Climate change has already made summer longer in some parts of the country, where heat extremes have been playing out in ways more typical of the Mediterranean.
That's according to a just published analysis that's used a new measure to document previously unobserved changes - and warns that we need to urgently rethink how we define "extreme heat" in our fast-warming world.
Along with finding that summer seasons have effectively been growing in certain places, Kiwi climate scientist Dr Luke Harrington's new paper has also turned up some heatwaves that earlier went unrecognised.
With its relatively cool, temperate climate, New Zealand has been one of many higher latitude countries thought to be less threatened by extreme heat driven by global warming, which is predicted to lift our temperature by several degrees this century.
Harrington argued this had meant metrics to gauge impacts of extreme heat here haven't kept pace with wider improvements in how heatwaves are defined.
Harrington, a research fellow in climate extremes at Oxford University's Environmental Change Institute, saw a particular flaw in using 25C as a catch-all threshold for extreme heat.
"Yet that is precisely what the vast majority of reports commissioned for the government and local councils in recent years have focused on."
A new measure
In New Zealand, 25C is formally defined as a "hot day" and purportedly stems from the claim that beef and dairy cattle start experiencing heat stress at temperatures above it, rather than any wider impacts on human health.
Part of the reason we haven't looked closer at this metric much, Harrington argued, was that most of us thought we'd never really experience temperatures hot enough to produce negative impacts - particularly compared with Australia next door.
Yet "extreme" was a relative concept - and there was ample evidence from overseas that people were adapted to the temperatures they were familiar with.
"For example, people have been hospitalised or even died from extreme heat in cool places like Finland, Stockholm or even Sheffield," he said.
"This is exposure to temperatures which, while relatively hot, might be considered normal summer weather to someone living in Sydney."
In his paper, Harrington began looking at past hot years using 25C metric and found three years in the last 40 - 1998/99, 2007/08 and 2012/13 - where that threshold was crossed in many regions, over more than three months of the year.
"Of course, when some major farming regions like Waikato experience a full summer of 'extreme heat' three times within 15 years, questions are raised as to how well the threshold even represents heat stress in livestock," he said.
"Also, because of the nature of the threshold, it was hard to figure out which of these three years were more extreme."
But when he applied a different metric, gauging relative heat, he found that 1998/99 was by far the most anomalous year on record, followed by 2012/13 and then 2007/08.
"My new metric works by digging into what temperatures we might expect in a specific part of New Zealand and for a specific time of year," he said.
"But rather than asking a binary yes-no question of whether some chosen threshold is exceeded during a spell of hot weather, this new approach instead quantifies how severe these temperatures were, relative to what we are familiar with in this part of the country at this time of year."
By removing the binary nature of the index, he said, we can better compare the relative differences, such as those between three different heatwaves which might all exceed some absolute temperature threshold.
That was crucial, given our notion of "extreme" heat under climate change remained heavily context-specific.
"If temperatures stayed above 10C all winter in Te Puke, for example, that would fit the bill as an extreme event for a kiwifruit grower there," he said.
"We also shouldn't be surprised, then, if the temperatures needed for a dairy cow to suffer heat stress will differ from that of a human, or if someone living in Hawke's Bay has a higher tolerance for experiencing a week of temperatures above 35C than someone from Westport."
It might also be the case that there are no bad effects for anyone in New Zealand experiencing a few summer days in the mid-30s.
"But the fact is we don't really know, because we haven't properly examined the chain from unusual temperatures to impacts before," he said.
"My concern is, what happens in the summer of 2030, when we have a week-long spell of temperatures approaching 40C in Auckland or Hamilton?
"Sure, this will fit the hot day criteria of 'above 25C'. But that doesn't help us understand what the impacts would be, which is the entire point of these metrics being developed.
"On the basis of overseas work, we could speculate that people will be hospitalised and some will even die because of the heat. But we can't yet provide any more detail than that, which is troubling."
Hidden heatwaves, longer summers
Fascinatingly, Harrington's analysis threw up some heatwaves from the past that had never been defined as such.
"I wanted to find events from our recent past which ticked the box using multiple heatwave criteria, and hence will be the best candidates to try and understand what the impacts of past heatwaves looked like."
The two most promising events were found in March 1999 and January 2018 - a month that went down as New Zealand's hottest on record, amid our warmest ever summer.
Harrington also compared how the number of extreme hot days in each month had changed between the 1980s and 2010s, a period that saw about 0.5C of global warming.
He found that February showed the most striking changes, with a more than five-fold increase in relative measures of extreme heat across the country.
More worryingly, November and March also show pronounced increases over Waikato and Otago, which meant an effective lengthening of the summer.
"Heat extremes are worsening faster than mean temperatures in these regions - this is the first time such changes have been seen in New Zealand observations," he said.
"This behaviour is actually more typical of the changes being seen in the Mediterranean, and suggests drying soils might be making hot days even hotter.
"But I stress though that these results are preliminary, and a lot more research is needed to get understand the details of what is going on."
His paper ultimately put forward several new ways to analyse extreme heat in cool climates like New Zealand, in addition to the one approach used already.
"But the key point is that no single metric will do a good job of capturing the full range of temperature-related impacts as our climate becomes increasingly unfamiliar over the next few decades."
Depending on the scenario, the number of Kiwis over the age of 75 would quadruple by the end of the century, while global temperatures will also continue to rise.
Both these trends increased the risks of extreme heat facing Kiwis.
"If we walk into this future without having a range of approaches - each of them tailored to understand how extreme heat will impact our people, our livestock and our horticultural industries separately - then we are walking down the road blindfolded."