SCOPE 21 -The Major Biogeochemical Cycles and Their Interactions  

ABSTRACT

When dissolved in the water, the chemical elements are transported over large areas by the ocean circulation. When incorporated into organic matter, they enter the food chain, settle, and are released at levels having a different circulation. The quantitative estimation of their net transport requires evaluation of these physical and chemical processes over a large range of time and space scales. The long timespace scale cycle deduced from the steady state distribution of tracers is the best known, but it is not adequate to explain the transient evolution of tracers introduced during the last decades by the industrial activity of man.

17.1 INTRODUCTION

Most of the elements mobilized by the industrial activity of man will eventually come to the oceans. Of fundamental importance are the elements involved in the composition of all living organisms: carbon, sulphur, nitrogen, and phosphorus.

In their oxidized and soluble forms, carbon (HCO₃ ) and sulphur (SO₄^2-) already exist in vast quantities in the ocean. The oceans contain more than 98% of the mobilized carbon at the Earth's surface. The partition of this carbon between the ocean and the atmosphere is a function of the pH and total inorganic carbon in the ocean surface layer. These are in turn, strongly controlled by the metabolic activity of plants and animals.

Phosphorus and nitrogen are the nutrient elements limiting production in the ocean. Their mobilization leads to a chain of reactions in which additional organic matter is synthesized, consumed, transported to the deep ocean (scavenging other elements on its way), and finally either dissolved or sedimented and buried. Nitrogen, unlike sulphur, is not most abundant in its stable and oxidized anionic form in the ocean. That our atmosphere is composed mainly of nitrogen gas is due to the denitrifying activity of the biota that keep just the amount of nitrate in the ocean necessary to use all the phosphate available.

In the sea, as on land, the biota are at the crossroads of the geochemical cycles of all these elements. Their influence is not only in producing a stoichiometry between its elemental components but, more important, the biota control the state of oxidation, mobilization and transport mechanisms of elements.

While dissolved in the water, the elements move with the water circulation which transports them mostly in an horizontal direction. When they are incorporated into organic matter first as living and later as dead particulate matter, they move vertically through the water. As they are regenerated by metabolism and bacterial activity, they are released into other patterns of circulation moving in different directions. The net transport of a given element is very much a function of the time that specific element spends in particulate and dissolved form.

Here, I will review some aspects of these two transport mechanisms. The physical and chemical processes can best be discussed under the framework of the tracer distribution equation. Conservation of mass leads to the mathematical model to determine the distribution of a tracer. Using the eddy diffusion hypothesis the tracer field is given by

where C is the time-averaged tracer concentration, V the mean velocity field, K the eddy diffusion tensor, a first-order decay constant for radioactive tracers, and J the net local rate of change of non-conservative tracers.

Accordingly, I will divide the processes under discussion in to two groups: circulation and mixing representing the physical processes in the left-hand side of equation (1); sources and sinks representing the chemical processes in the right-hand side of equation (1).

17.2 CIRCULATION AND MIXING 

The first problem that faces us in analysing the relevant physical processes for the distribution of an element in the ocean is one of time scales. The measured variability in the velocity and the tracer field shows a continuous spectrum of frequencies. The lower band of the spectrum, periods of more than a few years, can only be guessed.

For modelling purposes we like to think in terms of a large scale, steady state, `general ocean circulation' and mean tracer fields on which turbulent perturbations are imposed in a more or less random way, producing on the average an effect similar to diffusion. This is the basic assumption of the eddy diffusion hypothesis behind equation (1).

For this formalism to have any practical value, V and K in equation (1) should be independent of the tracer in question. But when equation (1) is applied to tracers with vastly different half-lives we cannot expect the same processes and the same time averaging to apply to both tracers. While Rn222 (t_1/2 = 3.825 days) conserves the memory of only 2 weeks, C-14 (t_1/2 = 5600 years) integrates a history of thousands of years. As a result, determination of vertical eddy diffusion coefficients in the abyssal ocean with stable tracers and C-14 usually gives values that are one or two order of magnitude smaller than the ones calculated with Rn-222 or Ra-228 (t_1/2 = 5.75 years).

The same problem with time scales confronts us when modelling the evolution of the oceanic distribution of C-14. We have the pre-industrial distribution of C-14, the decrease due to burning of fossil fuel CO₂ (the Suess effect) and the increase due to nuclear weapons testing. Three different time scales are involved, and the relevant parameterization is not necessarily the same for the three processes. On a short time scale of 10 years, the tritium distribution has shown that the important factors are: an early invasion of the sub-tropical surface water, the convective overturning of sub-arctic water, advective transport by the anticyclonic gyres in the surface layers of the sub-tropics and mixing by mesoscale eddies (scale about 100 km) along density surfaces (isopycnal). Jenkins (1979) successfully modelled the evolution of tritium in the thermocline of the Sargasso Sea by assuming a ventilation of isopycnal layers by winter water formation. In the Pacific Ocean, after an early penetration to about 300 m at GEOSECS St.347, the invasion into the thermocline has been very slow (Broecker and Peng, 1980). The current pattern is still revealed by strong tritium gradients within the gyres, and between the gyres and the equatorial region (Fine and Östlund, 1980; Fine et al., 1981). On a time scale of 30 years we can expect a diffusive penetration of the thermocline and cross basin exchange through the equatorial region. On a scale of hundreds of years, the exchange of surface with deep waters becomes the dominant factor.

The distinction between advection and eddy diffusion in short time scales is important because advection can carry tracers against the mean gradient while eddy diffusion cannot. This is particularly relevant for modelling the flushing of tracers produced during nuclear weapons testing, since there is a substantial latitudinal variation in surface values.

By necessity, most geochemical models use a simplified picture of the circulation. The tendency has been to use a few boxes (internally well-mixed) exchanging with each other by first-order coefficients. This approach brings additional problems: How can we simulate the ocean circulation with a few boxes? How can we choose the transfer coefficients to represent adequately the fluxes across the boundaries?

Figure 17.1 Division of the ocean into three layers based on a model by Jenkins (1979). is the renewal time of the water in isopycnal layers calculated with ³He or ³H. Reproduced by permission of W. Jenkins

The simplest description that can be made of the ocean is one of three layers. A warm surface layer (T > 15°, _t < 26.4), a cold deep layer (T < 5°, _t  > 27.4) and the layer in between (see Figure 17.1).

The surface layer occupies the central area of the basins between the sub-tropical convergences (40°N. to 40°S.). Its depth ranges from about 300 m in the centre of the gyres to less than 200 m at the equator. This layer is directly driven by the wind and rotates in anticyclonic gyres on both hemispheres. Mean velocities are of the order of 1 to 10 cm s^-1. The upper 50 m are well mixed and depleted of nutrients. The base of this layer has a seasonal thermocline, is renewed by winter convection and feeds the upwelling zones at the east sides of the basins and equator. The exchange of this layer with the atmosphere is rapid. Tritium and helium ages are less than 1 year. Conditions in this layer are highly variable in a short time scale. For long time scales it could perhaps be considered a well mixed reservoir.

The intermediate layer reaches to a depth of about 1000 m, lies under the upper layer and contacts the atmosphere between the sub-tropical convergences and the polar fronts (50°60^°). The residence time for exchange with the atmosphere increases with depth and is of the order of decades. It is the region of the permanent thermocline, main pycnocline and salinity minimum. It has strong eddy activity, and isopycnal advection and diffusion seem to be the predominant phenomena. In spite of several models of the thermocline, the exact dynamics of this layer are poorly known. Geostrophy indicates that the layer moves with the anticyclonic gyres and the horizontal velocity decreases rapidly with depth. Most of the baroclinic adjustment of the pressure field occurs in this layer. The thickening of the layer between isopycnals towards the equator can be due to vertical convergence from the deep layer below. In this case it is not difficult to visualize a uniformly distributed upwelling from the abyssal layer being channelled to the upwelling regions near the equator and the periphery of the gyres. This is one question that the distribution of tracers can help to elucidate.

The abyssal layer is driven by the pressure field created by the wind driven circulation and the vertical motion created by the downward buoyancy flux induced by mixing and convective processes associated with water sinking at high latitudes. Typical horizontal velocities are of the order of 10^-2 to 10^-1 cm s^-1, except in the western boundary currents. The movement is slow enough to enable the development of substantial horizontal gradients. Here the Atlantic and Pacific show considerable differences in properties. In the Atlantic the deep water is formed by sinking of surface water south of the Labrador and East Greenland Seas mixed with waters from the Norwegian and Mediterranean Seas. Due to its recent origin, the Atlantic deep water is rich in oxygen and C-14 and poor in silicate and nutrients. It loses C-14 southward by radioactive decay and mixing with the Circumpolar Water of the Southern Sea. In the Pacific and Indian Oceans there is no surface formation of deep water. The deep water in these oceans is a mixture of intermediate water and upwelled bottom water. The C-14 decreases northward from the Circumpolar Current and attains a minimum at a depth of 2.5 km in the northeastern region of the Pacific. In both oceans the bottom water at the base of this layer is of Antarctic origin.

Recent modelling of the C-14 distribution in the Pacific indicates some important characteristics of the thermohaline circulation. Figures 17.2, 17.3 and 17.4 show the comparison between the GEOSECS data and the solutions of a 3D model of the Deep Pacific under two sets of parameters. The upper sections are the observed data from Östlund and Stuiver (1980). The middle sections are from the solution of the model with a linear increase of the upwelling velocity from zero at the bottom to a value of 3 m yr^-1 at 1 km. The lower sections are from a solution in which the upwelling attains a maximum of 2.6 m yr^-1 at 3 km and decreases to a value of 1 m yr^-1 at 1 km. For both solutions, the horizontal and vertical eddy diffusion coefficients are 5 x 10⁶ cm² s^-1 and 0.6 cm² s^-1 respectively.

Figure 17.2 Comparison between the observed distribution of ¹⁴C in the western Pacific (from Östlund and Stuiver, 1980) and the western boundary solutions of a 3D model (from Fiadeiro, 1982). The middle section is from the solution with a uniform flow in the Deep and Bottom Water. The lower section is from the solution with the Deep Water flowing opposite to the Bottom Water. Reproduced by permission of `Radiocarbon'

The solutions show that to model even qualitatively the C-14 distribution, the vertical velocity has to increase from the bottom up to a maximum and then decrease upwards to the base of the thermocline to a value approximately half the maximum.

Figure 17.3 Comparison between the observed distribution of ¹⁴C in the north Pacific and the sections at 35°N. of the 3D model solutions. Reproduced by permission of `Radiocarbon

Figure 17.4 Comparison between the observed distribution of ¹⁴C in the eastern Pacific and the sections 60° E. of the WB of the 3D model solutions. Reproduced by permission of `Radiocarbon'

The water balance of the last solution provides good insight into the magnitude of the exchanges. The Circumpolar water feeds 12 Sv (1 Sv = 10⁶ m³ s^-1) of bottom water at the southern end of the western boundary current; 8 Sv upwell to form deep water and 4 Sv return to the Circumpolar water. Of the 8 Sv, 3 upwell to the thermocline layer while 5 return south. The deep water circulation forces an additional recirculation of 2.5 Sv between the Pacific and the Circumpolar. Therefore, of 14.5 Sv of water entering and leaving the basin only 3 Sv will eventually cross the thermocline to the upper layers of the ocean. The residence time of the abyssal layer with this flow is approximately 700 years.

It is clear that most of the exchange of the abyssal water in the interior of the basins is with the Southern Ocean rather than with the upper layers above. The Southern Ocean is a big reservoir connecting the three oceanic basins where the water becomes mixed by strong shears in the Circumpolar Current. The water is very homogeneous in its characteristics, vertically and zonally. It has an almost constant value of C-14 = 160 ‰. This water comes in contact with the atmosphere south of the Polar Front providing a direct exchange between the abyssal layer and the atmosphere. The view of the thermohaline circulation as a continuous path of deep water from the North Atlantic to the farthest end of the North Pacific is deceptive. It conceals the importance of the Southern Ocean as an independent motor of the thermohaline circulation. The heat diffused from the upper layer into the deep water in the Atlantic as it travels southward is lost near Antarctica, not in the North Pacific. A more correct picture of the Southern Ocean is that of a well mixed reservoir, exchanging with the atmosphere and the deep water in the three basins.

The most detailed model of the circulation in studies of the atmospheric CO₂ uptake is by Björkström (1979) who used 12 boxes for the ocean: a cold and a warm surface box, two for the intermediate water and eight for the deep water. The surface boxes are 75 m deep and the warm surface has twice the area of the cold surface box. The thermohaline circulation is modelled by a net flow from the cold surface to each of the deep ocean boxes, through the intermediate layers to the warm surface box and back to the cold surface box. Two cases were considered by fitting a C-14 profile: a'young ocean' with a net flow of 44 Sv and an `old ocean' with a net flow of 25 Sv. An equal amount of water is exchanged between the two surface boxes. The flow from the cold surface to the eight deep ocean boxes decreases with depth from a value of 283 to 71 x 10<sup>12</sup> m<sup>3</sup> yr<sup>-1</sup> in the `young ocean' case. Thus, both w (the vertical component of the velocity) and dw/dz increase upwards. The intermediate boxes do not receive any direct flow from the cold surface water, but exchange water with this box by mixing. Three values were considered for this exchange. They correspond to a residence time of water in the thermocline of 3000, 300 and 30 years.

The percentage of the excess CO₂ in the ocean for the three cases was: 10. 9(8.4), 13.1(11.3) and 21.4(20.4) for the `young (old) ocean'. Regardless of the circulation, the model shows that the most crucial parameter for the oceanic uptake of CO₂ is the least resistant path between the atmosphere and the thermocline reservoir. As in an electrical circuit, if the deeper layers are connected to the atmosphere in parallel with the surface layer, rather than in series, the total resistance of the system diminishes considerably. Given this fact, the exchange of the atmosphere with the thermocline and deep ocean should receive the utmost attention in modelling the evolution of tracers on time scales greater than a few decades.

The first order coefficients of box models can be calculated from the advection and diffusion fields when the boxes are small (Keeling and Bolin, 1967). In fact, this is one way to put the tracer equation in finite difference form for numerical integration (Fiadeiro and Craig, 1978). But when the boxes are large, the relationship between the boundary fluxes and the mean concentration in the boxes loses significance. With two boxes, not even a distinction between advection and diffusion can be made.

Box models assume a very characteristic time response. A signal is transmitted infinitely fast and the box performs a special kind of Laplace transform on the input signal (Keeling, 1973). Unless the input can be considered approximately constant in one residence time of the box, the output will be considerably distorted. This puts a practical limit on the size of the boxes to be used with a particular transient tracer. Although a box model can quantitatively reproduce the results of a steady state distribution, its time response can be completely erroneous.

There is also an historical prespective that must be taken into account in the evolution of transient tracers. Water mass formation is a process dependent upon climatic conditions and it changes from year to year. The initial spreading of an element is strongly controlled by the geographical distribution of its input and prevailing climatic conditions.

17.3 SOURCES AND SINKS

Elements that enter in the composition of the biota show a depletion in surface water relative to the deep waters. Nitrate, phosphate and silicate are virtually stripped from the top of the warm surface layer. Carbon and calcium, two other elements mobilized by the biota but that exist in large excess in the ocean, show also a gradient corresponding to the nutrient depletion.

These gradients in concentration are maintained by a downward flux of particulate matter from the euphotic zone to the abyssal ocean, where part of this material is converted back to dissolved inorganic compounds by the organic metabolism of bathypelagic fauna and microbial activity and part is incorporated into the sediments.

The quantitative evaluation of this flux and its change with depth has been the aim of many geochemists in the last few years. Different approaches have been used to estimate this flux: (1) direct measurement by sediment traps; (2) measurement of the concentration and size of the suspended particles with a model for the falling rates; and (3) measurement or calculation of the sources and sinks of the constituent elements either directly or by modelling. Unfortunately, estimates by these methods are sometimes difficult to reconcile.

Sediment trap results have been reported by many workers (Wiebe et al., 1976; Soutar et al., 1977; Honjo, 1978; Spencer et al., 1978; Hinga et al., 1979; Rowe and Gardner, 1979; Knauer et al., 1979; Honjo, 1980; Brewer et al., 1980; Deuser and Ross, 1980; and others). The variability in the results has been astonishing. While Hinga et al. (1979) and Rowe and Gardner (1979) collected a mean flux of 90 g m^-2 yr^-1, the PARFLUX traps (Honjo, 1980, Brewer et al., 1980) collected less than 7.5 g m^-2 yr^-1; their maximum was less than 25 g m^-2 yr^-1 at 390  m. in the equatorial region. Of two identical traps set side by side at 5363 m, one collected twice as much as the other (Honjo, 1978).

There seems to be more difference in the collecting efficiency of traps than can be explained by their geometry (Gardner, 1980). Many other factors can contribute to the difference observed between traps. Deployment time has varied from a few days to many months. Fragmentation, coagulation and flocculation of particles above and inside the trap can occur in long deployments. Big traps collect a high percentage of small particles with modal peaks at 36 µm. These particles should settle only very slowly and it is surprising that so many are caught. The traps probably catch particles suspended or flowing with the currents. A fact without proper explanation is the three-fold increase in the clay fraction from 1 to 5 km in the PARFLUX traps, even when the fluxes are standardized with Th-230 (Brewer et al., 1980). Investigations of the seasonal variation of the fluxes and also of the effect of lateral transport are under way (Brewer, personal communication). At the moment it is impossible to take the fluxes measured by traps at face value. Anyway, it will be very difficult to estimate regeneration rates from the difference between fluxes, because the particles that contribute most for the regeneration flux are small particles that fall slowly. Particles with large settling velocities have no time to dissolve.

A few facts are nevertheless evident from the experiments made so far. The traps reflect the qualitative distribution of productivity at the surface. The same may be said about the sediments but the fractions of organic matter, carbonates, clay and opal are different. Also, a sizeable fraction of the flux seems to reach the very deep layers of the water column. Thus, some metabolism must occur at the sedimentwater interface, and the water column regeneration would be expected to change from place to place. However, the distribution of the dissolved tracers: oxygen, nitrate, phosphate, silicate, inorganic carbon, and alkalinity give little evidence of these changes in the input function. Only rarely does a thin layer near the bottom reflect any anomaly that cannot be explained by water mass characteristics. The region under the equator shows no unusual features despite a doubling in the flux rate (there is a possible exception for the alkalinity in the Pacific).

Determination of fluxes from the size distribution of particles and settling velocities has been tried by Lal and Lerman (1973), Lerman and Lal (1977), and Bishop et al. (1977, 1978, 1980). Again, great variability has been found under regions of high and low productivity. Most of the particles do not fall below 1000 m; more than 95% of the organic matter is regenerated in the top 200 to 400 m. In oligotrophic areas the recycling is even more intense. Silica and CaCO₃ are also intensely recycled in the warm surface layer. Calculated values of the fluxes at 1500 m in the Panama Basin gave 10, 2 and 6 mM m^-2 yr^-1 for organic carbon, CaCO₃ and silica respectively. A comparison between the calculated flux at 1500 m and a sediment trap at 2600 m in the same region and deployed at the same time for 234 days showed that the trap collected 45 times more material (Bishop et al., 1980).

Metabolism in the water column has been estimated by measuring the electron transport activity (Packard, 1969; Packard et al., 1977). Within the scatter of the data, the oxygen consumption rates are compatible with the results of the model presented below. Direct measurement of the oxygen and nutrient fluxes at the sedimentwater interface have been made by Smith and Teal (1973), Smith (1974, 1978) Smith et al. (1978, 1979) and Hinga et al. (1979). Although the oxygen fluxes at any one depth show substantial variation, the general trend of a decrease with depth is similar to the oxygen consumption deduced by Fiadeiro and Craig (1978). Nitrate and phosphate fluxes are very erratic, do not correlate with oxygen, and sometimes have opposite directions in similar conditions. The silicate fluxes have been compatible with laboratory experiments and model results.

Model determination of sources and sinks give only an averaged quantitative estimate of the true regeneration rates. It is not possible to determine the exact horizontal variation because the effects are averaged over large areas. The general trend of the vertical variation can be deduced but it is also a mean trend and depends on the assumed vertical velocity profile. The integrated fluxes are however well determined because they represent the overall balance between inputs and outputs. Once the time scale of the circulation is determined with radioactive tracers the input necessary to produce a certain distribution becomes well established.

The value of the regeneration rates below 1 km given by the 3D model of the Pacific (Fiadeiro and Craig, 1978; Fiadeiro, 1980) in mmoles m^-2 yr^-1 are (with a 20% uncertainty): oxygen, 290; organic carbon, 230; nitrate, 27; phosphate, 2; silicate, 240; and carbonate, 150.

While the regeneration of organic matter clearly decreases with depth (halving in 1.3 km), the change in dissolution rate for silicate and carbonate is not evident. There is an indication that the P:N regeneration ratio increases with depth while the SiO₂:CaCO₃ ratio remains constant.

The model solutions are quite sensitive to a flux from the bottom, either from the sediment or the sediment water interface, so a reasonable estimation can be made of the importance of this source. The bottom flux of silicate cannot be greater than 20 mmoles m^-2 yr^-1. Reasonable solutions are obtained with a bottom carbonate flux of up to 5 mmoles m^-2 yr^-1, and a flux greater than 10 generates a maximum at the bottom. Trying to limit the dissolution domain to depths below 3 km produces a sharp gradient in alkalinity near the bottom in the North Pacific. This is a strong indication that most of the dissolution of silica and calcium carbonate occurs in the whole water column.

In view of the differences in saturation and dissolution kinetics, the similarity between the silica and alkalinity source distribution and vertical profiles suggests that the dissolution of the particles is due to some organic metabolism and proceeds regardless of the state of saturation of the water. An input ratio of CO₂:CO₃^2- equal to 1.77 in the deep water and 0.94 in the bottom water (Fiadeiro, 1980), adequately explains why the bottom water conserves its pH while the deep water becomes more acidic. However, I do not know of any model that can explain the alkalinity distribution with a dissolution rate based on the saturation state of the water and the inorganic kinetics measured in the laboratory (Morse, 1978; Keir, 1979).

Evidence has been accumulating that organisms destroy a great part of the annual production of CaCO₃ in very supersaturated waters, not only on reefs (Golubi and Schneider, 1979) but also in the open ocean (Bishop et al., 1980). Some kind of pseudo-equilibrium with CaCO₃ is crucial to maintain the pH of the sea-water. If organisms precipitate practically all the CaCO₃ in the ocean today, is it not very likely that they dissolve it also?

As a rule of thumb the biota (with the exception of man), produces, consumes and destroys any compound that it needs. The time scale for recycling elements by the biota is orders or magnitude smaller than any geological process. With this rate advantage the biota can exert a controlling influence on its environment. By just a small imbalance of fluxes, the biota can efficiently redirect elements into different geologic reservoirs and maintain the distributions by kinetic equilibrium. Thus, most of the chemical processes in the ocean are very likely to be biologically controlled, and seldom can principles of chemical equilibrium be applied.

17.4 CONCLUSIONS

With physical or chemical processes in the ocean we are faced with the same dilemma: How do we reconcile the large time and space variability in the fluxes with the large scale, steady state fields of the circulation and tracer distributions? What processes and what parameterization should be used for the evolution of the transient tracers on a short time scale?

We have been using parameters from steady state models to predict the time response of the system in very short time scales. More than half of all the inputs to the environment made by man have been done during the last 35 years. The distribution of the tracers produced during the 50s and 60s by nuclear weapons testing has shown that the layered structure of the ocean controls the initial spreading of the inputs. As more adequate models are developed to describe the distribution of these tracers, we will be better equipped to predict the fate of future emissions and their impact in the environment.

The role that the deep waters play in absorbing CO₂ must be assessed in a more precise way with the help of tritium, helium, freons, Kr-85, and bomb-produced C-14 (Broecker et al., 1980). The models show that the effective ventilation of the deep layers is the single most important parameter to determine the fraction of CO₂ in the ocean. If the airborne fraction stayed constant in the last 30 years, it means that the short turn-over reservoirs have passed their transient adjustment and are increasing in tandem with the input function. In the future, the exchange of the atmosphere with the large reservoirs will play the major role.

The biological control of the geochemical cycles has profound consequences for the environment. The abyssal ocean is a reservoir in which the essential nutrients are present in a stable and easily mobilized form. The bicarbonate in the ocean buffers it against large variations in pH. By maintaining a pseudo-equilibrium with CaCO₃, the biota has survived changes in circulation due to continental drift, periods of deep water anoxia, and different regimes of erosion. Presently, by transferring more CO₂ than carbonate into the abyssal layer, the biota maintains the ocean surface pH at high values, the CO₂ partial pressure in the atmosphere low, and the temperature of the globe at the fringes of glaciation.

Phosphorus seems to be the regulating element of the biological cycle in the ocean. In spite of observations that suggest a limiting role for nitrate in sub-tropical surface water, the biological cycles may be regulated to re-use as much phosphorus as is available because phosphorus has no other reservoirs or alternative pathways. Nitrogen deficiencies can be regulated by a balance between nitrification and denitrification, as it is in deep anoxic lakes like Tanganyika or closed basins like the Black Sea, where the deep water is entirely devoid of nitrate but rich in phosphate. The biota is extremely frugal in recycling phosphorus that is preferentially regenerated from the organic matter, as indicated by the increase in C:P ratios in suspended particles and decrease in regeneration rates with depth. The mobilization of phosphorus may produce the most significant alterations in the ocean environment.

The regulation of the geochemical cycles by the biota is done at the ecological level and has evolved during a few billion years. The immediate impact of man's disturbances is more likely to be felt at the level of populations and species. It is very difficult to predict what the response of the system will be and how long it will take to adjust. Man's involvement in the geochemical cycles is in a new phase of biological interaction with very unpredictable consequences.

17.5 REFERENCES

Bishop, J. K. B., Edmond, J. M., Ketten, D. R., Bacon, M. P., and Silker, W. B. (1977) The chemistry, biology, and vertical flux of particulate matter from the upper 400m of the equatorial Atlantic Ocean, Deep-Sea Res., 24, 511-548.

Börkström, A. (1979) A model of CO₂ interaction between atmosphere, oceans and land biota, in Bolin, B., Degens, E. T., Kempe, S., and Ketner, P. (eds) The Global Carbon Cycle, SCOPE Report No. 13, New York, Wiley.

Broecker, W. S., and Peng, T.-H. (1980) The distribution of bomb produced tritium and radiocarbon at GEOSECS station 247 in the eastern North Pacific, Earth Planet. Sci. Lett., 49, 453-462.

Denser, W. G., and Ross, E. H., (1980) Seasonal change in the flux of organic carbon to the deep Sargasso Sea, Nature, 283, 364-365.

Fiadeiro, M. E. (1982) Three-dimensional modeling of tracers in the deep Pacific Ocean: II. Radiocarbon and the circulation , J. Mar. Res. (submitted).

Fine, R. A., and Östlund, H. G. (1980) Exchange times in the Pacific Equatorial Current system, Earth Planet. Sci. Lett., 49, 447-452.

Gardner, W. D. (1980) Field assessment of sediment traps, J. Mar. Res., 38, 41-52. 

Golubi, S., and Schneider, J. (1979) Carbonate dissolution, in Trudinger, P. A., and Swaine, D. J. (eds) Biogeochemical Cycling of Mineral-forming Elements, Amsterdam, Elsevier.

Honjo, S. (1978) Sedimentation of material in the Sargasso Sea at a 5367 m deep station, J. Mar. Res., 36, 469-492.

Jenkins, W. J. (1979) Tritium and ³He in the Sargasso Sea, J. Mar. Res., 38, 533-569.

Keeling, C. D., and Bolin, B. (1967) The simultaneous use of chemical tracers in oceanic studies. Part IGeneral theory of reservoir models, Tellus, 19, 566-581.

 Keir, R. S. (1979) The dissolution kinetics of biogenic calcium carbonate: laboratory measurement and geochemical implications, Ph.D. Thesis, Yale University, New Haven.

Lal, D., and Lerman, A., (1973) Dissolution and behavior of particulate matter in the ocean: some theoretical considerations, J. Geophys. Res., 78, 7100-7111.

Morse, J. W. (1978) Dissolution kinetics of calcium carbonate in sea water: VI. The near equilibrium dissolution kinetics of calcium carbonate rich deep sea sediments, Am. J. Sci., 278, 344-353.

Packard, T. T. (1969) The estimation of the oxygen utilization rate in seawater from the activity of the respiratory electron system in plankton, Ph.D. Thesis, University of Washington, Seattle.

Rowe, G. T., and Gardner, W. D. (1979) Sedimentation rates in the slope water of the northwest Atlantic Ocean measured directly with sediment traps, J. Mar. Res., 37, 581-600.

Smith, K. L. Jr (1978) Benthic community respiration in the N.W. Atlantic Ocean: In-situ measurements from 40 to 5200 m, Mar. Biol., 47, 337-347.

Smith, K. L. Jr, White, G. A., and Laver, M. B. (1979) Oxygen uptake and nutrient exchange at sediments measured in situ using a free vehicle grab respirometer, Deep-Sea Res., 26, 337-346.

Soutar, A., Kling, S. A., Crill, A., Duffrin, E., and Bruland, K. W. (1977) Monitoring in the marine environment through sedimentation, Nature, 266,136-139.

Wiebe, P. H., Boyd, S. H., and Winget, C. (1976) Particulate matter sinking to the deep-sea floor at 2000m in the Tongue of the Ocean, Bahamas, with a description of a new sedimentation trap, J. Mar. Res., 34, 341-354.