Tuesday, August 26, 2014

Global Warming: Eyeballing the Data

Looking at the temperature data for the past century, the obvious interpretation is the combination of an upward trend with a roughly sinusoidal term of comparable magnitude and a period of about sixty years. When the additional term is going up, temperature rises at about twice the long term trend. When it is going down, it roughly cancels the trend, giving you stable temperatures from 1950 to 1980 and again from about 2000 to the present. I do not know if the oceanic model recently proposed to explain the additional term is correct or not, but that's the pattern suggested by the data.

If my interpretation is correct, there are two implications. One is that estimates of the rate of warming based on a period when the two effects were reinforcing each other will be too high. If we take 1940-2000 as one full cycle, the increase is just under .6°C, giving a trend of about .1°C/decade. That's one third as fast as the rate estimated in the first IPCC report, about half the rate estimated in later reports. If it holds, total warming by the end of the century, about half of which has already occurred, will be slightly under two degrees.

The other implication is that the current pause does not mean global warming is not real. If the pattern holds, warming will start again in about another fifteen years.

17 comments:

Eric said...

That model pretty much demolishes the anthropogenic theory, though.

Anonymous said...

Because?

Jesse said...

Eyeballing it a bit more, I would say it looks like the 1880-1910 is steeply down, the 1940-1970 is slightly down and so far 2000 on is basically flat. Each down cycle has less net down to it and so the steady up trend could be accelerating.

David Friedman said...

It doesn't demolish the anthropogenic theory unless you insist on single cause models.

Daublin said...

More precisely, it causes trouble for anthropogenic CO2 being the *main* driver of global temperature change. Your model is linear + sine, but CO2 levels increased significantly across the 20th century. If CO2 is driving the linear part of that model, then why was the slope the same at the beginning and end of the data period? If CO2 is dominant, or even major, you'd expect something more like quadratic + sine.

The sine part is a problem, too. Whatever is driving the sine part, it's large enough to offset CO2 entirely, at least temporarily.

Both aspects of your model imply that if we want to control the temperature in 2100, we will need to control at least two other forcings. If we only control one of the three, then the other two are surely still going to be enough to cause major temperature changes.

David Friedman said...

Daublin:

It's a pretty noisy graph, so I don't think you can be confident it's linear. Another commenter suggests that the effect is accelerating.

Beyond that, I'm not sure what you would expect from a more complicated analysis. Equilibrium temperature goes up not because CO2 is there but because it is increasing and depends on the log of CO2 concentration, so no reason to expect it to increase quadratically. On the other hand, getting to equilibrium takes a fair while.

Dan Pangburn said...

Search AGWunveiled to discover a corroborating analysis. The paper identifies the two natural drivers that explain measured average global temperatures since before 1900 with 95% correlation (R^2 greater than 0.9), credible hind-casting back to 1610 (Little Ice Age) and average global temperature trend prediction to 2037.

CO2 change is not a driver.

JWO said...

Way to little data to be much evidence of that.

Mac said...

This is an expansion and refinement of a previous post to your site.

AGW proof.

Carbon not causing warming.

Bodies radiate energy in the form of electromagnetic waves. The spectrum of this energy is determined by the temperature of the radiating surface. The surface of the Sun is at a temperature such that most of its radiation is in the frequency band we call visible light with some in the near infrared (which we sense as radiant heat) and some in the ultraviolet. Most of the UV is blocked high in the atmosphere. The visible (0.4-0.8 micron) and near infrared (0.75-1.4 micron) largely makes it to the earth’s surface.

The Earth having a much lower surface temperature will radiate energy back into space in the far infrared wavelengths (1-30 microns. Peaking around 10 microns. Most of the energy is in the 9-13 micron range.). Carbon dioxide and the other so called “greenhouse gasses” tend to absorb and re-radiate radiation in the far infrared and thus trap heat on the Earth.

Each vibrational mode of a particular type of molecule will act as a narrow band reject filter taking out one particular wavelength. This is based in quantum physics. The energy of a photon is inversely proportional to wavelength and only photons with the correct energy are absorbed. For CO2 these are 2.7, 4.3 and 15 microns. Two are base vibrations one is an overtone.

For an explanation of this see: http://www.wag.caltech.edu/home/jang/genchem/infrared.html

Note: Water as vapor, liquid and as ice has a very complex series of overtones and effects much of the far infrared.

Scientists like to assume things are linear, even when they are not. It makes modeling much easier. I suspect that many scientists involved in this modeling just assumed that the energy absorption was linear without checking.

I tried to research this. Starting in 2001. I Googled every form of the question I could think of for years without result. Finally I found the answer on my own bookshelf. The fifth edition of “Reference Data for Radio Engineers” (H.W. Sams & co. 1968) has a graph on page 26-28 showing the transmission of 300 meters (about 1000 feet) of air in the far infrared. The transmission of all three wavelengths affected by CO2 are already zero. The atmosphere is far thicker than 300 meters. Therefore it is safe to assume that almost no radiation at the wavelengths affected by CO2 was making it into space when the data for this graph was taken.

This says that CO2 has already done all the warming it can.

Case closed.

Mac said...

Continuing:

This brings up a second point. If the planet is being warmed by ANY greenhouse effect then the Troposphere (lower atmosphere) should be warming faster than the surface. Years of balloon and later satellite data do not show this. [This claim is widely made but I do not have hard data to reference.]

Likewise I should point out that the chart I referenced represents an upper bound on the distance to opaqueness. 300 meters is fully opaque but shorter distances may be also. I remember reading someplace that one of the CO2 absorbing wavelengths is opaque in 30 feet but I cannot point to that data and 300 meters is adequate to my case.

A second point: The modelers like to assume that the weak warming caused by CO2 is amplified by water in the atmosphere. This assumes a positive feedback with a loop gain near one.

As far as I can tell the modelers are making this as an assumption rather than basing it on data. The actions of water in the atmosphere is quite complex, complete with numerous non linear effects (cloud formation, precipitation etc.). I propose a simple observation to indicate the sign of this feedback. Is the day to night temperature swing larger or smaller with high humidity or with low humidity. This is quite unequivocal : Temperature swings are smaller with higher humidity and therefore the net feedback is negative and therefore water in the atmosphere is attenuating not amplifying other greenhouse gas effects.


Positive externalities.

Extra CO2 does have a major benefit. It fertilizes plants. The two most important plant nutrients are CO2 and water. The process of photosynthesis takes six CO2 molecules and six water molecules in the presence of chlorophyl and light and converts them into a glucose molecule and six molecules of oxygen. The plant then converts the glucose into other needed molecules.

Most of the photosynthesis takes place in the pores in the leaves. The inside of the pore is wet. On average about one hundred water molecules leave for every CO2 molecule entering. With higher CO2 concentrations plants tend to constrict the opening somewhat so less water is lost while the plant is still getting more CO2.

It was predicted in the 1970's and 1980's that we would see mass starvation in the 1990's. this did not happen. Part of the reason why is the fertilizing effect of the extra CO2.

Patrick said...

Mac, with regard to the saturation of CO2: the absorption of infrared by CO2 in the atmosphere will be a series of Voigt profiles (a convolution of the natural Lorentzian linewidth and the thermal Gaussian distributions). The optical density at a given wavelength will be proportional to the resonance cross section, times the column density of CO2, times a factor from the nearest absorption peak peak. So even if the peak of the distribution is saturated, increasing the density of CO2 will keep increasing total absorption.

If we simplify to say that any optical density above 1 corresponds to total absorption and any below 1 corresponds to no absorption, we get a "blackout window" in frequency around each absorption peak. Those windows will grow larger with higher CO2 densities, with a rate determined by the shape of the absorption curve.

Thus, there will be some effect from increasing CO2 density, no matter how high it gets. The current rule of thumb is that this causes a logarithmic heating effect of about 1 degree C per doubling of CO2. This estimate may or may not be plausible, but the general idea is at least self-consistent.

Daublin said...

David, I agree with the individual points you raise, though I would quibble with the claim that the "linear" component is definitely curving up a little. The data are too messy to say.

As you imply in another comment, this analysis demolishes a "single cause model". I am just calling out that even the linear part is hard to attribute wholely to CO2. If the linear part is mainly due to CO2, how do you explain the fast temperature growth in the early 20th century?

It might seem like a picky point, but consider the context. We are talking about humanity controlling CO2 as a way to control temperature. If CO2 is just one factor among many, then we won't have very good control over temperature after all.

Dan Pangburn said...

Terrestrial radiation absorbed by CO2 is immediately thermalized, i.e. the radiant energy absorbed by CO2 molecules is immediately (about 0.1 nanosecond) transferred (in a process similar to thermal conduction) to other atmospheric molecules which outnumber CO2 molecules 2500 to 1. CO2 can only absorb terrestrial EMR that has wave length 14-16 microns out of the significant range (mostly within 5-50 microns) of terrestrial radiation.

The absorption/thermalization quickly reduces the 14-16 micron radiation flux (there is only enough CO2 to reduce 14-16 micron radiation by about half). The warmed air rises in some places (appreciated by soaring birds and sail planes) and cooler air falls in others (pilots and passengers call these air pockets). Approximately 98% of the air cannot participate radiatively at terrestrial wave lengths.

The radiative participation (absorption/thermalization/reverse thermalization) of water vapor is not significantly affected.

renminbi said...

The various global temperature estimates should not be taken at face value since the controlling authorities have been less than forthcoming in explaining how their data was handled. Once upon a time the 1930s were the warmest decade for the continental US,but that was adjusted away.

Meanwhile in Oz:

http://wattsupwiththat.com/2014/08/26/boms-bomb-on-station-temperature-trend-fiddling/

Anonymous said...

Why does "eyeballing" the data suffice in lieu of actual analysis?

Tibor said...

I agree with Anonymous that one should make a better analysis than just visual inference on a descriptive statistic - especially since the data are indeed quite noisy. On the other hand, I assume that the purpose of this blog post is not to present a perfect analysis, but rather to start a discussion...in which case eyeballing is fine :)

In any case, I think we cannot confidently say based on the data whether temperature will keep rising (in the forseeable future) or not. While there is a clear upward trend in the recent past, there are factors which involve the temperature which can happen in very long cycles and we have reliable temperature data for only a couple decades and a bit less reliable for a couple centuries, but that could still not be enough to capture everything.

At the same time, it is not possible to assume from the data that the warming has stopped either. That is even easier to see as there have been times of even declining temperature lasting a couple of years in the recent past (1940s-1950s).

As long as the IPCC predictions were based on the trends just before this stagnation though (which I am not sure about, I am not interested in the topic enough to look it up), then they don't take such plateaus or even downward slopes into account and their prediction is then likely to be overblown...but are they not updating the projections every year based on incoming data (while using all the available data from the past)?

Anonymous said...

I think you might enjoy this: http://xkcd.com/1379/ .