SCOPE 21 -The Major Biogeochemical Cycles and Their Interactions  

ABSTRACT

The global river input of dissolved inorganic nitrogen and phosphorus to the ocean and its alteration by man's activities are first discussed briefly. Although uncertainties remain in estimations of natural and present-day fluxes of N and P, the contribution due to human activities is clearly several times larger than the natural fluxes. In the estuaries, chemical and biological processes may profoundly change the speciation of the nutrients and affect their transfer to the adjacent coastal zone. These processes are strongly influenced by the morphological and hydrodynamical properties of a given estuary, and may be altered in highly polluted rivers.

In the coastal zone, the N and P cycles are essentially linked to the C cycle, which presents distinct features when compared to the open ocean cycles. The coastal zones are characterized not only by a higher biomass, but more specifically by short food webs, high turn-over rates and large transfers of organic matter to the sediments. In turn, degradation of organic matter in the sediments plays an important role in the recycling of nutrients. Finally, the increase of N and P discharged by rivers in heavily populated regions may lead to modification of the phytoplankton composition of the coastal waters, particularly when dissolved silicon becomes limiting.

14.1 THE RIVER INPUT OF NUTRIENTS TO THE OCEAN SYSTEM 

The river input of dissolved inorganic nitrogen and phosphorus to the ocean system is obviously of great importance in a discussion of the interactions of these compounds in the estuarine and adjacent coastal zones. However, the evaluation of the fluxes is complicated by the facts that there may be large seasonal fluctuations and that human activities have largely disturbed the natural fluxes.

In order to evaluate the natural and present-day nutrient fluxes on a global basis, I have first selected a few well studied rivers that also cover a large range of population density in their drainage basins. The data selected are presented in Table 14.1.

Except for dissolved silicon, the concentration of the nutrients observed in the fresh-waters are extremely variable ranging from 0.4 to 60 µmoles litre^-1 for orthophosphate and from 3 to 800 µmoles litre^-1 for total dissolved inorganic nitrogen. These extreme compositional differences mainly reflect the influence of man's activities. This conclusion is documented in Figure 14.1 where the logarithm of the concentration of total dissolved phosphorus and nitrogen is plotted as a function of the logarithm of the number of inhabitants per unit of river discharge (litres s^-1).

In the case of the Hudson, we have calculated a theoretical concentration by dividing the mean annual input of each nutrient by the mean fresh-water discharge because a large fraction of the nutrient load is discharged in the estuarine zone where these nutrients are already diluted by sea-water.

Figure 14.1 shows for both N_T and P_T an increase in concentration as a function of the population density normalized to the water discharge. The slope of this increase, equal to 1, suggests that the concentration of nutrients in a river system can be approximately evaluated from

where C₀ is the concentration of nutrient in pristine water, a is the rate of production of nutrient per inhabitant, and x is the number of inhabitants per unit of water discharge of the river.

The values for C₀ and a obtained from this graph are reported in Table 14.2. However, most of the polluted rivers considered correspond to the more industrialized regions, and the rate of N and P input per inhabitant is probably over-estimated on a global basis. I have nevertheless calculated the global river input of nutrients assuming a mean global river discharge equal to 32 x 10¹² m³ per year and using the values computed in Table 14.2.

The calculated values (Table 14.3) are in fairly good agreement with recent evaluations of the mean global river fluxes of dissolved inorganic nitrogen and phosphorus.

The data of Van Bennekom and Salomons (1981) is based on fluxes due to various man's activities added to the natural fluxes, whereas Meybeck (1982) has estimated the fluxes from reported data for the rivers of the world in the 1970s. One main uncertainty concerns the concentration of inorganic nitrogen in pristine waters. It is possible that the higher values reported by Van Bennekom and Salomons (1981) and Meybeck (1982) have already been influenced by deforestation and by the increase of nitrate in rain-water. On the other hand, the present-day flux of nutrients quoted by Meybeck is one-half of the two other estimates.

Table 14.1 Nutrient concentrations in some typical rivers in µmoles litre^-1

	PO43-	NO3 + NO2	NH4	NTa	H4SiO4	N:P	Ref.
Amazon	0.4	3.7		3.7	187	7.5	(1)
Zaïre	0.8	6.7	0.4	7.1	165	8.4	(2)
Niger	0.4	7.6	1.0	8.6	250	21.5	(2)
Missisipi	2.8	156	< 4	160	83	57	(3)
Hudsonb	3.0	50	20	70	70	23	(4)
	(14)	(53)	(110)	(163)	(75)	(11.6)
Rhinec upper	0.5	30	2.5	32.5	61	65	(5)
lower	11.4	230	90	320	90	28	(5)
Scheldt	60d	150	650	800	230	13.3	(6)
Global mean	2.0			46	167	23	(7)
	0.7-1.2			22		3118	(8)
aTotal dissolved nitrogen. b Mean maximum concentration levels observed in the Hudson estuary. The values in parentheses have been calculated by dividing the mean annual input of each nutrient by the mean fresh-water discharge. cUpper Alpine Rhine at Schmitter and lower Rhine at Lobith.  d Including polyphosphate. References: (1) Gibbs (1972); (2) Van Bennekom et al. (1978); (3) Ho and Barret (1977); (4) Deck (1981); (5) Zobrist and Stumm (1981); (6) Wollast (1976); (7) Van Bennekom and Salomons (1981); (8) Meybeck (1982).

Figure 14.1 Evolution of total N and P concentration (moles litre^-1) in rivers as a function of the number of inhabitants of the hydrographic basin per unit of fresh-water discharge. Global mean after Van Bennekom and Salomons (O) and Meybeck ()

Table 14.2 Evaluation of the mean river concentrations and global fluxes of nutrients based on the data of Figure 14.1

As pointed out earlier, my calculations may tend to over-estimate the global fluxes. Also, the fluxes of nutrients due to man's activities, as assumed by Van Bennekom and Salomons, may not reach the estuarine region in their inorganic dissolved form. Therefore, the data of Meybeck are probably closer to reality. However, despite the uncertainty in this data, an obvious conclusion is that human activities have drastically modified the transport of nutrients from land to ocean.

Table 14.3The fluxes of dissolved inorganic N and P transported by the rivers to the ocean (in 10¹² g yr^-1)

Natural		Man's activities		Present		References
P	N	P	N	P	N
0.5	5	1.9	19	2.3	24	(1)
0.43	4.4	0.41.0	7	0.831.43	11.4	(2)
0.45	2.6	1.70	21	2.15	24	(3)
Reference: (1) Van Bennekom and Salomons (1981); (2) Meybeck (1982); (3) this work.

The increase of the dissolved nitrogen flux in rivers represents 30% of the nitrogen fixed annually by man mainly during combustion processes and fertilizer production (6 x 10¹² moles yr^-1 according to Simpson et al. (1977)). For phosphorus, the increase is only 15% of the total phosphorus mined annually.

14.2 TRANSFER OF MATTER THROUGH THE ESTUARINE ZONE 

The estuarine system is characterized by profound changes in the chemical properties of the water masses and usually by high biological activities, both of which significanctly affect the speciation of the elements and transfer to the adjacent coastal zones. This is particularly true in the case of nutrients and organic matter.

The transfer of materials in the estuarine zone is strongly influenced by the morphological and hydrodynamical properties of a given estuary, that in turn control water circulation, residence time of the water masses, and accumulation of particulate matter by sedimentation. Classifications of estuaries are mainly based upon the relative vertical and lateral stratification of the water masses identified by their salinities; most earlier classifications made a distinction between vertically homogeneous and stratified estuaries (Bowden, 1967). Vertical estuarine mixing is essentially controlled by the relative importance of river flow and tidal action. Intensity of mixing increases rapidly with decreasing depth, and stratification is therefore reduced in the shallow parts of estuaries. The residence time of fresh-water increases quickly with increasing vertical mixing due to its dilution in a large body of sea-water. For well stratified estuaries, fresh-water flows in a restricted surface layer without mixing with the underlying salt-water wedge, and thus rapidly reaches the sea.

Furthermore, most of the organic and inorganic particles transported by rivers are negatively charged, and are usually present in colloidal form. The increase in ionic strength during mixing of fresh- sea-water neutralizes the surface charges by adsorption of cations, and primarily as a result of this the particles flocculate and settle. The accumulation of sediments takes place in restricted flow zones and depends strongly on the type of estuary, but generally estuaries are areas of intensive sedimentation. In well mixed or partially stratified estuaries, the circulation pattern of the water masses induces the occurrence of a turbidity maximum which usually corresponds to the zone of accumulation of sediments.

It is difficult to give a general picture of the transfer of matter in such complicated systems. Consequently, I will briefly discuss some typical estuarine processes of particular importance for understanding the interactions of elements vital to life.

14.2.1 Production, Transport, and Degradation of Organic Matter in Estuaries

It is generally accepted that the particulate organic matter of terrestrial origin carried by fresh-water into estuaries is at least partly and often almost completely removed in the estuarine zone and does not reach the coastal zone. For instance in the Scheldt, a vertically homogeneous estuary, of the 152 x 10⁹ g yr^-1 of particulate organic matter carried by fresh-water, 115 x 10⁹ g yr^-1 are deposited in a restricted zone 30 km long, corresponding to the 110‰ range of salinity (Wollast and Peters, 1978). Calculations based on long term deposition rates in the Gironde estuary (Allen et al., 1976) gave similar results; 75% of the suspended load is trapped in the estuarine zone and does not reach the sea.

In vertically well-stratified estuaries, like those of most large rivers, the picture may be different. In the case of the Amazon, 95% of the terrigeneous sediment settles out within the river mouth before the salinity reaches 3‰ (Milliman et al., 1975). This intensive local sedimentation is mainly due to the existence of an extensive shoal occurring before the Amazon river mouth that gives rise to efficient vertical mixing (Gibbs, 1970).

In the Zaïre river, (Eisma et al., 1978), however, the out-flow of river water is concentrated in a narrow surface layer at the centre of the stream, which prevents rapid sedimentation. As a consequence, particulate organic matter settles and accumulates near the head of the Zaïre submarine canyon, inducing an oxygen-depleted zone in the bottom waters.

The removal of dissolved organic matter, particularly the humic materials, by coagulation and settling is also a well established phenomenon during estuarine mixing (Matson, 1968; Sieburth and Jensen, 1968; Nissembaum and Kaplan, 1972; Gardner and Menzel, 1974; Nissenbaum, 1974). However, the extent of removal of DOC during estuarine mixing appears to have been over-estimated in earlier studies (see comment by Burton, following this chapter).

Biological processes may also drastically affect the transfer of organic matter within the estuarine zone. On one hand, respiration of detrital organic matter may reduce the fluxes of this material through the estuarine zone and, on the other hand, large amounts of new organic carbon may be produced in systems characterized by high primary productivity.

In estuaries with long residence times, the non-refractory organic matter is almost completely decomposed by the heterotrophic bacteria. If the organic load is high, this leads to the formation of zones of anoxic water. For example, in the Scheldt estuary, about 40% of the organic load is respired, mainly affecting the dissolved organic fraction. As a consequence, a permanently anaerobic zone extends over 30 km of the estuary from the sediment surface to the top of the water column. However, in the Zaïre river the anoxic zone is restricted to the bottom water near the head of the canyon where the detritus of terrestrial plants accumulates. Approximately 1 % of this load is respired before settling (Van Bennekom et al., 1978), but this is sufficient to completely remove dissolved oxygen from the bottom waters.

The degradation of organic matter affects nutrient budgets by releasing dissolved NH₄⁺ and PO₄³. More important for the nutrient behaviour is the possible occurrence of an anoxic zone in which the bacterial activity or chemical processes drastically modify the speciation of some nutrients. These processes will be discussed in more detail in sections 14.2.2 and 14.2.3.

Finally, the high nutrient content of estuarine waters promotes high primary productivity located where the turbidity has dropped as a result of flocculation and sedimentation of terrigenous suspended matter. The diatom bloom in the Amazon estuary reaches its maximum at a salinity lower than 10‰. In the Zaïre, the algae start to develop at much higher salinities at which the dilution of nutrients has resulted in concentrations too low to produce an algal bloom of importance (Cadée, 1978).

In the Scheldt estuary (Figure 14.2), the large supply of nutrients produces phytoplankton blooms during spring and summer. During these periods the organic matter produced by photosynthesis in the lower part of the estuary nearly equals the amount of detrital organic carbon removed by respiration and sedimentation in the upper part (Wollast and Peters, 1978). Most of this spring and summer-produced organic matter is transferred to the coastal zone where it is often difficult to distinguish it from that produced by coastal primary production.

14.2.2 Behaviour of Nitrogen Species.

The three main processes that modify the speciation of nitrogen in aquatic systems (nitrification, denitrification, and biological uptake), are commonly very active in estuarine systems and may significantly affect the transfer of nitrogen to the adjacent coastal waters and to the atmosphere, in the case of denitrification. The source of nitrate is essentially related to leaching of soil and to surface run-off. The use of inorganic fertilizers has considerably increased the concentration of nitrate carried by rivers. On the other hand, the presence of anthropogenic NH₄⁺ is more directly related to domestic waste water discharges. As the estuarine zone is often heavily populated, high concentrations of ammonia are encountered there.

Figure 14.2 Longitudinal profiles of organic matter in the Scheldt estuary for various periods of the year (Wollast and Peters, 1978). Reproduced by permission of UNESCO

Denitrification has a pronounced effect on the transfer of nitrogen, because nitrate is released as N₂ or N₂O to the atmosphere. However, denitrification occurs only in sections of estuaries that exhibit oxygen depletion and where nitrate is used by heterotrophic bacteria as an oxidant. This is the case in heavily polluted rivers that have long residence times, or in stratified estuaries where organic matter, even of natural origin, accumulates. In the Scheldt for instance, denitrification is observed even during the winter when the bacterial activity is lowest. Figure 14.3 shows the situation in this river under these circumstances; most of the nitrate is consumed in the upper part of the estuary, where partially anaerobic conditions prevail. If these observations are extended over one complete year, approximately 30% of the total dissolved nitrogen input to the estuary from the river Scheldt is lost to the atmosphere by denitrification in the estuarine zone. Partial denitrification in the anoxic bottom waters of the Zaïre Canyon was also demonstrated by Van Bennekom et al. (1978).

Figure 14.3 Longitudinal profiles of nitrate, nitrite and oxygen in the Scheldt estuary during February (after Somville, 1980). Reproduced by permission of M. Somville

Nitrification is less important for the nitrogen budget, except perhaps for the release of N₂O to the atmosphere. However, nitrification may be an important factor in determining phytoplankton growth due to the preferential uptake of ammonia over nitrate.

In well oxygenated waters, NH₄⁺ is slowly oxidized to NO₂ and NO₃ by nitrifying bacteria; if the residence time of water masses in the estuary is long enough, NH₄⁺ is almost completely consumed by nitrifying bacteria. This is the case of the Scheldt, as shown in Figure 14.4, when aerobic conditions are restored in the estuary. It should also be noted that large concentrations of N₂O are observed in the nitrification zone. The net production rates of N₂O computed from this profile by Deck (1981) have maximum values of 0.1 µg m^-3 s^-1 and are very similar to those computed by McElroy et al. (1978) for the Potomac. These values seem, however, to be small with respect to the global land or ocean production rates.

Figure 14.4 Longitudinal profiles of dissolved nitrogen species and oxygen in the Scheldt estuary during October (Somville, 1980). Reproduced by permission of M. Somville

Primary production in the estuarine zone must be accompanied by a transfer of dissolved nutrients to the particulate phase. In the Chesapeake Bay McCarthy et al. (1975) have shown, that NH₄⁺ is consumed preferentially to NO₃ as long as the concentration of ammonia is greater than 0.1 µmole litre^-1. Thus one may expect that in most polluted estuaries, the phytoplankton growth will mainly affect the concentration of NH₄⁺. It is, however, not easy to distinguish this uptake from the nitrification process, which is often dominant in a region where extensive dilution of nutrients by sea-water occurs. Furthermore, McCarthy et al. (1975) have also shown that the turn-over time of NH₄⁺ in the euphotic zone ranges from only 3 to 20 hours and averages 8 hours in Chesapeake Bay. The net removal of this nutrient may thus be expected to remain low and difficult to quantify from concentration profiles. However, in estuaries having low concentrations of nutrients, the activity of phytoplankton is clearly demonstrated by removal of  the nutrients in the zone of high productivity. In the Amazon, Van Bennekom and Tijssen (1978) found that NO₃ , which is consumed in the absence of NH₄⁺ , was below 1 µmole litre ^-1 at 20 ‰ salinity. In the case of the Zaïre, which exhibits a much lower productivity, Van Bennekom et al. (1978) found that 1540% of dissolved nitrogen is consumed between 25 and 30 ‰ salinity and probably reflects the effects of phytoplankton growth.

Figure 14.5 Evolution of dissolved and particulate nitrogen concentration at the mouth of the Scheldt (km = 0) over one year (Wollast, 1976)

The nitrogen uptake by phytoplankton in the Scheldt has been evaluated by following monthly seasonal changes of dissolved and particulate nitrogen at the mouth of the river (Figure 14.5). The maximum N_part. occurs during May and corresponds to the intense Spring bloom, mainly of diatoms. One should remember that the decrease of N_diss is not only due to the uptake of nutrients during plankton growth but also to denitrification, which is also at a maximum during the Summer.

A tentative budget for nitrogen in the Scheldt is summarized in Table 14.4; these values should be considered only as first approximations. Also it should be recalled that the Scheldt represents an extreme case of a highly polluted estuary. Besides the importance of denitrification processes in such a system, it is also interesting to note that only a small fraction of the available nitrogen is consumed by phytoplankton.

Furthermore, as pointed out by McCarthy et al. (1975), little useful information is gained with respect to nutrient and phytoplankton dynamics from measurements of biomass and nutrient concentrations. Measurements of the rate of utilization of ¹⁵NH₄⁺ and ¹⁵NO₃ by phytoplankton are a more efficient tool for evaluating the dynamic aspects of plankton nutrition. A better understanding of the nutrientphytoplankton cycle in estuaries requires more direct measurements of the rate of utilization of the nutrients.

Specie	Input	Denitrif.	Nitrif.	Prod.	Sed.	Output
NH4	23		12	6		5
NO3 + NO2	9	11	+12			10
Npart	2			+6	3	5
Ntot	34	11			3	20

14.2.3 Behaviour of phosphate

Inorganic phosphorus occurs mainly as orthophosphate in natural waters. In polluted estuaries receiving untreated domestic waste water, polyphosphates may represent a significant portion of the inorganic dissolved phosphate, as shown in the case of the Scheldt estuary (Figure 14.6). This polyphosphate is slowly hydrolysed to orthophosphate within the river itself.

Figure 14.6 Longitudinal profile of dissolved phosphorus species in the Scheldt estuary during May (Wollast, 1976)

Processes affecting the behaviour of phosphate in estuaries are, however, very complex and probably not entirely identified and certainly not sufficiently understood. First, orthophosphate is a chemically active compound that may be involved in various dissolutionprecipitation or adsorptiondesorption reactions. The solid phases resulting from these reactions are so complex that basic properties like their solubility or exchange equilibria are poorly known. However, these chemical reactions seem to be reversible and rather fast. They appear to act as a buffering mechanism that maintains dissolved phosphate in a narrow range of concentration (around 1 µmole litre^-1) during the mixing of river- and sea-water (Liss, 1976).

Furthermore, the biological processes involving phosphorus are neither simple nor fully understood. The luxury uptake of phosphorus by phytoplankton is a well known phenomenon in rich nutrient zones but insufficient data in estuaries make it impossible to estimate departures from the C:N:P Redfield ratios (Redfield, 1958). Also, the direct or indirect role of bacteria in the estuarine phosphorus cycle has been completely neglected.

I will thus restrict my discussion to some qualitative aspects of the main processes responsible for modifying the distribution of phosphorus species in estuaries. I will first consider the case of an unpolluted large riverthe Zaïreand after that two heavily polluted estuariesthe Scheldt and the Hudson.

In the Zaïre, Van Bennekom et al. (1978) observed a maximum of PO₄^3- in the salinity range 0 to 25‰ whereas the concentration of phosphate in the particulate phase decreased with increasing salinity. The increase in dissolved phosphate in this salinity range thus may be the result of a desorption mechanism. At salinities above 25 ‰ the concentration of PO₄^3- decreases because of the mixing processes and phytoplankton growth. However, the changes in P due to planktonic activity are hardly detectable. In the case of the more productive Amazon estuary, Van Bennekom and Tijssen (1978) showed that at 20 ‰ salinity, PO₄^3-concentration was already below 0.1 µmole litre^-1 due to phytoplankton growth.

In the bottom waters of Zaïre Canyon, Van Bennekom et al. (1978) found a substantial decrease in PO₄^3- with decreasing dissolved oxygen content. Normally the reverse phenomenon is observed and is attributed to the release of phosphate-containing iron hydroxides during the dissolution of these minerals in anoxic conditions. These authors attribute the removal of phosphate in the surface water to adsorption on ferric hydroxide precipitated on quartz grains that settle through the water column in that region. Such a mechanism is possible due to the buffering capacity of the suspended matter with respect to dissolved phosphate reported by Liss (1976), if the anoxic conditions are furthermore insufficient to redissolve the ferric precipitate. The bacterial growth associated with the degradation of the large quantities of plant debris in these bottom waters may be another explanation for the uptake of PO₄³-.

Figure 14.7 Distribution of the concentration of orthophosphate as a function of salinity in the Hudson for low and high flow conditions (Deck, 1981). The maximum corresponds to the main sewage discharges from New York City. Reproduced by permission of B. L. Deck

The behaviour of phosphate in the Hudson is quite simple (Deck, 1981). The total upstream input equal to 1.8 moles s^-1 (1.5 moles s^-1. particulate and 0.3 moles s^-1 dissolved) is mainly related to agricultural activities. This input is small compared to the massive input of 12.6 moles s^-1 from human sewage loading localized in the New York Bight. Consequently a plot of the concentration of PO₄^3- versus salinity (Figure 14.7) exhibits a maximum corresponding to the urban input. The value of this maximum is a hyperbolic function of the river flow, which indicates a rather constant input, as may be expected. The linear change in the dissolved phosphate concentration on both sides of the maximum (Figure 14.7) indicates that this compound behaves as a conservative parameter during mixing and fresh- and sea-water. Phosphate consumption by primary producers during winter and spring is only 0.003 to 0.04 moles s^-1, whereas summer consumption may reach 0.5 to 0.7 moles s^-1. In any case, the rates of utilization of phosphate by phytoplankton are very small compared to the total input (14.3 moles s^-1) and are not detectable in the longitudinal concentration profiles.

The behaviour of phosphorus in the Scheldt estuary is much more complicated (Figure 14.6) because of the existence of an extended anaerobic zone, long residence times of the water masses, and intensive shoaling occurring in the upper part of the estuary. In the anaerobic zone, low redox potentials lead to the reduction of the most reactive iron hydroxides to soluble Fe²⁺ (Wollast, 1976). The phosphates eventually adsorbed by this particulate phase are then released to the dissolved phase. When the dissolved oxygen is restored by reaeration and by mixing with sea-water, Fe²⁺ is re-oxidized and precipitates again as iron hydroxide, sequestering large amounts of dissolved phosphate by co-precipitation and adsorption. This particulate material is removed from the water column by sedimentation and is not transported to the lower part of the estuary. The growth of phytoplankton in the aerobic zone is responsible for a supplementary uptake of dissolved phosphate. Although there are no direct measurements of the rate of consumption of PO₄^3- by plankton, if values of twice the Redfield ratio (Redfield, 1958) are assumed to take into account the luxury uptake of P this represents a maximum value of 750 T P y^-1 (0.75 moles s^-1). This is small relative to the total input of 7.100 T P y^-1 (particular + dissolved) (7 moles s^-1), a situation similar to the Hudson.

Other processes are possible for removing dissolved phosphorus at the high concentrations encountered in the Scheldt. Reactions such as precipitation of apatite (calcium phosphate), vivianite (ferrous phosphate) and magnesiumammonium phosphate are theoretically possible, but none have been identified in situ or in laboratory experiments simulating estuarine conditions. As with nitrogen, I will compute an annual budget for phosphorus in the Scheldt from the monthly longitudinal concentration profiles of dissolved and particulate phosphorus. These calculations, presented in Table 14.5, show that a significant fraction of the phosphate input to the estuary never reaches the sea, but is trapped in estuarine sediments.

Table 14.5 Tentative mass balance for phosphorus in the Scheldt estuary (in 10⁹ g yr^-1). 

	Input	Chemical	Plankton	Sediment	Output
		precipitation	uptake
Dissolved P	5.6	3.3	0.8	0	1.5
Particulate P	1.5	+3.3	+0.8	4.9	0.7
Total P	7.1			4.9	2.2

14.3 INTERACTIONS OF C, N, AND P IN THE COASTAL ZONE

The carbon and nitrogen cycles have been studied in great detail in the Southern Bight of the North Sea and we will use these data as a typical example of the transfer processes occuring in a coastal zone. Figures 14.8 and 14.9 show respectively the budget of organic carbon and nitrogen circulation for the coastal North Sea off Belgium, which represents a surface area of 4.5 x 10⁹ m² and a mean depth of 15 metres.

Figure 14.8 Schematic representation of organic matter cycling in the Belgian coastal region of the North Sea. Fluxes in a g C m^-2 yr^-1; reservoirs in g C m^-2

The values of the fluxes and of the sizes of the reservoirs presented in Figure 14.8 have been estimated on the basis of independent measurements performed over several years, and described more extensively in Joiris et al. (1979), Wollast and Billen (1981) and Joiris et al. (1982). The mass balance obtained for the various species are generally fulfilled within ± 20% or better, which may be considered as very satisfactory. In the case of the nitrogen cycle of Figure 14.9, the various pool sizes have been directly measured. The nitrogen fluxes into and out of the sediments were estimated by Billen (1976). The fluxes in the water column have been computed from the carbon fluxes of Figure 14.8, assuming a C:N ratio equal to 6.4.

It should be pointed out that very similar patterns have been described for various other coastal environments such as the Louisiana, Texas, and West Florida shelves of the Gulf of Mexico, and the central shelf of the Bering Sea (Walsh et al., 1981).

Figure 14.9 Schematic representation of nitrogen species cycling in the Belgian coastal region of the North Sea. Fluxes in g N m^-2 yr^-1; reservoirs in g N m^-2. Reproduced by permission of G. Billen

The net primary production of the Southern Bight of the North Sea (370 g C m^-2 yr^-1) is relatively high compared to the mean value for the whole oceanic system (about 100 g C m^-2 yr^-1) but this is generally the case for coastal environments (average: about 270 g C m^-2 yr^-1; Wollast and Billen, 1981). The river input of nitrogen is small compared to the requirements of the phytoplankton. This conclusion can be easily appreciated by considering the mean world river input of nutrients presented in Table 14.2 divided by the surface area of the coastal zone (30 x 10¹² m²). The global river inputs of nutrients are equal to 0.9 g N m² yr^-1 and 0.7 g P m^-2 yr^-1. The amounts of nutrients required for a mean net primary productivity of 270 g C m^-2 y^-1 are on the order of 50 g N m^-2 y^-1 and 7 g P m² y^-1. Thus, the input by rivers represents only 1 to 2% of the nutrient uptake by phytoplankton.

It is interesting to compare these values to the relative importance of upwelling and vertical diffusion of nutrients for the whole oceanic system. From my previous estimates (Wollast, 1981) the mean inputs of nitrogen and phosphorus to the euphotic zone by vertical diffusion, which is rather uniformly distributed over the pelagic zone of the ocean, are approximately equal to 0.5 g N m^-2 y^-1 and 0.1 g P m^-2 y^-1. These values are very similar to the actual river inputs, but the oceanic zones where the fluxes of nutrients are due to vertical diffusion alone are less productive areas (less than 50 g C m^-2 y^-1) than the coastal zone. In the regions of upwelling, the vertical flux of nutrients is much greater, and 5 g N m^-2 y-1 and 1 g P m^-2 y^-1 are reasonable mean values. It should be noted that upwelling frequently occurs in coastal zones and may at least partially explain the high productivity of the coastal zone. This situation is obviously not the case for the North Sea.

In all aquatic systems the external input of nutrients is only a small fraction of nutrient uptake because of the rapid turn-over of the plankton. In this case, primary production (P) is a function of both the external flux of nutrients (I) and the rate of decomposition (D) of the organic matter in the system. If a steady state is reached in the system, one can show that

    PI

The rate of decomposition (D) is usually much larger than the input or output of nutrients (I) and thus the productivity is strongly dependent on decomposition. In the case of the North Sea, the ratio D/I is close to 10 and it explains indeed very well the high productivity observed.

About 30% of the total net primary production is excreted by phytoplankton as dissolved organic matter and only 20% is grazed by zooplankton. The importance of direct grazing with respect to primary production commonly has been considered to be a characteristic feature of marine ecosystems in contrast with terrestrial ecosystems (Odum, 1962; Crisp, 1864; Wiegert and Owen, 1971). However, at least in coastal environments, zooplankton grazing does not account for all the phytoplankton mortality, and important pathways of organic matter recycling through detritus and bacteria exist in parallel to the pathways through herbivorus zooplankton (Banse, 1974; Pomeroy and Johannes, 1966, 1968; Joiris, 1978; Joiris et al., 1979; Walsh et al., 1981). The high rate of recycling is also confirmed by the turn-over time of the phytoplankton in the coastal zone, which is only 3 days, whereas for oceanic systems the turn-over time is usually between 7 and 14 days (Cauwet, 1977). The high rate of turn-over reflects in turn the species composition of the phytoplankton largely dominated by the colony forming microflagellate Phaeocystis poucheti. The unusually high productivity of dissolved organic matter in this zone is probably related to the excretion of polysaccharides by these organisms, which are, however, rapidly recycled by the bacteria.

Another characteristic of the coastal zone is the major role played by the benthic system. The shallow depth of this zone has two consequences: 

	highly reactive organic matter is deposited by sedimentation, inducing an intense biological activity in the benthos;
	the primary producing zone is in direct contact with the benthic boundary layer, which may become an important source of nutrients.

Figure 14.10 summarizes the available data concerning the quantitative importance of organic matter degradation in marine sediments with respect to primary production. The decrease in degradation observed with water depth is obviously explained by the degradation of a higher percentage of the organic matter within the water column with increasing depth. This leaves less organic matter available for recycling in the benthos at greater depths. In shallow environments, up to 50% of the primary production can be recycled in the benthos. Faecal pellets and dead zooplankton generally have been considered the most important source of organic material for the benthos (Steele, 1974). Balance calculations show, however, that for many shallow areas these sources cannot account for all the benthic activity. Direct input of phytoplankton material to the bottom occurs, either by passive sedimentation (Davies, 1979), by feeding of benthic macro-organisms (Daro and Polk, 1973) or by processes like wave-induced percolation of sea-water through the interstitial space of sediments (Steele et al., 1970; Reidl et al., 1972). The biological degradation of organic matter in sediments is described extensively by Jørgensen (Chapter 18, this volume).

Figure 14.10 Amount of organic matter degraded by microbial activity in the sediments expressed as percentage of primary production in the water column, for marine () and lacustrine () environments as a function of the water depth (Wollast and Billen, 1981)

In the Belgian coastal waters of the North Sea the intensive bacterial degradation of organic matter in the surficial layer of the sediments leads to a release of ammonia and orthophosphate to the pore waters, inducing a concentration gradient and an upward diffusion of these dissolved species into the water column. If the organic content of the sediments is not too high, aerobic conditions are maintained in the few first centimetres of the sediments, and nitrification of ammonia produced in the deeper parts of the sediments occurs. When strongly reducing conditions prevail, the sediments become a sink for NO₃ by denitrification. The net effect of the benthos on the fluxes of nitrogen species in the North Sea is summarized in Figure 14.9. Fluxes of nitrogen species from the sediments are about 10 times higher than the river input and are thus important sources of nutrients in these shallow environments. Nitrogen (as N₂) lost during denitrification, however, is higher than the river input.

Furthermore, there is a net export of nitrogen from the Southern Bight to the north mainly as nitrate and particulate organic matter. These fluxes are difficult to quantify due to uncertainties in knowledge concerning the residual water currents, but they are certainly greater than the river input. Nitrogen lost by denitrification and by export out of the zone must be compensated by a production process becuase the river input is insufficient to fulfill the mass balance. Nitrogen fixation by blue-green algae and photosynthetic bacteria is the most probable mechanism but has not been studied as yet in this area. The rate of nitrogen fixation should be of the order of 10 g N m^-2 yr^-1 in order to balance the budget. This value is similar to that for non-pelagic nitrogen fixation in the oceans reported by Söderlund and Svensson (1976).

The data collected in the Southern Bight are insufficient to evaluate a detailed cycle for the element phosphorus. However, most of the observations discussed above for nitrogen are valid for phosphorus. As a first approximation, the fluxes and the reservoirs of phosphorus species may be deduced from the carbon and nitrogen cycles by using the Redfield C:N:P ratios (Redfield, 1958).

Finally, the Southern Bight of the North Sea is an interesting case of eutrophified coastal waters now receiving a marked increase in nutrients discharged by the rivers Rhine, Meuse, Scheldt, Thames, and Ems. However, as pointed out in the first part of this review, the concentration of dissolved silica remains quite low in the more polluted rivers. Futhermore, eutrophication of natural and artificial lakes causes an increase in diatom growth. Some of the silica is deposited as skeleton debris in lake sediments and therefore the contribution of Si from river run-off may well be decreasing (Van Bennekom and Salomons, 1979).

An important factor in eutrophified estuaries and coastal waters may be the limitation of productivity by silicon that is not enhanced in a way corresponding to nitrogen and phosphorus. In the case of the Southern Bight of the North Sea, this limitation is indeed reflected in a decrease in diatom cell counts and increase in total cells other than diatoms (Gieskes and Kraay, 1977). After a short spring bloom dominated by diatoms, the phytoplankton consist almost entirely of flagellates (Phaeocystis sp.) because silica becomes limiting (Van Bennekom et al., 1975). Subsequent changes within the entire food chain must be expected but little data are as yet available.

ACKNOWLEDGEMENTS

14.4 REFERENCES

Banse, K. (1974) On the role of bacterioplankton in the tropical ocean, Mar. Biol., 24, 1-5.

Bowden, K. F (1967) Circulation and Diffusion, in Lauff, G. H. (ed.) Estuaries, Washington, D.C., American Association for the Advancement of Science, 15-36. 

Cadée, G. C. (1978) Primary production and chlorophyll in the Zaïre river, estuary and plume, Neth. J. Sea Res., 12, 368-381.

Crisp, D. J. (1964) Grazing in terrestrial and marine environments, in Crisp, D. J. (ed.) Br. Ecol. Soc. Symp. No. 4, Oxford, Blackwell Scientific Publ., 9-16.

Davies, J. M. (1979) Influence of sedimentation processes on biodegradation on the sea floor, Presented at the CNRS colloquium Biogeochimie de la matière organique à 1'interface eau sédiment marin, Marseille, France.

Eisma, D., Kalf, J., and van der Gaast, S. J. (1978) Suspended matter in the Zaïre estuary and adjacent Atlantic Ocean, Neth. J. Sea Res., 12, 382-406.

Gibbs, R. J. (1970) Circulation in the Amazon river estuary and adjacent Atlantic Ocean, J. Mar. Res., 28, 113-123.

Gieskes, W. W. C., and Kraay, G. W. (1977) Continuous plankton records: changes in the plankton of the North sea and its eutrophied Southern bay bight from 1949 to 1975, Neth. J. Sea. Res., 11, 334-364.

Joiris, C. (1978) Seabirds recorded in the northern North Sea in July: the ecological implications of their distribution, Gerfault, 68, 419-440.

Joiris, C., Billen, G., Lancelot, C., Daro, M. H., Mommaerts, J. P., Hecq, J. H., Bertels, A., Bossicart, M., and Nijs, J. (1982) A budget of carbon cycling in the Belgian Coastal Zone: relative roles of zooplankton, bacterioplankton, and benthos in the utilization of primary production, Neth. J. Sea Res. 16, 260-275.

Jørgensen, B. B. Processes at the Sediment-Water Interface. Chapter 18, this volume. 

Liss, P. S. (1976) Conservative and non-conservative behaviour of dissolved constituents during estuarine mixing, in Burton, J. D., and Liss, P. S. (eds) Estuarine Chemistry, London, Academic Press, 93-100.

McElroy, M. B., Elkins, J. W., Wofsky, S. C., Kolb, C. E., Duran, A. P., and Kaplan, W. A. (1978) Production and release of N₂O from the Potomac Estuary, Limnol. Oceanogr., 23, 1168-1182.

Meybeck, M. (1982) Nutrients (N, P, C) transport of world rivers, Amer. J. Sci. 282, 401-450.

Milliman, J. D., Summerhayes, C. P., and Baretto, H. T. (1975) Oceanography and suspended matter off the Amazon river, J. Sediment. Petrol., 45, 189-206.

Nissenbaum, A. (1974) Deuterium content of humic acids from marine and non-marine environments, Mar. Chem., 2, 59-63.

Odum, E. P. (1962) Relationship between structure and function in the ecosytem, Jap. J. Ecol., 12, 108-118.

Pomeroy, L. R., and Johannes, R. E. (1968) Respiration of ultra plankton in the upper 500 m of the Ocean, Deep Sea Res., 15, 381-391.

Riedl, R. S., Huang, N., and Machan, R. (1972) The subtidal pump: a mechanism of interstitial water exchange by wave action, Mar. Biol., 13, 210-221.

Simpson, H. J. et al. (1977) Man and the global nitrogen cycle. Group Report, in Stumm, W. (ed.) Global Chemical Cycles and their Alterations by Man, Berlin, Dahlem Konferenzen, 253-274.

Nitrogen, Phosphorus and SulphurGlobal cycles, SCOPE Report No. 7, Ecol. Bull. (Stockholm),

Somville, M. ((1980) Etude écophysiologique des métabolismes bactériens dans 1'Estuaire de 1'Escaut, Thése de Doctorat, Université Libre de Bruxelles, Belgium.

Steele, J. H., Munro, A. L. S., and Giese, G. S. (1970) Environmental factors controlling in the apipsammie flora on beach and sublittoral sands, J. Mar. Biol. Assoc. U.K., 50, 907-918.

Van Bennekom, A. J., Gieskes, W. W. C., and Tijssen, S. B. (1975) Eutrophication of Dutch coastal waters, Proc. R. Soc. London Ser. B, 189, 359-374.

Van Bennekom, A. J., and Tijssen, S. B. (1978) Nutrient distribution on and off the Guyana Shelf, related to upwelling and Amazon outflow, FAO Fish. Rep., 200, In suppl., 233-253.

Wiegert, R. G., and Owen, D. F. (1971) Trophic structure, available resources and population density in terrestrial versus aquatic ecosystems, J. Theor. Biol., 30, 69-81.

Wollast, R. (1981) Interactions between major biogeochemical cycles: marine cycles, in Likens, G. E. (ed.) Some Perspectives of the Major Biogeochemical Cycles, SCOPE Report No. 17, Chichester, Wiley, 125-141.

Wollast, R., and Peters, J. J. (1978) Biogeochemical properties of an estuarine system: the river Scheldt, in Goldberg, E. D. (ed.) Biochemistry of Estuarine Sediments, Paris, UNESCO, 279-293.

A COMMENT ON THE BEHAVIOUR OF DISSOLVED ORGANIC CARBON DURING ESTUARINE MIXING

J. D. BURTON

Knowledge of the behaviour during estuarine mixing of the dissolved organic material (DOM) transported by rivers is important for understanding aspects of the carbon cycle and the possible influence of terrigenous organic compounds on the productivity of coastal waters. Straightforward methods for the determination of dissolved organic carbon (DOC) have only recently become available, and earlier ideas about the fate of DOC were derived from indirect evidence, particularly from measurements on sediments. A substantial part of the organic material entering estuarine sediments is transported in particulate forms, however, making it difficult to infer the extent of removal of riverborne DOC via the particulate phase. The recent direct evidence, which is summarized below, does not support the widely-held view that a substantial fraction of terrigenous DOC is removed during estuarine mixing.

Sholkovitz et al. (1978) showed that the bulk components of DOC behave conservatively during mixing in the Amazon Estuary; although there was substantial removal of a humic acid fraction, which was probably colloidal, this material amounted to only about 5% of the DOM in the river input. Laboratory experiments on flocculation during the mixing of filtered sea-water and river-water, from the Amazon and from Scottish systems, support the environmental findings, with removal typically about 10% (Sholkovitz, 1976, 1978; Sholkovitz et al., 1978). Moore et al. (1979) found that DOC behaved essentially conservatively during mixing in the Beaulieu Estuary, England. For the Bristol Channel, Mantoura and Mann (1979) report a linear decrease in DOC over the salinity range of 1728 ‰, which extrapolates to a concentration of DOC of 6.9 mg litre^-1 for fresh-water. Concentrations of DOC at high salinities were above those predicted by the regression relationship for the lower range, which may be due to the effects of larger scale marine processes on the concentration in the sea-water end member. Data for the Tamar Estuary, presented by these authors and by Morris et al. (1978), provide no evidence for removal processes in the salinity range of 115 ‰ (see below with regard to processes at very low salinities). Subsequent studies have confirmed the essentially conservative behaviour of DOC in these estuaries and have shown similar behaviour in two other estuaries (Plym and Dart) in Devon (Mantoura, 1981, and personal communication). Laane (1980) has shown that DOC behaves conservatively during mixing in the Ems-Dollart Estuary and in the western Wadden Sea; Duursma (1961) had reported similar findings for the latter area. Mulholland (1981) states that about 20% of the DOC of a small stream in the south-eastern U.S.A. flocculated on mixing with sea-water.

A few measurements of DOC in samples of differing salinities from the North Dawes Inlet, Alaska, indicate essentially conservative behaviour in a system for which the concentration of DOC in the fresh-water input was lower than that in sea-water (Loder and Hood, 1972).

Thus, while a small fraction of high molecular weight humic material can undergo flocculation in estuaries, it appears that most of the terrigenous DOC is geochemically and microbiologically unreactive and enters the pool of DOC in coastal waters. Morris et al. (1978) have demonstrated that under certain circumstances, processes in the very early stages of mixing can lead to an increase in concentration of DOC; the findings were consistent with rapid heterotrophic utilization of this material. The reasons for this phenomenon, which may be related to exudation by micro-algae, and its generality require further investigation.

ACKNOWLEDGEMENT

REFERENCES

Laane, R. W. P. M. (1980) Conservative behaviour of dissolved organic carbon in the EmsDollart estuary and the western Wadden Sea, Neth. J. Sea Res., 14, 192-199. 

Loder, T. C. and Hood, D. W. (1972) Distribution of organic carbon in a glacial estuary in Alaska, Limnol. Oceanogr., 17, 349-355.

Mantoura, R. F. C., and Mann, S. V. (1979) Dissolved organic carbon in estuaries, in Severn, R. T., Dineley, D., and Hawker, L. E. (eds) Tidal Power and Estuary Management, Bristol, Scientechnica, 279-286.

Morris, A. W., Mantoura, R. F. C., Bale, A. J. and Howland, R. J. M. (1978) Very low salinity regions of estuaries: important sites for chemical and biological reactions, Nature, 274, 678-680.

Sholkovitz, E. R. (1976) Flocculation of dissolved organic and inorganic matter during the mixing of river water and seawater, Geochim. Cosmochim. Acta, 40, 831-845.

Sholkovitz, E. R., Boyle, E. A., and Price, N. B. (1978) The removal of dissolved humic acids and iron during estuarine mixing, Earth Planet. Sci. Lett., 40, 130-136.