The Two-Mile Time Machine

Ice Cores, Abrupt Climate Change, and Our Future

Richard B. Alley
Published 2000 by Princeton University Press

2004 Aug

This book documents the findings of ice core projects in Greenland:

Greenland Ice Sheet Project (GISP) Danish-Swiss-US, 1981
Greenland Ice Core Project (GRIP) European, 1989-1992
Greenland Ice Sheet Project 2 (GISP2) US, 1989-1993

Central Greenland is chosen because the ice there tends to accumulate (since it is the middle of the island) rather than flow.

Climate changes are deduced from:

• annual layers in the ice core created by summer-time melt (the melted snow is less dense and lighter colour)
• thickness of annual layers (a measure of snowfall that year)
• wind-blown dust and sea salt accumulates more in the winter, especially during ice ages when the winds are stronger
• electrical conductivity of the ice (atmospheric chemistry varies between summer and winter)
• major volcanic eruptions leave traces that are used to date a layer absolutely
• isotopic composition of precipitation varies with temperature, so by measuring hydrogen isotopes, the temperature at time of precipitation can be deduced
• oxygen isotopes taken up into carbonates (eg., shells) leave a layered record in ocean sediments (this gives a long-term history of earth's ice coverage, see figre 10.1)
• current temperature profile (heat flux from past changes is still travelling through the ice)
• hydrogen peroxide is produced in the atmosphere by sunlight; it is deposited in ice only in the summer (no sun in the winter)
• beryllium-10 isotopes are created by cosmic rays; their concentration records changes in the earth's magnetic field and solar output
• ice records dust blown (from Asia, it turns out), volcanoes, fires, human lead production
• gases trapped in air bubbles, eg., CO2, methane
• sunlight variations are predicted from orbital parameters

The book tells the story of how many of these indicators support each other. Researchers from around the planet have been turning up data from their specialties to provide further detail and certainty about the climate record.

Major conclusions (quoted):

• Climate in the past [>10K yrs ago] has been wildly variable, with larger, faster changes than anything argicultural or industrial humans have ever faced [the recent 10K yrs have been unusually stable and warm, compared with recent 1M yrs];
• Climate can be rather stable if nothing is causing it to change, but when the climate is 'pushed' or forced to change, if often jumps suddenly to very different conditions, rather than changing gradually. You might think of climate as a drunk: when left alone, it sits; when forced to move, it staggers.
• The 'pushes' that have caused climate changes in the past probably have included drifting continents, wiggles in the earth's orbit, surges of great ice sheets, sudden reversals in ocean circulation, and others.
• Small 'pushes' have cause large changes because many processes in the earth system amplify the pushes. Greenhouse gases have probably been the most important amplifiers.

Major results:

Earth's orbit has an eccentricity with a 100,000-year cycle, causing it to go from elliptical to circular. In addition, the axis of rotation tilts on a 41,000-year cycle, and wobbles on a 19,000- to 23,000-year cycle. These orbital changes affect the amount of sunlight received by the earth, and more importantly for climate, where the sunlight hits (north or south hemisphere). Sunlight variations don't translate directly to ice because of non-linearities -- for example, it takes longer to form an ice sheet than to melt one, because a larger ice sheet has a warmer bed, has a warmer surface, and can flow faster (into warm water around it). Here's the history of ice in the last 800,000 years, based on oxygen isotope variations in shells in ocean sediments:

Ice ages are associated with reduced carbon dioxide. The best guess is that stronger winds during ice ages (see next chart) cause more dust to fertilize algae in the ocean, absorbing CO2, reducing greenhouse effect, causing more ice.

Wind increased in Venezuela during colder periods -- winds carry dust that fertilize ocean algae, absorbing CO2:

(Winds there mix water, bringing nutrients to the surface to fertilize plankton, which make shells that color the sediment white -- the graph is of sediment color.)

Rocks are dropped in the north Altantic by melting icebergs. There are patterns over time of the droppings, finally explained as 'Heinrich events'. A Heinrich event occurs when an ice sheet covering Hudson's Bay gets so thick that the earth's heat starts melting its base, cause the Hudson Bay ice sheet to suddenly skate out into the Arctic and north Altantic oceans -- causing a big influx of fresh meltwater.

[It's interesting how the arrangement of the continents has such an impact. The role of Hudson Bay is astonishing! It reminds me of tidal action in the Bay of Fundy (an oscillation cavity).]

'Bond cycles' (related to changes in Gulf Stream circulation) lead up to a Heinrich event:

Heat is carried by the Gulf Stream into the far north. However, if there is enough fresh water in the north Atlantic, the salty Gulf Stream can sink prior to reaching the north. This has a major cooling effect that show up as Bond cycles. The ice core record shows that the Gulf Stream can switch modes abruptly. These switches are thought to be responsible for the 'Dansgaard-Oeschger cycles' (abrupt switching between warm and cold) observed in the ice cores.

Recent history:

The 'dryas' is a rose-like flower that grows today only in alpine regions. However, its pollen is found in layers of low-lying European forest bogs, indicating that at the time it was much colder. Three periods of dryas pollen have been noted in the bog records, and are called the Younger, Older, and Oldest Dryas periods. Radiocarbon dating of the layers sets their dates, which match other yardsticks, eg., ice cores, wind in Venezuela. "During the Younger Dryas, the north Pacific cooled, glaciers advanced from high peaks in NZ and the Andes, and lakes in Africa shrank as the Sahara Desert spread into former fertile regions".

[The end of the Younger Dryas seems to correspond to the end of the large mammals in North America. Climate change may have had a lot to do with that extinction, not just humans (who were expanding into N.America during that time).]

Recent history in a larger context:

Alley's conclusions:

• "The climate will change."
• "Climate change will give winners and losers."
• "In the short term, losers from climage change will outnumber winners" (because human settlement patterns are already optimised to match current conditions, and thus disrupting conditions will lead to more losers than winners).
• "In the long term, losers from climate change probably will outnumber winners" (because change is stressful to societies and systems).
• "Slowing down change may help us a lot" (because abrupt changes are harder to deal with and cause more disruption).
• "Saving some excess capacity may make our lives much easier in the future" (because having excess carrying-capacity gives us a safety margin).
• Having "too many people will use up our excess capacity" (ie,. the more people there are on the planet, the greater their effect on the planet, and the less our safety margin).

Of life decisions a person faces, the decision with probably the most consequence to the environment is whether or not to produce a child.

Subsequent developments (2004):

Nature article on NGRIP, a more recent ice core project