A lake as a self-regulating system
A healthy freshwater lake is not just a body of water sitting still — it is an ecosystem holding itself in a clear-water state through feedback. Rooted plants and clear water stabilise sediment and support the zooplankton and fish that graze down algae; that grazing keeps the water clear enough for light to reach the plants; the plants keep stabilising the sediment. Eutrophication is what happens when a nutrient, almost always phosphorus, is added faster than the system can absorb it, and that feedback loop starts working in reverse.
The forcing: phosphorus loading
Phosphorus is usually the limiting nutrient in freshwater — the one in shortest supply relative to what algae need — so even a modest increase in phosphorus input from agricultural runoff, sewage or fertiliser can trigger disproportionate algal growth. As loading rises, phytoplankton populations grow, the water column becomes progressively murkier, and less light reaches the rooted plants on the lake bed.
Why the collapse is sudden, not gradual
Ecologist Marten Scheffer's models of shallow lakes show this system has two stable states for a wide range of phosphorus loading, not one: a clear state with abundant rooted vegetation, and a turbid state dominated by phytoplankton, with almost no stable configuration in between. As loading creeps up, the clear state does not fade gracefully — it holds, propped up by its own feedback, right up until a threshold is crossed, at which point rooted plants lose too much light, zooplankton lose their refuge from fish predation among the plants, algal grazing collapses, and the lake flips to turbid within a single growing season. This is a regime shift, a qualitative jump between stable states rather than a smooth response to a smooth cause.
clear state: plants + zooplankton grazing keep water clear → light reaches plants → state reinforces itself
turbid state: algae block light → plants die back → sediment resuspends → state reinforces itself
a fold bifurcation separates the two — the system snaps between them, it does not slide smoothly
Hysteresis: why cutting phosphorus back is not enough
The most consequential feature of this model is hysteresis. Because both states are stabilised by their own feedback, the phosphorus level that triggers a collapse from clear to turbid is higher than the phosphorus level that would allow a recovery from turbid back to clear. Once a lake has flipped, simply restoring phosphorus loading to its pre-collapse level is not enough — the turbid state's own feedback (resuspended sediment releasing internally stored phosphorus, absent plants providing no refuge for zooplankton) keeps it turbid. Recovery typically requires cutting phosphorus well below the original collapse threshold, sometimes combined with directly removing planktivorous fish or replanting vegetation, before the lake will flip back.
The end state: anoxia and fish kill
Left unmanaged, the turbid state gets worse over time. Dense algal blooms eventually die en masse, and the bacteria that decompose the dead algae consume dissolved oxygen faster than it can be replenished, producing anoxic (oxygen-depleted) bottom water. Fish and other oxygen-dependent organisms suffocate, sometimes in mass die-offs visible as fish floating at the surface, and the decomposing fish add yet more nutrients back into the water, reinforcing the cycle. This simulation lets you slide the phosphorus loading up and down and watch both halves of that story: the sudden collapse when the threshold is crossed, and the reluctance of the system to recover even once loading is reduced again.
Frequently asked questions
Why does a lake collapse suddenly instead of gradually getting worse?
Because the clear and turbid states are each stabilised by their own internal feedback — rooted plants keep water clear, which lets more plants grow; algae keep water turbid, which kills off plants. The system resists change within either state until a threshold is crossed, at which point the feedback that was holding it together starts working against it and it flips within one growing season.
Why doesn't reducing phosphorus back to the original level restore a collapsed lake?
This is hysteresis: the turbid state has its own feedback, especially phosphorus stored in sediment that resuspends when plants are absent, which keeps the lake turbid even after external loading drops. Recovery generally needs phosphorus cut well below the original collapse threshold, sometimes with active intervention like removing fish or replanting vegetation.
What causes the fish kills associated with eutrophication?
Dense algal blooms eventually die and are decomposed by bacteria, which consumes dissolved oxygen faster than it is replenished. The resulting anoxic water suffocates fish and other oxygen-dependent life, and their decomposition then adds further nutrients back into the water, extending the problem.
Try it live
Everything above runs in your browser — open Eutrophication and change the parameters while it is running. Nothing is installed, nothing is uploaded, the whole model lives in one tab.
▶ Open Eutrophication simulation