Did cutting phosphorus starve Lake Geneva's algae?

Find your issue, and check it is measured

An environmental issue: something happening, to something, because of something, with proof that somebody measures it.

In the 1970s Lake Geneva was choking on phosphorus from sewage, detergents and farms, and algae bloomed. The phosphorus has since been brought down to less than a fifth of its peak, and the argument now is whether it has gone too far.

Proof it is measured

Plankton in Lake Geneva, opened: the INRAE lake observatory at Thonon sampled the deepest point of the lake once or twice a month from 1974 to 2010, and put phosphate, phytoplankton, water fleas and temperature in one file.

In the news

RTS, 30 March 2026: the lake's phosphorus fell from nearly 100 µg/L in 1978 to 15.6 µg/L in 2024, its lowest level since 1970, while the water set temperature records.

The system

Inputs: phosphorus from treatment works, from detergents (until 1986 in Switzerland, 2007 in France) and from farmland, carried in by the Rhône and smaller rivers. Stores: phosphate dissolved in the water; the phytoplankton. Flows: algae take up phosphate as they grow, water fleas graze the algae, fish eat the water fleas. Outputs: the outflow at Geneva, and dead algae sinking to the bed. Feedback: positive. Dead algae rot on the bed and use up the oxygen in the deep water, and without oxygen the sediment releases phosphorus back into the water, which feeds more algae.

Write your research question

One sentence naming the indicator and its units, the years and the places. No strategy in it.

The relationship between the change in winter dissolved phosphate (µg P/L, January to March) and the change in growing-season phytoplankton biomass (µg/L, May to September) from one year to the next, in the middle of Lake Geneva, 1974 to 2010.

Independent variable

The change in winter dissolved phosphate from one year to the next, in micrograms of phosphorus per litre, averaged over January to March: the phosphate on offer before the growing season starts. Most of it arrives in sewage and runoff. Human.

Dependent variable

The change in growing-season phytoplankton biomass from one year to the next, in micrograms per litre of the upper water, averaged over May to September. Natural.

Why winter

In summer the algae take phosphate out of the water they grow in, so a summer reading is partly the algae's own doing. January to March is the phosphate in the water before they start.

Why changes

Phosphate fell through most of these years, so its raw values would correlate with anything else that changed steadily over the same years, and the question would really be about time. Comparing each year with the one before takes the steady fall out.

Not obvious

The fishers' argument says less phosphate, less algae. But the algae also answer to light, temperature and grazing, and which kinds grow changes as the water clears. Phosphate may lead the algae closely, or hardly at all.

Find a real-world strategy and its tensions

One real-world strategy people are genuinely arguing about, two named positions, and what the tension does.

Lake Geneva's phosphorus target: 10 to 15 µg per litre, reached by removing phosphorus at sewage works and keeping it out of detergents, and now under review.

The strategy

CIPEL, the commission through which France and Switzerland look after the lake together, set a target of 10 to 15 µg of phosphorus per litre in its 2011 to 2020 action plan. Its 2021 to 2030 plan commits to setting a new phosphorus objective for the decade. The target is met at sewage works, which remove phosphorus, and in the shops: Switzerland banned phosphate from laundry detergent from 1 July 1986, France from 1 July 2007. The lake's phosphorus peaked at 89.5 µg/L in 1979.

Position 1 · economic, social

Mickaël Dumaz, president of the association of professional fishers on the French Alpine lakes (AAIPPLA), to France 3, 26 June 2025: "l'eau était de plus en plus claire, cela signifie qu'il y a moins de phytoplanctons": the water is clearer every year, which means less phytoplankton, the food of the zooplankton the fish eat. He asks for more of the phosphorus that sewage works remove to be let into the lake.

Position 2 · environmental, scientific

CIPEL's scientific council, in its report on the 2023 campaign (2024): now that algal blooms seem to be under control, "le seuil de 15 µgP·L-1 apparait comme étant un bon compromis", 15 µg per litre looks like a good compromise between a fish stock worth the name and the lake's other uses. Below 10, it found, fish populations are generally too weak.

The tension

Fish against clear water. Stéphan Jacquet of INRAE put it as a choice (RTS, 7 September 2025): swim all year with no algal blooms, or accept some days without swimming to help the fishing trade.

What it does

Delay: the 2021 to 2030 plan promised a new phosphorus objective, and CIPEL's scientists said it would reconsider the old one in 2025, and its report that December still gave no new number. Uneven costs: the fishers bear the smaller catches, while clearer water is shared by everyone who uses the lake.

For the environment

More phosphorus would mean more algae and, when they die and sink, less oxygen in the deep water, which has gone years without the complete winter mixing that refreshes it.

Link to the question

The fishers' case is a chain: less phosphorus, less phytoplankton, less food for fish. The question tests its first link, in the years the phosphorus came down.

Select and document your data

A dataset you can defend: counted, documented, and cut to a rule fixed before you looked.

Source

INRAE Observatory on Lakes (OLA), Biomass of phytoplankton assemblages in Lake Geneva 1974-2010, on Recherche Data Gouv: one CSV, downloaded with this link. Open it as semicolon-separated and Latin-1 (Western European), or every µ in the headings turns into strange characters.

The columns

Year, Month, PO4 (dissolved phosphate) and PhytoTot (total phytoplankton). Leave the rest.

Read the column

PO4's heading says µgP/L, but the values are milligrams: 0.034 is 34 µg per litre. Multiply by 1,000 and say so. NaN means missing, never zero.

Inclusion rule

A winter counts if at least two of January, February and March have a phosphate value: five winters are missing one month, and all 37 qualify. Every May to September is complete. The change from 2001 to 2002 is left out, because phytoplankton was sampled in the top 10 m up to 2001 and the top 18 m after, so that one change is partly the method. Fixed before looking.

Derived quantities

For each year: the mean PO4 of January to March, and the mean PhytoTot of May to September. Then each year's value minus the year before's.

How the numbers were made

The observatory's own staff sampled the deepest point in the middle of the lake. Phosphate was measured at a series of depths from surface to bottom by a standard laboratory method; the file does not say which depths its one value stands for. Phytoplankton was collected with a sampler that takes the whole upper layer at once, and its biomass worked out from the species found.

Controls

Pair: each year is compared with the year before, in the same place, so whatever stays the same from one year to the next drops out. Restrict: one site, the same months every year, and one change left out where the method moved.

Credit

Anneville, O., Dur, G., Rimet, F. and Souissi, S. (2019), Recherche Data Gouv, doi:10.15454/AOATKG, Etalab Open Licence 2.0, with the date downloaded.

Treat your data

Run a statistical test, leading to figures you built and a statistical result.

Spearman's rank correlation, two-tailed, at p = 0.05.

Why this test

A relationship between two measurements on the same 35 pairs of years. Year-to-year changes throw up a few large ones (a wet spring, a warm summer), so a rank test is safer than Pearson's.

Decided before running it

n = 35. Critical value about 0.33 (check it in the table you cite). A positive correlation would mean the algae rose and fell with the phosphate. Free calculators: Social Science Statistics, Geography Fieldwork.