1) Naturally, rains wash out street grime and small pieces of rubbish which may easily bypass the sewage system altogether and end up in the river.
2) Cities have systems in place to treat wastewaters before releasing them into the rivers. However, these systems need to store yet-untreated water in reservoirs, and heavy rain may fill these up. Then, the only solution (if you don't want sewage to blast out sewer holes) is to release the water without treating it. In fact, one of the main upgrades made by Paris for the recent Olympics was a massive, 30-meter-deep new basin [1] built exactly to mitigate that risk.
[1]: https://www.reuters.com/world/europe/giant-reservoir-ready-s...
I guess in Copenhagen its water is coming from the Baltic so maybe it's less of a concern (short of container ships having issues near the city). But the Seine river flows through many communities before it reaches Paris. Many opportunities for actors to illegally dump things in the river and for all sorts of nasty runoff agricultural and otherwise to make it in periodically along the whole stretch.
Continuous water testing at the city limits also sounds impractical
Why? Surely a small team of a few people could have that as their job, and it would be completely worth it, to help keep all the people that want to swim safe.
And like, aside from the manual testing of the water that this team would do once or twice per day probably, they could have a few different sensors to catch some of the most likely kinds of pollution.
On top of that add hefty fines and maybe even a year or two of jail time for anyone caught dumping pollutants into the river, and you have a good disincentive against polluting the water.
And have a strong sense of pride of nature in the country so that people will naturally want to be mindful of the environment we live in. Promote this kind of thinking in schools and in public service announcements (billboards, ads, etc) from the government.
https://www.msn.com/en-us/news/world/toxic-substances-flowin...
A further advantage of satellite remote sensing for water quality is that it can be done by a neutral third party, without permission or access to the land, which can help bring a bit of objectivity to the politicized narrative around water quality we are seeing at the moment.
- multiparameter sondes (submersible sensors): More expensive, maintenance heavy, only measure at one point.
- Manual grab-samples: Again more expensive when you factor in labour, unreliable, single point.
Remote sensing has its disadvantages though; cloud cover can be problematic for example.
Other than "don't put sewage in the rain collector".
https://ing.dk/artikel/ingenioerer-maatte-kaempe-haardt-faa-...
The increase in water discharge also causes sediment to lift from the bed where e coli bacteria live happily in the sediment (think of shaking a bottle with solids that have fallen to the bottom). The potentially harmful bacteria is now suspended in the water for a few days, also growing in number as a result of the runoff nutrients until the increase in the water volume drops and the water can no longer suspend the sediment.
This cycle is the largest reason why, it's not just the runoff (which can also contain animial feces with e coli, etc), it's the runoff with nutrients stirring up the water and lifting e coli which then feasts on the nutrients.
https://www.usgs.gov/publications/escherichia-coli-bacteria-...