The AP program to control the level of CO2 in the atmosphere is based on known technology to remove the dilute gas as biomass and convert it by anaerobic digestion to forms that can be segregated to keep it out; a large amount of methane, CH4, also results. It has been assumed that the biomass areas calculated to establish this control actually are able to do so, based on a statement by Lindzen in 1990 that 4 million tonnes (4.4 x 106 tons) of carbon emission avoidance was needed to stabilize CO2 level at the then-current emission rate; we have had no direct proof that this will really happen. Now we have identified a fundamental study in the literature which can be used to demonstrate that the AP concept will work, that the areas calculated are correct, and that the process can be monitored in real time.
This work was initiated by the late C.D. Keeling in the 1950's. It sought to measure the CO2 level of the atmosphere in a spot far from major emissions atop Mauna Loa in Hawaii. Measurement has been continuous since 1958, resulting in the famous saw-tooth curve shown here as Figure 1; selected for its large scale, this plot comes from Carl Sagan's book Billions & Billions. It is apparent from the general shape of the curve, as expected, that CO2 content is rising, and at an increasing rate. What astonishes here is this annual saw-tooth cycle, explained by the authors as due to deciduous trees, primarily in the northern hemisphere, going in and out of leaf.
Each spring and summer these trees generate a lot of extra biomass in the form of new foliage; this takes a corresponding amount of CO2 out of the air, shown by Keeling's data as a down-stroke in the curve in real time. In autumn and winter, the leaves fall and start returning their CO2, resulting in the annual up-stroke. Comparing the amplitudes of these swings in Figure 1 in the '90s with the '60s, one suspects that this feature is increasing in time. The scale of this curve is sufficient to allow peak and valley ordinates to be measured, and from them annual amplitudes and average levels; this has been done, with results listed in Table 1. The work continues, and we have included some more recent data in Table 1. Note that there is indeed a substantial rise in CO2 level of 43.4 ppm in the first 36 years, about 1.2 ppm/year, rising to 1.36 ppm/year over the full 48 years. Amplitude has increased from 5 ppm at first to about 8 ppm at the end. Let us see if we can get more out of this.
The level of CO2 measured is the result of all the global influences on this value, positive and negative; leafing may produce the saw-tooths observed, but other factors affect the amplitude in more subtle ways. For instance, equilibrium between atmosphere and oceans is under constant adjustment in the response to CO2 concentration in the air, water temperature and other parameters; volcanic eruptions can discharge CO2 directly or as undersea events. These are occasionally of historic magnitude, as was the Krakatoa eruption of the late 19th century. Nothing of this intensity appears to exist in the record now, but those who work more closely with such matters may be able to spot similar events if and when they occur; the record may be very useful then. Some recent undersea discharges in the Pacific could be of sufficient size to be discernible on this scale, and analysis in the near future might detect their effects.
AP can be applied to these data to show its effect as if it were done in practice; results are shown in Figure 2. Curve 1 is the average value shown in Table 1; it indicates that CO2 level is rising at a generally increasing rate, as expected, producing concave-upward shape with noticeable annual irregularities. To apply a correction, note that the values increase by 65.2 ppm over 48 years, or 1.36 ppm/year. If we were able to take out 1.36 ppm/yr and keep it out, we would have the numbers plotted as Curve 2; more realistically, assuming we might build up to this degree of removal over a span of 20 years, we would have the numbers plotted as Curve 3. These curves become horizontal at the times for which they are calculated; the AP concept works. The degree to which AP is applied could change in practice annually as CO2 emissions change; this is practical with real-time management.
In the upper region of Figure 3, amplitude (Table 1) is plotted against time, with annual readings on the left portion of the plot and average values on the right. A straight line can be struck through these points, corresponding to an annual increase in amplitude of 0.0458 ppm. This should logically result from more trees, more leaves per tree, larger leaves in general, or combinations of these factors, which appear generally to be increasing with time. The amplitude of these annual swings should be affected by the CO2 level if they are indeed due to the photosynthetic activity of deciduous trees; as CO2 level rises, trees should make more or bigger leaves, and more trees might even be encouraged to grow, all of which should tend to produce larger swings. This basic relationship is tested in the lower region of Figure 3, where amplitude of the annual swings is plotted against the average CO2 level; again, annual readings are on the left portion of the plot and averaged values on the right. A straight line can also be drawn here, showing an amplitude rise of 2.2 ppm as the average CO2 level rises 65.2 ppm, or 0.0337 ppm increase in amplitude for each ppm rise in CO2 level. To test the consistency of these relations of amplitude with time and CO2 level, divide 0.0458 ppm amplitude increase/year by 0.0337 ppm increase in CO2 level and find that CO2 level increase at the rate of 1.36 ppm/year. Compare this with the data from Table 1, 65.2ppm/48 years, or 1.36 ppm/year. This indicates that we have not distorted the data by manipulation.
The reason for featuring these two plots of amplitude versus time and CO2 level in Figure 3 is to show the degree of annual variation that exists. The plot of average CO2 level versus time in Figure 2 displays some annual irregularity, but the scale minimizes the effect; the amplitude plots of Figure 3 show large swings for virtually every year. It is also apparent that even modest degrees of averaging will remove many of these short-cycle swings (compare the right ends of these plots with the lefts). People who are familiar with unusual patterns in the oceans and the atmosphere, volcanic and other discharges of CO2, and natural disasters such as forest fires, etc., may be able to match these events with the record expressed in these curves. It seems rather certain that when a major disruptive event occurs, such as the eruption of a large volcano, this record will provide a base line to evaluate ecological damage as well as real-time measure of the progress of recovery.
What we show here is that the Mauna Loa data provide a background to which the removal and segregation of CO2 as AP proposes can be applied. When this is done, the concentration of CO2 in the atmosphere levels out as predicted by AP. Beyond this, we demonstrate that annual swings in CO2 level can be studied by examining their amplitude, revealing the effects of other influences in such a way that they may be identified.
