had 3 scenarios and the graph only shows one. Here are all three:

    How about that. How wrong is he? To understand it we have to know what the scenarios were:

    Line A was a temperature trend prediction based on rapid emissions growth and no large volcanic event; it was a steep climb through the year 2000 and beyond.

    Line B was based on modest emissions growth and one large volcanic eruption in the mid 1990s.

    Line C began along the same trajectory as Line B, and included the same volcanic eruption, but showed reductions in the growth of CO2 emission by the turn of the century — the result of hypothetical government controls.


    Also, check this to see whether the global warming has “stopped” in the last 10 years:

    The temperatures follow the linear trend incredibly well!

  11. @ bill tb

    “Computer models are not science.”

    I take issue with this statement, and I will explain why. Bear in mind that I generally support the “denier” position, since I acknowledge the existence of negative feedbacks which prevented runaway greenhouse warming during the time of the dinosaurs.

    “Models” are an integral part of science. We can never fully understand the world’s complexity so we have to build models, with certain built in, basic assumptions. If the assumptions are correct, then we will see complex, correct behaviour emerge from the model. We won’t capture all the complexity of the real system, because that is not possible.

    The climate models are dodgy because they assume the conclusion – that C02 is a significant greenhouse gas and is responsible for most of the slight warming of the past 150 years. If a false assumption is built in, then surprise surprise, the model loses its predictive power, hence the inability to predict the recent pause in the warming trend. If I suggested that listening to pop music caused global warming, I could build a model which explains very well past warming, if I build it carefully enough, but it would have little predictive power since its foundation is false.

    So models are the very foundation of science, since we can’t do better. Even the most successfully verified theory in science, Quantum Electrodynamics (QED), is still a model, and we know it does not correspond exactly to reality, but it comes darn close, because the built in assumptions (that particles obey the laws of special relativity and quantum mechanics) are sound.

    It does not matter whether the model is constructed with paper and pencil, or with a computer. The newest quantum field theory, Quantum Chromodynamics (QCD), is too complex to do on paper. There is no choice but to perform the calculations of paper, and we call this field of study Lattice QCD. However, it is still a model. Models are not wrong, they are essential, and do not become a farce just because the number crunching is done by a processor rather than by hand. The real worth of a model lies in its assumptions. If they are weak, the model is weak.

    Soem time ago, someone noticed that the planet was warming a bit, which should have been unsurprising considering that we had not long come out of a little Ice Age. However, that person was so dazzled by the high rise buildings and smoky factories that he or she fell right into the correlation is causation trap. This violates basic logic, but it is understandable. The trouble is, scientists then built this assumption (that CO2 causes most observed warming) right into their models. The models, which had to be built from scratch since climatology was a relatively new field, were flawed from their inception. CO2 is not a powerful greenhouse gas, and it is not the only one – and, negative feedbacks exist.

    It just so happens that anthropogenic CO2 emissions increased at the same time as the observed warming. We could have pinned any other factor against the warming, so long as it increased during the 20th century (maybe the viewing of naughty pictures or something?). This is bad science, and someone ought to teach climate “scientists” some basic statistics, rather than hand waving arguments.

    So, please don’t condemn computer modelling. Science is becoming so complex that soon no field will be able to avoid them. We are only at the very beginning of centuries of computer science. I am a theoretical physics student doing QCD, and I would be laughed at if I tried to do QCD calculations on a piece of paper. The climate system is even more complex than interacting quarks, since there are many more variables. The climate models need to be rebuilt with the correct assumptions, and this will not be possible for decades, since we have not even discovered some of the variables and factors influencing climate on the planet.

    In the mean time we are forced to endure climate alarmists disgustingly accusing “deniers” of being like the perpetrators of the holocaust, and equating the possibilty of disaster with truth value (I don’t need to explain how fallacious the disaster = truth argument is), and telling us that the explanation for the planets small warming trend over the last 150 years is CO2, with a bit of El Nino and La Nina thrown in to explain the times when things dont match their predictions. Soon this simplistic model will bite the dust, and people will get their history books out to try to find out who exactly suggested that CO2 was effecting the climate.

  12. Your dishonesty is truly disgusting. Only displaying Scenario A is really old and discredited, as Tuukka Simonen explained. You say “the only differences between the forecasts was in their CO2 output assumptions”, which is incorrect. I am quite sure you know it’s incorrect, you’re just lying to the weak-minded, who lap it up. Your graph entitled “high feedbacks greatly over-predict past warming” is pure fiction – your lines bear no relation to the predictions of any climate model. Your whole approach is idiotically stupid, and it’s obvious that you have neither the ability or inclination to realise what spectacular mistakes you’re making, over and over again.

  13. First time on this site and I like what I see. You seem to have a lot of critics which is a good indication you are doing something right. The criticism that seems to show up most common is the statistical significance, or lack of it, of a 10 year period. If the AGW crowd was looking at the “unprecedented warming trend” perhaps the cooling could be overlooked as just a slight deviation from the overall prediction and they could and would defend their position. BUT….. this cooling has been predicted (not projected) for some time based on the solar effects that the AGWA’s don’t recognize. Conveniently ignore is probably more accurate. Like most AGWA projections, this is based on linear regression. If you apply a fourth order polynomial analysis you will see that there is a downward temperature trend. This analysis is far more appropriate than linear which has no place in climate science. In a steady state situation it might be appropriate to make the claim that this cooling is part of a continuous progression but there has been identified a significant change in the game and that is the Sun. The Earth is getting cooler. Only a dishonest person could deny this. Now who are the denialists.

Comments are closed.