Bring the discussion of a couple of days ago further forward, here is a very interesting graphic from Ravelo and Ward [2004] showing the transition from Pliocene to the Pleistocene (read R to L), which gives a somewhat different perspective on the Pleistocene than just looking at the Vostok core back to 800K (although that’s interesting as well).
Figure 1. Ravelo and Ward Figure 1 with original caption.
In the Pliocene warm period, trees grew in high polar latitudes (e.g. Banks Island, Canadian Arctic Archipelago and the north coast of Greenland – both of which I posted on previously.) Again, it’s important to note that these fossils did not get to polar latitudes by continental drift, but were found there.
Ravelo and Ward show that tropical oceans were also warmer than in the Pleistocene. It’s interesting that the variations in the Pliocene on a millennial/10-millennial scale were much smaller than in the Pleistocene.
It’s also interesting to note that the depth of the most recent ice age (ending ~17/12,000 years ago depending on location) is the deepest in this record, which, in turn, would probably make it the deepest in the entire 50 million year Tertiary period. I’m not aware of any very convincing explanations for this. In passing, some present-day changes have plausibly been attributed to ongoing rebound from the most recent Ice Age – for example, distribution of trees in western U.S.A. are not in equilibrium and are still evolving from Ice Age distributions. I’ll discuss this some time.
Mostly, it seems to me that, when you step back and look at climate history from a geological perspective, the existence of change on all scales is really quite remarkable.
What caused the transition from a Pliocene climate to a Pleistocene climate? It seems plausible that tectonic changes have something to do with it, but there doesn’t seem to be any complete consensus about which changes or how. I’ve mentioned before that the closing of the Panama Isthmus in the Pliocene seems to be related to the transition, but I’ve seen arguments that it doesn’t yield the supposed impact. I’m not convinced that the arguments are right, but the question isn’t settled. Likewise, the Himalayas rose materially during the period, which changed its properties as a barrier to the monsoon. Perhaps there were other changes. The Samoan Passage is located on an important "hot spot" and its properties would change over this period. There are surprisingly few deep channels between the Southern Ocean and the North Pacific – perhaps changes here made a difference. The Australian Plate moved northward – closing of passages in the New Guinea area were discussed by Cane et al in various articles. Which tectonic changes "matter"? I suspect that they all make some sort of difference, but there certainly doesn’t seem to be any consensus on the details of exactly how they all inter-relate; my guess is that most specialists will emphasize the importance of "their" tectonic element, but that’s just a guess.
I’m not trying to make any magisterial pronouncements here; I’m just trying to draw attention to an interesting long record of climate change.
Ravelo, A.C. and M.W. Ward, (2004). The Role of the Tropical Oceans on Global Climate During a Warm Period and a Major Climate Transition, Oceanography 14. URL
Climate Audit 
10 Comments
Commenting on:
“It’s interesting that the variations in the Pliocene on a millennial/10-millennial scale were much smaller than in the Pleistocene”
It’s also interesting that nowadays the annual variations in the temperatures are much smaller in the tropics than in the north. On the other hand the recent increasing temperature trend (if really existing) is lower in the north than in the tropics, because summer temperature trend seems to be somehow buffered in the north. In Finland the summer temperature trend seems to correlate with northern hemispheric tree rings. In that way the tree rings seem to respond correctly to the measured temperatures in the north. They haven’t grown more because temperatures haven’t increased.
Larry Huldén
The most recent benthic stack that is supposed to represent climate is of Lisiecki and Raymo http://www.maureenraymo.com/current_projects.php#millennial
The graph is actually giving the variation in oxygen isotopes (d18O) of certain ocean floor dwelling (benthic)foraminifera, that are analysed from ocean sediment cores. some go back to much older strata. So if an ice sheet is building up it is made of light water with more 16O, because of the fractination during the evaporation. More 16O water evaporates so more 18O water stays behind. Now if the evaporated water does not return because it is used for ice sheet building then the amount of remaining 18O water in the oceans increases. So more 18O in the forams means: either the local water is colder or there is more 18O in the water, which means that there are polar ice sheets which means that it is cold.
However, there is a lot of “if-it-rains-the-streets-are-wet. The-streets-are-wet-so-it-rains” reasoning here. Foram d18O is also dependent on salinity and acidity, so a current change or other conditions changing may also cause d18O changes in forams. Then, ice sheets may only be building if the poles are land on a hundred million years scale. And due to of course of lot of plate tectonic every now and then a continent swarfs over the poles. Then a ice sheet may build but does it mean that the global temperature changes? I think not (that’s also regarding that greenhouse – ice house hypothesis in the other blog.
back to the benthic stack. Note also the big change around 900,000 years ago when the predominant weak cycle frequency shifted from ~41,000 years to the strong ~100,000 signal. The former is thought to be related to the obliquity cycle of the Earth spin axis. For the latter all current explanations do not suffice. This is a beautiful detective plot, waiting to be solved.
Looking on this scale the decline in temperature from 4 My BP looks almost linear. This would suggest the cause is a gradual change, such as the rise of the Hymalaya, or declining atmospheric CO2, rather than a sudden change, such as the closing of the Panama isthmus.
Andre, see my comments: http://www.climateaudit.org/index.php?p=191 #24 & #25. I think all sorts of hints can be found in that Nature article. I’ll add to #25 to say that an
Antartic bay, I forget which now, might have been broken up every 41K years before 850k years ago. This could be caused by sea level increase from melting, perhaps in other locations, lifting the ice off the floor of the bay. The pole was close to the bay before 850k years ago, but it has since moved inland. Warning I probably don’t know what I’m talking about, I have no formal education in this. But do look at the article for the data content.
Re #4; John, on your #25 mentiones the eccentricity cycle as predominant for the 100Ky. I know, you can find multiple textbook explications about this but still I don’t think it isn’t. The original Croll-Milankovitch eccentricity cycle is composed of two main components, the stronger cycle 410Ky and the weaker 90Ky. So it’s a mystery how it changed in the books to 100Ky after the discovery of that benthic foram signal and of course echoed in the Antarctic ice cores, 420Ky in Vostok, (Petit et al 1999) and 740Ky in EPICA Dome-C (Jouzel et al 2004). Richard Muller had another elegant hypothesis about this behavior, a cosmic dust band around the sun that the Earth would enter every 100Ky due to an inclination cycle.
http://muller.lbl.gov/papers/sciencespectra.htm
Now, dust indeed shows in the ice cores during the deepest pits in the cycles but closer investigation shows that it all Earth related, not cosmogenic. There is no other evidence like increased cosmogenic radioactivity so this hypothesis remains very tentative. Moreover if you really zoom into the abundant available details details of the last “termination” things are definitely not what they seem to be. There is still a lot to resolve.
Andre, some of the information on Milankovitich theory I was working with came from
http://www.usask.ca/geology/classes/geol206/glacial.html
It gives the cycle time for eccentricity changes as 92K to 100K years and changes in the tilt of the earth’s axis a cycle time of 41k years. de Gariedel-Thoron in the Nature article doesn’t mention Milankovitich but has in figure 2 of the article a couple of spectral analysis graphs of sea surface temperatures. One from 6 to 850kyrs, the other from 900 to 1,748kyrs ago. What they show is that the 100kyr cycle most prominent in the former and the 41kyr cycle in the latter. He has the 23kyr cycle also marked which is related to the precession of the equinoxes. The sliding of the Antartic continent over the pole saw the pole dive to the center of the Antartic land mass around 850kyrs ago. I wondered if that change might explain the switch from 41kyr to 100kyr glacial cycles 850kyrs ago. Back to the library! Thanks
Andre, Thank you for link to the Muler article.
Re #5; The inclination cycle would expose the earth more to the sun’s poles. The sun’s poles never get sunspots. Perhaps sunspots are new to the last 850Kyrs? You wouldn’t think the impact would be large enough. That the average irradiance could be so changed by 1.5 degree more or less of a solar pole view. A change in irradiance over solar latitude away from uniformity 850kys ago might do it if it was large enough.
I had confused axial tilt with precession in my earlier comments, so much of what I said belongs in the dust bin. Andre you are much too kind :).
Kenneth H. Schatten was almost a dozen years ahead of me on this. See Heliographic latitude dependence of the Sun’s irradiance, Journal of Geophysical Research, vol. 98, no. A11, p. 18,907-18,910
abstract: http://adsabs.harvard.edu/cgi-bin/nph-bib_query?bibcode=1993JGR….9818907S&db_key=AST
Seems like being on the sun’s spin axis is a cool place to be. More so 850Kys ago?
1. Do you buy the O18 proxy in these bugs?
2. Any danger that GW will plip us back into the Pliocene? (I would like it actually…alligators.)