Inaccessible

Figure.* 5,3. First-order model of ocean uptake of fossil fuel CX)>, The specific capacity of seawater to absorb excess CO2 is a fixed function of temperature and chemical composition. For today's thermocline and deep sea, the E(X)2 content increases h> about 0.5 /¿rnol/kg per ¿¿aim increase in pC'Oj. I Iowlvct, the uptake capacity drops as the CO» content of the atmosphere rises. Based on the distributions of natural radiocarbon and bomh radiocarbon and tritium, the volume of the ocean accessible to fossil fuel CO2 varies roughly as the square root of time. Based on the mean penetration depth of bomb l4G (i.e., 350 meters) one decade after the implementation of the ban on atmospheric testing, the penetration depth as a function mean age of fossil fuel molecules can be estimated.

of the thermo-haline circulation (deeper than 1750 meters) and the distribution of bomb UC (see Jiroccker ct aL, 1985) constrains the rate at which GO> is being stored in the thermocline (at depths less than 500 meters), we have as yet no strong constraint on the rate at which the ocean's intermediate waters (500—1750 meters) are being ventilated. Thus, we must either put our faith in the interpolation between the time scales for ventilation of the deep sea (centuries) and the time scales for ventilation of the thermocline (decades), or we must accept the as-yet somewhat shaky results of simulations made using ocean general circulation models (OGCMs) to estimate the storage in this all-critical mid-depth portion of the ocean. My impression is that we are still a long way from adequately modeling what goes on in this critical depth range.

When these two influences on the distribution of CO2 between the ocean and atmosphere are considered together, for most emission scenarios, the reduction in capacity will roughly balance the expected increase in the penetration depth of CO?. Thus, the current roughly 2 to 1 split between air and sea is likely to apply, at least tor the next several decades.

In light of the large number of calculations carried out in OGCMs, one might conclude that the simplified calculations of the ocean presented here arc unnecessary. However, one must keep in mind that although the models have been adjusted to yield reasonably good representations of the distributions of both natural and bomb radiocarbon, there currentlv is no way to confirm that their modes of ventilation of the intermediate depth range ocean arc correct. Inasmuch as this reservoir will become

Maiepi/ta/î, 3aLHwmeHHbiiï aBTopcKUM npaBOM

Figure 5.4. Estimates of the delivery of excess CO> to the deep sea via thermo-haline ventilation. The assumptions are (I) that 30 Sverdrups (i.e., 30 x IIV" m3 /sec) of water descend to the deep sea from the polar outcrops, (2) that these descending waters have achieved CO2 partial pressure close to that in the atmosphere (see Tahle 5.1), and f.i) that on the time scale of a century, little oft he excess CO}-chargcd deep water is recirculated. Because the rate of deep-sea ventilation may decrease as the globe warms and because new ly formed deep waters generally do not achieve equilibrium with the overlying atmosphere, the results of this calculation are likely to be upper limits.

Figure 5.4. Estimates of the delivery of excess CO> to the deep sea via thermo-haline ventilation. The assumptions are (I) that 30 Sverdrups (i.e., 30 x IIV" m3 /sec) of water descend to the deep sea from the polar outcrops, (2) that these descending waters have achieved CO2 partial pressure close to that in the atmosphere (see Tahle 5.1), and f.i) that on the time scale of a century, little oft he excess CO}-chargcd deep water is recirculated. Because the rate of deep-sea ventilation may decrease as the globe warms and because new ly formed deep waters generally do not achieve equilibrium with the overlying atmosphere, the results of this calculation are likely to be upper limits.

ever more important during the course of the next century, the current model-based projections could well be flawed.

One aspect of the entry of fossil fuel CO? into the ocean can be usefully isolated from the rest. Deep waters formed in the northern Atlantic and in the Southern Ocean carry excess CO? into the deep sea. Because of the very long residence time of deep water in the ocean, over the course of the next century little of the CO? carried away from the surface in this way will return to the atmosphere. Based on an estimate of 30 Sverdrups for the average deep water ventilation flux (Broecker ct al., 1998) and using the information listed in Table 5.1, a rough estimate of the amounts of CO2 delivered to the deep can be made (see Figure 5.4).

One other factor must be considered. To the extent that the planet warms, the rate of ventilation of the oceans interior is likely to decrease, reducing the fraction of uptake by the sea. I will return to this point.

5.3 Atmosphere-Terrestrial Biosphere Partitioning

The processes governing the partitioning of CO2 between the atmosphere and 1 he terrestrial biosphere are complex and not well understood, Most carbon cycle specialists agree that during the 1980s the terrestrial carbon inventory remained nearly constant, that is, greening roughly matched forest cutting. But, before 1989 there were no O2/N? measurements, so this conclusion is perhaps open to challenge. Monitoring of the rate of O2 decline (Keeling et aL, 1996) makes clear that during the early 1990s greening outstripped forest cutting. The net uptake of CO2 by the biosphere during this period balanced roughly 30% of the production of CO? by fossil fuel burning. Based on the geographic distribution of excess CO* in the atmosphere, most of this storage appears to have occurred in the North Temperate Zone. In a recent paper, based on a deconvolution of atmospheric CO? data Fan et al, (1998) suggest that the excess storage occurred mainly

Maiepwa/i, 3auiMiueHHbift asiopcKUM npaBpM

in North America. However, other treatments of the same data set do not support this conclusion.

What drives this excess storage? I;our possibilities exist:

1. Unusually favorable growth conditions during the early 1990s allowed photosynthesis to temporarily outstrip respiration.

2. Regrowth of forests on land previously used for farming (this regrowth is confined mainly to temperate North America and Europe).

3. Fertilization by the excess atmospheric CO2. (Because CO2 fertilization is global, to explain the observation that storage has occurred mainly in the North Temperate Zone, one would have to conclude that deforestation in the Tropics has roughly balanced the CO? growth enhancement.)

4. Fertilization with fixed nitrogen released from automobiles and farms (these releases have occurred mainly in temperate North America and Europe).

With regard to future expectations for biospheric storage, it makes a significant difference which combination of these drivers is responsible. Most écologiste attribute enhanced storage to some combination ofCO2 and fixed nitrogen fertilization. This explanation is consistent with a north temperate sink but not with dominance by a sink in North America. Rather, the storage should be more nearly equally split between Eurasia and North America. Direct estimates of forest regrowth suggest that it can account for no more than 20% of the required excess storage. The fact that a dramatic increase in the excess storage term appears to have commenced at the end of the 1980s suggests a météorologie explanation. If so, then according to the analysis by Fan et al. (1998), this impact was centered over North America, If indeed the explanation is meteorological, then these deconvolutions of atmospheric CO2 data do not provide us with the information we need if we are to predict the century-long changes in terrestrial storage.

My guess is that fixed nitrogen is a more important driver than most écologiste believ e it to be. Because the release and deposition of fixed nitrogen occurs mainly at the north temperate latitudes, this would fit the constraint based on the distribution of excess CO> in the atmosphere, I am suspicious that by thinking in terms of terrestrial "Redfield" ratios, we may be underestimating the power of nitrogen forcing. An analogy to an auto factory is useful in this regard. Fixed nitrogen may be more akin to assembly lines than to fenders. Thus, by increasing the supply of fixed nitrogen to forests, we are doing more than permitting carbon sequestration calculated as some set multiple of the number of moles of nitrogen. Rather, we are strengthening the apparatus that produces wood and humus. If so, we are elevating the steady-state carbon storage to a new plateau. But, as has been pointed out, the extent of this elevation has limits. Once the nitrogen supply saturates, as it has in most European forests, other ingredients (light, water, soil-derived nutrients, etc.) will become limiting. Hence, further addition of fixed nitrogen will no longer be effective.

All estimates of future storage are subject to question because of the potential influence of global warming on the humus inventory in soils. Everyone agrees that the warmer the temperature, the shorter the lifetime of soil humic compounds. But because both the magnitude of the expected global warming and the coefficient relating

Maiepnan, 3amwL4eHHbi(?i aBTopcKUM npaBOM

temperature and humus turnover times are uncertain, it is not clear to me whether, taken together, greening and warming will lead to a net increase or a net decrease in the carbon inventory oí soils.

So what can Ik said about the evolution of biosphcric storage over the next century? Can any useful limits be set? With regard to CO? fertilization, let us assume that CO2 growth enhancement can increase wood stocks by no more than yfPa"/yjp™- Were CO2 concentration in air to rise to 650 ppm by the year 2100, then the maximum increase in standing biomass w ould be 5(H) (^650/360 — 1), or 170 GtC. Of course, if there is a significant synergistic impact of fixed nitrogen, this number could be larger But, on the other hand, the temptation to harvest tropical forests and the impact of rising soil temperatures loom large. Thus, I find it hard to believe that at the end of the next century storage in the biosphere will have increased by more than 200 Gt, Indeed the increase may even prove to be much smaller than this.

5.4 Ocean Research

It is my opinion that estimates of CO» uptake using existing OGCMs are no more reliable than that obtained from the tracer calibrated box-diffusion model of Siegenthaler and Oesehgcr (1978). One has only to compare the plots of excess CO? versus bomb l4C obtained by the existing models to realize that modelers are a long way from producing simulations that adequately replicate thermocline ventilation, let alone intermediate water formation. Because this latter reservoir will assume ever greater importance in the uptake of CO? during the course of the next century, there is a way to go before estimates of future ocean uptake more accurate than ±20% can Ik obtained. Hence, my agenda for improvement of our ability to predict the uptake by the ocean of excess atmospheric CO? would have as its core the improvement of ocean general circulation models. Key to this exercise is the development of models that are capable of duplicating not only the 3D distribution of density but also the spatial and temporal evolution of the transient tracer distributions (i.e., 3H, 'He, l4C, CFCs, etc.). Of course, to strengthen the constraints placed by tracers, we must periodically update the global distributions of these substances. As time passes, the residence time of the bomb-produced tracers approaches that of the fossil fuel CO> molecules. Hence, the World Ocean Circulation Experiment (WOCE) surveys conducted during 1990 (i.e., 30 years after the major atmospheric H-bomb tests were conducted) will greatly strengthen the constraints placed by the earlier Geochemical Ocean Sections Study (GEOSECS) surveys (1972-1977).

This approach w ill be adequate only as long as the ocean operates in its preindustriaf state. As global warming progresses, there will be a tendency for the ventilation of the ocean's interior to slow. The reason is that both the warming and the "wetting" of the temperate latitude outcrops from which the thermocline and intermediate waters descend will reduce the density of these waters. We need a set of observations that will indicate that this slowdown is under way. This will prove to be a difficult task because tracers provide integral measures of ocean operation, and as a result ihejr distributions are not particularly sensitive to gradual changes in ventilation rate.

/laiepnan, 3amwmeHHbiti asTopcKi/iM npaBOM

Keeling et al. (19%) have suggested a window into this problem. It involves the measurement of the ratios of Oz to N:2 and Ar to N? in the atmosphere. The secular trend and hemispheric seasonality in the O? to N? ratio carries information regarding the global plant productivity. The secular trend and seasonal swings in atmospheric COj can be used to constrain the terrestrial contribution to global plant productivity. The residual O2 change is then a measure of the strength of the oceanic life cycles. Were ventilation of the thermocline to weaken, fewer nutrients would be brought to the surface and presumably the warm-season production of O2 would drop. By keeping track of the secular trend and the amplitude of the annual O2/N2 swing, one can perhaps obtain a measure of the strength of thermocline ventilation.

The Ar to N2 ratio in the air changes as the result of breathing by the ocean. For each calorie of heat taken up by the sea, both Ar and N2 are released to the atmosphere. The important point is that the ratio of Ar to N2 in the gas so released is not the same as that in the atmosphere. Hence, both the secular warming of the ocean and the seasonal heat exchange between ocean and atmosphere are recorded by the Ar to N2 ratio in the atmosphere. By monitoring the secular trend in this ratio for the globe as a whole, one can obtain a measure of the rate of warming of the ocean. Were ventilation to slow; the rate of heat uptake would also slow. Seasonal swings in Ar/N? will also be of interest. The more swamp-like the upper ocean becomes, the larger should be the seasonal cycle in its surface temperature and hence in the hemispheric Ar to N2 ratio.

Many of those who do research on the global carbon cycle push for surveys of surface water pCX>2 and upper-ocean EC(X> as a means of establishing the magnitude of an ocean's uptake. I deem both of these strategies to be flawed. The problem with the ApC02 (air-sea) is that the accuracy required to yield an uptake estimate accurate to 10 percent is ±1 /zatm. Not only would the seasonal swing in pC02 have to be accurately established everywhere in the world ocean, but also corrections would have to be made for the sea-to-land transport of CO? supplying the carbon carried by rivers to the sea and for the AT between the ocean's "skin" and the interior of its mixed layer. Estimates suggest that together these two effects may have maintained a prcindustrial ocean mixed-layer CO? partial pressure averaging as much as 4 /zatm higher than the atmosphere. If these corrections are neglected, the uptake of excess CO2 by the ocean would be underestimated bv 1 GtC/vr. I doubt whether sufficient accuracy can ever be achieved by this approach, and even if it could, the cost would be enormous.

This is not to say that surveys of surface ocean pCO? are not useful. Indeed, those who seek to deconvolve terrestrial sources and sinks from the distribution of CO? in the atmosphere must have this information in order to correct for the oceanic distribution of CO2 sources and sinks. This task is less demanding; the level of required accuracy is easily attainable.

The ICO2 approach has more promise. A repeat of the WOCE survey 20 years from now will allow the integral uptake of CO? over this period to be accurately assessed. Furthermore, the 31) distribution of the ZCOi excess w ill provide yet another powerful constraint on ocean circulation models. In combination with the distributions of bomb l4Cand the CFCs, these results will allow the influence of differing atmospheric equilibration times (—10 years for HC, —0.5 years for CO?, and —0.1 year for CFCs)

Maiepwan, 3amwmeHHbiPi aBTopcKWM npaBOM

to he addressed in model simulations. Hut, by itself, this integral uptake estimate has no predictive power. We must, as I've already said, emphasize model development.

5.5 Terrestrial Research

As already stated, developing the means to predict how storage of carbon on the continents will evolve is a very tall order - so tall that reliable predictions may prove impossible. In the absence of predictive power, the best we can do is to document the magnitude of the annual changes in storage and to better constrain the locations at which this storage is occurring. The key measurements to he made are those that document the secular trends in the atmospheric CO2 content and in O? to N? ratio and their interhemispheric gradients. To make the most effective use of this atmospheric data, it is important that vertical profiles of CO? be obtained. Modeling studies (Denning et al., 1996a,b) clearly show that without such data, rectification effects related to the dynamics of the atmospheric boundary layer can lead to significant biases in the interhemispheric gradient. In this regard, it is essential to quantify all the transports that generated gradients in atmospheric CO? content before the onset of fossil fuel burning. If these gradients are not properly isolated, they could easily lead to false conclusions regarding current terrestrial sources and sinks for fossil fuel CO2-There is one aspect of this problem with which I am quite familiar. It has to do with the interhemispheric transport of CO? via the Atlantic Ocean's conveyor circulation (Broecker and Peng, 1992). Peng and I estimated that this transport is on the order of 0.6 GtC per year (see Figure 5.5). Keeling and Peng (1995) confirmed the importance of this transport mode and demonstrated that it affects the distribution of O? as well. To date no ocean model has duplicated the strong natural uptake of CO2 that Broecker and Peng demonstrated to be occurring in the northern Atlantic. Hence, this phenomenon is not included in inversions of atmospheric CO2 distributions. To evaluate the possible impact of this northern Atlantic sink, Fan ct aL (1998) carried out simulations that suggest that were the uptake of CO? by the northern Atlantic to be increased by 0.6 GtC/yr, then the model-derived terrestrial storage in North America would drop by

Much emphasis is being placed on eddy correlation-based CO2 fluxes determined using towers (Wofsy et al,, 1993), For the first time, it is possible to estimate the net uptake (summer) and release (rest of year) of CO2 by forests, Although this is a major advance, one must keep in mind that the terrestrial biosphere is a crazy quilt consisting of thousands upon thousands of small patches, each with its own vegetation history. A majority of sites are likely to be storing carbon, for releases are in part catastrophic because of fire, blow down, or cutting. Only rarely has a true steady state been achieved. Thus, although the tower studies will help to elucidate the factors influencing photosynthesis and respiration, taken alone they are not likely to yield the answers we seek in our quest to predict carbon storage.

In my estimation, if we are to develop the needed predictability, then we will have to carry out a host of purposeful manipulations involving both fixed nitrogen and CO2. For nitrogen alone, purposeful fertilization of checkerboard plots within natural forests

Maiepwan, 3ammneHHbiM asiopcKUM npaBOM

Guide to Alternative Fuels

Guide to Alternative Fuels

Your Alternative Fuel Solution for Saving Money, Reducing Oil Dependency, and Helping the Planet. Ethanol is an alternative to gasoline. The use of ethanol has been demonstrated to reduce greenhouse emissions slightly as compared to gasoline. Through this ebook, you are going to learn what you will need to know why choosing an alternative fuel may benefit you and your future.

Get My Free Ebook


Post a comment