Middle Pleistocene Transition

1
Middle Pleistocene Transition

• James Collins

2
Introduction

• The Middle Pleistocene Transition (MPT) marks a
  change in the response of the climate to orbital
  forcing
• Benthic d18O record from equatorial Pacific
  (Shackleton and Opdyke, 1976)
• Pre MPT, climate was dominated by the 41-kyr
  cycle. Post MPT, climate dominated by the
  100-kyr cycle.
• Began at at 1250 ka and was complete by 700 ka
• Earths orbital parameters did not change

Taken from Clarke et al, (2006)
3
Timing of the MPT

• Increase in amplitude across the MPT associated
  with glaciations
• Mean d18O increases over the MPT some
  combination of cooler deep water and more land
  ice
• Standard deviation increases due to increased
  amplitude of the glacial cycles
• Filtered stack
• Amplitude of the 100-kyr component increases
  across the MPT
• Time-frequency spectrogram, shows the
  distribution of the low-frequency power

Taken from Clarke et al, (2006)
4
Ice volume vs. Temperature

• How much of the increase amplitude of d18O is due
  to ice-volume and how much is due to global deep
  water temperature?
• Once deep water temperature is accounted for
  using Mg/Ca, there is an extra 0.4 increase
  post MPT, which equates to an extra ice volume
  equivalent to 50m of sea level.
• Geological evidence suggests that at the start of
  the NHG the area covered by ice was similar to
  that after the MPT

Taken from Clarke et al, (2006)
5
SSTs

• North Atlantic SSTs based on foram census counts
  decrease through the MPT
• Tropical Atlantic SSTs from an alkenone record
  minimum at 900 ka
• Benguela current alkenone record secular cooling
  over the MPT
• South Atlantic Modern Analogue Technique
  planktonic foram record. Glacial-interglacial
  variability began to increase at 870 ka due to
  increasing interglacial SSTs.

Taken from Clarke et al, (2006)
6
Monsoons and Aridity

• Percent of eolian sediment. No change over the
  MPT
• Flux of eolian sediment (African source). Shows
  an increase over MPT
• Flux of eolian sediment (Arabian source). Shows
  an increase over MPT
• Chinese loess grain size used to infer monsoon
  wind strength. Shows an increase at the MPT.
• Biogenic silica percentage in Lake Baikal,
  Siberia high biogenic silica indicates warmer
  temperatures and an increase in the strength of
  the monsoon. Shows an increase at the MPT.

Taken from Clarke et al, (2006)
7
Deep ocean circulation evidence

• d13C values in NADW and CDW records become
  depleted at the start of the MPT
• Change in ocean circulation
• Erosion of organic soils by growing ice sheets?
• -This would be a continual process and so may
  explain the steady decrease
• Multi-proxy approach needed to tell us more

Taken from Clarke et al, (2006)
8
900 ka significance

• Signal first appeared at 1250 ka, disappeared for
  100 kyr and then re-emerged at 900 ka.
• First long (80 kyr) glaciation of Pleistocene at
  900 ka
• SST minimum
• d13C minimum
• -Could be explained by transfer of organic matter
  from shelf deposits - Rapid increase in ice sheet
  thickness, sea level fall and exposure of organic
  rich shelf sediments

Taken from Clarke et al, (2006)
9
Mechanisms

• Recap mechanism must allow for
• - Increase in ice sheet thickness (same area,
  greater volume)
• - Decreasing SSTs
• - Increasing aridity
• - 900 ka SST minima
• All mechanisms revolve around the long term
  cooling decrease in atmospheric pCO2
• Berger et al (1999)
• Ice sheets never grow large enough during minima
  to survive moderate insolation maxima
• Under lower pCO2, only deglaciate under maximum
  insolation forcing (high eccentricity high
  obliquity boreal summer at perihelion)
• Rial (2004)
• Colder world allows for greater ice sheet extent

10
Mechanisms

• Tziperman and Gildor (2003) Sea ice switch
  hypothesis
• Related the MPT to long-term deep water cooling
• Sea ice switch mechanism sea ice is able to
  rapidly switch climate from glaciation to
  deglaciation
• For cold deep-water, rapid sea ice growth has the
  effect of starving land ice sheets of moisture
• For a warmer deep water, sea ice does not form
  until a cooler atmospheric temperature already
  in a state of ablation

• Additional support for this hypothesis
• Deep Ocean was cooler (Ruddiman et al 1989)
• IRD rapid deglaciation

11
Mechanisms

• Clark et al (2006) - Regolith Hypothesis
• Ice sheet erosion of a thick regolith to expose
  unweathered crystalline bedrock
• Low friction regolith allowed development of thin
  but laterally extensive ice sheets
• High friction nature of unweathered bedrock
  allows thicker ice sheets to form
• Thicker ice sheets respond differently to orbital
  forcing
• Possible feedbacks
• An increase in silicate weathering would decrease
  pCO2 by 7-12ppm, possibly acting as a feedback
• Shield weathering could provide alkalinity for
  coral growth during high stands

12
Regolith hypothesis -evidence

• Regolith
• Evidence for a former regolith saprolite in
  Minnesota, Canada and the Appalachians
• Chemical weathering
• Sediments that have undergone chemical weathering
  (ie regolith) will have undergone removal of more
  reactive mineral species
• Roy (2004b) established that tills younger than
  780 ka show major element concentrations most
  similar to fresh shield rock.
• Sr isotopes
• Regolith soils should have a low 87Sr/86Sr ratio.
  Erosion of continents would raise ration of
  oceans more than erosion of regolith
• eHf
• eHf accelerated at 1200 ka. May reflect enhanced
  crushing an partial dissolution of zircons.
• Os isotopes
• Reflect continental weathering.
• Os isotopes in weathered soil is depleted
  relative to bulk soil. However, dates may be
  inaccurate.

Taken from Clarke et al, (2006)
13
Problems with the regolith hypothesis

• Sr may just reflect an overall increase in
  weathering
• Increase is early would take time to reach the
  basement
• Hf timing is better
• Why the plateau?
• Os differences in dating give different curves
• Mechanism does not explain the SST peak at 900ka