With the 10km marathon swim and the triathlon swim leg at the Paris Olympics both held in the Seine, there was a huge amount of talk about the river's water quality and the effort that went into cleaning it up to make it swimmable. It's a genuinely impressive story, a city spending well over a billion euros to stop raw sewage overflowing into a river that hadn't been swum in for a century. But it wasn't a completely clean result. Elevated bacteria levels forced the cancellation of practice swims after heavy rain, and Belgium's triathlon team ended up withdrawing from the mixed relay after one of their athletes, Claire Michel, fell ill and was hospitalised following her swim, reportedly with an E. coli-related infection. A Swiss triathlete who'd also swum the Seine withdrew sick as well. So it's not quite the fairy tale ending some of the coverage made it out to be, it's a real example of how quickly water quality can turn on you, even with world-class monitoring in place.
Now, what do I actually know about water quality? I'm a swimmer, but before I started this wetsuit brand, I completed a science degree majoring in environmental geoscience, with honours in hydrogeology and hydrogeochemistry. I then worked for five years in the industry across mining, water supply and remediation. So I know a reasonable amount about water quality, and about rocks. What follows is how I personally think about my own risk, and the process authorities actually use to make these calls, not medical advice. If you're reading this as a swimmer, you swim at your own risk, and you have to make your own judgement call every time you get in.
What EPA Victoria (and most Australian authorities) actually measure
Australian water quality guidelines, set nationally by the NHMRC, are built around a bacteria called enterococci. It's found in the gut of humans and animals, and it's used as an indicator of faecal contamination because it survives longer in salt water than other bacteria like E. coli, and it correlates more reliably with actual illness risk in marine environments. Authorities aren't testing for every possible pathogen in the water, that would be neither practical nor affordable, they're testing for this one indicator bacteria as a proxy for overall contamination risk.
The NHMRC guidelines break the results into bands. Category A, at 40 organisms per 100mL or under, carries a gastrointestinal illness risk of under 1%, and this is generally treated as the “good to swim” threshold. From there it climbs through Category B and C up to Category D, above 500 organisms per 100mL, where the risk is high enough that swimming isn't recommended at all. EPA Victoria monitors around 36 Port Phillip Bay beaches through summer, reporting results twice daily, but the actual lab result for a water sample takes about 24 hours to come back. That's an important detail: the number you see on the app or the website today is often based on a forecast model, not a fresh sample from that exact hour.
Why the forecast isn't the same as a real-time test
The forecast model is built on genuinely good scientific data, historical sampling, rainfall, wind and tidal patterns all feed into it, but it isn't based on site-specific testing at the exact time and place you're about to swim, because that would be far too expensive and slow to be useful. Rainfall is the single biggest factor. EPA Victoria advises against swimming for 24 to 48 hours after rain, because stormwater runoff, carrying everything from animal waste to litter to overflow from ageing sewer infrastructure, is the main way bacteria levels spike. This is exactly what happened in the Seine before the Olympics, and it's the same mechanism, just at a different scale, that drives a swim advisory going up at your local Australian beach after a big storm.
This is also why a triathlon organiser running an event in Port Phillip Bay can't simply ignore an EPA swim advisory. If the EPA has recommended against swimming based on their data and an organiser runs the swim anyway, and a competitor gets sick, the event's insurance is very likely voided. An organiser can commission their own independent testing, using a lab and a consultant to interpret site-specific results, which could technically override a general advisory since it's specific to that location and time. But for most organisers, that's not practical. It costs real money, and even a sample taken the night before can be out of date if conditions change overnight. In practice, it's simply easier and safer for organisers to cancel or postpone when the advisory is against them.
Where I sit with it personally
For myself, I'm pretty relaxed. I've swum after rain before, but that's a risk I'm choosing to take with a decent understanding of what's actually driving that risk, not a blanket recommendation for anyone else. If you're newer to open water or you're swimming with kids, I'd lean on the official advisories rather than my own comfort level. Check your local water quality reporting before you swim (EPA Victoria's Beach Report is the one I use), give it extra time after heavy rain, and stay away from stormwater drains and river mouths, that's where contamination concentrates. The water is very likely fine most of the time. But “likely” is doing some real work in that sentence, and it's worth understanding why before you decide how much risk you're comfortable with.
Useful references
● Check today's water quality (EPA Victoria's Beach Report) — epa.vic.gov.au/check-air-and-water-quality
● How EPA Victoria monitors and forecasts water quality — epa.vic.gov.au/how-we-monitor-water-quality
● Water quality forecast ratings explained (Good/Fair/Poor, rain advice) — epa.vic.gov.au/water-quality-forecast-ratings
● NHMRC guidelines for managing risks in recreational water — nhmrc.gov.au/about-us/publications/guidelines-managing-risks-recreational-water