Evidence of glaciation in 'super greenhouse' world

Large ice-sheets existed on Earth about 91 million years ago, during one of the warmest periods since life began, an international team of scientists reports this week.

The findings, published in the journal 'Science', challenges the popular assumption that large glaciers could not have existed in the 'super greenhouse' climate, when tropical surface ocean temperatures reached as high as 35-37C (95-98.6F) and alligators lived in the Arctic.

Scientists from the USA, UK, Germany and Netherlands found evidence of an approximate 200,000 year period of widespread glaciation during the Turonian 'super-greenhouse' period of the Cretaceous, with ice sheets about 60 per cent the size of the modern Antarctic ice cap.

The team obtained their evidence from detailed analyses of sediments that were deposited in the western Equatorial Atlantic Ocean at that time.

The sediments, recovered by drilling into the ocean floor off Suriname in South America, contained fossil shells of tiny sea creatures, foraminifera, that lived in the Cretaceous seas.

These shells 'captured' chemicals that were present at the time, providing the researchers with clues about the temperature, composition and salinity of the seawater. Scientists at the Scripps Institution of Oceanography in the USA were able to use this information to reconstruct sea temperature, both at the surface and at depth.

Meanwhile, a European team at Newcastle University in the UK, the University of Cologne in Germany and the Royal Netherlands Institute for Sea Research (NIOZ), studied the composition of organic molecules in the same sediments, providing separate evidence about the temperature of the surface water during this period of time.

By combining these two lines of data, the team was able to identify temperature and chemical changes in the ocean that are consistent with periods of glacial formation.

Professor Thomas Wagner, of the School of Civil Engineering and Geosciences and the Institute for Research on Environment and Sustainability at Newcastle University, said: 'Speculation about whether large ice caps could have formed during short periods of the Earth's warmest interval has a long history in Geology and climate research, but there has never been final conclusive evidence.

'This uncertainty remained, as there is very little direct evidence from high latitude rocks supporting or disproving the concept; also computer simulations have difficulties to accurately model climate conditions at polar latitudes during past greenhouse conditions'.

'Our research from tropical marine sediments provides strong evidence that large ice sheets indeed did exist for short periods of the Cretaceous, despite the fact that the world was a much hotter place than it is today, or is likely to be in the near future'.

Professor Jaap S. Damste from the Royal NIOZ added: 'The results are consistent with independent evidence from Russia and the USA that sea level fell by about 25-40 metres at this time. Sea level is known to fall as water is removed from the oceans to build continental ice-sheets and to rise as ice melts and returns to the sea. Today, the Antarctic ice cap stores enough water to raise sea level by about 60 metres if the whole mass melted and flowed back into the ocean.'

Dr Andre Bornmann, who led the research at Scripps Institute of Oceanography, University of California, together with Professor Richard Norris, and who has since moved to Leipzig University in Germany, said it was not clear where such a large mass of ice could have existed in the Cretaceous period or how ice growth could have started. 'This study demonstrates that even the super-warm climates of the Cretaceous Thermal Maximum were not warm enough to always prevent ice growth. Certainly, ice sheets were much less common during the Cretaceous Thermal Maximum than they are during more recent 'Icehouse' climates, allowing tropical plants and animals like breadfruit trees and alligators to frequent the high arctic. However, paradoxically past greenhouse climates may actually have aided ice growth by increasing the amount of moisture in the atmosphere and creating more winter snowfall at high elevations and high latitudes,' he said. The findings of this study provide compelling support for another related study published by Fletcher and co-authors from The University of Sheffield and Yale in the January 2008 issue of the journal, Nature Geoscience (first published online, 9 December 2007). In their study, Fletcher and co-authors reconstructed atmospheric carbon dioxide concentrations for the Mesozoic and early Cenozoic below the simulated threshold for the initiation of widespread glaciation on several occasions and speculated on the repetitive occurrence of cold intervals in a general greenhouse world.

Technical notes on methodology:

The research team obtained their evidence from detailed geochemical and isotopic analyses of organic carbon-rich sediments that were deposited in the western Equatorial Atlantic at Demerara Rise off Surinam during the Cretaceous.

The sediments were recovered during the Ocean Drilling Program Leg 207 and contained glassy carbonate shells of tiny sea creatures, foraminifera, that lived in the Cretaceous seas. The fossil shells consist of pristine carbonate, which contain oxygen and other elements. By analysing the different types of oxygen atoms (isotopes) in these shells scientists at the Scripps Institution of Oceanography in the USA were able to reconstruct sea temperature, both at the surface and at depth. Meanwhile, a European team at the Universities of Newcastle and Cologne in the UK and Germany, and the Royal Netherlands Institute for Sea Research (NIOZ) in the Netherlands studied the composition of organic molecules of membrane lipids from archea in exactly the same sediments, providing an independent temperature record of surface waters for the Cretaceous western tropical Atlantic. Because the growth of continental ice enriches seawater in oxygen-18 (the isotope of oyygen with an atomic mass of 18), the ” oxygen-18 chemistry, when constrained by biomarker temperature estimates, was used to estimate the size of continental ice sheets. By combining these two lines of data, the team was able to show that differences in the records of the tropical oceans were consistent with periods of glacial formation.