Transcript Summary: Rapid Climate Change

What have we covered so far – the Basic Questions
1. Climate has not always been similar to the present; in fact has rarely
been like the present Holocene climate.
2. Climate depends on a large number of variables - with abundant
feedbacks, and climate change is not necessarily intuitive.
3. The scientific community does not understand some basics about
climate, even for recent periods where the data are good. This includes
the causes of the glacial/interglacial cycles.
4.Climate change can be both dramatic, and fast: the return to glacial
climate during the Younger Dryas may have happened in just a few
decades.
5. Computer models are just (digital) plausibility arguments, and are
limited by our understanding of what are the important variables, in
resolution, and in ability to predict the future.
Summary: Climate Conditions during
Last Glacial Maximum
• Insolation rates about the same as
today.
Solar Insolation Changes
• Summer insolation starts
to increase at ~20K years
and reaches a maximum
at ~10K years.
Not much ice can
remain over the summer
at 10,000 years BP.
Raymo’s thermal gradient argument?
She didn’t do her calculations forward to
the present time….
What do present solar insulation models predict for the future?
Summary: Climate Conditions during
LGM
• Insolation rates about the same as today.
• Colder (~ -4 º C globally and ~ -10 ºC (or –
20) near the poles and ~-2 to -3 ºC in tropics).
• Ice Sheet volume was ~ twice today.
• Sea Level lower by ~ 125m. Exposed large
continental shelf areas.
• Drier and dustier (globally).
• Reduced atmospheric CO2 and CH4 levels
• Vegetation more arctic like (tundra, steppe).
• Deep Ocean circulation more sluggish.
colder
Millenial Scale Oscillations
in Temperature DURING the
Last Glacial Period.
It wasn’t just cold back then….
During last glacial period (20100 K yrs BP) there were
many large and fast
temperature swings recorded
in the Greenland ice cores
and deep sea sediments from
N. Atlantic.
warmer
Summary: Rapid Climate Change
• During the last 100K yrs there have been repeated
oscillations between warm (Dansgaard-Oeschger: D-O)
and cold (Heinrich) conditions, with very fast
temperature changes (7ºC in 50 yrs in Greenland).
• The most recent (strong) cold event occurred about 12K
yrs ago (Younger-Dryas) during the transition from LGM
to current interglacial period.
• Current hypothesis is that variations in deep water
formation rates in N. Atlantic driven by salinity, and thus
poleward heat transport, is a likely mechanism for YD.
• These rapid climate change events are strongest in the
N. Atlantic, but some evidence that they occurred
globally.
• Generally, there is an anti-phasing of temperature
fluctuations between N. and S. Hemispheres.
Rise in Temperature and Atmospheric Gases
• Temperature, CO2 and CH4 start increasing ~18K yrs.
• Implies changes in radiation budget (temperature), ocean
circulation /biology/chemistry (CO2) and precipitation (CH4).
Ice Sheet Retreat
Retreat begins
~18K and BOTH
LIS and CIS ice
sheets are gone
by ~6K.
(Real calendar
age = 14C age
plus ~ 1700
yrs.)
Sea Level Rise
• Use 14C and
230Th/238U to date
the age of a
sequence of
submerged corals
that lived close to
the sea surface.
• The rate of sea
level rise has
pulses.
(14C ages are too
young by up to ~3K
yrs.)
Short discussion on why 14C years differ from calendar years
If you are dating corals
(say from Barbados),
you can use both 14C
and Th/U age dating
techniques.
BUT they don’t agree.
14C
dates are systematically
younger than the Th/U dates.
On the same samples.
Tree rings (count rings and do 14C dating on
wood) show the same effect, the 14C dates
are TOO YOUNG.
Here’ s why.
14C
is generated in the atmosphere by cosmic rays hitting 14N.
The 14C decays at a constant rate, but the rate of PRODUCTION of 14C depends on
the strength of the geomagnetic field.
If the field is strong, fewer cosmic rays hit the atmosphere. If it is weak (see below),
then MORE 14C is generated.
Formation:
14N
+ neutron => 14C + proton
Decay:
14C
=> 14N + b- + a neutrino
So – if the magnetic field was WEAKER than present, more 14C would have
been produced then, and more would still be around (in the corals) now.
This would make the corals appear YOUNGER than they are.
This means that you have to
correct 14C dates for the
changes in the geomagnetic
field.
If you do this, then 14C is a
reliable dating technique.
Transition from Glacial to Interglacial Conditions
• Rapid Bolling-Allerod warming
event at ~14.5 Kyrs.
• Transition from the LGM
climate to present interglacial
climate was not smooth.
• Younger-Dryas is a period of
cooling at ~12 Kyrs that lasted
for ~ 1000 yrs.
Greenland Ice Core
Younger-Dryas
Event:
atmospheric,
marine and
continental
proxies
Y-D
. climate signal is
strongest in N.
Atlantic region.
Rapid Temperature Change during the Y-D
Sawtooth
Pattern of Y-D
Heating is rapid,
Cooling is slower
At end of Y-D, temperature in Greenland increased by 7ºC in 50 yrs.
Role of Proglacial Lakes in
Climate Change
Possible Pathways of Meltwater Flow
Also out the Hudson
River valley
Appearance of Meltwater Pulses
From sea level record
From d18O-CaCO3 record
Why do meltwater pulses show up in d18O-CaCO3 record?
Meltwater Pulses as a Climate Change Trigger
• Pulse of freshwater discharge into the N.
Atlantic would reduce the formation rate
of deep water (N. Atlantic Deep Water) by
reducing salinity (density) of surface
water.
• This would reduce the ocean’s transport
rate of heat via Gulf Stream to N. Atlantic
region and cause cooling in the region.
Deep Water Formation: Present vs LGM
Possible Impact of Reduced NADW Formation Rates on Air
Temperatures
Millenial Scale
Temperature
Oscillations
During last glacial period
(20-100 K yrs BP) there
were lots of large and
fast temperature swings
recorded in the
Greenland ice cores and
deep sea sediments
from N. Atlantic.
Heinrich Events: Ice Rafted Debris. Remember, these
areEvents
COLD events
Heinrich
Dansgaard-Oeschger and Heinrich Events
recorded in the N. Atlantic Region
Fairly consistent ~1500 yr period between warm periods
(D-O events). Several of the coldest periods are associated
with Heinrich events (e.g., Younger-Dryas). Sawtooth
pattern of fast warming and slow cooling.
NOTE that there are a lot more D-O warm cycles than there are H-events.
That is because in order to have H-events, you need the ice sheets to be
advancing – which requires both cold temperatures AND moisture back on land
to force them to advance (which takes time). I.e., the H-events come at the end
of a cold period that is preceded by a warm period.
Possible Mechanism: Salt Oscillator
Hypothesis: Changes in
the salinity of the N. Atlantic,
resulting from ice melting or
ice formation, determines
the strength of NADW
formation and,
as a result, changes the rate
of heat transport to the N.
Atlantic region.
The H-event
Salt Oscillator: Ocean circulation has two
modes
Heinrich events are defined as ICE-RAFTED debris, found in
sediments in the mid-North Atlantic sediments.
How are D-O events defined?
What parameter defines a WARM D-O event?
Model Simulations
Does S. Hemisphere warm when the N. Hemisphere cools?
Salt Oscillator: Explaining the time
sequence of D-O and Heinrich events
Hypothesis: The
sawtooth pattern of
temperature change
is caused by a slow
decrease in rate of
NADW formation,
until it eventually
stops,
which is then
followed by rapid
return of NADW
formation after a
critical value of
salinity is reached.
small
big
Model Simulations
Does S. Hemisphere warm when the N. Hemisphere cools?
Antarctic warmed during Younger Dryas
• Antarctica temperature
increases during Y-D
cooling. Ice core data.
• Globally, atmospheric
CO2 levels increase.
• Globally, atmospheric
CH4 levels decrease.
Inter-hemispheric Seesaw
• Model simulations of Heinrich events indicate that
reduced heat transport into the N. Atlantic yields less
heat loss from S. Hemisphere and thus warming.
• Models indicate that when NADW formation is reduced,
then Antarctic Bottom Water formation rate increases,
which in turn means higher ocean to atmosphere heat
transfer and warmer temperatures in Antarctica.
• Temperature records during Y-D from Antarctic ice
cores indicate warming while Greenland cooled.
Were D-O events and Younger-Dryas global?
Map showing locations where abrupt climate changes (i.e., D-O events) have been
documented in marine sediments (red) or polar ice (blue). Yellow dots show those locations
where the last of these events (i.e., Younger Dryas) is recorded by major advances of
mountain glaciers. While for most of the globe, these events are in phase, in parts of the
Southern Ocean and of the Antarctic ice cap, they are clearly anti-phased.
Rapid Climate
change evidence in
Santa Barbara Basin
• Warming events
associated with negative
d13C, which author
interprets is a result of
methane hydrate
release.
• Did ocean circulation
rates change between
warming and cooling
events?
Conclusion: a LOT of methane was
released during the D-O interstadials
into the BOTTOM water of SB Basin.
Summary: Rapid Climate Change
• During the last 100K yrs there have been repeated
oscillation between warm (D-O) and cold (Heinrich)
conditions, with very fast temperature changes (7ºC in 50
yrs in Greenland).
• The most recent (strong) cold event occurred about 12K
yrs ago (Younger-Dryas) during the transition from LGM to
current interglacial period.
• Current hypothesis is that variations in deep water
formation rates in N. Atlantic driven by salinity, and thus
poleward heat transport, is a likely mechanism.
• These rapid climate change events are strongest in the N.
Atlantic, but some evidence that they occurred globally.
• Generally, there is an antiphasing of temperature
fluctuations between N. and S. Hemispheres.
How did the last Interglacial Period (the Eemian) end?
The Eemian was warm for about 10,000 years. No ice sheets in N. America or
Europe.
But at the end of the
Eemian, ice sheets grew
slowly (over 11K years),
and these were
punctuated by millennialscale cold/ice rafting
periods (H-events).
This has long-term
implications for the future:
AFTER the present
period of greenhouse gas
warming is over.
Compiled record of temperatures in Northern Europe, including tree-ring
data, black line is ‘instrumental’ records, growing (vineyard) seasons.
The “Little Ice Age”
Sea Ice off southern Iceland, compiled from records of the days that the fishing
fleet could actually leave port – and had to return to port.
Record of the
advance/retreat of
mountain glaciers
Derived from repeat
drawings of the
same scene in
European paintings.
We actually don’t
know the cause of the
Little Ice Age.
Other proposed
causes are the Plague
(by Ruddiman), that
caused massive
depopulation of
Europe and the regrowth of farmland into
forests.
Which pulled CO2 out
of the atmosphere and
cooled the climate.
Ice core record of Little Ice Age from
Peru (mountain) ice cap.
Top graph, showing dust and d18O
record of low temperatures (high
winds) during interval 1600 to 1900
period.
Lower plots showing (left) ice core
record taken in 1980, showing annual
ice core temperature record.
Lower plot (right) showing same coring
area in 1993, with ice melting. The
first melting of the Peruvian ice cap in
1000 years.
Moving into Modern Climates – compiled record from the last 1000 years.