Allelopathy

For a general introduction to water chemistry, see Chemistry. Allelopathy is the chemistry behind one of the most satisfying things a planted tank can do — clear its own algae — and one of the most insidious things a cyanobacterial bloom can do: poison the very animals that would otherwise eat it.

My hornwort cleared my algae bloom

Ask around the planted-tank community and you will hear the same folklore again and again: someone fighting green water dumps in a handful of hornwort, and within a couple of weeks the haze lifts. Stem plants "out-compete" algae, the story goes. Part of that is true and ordinary — a dense thicket of fast-growing plants shades the water and strips it of nitrogen and phosphorus, leaving algae starved and dim. But part of it is chemical warfare. Many aquatic plants leak compounds into the water that directly suppress algae, and many cyanobacteria leak compounds that sicken and kill the small grazers that would otherwise crop them back. This is allelopathy — competition not over resources, but by poison.

It matters because it is a positive feedback, and positive feedbacks are what give an ecosystem its character. A planted tank that suppresses algae chemically tilts the whole system toward staying clear and plant-dominated; a cyanobacterial bloom that poisons its grazers tilts the whole system toward staying turbid and bloom-dominated. Without this chemistry, a simulator can only explain plant-versus-algae outcomes through shading and nutrient drawdown — and it under-predicts just how decisively a dense planted tank can shut algae down, and how a bloom can escape the grazers that should have ended it.

Two weapons, two stories

The model represents two distinct allelochemicals, each produced by one side of an old ecological rivalry and aimed at a different target.

Polyphenols are the plants' weapon. Rooted and submerged macrophytes release phenolic compounds — the same broad family of molecules that make tea and oak leaves astringent — that suppress the growth of competing algae. They are the chemistry behind "my plants cleared the water." Hornwort (Ceratophyllum) is the archetype, the most chemically aggressive plant in the model; Vallisneria and Cryptocoryne are milder; and floating plants like Salvinia and duckweed commit very little to chemical defence, because their real weapon is the shade they cast from above (Hilt & Gross 2008).

Cyanotoxins are the cyanobacteria's weapon. Blooms release microcystin and related toxins that kill small grazers — Daphnia, copepods, rotifers, shrimp — and, at lower doses, stop them feeding before they die. This is the chemistry behind a bloom that should have been eaten but wasn't: in many real blooms microcystin kills Daphnia faster than copper does (Rohrlack 2003), releasing the cyanobacteria from the top-down grazing control that limits ordinary algae.

The two weapons point in opposite directions — plants suppress producers, cyanobacteria suppress consumers — but mechanically they are the same idea: a dissolved chemical whose concentration in the water gates someone else's growth, feeding, or survival.

Where the chemicals come from, and how they fade

Both allelochemicals are made of carbon the producer fixed in photosynthesis and then chose to spend on chemical defence rather than growth. A small fraction of each producer's fixed carbon is diverted into its allelochemical pool — a fraction of a percent for the plants, a little more for cyanobacteria, which keep the great majority of their toxin locked inside their cells and leak only a trickle into the water. Because the chemical is fixed carbon, releasing it costs the producer carbon, and when it later breaks down that carbon returns to the dissolved inorganic carbon pool. Nothing is created or destroyed; the carbon budget closes on its own.

Neither chemical lasts forever, and the two fade in different ways:

• Polyphenols are bleached by light. They photodegrade, breaking down under sunlight on a timescale of roughly a week — faster in bright surface light, slower in the shaded depths (Wetzel 1992). A planted tank therefore has to keep producing polyphenols continuously to maintain a suppressive level; switch off the lights or thin the plants and the chemical defence fades within days.
• Cyanotoxins are eaten by bacteria. Microcystin is broken down by an adapted microbial community over a couple of weeks, and the breakdown speeds up in warm water the way most microbial processes do (Edwards & Lawton 2009). This longer persistence is part of why a bloom's toxins can keep suppressing grazers even in the lulls between bursts of release.

The exact release fractions, decay rates, and half-lives are tabulated in the Parameter Reference.

Polyphenols: the planted-tank advantage

A polyphenol does not kill algae outright. It throttles their photosynthesis, and it does so in a threshold-like way: below a certain dissolved concentration the algae barely notice it, and above that concentration their growth is sharply curtailed. The same idea governs every concentration-dependent gate in the model — the Monod and dose-response kinetics used for nutrients and for copper toxicity — so a polyphenol behaves much like a mild, plant-made herbicide whose strength rises steeply once it crosses the effective threshold.

Different algae have different tolerances. In the model, delicate planktonic greens and diatoms are the most readily suppressed, benthic films somewhat more resistant. Crucially, the macrophytes are immune to their own polyphenols — a plant does not poison itself — so the chemistry is purely an attack on the competition. (Cyanobacteria, for their part, are insensitive to polyphenols and fight on a different axis entirely.)

This is the mechanism behind the alternative stable states that make shallow planted systems so bimodal. A clear, macrophyte-dominated tank and a turbid, algae-dominated one are both self-reinforcing: the planted state keeps algae down through shade, nutrient drawdown, and polyphenol suppression, while the turbid state shades out the plants that would otherwise fight back. Allelopathy is one of the feedbacks that locks a system into the clear state and makes the flip between the two abrupt rather than gradual (Scheffer 2004; Hilt & Gross 2008). It is, in miniature, why throwing hornwort at green water can tip a whole tank.

Cyanotoxins: chemical warfare on the grazers

Cyanobacteria face a different problem from algae. They are not especially good competitors for light or nutrients, but they are nearly inedible — and where they are eaten, they fight back. Their toxins act on the grazer guild in two stages.

First, at sub-lethal doses, the toxin suppresses feeding. A grazer exposed to dissolved microcystin slows or stops filtering before it shows any other sign of harm (Gilbert 1990). This alone is enough to release the bloom: grazers that have stopped eating cannot crop it. Second, at higher doses, the toxin kills. Cyanotoxin mortality is added to the same budget as the model's other chemical killers — copper and hydrogen sulfide — and for the most sensitive animals it can be the dominant cause of death during an intense bloom.

Sensitivity runs in a characteristic order. Daphnia and rotifers are the most vulnerable; copepods and dwarf shrimp like Neocaridina are somewhat hardier. The cyanobacteria themselves are immune to their own toxin. The result is the canonical top-down release: a bloom mass-produces a chemical that disables and then kills the very animals whose grazing would have ended it, and so the bloom persists far longer than nutrient exhaustion or self-shading alone would allow. Reproduce this in a simulator and a cyanobacterial bloom stops behaving like ordinary algae and starts behaving like a real one.

The dose it actually takes

The two allelochemicals are calibrated on different evidence, and it is worth being explicit about which, because it changes how the numbers should be read.

Polyphenols are keyed to the measured laboratory dose. Nakai and colleagues (2000) isolated the four compounds Myriophyllum spicatum releases and measured what each does to a culture of Microcystis. Over 72 hours, half-inhibition arrived at 0.65 mg/L for pyrogallic acid, 1.0 mg/L for gallic acid, and 5.1 and 5.5 mg/L for ellagic acid and catechin. Those are the numbers this model uses, and the simulator reports its polyphenol pool in the same unit they are measured in, gallic acid equivalents, so the two are directly comparable.

For years the standard reading of the field data was that real weed beds suppress algae at only tens of micrograms per litre, a hundred times below the lab dose, and that synergism, chronic exposure and boundary-layer hot spots make up the difference. That reading has not held up. Bergeron and colleagues (2025) looked for the effect directly in two Québec lakes heavily infested with Eurasian watermilfoil and found no causal link to cyanobacteria at all. The polyphenols they could detect in the water sat 30 to 35,000 times below the concentrations that inhibit growth. So the model no longer assumes a hidden potency multiplier. A living plant's exudate is a real but weak channel here, and the shading and nutrient drawdown a dense planted tank provides do nearly all of the work of holding algae down.

Where polyphenol suppression does become strong is when you add the chemical in bulk rather than waiting for a plant to make it. That is what leaf litter does, and it is why barley straw and decomposing leaves are long-standing pond treatments for nuisance algae (Ridge et al. 1999).

Cyanotoxins are still keyed to effective dose. Purified microcystin needs single-digit micrograms per litre to kill Daphnia in a 24-hour assay (Rohrlack 2003), yet the dissolved toxin measured in the open water of a bloom is usually well below one microgram per litre (Burford 2014). Here the gap is mechanically explained: most of a bloom's microcystin stays locked inside the cells until something eats them, so a grazer's real dose is set by what it ingests rather than by what a water test reads. The cyanotoxin numbers in this model should be read as effective, bioavailable dose, not as something you could match against an assay of a water sample.

Which algae are hit hardest

Cyanobacteria. This is the consistent finding across the polyphenol literature, and it is the reason barley straw and leaf litter are usable pond treatments rather than blanket sterilisers: they knock back the blue-greens and largely spare the eukaryotic algae. In the model, cyanobacteria are half-suppressed around 1 mg/L of dissolved polyphenol, diatoms around 2, planktonic green algae around 3, and surface-attached green algae around 4. Attached algae are the least affected because the biofilm they sit in is itself a diffusion barrier.

Polyphenols versus cyanotoxins at a glance

How allelopathy interacts with the rest of the tank

• Producer competition — the headline coupling on the plant side. Polyphenol suppression sits on top of the ordinary competition for light and nutrients, so a dense planted tank fights algae on three fronts at once. This is what makes the clear, macrophyte-dominated state so stable, and the flip to a turbid state so abrupt. See Photosynthesis and Aquatic plants.
• Grazing and the food web — the headline coupling on the cyanobacterial side. Cyanotoxins remove or disable the grazers, changing who eats whom and letting a bloom run unchecked. The feeding-suppression effect bites before the lethal one, so a bloom can release itself from grazing long before any animals actually die.
• Carbon — both allelochemicals are spent fixed carbon, and both return their carbon to the dissolved inorganic carbon pool when they break down. The amounts are small, but they are part of the carbon book and are accounted for there.
• Other chemical stressors — cyanotoxin mortality is added alongside copper and hydrogen-sulfide toxicity on the grazers, so a bloom in a tank already carrying another stressor compounds the pressure on the animals.

How allelopathy sits in the budget

Both allelochemicals are pure-carbon tracers: a producer spends a little of its fixed carbon to make them, and that carbon returns to the dissolved inorganic carbon pool when the chemical decays. There is no separate nitrogen or phosphorus bookkeeping, because at realistic release rates the nutrient content of these pools is a negligible sliver of the producer's budget. A scenario with no plants and no cyanobacteria carries no allelopathy at all — the pools simply stay empty until a producer that makes them is present and growing.

Further reading

• The Carbonate System — where the carbon spent on allelochemicals returns when they decay
• Aquatic plants — the macrophytes that produce polyphenols, and the shading that is the floating plants' real weapon
• Photosynthesis — the producer growth that polyphenol suppression throttles
• Stability and Failure — where the clear-vs-turbid flip and the runaway bloom sit among the ways a tank can swing
• Parameter Reference — every release fraction, decay rate, and tolerance threshold behind this page, with citations

Key references

• Bergeron, C., Proulx, F., Behmel, S. & Lavoie, C. (2025). No evidence of an allelopathic effect of Eurasian watermilfoil (Myriophyllum spicatum) on cyanobacteria in two Canadian lakes. Journal of Freshwater Ecology 40, 2520925.
• Burford, M.A. et al. (2014). Understanding the winning strategies used by the bloom-forming cyanobacterium Cylindrospermopsis raciborskii. Harmful Algae 38, 70–79.
• Edwards, C. & Lawton, L.A. (2009). Bioremediation of cyanotoxins. Advances in Applied Microbiology 67, 109–129.
• Gilbert, J.J. (1990). Differential effects of Anabaena affinis on cloning and sexual reproduction in the rotifer Brachionus calyciflorus. Journal of Plankton Research 12, 1023–1033.
• Gross, E.M. (2003). Allelopathy of aquatic autotrophs. Critical Reviews in Plant Sciences 22, 313–339.
• Hilt, S. & Gross, E.M. (2008). Can allelopathically active submerged macrophytes stabilise clear-water states in shallow lakes? Basic and Applied Ecology 9, 422–432.
• Nakai, S. et al. (2000). Myriophyllum spicatum-released allelopathic polyphenols inhibiting growth of blue-green algae Microcystis aeruginosa. Water Research 34, 3026–3032.
• Ridge, I., Walters, J. & Street, M. (1999). Algal growth control by terrestrial leaf litter: a realistic tool? Hydrobiologia 395/396, 173–180.
• Rohrlack, T. et al. (2003). Role of microcystins in poisoning and food ingestion inhibition of Daphnia galeata caused by the cyanobacterium Microcystis aeruginosa. Applied and Environmental Microbiology 65, 737–739.
• Scheffer, M. (2004). Ecology of Shallow Lakes. Springer.
• Zhu, X. et al. (2020). Combined effects of allelopathic polyphenols on Microcystis aeruginosa and response of different chlorophyll fluorescence parameters. Frontiers in Microbiology 11, 614963.
• Wetzel, R.G. (1992). Gradient-dominated ecosystems: sources and regulatory functions of dissolved organic matter in freshwater ecosystems. Hydrobiologia 229, 181–198.