• The Listener
  • North & South
  • Noted
  • RNZ

What's in our water? The new threats to New Zealand's drinking water

Kerry Mackintosh can move her legs again and her eyesight has improved a little. But she still can’t stand unaided and experiences terrible pain in the soles of her feet. Photo/Tim Whittaker

Havelock North has had one of the largest outbreaks of water-borne disease in the developed world, which is raising red flags around the country.

Kerry Mackintosh thought it must have been the butter chicken pie she’d eaten for lunch that made her sick. She rarely eats takeaways, but that day her husband, David, had bought her the pie for lunch from a local BP service station.

That was at 1pm on July 6. By 5.30 that evening, the Havelock North woman was violently ill. She and David were swiftly on the phone to the oil company, complaining that the pie must have been off.

By July 20, she was so dehydrated she had to be hospitalised and put on a drip. Tests taken at the hospital showed she had campylobacter. The short time frame between eating the pie and the onset of illness led doctors to conclude it was extremely unlikely to be the cause, as the incubation period for campylobacter ranges from one to 10 days. Mackintosh says because she has osteoarthritis, she is careful with her diet and seldom eats out. Other than the pie, there had been no variation from her normal regime of home-cooked meals.

She was sent home with a prescription for antibiotics, but within 24 hours of finishing the course, she was as sick as ever. By late July, she had numb fingers and toes, blurred and double vision, and every joint in her body felt inflamed. Her medical centre told her the numbness could be caused by hyperventilating. She was too crook to argue with them.

Not until August 12 did it occur to her that the tap water might have caused her illness. That was the day when routine water-safety testing for E coli by Hastings District Council showed the town supply was contaminated. By that time, the Hawke’s Bay District Health Board was becoming aware that gastrointestinal sickness was ripping through the town.

That evening, the council started chlorinating the supply, and a boil-water notice was issued.

In the week after the authorities declared the water crisis, Mackintosh’s GP began to suspect the campylobacter infection had triggered Guillain-Barré syndrome, a disorder in which the body’s immune system attacks part of the peripheral nervous system. At 2am on August 19, that suspicion was confirmed when she had a major attack – her limbs were flailing uncontrollably, she couldn’t speak and the electrical circuitry of her heart and lungs began to fail. David had to do CPR, and she was swept back to hospital by ambulance.

Mackintosh counts herself lucky – she was diagnosed early enough to be treated with immunoglobulin products. She has had five treatments and is slowly getting better. She can move her legs again and her eyesight has improved a little. But she still can’t stand unaided and experiences terrible pain in the soles of her feet – “like walking on coral” – when she puts weight on them.

“They think I will recover, but they are making it clear it’s a day-by-day situation,” she says.

Mackintosh at a Brookvale bore. Photo/Tim Whittaker

So many questions

Could Mackintosh have been sickened by her town water supply way back in early July – more than five weeks before the contamination was officially declared? The DHB regards her illness as unrelated to the outbreak, but her case raises just one of many questions swirling around this disaster.

Water testing results from June 14 until August 16 – which Mackintosh has been provided by Hastings District Council – show no sign of contamination up until August 12, the day the crisis was declared. But even setting aside the anomalous timing of her illness, serious doubts now hang over the accuracy of the testing regime, given that many people were falling ill days before the contaminated supply was detected.

Residents at Anglican Care’s retirement village and rest home in Havelock North were sick from August 8, and the facility went into lockdown on August 11. The Mary Doyle Lifecare retirement centre is also understood to have gone into lockdown on August 11, although the complex refuses to confirm this or answer any questions about how its residents have been affected by the outbreak. At Woodford House school, girls started presenting with symptoms on August 8.

Carol Winters, who runs Age Concern in the town, has built a detailed picture of how the contamination unfolded. Her organisation has been serving as an emergency distribution centre since August 16, and she estimates 600-700 people have come through the door seeking bottled water, handwash and anti-bacterial supplies. From the first days of the crisis, she started gathering stories of when people became ill and how they had been affected.

Carol Winters: many people told her they were sick significantly before the outbreak was officially declared. Photo/Tim Whittaker

She says many told her they were sick significantly before the outbreak was officially declared. She has met families who were ill for a week before the official declaration, including families with elderly members who became severely dehydrated, with one subsequently suffering a stroke. One household with members who had been sick for week before the crisis was recognised included a man suffering cancer, who was being rehydrated with “poo water” through a tube, says Winters.

She says many people had been trying to cope at home, believing they simply had a “tummy bug”. Some – particularly elderly people – didn’t even know of the water contamination and boil-water notice until they heard it through the grapevine over the weekend of August 13 and 14.

Given the one- to 10-day incubation period for campylobacter, these reports raise the possibility that the water may have been contaminated since early August or even earlier – and yet the test results from the two town bores and reticulation system reveal no sign of trouble until August 12.

By last week, Winters was seeing people who were struggling through a second wave of illness – in many cases having felt well enough to go back to work, only to relapse. This week, many of the people through the door have been under severe economic stress, including small-business people deprived of cash flow because of the crisis and large families struggling to cope with big doctors’ bills.

Photo/Tim Whittaker

Huge cost

Three weeks on from the discovery that the town supply was contaminated, two people have died of other causes while sick with campylobacter, 612 people are confirmed as having had the illness and 5198 people – a third of the town and half of all households – are estimated to have been affected.

One of New Zealand’s prettiest and most prestigious towns has been brought to its knees by a preventable breakdown in the supply of that most basic of human requirements: safe water.

Aside from the considerable human cost – particularly for the small proportion of people such as Kerry Mackintosh who may develop serious complications – the economic cost is likely to be large. Until the Havelock North disaster, one of the worst waterborne outbreaks of campylobacter in New Zealand was in the small mid-Canterbury town of Darfield. In 2012, it experienced an outbreak of campylobacter caused by the failure of a chlorination system, with 138 confirmed or probable cases out of a population of 1790. The cost of lost production and time off work has been estimated at up to $1.26 million.

Although Health Minister Jonathan Coleman declined to use his power to declare a drinking water emergency in Havelock North – which would have ensured efficient decision making and deployment of resources, and the opportunity for businesses to make insurance claims for loss of trade – health professionals say the crisis is of international significance. Canterbury medical officer of health Alistair Humphrey, one of the country’s most active defenders of drinking water quality against the increasing risk from agricultural contamination, says Havelock North ranks as one of the largest such outbreaks in the developed world. “There is a lot of international interest. This will be talked about at the big water conferences.”

Nigel French, director of the Infectious Disease Research Centre at Massey University, says there are parallels between the Havelock North crisis and the seminal water safety failure in the Canadian town of Walkerton in 2000. There, the town’s shallow bore was polluted with cow manure after heavy rain, and residents were infected with campylobacter and the toxic bacteria E coli O157:H7. By the time a boil-water notice was issued, the town had been exposed to contaminated water for days. Seven people died, 27 developed haemolytic uraemic syndrome (a serious condition characterised by renal failure) and 2300 became ill – some of whom continue to suffer from long-term illness to this day.

In Havelock North, the town’s shallow aquifer has been contaminated by faeces from ruminant animals such as cattle, sheep or deer.

Tracking down the cause of the system failure that has allowed Havelock North’s supply to be poisoned, with such far-reaching consequences, will require a detailed forensic examination by the inquiry to be set up by Attorney-General Chris Finlayson. Bruce Davidson, who led a grass-roots community response following the Walkerton water crisis and continues to advocate for community water quality, cautions against rushing to judgment and blame about the cause of the failure in Havelock North. He says the crisis needs to be investigated by a judicial inquiry capable of unpicking all elements of the system to find out where the breaches have occurred. He says it’s unlikely there will have been a single cause.

Alistair Humphrey: “Havelock North is one of the largest such outbreaks in the developed world.” Photo/Martin Hunter

Red flags

However, some early red flags appear to warrant closer investigation. Humphrey says New Zealand’s drinking water safety regime follows the “multi-barrier” approach, with the regulations requiring a series of defences to protect water supplies from contamination. The philosophy is equivalent to the Swiss cheese model of health and safety, whereby layers of defence – all of which will be fallible as long as they are operated by people who can make mistakes – must be maintained, so that failure at one point in the system can’t cause a catastrophe. The objective is to prevent the “holes in the cheese” from lining up and allowing calamity to occur.

The fact that almost 5200 people have been sickened in a public health disaster means that, by definition, the holes in the cheese of Havelock North’s water safety have lined up. Not only was the supply contaminated, but it appears the contamination went undetected for some time despite routine water safety testing by the district council, during which residents continued to drink the polluted tap water. And despite the suffering of the community over the past three weeks, French says the town has been saved by an element of luck – New Zealand has a high prevalence of the toxic E coli O157 in farm animals, and if that pathogen had been the contaminant in Havelock North’s water, it’s likely there would have been many more fatalities, as occurred in Walkerton.

Humphrey says the key lines of defence in public water safety operate at three levels: source protection (keeping the source of the water – in this case groundwater – from contamination); treatment; and reticulation and distribution.

Havelock North’s water was not chlorinated before the disaster because the aquifer it draws from was deemed under the drinking water standards to be a “secure supply”, unable to be penetrated from the surface by contaminants. Key questions for the inquiry will include how robust that assessment was, and what other shallow unchlorinated groundwater supplies around the country may also be at risk. The Ministry of Health currently has no idea of the number or location of such wells, although it has asked the Institute of Environmental Science and Research (ESR) to report on that.

Hastings District Council, as the supplier of the water, is responsible for complying with the drinking water standards and carrying out a range of stipulated quality tests, under the oversight of drinking water assessors employed by the district health board. Hawke’s Bay Regional Council, meanwhile, is responsible for the protection of the catchments of public water supplies and has the power to decline resource consents that can endanger those supplies.

Cattle on bare, muddy soil near the Tukituki River in Central Hawke’s Bay. Photo/Peter Scott

Consents queried

Retired water engineer David McBryde – who has worked for both Hastings District Council and Hawke’s Bay Regional Council – says one of the questions he would be asking in trying to track down the cause of the contamination is how the regional council has exercised its responsibilities under the National Environmental Standards for the protection of drinking water sources when granting resource consents for activities in the area around the town’s three water bores in Brookvale Rd.

One such consent that has him “scratching his head” was granted in 2015 for a 120-cow barn and the distribution of shed effluent, a few hundred metres upstream (in terms of the flow of groundwater) of the Brookvale Rd bores. The consent was issued to Te Mata Mushroom Company.

The dairy operation has not yet been developed, so it is not the source of the contamination, but McBryde believes it raises questions about the regional council’s judgment and how it has been administering its duties to protect the drinking water source.

Also falling under the regional council’s ambit are several other bores close to the district council’s town supply bores. A public meeting in Havelock North this week heard from the regional council’s Stephen Swabey that its investigations since mid-August had uncovered “quite a few bores” it hadn’t previously known about, which have been capped.

The aquifer is described as “semi-confined”, protected by a relatively thin layer of silt and clay. “So you are depending on the silt layers to keep contaminants and surface water out,” says McBryde. Any disturbance to that layer – such as earthworks – could therefore provide a potential pathway for contaminants into the aquifer.

Hastings Mayor Lawrence Yule. Photo/Tim Whittaker

Again, there are some potential red flags about the interaction between activities on the surface and the safety of the shallow groundwater supply just 20-odd metres below the ground. In 2013, one of the town’s three bores – Brookvale Road Number 3 municipal well – tested positive for E coli. Mayor Lawrence Yule says it was a brief period of contamination and no one got sick. E coli was detected again last October, and the council decided to close the bore and bring in engineering consultancy Tonkin & Taylor to investigate.

In a draft report released following the onset of the water crisis, Tonkin & Taylor says the thin layer of silts and clay overlaying the aquifer may have been breached, allowing infiltration of pollution from the surface. A large woodlot had been removed near the borefield, and there had also been earthworks at the nearby Te Mata Mushrooms site, which involved cutting a new drain and making a parking area.

Tonkin & Taylor says there has been an “increase in pollution potential” because of the earthworks, and “relatively shallow earthworks have the potential to provide a direct pathway for contaminants to directly enter the gravels above the borefield screens”.

Te Mata Mushrooms owner Michael Whittaker, whose business employs 120 people, has reacted angrily to the Tonkin & Taylor report, saying the suggestion that earthworks on his site could have changed the hydrology of the area is “bunk”. He says Tonkin & Taylor never visited his site or asked him any questions before writing its report. He says the mushroom company has suffered a major loss of business since the crisis, after it came under early suspicion as the source of contamination from chicken manure used in its compost. That was ruled out when interim genotyping by ESR showed ruminant animals were the source of the pollution.

Age Concern’s Carol Winters. Photo/Tim Whittaker

“Unusually high” E coli

Another potential warning sign lies in the results of water tests by Hawke’s Bay Regional Council at one of its monitoring bores near the Brookvale Rd water supply bores. On December 2 last year, the regional council detected “unusually high” E coli in the bore, and again on December 14, although at lower levels.

By the time those tests were taken, Hastings District Council had discovered the contamination of its Number 3 well, and had shut it down and instituted the Tonkin & Taylor inquiry – yet the dots between these two events don’t appear to have been joined.

Hydrogeologist Gil Zemansky, who is very familiar with Hawke’s Bay’s groundwater systems, questions whether there has been a pathway for contamination from the nearby Tukituki River. He notes that the soils above the shallow Te Mata aquifer are “at the high end of transmissibility”, with coarse gravels “that are not going to act as much of a filter”.

A water-contamination meeting in Havelock North on August 30. Photo/Tim Whittaker

Zemansky thinks it’s possible that during the dry winter, some of the silty clays in the top 3m could have cracked, enabling contaminated surface water to be carried by heavy rain along small natural drainage channels that run towards the borefield and to find a vertical path down through cracks into the aquifer.

Whether or not Zemansky is correct, it’s clear that water from the surface is getting in somewhere. Interim results of water-dating tests carried out by GNS for Hastings District Council, as part of its routine five-yearly monitoring of the security of the supply for town water purposes, has revealed the presence of very young water – under a year old. The previous round of testing, in 2011, put the age of the water at the Number 3 borehole at 49 years – meaning it has taken that long to filter through the soil from the Tukituki River and other recharge zones.

In a bizarre outbreak of internecine war in the midst of the water crisis, Hawke’s Bay Regional Council seized on the GNS age testing results last week as the basis for an investigation into Hastings District Council. While technical staff at the latter council were racing to get a temporary new water supply from Hastings city up and running, officers from the regional council turned up at the district council’s door and took them aside for questioning for several hours. District council insiders were continuing to complain this week about the amount of time key staff were being distracted by the regional council’s inquiry from the job of getting a safe water supply established.

A water-contamination meeting in Havelock North on August 30. Photo/Tim Whittaker

Agricultural pathogen risk

The Havelock North crisis also draws attention to the increasing risk to communities from agricultural pathogens spread by water, which animal and public health experts such as Humphrey, French and veterinarian Alison Dewes have been highlighting for years.

French lists the major waterborne zoonotic diseases (transmitted from animals to humans) in the New Zealand environment as campylobacter, salmonella, E coli O157 and leptospira, as well as the cryptosporidium and giardia.

Cattle are particularly rich reservoirs of pathogens harmful to human health. A study last year found 50% of New Zealand farms surveyed had calves with cryptosporidium in their faeces, 70% had rotavirus and 4% had salmonella.

The number of cattle in New Zealand’s agricultural landscape has increased significantly in recent years, and each animal excretes about 25kg of faeces each day – a total of about 84 million tonnes a year nationwide. Unlike human effluent, it is deposited straight onto the ground, where it sits untreated.

There is substantial evidence linking this to heightened public health risk. The rate of illness from the toxic E coli O157:H7 is increasing in New Zealand, and is more prevalent in dairy farming areas, according to a 2013 study. New Zealand also has a higher reported incidence of cryptosporidium and giardia than most other developed countries (see table below), and the rates are highest in dairy farming areas.

In expert evidence on Canterbury’s Land & Water Regional Plan, Dewes cited rates of infection from enteric diseases such as cryptosporidium and salmonella in the dairy-intensive Ashburton district that are twice as high as in the rest of New Zealand. She says the intensification of farming “amplifies mechanisms for the rapid evolution and spread of novel viruses”.

She says there are multiple pathways for animal pathogens to reach waterways, including being flushed by rainfall from cow tracks and pugged-up winter-cropping paddocks, leaching through stony free-draining soils, and passing through underground farm drainage systems.

“Water provides a key source of connectivity from the agricultural activity to the human populations,” she said in expert evidence this year on a proposal for the massive expansion of a dairy factory at Studholme, South Canterbury.

Further, several bugs are not detected by the E coli test relied on by public water suppliers as a proxy for contamination, including cryptosporidium, giardia, yersinia and leptospirosis, Dewes says. And some bugs remain viable in the environment for months, or even years, by adhering to sediment.

Alison Dewes: tests don’t catch all bugs. Photo/NZ Farmers Weekly

The nitrate problem

There is also widespread concern about nitrate levels in groundwater, particularly in Canterbury where 30% of shallow wells show increasing trends. High nitrates in drinking water can cause blue baby syndrome, where oxygen is not transported successfully around the infant’s body. The World Health Organisation stipulates 11.3mg/litre as the maximum level of contamination for drinking water supplies.

Humphrey says aside from the risk to small babies, “nitrates are the canary in the mine”, because when they are rising, other agricultural pollutants will also be rising. Though deeper groundwater tends to be safer because bugs have been filtered out, he says communities can’t be expected to continue drilling further underground to escape the consequences of contamination.

“Our drinking water is a precious resource … we can’t just keep going deeper. We have to protect the source.”

Agricultural pollution that requires ratepayers to pay for deeper wells or the installation of chlorine or UV treatment also means that those benefiting from the polluting activities – farmers and their investors – are pushing the cost of that pollution on to the public.


Large-scale irrigation has enabled the massive expansion of dairy farming in areas such as Canterbury. Even before the Havelock North water crisis, there was concern among some in the Hawke’s Bay community about the proposed Ruataniwha irrigation scheme and the prospect of more intensive farming through the district. The campylobacter outbreak has heightened those fears, with lobby group Transparent Hawke’s Bay calling for a moratorium on the Ruataniwha scheme and others questioning whether pathogens are flowing into the environment from intensive feedlot farms – where cattle have been photographed on bare, muddy soil – near the Tukituki River in Central Hawke’s Bay.

The latest State of the Environment report published by Hawke’s Bay Regional Council – whose investment company will be the developer and foundation investor in the proposed Ruataniwha scheme – shows water pollution is already on the rise. Twenty per cent of recreational water quality monitoring sites have worsening faecal contamination. Seven out of 22 sites were unsafe for swimming, and six showed significant deteriorating trends in E coli. Parts of the Tukituki catchment rank as “poor” for aquatic health, and two-thirds of monitoring sites showed excess nitrogen levels.

Walkerton’s Bruce Davidson warns against vilifying agriculture in the wake of the Havelock North crisis, but says communities have to think hard about how to protect the sources of their water “without cutting off economic activity”.

“Let’s be sensible about where we place or allow certain economic activities that may pose a threat to our water systems … What I’ve come to realise about these disasters is that the precautions we take are commensurate with the threat that we perceive. And in the developed world, we still don’t in many cases perceive a genuine threat to our drinking water. We think it’s something that exclusively happens in the developing world and in underprivileged countries where they don’t have the training and technology to keep water safe. We tend to think if it hasn’t happened here yet, it’s not going to, that we are somehow magically protected. But in many cases we are setting the table for these things with complacency, under-regulation and inappropriate or lack of source protection, and unfortunately they are going to happen.

“We don’t need to be doing this to our communities. We have the science. We have the technology. And it’s a lot cheaper to prevent these things.”

Follow the Listener on Twitter or Facebook.