THE controversial idea that global warming could trigger a sudden drop in temperatures 鈥� maybe not in a matter of days as portrayed in the recent disaster movie The Day After Tomorrow, but possibly within a century 鈥� has finally been put to rest.
The latest ice core drilled from northern Greenland is showing that the last interglacial period, despite being warmer than today, did not end in a sudden freeze. Rather, it took thousands of years for the warm temperatures to give way to the next ice age.
The Greenland ice sheet is made from layers of snow that have compacted into ice over millennia. By drilling a core of ice, researchers can look back in time and determine the temperature when the snow fell by analysing the ratio of oxygen isotopes in the ice. Two previous Greenland ice cores, one known as GRIP extracted by European scientists in 1992, and another called GISP2 retrieved by Americans a year later, gave climatologists their best ever records of temperatures going far back in time.
Advertisement
The two cores agreed almost perfectly all the way back to 113,000 years ago, but then diverged dramatically. GRIP showed that temperatures in Greenland, and presumably worldwide, underwent many sudden fluctuations between 113,000 to 125,000 years ago. In one instance, temperatures appeared to plummet by up to 14 掳C within 70 years.
This sparked alarm because the last interglacial period, known as the Eemian, lasted from about 130,000 to 115,000 years ago, and conditions then are thought to closely parallel today鈥檚 climate. Scientists worried that warm temperatures during the Eemian could have shut down the Gulf Stream, which keeps the north-eastern US and northern Europe relatively warm for their latitudes.
But controversy erupted when GISP2 found no record of such fluctuations. It soon became clear that at least one team, and possibly both, had drilled in a region where the underlying rock is very hilly, potentially jumbling the bottom 10 per cent of the ice. To resolve the debate, European researchers went back to northern Greenland in 1996 and started drilling in a region with flat bedrock, which they reached in July 2003.
The new core, known as NGRIP, goes back 123,000 years, and at 3085 metres it is the longest ice core recovered from Greenland. Besides analysing the oxygen isotopes in the ice, the Europeans also looked at levels of methane trapped in air bubbles. Methane levels rise during warm periods and fall when it gets cold, and the variations back up the oxygen-isotope data.
鈥淭his time we are 100 per cent certain that the ice core is reliable,鈥� says team member J酶rgen Peder Steffensen of the University of Copenhagen in Denmark. 鈥淭he new analysis also shows that the two older ice cores are only reliable to 105,000 years.鈥�
The NGRIP core reaches into the final 8000 years of the Eemian. The team found that Greenland was then about five degrees warmer on average than today, and that the climate was stable. The warm period ended with a slow cooling over 5000 years (Nature, vol 431, p 147). 鈥淭his is important information,鈥� says Eric Wolff of the British Antarctic Survey in Cambridge. 鈥淭he Eemian may not be a perfect analogue of a future warmer world, but it is the best we鈥檝e got. A crucial point is that this part of the Greenland ice sheet apparently did not melt substantially in spite of the high temperatures.鈥�
NGRIP contains nearly a centimetre of ice for each year towards the end of the Eemian and the inception of the ice age, enough to reveal air temperatures and atmospheric chemistry for each year during that period. Such data will be invaluable for understanding how an ice age starts. 鈥淭he NGRIP ice core gives us a climate record of unsurpassed detail from high latitudes where the ice sheets start to grow,鈥� says Wolff.
Kurt Cuffey of the University of California, Berkeley, agrees. 鈥淭he interest in this ice-core is tremendous,鈥� he says. 鈥淜nowledge of the onset of ice ages will tell us a lot about the climate system in general and make it possible to test climate models very rigorously.鈥�
Testing and verifying climate models is essential for understanding how long our interglacial period will last. The main trigger for a new ice age is a shift in the Earth鈥檚 orbit from being more elliptical to being more circular, which happens in cycles lasting roughly 100,000 years. Ignoring the effect of human activity, the next ice age is at least 10,000 to 20,000 years away. However, 鈥渢he anthropogenic greenhouse effect is a joker in the pack鈥�, Steffensen says.
The core has turned up other oddities. When the drill reached bedrock, liquid water flooded the bottom 45 metres of the borehole. This summer the team recovered a core of the refrozen water, and the muddy ice contains what appears to be organic material that looks a lot like pine needles. 鈥淭his material may be several million years old, from a time when trees covered Greenland,鈥� says NGRIP team leader Dorthe Dahl-Jensen, also of the University of Copenhagen. 鈥淚t is even possible that unknown microbes adapted to extreme conditions are alive in the bottom water today.鈥�
