Lecture 32: Ice Sheets and ice-core records (31 was on Coral Reef bleaching)

Ice sheets: ca. 18,000 years ago the ice sheets over the Northern hemisphere were significantly large. In the southern hemisphere not as big a difference.

Ice flows and ice core depth-age relations
Ice flows very slowly in an ice sheet. Similar to drainage divides, ice sheets have boundaries from which the ice flows away. As the ice flows horizontally, the sheet thins vertically and so the age –depth profile of the ice is exponential, rather than linear.
depth – age curve


Gas bubbles trapped in ice further complicates the ice age because of the transition of snow to firn to ice.
A complication in interpreting ice core records, and in defining the depth-age relations, is how snow transforms to ice 50 to 100 m below the surface of the ice sheets. Gas trapped in ice is actually younger than the solid ice at the same depth, and a variety of processes can transport and redistribute gases in this snowy upper layer, FIRN. At the ice-sheet surface, motion of water vapor rapidly converts faceted and angular snow to solid ice spherules. The weight of the overlying accumulation forces these into a more compact configuration by slip at their boundaries and creep deformation. The density of a given layer steadily increases through time.

Ice cores have been collected from glacial ice globally: including from Greenland ice sheet (several cores collected from Greenland); Antarctica and from low latitude glaciers as well, Quelcaya (tropical Andes) and Dunde (Tibetan Plateau in China) ice caps.
Accumulation arates on different ice caps varies depending on the climate of the region. E.g., On Quelcaya ice cap (Peru) annual accumulation (precipitation minus evaporation and sublimation) averages 1.15 m of water equivalent (snow is largely made up of air, often containing less than 10% water; think of water equivalent as the amount of liquid water produced when snow is melted). When scientists drill into the ice cap, they recover what are essentially fossil pieces of ice containing crucial information about past climate.


Ice Core records:
Can often see annual visual layers in the cores
GISP2 ice core project – 3053 meter long core, covering the last 200,000 years of climate history
Data:
ECM – Electrical Conductivity Measurements are done on the entire length of the core. ECM is a fast and high-resolution way of measuring ice acidity. Two electrodes are drawn along the surface of the ice core while the electrical resistance (which varies as a function of acidity) between them is measured.
ECM is important for three reasons. First, it is well suited to detecting volcanic events in the core record. Volcanic eruptions emit large amounts of sulfur gases that react in the atmosphere with water to produce sulfuric acid (H2SO4). Given the right atmospheric conditions, some of this acid-enriched moisture may be transported to Greenland and precipitated onto the Greenland Ice Sheet, where it is incorporated into the glacier. These volcanic events often appear in the ECM record as large values reflecting high acid content because of the SO42-.
Second, ECM is relatively easy to measure and is performed on the entire length of the core at a resolution of a few millimeters. Scientists look at ECM data to determine the location of interesting or significant climatic events that warrant further analyses, which ensures that researchers spend their time and money as effectively as possible.
Third, ECM varies seasonally, providing a high-resolution dating mechanism that, in conjunction with measurements of dust concentration and visual counting of annual dust bands (visual stratigraphy), allows the GISP2 core to be dated with unprecedented accuracy.
Dust from Arctic Canada and Greenland is alkaline, so precipitation deposited on the ice sheet during dusty periods is less acidic than precipitation deposited at other times. Dusty Arctic summers show up as lower readings in the ECM record. Scientists correlate large ECM peaks with historical records of volcanic eruptions. The 1660 eruption of Katla in Iceland and the 1667 eruption of Japan's Tarumani appear in this core segment. The volcanic signal in the ECM record greatly improves dating accuracy by providing absolute dates, unambiguous benchmarks upon which a reliable core chronology depends. ECM and other parameters that vary seasonally such as dust concentration, d18O, and visual stratigraphy have been used to date the core to 40,000 yr B.P. with an estimated age error of ±10%; currently, age models are being used to provisionally date the deepest core, and work is continuing to extend this dating to even greater depths.

ECM varies seasonally because dusty summers result in lower ECM readings, windy and dusty periods in the Earth's climatic history appear in the ECM record as periods of near-zero readings.
Close correlation between cold climatic events (the Younger Dryas and Wisconsin Glacial [the Wisconsin is the term used in North America to refer to the last full ice age], and low ECM readings. Paleoclimatologists postulate that dust fluxes increase during colder periods because the glacial atmosphere is drier. Since dust stays in the air longer when the climate is dry, it is transported greater distances in the atmosphere, resulting in increased dust fluxes to sites like the Greenland Ice Sheet that are without local dust sources.
Volcanic eruptions and ECM - Signal of a 1479 A. D. eruption of Mt. St. Helens using data from. Researchers found a clear signal of a 1479 A. D. eruption of Mt. St. Helens (first identified in ash deposits in the Pacific Northwest and dated using tree rings) in both the particulate (dust) and sulfate (SO42-) records.

Calcium Concentrations
Calcium concentrations, like ECM, serve as a proxy measurement of dust flux onto the ice sheet. Glacial periods are marked by high calcium concentrations, while warmer periods are notable for very low calcium concentrations. As you can see from this slide and the previous one, the climatic regime underwent extremely rapid transitions during the period from 18,000 to 10,000 yr B. P.

Stable isotopes of the water molecules composing the ice, del 18O and del D.

The ratio between the heavier H218O and lighter H216O water molecules in the ice is expressed as the departure from a standard --d 18O=1000x(18O/16Oice -18O/16O std). Snow falling on the Greenland Ice Sheet under colder temperatures is more negative, that is, it contains more of the light H216Omolecules. Thus, a plot of the d18O of snow versus temperature shows an excellent correlation. Since d18O is primarily temperature-dependent, it serves as a paleo-thermometer providing important information about past climate.

Greenhouse Gas in bubbles
Measurements of greenhouse gas CO2 have been taken from several ice cores. Data from Antarctica cores show that CO2 concentrations are at their highest levels in 160,000 years.
measurements of methane concentration from the GRIP (the European counterpart to GISP2) and Vostok cores agree very well in broad terms, illustrating large-scale climatic events such as the Younger Dryas cold event at ~11,000 yr B. P.
The Vostok core illustrated for the first time the strong correlation between paleo-temperature and the concentration of greenhouse gases in the atmosphere. Concentrations of carbon dioxide (CO2) and methane (CH4) have moved in tandem with paleotemperatures derived from the stable isotope record. The mechanisms of these relationships are poorly understood, and it is not known with any certainty whether increased temperatures are, to use an old paradox, the chicken or the egg. To what extent did increased temperatures bring about higher greenhouse gas concentrations? On the other hand, to what extent did higher greenhouse gases cause greater radiative warming of the Earth's atmosphere? To paleoclimatologists hoping to provide answers about global climate change (global warming), this has sparked intriguing scientific debate.

Lecture 33 - Ice -core records (cont’d)

Ice core records about climate: the ECM and dust measurements describe changes in precipitation amount.

Ice core records have been used to develop history of temperature changes.
Stable isotopes of ice water – already discussed.

Reasons to suspect that stable isotopes may not be a faithful quantitative thermometer:
- isotope value of the initial source water(s) can change through source changes (ocean composition changes due to global ice volume, or the actual source location can change due to changes in atmospheric circulation)
- isotope value is primarily sensitive to the temperature difference between the source and the site (remember the path the air parcel took). Changes that occur at the source will dampen the isotope signal changes at the core site.
- Complications at the ice core site. Seasonal variation in the isotope value is larger than the isotopic value change associated with climate change, even for the big transitions between glacial and interglacial. Ppt doesn’t necessarily fall uniformly throughout the year. Small changes in seasonal timing of ppt can cause changes in isotope values that may be misinterpreted as significant changes in climatic temperature.
These complications point to the need for independent temperature histories. None match the resolution of the isotope records.

Other temperature records:
Borehole temperatures – distribution of temperature through an ice sheet (vertical) can be a relatively direct measure of past temperature at the ice-sheet surface.
How this works: Heat is transported downward in ice via conduction and advection. Given constant surface temperature, temperature at depth will evolve to have a relatively isothermal upper layer, due to downward transport of ice, and warmer ice at depth due to heat flux from Earth and to heat generation in rapidly deforming basal layers.
If climate warms abruptly and maintains the new warmer temperature, the heat will be sent downward as a wave into the ice. Later measurements of the temp-depth profile (distribution) will reveal a cold spot, a direct remnant of the previous cooler climate. The natural heat flow can be reversed mathematically using heat and ice flow models to produce a history of temperature variations. (image on board of diagrams – 18-8)
Limitations with this method: reconstructions are only for long-term average temperatures and restricted to the last glacial period to recent times.
Strength of technique: the Temp – Depth profile is direct remnant of paleotemperatures at the ice-sheet surface. It provides a quantitatively accurate measure of long-term average temperatures allowing stable isotope records to be calibrated for major climate events.

Firn gas thermometry – gases are mobile in the firn layer, so that temperature-dependent changes in the composition of the gas is recorded in the gas trapped in the ice at the base of the firn.
A rapid temperature change in climate at the ice-sheet surface will lead to a temporary, steep temperature gradient throughout the Firn, causing, temporarily, thermal fractionation of the isotopes in the gas, which will be locked into the gas trapped in the ice at this time. So, the isotope ratios of the gases in the ice will have unusual values, spikes, which correspond to abrupt temperature changes occurring in the past. Useful for paleotemperature information: 1. magnitudes of the isotopic spikes are measures of the magnitude of change; 2. along core separation of the gas isotope values versus the solid ice isotope values gives a direct measure of the gas-ice age difference. This then allows an estimate of the temperature prior to the climate change, because the gas – ice age difference depends on temperature and accumulation rate.

Results of Thermometry
Ice core isotope records from No and So hemispheres clearly show cycling of Earth’s climate between cold glacial and warm interglacial conditions. Interglacials have been brief (1/4 duration of Glacials). Glacials tend to become progressively colder then terminate abruptly. The long term temperature patterns of variation is identical to the global ice volume patterns (seen in the ocean sediments). The polar temperature histories also show significant variations at ca. 20 and 40 kyr periodicities (similar to precession and obliquity cycles), and the 100 Kyr glacial-interglacial cycle dominates.
Average ice sheet surface temperatures during the last glacial period were
15 degrees colder than present on Greenland and
8 degrees colder than present in central Antarctica.

Millennial Variability patterns
Contrast the millennial variability of polar interglacial and glacial temperature histories. Throughout the glacial period in both hemispheres, polar climate changes were frequent and large. During the interglacial period polar climate has been relatively stable.
The climate change events during the glacial period are called Dansgaard-Oeschger (D-O) events and were rapid, millennial scale events. They have been recorded as abrupt events in all the ice cores from Greenland, and also appear in Antarctica cores (though muted).
The D-O events occurred in groups that outline a saw-tooth pattern of initial abrupt warming, followed by gradual cooling, and each group lasting on average 7,000 years, and called a Bond Cycle. Within each Bond Cycle, the D-O events show generally decreasing duration and maxima.
These events represent rapid environmental changes that had global significance, as seen in the fact that:
- the Bond cycles are correlated to changes in sea-floor deposits, the cold extremes of Bond cycles are correlated to massive iceberg discharge events from North American ice sheet into the North Atlantic
- coincident changes in methane concentrations in the ice gas bubbles – higher concentrations during warmer events. Natural sources of methane are wetlands, methane increases in ice gases implies increases in atmosphere and so are indicative of global temperature impacts because of the wide geographical distribution of wetlands.

Cause of D-O events.
Too rapid to be due to insolation changes. Most likely result from changes in ocean circulation. This is supported by the clarity of the Greenland ice cores relative to the Antarctic cores as Greenland is closer to the site of deep water formation.