New Zealanders love the coast.
We’ve spent decades building stuff hard up against it - underground and on the land. But it’s getting too close.
Right now, the sea is advancing on the tens of thousands of people who live within 50cm of the high tide mark.
And it won’t stop.
Insurers will give up on whole communities. Taxpayer bailouts will almost certainly follow.But we’re nowhere close to figuring out how we’ll manage the reshaping of our country.
When the sun is out, Tracey Reeves perches atop a seawall with her fishing rod, resting her coffee on one of the large, jagged boulders holding back the sea.
On a weekday morning, locals are walking their dogs along the concrete footpath trailing the top of the seawall. Waves on a recent high tide scattered some of the rocks, which are being replaced by contractors with a digger.
One of the great attractions of life in Moanataiari, a suburb of Thames on the Coromandel coast, is that it's surrounded by water. For an angler like Reeves, who lived in Auckland for decades where she built boats, the appeal of a quiet life by the water was obvious.
The seawall is a dominating feature of Moanataiari because it's so tall. From her fishing spot, Reeves can keep an eye on the television in her living room, which is several metres below her.
The suburb was built on reclaimed land, which is slowly sinking; Some of the land is now below sea-level, meaning that at high tide, the water level can rise above the houses. From above, it has a peculiar effect: Moanataiari looks like a mixing bowl floating in a sink full of water.
For Reeves, her new life in Moanataiari was perfect until the first major storm. It arrived on the morning of January 5, 2018, at the same time as a king tide.
It was expected to be a big one, but at first, it seemed to have been over-egged. Although the waves steadily grew higher, there was no rain, just an ominous grey shroud. Then it started to pick up. Northerly winds fuelled waves that screamed down the firth and smashed against the seawall, every so often flooding onto the road and leaking through the top of the seawall.
At its height, waves were crashing over the wall and down onto the road.
"It was the first storm we've experienced of that size," Reeves says.
"We sit here and we're below the water when the tide comes in, and that day I think every seventh wave was higher than the wall. It was quite frightening."
Reeves took several videos before the storm hit its peak. In one video, she scans the top of the seawall where the rocks are scattered everywhere, including her coffee table. "It'll never be the same," she says.
Another video shows a council officer rushing to his car and driving away as a large wave crashes over the wall.
Moanataiari has long been on a war-footing with the sea - it has a dedicated pumping station and industrial sized drains on each side of the road. When a storm comes, locals know to park their cars further away, so they don't become trapped. The day after the storm ended, the tide receded, and everything was quiet. Reeves was mowing the lawn, and business continued as usual.
In a time where extreme weather events are becoming more frequent as a consequence of human-induced climate change, the long-term fate of communities like Moanataiari is uncertain.
Buying into a suburb below sea-level wasn't a concern for Reeves. She did her research, told the bank, and felt it wouldn't be a problem in the near future. She acknowledges the sea is rising, but wants to live in her house forever, so it doesn't matter if it loses value.
The storm, however, was a clear harbinger of what was to come. Although it won't worry her, she can see the impact on future generations, as the storms get worse.
"I sort of worry for the kids," she says.
"I do believe [the water] is probably going to come in. In the next 30 years I think the high tide will become an issue."
The damage at Moanataiari rippled further up the firth, too.
A long, narrow road traces the coastline for dozens of kilometres, linking Thames and the various bays along the Coromandel peninsula.
Much of that road is protected by seawalls, joined together in a long chain. Near Moanataiari, in a community called Te Puru, the seawalls couldn't stop the waves, which shredded the tarsealed road like soggy Weet-bix. One resident, at the time, said she "hasn't seen it this bad for 20 years".
On the other side of the firth, the coastal community of Kaiaua was devastated by flooding. Waves swept into houses in the low-lying community: "[I]t's never been this bad," a local firefighter said. At the bottom of the firth, between Kaiaua and Thames, seawater flooded the farms on the low-lying Hauraki Plains; the salt in the water razed the pasture, leaving some farmers unable to milk their cows.
Many months later, the coastal Thames road is still being fixed. The repairs to the road alone so far have cost nearly $20 million.
The insurance cost of that storm was around $35m - it remained the most damaging storm of the year for just three weeks, when ex-Cyclone Fehi caused around $46m of insured damages.
2018 is on track to be the most expensive year on record for insurance losses due to extreme weather, nearing a quarter of a billion dollars and close to surpassing the previous record set in 2017.
Moanataiari is one of many New Zealand communities confronting a growing problem: The steadily rising sea.
As the climate warms, the sea will rise, likely with increasing speed. It is not a matter of if, or when, but how fast.
Sea-level rise is certain because there's a clear, historical link between a warmer climate and higher seas. The cause is two-fold: The ocean is warming, and water expands when it's warm, a process called thermal expansion. The other is that land-based ice - mostly glaciers and ice-sheets - has melted dramatically, transferring water once tied up on land into the sea. In Antarctica, where much of the world's land-based ice is found, around 2.7 trillion tonnes of ice has melted into the ocean since 1992, according to recent research.
Over the last century, the sea hasn't risen much. Around New Zealand, the average annual rate for much of the century was around 1.5mm.
It has long been expected that the sea would start to rise faster, which is shown in the latest data - seas around New Zealand are now rising 3-4mm a year.
"The trend has accelerated," says Dr Rob Bell, a coastal oceanographer at Niwa, which supplies sea-level rise data to the government.
"We've doubled the rate in the last 25 years."
There are nuances. Weather patterns such as El Nino and La Nina can counteract sea-level rise; parts of New Zealand are subsiding, meaning effective sea-level rise can be higher than average, and other areas are being uplifted, partly negating sea-level rise.
But there is a clear, long-term trend: The seas are rising, and doing so more quickly than they once did.
We know that millennia ago, when temperatures were similar to what they are today, the sea was around 5m higher. The speed at which the world has warmed in the last century is unprecedented, and the sea is slowly catching up.
The main thing to worry about with rising seas are storms. The sea-level is the base from which major events like storm surges spring. When the base is higher, so too is the surge.
The January storm was particularly bad due to an unfortunate combination of events.
A low pressure system had moved from the sub-tropics, over unusually warm seawater, and coincided with northerly winds and a king tide. The result was the highest sea-level recorded in Thames since at least 1938.
A few weeks later, a similar situation happened along the Tasman coast, near Nelson - ex-Cyclone Fehi became much more damaging when it arrived at a particularly high tide, causing storm surges to flood over seawalls.
These events interest climate change researchers because they show what happens when sea-levels are higher, particularly in coastal areas. A higher sea means coastal flooding becomes more common, to the point where it will become the major source of flood damage in the coming decades.
"Waves, tides, everything are riding on the back of this higher baseline sea-level," Bell says.
"By the middle of this century, as sea-level increases, coastal flooding will become the dominant coastal hazard risk. Once you start to get to half a metre or so, extensive areas will get flooded."
Ken Richard's Granity home was swamped by the ocean.
Ken Richard's Granity home was swamped by the ocean.
New Zealand will be spared some of the most severe impacts of a warmer climate - such as extreme high temperatures - but it is particularly exposed to sea-level rise.
New Zealand has the world's ninth longest coastline, and millions of people living within 5km of the coast.
More than three-quarters of the major population centres are coastal, and tens of thousands of people live within 50cm of the mean high tide mark, the area most at risk of sea-level rise. New Zealanders love the coast, and have manoeuvred their communities to live as near to it as possible.
Perhaps the best evidence of this is found in the country's most common street name: Beach Road.
The consequence is that billions of dollars of infrastructure, including residential housing, is in areas most immediately threatened by rising seas. They include airports, roads, underground pipes, parks and reserves, community facilities and social housing.
The most recent estimates, based on a draft report set to be published next year, show that around 125,000 buildings - both commercial and residential - would be at risk of flooding with 1m of sea-level rise. The replacement value of those buildings is around $38 billion.
Another metre of sea-level rise on top of that would put a further 70,000 buildings at risk, with a replacement cost of $26b.
The speed at which the sea rises is vitally important, because it dictates how much time is available to relocate these assets.
Based on the amount of greenhouse gases already in the atmosphere, scientists project with high confidence around 20-30cm of sea-level rise by 2065, regardless of how emissions are reduced.
The uncertainty is in what happens next. In a best case scenario, where emissions are quickly reduced, sea-level rise could be slowed to around half a metre by 2100. In a worst case scenario, where business as usual continues, the rise could reach between 1m and 2m, or possibly even higher, by 2100.
Public planning documents typically end at this point, about a century from now. That becomes a problem for future generations.
But the thing that makes sea-level rise such a concern is that it likely won't stop there; it will keep going, long after everyone who is currently alive is dead, rising higher and higher for centuries, reshaping the coastal boundary and dictating where future New Zealanders can and can't live.
"The critical thing around sea-level rise is that it's not going to stop," says Dr Judy Lawrence, a climate change researcher at the Victoria University of Wellington.
"It's going to keep going for centuries, and we don't know exactly when the reduction in emissions will start to kick in and slow that down."
It has prompted a flurry of questions among scientists and researchers: How do you fairly plan for a future that no one alive today will experience? And who will bear the cost of a slowly unfolding disaster that, unlike the disasters we're familiar with, is entirely predictable?
Experts say that the impact of sea-level rise will become clear long before the sea washes over the land. And it will probably start underground.
New Zealand has a vast network of underground stormwater and wastewater pipes, many of which were designed long ago and have aged considerably since they were installed. Like the communities around them, they were built near the coast, under the assumption that the sea would stay where it was.
Most of the pipe network is gravity fed, meaning the pipes start high and tilt downwards, ending at the lowest point, usually at the coast or near a river to discharge into water. They are particularly exposed to sea-level rise, simply because they are closer to the sea, or tidally influenced rivers, than anything else.
When the sea rises, corrosive saltwater is more likely to reach the ends of the pipes and infiltrate the network, causing flooding and pipe damage. In some areas, such as South Dunedin, the integrity of the pipe system is crucial.
South Dunedin is more exposed to sea-level rise than any other large community, despite the fact it's not connected to the sea. The problem there is groundwater; like the sea, it rises and falls with the tides, and has been steadily rising.
Because the groundwater is so high, rain has nowhere to go, and ponds on the surface. If the pipes break, the water will amass.
Adding to this problem is that rain events are projected to become more extreme in most parts of New Zealand, particularly in winter months. When high seas, storm surges, and heavy rain combine, the effect can be too much for existing infrastructure to handle.
Replacing these at-risk assets will be a heavy burden for local authorities - the value of the pipes at risk is unknown, but those most likely to be affected include large areas of Christchurch and Wellington.
If they are not fixed, or moved, broken pipes can cause a variety of environmental and health issues. Another issue is that it would mean greater use of pumps, which are expensive, and usually run on diesel, which further pollutes the atmosphere.
Above land, a major issue won't be infrastructure, but finance.
Insurance, in the most basic sense, is protection against uncertainty; it's a way to transfer low-probability risks, like damage from a fire, an earthquake, or a major flood to someone else.
By definition, insurance covers events unlikely to happen. That the sea will rise is certain.
There is clear evidence that extreme weather events causing damage are becoming both more severe and more frequent. The Insurance Council releases its claims data every year, which show that storms causing tens of millions of dollars in damage now happen several times a year.
The greatest threat is when a community as a whole loses insurance, a phenomenon known as "insurance retreat".
There is no evidence this has happened anywhere in New Zealand, but there are growing signs it will. Insurance premiums have risen in some areas following major storms, like the one that hit Edgecumbe in 2016. When premiums start to rise more broadly in coastal areas, it's a price signal - the market is saying that living on the coast is riskier than it used to be.
Sea-level rise researchers believe insurance retreat will be among the first major consequences of sea-level rise in New Zealand, because it unleashes a cascade of other issues.
For insurers, it's a mathematical problem: At what point does an unlikely event become too likely, turning a profit into a loss? Insurers do their own risk modelling, and likely have a sense of when and where retreat will happen, but are reluctant to share that data.
It has left policy makers and researchers working on their own estimates.
Belinda Storey, a researcher from the Victoria University of Wellington, is examining the point at which insurance retreat is likely around the country. She says it's likely sooner than many might expect.
"Insurance won't provide cover for a certainty; they'll only provide for probability," she says.
"Once that probability reaches a point where it's very likely, then they simply won't provide insurance at all. So insurers are going to pull out of those locations that are most affected by climate change well before people are ready to move."
A major problem is that insurance policies are annual, but the mortgages on the houses they apply to last decades. A house can have insurance one year and not the next, but its mortgage will remain.
That shackles insurance to the banking sector: When a house loses insurance, it leads to a technical default, because having insurance is often a requirement for a mortgage. The bank is left with the house, which has likely lost value.
For the person living in the house, getting another mortgage is tricky, because they can't use an uninsured property as collateral for a loan. If the house is damaged in a storm, they'll have to pay the repairs; without private insurance, they are not covered by EQC, either. Continuing to protect the house may require costly defences, such as a sea-wall, paid for out of pocket or through targeted rates. The cost of coastal living grows, while the value of their asset drops.
The local authority, which makes infrastructure decisions based on long time-periods, may decide to stop investing in the area; the roads decay and the pipes rot. There's a possibility that the only people living on the coast become those with no other options.
"You may have people moving into those locations who can't afford to live anywhere else," Storey says.
"You may end up having more vulnerable populations in the most hazardous locations."
As part of her research, Storey has been mapping maximum credible weather events onto different parts of the country, accounting for sea-level rise.
One way in which insurers decide whether to provide cover is a factor called the Annual Exceedance Probability (AEP). It's a confusing term, but it's basically an estimate for the likelihood of a major event – like a storm or an extreme water level – happening in a given year.
It's usually framed as a time period. The storm in Thames, for example, was a '100 year storm'. It means the likelihood of a storm that size happening in any given year is 1 per cent.
A 100-year storm has an AEP of 1 per cent; a 50-year storm has an AEP of 2 per cent; a 20-year storm, AEP 5 per cent; and so on.
The major concern for both insurers and authorities is that when the sea rises, these events become much more frequent, alarmingly fast.
Just how frequent varies across the country.
NIWA research, commissioned by the Parliamentary Commissioner for the Environment in 2015, looked at how likely a '100 year extreme sea-level' (1 per cent AEP) today would be in the future, with different levels of sea-level rise.
Even though sea-levels are rising fairly evenly around the country, two cities are particularly disadvantaged due to their short tidal ranges: Wellington and Christchurch.
They will face these more common extreme sea-levels long before Auckland, which is the city least sensitive to sea-level rise, but all major cities will reach that point eventually.
The NIWA research showed that in those cities, today's 100-year event will happen every year with around 16cm of sea-level rise (likely reached within 20 years) and every fortnight after 33cm (likely reached within 35 years). By the end of the century, it is likely every high tide in Wellington and Christchurch will be as high as the most extreme tide we can feasibly imagine today. The new extreme tides then will be, much, much higher.
Auckland and Dunedin are slower: But a 100-year event could be expected annually after 30cm of sea-level rise in all locations.
Belinda Storey believes that insurers are likely to withdraw cover when a claim is likely after a 50-year event or a 20-year event.
For some parts of the country, that threshold will be reached with 10cm of sea-level rise, which is possible within a decade or two: "We would argue that 5 per cent AEP [20-year event] is the absolute latest you can expect to retain insurance - you're more likely to lose it before then," Storey says.
It's a dilemma for both coastal residents and local authorities, Storey says, because it means insurance retreat is likely to happen quickly in some places, long before people are ready or able to move.
"The people that own the properties don't want to talk about it, councils don't really want to talk about it, insurers certainly aren't sharing their information," she says.
"We know it's happening but it's hard to get a feel for where and how quickly it is happening. It's one of those things no one really wants to talk about."
If a community loses insurance, and is then wiped out by a storm, what comes next? There's no answer to that question.
How this would happen is unclear. In previous disasters, like the Canterbury earthquakes, nearly everyone had insurance - those without it were not compensated by the Government unless they were red-zoned.
When an entire community doesn't have insurance - not because of negligence, but because it's not available, leaving them unable to move elsewhere - it presents a much thornier mess.
Sea-level rise is not just a scientific problem, or even a political one: It's philosophical.
Many of the most at risk communities already rate poorly on the socio-economic index, but researchers say there's a risk the economic gaps could widen in a world with higher seas.
"In order to respond to something that's a predictable, ongoing threat, we need to do something different than we already do with natural hazards," says Dr Lisa Ellis, a political theorist at the University of Otago.
Dr Lisa Ellis
"With natural hazards, New Zealand has a beautiful tradition of solidaristic response to fellow citizens in need. The problem in going forward with that is it puts us all in really dangerous relations with each other."
Ellis has been specifically studying how the risk of sea-level rise is distributed - how best to ensure that everyone shares the burden equally.
The major issue she sees is that communities with less economic power could end up being forced to move, while wealthier communities use their money to fortify the coasts, in some cases with public money.
"The rich get seawalls and the poor get moved"
When Ellis phrased this phenomenon in simple terms when conducting her research - "The rich get seawalls and the poor get moved" - she said no one agreed it was fair. But under the existing framework, it's likely to happen.
"That's a pretty likely policy if no action is taken," she says.
"People are able now under our current unregulated, decentralised framework, to use private power to adjust local policy actions."
Sea-level rise as a whole, she says, poses the threat of a "perfect moral storm of risk transfer" - without a solid, fair legal framework, the disadvantaged would bear the brunt of the problem.
"It's not as if people are saying I want to take advantage of my fellow citizens: The problem is we don't have a predictable, legal framework outlying where those risks should lie," she says.
"The legal framework has to come from central Government."
One of the most obvious ways this risk transfer could happen is through compensation.
Because of insurance retreat, it is likely a community unable to get insurance will, at some point, be damaged by a storm. The question becomes: Who pays for the damage?
A likely answer is the taxpayer. The problem is that signalling this ahead of time could encourage people to not buy insurance, even if it is still available, knowing they'll be bailed out.
Another risk is that it could prompt expensive, extravagant developments on risky parts of the coast. It would allow a developer to keep all the gains, while externalising the risk to the taxpayer, who would have to pay for the damage.
"Generally speaking, people who make risky coastal investments are relatively advantaged, and the average ratepayer or taxpayer is relatively socially disadvantaged compared to those investors," Ellis says.
"You have a transfer of risk from the least to the most vulnerable in that case."
This is a problem with no easy answer. It's also something that's almost certainly happening.
Late last year, much-awaited coastal hazard guidance for local authorities was officially released by the new government.
It was a significant step in planning for sea-level-rise. The guidance is effectively a coastal hazards bible for councils, which had found themselves struggling to keep up with the evolving science around coastal hazards, and increasingly taken to court by residents for their planning decisions.
The 300-page document has no statutory power - it's advice, not a rule book - but it provides thorough, clear information about what councils need to do to manage the impact of coastal hazards, including how best to engage with their communities sensitively.
The document had been set for release in December of 2016. It was so far advanced a press release had been drafted.
The delay was political - the Government knew that stricter planning guidelines would hurt developers, and possibly affect property prices. Although councils were regularly asking for the guidance to be released, it took a change of government for that to happen. It was ultimately unveiled last December by the new coalition government, a year to the day after it was ready to publish.
Its co-lead authors, Judy Lawrence and Rob Bell, have become evangelists for climate change adaptation. They have been travelling the country, speaking to councils about how best to manage the many challenges posed by sea-level rise.
"[The guidance] is very clear in what councils can do and how they can go about doing it," Lawrence says.
"There's a real appetite to use the guidance - they [councils] understand it now. It's setting out a fairly standard approach for councils to take for working with their communities."
The delay in publishing the guidance, however, has created a problem.
The document was an update to similar guidance released in 2008, which was clearly outdated. Its projected sea-level rise scenarios were significantly lower than estimated by the latest science; but it was also, effectively, an engineering manual, which took little account for the social implications of planning decisions.
The new guidance, for example, suggests councils consider an extreme sea-level rise scenario of 1.9m by 2150 for long-lasting developments, like subdivisions. The most extreme scenario under the old guidance was 0.8m by 2100.
The gulf between the documents, and the delay in releasing the updated guide, means councils have likely been rubber-stamping development in risky areas by referring to outdated advice.
That includes not just housing developments, but important infrastructure like sea-walls, stopbanks, and roads. Once a development gets consent, it's very difficult to stop. There are likely many consented developments on the coast that have yet to be built, which were approved under outdated standards.
Just how many of these developments are going ahead is unknown, because no-one had thought to make a stocktake, Lawrence says.
"There is quite a lot of land around New Zealand, and no one seems to know how much yet, which is already consented but not built on in low-lying areas," she says.
"We don't know how much of it there is, and that in itself is a problem."
It's an important question because it may fall on taxpayers to bail out those developers, effectively subsidising their recklessness.
"We don't want to create moral hazard, where the taxpayer is funding people who live in silly places, or to undertake activities that aren't sustainable," Lawrence says.
"We don't want taxpayer and ratepayer money going to people who have made bad decisions because they haven't done their due diligence, in terms of going into an area and building a great big flash house and a swimming pool."
There are a few known examples. In Whitianga, a development featuring man-made canals is expanding, despite being consented for much lower sea-level rise projections in 1998. Already, the community is flooded during king tides.
A new development in nearby Thames flooded during a storm in January, before it had even been completed. In Dunedin last year, a new suburb with partially completed houses flooded.
Even the new guidance hasn't been enough to stop some new coastal development going ahead.
One way this has happened is through Special Housing Areas [SHAs], the result of legislation passed under the last government to accelerate the building process.
The legislation, Lawrence says, did not appear to link back to the section of the Resource Management Act on natural hazards, meaning they could be built in low-lying, flood prone areas.
Several proposed SHAs analysed by Stuff are in areas that will become increasingly flood-prone. One example is a 190-house development planned in Ngongotahā, near Rotorua, which is intended for first home buyers and people with low incomes.
During an unprecedented storm in July, part of the site - which is on a floodplain - was underwater. Locals have long complained of flooding in the area.
Several SHAs in central Nelson are planned for flood-prone areas, according to council maps, and others are planned further along the Tasman coastline, where erosion is whittling away the land.
Others are in Papamoa East, near Tauranga, where large areas of intensification continue on the shore. One new development flooded extensively after rainfall in June.
"[The legislation] is generating a lot of development, some of which is in low-lying coastal areas," Lawrence says. "So we have a legislative mismatch."
"It means we have an ongoing legacy where councils, because of the way their plans are written and these new bits of legislation, they can't stop an ongoing development occurring in low-lying areas."
It's frustrating because the easiest, most cost-effective way to adapt to sea-level rise is to plan well in advance and stop putting assets in places where they're likely to be stranded.
This requires tough decisions, particularly by councils and central Government, which need to ease the transition for their communities.
"You've got to allow communities, I think, to transition," Rob Bell says.
"The thing I think we really need to get through in that transition is to put in planning controls to stop intensification. If we intensify too much, then further down the track it just becomes more intractable to retreat or deal with those abandoned assets."
Bell, who is an engineer, has seen firsthand how the technocratic approach towards coastal hazards has made way for something more empathetic.
In recent years, there have been high profile battles between residents and councils over coastal planning decisions, several of which have ended up in the courts.
Authorities need to be careful in how they approach these issues, Bell says - it requires a human touch.
"It's 90 per cent human dimensions," he says.
"I recognise we've had some failures. In the Ministry of Works days, the old consultation approach - we've worked up options A, B and C, and we think B is the best for you - doesn't cut it any more. And I've seen that evolution."
A glimpse of the future of coastal planning can be seen in Hawke's Bay.
A long part of the coastline south of Napier is eroding, putting hundreds of houses along the coast at risk. Local authorities have combined to create a long-term strategy, which extends out to 2120, for each of the individual communities.
It uses an approach called a Dynamic Adaptive Planning Pathway, which looks, on paper, a bit like a subway map. A panel of people representing each community decides a priority for the short term (within 20 years), medium term (20-50 years) and the long term (50-100 years).
In Haumoana, for example - which is at high risk of erosion - the panel recommended beach renourishment and groynes for the short and medium terms, and managed retreat for the long term.
These can be changed, if the hazards turn out to be more or less severe than expected. None of the Hawke's Bay communities opted for managed retreat in the short term - all chose to defend their current position - which poses some risks. Hard defences cause beaches to disappear, and have flow on effects for ecosystem health.
It has been, however, a less hostile process than seen elsewhere, largely because the community had most of the control.
"In the short term, I think quite a few of our communities … they want to buy some time to sort of transition themselves, and a lot of them are still in that 'hold the line' paradigm," he says.
"While we may not necessarily like to see lots of sea-walls go in, we can see that you have to have an engagement process where people work collaboratively and they can see the longer term picture."