I think the “Big Cheese” idea looks closer.

Like Josh, I also am British and know the phrase “sponger”. But they don’t tend to obtain money through threats. ]]>

…

No, it implies that 3 out of 5 are too busy to respond, and the other one threw the letter away thinking it was junk mail ]]>

Imagine you go to the doctor and your blood pressure is 140/90. He tells you that you must immediately eliminate all salt in your diet and go on blood pressure medication for the rest of your life.

What do you do?

Harry Flashman, how about policy decisions be made by policy experts and elected officials? I know climate scientists and their supporters are all geniuses, but I suspect their approach to political policy-making is a bit…self-centered.

]]>The models are not wrong. They fit the past data very well.

What is wrong are the predictions. If temperature anomalies stay about what they are now there will be more deviations between the predictions from the models and reality. It is beginning to show up but not enough so far for the modelers to adjust the models. They simply claim not enough time yet.

If you want to claim that the models are wrong, you need to run your analysis with both the CO2 concentrations vs. the temperature anomalies using a curve fitting analysis other than simple linear regression. You need an analysis that will show you a deviation from a steady increase of temperature while CO2 continues to be steadily increasing, that is a plateau will start because temperature anomalies are no longer increasing. From my simple linear regression analysis, the deviation will start at about 1990 and accentuates around 1998 to then becomes a flat line.

This is why my R squared values started decreasing for the data 1990-2014 (24 years span) and for 1998-2014 (16 years span). They decreased not because the lower number of data points (as suggested by rooter at 4.25 below) but because the increase in temperature anomalies themselves is no longer happening. Clearly the temperature anomalies between 1990 and 2014 are higher than between 1959 and 1990, so yes global warming (so to speak) did increase.

I see that you always plot the CO2 data from the same source I used when plotting your temperature anomalies. You need plot the annual average, not the monthly average. The monthly average is just a distraction. Plotting the annual average will give you a nice straight line, calculate the R squared value. It will a very high value, close to 1.0. It will then become easy to compare both trends, there is obviously no trend (your R squared values for temperature anomalies are zilch) for temperature but a very strong one for CO2.

In my assessment, the issue is not the models and not the warming. The issue is did CO2 caused it? It may have caused it at the beginning or it may just be a coincidence but since 1990 its influence, if indeed it was the cause, has greatly diminished and perhaps about to disappear completely.

]]>Have the greens yet worked out how to heat our homes/run our factories when the wind doesn’t blow and they have managed to shut down all our real poewer generators!!

Apologies no link to the article, but it should be available on the DT’s web site

Let us just for grins suppose that the atmosphere is dynamically disequilibrated in CO_2. That’s a pretty harmless assumption, since it is manifestly not stationary. Some would even call it a simple observational fact. Second let’s assume, as Bart does, that it is trying to relax to some (possibly mythical) “true equilibrium”. I say possibly mythical because Bart’s equations are in some fundamental sense a mean field theory projecting what is almost certainly complex multidimensional dynamics onto the time axis, where the actual dynamics could be poincare cycles around a fixed point in N dimensions, not relaxation, but because he is linearizing in 1 D it will look like simple relaxation across any sufficiently short time simply because that’s how Taylor series work. Note well that I am not stating that this step is “right” or “wrong” — when modelling a complex phenomenon at some point or another after studying the data one has to choose a model, unless one is using an unbiased but meaningless universal model such as a neural net.

Bart has selected a model that says “the model is disequilibrated in CO_2 because the CO_2 equilibrium fixed point is determined by the temperature”. When the temperature rises, the equilibrium CO_2 concentration rises, disequilibrating the atmosphere. The sum totality of sources and sinks and feedbacks then conspire to increase CO_2 to relax to the new equilibrium. This relaxation proceeds at a rate k, proportional to the temperature increase relative to the equilibrium temperature.

This model is not without problems. For one thing, what exactly is , both conceptually and numerically? For another, is itself stationary? It is safe enough to assume (again, really an observable fact) that , the temperature itself is a function of time and is very much not stationary. As a consequence

basically says that CO_2 will increase without bound as long as the temperature remains above this equilibrium temperature. And what determines this temperature? If it varies independent of itself, of what use is this equation for projecting the model into the future or the past? How can one compute it, or at least argue that it should have some particular value? If (for example) it is 290 K, cCO_2 should be falling. If it is 280 K, we are cosmically screwed, as cCO2 will increase without bound until the next ice age (according to this formula).

Since the latter seems (to me) to be more than a bit implausible, it appears (again, to me) that Bart’s model, however well or poorly it fits a (remarkably short) segment of the data from the very recent past, is at the very least *incomplete*. It contains absolutely nothing to describe some sort of eventual equilibration even if the temperature becomes constant, as long as that constant is higher than . It cannot possibly work in the distant past or the distant future, or else the atmospheric chemistry would be fundamentally unstable, and if there were *any* sort of greenhouse effect either positive or negative feedback would cause increase or decrease without bound.

In order to be taken seriously, then, the model requires at the very least and an equation that predicts it (because, as I said, as a universal constant is directly contradicted by billions of years of atmospheric stability against runaway CO_2 concentration). We then are led to a very interesting fork in the road.

Does this *temperature* that controls whether or not CO_2 *locally* increases or decreases in time depend on CO_2 concentration? Does it depend, in any way, on the total amount of carbon in play in the entire system of reservoirs? Does it depend in any way on the *details* of how those reservoirs and their “capacity” for CO_2?

Again, one has to be very careful with one’s answers. Basically, the answer to one or more of these had *better be yes* or one cannot prevent the implicit runaway catastrophe in the model. That is because there must be, somewhere in the system of equations, a projection that looks like this:

In words: There must exist a non-stationary equilibrium CO_2 concentration for the atmosphere. This equilibrium concentration can and probably does depend on many things — time, air temperature, ocean temperature, ocean *state* (as Bart himself has noted) and yes, sure, one of those parameters could well be “total carbon in play in the system” and hence be changing due to anthropogenic input.

If the current CO_2 concentration is less than the equilibrium, its rate of change is positive, driving it towards equilibrium. If it higher, the rate of change is negative, driving it back to equilibrium. The system is then *locally stable*, and unless internal feedbacks in the *still omitted* equations that determine the time variation of itself, which very likely is coupled to itself among many other things, are themselves positive in certain ways, it is *globally stable* as well. I’d obviously argue strongly for global stability because of the finitude of the carbon available in the entire system and the simple fact that while it has varied considerably on geological time, CO_2 has never evolved to either trivial fixed point of all CO_2 in the atmosphere or all CO_2 bound up in reservoirs in a CO_2 free atmosphere (basically, eternal hothouse Earth or eternal snowball earth, in all probability).

Even this equation, with all of the implicit power of making the equilibrium concentration a function of the kitchen sink as necessary, is probably not sufficient. It really doesn’t allow, for example, for adding a simple, direct bolus of CO_2 to the atmosphere from carbon sources outside of the existing system. It omits *us*. If we add us back in:

where represents the direct addition of CO_2 to the system by means not described in the shift of equilibrium CO_2 concentration and relaxation processes within the system, then the system *might* be sufficient. Without this, one couldn’t (say) double atmospheric CO_2 overnight and then watch the system relax to equilibrium (the same or a different one).

To conclude, this latter equation doesn’t mean that Bart’s description is contradicted. For Bart’s description to work locally, it suffices to make a function of atmospheric temperature but wait — *it already is,* obviously, a function of temperature, and beyond that is just a product of and the linear term projected into the dimension from some multidimensional Taylor series of about a more or less arbitrary point (hence the lack of meaning of ).

I’m not arguing for any particular form for the equilibrium concentration as a function of kitchen sinks. Nor am I taking a position concerning what is the *best local projection* of this multivariate differential into a restricted subspace of variables, although I do suspect that nearly all such projections will be naive and of limited utility outside of an interval where the linearizations happen to work (which we probably cannot compute a priori, making linearized models nearly useless as predictors as they’ll work — until they don’t). It is just to show that both you, Bart, and you, Ferdinand, can be entirely correct about assertions of agreement of your simple models and just be looking at different projections of the same basic *general* description of the dynamics.

This means that there is really no good reason for either of you to call the other a “liar” or even mistaken. What you are arguing about is what terms are the most important, which views of the data are most informative, given highly incomplete, noisy, and in the period over which Bart is looking, boringly monotonic data.

If I were to participate myself — and I’m not — I’d note that there is a very good correspondence between the rate of increase of atmospheric CO_2 over the last 10-15 years and the anthropogenic CO_2 contribution. A very, very good correspondence. Like, unity. If one assumes we are adding roughly 10 GT/year, that works out to 2 ppm/year, and gee, that’s a very good approximation to the slope of Mauna Loa data. This doesn’t completely contradict a 50 year relaxation rate, but it does mean that we are adding CO_2 at a rate that far exceeds this relaxation rate so that the net concentration is essentially perfectly tracking the additions as we remain far from equilibrium — the term in the equation above is dominant even if the system is disequilibrated high instead of low because is too small for us to see more than *modulation* of the response (what Bart is picking up on).

My last suggestion is that you two start a thread on this topic alone, independent of the ice core stuff. Put together a *complete* statement of your position — not just one equation in an obviously incomplete theory — and make sure that there are *equations*. Too much of the argument above is presenting a graph of this or graph of that where one cannot see where the data came from and what is data and what is a computation and what equation is supposed to be a sufficient explanation. Also, be prepared to defend your equation/hypothesis outside of the range of Mauna Loa. For better or worse, the Mauna Loa data is bo-ring and monotonic, and I can fit a monotonic function on a short interval lots of ways without the fit having the slightest extrapolative virtue. Any model proposed should have at least some virtue across all of HadCRUT4 and all of the combined ice core and ML data. And no omitted detail! If the model is manifestly unstable (e.g. Bart’s) you MUST address this. I’m fine with local linearization, but if it is globally unstable it surely isn’t crazy to require some statement about what *nonlinear* terms or multivariate behavior will eventually emerge to quench it.

rgb

]]>If so, is there anyone here with enough time and computer power to do a nice straightforward analysis using simple rules:

There is a MUCH simpler method to calculate average temperature.

1. Start with the very basic observation that your stations change over time.

2. That any adjustments to make your stations appear “unchanged” is thus a source of error.

3. Therefore, it is useless to try and build a temperature record based on fixed stations, as you can never be sure of how much error you introduced.

4. Instead, assume that your station readings are simply random samples in time and space.

5. apply sampling theory to pick a random samples that accurately recreate the spatial and temporal distribution of the earth surface over a year.

6. these samples should fit a normal distribution – check this assumption

7. calculate the average temperature and standard deviation and standard error for the year.

This result should be at least as accurate as any gridding method and has huge computational advantages. Anyone with a modern PC and a good sized drive should be able to tack this. All that is required is a small bit of custom programming to build and analyze the samples. I’ll probably use sql, as the problem lends itself readily to analysis on a database, but many different tools should be able to do the job.

]]>…

Imagine you are feeling ill.

You visit 10 doctors, and they examine you and do tests.

9 out of the 10 say you have cancer, the other one gives you some aspirin and sends you away.

..

What do you do? ]]>

Don’t keep saying that. The greenhouse effect, as defined by the IPCC, is not rteal. It is falsified by observations of nature. There has been no warming for 18 years while carbon dioxide steadily increased. Their greenhouse theory requires that addition of carbon dioxide to the atmosphere will cause warming and this has not happened.From this it follows that the greenhouse effect is not real. Their claim that the greenhouse effect as defined by IPCC exists is simply pseudo-science, an assertion contrary to the laws of nature.

]]>