SCOPE 21 -The Major Biogeochemical Cycles and Their Interactions 

ABSTRACT

Analyses of the global carbon cycle based on oceanic models suggest that the earth's terrestrial ecosystems currently accumulate carbon. Analyses based on changes in the area of world vegetation show that terrestrial ecosystems are releasing carbon as a result of changes in the use of land. Both analyses might be correct and compatible if the biota and detritus of the unmanaged systems of the earth are accumulating carbon per unit area in proportion to the increase in atmospheric CO₂. This paper reviews the evidence for an increase in the amount of carbon per unit area in terrestrial and marine systems. Direct evidence is almost non-existent, although a significant global change might be so slight as to be undetectable for several decades or more. Indirect evidence is based: (1) on experiments showing that increased levels of CO₂ enhance productivity and nitrogen fixation; (2) on arguments that industrialization causes a eutrophication of terrestrial and marine systems through a release of nutrient elements; and (3) on arguments that changes in climate may affect the distribution of terrestrial ecosystems with different amounts of carbon. Each of these lines of evidence is based to some extent on the concept of limiting factors, and none of them is without counter arguments. Much of the current discussion of this topic has addressed the wrong question, however, for it is not what limits the fixation of carbon but what limits the storage of carbon that is important in the context of the global carbon question. The authors conclude that there is little evidence for an increased storage of carbon in the unmanaged ecosystems of the earth.

11.1 INTRODUCTION 

Analyses of the global carbon balance based on oceanographic models suggest a small net storage of terrestrial carbon in the range 01 x 10¹⁵g C annually (Bacastow and Keeling, 1973; Siegenthaler and Oeschger, 1978; Broecker et al., 1979) In contrast, recent analyses based on changes in the use of land world-wide indicate a net reduction in terrestrial carbon of the range 25 x 10¹⁵g annually (Moore et al., 1981; Houghton et al.,1983). Included in the latter analyses are the transformations of natural lands to agriculture, the harvest of forests, and the regrowth of forests following harvest or abandonment of agriculture. The results imply either that the ocean models are in error or that land-use changes are responsible for only a part of the net flux between terrestrial systems and the atmosphere. The possibility exists that there is an opposite flux, perhaps to undisturbed systems, responsible for a net storage of carbon in the earth's terrestrial biota, in soils, or in marine sediments. It is the purpose of this paper to evaluate the current evidence for such a storage.

The increased storage of carbon in the biota as a consequence of increased CO₂ in the atmosphere has been called the biotic growth factor or, loosely, the -factor (Bacastow and Keeling, 1973). In a broad context the biotic growth factor applies not only to the direct effect that increased CO₂ might have on plant growth but to any factor, correlated with increased CO₂ or increased industrial activity, that might increase the amount of carbon in the earth's biota, soils, or sediments. Such factors include the increased availability of nutrients and changes in climate. What is the probability that such factors are important? 

11.2 DIRECT ANALYSES

Direct measurement of the biotic growth factor would be very difficult. If the 26 x 10¹⁵g carbon needed to reconcile the two types of analyses described above were spread evenly over the land surface of the earth, where the vegetation contains 590 x 10¹⁵g C and the litter and soils contain another 16602060 x 10¹⁵g C (Atjay et al., 1979), the increase would amount to between 0.0008 and 0.002 of the standing stock of carbon (Table 11.1). The variability of biomass or of soil carbon per unit area is such that measurements in the field generally have a standard error of the order of 0.10 of the mean, and 100 years' accumulation might be required to observe an increased storage of carbon above background.

The problem appears insoluble. The change we are looking for can be, at once, both too small to observe and large enough to be important. The problem becomes more tractable if one seeks to document a change in the woody component of the vegetation or in the net flux of carbon to the biota annually (net primary productivity). The possibility of measuring such changes is about 50 more likely than with standing stocks (Table 11.1), but not with measurements of flux directly. The width of tree rings provides a means of measuring the production of woody tissues, and there is evidence from a few regions of the world of an increased rate of growth in recent years.

Table 11.1 A comparison of the fractions of carbon stocks and fluxes required to reconcile the oceanic and terrestrial carbon budgets

Records of growth rings in trees from central Europe show a slight, but statistically insignificant, increase in ring width in the last 115 years relative to a 2700-year average (Rebello and Wagener, 1980). Also, the width of growth rings in New Zealand trees indicates an increased rate of growth during the past century (John Ogden, personal communication). These results suggest a positive effect on diameter growth of trees, but results from other locations suggest a negative effect. For example, Whittaker et al. (1974) found that the productivity of hardwood forests in the northeastern U.S.A. had decreased 18% between the periods 19561960 and 19611965, and A. H. Johnson et al. (1981) found reduced rates of growth in the Pine Barrens of New Jersey, U.S.A. The results of the studies of tree rings are not consistent from place to place. They suggest that the biotic growth factor, if it operates in nature and can be measured in this way, is not universally positive. These analyses, however, address only a part of the question of carbon storage in terrestrial ecosystems. They do not address the possibility of increased litter production and changes in the organic matter of soils.

11.3 INDIRECT ANALYSES BASED ON LIMITING FACTORS 

11.3.1 The Concept of Limiting Factors

The question of whether or not the amount of organic matter on land is increasing is often addressed by a consideration of what limits the accumulation of carbon in terrestrial systems and whether those limiting factors are changing. The concept of limiting factors is an extension of the Law of the Minimum, attributed to Liebig but recognized by others prior to Liebig's formulation in 1840 (Browne, 1942). The Law of the Minimum states that if all of the mineral nutrients but one are available in the quantities required for the growth of a plant, the deficiency of that one nutrient will prevent growth. The concept of limiting factors in ecology has been broadened to consider not only plant growth and mineral nutrients but the decomposition of organic matter, the distribution of organisms, and the effects of temperature, moisture, and light, as well. In fact, all processes can be thought of as controlled by a limiting factor. The concept appears in many forms. In the paragraphs below we use the concept to consider whether the storage of carbon in the terrestrial biota and soils could be increasing currently.

11.3.2 The Effects of Increased CO₂ on Productivity

Evidence that the terrestrial biota and soils might be accumulating the carbon required to reconcile the analyses of the global carbon cycle is based on assumptions drawn from laboratory and greenhouse experiments. Increased concentrations of CO₂ in air increase photosynthesis and the growth of potted plants (for reviews see Lemon, 1977; Strain, 1978; Goudriaan and Ajtay, 1979; Kramer, 1981; Rosenberg, 1981). Some commercial greenhouses routinely use CO₂ to increase productivity. The conditions for growth are kept close to optimal in greenhouses: plants are watered, fertilized, and illuminated. Under such conditions it is not surprising that CO₂ can be shown to limit growth. The question is whether plants in natural ecosystems are limited by CO₂. The argument that they are not, and that increased concentrations of CO₂ will not enhance productivity, is based on the overwhelming evidence that factors other than CO₂ generally limit productivity. These other factors, such as light, water, temperature, or mineral nutrients, normally limit growth of plants before the concentration of CO₂ does. Large interannual variability in the primary productivity of ecosystems and in the width of tree rings makes it clear that factors other than CO₂ have measurable effects.

The counter-argument is that the Law of the Minimum is not strictly accurate. According to the Law of the Minimum, plant growth should be limited by the factor in least supply, no matter how abundant the other substances. In recent laboratory experiments, however, Gifford (1979a, b) showed that plants limited by water or light responded proportionately more to a CO₂ increase than plants not limited by those factors. The results seem inconsistent with the Law of the Minimum, but the Law is not strictly applicable here. Stomates partially closed to reduce water loss can conduct a greater amount of CO₂ into a leaf if the ambient concentration of CO₂ is elevated. This observation explains why increased levels of CO₂ increase the water-use efficiency of plants (the mass of water required to produce a unit mass of plant tissue) (Rosenberg, 1981). The point is that increased concentrations of CO₂ can enhance productivity even when another factor is limiting it.

A similar experiment where nitrogen was limiting showed the increase in growth with elevated CO₂ relative to growth in ambient air was similar under four different levels of nitrogen (Wong, 1979). The results from these experiments suggest that plants normally limited by water, light, or mineral nutrients may, nevertheless, respond to an increase in atmospheric CO₂; the plants need not be growing under optimal conditions.

Most of the experimental evidence is from short term studies in growth chambers, however, and its application to natural systems is unclear. While increased levels of CO₂ might be expected on the basis of molecular or biochemical evidence to increase photosynthesis, a hierarchy of other plant processes interact such that the increased photosynthesis does not necessarily result in increased growth or dry matter production (Evans, 1976; 1980; Lemon, 1977; Goudriaan and Ajtay, 1979). Kramer (1981), in a review of the effects of CO₂ on photosynthesis and dry matter production, stressed that the long term responses of plants to increased concentrations of CO₂ vary widely among species, and that rates of photosynthesis are limited by various other factors in addition to CO₂ concentration. He concluded:

Overall, there seems to be a tendency to overestimate the effects on vegetation of doubling the CO₂ concentration of the atmosphere. This is so because present knowledge of the effects of CO₂ concentration is based chiefly on short-term laboratory experiments and measurements of plant growth made in greenhouses where water and mineral nutrition are seldom limiting.

While it is unclear how large the direct effect of increased CO₂ might be on productivity, there are at least two ways in which the increased CO₂ concentration in the atmosphere might stimulate productivity indirectly by mobilizing other elements: one way is through the enhancement of nitrogen fixation; the other is through a decreased pH of precipitation and hence an increase in weathering rates with concomitant releases of P, S, or other elements from rock.

CO₂ enrichment has been observed to increase the fixation of nitrogen by soybeans (Hardy and Havelka, 1973, 1976). Presumably the elevated level of CO₂ decrease photorespiration so that more photosynthate is available to root nodules where N-fixation takes place. To the extent that increased CO₂ enhances N-fixation in natural systems, and we know of no evidence that it does, increased CO₂ may increase the availability of both C and N.

While CO₂ dissolved in water forms carbonic acid and lowers the pH of rain, the significance of the effect is small for two reasons. One is that carbonic acid is buffered, unlike either sulphuric or nitric acid, the major contributors to acidification of rain in many parts of the world (Johnson et al., 1972). The second reason why increased atmospheric CO₂ does not affect the weathering of rock significantly is that most weathering takes place in the soil atmosphere, where the concentration of CO₂ due to biological activity is many times the concentration in the atmosphere (Garrels and MacKenzie, 1971). At present the CO₂ in the atmosphere is about 20% above the pre-industrial value, and it seems unlikely that the availability of mineral nutrients has been affected through increased weathering by CO₂, although the other components of acid rain have probably done so (N. M. Johnson et al., 1972, 1981).

11.3.3 The Effects of Eutrophication on Productivity

The ratios of C:N:P by weight in the earth's biota and in industrial by-products have been used to calculate how much carbon could be stored in the biota or soils if one or another of the elements is assumed to be limiting to growth. Sulphur is not generally thought to limit growth in forests and is not considered here. The ratios of elements released from the combustion of fossil fuels bear little resemblance to the ratios in living plants; carbon is disproportionately abundant in fossil fuels. (Table 11.2). The ratios for marine plants are even less similar to the ratios for fossil fuels. When the amounts of N and P associated with industrial fixation and mining are added into the fossil fuel quantities, the elemental ratios of man's releases are enriched in N and P relative to C (Table 11.2). Making the assumption that all of the carbon, nitrogen, and phosphorus released by man's activities are distributed to the forests of the world in the ratios given in the last row of Table 11.2, a release of 5.2 x 10¹⁵g C from the combustion of fossil fuels in 1980 (Rotty, 1981) would result in an additional fixation of 5.2, 9.6, or 17.6 x 10¹⁵g C if C, N, or P, respectively, were limiting the fixation of carbon. Such enchanced productivity is unlikely to occur since most of the nitrogen and phosphorus mobilized is applied to agricultural lands and not available to stimulate growth in forests. The enhanced productivity of crops is not pertinent to this discussion because it does not represent a net storage of carbon; crops are generally consumed the same year they are grown.

Table 11.2 Weight ratios of elements in the biota of the earth and in the by-products of man's activities

Deevey (1970) was one of the first to recognize the potential for eutrophication of terrestrial ecosystems with the releases of N, P, and S from industrial activities. He used the review of world biomass by Rodin and Basilevich (1968) to compute whether the world's forests were in steady state or growing. By subtracting estimates of litter production and mortality from estimates of net primary production, Deevey determined that the biomass of forests was increasing, at an average rate of 2148 kg organic matter ha^-1 yr^-1. Given a world-wide area of forests of 48.5 x 10⁸ ha (Whittaker and Likens, 1973), this rate of carbon storage is equivalent to 4.7 x 10¹⁵g C yr^-1. The amount is enough to cancel the net release of carbon from land as a result of changes in land use. The value is high, however, because the production values given by Rodin and Basilevich included estimates for below-ground material while the litter and mortality values did not.

Using a more detailed analysis Melillo and Gosz (Chapter 6, this volume) concluded that the forests of the world might be storing as much as 0.3 x 10¹⁵g C yr^-1 as a result of eutrophication with nitrogen fixed during the combustion of fossil fuels. Melillo and Gosz assumed that 25% of all the nitrogen fixed was available to forests, and that nitrogen is the factor limiting forest growth. Similar calculations based on the release of phosphorus from the combustion of fossil fuels showed that the potential effect of phosphorus would be smaller than that of nitrogen.

The question of limiting factors in fresh water systems has been addressed in a symposium of the American Society of Limnology and Oceanography (Likens, 1972). While carbon may be limiting during short periods of time, the exchange of both CO₂ and N₂ across the airwater interface and the fixation of nitrogen occur readily enough that P is generally thought to limit productivity in lakes (Schindler and Fee, 1974; Schindler, 1977).

Peterson (1981) and Walsh et al. (1981) have independently addressed the possibility that the increased mobilization of nitrogen and phosphorus by man's activities has increased the concentrations of these nutrients in river run-off, enhanced the productivity of coastal waters, and led to an increase in the burial of carbon in marine sediments. Walsh et al. (1981) concluded that the current burial of carbon in slope sediments might be up to 0.75 x 10¹⁵g C yr^-1 greater than the pre-industrial burial. Peterson (1981) concluded that recent increases were unlikely to have exceeded 0.1 0.2 x 10¹⁵g C yr^-1. Peterson argued that, although coastal eutrophication might increase productivity locally, the major source of phosphorus and nitrogen to the euphotic zone is the upwelling of deep water, and that the amount of N and P mobilized by man's activities is small relative to the amount available for productivity from this deep water source. It seems unlikely that the eutrophication of aquatic systems has buried enough carbon to balance the equations of the carbon cycle.

On the other hand, if oceanic phytoplankton communities are sensitive either to subtle changes in elemental ratios, as lacustrine communities have been shown to be (Schindler et al., 1971), or to the addition of novel compounds (Greve and Parsons, 1977; Officer and Ryther, 1980), the dominance of diatoms and flagellates and the structure of marine food chains may have already changed. Such a change could alter the proportion of silicate-to carbonate-forming organisms, change the alkalinity of surface waters, and, hence, affect the ability of the oceans to absorb atmospheric CO₂. The effect would be similar to the dissolution of carbonate sediments, a process that has been assumed in most oceanic analyses not to be important in the time scale of tens of years.

Calculations with elemental ratios allow speculations of how much carbon might be stored if certain elements are assumed to limit production. The ratios are global averages, however, and no ecosystem or organism experiences a global average. Carbon in the form of CO₂ is distributed globally, but oxides of N and S tend to fall out of the atmosphere near their sources of origin, as isopleths of acid rain indicate. And phosphorus released from the combustion of fossil fuels has probably a local, concentrated distribution since it has no readily-produced gaseous form. Further studies of the response of the biota to releases of various elements would benefit from a regional approach. The appropriate scale is probably related to atmospheric transport processes and to the rates at which substances are removed from the atmosphere, either physically or chemically.

11.3.4 The Effects of Toxification on Productivity

Not all compounds released as a consequence of man's activities act as fertilizers. Numerous toxic substances are released along with C, N, P, and S in the combustion of fossil fuels. The combined effects of sulphur dioxide, ozone, and nitrogen oxides as well as of acid precipitation on agricultural production in the Ohio River Basin have recently been investigated (Loucks et al., 1980). These substances are apparently responsible for reductions in crop yields ranging from 6 to 14%. Such reductions have been demonstrated in other experiments (for example, Heggestad and Bennett, 1981). Acid rain is also thought to have been responsible for reducing the productivity of trees in the New Jersey Pine Barrens (A. H. Johnson et al., 1981) and, perhaps, in the hardwood forests of New England (Whittaker et al., 1974).

Another by-product of industrialization is the release of fine particulates and aerosols to the atmosphere. These have the potential to change the radiative climate of the earth as a whole or of particular regions through the absorption or scattering of solar radiation (Bolin and Charlson, 1976). A reduction in the solar radiation reaching the earth's surface has the potential to shorten growing seasons and to affect photosynthesis as well, especially if light is a limiting factor. Increased atmospheric turbidity does not necessarily decrease photosynthesis, however; diffuse light may enhance photosynthesis of leaves that are not light saturated (Lemon, 1977). Just how the factors that impair primary production will balance with those that enhance it is unclear.

11.3.5 The Effects of Increased Productivity on the Storage of Carbon

The majority of work on the question of carbon balance has been concerned with changes in photosynthesis, two steps removed from carbon storage. The point has been made that increased rates of photosynthesis do not correspond to increased rates of growth or dry matter production (Evans, 1976, 1980; Lemon, 1977; Goudriaan and Ajtay, 1979; Kramer, 1981); and it is equally important to recognize that increased productivity does not correspond to increased carbon storage. Storage of carbon results from the balance between the production of organic matter and its decomposition or oxidation. The question is not what limits the production of organic matter, or what limits its decomposition, but how the effects of man's activities alter the balance of these two processes. While a number of researchers have reviewed the possible effects of increased CO₂ on productivity, a similar review concerned with the effects of increased productivity on carbon storage has not been completed, although Melillo and Gosz (Chapter 6, this volume) outlined an approach to such a study.

Except for the atmospheric concentration of CO₂, most of the factors presented above as possibly limiting productivity can be shown in certain locations or times to limit decomposition as well. For example, the increased availability of nitrogen may not only increase the production of organic matter but may increase its rate of mineralization (Alexander, 1961). And, similarly, the toxic substances that reduce productivity are, qualitatively at least, likely to reduce decomposition (Strojan, 1978).

Thus the processes that produce fixed carbon may be cancelled by the processes that release it. Even if an increase in CO₂ enhances production but does not directly affect decomposition, the effect of increased litter production may alter the rate of decomposition indirectly. The question is no easier to resolve than the problem posed in section 11.2: the change we are looking for can be, at once, too small to observe and large enough to be important. Eventual resolution may rest with analyses of carbon isotopes in tree rings or air, or in interpretation of changes in the seasonal oscillation of CO₂ as recorded at Mauna Loa and elsewhere. Alternatively, if the annual release of carbon from the biota grows in proportion to fossil fuel use in the furture, or when the atmospheric concentration of CO₂ doubles or triples, resolution of the question will be possible. The fact that large releases from the biota are accompanied by a reduction in the area of forests will either reduce the potential for storage or make it more visible in the residual forests.

11.3.6 The Effects of Changes in Climate on the Storage of Carbon

There are several ways in which the increased concentration of CO₂ in the atmosphere can modify the climate and, hence, affect the amount of carbon stored in the biota and soils of the earth. The most well known of these is the greenhouse effect, or the warming of the earth's surface by the absorption of out-going terrestrial radiation by tropospheric CO₂. Current estimates are that a doubling of the pre-industrial CO₂ concentration will occur near the middle of the next century (WMO, 1981; see Chapter 1, Section 1.3.2B), and that the effect will be to warm the earth an average of 23°C (Manabe and Stouffer, 1979; Manabe and Wetherald, 1980). There is evidence that the warming trend has already begun (Hansen et al., 1981; Kukla and Gavin, 1981). The warming is expected to be greatest at the poles (68°C) and least at the equator (1^°C) (Manabe and Stouffer, 1979; Manabe and Wetherald, 1980). The question of how this warming will affect the amount of carbon stored in the terrestrial biota and soils of the earth, therefore, becomes crucial in boreal systems, where the temperature increase will be greatest, where temperature has a profound effect, and where a large amount of carbon is stored in peat.

The question was addressed at a recent conference (Miller, 1981), and the results from three lines of evidence were inconclusive. The latitudinal distribution of carbon in northern ecosystems seems to indicate that a warming should lead to a greater storage of carbon in the Arctic. Paleoclimatic studies also indicate a greater storage of carbon in forests during past interglacial periods. Analyses based on the temperature dependence of photosynthesis and respiration, however, suggest that for the entire ecosystem respiration would increase more than photosynthesis with warmer temperatures, and, hence, that there would be a net reduction in the amount of carbon stored in boreal ecosystems. This reduction seems especially likely in tundra since the depth of thaw and soil respiration are both temperature dependent. One of the interesting points to be drawn from the three approaches used at the conference is that the conclusions based on short term physiological considerations were opposite to those based on long term palynological evidence. In the context of the global carbon balance, the time scale of interest is tens and hundreds of years. Changes in the distribution of vegetation type are on a longer time scale.

The ability to predict changes in the storage of carbon in terrestrial ecosystems is still more speculative when one considers water as a limiting factor. It seems certain that a change in the average temperature of the earth and in the latitudinal distribution of sensible heat will affect the earth's climate in many ways, but exactly how the patterns of evaporation and precipitation will be changed over the different regions of the globe is difficult to specify. Regions that are currently arid or wet will not simply move poleward (Manabe and Wetherald, 1980). Middle and high latitude regions may experience drier summers, for example (Manabe et al., 1981). A discussion of how temperature, moisture, and other climatic factors are expected to change as a result of increased atmospheric CO₂ is beyond the scope of this paper. It is important to recognize, however, that changes in the global distribution of temperature and moisture are likely to have as great, or greater, effects on the storage of carbon on land than does the increased CO₂ itself. A crude calculation shows the potential effect.

The net primary production of terrestrial systems is about 60 x 10¹⁵g C yr^-1 (Ajtay et al., 1979). If carbon is neither stored nor released, this production must be balanced by a heterotrophic respiration of the same amount. If respiration has a Q₁₀ of about 2, an increase in temperature of only 1°C would contribute an additional 6 x 10¹⁵g C to the atmosphere annually. To the extent that photosynthesis has a similar Q₁₀, both production and respiration would be increased, carbon would cycle more rapidly, and the storage of carbon would not change. Although predictions of future climates based on current understanding of the ocean-land-atmosphere system are tentative, global analyses of the interactions of biogeochemical cycles will have to take climate into account.

11.4 DISCUSSION

Table 11.3 summarizes the potential that man's activities may have on the storage of carbon in the unmanaged biota and detritus of the earth. The ranges are guesses of what might be happening; there is little direct evidence that any change has occurred. The summary is of limited use in reconciling the analyses based on oceanic models with those based on the managed lands of the earth. The possible range includes the range required for agreement but it also extends beyond that range in either direction. Perhaps the only conclusions to be drawn from such a speculative summary are: (1) that the potential for change in the natural flows of carbon and other elements is large; (2) that if large changes had occurred they would be observable now; and (3) that the absence of large changes thus far suggests a stability maintained by mechanisms largely unknown. One wonders whether mechanisms exist at a global level, and whether research at any level less than a global level will be able to explain the apparent homeostasis of the planet (Lovelock, 1979).

Much of the research concerned with the possible enhancement of photosynthesis by increased atmospheric CO₂, for example, has centred on the wrong question. What matters is not what limits the fixation of carbon but what limits its storage. Carbon cannot be stored before it is fixed, but the controls of respiration and decomposition are as important as those of photosynthesis and growth. The balance between net primary productivity and heterotrophic respiration determines the accumulation of carbon in an ecosystem (Woodwell and Whittaker, 1968).

Table 11.3 The potential effects of various factors on the storage of carbon in the world's biota and detritus. Values reflect possible net changes in both productivity and decomposition. Negative values represent net losses of carbon from the biota, soils, or sediments

Short term responses are not adequate to predict long term responses. Fertilization experiments, for example, are performed routinely to determine the factor or combination of factors limiting productivity, the factor(s) that, when added, cause an increase in productivity. But what are the consequences of increased productivity to plants? The advantage of rapid growth under some situations seems clear: rapid growth in height may enable a plant to avoid being shaded by other plants. Under other situations, however, plants responding to fertilization with increased growth are killed by frost, high salinity, or some other stress that does not kill control plants. In fact, the observation is common enough that some workers describe the lack of a response to fertilizer as an adaptive response (Jefferies, 1977; Jefferies and Perkins, 1977) because a response would presumably have caused death. The effects of short term responses may turn out to be reversed in the long term.

The short term responses of communities or ecosystems to additions of mineral elements are also likely to be misleading, for the species that occur in an ecosystem are affected differentially by different factors. Phosphorus, for example, has frequently been shown to limit the productivity of lakes; the addition of phosphorus to a lake causes an increase in net primary production. The increased productivity, however, is often from a different assemblage of phytoplankton species from what existed prior to enrichment. Phosphorus was limiting productivity, but was it limiting the productivity in the original phytoplankton community? What else has been changed besides the productivity and the structure of the community? When the addition of phosphorus to a lake leads to eutrophication, it is frequently accompanied by an anaerobic hypolimnion, loss of fish, and changes in the cycling of carbon, oxygen, and sulphur as well as phosphorus. Experiments with whole lakes have shown the importance of subtle shifts in elemental ratios in causing large changes in the composition of phytoplankton populations (Schindler et al., 1971; Schindler, 1977). These findings, together with studies showing the differences in nutrient concentrations among forest species (Woodwell et al., 1975) suggest that the industrial release of elements in ratios only slightly different from natural ones may cause changes in the structure and function of other ecosystems.

Focusing on productivity, carbon storage, or any one process to the exclusion of others may miss other important changes. The functions of an ecosystem (for example, productivity, mineral cycling, and decomposition) are not independent of the kinds of organisms present. If ecosystems are important in regulating the cycles of C, N, P, and S, how important are species interactions in carrying out the function of various ecosystems?

The papers in this volume attest to the importance of the biota in the cycles of C, N, P, and S. The biota both affects and is affected by its environment. The reciprocal relationship implies that a change in one will lead to a change in the other. Man's activities are changing both. He is not only modifying the chemistry of the earth directly but is reducing the number of species and the area of natural ecosystems. Will these changes reduce the ability of the earth to support man? Answering the question is one of the most important tasks facing us today.

11.5 CONCLUSIONS 

There is little direct evidence that the storage of carbon in the earth's biota or detritus is increasing as a result of the increased concentration of CO₂ in the atmosphere or as a result of any other factor associated with industrial activity. What evidence there is is based on tree rings and balanced by equally sparse evidence showing that the storage of carbon has decreased. Indirect evidence based on laboratory experiments or on assumptions of limiting factors has shown the possibilities for a large increase in the storage of carbon but cannot demonstrate that such an increase has occurred. Equally plausible counter-arguments and experimental results suggest a net loss of carbon. The possibilities include large fluxes of carbon, but neither the magnitude nor even the direction is known currently (Table 11.3). Most of the experimental evidence is based on short term experiments that are insufficient for long term predictions. The telling analyses or experiments have yet to be carried out, and our understanding of the equation for the global carbon cycle is incomplete.

The emphasis thus far has been an analysis of what limits primary productivity, while it is the storage of carbon in ecosystems that is important. Even when the process appropriate for the global carbon issue is identified, however, it is a mistake to expect that changed productivity or storage is the only change. Changes in one process or function are likely to lead in a longer term to changes in structure and, concomitantly, to other changes in function. Predictions about biogeochemical cycles require not only a knowledge of the current system and its behaviour but also a knowledge of how tightly coupled a system's function is to its structure, and how changes in either structure or function might be expected to affect subsequent changes.

11.6 REFERENCES

Alexander, M. (1961) Introduction to Soil Microbiology, New York, Wiley.

Bolin, B., and Charlson, R. J. (1976) On the role of the tropospheric sulphur cycle in the shortwave radiative climate of the earth, Ambio, 5, 47-54.

Browne, C. A. (1942) Liebig and the law of the minimum, in Moulton, F. R. (ed.) Liebig and After Liebig: A Century of Progress in Agricultural Chemistry, Washington, D.C., AAAS, 71-82.

Delwiche, C. C., and Likens, G. E. (1977) Biological response to fossil fuel combustion products, in Stumm, W. (ed.) Global Chemical Cycles and their Alterations by Man, Berlin, Dahlem Konferenzen, 73-78.

Evans, L. T. (1980) The natural history of crop yield, Am. Scientist, 68, 388-397. 

Garrels, R. M., and Mackenzie, F. T. (1971) Evolution of Sedimentary Rocks, New York, Norton.

Gifford, R. M. (1979b) Growth and yield of CO₂-enriched wheat under water-limited conditions, Aust. J. Plant Physiol., 6, 367-378.

Greve, W., and Parsons, T. R. (1977) Photosynthesis and fish production: Hypothetical effects of climatic change and pollution, Helgolander wiss. Meeresunters, 30, 546-564.

Hardy, R. W. F., and Havelka, U. D. (1973) Symbiotic N₂ fixation: multifold enhancement of CO₂-enrichment of field-grown soyabeans, Plant Physiol., (Suppl.) 48, 35.

Heggestad, H. E., and Bennett, J. H. (1981) Photochemical oxidants potentiate yield losses in snap beans attributable to sulphur dioxide, Science, 213, 1008-1010. 

Houghton, R. A., Hobbie, J. E., Melillo, J. M., Moore, B., Peterson, B. J., Shaver, G. R., and Woodwell, G. M. (1983) Changes in the carbon content of terrestrial biota and soils between 1860 and 1980: a net release of CO₂ to the atmosphere Ecol. Monogr. (in press).

Jefferies, R. L., and Perkins, N. (1977) The effects on the vegetation of the additions of inorganic nutrients to salt marsh soils at Stiffkey, Norfolk, J. Ecol., 65, 867-882. 

Johnson, A. H., Siccama, T. G., Wang, D. R., Turner, S., and Barringer, T. H. (1981) Recent changes in patterns of tree growth rate in the New Jersey Pinelands: a possible effect of acid rain, J. Envir. Qual., 10, 427-430.

Johnson, N. M., Reynolds, R. C., and Likens, G. E. (1972) Atmospheric sulphur: Its effect on the chemical weathering of New England, Science, 177, 514-516.

Kukla, G., and Gavin, J. (1981) Summer ice and carbon dioxide, Science, 214, 497-503.

Likens, G. E. (ed.) (1972) Nutrients and Eutrophication: The Limiting-Nutrient Controversy, Lawrence, Kansas, Allen Press, Inc.

Lovelock, J. E. (1979) Gaia: A New Look at Life on Earth, Oxford, Oxford University Press.

Manabe, S., and Wetherald, R. T. (1980) On the distribution of climate change resulting from an increase in CO₂-content of the atmosphere, J. Atmos. Sci., 37, 99-118.

Manabe, S., Wetherald, R. T., and Stouffer, R. J. (1981) Summer dryness due to an increase of atmospheric CO₂ concentration, Climatic Change, 3, 347-385.

Miller, P. C. (ed.) (1981) Carbon Balance in Northern Ecosystems and the Potential Effect of Carbon Dioxide Induced Climatic Change, Report 15, Carbon Dioxide Effects Research and Assessment Program, Department of Energy, Washington, D.C.

Officer, C. B., and Ryther, J. H. (1980) The possible importance of silicon in marine eutrophication, Mar. Ecol. - Progress Series, 3, 83-91.

Rebello, A. de L., and Wagener, K. (1980) Growth trend in recent tree rings, in Jacoby, G. (ed.) Proceedings of the International Meeting on Stable Isotopes in Tree-Ring Research, Report 12, Carbon Dioxide Effects Research and Assessment Program, Department of Energy, Washington, D.C., 128-132.

Rodin, L. E., and Bazilevich, N. I. (1968) Production and Mineral Cycling in Terrestrial Vegetation, Fogg, G. E., translation editor, Edinburgh, Oliver and Boyd Ltd.

Rotty, R. M. (1981) Distribution and changes in industrial carbon dioxide production, in Papers presented at the WMO/ICSU/UNEP Scientific Conference on Analysis and Interpretation of Atmospheric CO₂ Data, World Climate Programme 14, Geneva, Switzerland, WMO, 123-133.

Schindler, D. W. (1977) Evolution of phosphorus limitation in lakes, Science, 195, 260-262.

Schindler, D. W., and Fee, E. J. (1974) Experimental Lakes Area: Whole-lake experiments in eutrophication, J. Fish. Res. Bd. Can., 31, 937-953.

Strain, B. R. (ed.) (1978) Report of the Workshop on Anticipated Plant Responses to Global Carbon Dioxide Enrichment, Duke University, Durham, North Carolina. 

Strojan, C. L. (1978) Forest leaf litter decomposition in the vicinity of a zinc smelter, Oecologia, 32, 203-212.

Whittaker, R. H., Bormann, F. H., Likens, G. E., and Siccama, T. G. (1974) The Hubbard Brook ecosystem study: Forest biomass and production, Ecol. Monogr., 44, 233-254.

WMO (1981) On the Assessment of the Role of CO₂ on Climate Variations and their Impact, Geneva, WMO.

Woodwell, G. M., and Whittaker, R. H. (1968) Primary production in terrestrial ecosystems, Amer. Zool., 8, 19-30.

Woodwell, G. M., Whittaker, R. H., and Houghton, R. A. (1975) Nutrient concentrations in plants in the Brookhaven oak-pine forest, Ecology, 56, 318-332.