Geoscientist 17,1 January 2007
In the first of a series of historical second features to appear in 2007, Sue Bowler examines the role of the Society in the first glreat climate debate – over Ice Ages
Climatology was not a primary object of the eighteenth century gentlemen who founded the Geological Society to investigate “the mineral structure of the Earth”. Yet is it now a discipline that engages many Earth scientists, working in business, society and industry as well as research. Among the disciplines needed in climatology, geology stands out as a source of solid data from Earth’s past climates. That the Earth sciences can provide such valuable information is a tribute to the enormous advances made over the past 40 years in understanding both climate processes and the ways in which they are recorded in the geological record.
The clues that rocks and fossils hold about climate have long been recognised. Wegener used the different climates indicated by rocks such as red beds as evidence for continental drift. A century before, James Hutton understood the cyclic nature of erosion and rock formation, as processes he could observe on the Earth’s surface formed ancient lithologies. Lyell’s formalisation of these ideas as uniformitarianism led to the realisation, in the late nineteenth century, that the Earth had to be immensely old if there were to have been enough time to form the rocks. The development of radiometric dating put numbers onto this immense age and set off a cycle of technological development, scientific advance and further technical refinement that characterised geochemical research and radiometric dating through the 20th Century.
In the 1950s and 1960s geologists realised that variations in the proportions of stable isotopes could be used as a type of stratigraphy. Isotopes of oxygen, carbon and hydrogen are present in ice, sea water, rocks and air; chemical reactions involving these species alter the proportions of the different isotopes, so that they act as markers for a range of earth surface and biological processes. In 1951 Harold Urey had picked up palaeoclimate data based on carbon isotopes in geological samples, after he had realised that the fractionation he was observing depended on temperature. Changes in the fractionation of isotopes in calcite shells formed from seawater could be used to calculate the temperature of the sea. But this method did not prove immediately useful because temperature was not the only variable affecting fractionation: sea water isotope ratios varied, notably with the volume of water locked in ice during glaciations, so the measurement could not give a reliable temperature estimate. The developing field of isotope stratigraphy needed a means of identifying a temperature signal.
Nicholas Shackleton (Box) provided a solution – and developed techniques precise enough not only to get a temperature scale, but also to document the comings and goings of ice ages. Shackleton was a research student at Cambridge when he started to tackle this problem. He never really stopped, in a career that established new and better timescales for the Earth’s recent past, showed the major driving influences on our climate and demonstrated the value of rigorous quantitative analysis of high quality data. Shackleton’s solution to the temperature problem was to treat the temperature changes of sea water through ice ages as minor; most of the changes in the isotope ratios, he reasoned, came about from the huge volumes of water frozen into ice sheets as ice ages started. So species of shelly plankton that lived in the deep ocean, at fairly constant temperature, would have different oxygen isotope ratios from those living in shallow ocean waters, where the waters cooled; comparing deep water and shallow water species separates the temperature signal from that of ice volume.
Shackleton’s work contributed to the first global compilation of climate data, CLIMAP1, in 1976 and added to the growing research interest in high resolution isotope stratigraphy. He, John Imbrie of Brown University and Jim Hays of the Lamont Doherty Geophysics Observatory, showed that there was a strong signal in the ocean oxygen isotope record from the volume changes associated with ice ages2. There were cyclic changes in the signals that took place over familiar periods of 23000 and 41000 years, timescales familiar from theoretical work in the 1930s by the Serbian climatologist Milutin Milankovitch. These and the 100,000 year cycle were identified with ice ages by Milankovitch, on the basis of on solar insolation theory, changes in the radiation reaching Earth from the Sun as a result of regular changes in the precession, obliquity and eccentricity of the Earth’s orbit.
Shackleton, Imbrie and Hays’s confirmation of the Milankovitch cycles in the ice age record was a key finding: changes in our climate are induced by processes outside the Earth itself. It implied that these orbital cycles could be found in a range of palaeoclimate data. How and why the Earth’s orbital changes affect the climate remained – and remain – difficult questions, but the very existence of such cycles in the recent rock record sparked new interest. When researchers looked for these periodicities, in tree ring records or annual sediment layers in glacial lakes, they found them. Palaeoclimatology was born.
The discipline grew fast, with abundant sequences showing the orbital signals, albeit sometimes with a bit of a time lag, as for ice ages and the 100000 year cycle. The orbital signals formed the basis of a ocean timescale based on oxygen isotopes, the Spectral Analysis and Mapping Project (SPECMAP) of the 1980s3. Shackleton went on to work on deep ocean cores from the Ocean Drilling Project, in which he found evidence for many glacial and interglacial periods in the past 2.5 million years, as a time when land-based evidence pointed towards just a handful of glaciations.
Some of the most interesting developments in this period came when the overarching orbital cycles allowed different datasets to be calibrated and compared. Shackleton used the astronomical cycles to ‘tune’ timescales that showed orbital cycles, sometimes in conflict with established dating methods. For example, the sequence of magnetic reversals had been successfully extended and dated using potassium argon dating, reaching back around 2 million years. But Shackleton with Andre Berger and Richard Peltier4 analysed this timescale and suggested corrections of a few percent, based on the orbital cycles. The suggestion was not well-received at first, but later accepted, when analysis showed systematic argon loss had produced incorrect radiometric dates. Later work on ODP cores produced a high resolution timescale reaching back 15 million years5. Comparison of such tuned datasets also brought rewards. Climate proxies from ice cores drilled in Greenland and Antarctica were especially useful. The ice can contain bubbles of air sealed when the snow fell; these inclusions allow direct measurement of the levels of carbon dioxide in the atmosphere, and atmospheric isotopes to give air temperature, at the same time as the isotope record in the ice.
Shackleton used this data not only to revise and recalibrate the ice volume chronology for the past 400000 years, but also to shed light on the key role of carbon dioxide6. The combined dataset showed that air temperature, deep water temperature and carbon dioxide levels follow the 100000 year orbital eccentricity cycle closely, but that ice volume measures lag behind. He interpreted this time lag as evidence that the orbital eccentricity changes did not directly drive ice sheet changes, but that they did so indirectly, through changes in carbon dioxide levels. In other words, it is not the dynamics of ice sheets themselves that trigger glaciations, but more probably the response of the global carbon cycle to the insolation changes, altering the concentration of carbon dioxide in the atmosphere.
It is the combination of stratigraphic principles, accurate measurements and systematic thinking that brought such speedy progress to the young field of palaeoclimatology – with Nick Shackleton in the vanguard. It is a field where UK geosciences are now strongly placed, with world-class quaternary research and a growing pool of talented and capable researchers. Less tangible, but arguably more beneficial, is the approach that this field has championed, where quantitative rigour and high precision data, demanding an understanding of the whole geological process of formation, deposition and preservation, are used to the full in understanding and modelling the climate of the past. It is a striking example of the value of earth systems science – in this case including the Sun – and a driver for the growth of integrated earth and environment institutions, such as the Tyndall Centre for Climate Change.
And this new research discipline remains in the tradition of the geologists of 200 years ago, with a slight change of emphasis. Archibald Geikie summed up the uniformitarianism expounded by Lyell as “the present is the key to the past”. As far as the climate goes, the past is the key to the future and geologists are cutting the keys.