Amid one of the biggest dry spells in New Zealand history, many people have been asking whether we're seeing climate change in action.
The simple answer is yes - but to an extent.
Drought that's engulfed much of the North Island has been put down to a mix of climate factors - notably a ridge of high pressure that spent weeks parked above the country and blocked rain-making systems delivering much-needed moisture.
But long-term patterns are showing how drought events like this one are becoming increasingly more likely to play out under climate change.
Right now, we know that the global-average temperatures are warming at a rate of about two two tenths of a degree per decade - with all of this being driven by continuing greenhouse gas emissions.
If we consider mean changes to the summer climate over the Southern Hemisphere, this warming encourages an expansion of the dry latitude belt which normally sits north of New Zealand - along with a slight southward shift of the westerly wind belt which normally brings a lot of our rain-bearing weather systems across the country.
And while these average changes in are quite simple to describe, it's those extreme events which matter, Kiwi climate scientist Luke Harrington said.
"The fact is, we as a nation are remarkably vulnerable to dry spells as on the order of only six weeks long – this is really quite short by global standards," said Harrington, a research fellow in climate extremes at Oxford University's Environmental Change Institute.
"The risks are compounded by the fact that we often rely on rainfall from ex-tropical cyclones to help ease dry conditions towards the end of the summer – these are also notoriously unpredictable."
Collectively, scientists already know that the chances of having longer consecutive spells of not much rainfall are increasing.
They also also know that summertime maximum temperatures are rising significantly over the North Island as well.
2013 vs 2020
In a 2016 study, Harrington and Victoria University climate scientist Professor Dave Frame explored the 2013 drought that ultimately cost the economy at least $1.3b.
The research indicated weather patterns such as those seen in that event were much more likely to occur in the present day than would have been the case without climate change.
So what about this drought?
Meteorologists have already drawn some interesting parallels with that event and this one - notably, the influence of blocking systems, a dramatic contrast between a dried-out Northland and a drenched Southland, and the absence of a major climate driver like El Nino or La Nina.
But Harrington added there were some important differences.
"It seems that the high-pressure systems this summer have been centred slightly to the north-west of the North Island, whereas there were many more days in 2013 where the high-pressure system was centred off the South Island's East Coast," he said.
"These differences sound minor, but they can affect which regions are buffered by the high-pressure system, and which are not."
Because the blocking systems were broadly covering the middle of the country more often in 2013, this prevented the usual incoming flow of storm fronts from making landfall over the West Coast.
"Whereas it seems this year, the blocking days have been far enough north that they have been preventing the rain-bearing systems from reaching the North Island," he said.
"But in doing so, these severe atmospheric river events have instead been deflected straight into the southwest of the country, causing major flooding issues."
Running the numbers
To gauge the impact of climate change more generally, Harrington and colleagues run physical models which simulate the climate system as we see it today, using well-known laws of physics.
These use the same underlying principles as the models used to forecast the weather for tomorrow, but, to be able to look over a longer time period than one week, they're of a slightly lower resolution.
"When we want to look at whether and to what extent climate change has made an extreme event more likely, we first run a sequence of tests to confirm whether the climate model is doing a good job of representing the relevant features of New Zealand's observed climate today," Harrington said.
"If so, we then run two types of experiments, many thousands of times over. The first is trying to replicate the evolution of today's climate: including things like present-day greenhouse gas concentrations, aerosol and ozone concentrations, any changes in volcanic activity, solar variations, and sometimes even considering the specific patterns of sea surface temperatures and polar sea ice concentrations.
"We then have an alternative scenario, which has removed the human contribution to each of these features. This is our 'world without climate change'."
Next, they ran the climate model for a full year in each experiment, and repeated it thousands of times over.
"That way we get many possible realisations of, say, what the summer of 2013 could have been, given the background state of today's climate, as well as in a world 'without humans'."
They then looked at how often we experienced summers which could be deemed similar to the observed extreme event.
"In 10,000 simulated years of the present-day climate, we might witness a 2013-like summer 100 times," he said.
"But in 10,000 simulations of New Zealand summers in the climate 'without humans', we might see a drought like 2013 occur only 50 times.
"Therefore, we would say that the 2013 drought was twice as likely to have occurred today because of those human influences that we removed between the two experiments."
The future of drought
How might those figures change in the future?
That depended on how fast temperatures continued to rise in the next several decades - or, of course, how quickly the world decarbonised.
But that wasn't all. One complicating factor for New Zealand, particularly, was the recovery of the ozone hole over the Antarctic.
This could temporarily counter some of the changes which were expected with continuing greenhouse gas-driven warming over the coming years - but the details around this were still unclear.
There also remained many unknowns about the specific physical feedbacks which constrained drought processes here.
"A lot of work has been done on these questions in Europe, the US and Australia, but many remain over New Zealand," Harrington said.
"These questions will only be solved by innovative projects in high-resolution climate modelling, like the Weather@Home project being led by Dr Suzanne Rosier at Niwa."
A further big question was the potential impacts of extreme heatwaves, which often accompanied hot, dry summers.
"There are a myriad of known impacts – particularly related to human health – which are associated with extreme heat, even in relatively cool climates like New Zealand," Harrington said.
"Examples can be found in the UK, Canada and even Scandinavia. Yet, these potential impacts have almost never been explicitly quantified in Aotearoa.
"In fact, when you look around the world for similar examples of misclassifying a compound extreme event - that is, combined extreme heat and aridity - as only a drought, New Zealand actually most closely resembles countries in sub-Saharan Africa."
The bottom line, he said, was that we needed to improve our resilience to extreme weather.
"We should be ready for a summer in the next 10 or 20 years where a decent spell of rain might not turn up for 10 weeks, instead of six," Harrington said.
"This is not a prediction. But the fact is the plausible range of variability in our summer weather is massive, and I'm not convinced we have seen the worst-case scenario just yet."