Continuing to look at the effects of carbon dioxide

One of the useful things about blogs is that it allows debates to occur in front of the public, and thus to allow us to learn more about the underlying science on which predictions are made. Thus it appears that there is a focus beginning to develop on debating some of the facts that relate to the actual mechanisms that cause “the greenhouse effect.” In my post last week I tried to provide some underlying information, not expecting that there would be a couple of more detailed posts on the subject this week.

At the beginning note that what is being called “the greenhouse effect” doesn’t actually follow the greenhouse model, since the world is not capped with a glass ceiling, but otherwise the equivalent effect does occur. Bear in mind, however, that the mechanism is a radiative absorption of energy by water vapor, carbon dioxide, and methane and its transfer, that gives the warming of the lower atmosphere, but that this warmer air can also be carried up higher into the atmosphere and radiated outwards (the glass traps the warm air in the greenhouse).

So to go to the two posts of interest. Early in the week Richard Lindzen, who is a Professor of Meteorology at MIT, put up, what has turned out to be a controversial post at WUWT trying to explain negative climate feedback. He begins with a simple explanation of energy absorption along the lines that I and Dr Spencer both gave, and notes that doubling the CO2 concentration alone, there is enough uncaptured energy that was escaping in the relevant wavelengths to increase the atmospheric temperature by about 1 degree Centigrade. However carbon dioxide does not exist by itself in the atmosphere, there are also considerably greater quantities of water vapor, and somewhat lesser amounts of methane, both of which also act to absorb radiation in different wavelengths. Water vapor itself having a broad spectrum over which it captures energy.


So, to simplify the argument: if the increased concentration of carbon dioxide in the atmosphere were to cause the temperature to go up, and this were, in turn, to increase the amount of water vapor in the atmosphere, then the temperature would go up, not just because of the increase in CO2, the one degree rise for doubling, but also due to the increased amount of water vapor in the air, and this is projected by the IPCC models to cause an increase that can vary between 1.5 degrees and 5 degrees. Thus one can reach the increased temperature levels that one gets that produce the catastrophic forecasts that have been prevalent over the past decade. This increase in water vapor is called a positive feedback since the change positively helps the effect of the carbon dioxide increase.

The initial point that Professor Lindzen makes is that the increase in water vapor is as a result of an increase in temperature (i.e. it is a secondary effect) and the increase in carbon dioxide, does not directly change the amount of water vapor in the air. Now the change in the atmospheric condition does not, in itself generate heat, what the above description provides is how both carbon dioxide and water vapor inhibit (as they increase in concentration) the transmission of energy from the Earth back out into space. And thus, as their concentration increases, so the amount of energy that the Earth is radiating out into space should go down. (While the argument is simple, there are a number of conditions that make simple measurements not that easy to judge since the presence of volcanic dust in the atmosphere from eruptions such as we just saw from Mt Redoubt, and the much greater one which occurred in 1991 from Mt Pinatubo can cause confounding effects).

Now it turns out that there was a satellite monitoring the Earth’s radiation over the years from 1985 to 1999, and this information had been published in a paper in Science (Wielicki, B.A., T. Wong, et al, 2002: Evidence for large decadal variability in the tropical mean radiative energy budget. Science, 295, 841-844.) I am borrowing the relevant part of that graph from the post:



In the plot the long-wave emission from the Earth is monitored by the Earth Radiation Budget Satellite (ERBS) and compared with model predictions. And it is appropriate to note here that the authors of the paper have, in earlier work, used some of this data to argue against some of the earlier theories (known as the Iris hypothesis) that Professor Lindzen has proposed, in regard to cloud effects. That debate has continued and at this stage tends to overcomplicate this current discussion – but I mention it just to point out that these are not totally disinterested discussions that just happened to pop up.

OK, to get back to the plot. You can see that the models and the data follow one another fairly well until 1992. (The models were initially tuned to the data until 1989). Now there was an increase in temperature from 1975 through 1998, but given the increase in carbon dioxide and water vapor generated, this should, if the models are correct, create a reduction in the amount of heat radiated, by a factor of 2-3 over that if there were no feedbacks. However, as you look at the graph you can see that the readings showed that the amount of radiation went up, instead of down.

Thus instead of the water vapor enhancing the trapping of radiation with increase in temperature, (a positive feedback) in fact the amount of radiation trapped went down (a negative feedback). This, argues Professor Lindzen, shows that the climate models that anticipate a positive feedback are wrong.

The post was answered over on the Climate Change website, where it was pointed out that there was a problem with the satellite. Apparently over the time of the readings the satellite was gradually losing altitude. Since the amount of radiation measured varies as the inverse square of the distance, the change in altitude (about 25 km over the initial height of 611 km) is apparently sufficient to explain the change in data, and when the correction is made, then there is no discrepancy between the models and the actual result.



I am presuming that the drop in radiation in 1991 occurred right after the Pinatubo eruption and continued to affect the data through 1993. (In essence the curve pivots around 1989 to remove the steady increasing slope to the Lindzen curve and put the data back along the zero anomaly line).

There has been no response as yet from Dr Lindzen, though he was aware of this change when he wrote a presentation on Climate Change that he gave at Colgate in November 2007, when he said

Satellite studies of outgoing long wave radiation by numerous groups show that OLR corresponding to warm periods is much greater than inferred from climate models. This implies negative rather than positive feedbacks. Studies of the rate of heat uptake by the ocean suggest tight coupling of the atmosphere and ocean which demands low sensitivity (Schwartz, 2007). Observational studies show the presence of negative feedbacks that current models fail to replicate (Lindzen, Chou and Hou, 2001, Spencer, Braswell, Christy and Hnilo, 2007)

This ‘anomaly’ is quantitatively greater than what one would expect from the iris effect (that is to say, represents a larger negative feedback)(Chou and Lindzen (2002) Comments on “Tropical convection and the energy balance of the top of the atmosphere.” J. Climate, 15, 2566-2570.), but, except for the last paper, all attempted to argue a different origin for the observation. The last showed that the alternative explanations were inconsistent with existing models.

Recently, Wong et al (Wong, Wielicki et al, 2006, Reexamination of the Observed Decadal Variability of the Earth Radiation Budget Using Altitude-Corrected ERBE/ERBS Nonscanner WFOV Data, J. Clim., 19, 4028-4040) have reassessed their data to reduce the magnitude of the anomaly, but the remaining anomaly still represents a substantial negative feedback, and there is reason to question the new adjustments. For example, a more recent examination of the same datasets explicitly confirms the iris relations at least for intraseasonal time scales (Spencer, R.W., W.D. Braswell, J.R. Christy and J. Hnilo, 2007, Cloud and radiation budget changes associated with the tropical intraseasonal oscillations, Geophys. Res. Ltrs.)

Aside on the Iris Effect:
In the tropical half of the earth (30S-30N), heat and moisture transport from the surface to the atmosphere is dominated by cumulonimbus convection. The convective elements consist in deep, rapidly ascending cumulus towers. Although these towers occupy a relatively small area, the ice detrained from these towers is responsible for the extensive cirrus decks in the tropics, and the evaporation of precipitation from these decks is the major source of water vapor for the tropical atmosphere. Moreover, both the water vapor and the cirrus decks are very powerful greenhouse substances.

The amount of ice available to form cirrus and humidify the atmosphere depends on the efficiency of precipitation formation within the towers. The more efficiently precipitation forms in the towers, the less ice is available for producing cirrus. It has long been known that precipitation efficiency increases with temperature. Thus, warming leads to contracting cirrus coverage while cooling leads to expanded cirrus. This effect, which is called the Iris Effect (by analogy with the behavior of the eye’s iris), powerfully resists changes of temperature. In other words, it provides a strong negative feedback.

So what does all this say? Well the initial argument by Professor Lindzen seems significantly weakened by the change in the data. Since it is not clear what remaining negative feedback he is referring to following the correction of the data, it is hard to continue to accept the argument that he makes. But then I may have missed something (certainly there is a lot to this argument that one can follow if one goes off into the comments that talk about it at the various sites I found exploring the topic) yet I get the impression that there is sufficiently more to the story that I will have to go and read more before coming back to write on this again. It is not as simple as presented.