Carbon Cycle

For a high-level tour of how every element moves through the tank, start with Nutrient Cycling. Carbon is the element that ties everything else together — it is the same thread running through your tank's oxygen, its pH, and the energy budget of every living thing in it.

Why carbon is the element that connects everything

Carbon is the universal connector in aquatic chemistry. It is not just a building block of organic molecules — it is the element whose transformations directly control dissolved oxygen, pH, and the energy budget of the entire ecosystem. When an alga fixes one molecule of carbon from CO₂, it produces exactly one molecule of O₂ — a relationship called the photosynthetic quotient. When any organism respires one molecule of organic carbon, it consumes exactly one molecule of O₂ and releases one of CO₂ back into the water. This one-to-one coupling means that carbon flow and oxygen flow are two views of the same process. You cannot understand one without the other.

Carbon also controls pH. The dissolved inorganic carbon pool — CO₂, bicarbonate, and carbonate — is the water's primary acid-base buffer. Photosynthesis draws down CO₂ and pushes pH up; respiration adds CO₂ and pushes pH down. In a small sealed container these swings can be dramatic — pH rising above 9 on a sunny afternoon as algae strip CO₂ from the water, then falling below 7 overnight as the whole community breathes. The carbonate system translates every molecule of carbon fixed or respired into a pH change, which in turn touches every organism's physiology.

This triple coupling — carbon to oxygen, carbon to pH, carbon to energy — makes the carbon cycle the integrating framework for the entire model. Diagnosing why oxygen is declining, why pH is swinging, or why a species is energy-starved almost always comes back to following the carbon.

The dissolved inorganic carbon pool

Dissolved inorganic carbon (DIC) is the sum of dissolved CO₂, bicarbonate, and carbonate in the water. The model tracks DIC as a single pool and splits it into the three forms based on pH whenever it needs the breakdown. Only dissolved CO₂ participates in gas exchange and is the main raw material for photosynthesis, though some species can also use bicarbonate through their carbon concentrating mechanisms.

For the full treatment of carbonate speciation, pH calculation, equilibrium constants, alkalinity, and buffering, see The Carbonate System.

Gas exchange

CO₂ moves between the dissolved pool and the headspace above the water. Only the dissolved-CO₂ fraction can cross the air–water interface — bicarbonate and carbonate are charged ions and stay in solution. The direction and rate of transfer follow Henry's law. For the complete gas-exchange model — solubility constants, temperature and salinity corrections, sealed versus open systems, and headspace dynamics — see Dissolved Gases and Gas Exchange.

What adds carbon to the dissolved pool

Respiration by every organism. Everything alive in the model — algae, periphyton, macrophytes, grazers, and bacteria — breathes around the clock. Respiration consumes O₂ and releases CO₂ at a one-to-one ratio, scaling with biomass and throttled by available oxygen. Organisms living in uncomfortable salinity breathe harder, paying the energetic cost of osmoregulation.

Decomposition. When detritus breaks down, a portion of its carbon is released straight to CO₂ — the fast, abiotic fraction of decay — while the rest passes through dissolved organic matter first.

Bacterial respiration of DOM. Heterotrophic bacteria consume dissolved organic matter, and how much of its carbon they keep depends on the kind. Labile DOM — fresh exudates, sugars, amino acids — is processed efficiently, with most of the carbon becoming new cells and the rest respired. Refractory DOM — humic and fulvic polymers — yields far less: the great majority of its carbon is respired as CO₂. When oxygen is scarce and bacteria cannot fully process their uptake, the carbon that would have become biomass is returned to the dissolved pool as fermentation waste instead. Bacteria also carry a baseline maintenance cost that releases CO₂ even when they are not growing.

DOM photodegradation. Sunlight breaks dissolved organic molecules apart, releasing their carbon as CO₂. This runs only in daylight and saturates at high light.

Denitrification and DNRA. Both of these anaerobic pathways burn organic carbon as an electron donor and oxidize it to CO₂. DNRA burns more carbon per unit of nitrate than denitrification does, because reducing nitrate all the way back to ammonium moves more electrons than reducing it to nitrogen gas.

What removes carbon from the dissolved pool

Photosynthesis. Algae, periphyton, and macrophytes fix dissolved carbon into organic matter using light, and this is by far the largest carbon sink. The rate depends on light, on nutrient availability through Liebig's law of the minimum, on how much CO₂ and bicarbonate are available, and — for periphyton — on surface area. Algae prefer dissolved CO₂ but can fall back on bicarbonate at reduced efficiency through their carbon concentrating mechanisms. Rooted macrophytes have a second supply line: they draw CO₂ directly from sediment pore water through their root aerenchyma, bypassing the water-column carbonate equilibrium entirely (see Pore-water CO₂, below). At high pH, when most of the DIC has shifted into bicarbonate and carbonate, photosynthesis can become carbon-limited. For light limitation, CO₂ limitation, photorespiration, and species differences, see Photosynthesis.

Carbon fixation by nitrifiers. Nitrifying bacteria are chemoautotrophs: they fix CO₂ into biomass using the energy from oxidizing ammonium rather than light. This is a small but genuine dark-carbon sink.

The organic carbon pools

Carbon exists in organic form across several pools:

Living biomass. Every species carries a carbon pool — its body mass — with a characteristic carbon-to-nitrogen ratio. Algae and periphyton hover around a structural C:N ratio near 6.6 but let it drift: when nitrogen is scarce they keep photosynthesizing and bank the surplus carbon as starch and lipids, driving the ratio up toward 14–20, then ease back down when nitrogen returns. This dynamic tracking is a key feature of the model. Consumers, by contrast, hold a fixed C:N and N:P through stoichiometric homeostasis, excreting any surplus nitrogen and phosphorus; bacteria hold a fixed ratio near 5.

Detritus. Dead organic matter, tracked as both carbon and nitrogen and split between suspended particles in the water column and settled material on the bottom. Suspended detritus decomposes faster; settled detritus decomposes more slowly, with reduced oxygen access.

Dissolved organic matter. Organic molecules dissolved in the water, split into two fractions with very different turnover. Labile DOM — fresh, low-molecular-weight compounds from algal exudates and cell lysis — is consumed rapidly by bacteria. Refractory DOM — humic and fulvic polymers — accumulates slowly, is consumed about ten times more slowly and at much lower efficiency, and absorbs light, darkening the water as it builds. DOM is the intermediate step between dead biomass and full mineralization back to CO₂ and inorganic nitrogen.

How carbon flows between the pools

The full path traces carbon from inorganic fixation, through the food web, and back to CO₂:

Fixation into biomass and DOM. Algae and periphyton fix dissolved carbon by day. The gross rate is trimmed by photorespiration — when oxygen runs high relative to CO₂, the enzyme RuBisCO mistakenly grabs O₂ and wastes some of the effort. The net fixation is then split: a small fraction is excreted as dissolved organic carbon (exudation) into the labile pool, and the rest goes to growth.

Grazing. Grazers eat algae and assimilate part of the carbon. The unassimilated remainder is egested as fecal pellets that become detritus, some staying suspended and the rest settling. Digestion itself carries a metabolic overhead — specific dynamic action — that burns roughly a fifth of the assimilated carbon, consuming O₂ and releasing CO₂.

Death. When organisms die their carbon enters the detritus pools, split between suspended and settled by body size. Bacteria are the exception: most of their dead biomass goes to DOM rather than detritus, because bacterial cells tend to lyse and spill their contents.

Decomposition. Of the detritus carbon that decomposes, most becomes DOM and the rest is mineralized straight to CO₂. The DOM share is further divided between the labile and refractory fractions, with settled material humifying more (and so producing proportionally more refractory DOM) than suspended material.

The microbial loop. Heterotrophic bacteria consume DOM — preferring it about two-to-one over suspended detritus — keeping a fraction of the carbon as new cells and respiring the rest. Sunlight runs a parallel abiotic shortcut, mineralizing part of the DOM it degrades directly to CO₂.

Detritus exchange. Suspended detritus slowly aggregates and sinks to the settled pool, while settled detritus can be resuspended by mixing — a slow two-way trickle.

The exact decomposition rates, growth efficiencies, exudation fractions, and stoichiometric coefficients behind all of these flows are tabulated in the Parameter Reference.

Pore-water CO₂ in planted tanks

In a tank with an organic soil substrate, the CO₂ produced by aerobic breakdown of soil organic matter is not released straight into the water. It is generated centimetres down in the sediment and has to diffuse up through the soil and sand cap before it can reach the water column. That physical lag lets sediment pore water hold many times more dissolved CO₂ than the overlying water — in an open tank the water sits near the trickle of atmospheric CO₂, while the pore water beneath it can be ten to fifty times richer.

From there, pore CO₂ has two ways out. Most of it diffuses slowly upward along the concentration gradient, subsidizing the water column (dissolved CO₂ diffuses faster through the sediment than ammonium does, so this flux is substantial even at modest concentrations). The rest is taken directly by rooted plants: species like Cryptocoryne and Vallisneria draw a large share of their photosynthetic carbon straight from the pore water through their root aerenchyma, bypassing the water-column carbonate equilibrium entirely. This is the primary reason Walstad-method tanks can sustain dense plantings of carbon-hungry species without any CO₂ injection. The diffusion can in principle reverse — if the water column ever holds more CO₂ than the pore water — but in an active soil tank it rarely does.

Key ecological dynamics

The day/night carbon oscillation

Every day the ecosystem swings between two metabolic states. By day, photosynthesis draws CO₂ out of the water faster than respiration puts it back: DIC drops, pH rises, dissolved O₂ climbs. By night, photosynthesis stops but respiration does not — every organism is still breathing, decomposition is still running, nitrifiers are still working — so CO₂ accumulates, DIC rises, pH falls, and O₂ declines.

In a well-buffered system (see buffering capacity) with moderate biomass these swings are gentle — perhaps a few tenths of a pH unit and a few mg/L of O₂. But in a small sealed container with dense algae and low alkalinity they can be severe. Daytime pH can climb above 9, stressing animals and shifting the carbonate balance so far toward carbonate that algae are starved of usable CO₂; nighttime pH can fall below 7, dissolving calcareous substrates and stressing shell-builders. The amplitude of this oscillation is one of the best signs of whether an ecosystem is in balance — large swings mean production and respiration are poorly matched.

Carbon limitation at high pH

When photosynthesis drives pH above about 8.5, most of the DIC shifts into carbonate, which algae cannot use directly, and the dissolved CO₂ that diffuses freely into cells can drop to near zero. At that point carbon itself becomes the limiting nutrient for photosynthesis, even with nitrogen and phosphorus to spare. Some species partly compensate by using their carbon concentrating mechanisms to reach bicarbonate, but even those become less effective as pH climbs.

This sets up a natural negative feedback: intense photosynthesis raises pH, which depletes CO₂, which slows photosynthesis. In open systems the feedback is weak, because the atmosphere continuously replenishes CO₂. In sealed systems it is much stronger — the headspace reservoir is finite, and once it is drawn down, carbon limitation can cap primary production well below what light and nutrients would otherwise allow.

How planted tanks bypass carbon limitation

In a Walstad-style planted tank, carbon limitation is partly sidestepped by the pore-water pathway described above. Soil organic matter decomposes below the sand cap and charges the pore water with CO₂ far richer than the water column; rooted plants tap that reserve directly through their root aerenchyma. Cryptocoryne draws the great majority of its photosynthetic carbon this way, Vallisneria a smaller but still substantial share, so both can keep photosynthesizing at near-full rate even when the water column is carbon-depleted at high pH. Whatever the roots do not take diffuses slowly upward, a steady carbon subsidy to the rest of the tank.

Methane: the anaerobic branch of the carbon cycle

Everything above describes aerobic carbon flow, where every unit of organic carbon ends up oxidized to one of CO₂. Anaerobic decomposition cannot finish that oxidation — there is not enough oxygen left to carry the electrons all the way. Instead, methanogens fork the carbon: roughly half goes to CO₂ and half goes to methane (Conrad 1999; Whiticar 1999). Methane is therefore the second terminal outlet for carbon in freshwater, alongside CO₂.

This is the carbon that bubbles. A hobbyist watching a freshly set-up planted jar will see, after a few weeks, small bubbles rising from the substrate at random intervals — not the reliable daytime oxygen of photosynthesis, and not the near-invisible nitrogen of denitrification. Those bubbles are largely methane, the visual signature of a maturing anaerobic substrate. Anywhere a sediment goes anoxic and exhausts its higher-energy electron acceptors — oxygen first, then nitrate, then iron oxides, then sulfate — methanogens take over and convert the leftover carbon to methane.

Where the methane sits

The model follows methane through three places. Pore-water methane is where methanogenesis output first accumulates, deep in the sediment. From there some diffuses up into the water column, where it dissolves; methane is only sparingly soluble, so in an open-top tank this stays near the trace atmospheric level almost regardless of how hard the substrate is making it. The third place is the headspace — the gas pocket above the water in a sealed jar, which is where the visible bubbles end up. There is no separate biomass currency for methane; it is a chemistry-only intermediate that exists only where carbon has been fermented out of detritus or soil organic matter under anoxia.

Where the methane comes from

Methane has two sources in the model. The first is the carbon that anaerobic soil-organic-matter breakdown always produced but historically had nowhere to go. Aerobic mineralization oxidizes every unit of carbon to CO₂; anaerobic mineralization, in real sediments, sends about half of it to methane instead. The model interpolates between the two on the basis of how oxygen-starved the substrate is — fully oxic soil routes all of its mineralized carbon to CO₂, fully anoxic soil splits it evenly between CO₂ and methane, and partial conditions fall in between.

The second source is settled detritus consumed directly by sediment methanogens, in parallel with — and competing with — sulfate reduction. In real lake sediments the terminal electron acceptors are used in a strict order, each extracting more energy per unit of carbon than the next: oxygen, then nitrate, then iron oxides, then sulfate, and finally CO₂ reduction to methane at the very bottom of the ladder. The model captures this ordering with two gates on methanogenesis — one that switches it off while nitrate is still around, and one that switches it off while sulfate reducers still have substrate. Methanogens take over only when both are exhausted. They are also noticeably more temperature-sensitive than denitrifiers or sulfate reducers (Bastviken 2004), so methane production ramps up sharply in a warm substrate. This pathway runs only where there is a soil layer to host it; a bare-bottom tank lacks both the pore volume and the deep anoxia required.

Methanotrophy: the oxic sink

Once dissolved methane reaches the oxygenated water column it is short-lived. Methanotrophic bacteria oxidize it back to CO₂ — methane and oxygen react to carbon dioxide and water — consuming two units of oxygen per unit of methane, with no effect on alkalinity. The half-life of dissolved methane in well-oxygenated water runs from hours to a couple of days (Hanson & Hanson 1996; Bastviken 2004), and the model places it in the middle of that range. The model captures the methanotrophic kinetics without tracking the bacteria themselves as a separate guild.

The strong oxygen dependence has a structural consequence: in a fully oxic water column almost every dissolved methane molecule is re-oxidized to CO₂ before it can reach the surface. The methane a keeper actually sees venting to the air or building up in the headspace is overwhelmingly the bubble fraction, which leaves the sediment as gas and bypasses the dissolved-phase oxidation entirely. A second methanotrophy step runs at the thin oxic skin at the top of the sediment, where oxygen from the water meets methane rising from below — a "cryptic" trap that re-oxidizes a large share of the pore methane before it can escape, mirroring the iron and sulfide traps described in their own cycle pages (Frenzel 2000; Bastviken 2004).

Ebullition: the bubbles

When pore-water methane climbs past its solubility limit, bubbles nucleate on substrate grains and rise, bypassing both the slow diffusion path and the cryptic re-oxidation trap. This is what aquarists see: small bubbles rising from a Walstad substrate after a few weeks, often a column of pearls along the glass when the substrate is disturbed. The model keeps pore methane pinned near saturation in a steady-state methanogenic substrate — bubbling is fast enough that methane does not build up indefinitely, but slow enough that a disturbance or a sudden warming triggers a visible burst rather than instant degassing.

In a sealed tank the bubbles accumulate in the headspace, a tracked pool that grows over the run. In an open-top tank they vent to the air and the carbon is genuinely lost from the system. Over a long open-top run, a methanogenic substrate becomes a real export route for carbon — comparable in magnitude to the nitrogen that denitrification bleeds away.

What the model leaves out

The methanotrophic bacteria are not modelled as their own species, so the grazers that feed on them in real methane-rich sediments have no biomass to draw from here. Hydrogenotrophic methanogenesis — a second route to methane that competes with the acetate route in real systems — is folded into the single pathway the model resolves, which makes the half-to-CO₂, half-to-methane split only approximately right where the hydrogen route is active. And like the rest of the anaerobic family, methanogenesis is gated on bulk-water oxygen as a stand-in for true pore oxygen; real sediments go anoxic within millimetres of the surface even under saturated water, so the model can under-call anaerobic activity in tanks whose water column stays oxic.

Key references

Bastviken, D. et al. (2004). Methane emissions from lakes: dependence of lake characteristics, two regional assessments, and a global estimate. Global Biogeochemical Cycles 18, GB4009.
Conrad, R. (1999). Contribution of hydrogen to methane production and control of hydrogen concentrations in methanogenic soils and sediments. FEMS Microbiology Ecology 28, 193–202.
Frenzel, P. (2000). Plant-associated methane oxidation in rice fields and wetlands. Advances in Microbial Ecology 16, 85–114.
Hanson, R.S. & Hanson, T.E. (1996). Methanotrophic bacteria. Microbiological Reviews 60, 439–471.
Jähne, B. et al. (1987). Measurement of the diffusion coefficients of sparingly soluble gases in water. Journal of Geophysical Research 92(C10), 10767–10776.
Sander, R. (2015). Compilation of Henry's law constants for water as solvent. Atmospheric Chemistry and Physics 15, 4399–4981.
Smits, A.J.M. et al. (1990). Internal aeration of aquatic macrophytes. Aquatic Botany 38, 1–17.
Walstad, D. (1999). Ecology of the Planted Aquarium. Echinodorus Publishing.
Whiticar, M.J. (1999). Carbon and hydrogen isotope systematics of bacterial formation and oxidation of methane. Chemical Geology 161, 291–314.