New ice core records rewrite volcanic history

Active volcano, Mount Erebus, on Ross Island, Antarctica.
Mt Erebus on Ross Island, is the most active volcano in Antarctica. (Photo: Noel Tennant)

The Earth’s volcanic history has been rewritten thanks to a new and more extensive array of Antarctic ice cores containing detailed eruption records.

The findings will permit more stringent tests of models against past climate variations. This will help reduce uncertainties in model projections for the future, which are used to inform climate policy decisions.

The research was conducted by an international team of scientists led by Dr Michael Sigl from the Desert Research Institute (USA) and including Dr Mark Curran from the Australian Antarctic Division. The team measured sulphate aerosols – a by-product of volcanic eruptions – in 26 ice cores taken from 19 different sites across Antarctica.

From these aerosol measurements the team identified 116 eruptions over the past 2000 years and obtained the most accurate estimate yet of the eruptions’ strength and timing.

‘We’ve developed a more accurate picture of volcanic activity by looking at the amount of sulphate aerosols deposited in a much larger number of ice core samples than were previously available,’ Dr Curran said.

‘These cores included some with highly detailed and well dated records, such as the core Australia collected from Law Dome in 1993, and this allowed us to better synchronise the cores and ensure that common events were correctly dated.’

Volcanic eruptions spew large amounts of sulphur dioxide into the atmosphere, where it converts to sulphate aerosols. These aerosols deflect sunlight away from the Earth’s surface leading to short-term global-scale cooling. Bigger eruptions result in more sulphate aerosols and therefore more ‘forcing’ of the climate towards cooling.

Leader of the Australian Antarctic Division’s climate program, Dr Tas van Ommen, said this forcing effect is used in climate models to simulate the climate of the past, and provides a test for determining the sensitivity of the climate system to natural and man-made climate forcing.

‘Part of the uncertainty in future climate projections comes from model-to-model variability,’ Dr van Ommen said.

‘The ability to test models against past changes allows us to evaluate which ones do a better job, and provide the best basis for policy decisions.’

Prior to this study, estimates of volcanic forcing were measured in a smaller number of ice cores taken from low snowfall sites. Low snowfall causes irregular snow accumulation patterns and, as a result, uncertainty in sulphate aerosol deposition estimates. Also, as aerosol deposition in ice is not uniform across Antarctica, a wide array of ice cores is needed to get a representative and more accurate measure of volcanic fallout.

‘We had very few good ice core records extending prior to 1500,’ Dr van Ommen said.

‘Tree ring records pointed to global-scale volcanic events that would have had a pronounced impact on climate, but we had difficulty identifying these in ice cores, particularly during the first 1000 years AD.’

After adding many new records and correctly dating older records, and then comparing them to the best annually dated 2000 year ice cores, the team found that the existing volcanic reconstruction after 1500 was excellent, but substantially different prior to this time. Global aerosol forcing from some of the largest eruptions prior to 1500 were overestimated by 20-30% and others underestimated by 20-50%.

Aerosol loads in the ice cores showed that the largest eruptions occurred in Indonesia in 1257 (Samalas) and 1815 (Tambora) and in Vanuatu in 1458 (Kuwae). In the first millennium three smaller but significant eruptions of unknown origin occurred in 531, 566 and 674.

‘Our research allowed us to re-date the Kuwae eruption from 1452 to 1458, and we also found that the Samalas and Kuwae eruptions were less intense than previously estimated, translating to 15% and 25% less global aerosol loading respectively,’ Dr Curran said.

‘This explains in part a mismatch of temperature reconstructions and climate simulations for these events.’

While further work is needed to identify climatically important eruptions in the northern hemisphere that affected the southern hemisphere, the research team said the current study provides a ‘step-change’ improvement in existing reconstructions of volcanic aerosol loading for the southern hemisphere.

Wendy Pyper
Corporate Communications, Australian Antarctic Division