by Leonard Weinstein, ScD. (See short bio on Scientists page.)
Background and the Issue
The material in this first section tries to define the issue clearly rather than get into specific numbers and results. The comments may include discussion of specific results, but interpretation of available data is given in later sections. Before I describe the atmospheric greenhouse effect, I need to make three points:
The first is that the effect is not like a greenhouse for growing plants. That type of greenhouse actually is convection limited (trapped gas) volume, where the interior gas is heated, but can’t convectively carry the heat away. Heat conduction in a gas is far less effective that convection to transport energy, and radiation is limited in the amount of heat transfer for these cases. It is the same effect you get in an enclosed car in sunlight, where the interior gets very hot. Open the windows and the car cools quickly due to the ability for convection to the atmosphere.
The second point is that the atmospheric greenhouse gases act as radiation insulators but the atmosphere can freely convectively transport thermal energy. This radiation insulation effect is often described as causing additional heating from back radiation. While there is absorption and re radiation of thermal energy, including back radiation, that description is misleading and in fact wrong in its implication. A later section shows the result, which requires an addition effect to result in the heating.
The third point is that the concept of average planet temperature is quite flawed. Since temperature varies a lot day to night, at different latitudes, and different seasons, the fact that the black body radiation is highly non-linear (fourth power effect), and that local regions vary so much, just doing a linear average of temperatures is generally inaccurate in characterizing what is going on. However, it is the metric used, and will be assumed here to be adequate for the comparisons, although I have serious doubts on that assumption.
Earth’s atmospheric greenhouse gases
There are several gases (and particles) in Earth’s atmosphere that are relatively transparent to incoming sunlight but optically absorbing of most of the thermally radiating outgoing energy. The principal one is water vapor, which absorbs about half of the thermal energy from the surface. In addition, water droplets in clouds absorb about half of the remainder, and also absorb some of the incoming sunlight. Dust particles and aerosols also absorb some incoming and some outgoing energy. The next main gas is CO2, which has been described as absorbing 8% to 20% of the outgoing energy (by different authors). Methane and other gases absorb the rest that is not directly transmitted. In addition, about 8% of the surface and atmospheric absorbed radiation is directly radiated to space through optical “windows” in the GHG (spectral regions which are not absorbed).
Atmospheric greenhouse effect
Once a sufficient amount of atmospheric long wavelength IR absorbing gases (called atmospheric greenhouse gases by convention) are present, they reduce the direct radiation transmission to space enough so that convection and relayed radiation becomes the dominant modes to transport energy from the surface (and from absorbed incoming energy) to a high altitude where it is radiated to space. The only way the energy can be radiated to space from the high altitude is from the absorbing and radiating gases. Once this situation is encountered, adding more absorbing and radiating gases (but not enough to significantly increase the total mass of the atmosphere) do not change the fact that convection is still the dominant heat transport mechanism. In fact, this reduces the radiation conduction, so that convection carries even more of the total energy. However, adding more absorbing gas does raise the altitude of outgoing radiation somewhat (not because of the tiny added mass, but because the height where the radiating gas concentration is suitable to radiate to space is increased). It is this increase in altitude of the outgoing radiation that results in the slight temperature increase.
Once the dominant mode of heat transport is convection (buoyancy, winds, and turbulent mixing), the atmosphere will form and maintain (on the average) an adiabatic lapse rate. This lapse rate is a temperature gradient due to the cooling effect of rising gas in a dropping pressure (due to gravity). Evaporation from the surface and condensing water vapor in the atmosphere change the level of the lapse rate from the dry air value, and this is called the wet adiabatic lapse rate. The outgoing radiation has to equal the incoming absorbed radiation unless the temperature is changing, but here we consider the case where the temperature has leveled off for simplicity (as it has on Earth for the last decade or so). In that case, the match of radiation out to space, from a particular effective altitude, to absorbed input radiation, determines the effective temperature of the gas at that altitude. This temperature is then added to the lapse rate times the effective altitude of outgoing radiation, and this gives the ground effective temperature. The combination of moving the location of the effective level of the atmosphere, where the radiation to space occurs, to a higher altitude, and adding the effect of the lapse rate times increased altitude is the source of higher low altitude and surface temperatures.
Feedback effects and sensitivity
The issue is actually more complex than indicated above due to what are called feedback effects in the sensitivity (sensitivity is defined as the change in average temperature caused by doubling the concentration of CO2 in the atmosphere, including feedbacks, and is thought to be approximately constant for CO2). The increase in CO2 comes mainly from burning of fossil fuels, manufacture of cement, and land use practices. The argument goes this way:
The doubling of the CO2 concentration would cause a temperature rise of about 1.2 C if that were the only change. However, the small temperature rise causes increased water vapor content, since vapor pressure increases with temperature. In addition, the higher temperature in the oceans would cause more CO2 to be released into the atmosphere, since solubility of CO2 decreases at higher temperature. An additional effect would come from more melting of sea ice at the high latitudes from the warmer temperature, which decreases the average albedo of Earth, and thus increase absorbed solar energy. All of these effects drive the temperature even higher, until the rapid nonlinear increase in black body radiation with increasing temperature stops further increase. However, there are other effects that also occur. The main ones are:
- Clouds may increases, increasing albedo and decrease rather than increase sensitivity.
- Aerosols, also produced by the burning of fossil and other fuels, reduce incoming solar energy.
The main issue of the anthropogenic global warming debate is what is the net effect of all feedbacks. If they were strongly positive, the sensitivity would be much larger. If negative, the sensitivity would be smaller. Many scientists posit a net sensitivity of 2 to 6 C, while some skeptics think it is 0.2 to 1 C. The entire debate hinges on the actual value.
Possible other causes of temperature change
In addition to atmospheric greenhouse gas changes and feedback effects, there are numerous other possible causes for short term and longer-term temperature changes. Some of the main ones in order of increasing time scale are:
- Large volcanic eruptions (typically negative effect lasts one or at most a few years)
- Solar activity (spots, spectral variation, and insolation)
- Cause of cloud forcing from cosmic radiation variation
- Long period ocean current variation (PDO, ENSO, AMO, etc.)
- Planetary tilt and orbit variation
- Land movement, which also changes land at high latitudes and ocean circulation
The Logical Flaws in the CO2 Positive Feedback Argument
The increase in the atmospheric CO2 concentration in the last 150 years can reasonably be blamed on human activity such as the burning of massive quantities of fossil fuels. CO2 is an optically absorbing (and emitting) gas in part of the wavelength range of thermal emission from the Earth.
The presence of atmospheric gases that absorb in the thermal wavelengths has the effect to raise the location of outgoing radiation to space that matches the absorbed solar input radiation. This effect, in conjunction with the formation and maintenance of the atmospheric adiabatic lapse rate, causes the surface and lower atmosphere temperature to be higher than for the case with no absorbing gases. This effect is called the atmospheric greenhouse effect (AGE).
The effect of increasing the quantity of any absorbing gas in the AGE is to raise the location of outgoing radiation, and this in conjunction with the lapse rate increases the surface temperature. Calculations based on the absorption spectra of all the absorbing gases indicate that water vapor is by far the main greenhouse gas, and clouds have the next highest effect. CO2 is third (although aerosols have a large and less known effect).
Calculation were made assuming that the quantity of CO2 in the atmosphere was doubled from recent levels, and also assumed that all other contributors to AGE were not changed, to see what effect this would have on the temperature. The calculations indicated that about 1.2 C increase would occur for each doubling of CO2. In fact, other factors might occur if the doubling of CO2 occurred. The vapor pressure of water would increase at the slightly higher temperature, so potentially more water vapor would amplify the effect. This could cause more clouds, which could decrease or increase the effect depending on details. Also more CO2 could be driven out of the ocean due to the higher temperature, increasing the effect. Finely, ice could melt over large enough areas, so solar absorption could increase. The net effect of all these contributions is a temperature increase that may be higher (or possibly lower) that due to CO2 alone. The net increase from this doubling of CO2 is called the sensitivity of temperature to CO2 doubling.
There are three main factors I will examine here as relating to the CO2 sensitivity. They are the water vapor feedback effect, the CO2 from seawater feedback, and the melting ice feedback. The cloud effect and aerosols are not well enough defined, and are not considered here, but note that both are probably significant negative feedback causes.
Before discussing the three, I want to point out that since the adiabatic lapse rate is maintained even though there is some net radiation flux through the atmosphere, that whatever happens in the lower several km of atmosphere does not matter by itself. Only the change in effective location of outgoing radiation determines the surface temperature. Thus the only effective greenhouse gas concentrations that matter are the ones located at the altitude of outgoing radiation. The only net effect of adding absorbing gases is to move this effective location up a bit.
The first example is the so-called water vapor positive feedback effect. For this effect to be active, the absolute concentration of water vapor must increase along with the CO2 increase at higher altitudes. It has been clearly shown that this water vapor concentration does increase near the surface, but contrary to expectations, it not only did not increase, it slightly decreased in the upper troposphere and lower stratosphere in recent times. The limited data accuracy and short time for data do make this not the final word. However, this reduction at altitude rather than increase refutes any present positive water vapor feedback claim until contradicted by new data.
The second so called positive feedback is the addition of more CO2 to the atmosphere from seawater due to warming. However, there is also a claim the increasing CO2 dissolving in seawater is acidifying it. That is, the seawater is a sink, not source. You can’t have it both ways. It presently is a sink and will remain so for the future of excess CO2 production. This makes it a negative feedback.
The third so called positive feedback is exposure of more absorbing area to sunlight due to sea and land ice area decreasing from the warming. It is true that the Arctic sea ice and some lower elevation Greenland ice has reduced during summers, although black carbon seems to be at least partly to blame, and also the ice seems to be partially recovering from the recent melt. Also many smaller land glaciers have been reducing since the end of the LIA. However, the Antarctic land and sea ice has been generally increasing over most of this time. The total global ice coverage has only reduced by a very small amount over the last 50 years, and the exact amounts were not known accurately prior to that to determine the level. Since the high latitude ice is exposed to a low Sun angle at most, the ice area effect is not a significant factor in feedback for periods short as we are considering.
We have thus shown that the three main positive feedback terms are not doing what is required, and we know clouds and aerosols tend to reduce heating, so the effective sensitivity is very likely less than 1.2 C per doubling.
Long Term CO2 and Water Vapor Effects on Heating
It has been pointed out and emphasized that it is necessary for the sensitivity to have significant positive feedback in order to explain the temperature CO2 correlation that has been observed over Geological time scales. The three main sources of possible positive feedback are water vapor feedback, ocean degassing of additional CO2 with rising temperature, and ice melting, decreasing the effective albedo of Earth.
Since many of the periods in the last 600 My were essentially glacier and polar ice free, the change in ice is not a viable feedback for these periods. Since the CO2 level that actually was present is supposed to have been determined, the source, and possible ocean feedback on CO2 level can’t be isolated, so any ocean contribution has to be considered as part of the source, not a separate feedback. The only remaining positive feedback is increased water vapor that increased at higher altitudes. We know that at the present, even though CO2 has increased about 40% from 150 years ago, and the temperature has increased about 0.8 C over that time, that the upper atmosphere water vapor did not increase (although there are accuracy of data questions). This does not support the supposition of positive water vapor feedback.
However, a more direct question occurs. If a small temperature increase from any source occurs (changing solar heating, biomaterial changing albedo, very long term ocean current variation due to land mass movement, etc.), why would this not also cause a positive water vapor driven feedback. What is so special about CO2? One argument given is that CO2 has a much longer persistence in the atmosphere and thus is not directly coupled to temperature variation. This may be a factor for a period of a few years or even a few decades, but we are talking about Geological periods of Millions of years. It is pure nonsense to talk about the CO2 persistence effect at those time scales.