Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

The missing player

  1. Jan 29, 2007 #1
    In the NERC discussions and somewhere here in a thread I remarked that the great mysteries of the Pleistocene Ice Age cannot be solved without finding the "missing player"

    We got him.

    Anyone curious? I'm ready to discuss.

    BTW my earlier threads about the role of clathrate for the extinction of the megafauna remains roughly valid, we're only one "why?" further. However Many "why's" to go before I sleep.

  2. jcsd
  3. Jan 29, 2007 #2
    What the hell are you talking about?
  4. Jan 29, 2007 #3
    So you still think that there is no problem or two or twenty with the hypotheses of the ice age? In that case we still need to do a little demolishion, like this fragment from the NERC discussions:

    http://home.wanadoo.nl/bijkerk/refuting the Greenland paleo thermometer.pdf

    I mentioned the missing player here: post #361

    It's about that missing player
    Last edited: Jan 29, 2007
  5. Jan 29, 2007 #4
    Andre, did you write or can cite a paper which would coherently describe and present your point and what you are trying to tell us here?

    I am not an expert on climate so i would appreciate something explanatory and cohesive which i could read and understand. These links, and online discussions are just getting accross ideas. That does not count much as anything without the proper work. To be honest i have only slight idea what you trying to say but i fail to see any connection to GW, glaciers, radiation, etc from those separate, unconnected links.

    Would something like that be in your powers to write?
    Last edited: Jan 29, 2007
  6. Jan 29, 2007 #5
    You are saying,Andre,that people should not base current models of anthropogenic greenhouse forcing on Pleistocene data because co2 data and radioisotope data from the Pleistocene do not agree.I have also read on Wikipedia that Milankovitch cycles cannot account for all the glacial-interglacial periods in the Pleistocene, on their own.Coral died in the Pleistocene because of something that changed in the deep ocean?
    Last edited: Jan 29, 2007
  7. Jan 29, 2007 #6
    I have been reading papers all day, none of them made any sense!!!
    (note: nothing to do with this discussion)
    Why did I chose geophysics??

    But yes, sneez is right, you're a specialist in a field that I personally have been to a few 1 hour seminars on/skimmed past in other reading. If you want us to make intelligent comments you need to give us something a little more solid, something that gives us the relevant background, and is accessible to someone who doesn't necessarily know all the nomenclature.
  8. Jan 29, 2007 #7
    Okay that makes sense. It makes a difference talking to a quartenary specialist or somebody whose textbook says in two paragraphs that there aint no mysteries in the ice age.

    I can cite loads of assumed facts that are proven wrong, papers that actually contradict but I don't think that anybody is waiting for that.

    One of the major problem areas is the ocean. ocean flows are slow and the body of water has a tremendous inertia. Nevertheless, ocean ice age papers are all about shift ands quick reactions of the ocean to any changes. If you compare the deep sea bed sediment proxies with the ice core proxies, they are not only virtually identical but also it appears that the ocean could be leading. Nobody appears to think that is this weird.

    Then there are the sea level problems. Only along the equator and for the last transition from the glacial there is some consistency in equatorial proxies but not elsewhere. That's even in the TAR of the IPCC. However strange things have happened before the former interglacial, the sea level rise appeared to be completed before the deglaciation started. There is plenty plenty more. I hinted to a few in several threads.

    The most aggravating non consistency with any hypothesis is probably the reason why so many easy errors have been introduced like the fake 100ka Milankovitch cycle. Things need to be in an orderly explainable fashion, but they aint.

    Here is the idea in a nutshell (let me know if there are difficult words in there):

    here is a modest animation that may help visualizing what has happened, advise to use page down/up keys.
    Last edited: Jan 29, 2007
  9. Jan 29, 2007 #8
    I also said: "And when the picture is complete, the reason of the passive CO2 reaction will be obvious"

    And indeed, Verdigris, With this completely different picture in mind, it will be clear why I think that the role of CO2 for climate thoughout the Pleistocene was (and is) completely insignificant.
  10. Jan 29, 2007 #9
    That's quite a neat little mechanism, have you checked the timescales? what kind of strain rates would you need, wouldn't there have been an increase in seismicity? have you run any computer simulations? have you published anything on this, I'd be interested in checking it out.
  11. Jan 30, 2007 #10
    Yes, course of events appear to match, but of course a lot of things need to be addressed

    It would be nice if specialist would be interested to judge that.

    That would have been a neat prediction, if it wasn't for the puzzling volcanic activity peak during the last glacial transition we already know about. For instance both the German Eifel and the French Massiv Central, mostly dormant, had very active volcanism during that period.

    It's a pristine idea however with dazzling matches of geologic features.

    You're most welcome. A lot needs to be done.

    first witness: The Mid Pleistocene Stilostomella extinction tells us when this mechanism started. These species could not cope with the dramatic changes in deep water waterflows which this mechanism triggered.

    When things seems logical all of sudden, you know that you're on to something.
  12. Jan 30, 2007 #11
    ANDRE said:

    "These species could not cope with the dramatic changes in deep water waterflows which this mechanism triggered"

    What in particular did the stilostomella have trouble coping with - water temperature change,nutrient availability changing?

    There is a nice photo of stilostomella here:
  13. Jan 30, 2007 #12
    Who knows. It's just likely that a typical darwian selection took place, when conditions changed, the species that were too specialized disappeared, while the most adaptable survived.

    As mentioned, one thing that did change was isotopes captured by those creatures in the deep sea beds, almost exactly at the same time as isotopes of Antarctic ice cores like this:


    It could be suggested that the oceanic isotopes were leading the ice cores, although the error bars are rather uncertain and the dating methods may easily get into circular reasoning when time scales are tuned to each other.

    However, ice sheets take centuries to millenia build up as can be derived from geologic data. This would determine the slope of the ice core proxies. However the THC circulation also takes more than a millenium to distribute that signal over the sea beds but also it mixes with the deep water so it would take even longer to get the ice build up signal to the deep sea. But it did not. The sea beds knew that the deglaciation started, practically before it actually did.

    Consequently the idea that both those isotopes represent ice sheet build up is clearly wrong. Whatever it is, it started at the sea beds and now we can assume that it represent mostly currents generated by the pulsating equator
    Last edited: Jan 30, 2007
  14. Jan 30, 2007 #13
    ANDRE said:
    "Whatever it is, it started at the sea beds "

    What if material from the surface sank to the sea beds e.g plankton,because of a high mortality rate from ozone depletion at the poles and increased uv light.
  15. Jan 30, 2007 #14

    The d18O is measured from shells of identified foraminifera species which are bottom dwellers (benthic). Isotopes are usually considered temperature proxies. Since the temperature at those depth around 2-4km are rather constant, be at at the equator or pole, it is thought that these isotope spikes represent the actual isotope ratio of the deep water. Then it was recognised that the ice sheets have very little heavy isotopes due to the fractination of the physical processes involved (evaporation - condensation). So the idea was that the oceanic d18O ratio was a measurement for the amount of ice locked on the poles. A sort of distilling process.

    However, if you consider the supposed global (eustatic) sea-level difference of ~125 meters then you're still about half a mil out of ~1,5 mil short of the required values, but you'd need already to put 15.4 Greenland ice sheets equivalents away somewhere. There is not too many room for that on Antarctica, already loaded with ice, perhaps 2-3 ice sheets. North America can have some 5-6 Greenland ice sheets equivalents before it is filled up like it was at the hight of the last Wisconsan ice age. So we need to tuck away ~7 Greenland ice sheet. No problem: in the 1980ies when this was all invented; there was a large and very unknown Siberia, lots of space out there. Plenty for so many Greenland ice sheets.

    It was not before the turn of the century when the researchers were allowed to swarm over that area that it became clear that had been no ice on Siberia at all during the Last Glacial maximum. This is all you get, perhaps just enough for two Greenland ice sheet equivalents.

    So I still have about 5 Greenland ice sheets left to stow away somewhere and that only covers the ~120 sea level change, not enough to balance the isotopes. That would have required some 20.2 Greenland ice sheet equivalents in total as well as a sealevel lowering of 164.5 meters which is not attained anywhere except perhaps for a certain lost Cuban city.

    So both the sea level yoyo as well the isotope balance hypothesis (Rutherford) do not match the actual numbers and can be considered falsified. Not that they are considered that, it's still used every day because it's the best there is perhaps.
    Last edited: Jan 30, 2007
  16. Jan 30, 2007 #15
    Looking again here it appears that the bulging reaction of the North Pole was not symmetrical. Plenty of ice on N-America, little ice in the east.

    Now, a little bit more to the East, just off the map at #1 the Laptev Sea, close to the coast, the Yukagir mammoth was found, with a calibrated age of some 22,000 years, just at the onset of the Last Glacial Maximum. As I elaborate upon here, page 5 in his intestine were remains of the greater burnet, a herb with limited winter hardiness, basically trashing the ice age idea.

    Now if we assume orographic effects, Scandinavia way above sea level in a less elliptical world then there would not be a requirement for Siberia to be that much colder because evidently it was warmer then as today.

    Interesting study:

  17. Jan 30, 2007 #16
    Nakada et al. 2000 gives a current rate of isostatic rebound for the continent of Antarctica as 1.7 mm yr-1. I don't know how long you propose these cycles take, I guess between 10,000 - 100,000 yrs. Which if we take the current rate and be a little generous, let's say 2 mm yr-1 you're getting about a 20 - 200 m change in elevation over those time scales. I guess maybe you'd have to half that to find the amplitude of a cycle, so maybe 10 - 100 m actual elevation change. Do these changes seem reasonable? Could they account for the orographic effects you mentioned?
  18. Jan 30, 2007 #17
    ANDRE said:
    "However the THC circulation also takes more than a millenium to distribute that signal over the sea beds "

    Why does it take this long? Are you talking about the entire surface of the sea bed around the world?

    ANDRE said:
    "Consequently the idea that both those isotopes represent ice sheet build up is clearly wrong. "

    Perhaps not - if the ice sheets built up in the sea around Antarctic.
    Then the weight of ice on water would influence currents and the flow of information just as the Odden feature does in Greenland.I have also read that Stilostomella thrive when the carbon flux is high.In a colder climate with low biomass - and low carbon flux - the stilostomella could have been starved and the increased sea water salinity may have killed them.Also the circumpolar current would have been weak because of less wind energy in the atmosphere.This would have helped ice to form on the sea.

    At this link: http://www.space.com/scienceastronomy/equator_bulge_020801.html
    Christopher Cox suggests that shifting ocean currents can account for the pulsing equator.Would this also account for the satellite orbital anomaly of 1998 (there were also anomalies in 1985 and 1989)?
    Last edited: Jan 31, 2007
  19. Jan 31, 2007 #18
    Isostatic rebound it a local process with tops at the North Baltic sea of 9mm a year. The pulsating equator would probably match that order of magnitude, seeing the speed at which equatorial sea levels change, say 120 meters in 12,000 years. Half of that probably to balance the water captured in the ice sheets, the other half for pulsating. Just ballpark figures. Mind that in a perfect ellipsoide for each procent that the equator bulges, the poles have 1,96% change, so that would be around 120 meters for bulging out and some more for the sea level lowering when the oceans rebalance. Seeing the difference of the Scandinavian and North American ice sheets versus the productive Siberian Mammth steppe, then it is clear that the actual situation is less than symmetrical. But I have yet to see another attempt to approximate a possible mechanisme that could explain those differences.

    It appears to be a very irregular cycle, but we cannot possibly says which part of which proxy is result of which effect. N equations with M variables and M >> N for the moment.

    I gues that the average total change - geoide - ellipsoid form change plus resettling sea level is ROM 120-200 meters. That's good for for max 1,3 degrees in a standard atmosphere. Doesn't seem much, we need local effects. Some more elevation change at some places than others. You need only a few domes to generate glaciers, plenty available in Scandinavia and NE Canada.

    other evidence is the volcanic tracer proxy at the Greenland ice cores showing "high activity during the climate changes"


    How would the East African rift valley fit in the scenario?
  20. Jan 31, 2007 #19
    The benthic stack of which I'm talking, check the paper here:


    Data of 57 different stacks at evenly divided over the world, all show the same. The THC needs some 1500 years to complete a cycle, where is the difference? The isotopes of the benthic stacks should register both the slow change in ice accumulation, the slow distribution if isotope ratio change and the slow mixing process. They show none of the latter two.

    Sorryu someone is hovering around to take over thisd computer. Back later
  21. Jan 31, 2007 #20
    How do we know the thermohaline cycle takes 1500 years?
    It's interesting to note that one of the mass extinctions of foraminifera targeted cylindrical forms in particular.
    This link gives some ideas researchers have about foriminifera extinction:

    During the Pleistocene glacials there were "pluvials" in which hot parts of the world got a lot of rainfall.Perhaps the surface currents in tropical regions became too warm and killed coral - and coral fragments sank from the surface and carried calcium carbonate to the deep sea,the calcium carbonate then becoming part of the foraminifera shells.This could account for the absence of a delay between the ice-sheet and benthic isotopes.
    Last edited: Jan 31, 2007
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Have something to add?

Similar Discussions: The missing player