5. Changes in carbon dioxide levels actually don't cause climate changes

Some skeptics noticed that, during past climate changes, atmospheric CO2 levels actually followed the changes in temperature by about a thousand years.

If CO2 levels didn't cause the temperature to rise then, they think we can put as much into the air as we want now.


One reason given is that it is known that the Earth's atmosphere is already opaque to infrared radiation ("saturated"), so they think an increase in CO2 can't cause an increase in surface temperature.

Skeptics don't argue against the fact that the CO2 level has increased dramatically during the last 200 years. As the figures immediately above and below show, the evidence is just too overwhelming to deny. However, they have introduced several arguments against the scientists' interpretation of the causes and results of this change. Unfortunately, their arguments contain a grain of truth, one possible reason that the arguments have persisted.

First, it is true that during ice ages, for example, the changes in CO2 levels followed the change in temperature by a considerable time. Therefore, CO2 levels actually didn't force the changes during the ice ages. Rather, CO2 changes actually may have acted to stabilize the climate in these cases.

But, it has been believed for some time that the "forcing" factor, that factor that actually initiates changes during the ice age cycles, has been astronomical, not atmospheric.
The ice age cycles follows a certain repetitive pattern due to changes in the Earth's orbit and inclination.

The cycles were discovered and tied to ice age changes by Milankovitch in 1920. His theory has been well accepted since 1924, and has stood the test of time. No scientist has seriously proposed that the previous ice age cycles were caused by changes in CO2.

But that was then. This is now. Just because in the past the CO2 levels didn't cause climate change, but followed the changes, does not mean that a change in CO2 caused by human activity cannot force a change in climate. It can.

How an increase in carbon dioxide causes an increase in surface temperature.
To see how an increase in atmospheric CO2 tends to cause an increase in the surface temperature of the Earth, we must look in more detail about how the greenhouse effect actually works in Earth's atmosphere. The figure below shows the spectrum of radiation emitted by a body in thermal equilibrium at different temperatures. That is, it shows the energy contained in a unit frequency interval as a function of the frequency of the emission.

First, look at the spectrum of the sun's output in the figure above. Both axes are logarithmic. The vertical axis is the energy output per unit frequency, with each tick denoting a factor of ten higher in energy than the one below. The horizontal scale denotes the frequency in Hz (cps) where "9" means a one with 9 zeros following, i.e., a frequency of 1 gigahertz (GHz).

We can see that what we call light is just a narrow part of the spectrum of electromagnetic waves emitted by the sun. Frequencies of electromagnetic waves run from radio and microwaves at low frequencies (left side), to infrared radiation (heat), and then to light, ultraviolet and X rays at higher frequencies. At a temperature of 6000 C, the sun's energy output peaks in the visible spectrum. It is probably not an accident that our eyes have evolved to see this range of frequencies.

The Earth's atmosphere is essentially transparent at frequencies for waves near the maximum of the sun's output (see top bar in figure) so light and near-infrared radiation in particular come straight through our atmosphere. On average, when this radiation hits the surface about 30% of it is reflected, unchanged, right back into space. The rest is absorbed by the surface of the ground or ocean, where its energy is converted into heat.

Any object, at any temperature except absolute zero, emits electromagnetic radiation. The surface of the Earth is no exception. The shape of a thermal emission spectrum is always the same, but at a temperature of about 20 C, the peak in the energy output is at a much lower frequency than is the peak for solar radiation. The peak from a warm body is in the infrared range as can be seen in the figure. As shown, such a cooler body doesn't produce enough energy in the visible range for us to see it (without night-vision goggles). However, nearby objects warmer than the background we can sense as heat on our skin. In addition to radiation, of course, significant heat can be transmitted by conduction and convection depending on circumstances.

In the model,therefore, heat radiated from the ground and ocean surface heads skyward. If the Earth's atmosphere only consisted of oxygen and nitrogen, all this radiation would escape into space and Earth's surface would be much colder than it is. Instead, the Earth's atmosphere contains a small percentage of gasses which are not transparent in most of the infrared range. I
n order of effectiveness, these "greenhouse gasses" are water vapor, carbon dioxide, methane (CH4), and ozone (O3).

These four gasses stop the outbound infrared radiation in the lower atmosphere, where the radiation energy is converted into heating the air at the point where it is stopped. This air then re-radiates infrared radiation in a spectrum characteristic of the temperature at that point, and some of the re-radiated energy continues upward, to be absorbed and re-emitted again and again.

When the system is in equilibrium, as the radiation continues upward, the atmosphere slowly cools and thins out. Finally there comes a point where radiation can escape into space without being stopped by the greenhouse gasses. Note from the figure that even at the lowest temperature in the atmosphere, about -100 C, the spectrum of the radiation is much the same as from the warmer surface.

It is at this point, where the radiation is starting to escape the Earth's atmosphere, that a little bit of added greenhouse gas can make a big difference. It stops a little more of the escaping radiation and causes the Earth's atmosphere in this region to heat up and expand.

Note, in general, it is the rate of decline in temperature with increasing height which actually causes heat loss up through the atmosphere. That is, heat conduction always flows in the direction from hotter to colder. The steeper the gradient in temperature, the faster heat is transmitted. A little more greenhouse gas therefore acts to decease the overall slope of the temperature gradient. In a very real sense, it makes the "blanket" around the Earth a little thicker. Less heat is lost. The Earth's surface gets warmer.

As can be seen from the Wikipedia figure above, a substantial part of the heat from the Earth's atmosphere escapes into space (white upward arrow). A small change in this heat loss can make a big change in the Earth's surface temperature. Obviously, a change in the CO2 concentration CAN cause a difference in the temperature at the Earth's surface. Whether the amount of CO2 change so far has actually done so or not is examined in section 6..

We have also included in the figure at the above right an exponential extrapolation to the observed levels, under the simple assumption that the present growth rates stay the same. That is, we assume that we follow the advice of doing nothing as advised by many skeptics. The extrapolation suggests that, by 2050, there will be a doubling of the CO2 concentration compared to the pre-industrial maximum of 275 ppm measured in 1800. Even if it were true that the CO2 increase up to now hasn't caused an increase in temperature (and it probably has), the expected doubling certainly will! (For more information on the source of the measurements and the validity of the exponential extrapolation, see the Details section below.)

After revising this section, one begins to wonder how the skeptics can possibly maintain their point of view. They must either think that scientists all over the world are so stupid as to not know that the lower atmosphere is opaque to infrared radiation, or they must think that all of the world's scientists (not just the climate scientists) are members of a truly vast conspiracy. One without any real benefit for themselves. Except, of course, for the few "tobacco industry" scientists.

Overall, I think that most skeptics could not and would not have maintained such strange beliefs in defiance of logic without the influence of the "disinformation" campaign (lies) started by big tobacco in 1993, and maintained since by "scientists" and others supported covertly by special interests such as the coal and oil industries. They have had recent help from the religious creationists, who have developed and propagated good techniques for teaching people how to deny clearly established, accepted scientific facts. A separate short post has been added elsewhere that discusses how one can tell hoax from truth, and how scientists (and others) can and do distinguish controversial theory from established fact.

How long does the CO2 stay in the atmosphere?
It is also very important to ask how long the added CO2 stays in the atmosphere before it is removed by natural processes. It turns out that a significant fraction of it stays for a very long time.
The question is scientifically posed as follows. If we were able to completely stop emitting greenhouse gasses today, how long would it take for the atmosphere to clear, by natural processes, the excess CO2 we have already added? The figure below right shows the theoretical decay with time of a large pulse of CO2 injected into the Earth's climate system. This figure demonstrates how one can get very different answers depending on exactly what question is asked and what assumptions are made.

The time to get down to 80% of the original pulse is only a few years. The time to remove 50% of the initial CO2 pulse (the first "half time") is between 30 and 70 years (added dashed blue lines) depending on assumptions. However, note that this is NOT at all a linear system. The fractional decline in the figure increases rapidly with time, while it would be constant with time if it were a linear system. As the figure shows, the time to get down to 25% of the original pulse could take more than 300 years. If the system were linear, the total time to get to 1/4 of the original would only be twice the time to get to1/2 of the original, i.e., it would be 60-140 years.

Unfortunately, the removal of the last 15-30% takes thousands of years. The reason for the non-linearity is that the initial removal, over the first few hundreds of years, is accomplished by adsorption of CO2 into the ocean and incorporation into plants. However, not all of the CO2 can be removed this way, because the levels absorbed depend on some percentage of the excess remaining in the atmosphere. That is, the level of CO2 dissolved in the ocean will remain higher than normal only as long as some proportional excess level of CO2 remains in the atmosphere. The added CO2 is entirely removed from the system only when it becomes incorporated into carbonate rocks over thousands of years.

The idea of "stopping" global warming sounds good but unfortunately it is an impossible goal -- we have already put enough CO2 into the atmosphere that the excess "signature" will be with us for thousands of years -- as will the corresponding global temperature increase. What we can hope to do now is to slow down the very rapid rate of rise of the gas, so that our climate, our society and everything else that lives on our planet can have a little more time to adjust to the changes to come. Given how long CO2 remains in the atmosphere, of course, our eventual goal must be to completely stop putting significant CO2 in, or at least arrange to take it out as fast as it is put in.

How the atmosphere is modeled to produce predictions of future behavior
Some global warming skeptics apparently think that the climate modelers "put" anthropogenic global warming (AGW) into their models from the beginning. If they did, of course, that is what they would find. They don't. Because of the widespread misconception, we will here detail briefly how the modeling actually works.

The simplest possible yet useful atmospheric model is one dimensional. At the bottom is the ocean and the ocean surface. Next is the atmosphere, consisting of the appropriate mixture of oxygen, nitrogen and the greenhouse gasses carbon dioxide, methane and ozone. Water vapor concentrations are actually determined by the physics of the ocean surface and the atmosphere, so these concentrations will change as the climate does. At the top is the sun, putting out the appropriate solar forcing. A uniform, downward gravity field is assumed. The appropriate physics are input for the ocean surface and all of the gasses, that is, their absorption and transmission characteristics at different frequencies, the water vapor rising off the ocean as a function of temperature, etc. The model is "started" and the calculation continues until the model stabilizes.

In these calculations, the ocean absorbs and re-emits radiation according to its temperature.
At each wavelength and altitude, the absorption and re-emission of radiation is calculated along with other heat transport mechanisms. The atmosphere evolves to become stable, so that its density, pressure and temperature are in agreement with the heat transport from sun to surface and back up again.

Once the model atmosphere is stable, it is checked to see that it is a reasonable fit to the observed atmosphere. If not, there may be some aspect of the physics that has to be added in that was originally thought to be unimportant and was therefore ignored. The initial phase of modeling ends when the correct ocean surface temperature is obtained and the atmosphere is stable and a good fit to the present atmosphere (or to, say, the conditions in 1800).

Then, additional carbon dioxide is "forced" into the system. A pulse of CO2 may be assumed, or a predetermined behavior with time. In the ideal case, CO2 would be added at the bottom and allowed to diffuse naturally into the atmosphere. The behavior of the model is then followed to see the result. If, in the end, the ocean surface is heated up from the added CO2, then the modeler can say that the model predicts AGW. The result may be given as a prediction of the degree of warming for each megaton of carbon input to the atmosphere, etc.

No modeler starts out assuming AGW is occurring, and no modeler is so stupid as to ignore the saturation in the lower atmosphere, or to make many other of the mistakes that the skeptics seem to assume were made. And you can't get scientists all over the world to agree on anything, especially to be part of a vast conspiracy to fake results. Although a recent poll did show that an amazing 97% of climate scientists agreed that AGW had been shown to be likely!

As computer power has increased, model calculations have increased in complexity. Now calculations are being made using three dimensional representations of Earth and its atmosphere, with oceans and land being present, accurate to grid sizes down to a few hundred kilometers square, and even including clouds. Their calculations have improved, but all continue to show an AGW effect to a varying, but serious, degree.


The graph in the final section of "Our Earth in 2050" shows some of the variation in the models according to different assumptions about how mankind will react. We are not such optimists as the IPCC! Our prediction is that, partly because of the activities of the covert groups propagating global warming skepticism, mankind will do nothing useful in time to slow the process. We predict the consequences in "Our Earth in 2050."


DETAILS

The observed CO2 levels in the top graph from 1958 to 2010 came from accurate measurements at Mauna Loa, Hawaii, by C. D. Keeling of the Scripps Institute of Oceanography and by the National Oceanic and Atmospheric Administration.

The graph at the above right combines these data with CO2 levels from measurements of air bubbles trapped in ice cores (see review by Wikipedia).
The exponential extrapolation was done by matching an exponential function to the accurate data from 1958 on. An exponential response is the most natural match because: 1) it matches the appearance of the curve very well, and 2) it is the way things (bank accounts, national economies) naturally grow when the growth rate is constant. It should be no surprise that the CO2 emission matches an exponential which has an annual growth rate of 2.2%.

The energy spectra in the next figure were calculated from the formula developed by Planck for the "blackbody" radiation curve.

The next figure is from the Wikipedia commons.

The lowest right figure is from Global Warming Art.

COMMENTS
:


Feb. 8, 2010, Mike said...

1 Kilometer of atmosphere:

The first 770 meters are Nitrogen.

The next 210 meters are Oxygen.

That's 980 meters of the 1 kilometer. 20 meters to go.

The next 10 meters are water vapor.
10 meters left.
9 meters are argon.
Just 1 more metre.

A few gases make up the first bit of that last meter.

The last 36 centimeters of the kilometer - that's carbon dioxide.
A bit over one foot.

97% of that is produced by Mother Nature. It's natural.
There are just 14 millimeters left. Just over a centimeter - about half an inch.
That's the amount of carbon dioxide that global human activity puts into the atmosphere.


Feb 9, 2010. David Mills replied:
You don't say so, but I assume your point is: "How can 3% of the total CO2 matter?"
I am not sure where you got your CO2 numbers. (Please do give a reference if you have one.) I used Wikipedia as the first place to look for information, then checked important numbers in original sources. In sum: The pre-industrial CO2 levels were estimated to be 275 parts per million (ppm) in 1800. This is, by definition, the natural level. It is, in fact, the normal maximum level reached during each of the large number of interglacial periods over the last 1.3 million years.

In 1800, this natural CO2 was in equilibrium, that is, each year there was the same amount being added by natural processes as was being taken out of the atmosphere by natural processes.
In contrast, the most recent measurements at Mauna Loa in 2010 put the CO2 level at 388 ppm. The addition of 113 ppm isn't a natural variation, it has been shown to be from human sources by isotope ratios and other tests. The total is also much higher than at any time in the last 1.3 million years. That means the current amount of CO2 of human origin is now 113/388 of the total, or almost 30%. Much more than 3%!

The contribution to the total by human activities is so high because we have been putting it in so fast that there has not been time for it to be removed by natural processes, as discussed above..
Obviously, a 30% rise in CO2 can affect the Earth's climate a great deal. The exponential projection predicts that the human-added amount will be equal to the initial natural level by 2050.

There are expected to be many unpleasant effects from doubling the Earth's atmospheric level of CO2 over a period of only 100 years or so, given what an excellent greenhouse gas it is. No matter how small a component of the total atmosphere it might be.


Feb. 13, 2010: Dr M.Cejnar wrote...

Dear Sir,
Your use of the Mann team Hockey stick without even mentioning the MWP and the controversy surrounding the graph (numerical methods, Yamal trees, divergence problem etc) makes your blog look disingenuous to even a casual observer of the debate. The debate has moved well beyond the superficial treatment in your blog. As a physicist, you may however be able to review and answer one question bugging me - CO2 residence time. This is central to AGW and is reported as short as 5 years (R.H.Essenhigh Energy & Fuels 2009) or as long as 100 to 1000 years (many alarmmists, and IPCC, I believe). What is the evidence?

Feb. 14, and Mar. 21, 2010. David Mills replied...
Dr. Cejnar,
this blog is not intended for specialists, but for people who may be intimidated or confused by the esoteric claims and counterclaims. It is an introductory treatment. Also, I think the argument over the temperature is a big (and deliberate) misdirection. It is the CO2 level which is easy to measure and which has undeniably gone up dramatically, and which will drive all the other changes to come. So your question has particular pertinence. As a consequence, I have revised this section to include the answer to your question in the main body of the post.

We have had other comments which included arguments that atmospheric CO2 levels cannot affect the climate, mostly in posts on the Introduction page. Relevant excerpts from some of these posts are repeated here.

Feb 13, 2010...
There may be numerous reasons other than anthropogenic causes. We are, are we not, in an interglacial period? When warming occurs? And does not the CO2 in the atmosphere trail rather than lead increases in temperature?

Feb 17, 2010...
Warmer is better for life, lack of CO2 is one of the greatest limitations on the food chain, I am frezing my butt off 6 months a year and there is a lot more land between me and the pole than between me and the equator. I am unconvinced about data on past CO2 levels, this work looks doubtful. But if you accept it at all than the CO2 being a climate driver of any significance is a long dead idea. Causality violation is a big deal in the physics I know.

March 11, 2010....:
*2c. plants grow twice as much biomass under twice present [CO2]atm ( [X]atm =atmospheric concentration ). If anything, deserts would shrink, and the benefits to agriculture would be a massive assist to the poor. *2f. the blatant failure to consider that complex natural numerical progressions are rarely linear. No-one has examined whether [CO2]atm of 0.027% is at or above the maximum level at which CO2 can act as a 'greenhouse gas' under true atmospheric conditions. What if greater [CO2] has no further effect? March 15, 2010: . if he/she believes CO2 is the major driver of climate (it is not) than it's effect is going to be overstated (which it obviously is) and changes in CO2 levels will show far more change in the model than they have been observed in the real world. The failure of models to reflect real world observations is a major arguement against their use for projecting the future.

March 15, 2010. David Mills replies...
Most of these comments have been answered by revising the text of the main post above, and adding two figures.

March 15, 2010, Anonymous said... remember the Permian era?? anyone out there?? or more exactly, the end of the Permian era?? there are now real good models of what exactly it was that caused the GREATEST DIE-OFF of life in any one extinction event in the history of our planet, an extinction event that made the end of the dinosaurs look like Earth just had a cold....... well, pals, computer models of our future atmosphere put the Permian extinction event squarely in mind when you look at CO2 levels.....