Lake Mendota is a roughly 9,800-acre lake in Madison, Wisconsin, and arguably the most intensively studied lake in the Western Hemisphere. Sitting at the head of the Yahara chain of lakes, with the University of Wisconsin campus lining its southern shore, it has generated more than a century of continuous ecological data. That record has made Mendota a living textbook for understanding how nutrient pollution, invasive species, climate change, and human management decisions interact in freshwater ecosystems. The story of the lake is, in many ways, the story of every overfertilized lake in the temperate world, played out in unusually sharp detail.
How Phosphorus Took Over the Lake
Lake Mendota’s central ecological problem is phosphorus. The lake’s watershed is dominated by agricultural land, and for more than a century, fertilizer and manure runoff have delivered phosphorus to the water. That phosphorus fuels the growth of algae and cyanobacteria, turning Mendota into what limnologists call a eutrophic lake. Research on the Mendota watershed found that the lake’s eutrophication is caused by only a small fraction of the total phosphorus moving through the landscape, but the amount of phosphorus that accumulates in the surrounding soil acts as a long-term reservoir, described by researchers as a “chemical time bomb” that can be mobilized by heavy storms, changing land practices, or chemical shifts in the soil.1Ecosystems. A Phosphorus Budget for the Lake Mendota Watershed
This is not just a matter of what enters the lake right now. The soil across the watershed has been accumulating excess phosphorus for decades, and even aggressive reductions in fertilizer application will not quickly drain that built-up supply. Studies of phosphorus flow in the Mendota ecosystem have concluded that only reducing the total amount of fertilizer phosphorus imported to the watershed will bring the system into balance at steady state. Soil phosphorus, because it turns over so slowly, controls the long-term flux to the lake and must shrink for eutrophication to be durably controlled.2Ecosystems. Phosphorus Flow in a Watershed-Lake Ecosystem
Making this worse, the lake itself has its own phosphorus memory. When algae and organic matter settle to the bottom and decompose, they consume oxygen. Once the deep water (the hypolimnion) goes anoxic, phosphorus locked in the sediments is released back into the water column, a process called internal loading. A modeling study that reconstructed Mendota’s water quality over many years found that even after nutrient loads from the watershed are cut, different aspects of water quality recover on very different timescales. Water clarity and algal biomass can improve within a few years of nutrient reductions, but deep-water oxygen levels take decades to rebound because the sediment phosphorus pools turn over so slowly.3Journal of Geophysical Research: Biogeosciences. Legacy Phosphorus and Ecosystem Memory Control Future Water Quality in a Eutrophic Lake Thermal structure also plays a role: how the lake stratifies in summer, and how long that stratification lasts, drives year-to-year variation in how severe the oxygen-depleted dead zone becomes.4Hydrology and Earth System Sciences. Lake thermal structure drives interannual variability in summer anoxia dynamics in a eutrophic lake over 37 years
The Algal Blooms That Close Beaches
For anyone living in Madison, the most visible symptom of Mendota’s phosphorus problem is the thick, green scum that can coat the lake’s surface on summer days. These are blooms of cyanobacteria, sometimes loosely called blue-green algae. They follow a predictable seasonal succession. After the lake stratifies in late spring, nitrogen-fixing genera like Aphanizomenon and Anabaena tend to dominate first, with Microcystis becoming more abundant by midsummer.5Limnology and Oceanography. Planktonic blue-green algae: Production, sedimentation, and decomposition in Lake Mendota, Wisconsin
Microcystis is the one that grabs public health attention, because it produces microcystin, a liver toxin. Research on Mendota found that microcystin concentrations rose sharply a few days after the first significant nitrogen-fixation events in the lake, then peaked when Microcystis became the dominant species. The handoff between Aphanizomenon (which fixes atmospheric nitrogen, enriching the water) and Microcystis (which exploits that extra nitrogen to bloom and produce toxin) makes the timing of toxic episodes tightly coupled to the internal nutrient cycling of the lake.6PLoS ONE. The Role of Nitrogen Fixation in Cyanobacterial Bloom Toxicity in a Temperate, Eutrophic Lake
Beach closures are a routine consequence. Local officials have been seeking better tools to anticipate blooms, and recent work has explored season-ahead forecasting models for cyanobacteria abundance on Mendota, partly to give beach managers the lead time to warn the public before conditions become hazardous.7MINDS@UW Madison. Cyanobacteria Abundance Modeling: Development and Assessment of Season-Ahead Forecasts To Improve Beach Management on Lake Mendota High-frequency sensor data collected over a decade from a buoy at Mendota’s center station revealed that cyanobacteria concentrations flip between two alternative states, one low and one high, and that transitions between these states are abrupt and largely driven by random factors. At daily timescales, cyanobacteria did respond to phosphorus input, precipitation, and wind, but the big-picture year-to-year pattern was dominated by noise.8Limnology and Oceanography Letters. Stochastic dynamics of Cyanobacteria in long-term high-frequency observations of a eutrophic lake That unpredictability is part of what makes management so frustrating.
The Biomanipulation Experiment
In the late 1980s, researchers and state resource managers tried something ambitious: rather than focusing solely on nutrient reductions, they attempted to restructure the lake’s food web from the top down. The idea was to stock large predator fish that would eat the smaller fish, which would allow populations of large-bodied water fleas (Daphnia) to grow, which would in turn graze on algae and improve water clarity. This approach, called biomanipulation, was one of the most high-profile lake restoration experiments ever attempted in North America.
Between 1987 and 1999, about 2.7 million walleye fingerlings and 170,000 northern pike fingerlings were stocked in Mendota. The project got a lucky boost at the start: in 1987, the cisco population suffered a massive natural die-off. Total planktivore (small fish that eat zooplankton) biomass dropped from roughly 300 to 600 kilograms per hectare down to about 20 to 40 kilograms per hectare. With fewer small fish eating them, the larger-bodied Daphnia pulicaria replaced the smaller Daphnia galeata mendotae as the dominant grazer. In many years that D. pulicaria dominated, water clarity improved during spring and summer.9Freshwater Biology. Stocking piscivores to improve fishing and water clarity: a synthesis of the Lake Mendota biomanipulation project
The project was considered a qualified success. Harvest regulations were essential to protect the stocked predators, because fishing pressure on walleye especially surged once the word got out. The large Daphnia persisted for more than a decade, and fishing diversity improved. But phosphorus levels in the lake also shifted during the same period, making it hard to separate the effects of the food web change from other environmental drivers. The biomanipulation showed that you could nudge a eutrophic lake toward clearer water through food-web management, but it could not substitute for reducing the phosphorus coming in from the land.
An Invader Unravels the Food Web
In 2009, the spiny water flea (Bythotrephes longimanus), an invasive predatory zooplankton from Eurasia, was first detected in Lake Mendota. It quickly reached densities higher than had been recorded in any other lake.10PubMed Central. Invasive species triggers a massive loss of ecosystem services through a trophic cascade The consequences were swift and severe. Bythotrephes feeds on Daphnia, the very grazer that the biomanipulation project had worked so hard to protect. With the spiny water flea consuming Daphnia pulicaria, the grazing pressure on algae weakened, and water clarity dropped by nearly a meter.11PubMed Central. Invasive species triggers a massive loss of ecosystem services through a trophic cascade
Analysis across a 40-year time series showed that the spiny water flea effectively reversed the trophic cascade that had been the backbone of Mendota’s clearer-water years. Zooplanktivory, the rate at which zooplankton are consumed, jumped back to about 47% of the level seen during the old cisco-dominated era, before the biomanipulation began. Both Daphnia abundance and water clarity showed strong negative relationships with this increased predation pressure.12Limnology and Oceanography. Invasive invertebrate predator, Bythotrephes longimanus, reverses trophic cascade in a north-temperate lake Researchers studying both Mendota and its downstream neighbor Lake Monona found that after invasion, diatom populations increased in both lakes, driving the clarity loss, while cyanobacteria levels did not change significantly.13Limnology and Oceanography. Uncoupling indicators of water quality due to the invasive zooplankter, Bythotrephes longimanus The invasion effectively decoupled two indicators that managers had been tracking together: nutrients and clarity no longer moved in lockstep, because biology had introduced a new driver.
There is no established method for removing spiny water fleas from a lake once they colonize it. The Mendota experience has become a cautionary example used to illustrate how a single invasive species can undo years of expensive ecological management.
Big Storms, Manure, and the Phosphorus Spikes
Even if baseline phosphorus loading were steady, the delivery of nutrients to Mendota is anything but smooth. Extreme rainstorms flush enormous pulses of phosphorus off the landscape and into the lake. Research tracking phosphorus loads and cyanobacteria over more than a decade found that about three-quarters of extreme phosphorus-loading events were followed, within one to sixty days, by an extreme bloom of cyanobacteria.14PubMed Central. Long-range dependence and extreme values of precipitation, phosphorus load, and Cyanobacteria The pattern was clear: a heavy rain event correlated with a phosphorus spike about a day later, and a cyanobacteria spike followed roughly two to four weeks after that.
What makes this especially concerning is that the effect of rain on phosphorus runoff is not just additive; it is synergistic with the amount of manure on the landscape. Field, stream, and lake-scale measurements in the Mendota watershed showed that when manure supply was high, the impact of intense rainfall on dissolved phosphorus concentrations was disproportionately large. The interaction was not significant for soil phosphorus or commercial fertilizer alone, pointing specifically to manure as the trigger for the worst pulses.15Environmental Research Letters. The synergistic effect of manure supply and extreme precipitation on surface water quality In a region where dairy farming is a major industry and climate models project heavier rainfall events, this synergy is a problem that is likely to intensify.
Ice, Heat, and a Changing Climate
Lake Mendota has one of the longest ice records of any lake in the world, dating back to the 1850s. Over a 96-year analysis period, ice cover duration and the date the ice melted in spring both trended significantly downward, while winter air temperatures and snowfall increased.16Journal of Hydrology. Coherence between lake ice cover, local climate and teleconnections (Lake Mendota, Wisconsin) The lake has been freezing later and thawing earlier, a pattern consistent with warming trends across the Northern Hemisphere’s lakes.
Shorter ice seasons are more than a recreational inconvenience. The timing and duration of ice cover influence how long the lake stratifies in summer, how deeply it mixes in spring and fall, and how much oxygen reaches the deep water. As described earlier, the thermal structure of the lake is a major driver of whether the deep water goes anoxic in summer and how much phosphorus gets recycled from the sediments. A warmer climate that extends stratification and shortens mixing can amplify the internal loading that already plagues Mendota.
A Hidden World of Microbes
Beyond the visible algal blooms, Mendota harbors a complex microbial community that researchers have tracked for years using DNA-based methods. A six-year study of bacterial community composition found that the lake’s microbes follow a repeating seasonal pattern, or phenology, that was more predictable than scientists had expected. Water column mixing, temperature, dissolved oxygen, and nitrate concentrations were the strongest drivers of change in bacterial communities from season to season.17Limnology and Oceanography. Interannual dynamics and phenology of bacterial communities in a eutrophic lake During the stratified months between May and September, communities from different years became progressively less similar, suggesting that random ecological drift plays a larger role when the lake’s physical environment is stable and warm.
More recent genomic work has gone deeper. Researchers assembled thousands of species-representative genomes from Mendota’s water column and found that many species and strains follow cyclical seasonal patterns. About one in five species showed a shift in which strains dominated over the span of a decade, indicating that evolutionary turnover happens on timescales that overlap with the ecological monitoring.18PubMed Central. Bacterial ecology and evolution converge on seasonal and decadal scales Lake Mendota’s long observational record makes it one of the few places where this kind of ecological-evolutionary overlap can be studied in nature.
The Lake as a Greenhouse Gas Source
Eutrophic lakes are not just recipients of pollution; they are also active participants in the global carbon cycle. During the 2016 ice-free season, researchers mapped surface concentrations of carbon dioxide and methane across Mendota on a roughly weekly basis. Carbon dioxide in the surface water was generally below what the atmosphere would dictate, meaning the lake was absorbing CO2 from the air during the stratified months. Methane, by contrast, was routinely far above atmospheric saturation and showed much greater patchiness across the lake’s surface.19Journal of Geophysical Research: Biogeosciences. Large Spatial and Temporal Variability of Carbon Dioxide and Methane in a Eutrophic Lake During fall mixing, when the stratification breaks down and deep water reaches the surface, both gases increased and became more spatially variable. The methane that had accumulated in the oxygen-depleted bottom water during summer was effectively released to the atmosphere in a seasonal burst.
This pattern means that the same eutrophication driving algal blooms and fish kills also makes the lake a stronger emitter of methane, a potent greenhouse gas. It is one more dimension of the phosphorus problem that tends to be invisible to most residents.
Why Decades of Work Have Not Fixed the Problem
Anyone looking at the history of Lake Mendota might reasonably wonder why, after more than forty years of awareness and active management, the lake is still green. The answer comes down to the mismatch between the speed of human intervention and the inertia of the phosphorus cycle. The watershed soil has been accumulating phosphorus for generations, and that reservoir releases nutrients to the lake slowly but persistently.20Ecosystems. A Phosphorus Budget for the Lake Mendota Watershed The lake’s own sediments hold their own legacy pool. Modeling work has shown that even after nutrient inputs are cut, the sediment pool takes decades to draw down enough for deep-water oxygen to recover.21Journal of Geophysical Research: Biogeosciences. Legacy Phosphorus and Ecosystem Memory Control Future Water Quality in a Eutrophic Lake
Meanwhile, the goalposts keep moving. The spiny water flea scrambled a food web that managers had spent a decade engineering. Climate change is delivering heavier rainstorms that mobilize more phosphorus per event, especially when manure is present on the landscape. And the fundamental economic driver of the problem, intensive dairy agriculture in the watershed, has not disappeared. Reducing fertilizer imports remains the only strategy that will bring the system into long-term equilibrium, but it requires sustained changes in farming practices across thousands of individual operations.22Ecosystems. Phosphorus Flow in a Watershed-Lake Ecosystem
Mendota as a Scientific Observatory
What sets Lake Mendota apart from thousands of other troubled lakes is the depth of its monitoring infrastructure. As part of the North Temperate Lakes Long-Term Ecological Research program, sponsored by the National Science Foundation, the lake has been monitored continuously since 1981. High-frequency sensors on a central buoy collect measurements at one-minute intervals during the ice-free season. Researchers have used that firehose of data to develop machine learning models for predicting dissolved oxygen and other water quality indicators, with performance comparable to traditional process-based models.23MINDS@UW Madison. Advancing Data and Tools for Machine Learning Modeling of Lake Water Quality This pairing of long historical datasets with modern computational tools has made Mendota a proving ground for techniques that could be applied to far less well-studied lakes around the world.
The ice records stretching back to the mid-nineteenth century, the phosphorus budgets, the bacterial genomics, the food-web experiments, the invasion response: each of these threads exists in richer detail at Mendota than at nearly any comparable lake. That density of information is both the scientific legacy and the practical value of the place. Lessons learned here, about how long legacy nutrients persist, how invasive species disrupt restoration, and how climate amplifies nutrient pulses, inform management of eutrophic lakes everywhere, even those with far less data to work from.