Transcript 1. Thermal Fractionation - Georgia Institute of Technology

Timing of Abrupt Climate
Change of the Younger Dryas
H. Merritt, I.S. Nurhati, A. Williams
Paleoclimatology & Paleoceanography
Spring 2006
Overview
Severinghaus, J.P., Sowers, T., Brook, E.J., Alley, R.B., and M.L.
Benders. 1998. Timing of abrupt changes at the end of the Younger
Dryas interval from thermally fractionated gases in polar ice. Nature
391:141-146.




The Younger Dryas
GISP2
Gases in ice cores
Climate Implications
The Younger Dryas Stadial
o
o
o
o
o
Brief cold climate period (~1300 years)
Named for an Arctic Scandinavian flower
After Pleistocene and before warmer Holocene
Debated spatial extension (hemispheric or global?)
Some believed to be caused by Lake Agassiz
freshwater influx (=hampered thermohaline
circulation in the Atlantic)
Lake Agassiz
Evidences of Worldwide Impact
o Scandinavian forest turned to tundra
o Higher snowfall and glaciation rates in the
mountains of the world
o Higher amounts of dust from Asian deserts
o Drought in the Middle East (which may
have inspired the creation of Agriculture)
GISP2
o 3000m-deep ice core on the
summit of Greenland, drilled
near the European GRIP
core
o Back to >100,000 years,
and are believed to be valid
and agree down to a few
meters above Greenland’s
bedrock
o Have been used extensively
in recreating the climate of
the North Atlantic and the
world
Greenland Ice Core Records
o Drastic change about 11.6 ky bp that is well preserved in
the ice core
o Change came at the end of the Younger Dryas
o Due to the abrupt nature of change common methods of
climate reconstruction are not as effective as usual
Methane
o In the Greenland ice core, very high levels
of methane were found along this time
period
o Methane suggests high precipitation in
methane producing regions
o In order to better understand what
mechanisms are driving this, the
chronology of these events is key
Limitation
o The relationship of the δ18O ratio of ice and the
paleotemperature has been shown to change
over time, and may not be useful in certain
situations of abrupt temperature change
o Using δ18O, the temperature change leading
into the Holocene is underestimated by a factor
of 2
o Leads to search for independent
paleothermometer
Limitation (contd.)
o The air trapped in the ice is younger than the ice
30y (Law Dome, coastal)
7,000y (Vostok, interior)
o In times of rapid change like the end of the
Younger Dryas, this becomes an issue because
the slight difference in age of the air compared
to the age of the ice can make them have very
significant differences in composition
A New Way
o The way to confront the gas-age—ice-age
issue is to compare the composition of
gases to other gases
o By examining the thermal diffusion of
stable isotopes of atmospheric gas
trapped in ice, temperature can be found.
o This relies on the fact that gas mixtures
will fractionate in a temperature gradient
according to their mass
Obtaining Data
o Once ice core is drilled, the gases are
extracted and their isotopic compositions
are found through a melt-refreeze
technique that releases gases
o Mainly the center of these cores are used
to minimize the effect of the loss of gas
during retrieval and the handling of ice
samples
Analysis
o Once gas is collected, it is isolated from other
elements/molecules and then analyzed with a
mass spectrometer to determine how much of
each isotope is present in the sample
o For gases such as argon, which are much less
abundant than nitrogen, other gases may be
added to create a “solution” much like a
chemical in water so the sample has an
appropriate volume for the analytical apparatus
Air-Ice Core Gas Fractionation
Mixing with the
atmosphere
(~10m)
Thermal Diffusion
Fern (unconsolidated snow)
Diffusion and compaction occurs
Gravity Settling
ice bubbles sealed off
~70m in Greenland
~96m at Vostok
Air-Ice Core Gas Fractionation
1. Thermal Fractionation
- Thermal gradient drives
diffusive molecular transport
2. Gravitational settling
HEAVIER GAS IS ENRICHED
ON THE BOTTOM
HEAVIER GAS IS ENRICHED IN
COLDER REGION
Mass difference
Depth
AIR
Fractional deviation
of R and Ro
Example:
298K
δ15N (15N
15N
Temp
ratio
and
Thermal
diffusion
factor
14N)
ICE
80m,
236K
308K
δ15N=+0.2‰ on the cold-end
δ15N=+0.4‰
relative to top
15N
Heat & Molecular Diffusion in Firn
5ºC warming
15N
o With a +5ºC step function
o Gas diffuses 10x faster than
heat
o Diffusion rate depends on
the mass, ~7% faster for
heavier 15N14N
0.4‰ during a stable cold
period
~70m in Greenland
~96m at Vostok
+0.15‰ at 11.6kyr bp
followed by a decline
(recall +0.2‰ for our 10K
example)
X : previous study
Bad data points
excluded
Replicates pair of
data
Inflection point:
1700.3m = 11.64 kyr bp,
with ±20 yr uncertainty
Separating the thermal vs. gravity effects
A dynamic densification model predict a 6m deepening in fern column
= ↑ gravity settling  ↑ δ15N by 0.03%
Use δ40Ar (40Ar/ 36Ar)
- δ40Ar is not affected by
glacial-interglacial change
(unlike δ18O)
-
Ar is half sensitive to
thermal diffusion than N2
-
δ ~ Δm
δ15N (15N/ 14N), Δm N=1,
δ40Ar (40Ar/ 36Ar), Δm Ar=4
Hence, δ40Ar/4=δ15N
IF ONLY THERMAL EFFECT 
Change in: 2 x δ40Ar = δ15N
IF ONLY GRAVITY EFFECT 
Amplitude of: δ40Ar/4 = δ15N
Separating the thermal vs. gravity effects
IF ONLY GRAVITY EFFECT 
Amplitude of: δ40Ar/4 = δ15N
IF ONLY THERMAL EFFECT 
Change in: 2 x δ40Ar =δ15N
Ar amplitude is about ¾ instead of ½,
The anomaly in Ar is less than N2
suggesting gravitation effect through deepening suggesting the thermal effect
Abrupt warming temp (& corrected)
Severinghaus et al. (1998)
5-10°C of abrupt warming (highly tentative)
~ high analytical uncertainties
~ unknown thermal diffusion factor for N2
and Ar at -40°C
Grachev & Severinghaus (2004)
Revised to 10±4°C
~ acquiring the thermal diffusion factor
~ three different approaches involving δ15Nexcess,
δ15N, δ40Ar, and δ18O
www.aquatic.uoguelph.ca/wetlands/page1.htm
Methane and Warming at the End of the Younger Dryas
http://www.nasa.gov/centers/goddard/news/topstory/2005/methane.html
o
o
o
o
o
o
o
Pre-industrial source of methane was wetlands
Heavy rainfall increases standing water in bogs, which
increases methane production
Abrupt climate change at the end of the Younger Dryas was
thought to have been hemisphere wide
Amount of methane found was too high to be local; the
residence time of methane in the atmosphere is very short
Wetlands that produce methane are found hemisphere
wide.
Methane is not a very strong greenhouse gas.
Does methane cause climate change?
Methane seems to RESPOND to climate change, not CAUSE climate change
Methane and the Tropical HydrologyNADW Link
o There is a proposed link between changes in the
tropical hydrological cycle and North Atlantic
deep water (NADW)
Theory: Increased evaporation over the tropical
Atlantic would produce methane rise shown in
core, followed years later by an increase NADW
formation and Greenland temperature shown in
δ18O.
↑ Evaporation over tropical Atlantic (or increased precipitation in tropics)
Hemisphere increase in methane atmospheric concentration
Increase salinity of water, saltier warm water gets to poles decades later &
is cooled
Salty water sinks
Increase in NADW formation
Increased heat budget
More precipitation
Increase temperature in Greenland
According to this theory, the
methane rise would precede the
increase in temperature
indicated by δ18O by several
decades.
Conclusions
o Abrupt warming at the end of the YD (11.6 ky
bp) can be shown using δ15N and δ40Ar,
because δ18O is less useful for rapid change
o The diffusion of gas in ice core can be modeled
by the thermal and gravity gradient mechanisms
o 5-10°C (with revised=10±4°C is estimated for
the increase in temperature)
o Methane has proven not be the cause of this
abrupt warming event, rather a consequence
References
Grachev, A. M., and J.F. Severinghaus. 2004. A revised
+10±4°C magnitude of the abrupt change in Greenland
temperature at the Younger Dryas termination using published
GISP2 gas isotope data and air thermal diffusion constants.
Quaternary Science Reviews 24: 513-519.
http://en.wikipedia.org/wiki/Younger_Dryas
http://www.agu.org/revgeophys/mayews01/node6.html
http://www.ldeo.columbia.edu/res/pi/arch/examples.shtml