Bugs in the system: How do we make sense of recreational water quality?

By Waiology 23/10/2013 6


By Gary Bedford

Un-muddying the Waters : Waiology : Oct-Dec 2013When it comes to recreational use of freshwater, are we swimming in mixed messages?

Fairfax reported that ‘More than 60 per cent of monitored rivers in New Zealand are unsafe for swimming according to Environment Ministry figures’. Dr Mike Joy from Massey University reportedly stated that 95% of New Zealand’s lowland rivers fail the bathing standard due to pathogens. And the Consumer organisation claims that gradings applied to bathing spots are more meaningful than actual water quality analyses[1].

To add to the confusion, monitoring is not undertaken or reported consistently inter-regionally or at a national level. If a site fails to meet the bacteriological standard, that doesn’t actually mean it is ‘unsafe’. And there’s a degree of ignorance, and/or political grandstanding, over how good or bad NZ’s freshwater recreational spots actually are.

So what’s the real story?

The bible for freshwater recreational monitoring is Microbiological Water Quality Guidelines for Marine and Freshwater Recreational Areas, published by the Ministry for the Environment and the Ministry of Health in 2003.

The guideline is that no single sample should exceed 550 E coli per 100 mL. If this number is exceeded, then the public is to be warned of a potential health risk. At this level, the calculated health risk is that 1 bather in 20 would pick up a Campylobacter infection. Or to put it another way, 19 in 20 bathers wouldn’t. The criterion is a single point along a continuum of incremental risk. There is no sudden switch from ‘safe’ to ‘unhealthy’.

Swiming water quality samples are to be collected when people are likely to go swimming (K. Mischefski, Taranaki Regional Council).
Swiming water quality samples are to be collected when people are likely to go swimming (Taranaki Regional Council).

Samples are also to be collected ‘when people are bathing’. Hence there is debate over whether sampling should occur during just fine weather or any weather, and during the bathing season or all year round.

The Guidelines also set out a methodology for categorizing sites. It considers the surrounding land use, catchment inspections, and the worst of sampling results found over the preceding five years. Sites are then labeled from ‘very good’ to ‘very poor’, language which is meant to indicate the worst case (precautionary) scenario for health risk, but unfortunately implies the actual prevailing quality of bathing sites.

A recent report on bathing sites in NZ by MfE focused only on categories of freshwater sites, not the measured data. This became the trigger of some headlines noted at the start of this article. The Ministry did advise within its report that the categorisation:

  • “does not represent an accurate picture of water quality in the catchment
  • reflect a precautionary approach to managing health risk
  • are not designed to represent health risks on a particular day.
  • tend to reflect the poorest water quality measured at a site rather than the average water quality.
  • A site may be graded as poor but still be suitable for swimming much of the time
  • does not replace the site-specific information available on council websites”

There can be big discrepancies between categories and quality. For example, sites within intensive agricultural catchments will inevitably be graded ‘poor’ or ‘very poor’, even when the water quality is excellent. The categorisation processes do not account for land management practices that enhance water quality (e.g., riparian exclusion, waterway crossings, and riparian plantings). In Taranaki, four sites have never exceeded water quality standards in the last five years, yet all must be rated ‘poor’ or ‘very poor’ as bathing sites according to the Guidelines.

Where to from here?

There are a number of issues, some around science and some around communication.

  • E coli counts are probabilistic, not absolute, indicators of something nasty in the water. Emerging techniques for directly detecting pathogens take more time, cost much more, and are still only probabilistic, not absolute, in nature.
  • There are bathing water quality risks other than faecal microbial contaminants (e.g. some cyanobacteria) which aren’t well-sorted yet.
  • The Guidelines are for popular high-contact recreation. What about kayakers or kids paddling in the shallows?
  • How do we usefully and clearly communicate risk?

In a world demanding ever-greater certainty and less risk, there is obviously much to do. A review of the 2003 Guidelines is underway. Will it eliminate the bugs?


Gary Bedford is Director Environment Quality, Taranaki Regional Council.


[1] Correspondence: Suzanne Chetwin CE Consumer to TRC, 17 December 2008.


6 Responses to “Bugs in the system: How do we make sense of recreational water quality?”

  • Thank you Gary for the well balanced review of the recreational water quality guidelines and MfE’s reporting. It is a pity the mainstream press dont seem able to provide a similar balanced analysis. We have swum in the Rangitikei river east of Marton for decades and my daughter now spends many happy hours swimming and playing in the river. I have no idea what rating MfE gives the Rangitikei.

  • Thanks for feedback David. I had to trim my views on the mainstream media from the article due to pressure of space.
    Re the Rangitikei: I understand Horizons Regional Council have re-graded it recently- you could give them a call and ask for past and current gradings and how they were derived, if interested. It is actually regional councils that calculate ratings, but following MfE’s prescribed methodology.

  • Very Very good Gary.
    Best water quality article I have read in some time. Long have we needed this context.
    I’ve had questions around SIC values consistency and bias. It seems to me that once farming was viewed as a pristine environment and increasingly this view has changed. These eprceptions must have altered the SIC values and the outcome/trend of the data. I fear we view our swimming sites declining due to peoples perceptions. Not because of any scientific reasoning.
    I’ve queried MFE on this in the past. They’ve acknowledged a need to improve.
    I’d like MAC and SIC values to be presented separately. If they trend in opposite directions we need to ask questions.

    Mr E

  • Ten days ago, thousands of people were spending their feast holidays in the coast of the Red Sea.
    Unfortunately, several cities run their sewerage water into the sea and all people were swimming in dirty lagoons
    God Help

  • Gary I was told to read it and I am glad I have. So impressed at the clear way you unmuddy things I have just tweeted it!

  • Hmmm. Would you let your kids swim in water where they had a 1 in 20 chance of picking up Campylobacter? And if there’s Campy present, chances are there’s something even nastier like Salmonella. I note the limits for “very good” according to MfE are 130 E coli/100ml. I further note the TRC’s propensity for glossing over poor water quality in Taranaki, it’s poor standard of science and its unwavering support for the oil industry.