Search This Blog

Friday, December 31, 2010

Have we recovered yet (Pt 3)

This is the third and final part of my critique of Syun-ichi Akasofu's recent article.  In the first part I showed that temperature variations in the 20th century can not be accurately described as a "recovery" from the Little Ice Age; while in the second part I showed that no consistent assignment of sensitivity to solar forcing can plausibly explain both the recovery from the LIA and the late 20th century warming, even with the aid of oceanic oscillations.  Together, the two posts refute Akasofu's central claims.

What was missing from those two posts was an analysis of greenhouse forcing over the period since the LIA, a lack which will be made up for in this post.  Climate Change "sceptics" nearly always leave out from their accounts the fact that they assume, either tacitly or directly, that the climate sensitivity of solar (or other natural forcings) have a much greater effect on global temperatures than equivalent anthropogenic forcings.  This is actually a crucial fact.  If a scientist wants to assume for their theory that one watt of warming from a natural source will heat the Earth ten times as much as one watt from an anthropogenic source, they are entitled to do so.  But it is an extraordinary claim.  It is a claim that, should they want to make it, they should be up front about, and explicitly defend.

So long as they are not up front about such claims, then they are no longer doing science.  They are instead practising rhetoric.  Instead of presenting you with their claims, and hoping to persuade you by logic and evidence; they are concealing from you the major part of their claims, and hoping to persuade you by keeping you ignorant of what their claims actually entail.

So, what do Akasofu's claims entail about the relative sensitivity of natural and anthropogenic forcings?

As in part 2, we can determine the solar forcings on which Akasofu's theory relies over three benchmark periods:  1620 - 1850;  1850 - 1940; and 1940 - 2004.  Again using Lean, 2000, we find the change in TSI for these periods were respectively 1.5, 1, and 0.05 w/m^2.  However, the TSI is the amount of energy absorbed by a totally absorbing plate held perpendicular to the Sun's rays at the distance of the Earth's orbit.  It is not the amount of energy absorbed from the Sun by the Earth's surface at any given time or place.  That amount depends on the albedo, on the latitude and longitude, and on the time of day and year.  Fortunately, we do not need to calculate all these factors.  Instead we can rely on the fact that the area of a circle with a given diameter is 1/4th of the area of a sphere with the same diameter, and on the fact that the average albedo of the Earth is 0.7.  Consequently, the solar forcing, averaged over the Earth's surface  is 0.7 times TSI/4; or for our three time periods, the change in solar forcing is, respectively  0.265, 0.175, and 0.00875 w/m^2.

To find an anthropogenic forcing for comparison, I will only use CO2.  This is for convenience only, as CO2 only represents about 60% of anthropogenic greenhouse forcings.  Consequently I will be underestimating the relative strengths of natural to anthropogenic climate sensitivities assumed by Akasofu.

In 1620, there were 272.3 ppm of CO2 in the atmosphere, determined from ice cores at Taylor Dome in Antarctica.  In 1850 there were 286.8 ppm; in 1940 there was 310.5; while in 2004 there was 374.7 (as measured at Cape Grim).  The forcing from CO2 concentrations is 5.35 * ln(CO2/initial CO2).  From this we can work out the forcings for our respective periods as being   0.28,  0.42, 1 w/m^2 respectively.

These figures will come as a surprise to most people.  If we assume equal climate sensitivity to CO2 and solar forcings (the null hypothesis, for those who like those sorts of things), then by a reasonable estimate, CO2 forcing accounts for 50% or the temperature rise between the lowest point of the LIA and 1850.  Once methane and NO2 are taken into account, green house gases probably account for more than half of that warming.

The reason this will surprise most people is because of two false assumptions.  The first is that CO2 forcings respond linearly to CO2 concentrations.  But of course, CO2 forcings are a function of the natural log of relative concentrations.  Consequently, the smaller early increases in CO2 have a greater relative importance than intuition suggests.

The second is that CO2 levels were more or less constant prior to the industrial era.  That is simply not true, with variations of up to 14 ppm occurring in the period from 1 to 1750 AD.  What is more, much of this variation was due to human activity.  Approximately half of the variation between 1620 and 1850 was because of human activity, primarily from changes in land use and deforestation; but also from the burning of coal, which by 1650 had already turned the streets of London black from use as a domestic heater.   The other 50% of the increase in CO2 concentrations comes from out gassing from the warming ocean over that period.

The relative strength of CO2 to solar forcings becomes very apparent between 1850 and 1940, with approximately 70% of the forcing coming from CO2.  Of course, even over this period, changes in the Sun's radiation was a dominant forcing agent.  This is partly because it had a much larger variability than GHG forcings, but also because GHG forcings came to be matched by an approximately equal negative forcing from industrial emissions of aerosols, particularly sulfur dioxide.  Of course, in the later half of the twentieth century, anthropogenic forcings dominate, with the sun providing only about 1/100th of the change in forcing provided by CO2.

Of course, this is the way an AGW proponent would interpret the data.  Akasofu, on the other, would have you believe the sun was the dominant forcing, even in the later half of the twentieth century.  Presumably then, he believes the solar forcing times the solar climate sensitivity is at least equal to CO2 forcing times CO2 climate sensitivity over that period.  But that is only possible if the solar climate sensitivity is at least a hundred times larger than CO2 climate sensitivity.  But this mammoth difference, though necessary to Akasofu's argument, goes unmentioned, unscrutinized, and undefended.  It is passed over in silence, either because Akasofu has not thought out the implications of his position, or because silence is the only hope he has of defending it.

(Return to Part 1)
(Return to Part 2)

For further reading:
What ended the LIA, at Skeptical Science
A detailed look at the Little Ice Age, at Skeptical Science


  1. Tom,
    You use the assumption of direct solar insolation as the basis for you rejection of strong solar effects on Earth's temperature. That is not what many skeptics use as a basis for the solar effect. The solar activity has two other effects. The change in UV is a much larger change than in insolation. This affects Ozone production, which has a large effect. Also, solar activity is associated with solar magnetic field intensity, which affects cosmic ray intensity to Earth, and this has been shown to affect cloud activity. These two factors have a larger effect that just insolation alone, but are generally related to the insolation level, so the analysis is not as simple as you indicate.

  2. Leonard, I think it is straight forward that any UV not absorbed in the upper atmosphere will reach the Earth's surface in approximately the same proportions as visible radiation, so that a change in the proportion of UV reaching the surface is not the basis of a change in solar climate sensitivity. On the other hand, energy from UV absorbed in the upper atmosphere would appear to lower, not raise the solar climate sensitivity.

    This may not be immediately apparent, but that energy is redistributed by radiation. Approximately 2/3rds of it is radiated as IR by CO2 (and to a much lesser extent, H2O and CH4), and about 1/3rd of it as IR from ozone. Of that radiated by ozone, about half is radiated out to space, and the remaining sixth is radiated to the surface because of the lack of significant ozone concentrations in the troposphere. Of that radiated by CO2, half is again radiated out to space, and the remaining third is radiated into the upper troposphere where it is absorbed by CO2 at the level of effective radiation. Consequently, only half of any UV radiation absorbed in the upper atmosphere contributes to increased temperatures on the surface/troposphere.

    That being the case, increasing the proportion of UV radiation in incoming radiation increases the proportion of that radiation that does not contribute to surface heating. In other words, while an increased proportion of UV may well effect the sensitivity of solar radiation, it woud do so by reducing it, not increasing it.

    As to Svenmark's hypothesis, the evidence of the correlation between cosmic rays and clouds is spotty at best. What is more, the assumption that cosmic rays are necessary to the formation of cloud condensation nuclei is dubious. Far more likely is the possibility that it will increase the number of cloud condensation nuclei from a very large number derive from other sources. If it does that, then increased cosmic rays will not cause clouds to form that otherwise wouldn't, but it will cause clouds to have a greater albedo, and to last longer (because they produce less precipitation). That would produce an increased sensitivity for solar forcings, but no where near large enough an increase to sustain the idea that the late twentieth century warming is primarilly driven by the sun.

  3. Tom,
    Your response on UV sounded reasonable, but I don't think we have enough information to conclude anything on the cosmic ray connection. However, I want you to respond to the question: If the temperature has gone up and down and as fast by as much or more several times during the Holocene, and was likely higher by as much as a degree or more, that the present, what do you claim were possible causes? These are not explainable by CO2 or human activity.

  4. Tom,
    Ignore my last post here. I see you already responded on a different entry, and I did not see it before I made the above comment.

  5. Thanks for this critique of the Syun-Ichi Akasofu paper. I am also more philosopher than climate scientist, but found your posts to be coherent and convincing.

    I'm writing to say thanks, but also to make sure I hear about it if and when someone publishes a rebuttal to this paper in the literature. Make sure you let us know if/when that occurs.