It's an intriguing puzzle that has so far eluded climate scientists -- why is the extent of sea ice surrounding Antarctica increasing when the world is warming?

A New Zealand study will now unravel the mystery, and how it fits into the wider picture of the earth's climate.

Overall, the total sea ice extent around the Antarctic coast is increasing by around one per cent each year and in recent times the rate has been even above the trend.

As at last September, there was around 20 million square kilometres of sea ice -- the most ever observed in a record stretching back 36 years.


The changes, however, were far from uniform -- near the Antarctic Peninsula, and especially to the west of it, sea ice was disappearing as fast as it is in the Arctic -- a rate of around 11 per cent per decade.

At the opposite extreme, sea ice in the eastern Weddell Sea and over to the Ross Sea to the south of New Zealand, the extent was increasing rapidly -- up to about 8 per cent per decade.

Most of the ice was relatively thin, at between one and two metres thick.

Rates of change also varied strongly with each season -- the largest increases happened in spring and autumn, and the smallest in summer and winter.

"It is clear that the changes line up well with trends in winds around Antarctica," said Victoria University climate scientist Dr James Renwick, who will be leading the Marsden Fund-supported study.

"Where there are more cold southerlies there is more ice, and vice versa -- this is a pretty simplistic picture that doesn't hold everywhere all the time, but it's a fair rule of thumb."

Dr Renwick suspected the winds had changed because of natural influences such as El Nino and the Pacific Decadal Oscillation.

The implication was that these natural influences could go away or change their form any time -- so the increase could become a decrease any time. "Ultimately, with continued warming of the atmosphere and oceans, the sea ice must start to decrease," he said.


"It certainly can't keep increasing for much longer, as there is no way we'll have sea ice around New Zealand any time soon."

Beyond how the sea ice increase stood against the background of climate change, Dr Renwick felt it was important to understand what it would mean for the future.

"Sea ice is important for marine life, for the ocean circulation and for the overall circulation of the atmosphere, as it is one of the things that controls the temperature difference between the tropics and the pole -- which itself controls the westerly wind circulation and the distribution of storms and rainfall in the middle latitudes of the Southern Hemisphere, where New Zealand sits," he said.

"Knowing the future evolution of Antarctic sea ice extent will help us better understand all these things."

To crack the puzzle, Dr Renwick would draw on computer-based analysis and modelling.

The work was indirectly linked to recent work by fellow Victoria University scientists which showed that changes to the Southern Ocean's sea ice belt could mean future ice sheet melt and global sea level rising several metres in coming centuries.


"There is now recognition that the sea ice supports the fast ice -- the "permanent" sea ice -- which supports the ice shelves which support the ice sheets," Dr Renwick said.

"So, losses at the end of the chain have implications all round."

As for any comparisons with the Arctic and dramatic processes being observed there, Dr Renwick pointed out the two hemispheres couldn't be more different in terms of geography.

"The Arctic ocean is almost completely enclosed, so much of the ice grows over the polar ocean and sits there from one year to the next," he said.

Antarctic sea ice, meanwhile, grew around the edge of a continent, over open and turbulent oceans -- and at lower latitudes than in the Arctic -- and mostly fresh every year.

"The Arctic losses are mostly about thermal changes, changes in energy balance -- more heating from greenhouse gases means ice melt, which increases heat absorption, which melts more ice.


"In the Antarctic, it's dynamical factors that seem to be more important -- what the winds and currents are doing, rather than the straight radiative heating. "So, while both hemisphere have the same kind of ice over their oceans, most of the details of how the ice forms and melts, and how long it lasts, is quite different between the hemispheres."

Dr Renwick said it was false to use Antarctica's sea ice increases to argue against global warming, which -- at a chance of at least 50 per cent in the highest emission scenario -- could push global surface temperatures 4C above pre-industrial times by the close of this century.

The straight thermal signal from greenhouse gas warming was over-ridden by dynamical effects, among them how the winds were changing, the "massive thermal inertia" of the Antarctic ice sheets, and the fact the southern oceans were very turbulent and good at sucking heat down to depth, he said.

This was why the surface ocean around Antarctica hadn't warmed at all in the past five decades, yet, at depths below 1000m, the southern oceans were one of the regions of the planet that was warming fastest.

"Basically, singling out Antarctic sea ice to 'prove' there's no warming is a very blinkered approach," he said.

"Taken all together, the ocean warming, atmosphere warming, land ice melting ... we end up overwhelmingly with a picture of warming."


Robotic floats confirm ocean warming

Thousands of instruments deployed across the world's oceans have shown the temperature of the world's oceans has risen steadily between 2006 and 2013.

One of the world's leading oceanographers, Professor Dean Roemmich of Scripps Institution of Oceanography in the US, has been basing himself at Wellington over the summer to analyse data being gathered by the global network of 3750 Argo floats.

The Argo floats are automated devices which periodically descend and ascend through the top 2000m of water, taking measurements of temperature, salinity and velocity as they go.

They transmit their data via satellite when they reach the surface, before commencing a new cycle of measurement.

A study led by Dr Roemmich and just published in the journal Nature Climate Change found the top 500m of the world's oceans warmed at a rate of approximately 0.005 C per year, while at depths of between 500m and 2000m the warming was at approximately 0.002 C per year.


NIWA oceanographer Dr Philip Sutton, who has been working with Dr Roemmich at the institute's Wellington offices, said this was equivalent to adding the heat of two trillion continuously burning 100-Watt light bulbs to the world's oceans.

Yet the rate of ocean heat gain during the eight-year period was not unusual, Dr Sutton said.

"What is new is that the rate and patterns of heat gain are revealed over a period as short as eight years thanks to the Argo float network, the warming extends to 2000m and deeper, and is occurring predominantly in the Southern Hemisphere south of 20?S."

It added to a picture which had left Dr Roemmich to conclude that changes so far seen in ocean circulation in the southern Pacific were linked to climate change, which had contributed to a heavily evidenced warming of oceans over the past 50 years.

NIWA has played an important part in the Argo programme, deploying more than 1200 floats from its Kaharoa research vessel, and last month Dr Roemmich said New Zealand had been "absolutely critical for achieving a global array".

The Argo programme would soon be expanded with the deployment of "Deep Argo" floats, which could sink much deeper and give insights into how warming was affecting oceans at such depths where changes were less understood.