Sources and fluxes of available inorganic carbon

The sea-level, air-equilibrated concentration of carbon dioxide in water at 0 °C is ~23 |M, falling to ~13|M at 20°C (say, 0.5-1.0mg CO2 L-1, or ~0.15-0.3mgC L-1). Besides being sensitive to temperature, the equilibrium concentration depends upon the atmospheric partial pressure of CO2, which is, in turn, affected by altitude. In many natural waters, carbonic acid is the only free acid present and, thus, the concentrations of alkalinity (base ions), carbon dioxide

Hco3 Determination Water

Figure 3.17

The pH-carbon dioxide-carbonate system in natural waters. The relative quantities of the three components, CO2, HCO3- and CO32-, determine the pH of the water, as shown in the inset. Changes in the concentration of one component shifts the equilibrium. Photosynthetic withdrawal of CO2 can raise pH to the point where CO32- is precipitated as a calcium salt. For a more complete explanation, see Stumm and Morgan (1996). Redrawn from Reynolds (1984a).

Figure 3.17

The pH-carbon dioxide-carbonate system in natural waters. The relative quantities of the three components, CO2, HCO3- and CO32-, determine the pH of the water, as shown in the inset. Changes in the concentration of one component shifts the equilibrium. Photosynthetic withdrawal of CO2 can raise pH to the point where CO32- is precipitated as a calcium salt. For a more complete explanation, see Stumm and Morgan (1996). Redrawn from Reynolds (1984a).

and pH (acidity) are continuously interrelated. In this range, bicarbonate is generally the dominant anion (0 to ~3.5 meq L-1; higher alkalinities may be associated with alternative solute composition, sodium or potassium being the dominant cation, rather than calcium). Potentially, bicarbonate may dissociate to release free CO2, according to the reversible reactions shown in Fig. 3.17. In this way, bicarbonate in solution represents an exploitable store of dissolved inorganic carbon (DIC) of up to 42 mg CO2 L-1. For much of the time it serves to buffer the water to the mildly alkaline side of neutrality (pH ~8.3), potentially to the point of calcium carbonate precipitation (see Fig. 3.17). In base-poor lakes, this buffering capacity is proportionately weaker. In extremely acidic sodium-sulphate waters (pH < 4.5), there is (by definition) no alkalinity at all and DIC is present only as dissolved carbon dioxide or carbonic acid. Here, as in the first case considered, the equilibrated mass of gas in solution (again, generally ~0.3mgCL-1) complies with Henry's law and does not exceed the proportionality of the partial pressure of the gas in contact with the liquid.

At face value, the instantaneous carbon capacity of natural waters to support phyto-plankton is unlikely to exceed 0.3mgCL-1 (or about 0.02molCm-3). Where present, bicarbonate raises the DIC reserve up to 2 orders of magnitude greater. Supposing a C : Chla of 50, these capacities are equivalent to the supportive capacity for 6 to 600 |g chla l-1. Plainly, carbon limitation of the phytoplankton supportive capacity is hardly likely to arise among the many water bodies in the world in which biomass is severely restricted by 1 or 2 orders of magnitude (~0.6 to 6 |g chla L-1). Neither is the standing biomass of non-calcareous waters prevented from considerably exceeding 6 |g chla L-1. The instantaneous carbon dioxide availability in even the soft, non-calcareous waters may significantly exceed the air-equilibrated concentration and some production may be maintained when the DIC reserve is exhausted. This inspires queries about the internal sources of carbon dioxide and the rates of their replenishment.

The proximal sources of 'new' carbon include the solution of CO2 at the air-water interface, not just at the surface of the water body in question but in the rainfall leaching the atmosphere and falling directly or, indirectly, in the overland flow discharging into it from the surrounding watershed. The quantities transported to lakes can be considerable but the concentrations are still subject to equilibrium constraints. That fraction of the inflow made up from groundwater sources can become relatively enriched with CO2 under pressure (to the extent that some may vaporise when normal air pressure is encountered). Direct vulcanism (through fumaroles) can provide additional sources of carbon dioxide to sea water in certain locations.

Usually, the major source of DIC is derived from chemical weathering of carbonate rocks and debris in soils, including terrestrially sequestered atmospheric CO2, which is transported in run-off. Anthropogenically increased CO2 levels and accelerated erosion have contributed to historic sharp increases in alkalinity in some major catchments, including that of the Mississippi River (Raymond and Cole, 2003). Significant additional sources of carbon in lakes may come from deliveries of readily oxidisable organic carbon, both particulate (POC) and in solution (DOC). Anthropogenic sources (sewage, acid deposition, mine discharge) may be of local importance.

These various carbon sources are available to primary producers and, thence, to assimilation in aquatic food webs. In lakes, some of this carbon may be removed as organisms, their wastes or cadavers, either to the sediments, or to downstream transport, eventually becoming part of the POC flux to the sea. What is usually rather a larger part of the biogenically assembled carbon is respired by the producers or metabolised and respired by their heterotrophic consumers (grazers and decomposers), mostly back to carbon dioxide. This gas can now be vented to the atmosphere by equilibration. In lakes, especially, and at times of low biological activity, carbon dioxide is present in solution at concentrations considerably over those predicted by Henry's law (Satake and Saijo, 1974). In deep, oligomictic and meromictic lakes, hydrostatic pressure adds to the level of carbon-dioxide supersaturation that is possible. Mechanical release of such reserves, such as occurred at Lake Nyos, Cameroon, in 1986, carries dire consequences for people and livestock in the adjacent hinterland (see Loffler, 1988).

It becomes clear that there is a wide range of carbon availabilities among lakes and seas, as there is in the principal carbon sources. The stores and supplies can often be adequate but, at times, demand is capable of exhausting them faster than they can be replenished. This can be especially true in individual lakes having high biomass-supportive capacities and making strong seasonal demands on the carbon flux. Maberly (1996) constructed a balance sheet of annual carbon dioxide exchanges in Esthwaite Water (Cumbria, UK), a small (1.0 km2), stratifying (15 m), soft-water (alkalinity: 0.4 meq L-1) but eutrophic lake. During the autumn, winter and spring, free-CO2 concentrations of up to 120 |M (1.4 mg C L-1), almost seven times the expected atmospheric equilibrium, were observed. At such times, the lake would have been losing CO2 to the atmosphere. In contrast, photosynthetic carbon consumption in the summer typically depletes the epilimnetic DIC to very low levels (Heaney et al., 1986), occasionally to zero (pH ~10.3). At these times, the atmosphere becomes the main photo-synthetic carbon source.

Over the year, this lake probably loses three to four times more CO2 to the atmosphere (up to 2.8 mol m-2 a-1) than it absorbs (Maberly, 1996). No more than 4% of the annual production of biomass was found to be attributable to CO2 solution across the water surface. Most of the net resource influx arrives in the lake in solution in the inflow streams. As elsewhere, the main part of the annual load of free CO2 is roughly proportional to the hydraulic load and the bicarbonate load is, approximately, the product of the mean bicarbonate alkalinity in the inflow and the number of annual hydraulic replacements.

The idea that smaller lakes and rivers are not necessarily net sinks for atmospheric CO2 but, rather, may often be outgassing CO2 to the atmosphere, is relatively new (Cole et al., 1994; Cole and Caraco, 1998). In the wet tropics, catchment sources of CO2 can make an especially significant contribution to the dissolved content and to losses back to the atmosphere (Richey et al., 2002). In yet another study, Jones et al. (2001) calculated that net CO2 efflux from temperate Loch Ness may represent around 6% of the net ecosystem production of the catchment.

Proportionately, the amount of carbon dioxide loaded hydraulically must diminish with increasing size of the water body, as (presumably) the proportion of the carbon dioxide influx contributed by net inward invasion across the water surface increases. Yet it is plain that, even in Esthwaite Water, there are times of high carbon demand and low resource-renewal rate, marked by high pH values (~10) when the accelerated absorption of atmospheric carbon dioxide across the water surface must supplement the truncated terrestrial and internally recycled sources.

In water bodies much larger than Esthwaite Water, the hydraulic loads are relatively very much smaller and the oxidation of external POC

may be equally diminished on a relative scale. In this case, the exchange of carbon dioxide is mediated mainly by respiration and the dynamics tend to be dominated by metabolic turnover, the loss to sedimentary depletion of particulate carbon and the supplement of 'new', invading atmospheric CO2.

The direction and rate of gas-exchange flux (FC) across the water surface is governed by the relationship

where f is the solubility coefficient (in mol m-3 atmosphere-1), ApCO2 is the difference in partial pressure of carbon dioxide between water and air, and GC is the gas exchange coefficient, or linear migration rate (m s-1). In fact, the magnitude of actual exchanges is difficult to establish. However, the work of Frankignoulle (1988), Upstill-Goddard et al. (1990), Watson et al. (1991) and Crusius and Wanninkhof (2003), who are among those who have attempted to determine gas-transfer rates by reference to models or to the movements of sulphur hexafluoride tracer (SF6), provides important verifications. The seasonal variability in pCO2 becomes a crucially powerful driver in those instances when, as in Esth-waite Water, photosynthetic withdrawal from the aquatic phase takes the air-water difference to its maximum (up to 9 x 10-4 atmosphere, i.e. the fastest consumption stimulates the most rapid invasion of the lake). However, the transfer velocity is accelerated as a function of wind speed and surface roughness, from ~10-5ms-1, at wind velocities beneath the critical value of 3.5-3.7 ms-1, to an order of magnitude greater, at 15 ms-1 (Watson et al., 1991; Crusius and Wan-ninkhof, 2003). Given high values of ApCO2, the corresponding invasion fluxes are calculated to be in the order of 3-30 x 10-8 molm-2 s-1, or between 31 and 310mgCm-2d-1. Once again, using the approximate 50 : 1 conversion, this is theoretically sufficient to sponsor a productive increment of only 0.6 to 6 mg chlorophyll m-2 d-1.

These calculations fully amplify the observation that large crops of algae, especially in lakes of low bicarbonate alkalinity, do not just deplete the store of CO2 available, with a sharp rise in pH, but they create the conditions for carbon limitation of their own photosynthesis, at least until the demand falls or other sources of carbon can assuage it. In the open sea, where, it is alleged, atmospheric dissolution represents the major resource of new carbon, frequent strong winds may well fulfil one of the criteria of gaseous invasion. The typical low biomass represented by oceanic phytoplankton assemblages rarely raises the pH far above neutrality. Even so, the maximum rates of invasion under the conditions envisaged here can hardly be expected to supply much more than 100gCm-2 a-1.

0 0

Post a comment