Acabo de encontrar este paper de unos españoles, publicado en Pshysics letters del Institut de Ciencies del Mar en Barcelona sobre que los nucleos de hielo confirman los ciclos de Milankovith. Lo he encontrado citado en Solarcycle 24 como prueba de que (según los nucleos de hielo) el CO2 estaba en un nivel máximo justo en la parte media o final de una glaciación
http://www.icm.csic.es/files/oce/almacen/papers/AR-2007-09.pdfExtracto del post de solarcycle24
"Now, on the efficacy of CO2 as a climate trigger, there is some serious denial to be performed by known members here as regards the results of Sole, Turiel and Llebot writing in Physics Letters A (366 [2007] 184–189) which identified three classes of D-O oscillations in the GISP2 Greenland ice cores; A (brief), B (medium) and C (long), reflecting the speed at which the warming relaxes back to the cold glacial state:
“In this work ice-core CO2 time evolution in the period going from 20 to 60 kyr BP [15] has been qualitatively compared to our temperature cycles, according to the class they belong to. It can be observed in Fig. 6 that class A cycles are completely unrelated to changes in CO2 concentration. We have observed some correlation between B and C cycles and CO2 concentration, but of the opposite sign to the one expected: maxima in atmospheric CO2 concentration tend to correspond to the middle part or the end the cooling period. The role of CO2 in the oscillation phenomena seems to be more related to extend the duration of the cooling phase than to trigger warming. This could explain why cycles not coincident in time with maxima of CO2 (A cycles) rapidly decay back to the cold state. ”
The evidences discussed above could justify why A cycles
decay faster, but they do not explain the differences in decay
between B and C cycles, however, as CO2 maxima seem to
correlate equally well with both types of cycles. What could
explain such a difference is astronomical cycles [19], which are
known to affect to total budget of solar radiation received by
the Earth [14]. The relative abundance of class B (7 cycles) and
class C (2 cycles) suggest that the anomalously long cooling
phases for class B and class C cycles could be coupled with
an appropriate astronomical cycle. The existence of only two
C cycles separated away about 40,000 years suggest that C cycles
are coupled with the tilt cycle [24], while B cycles could
have an important coupling with the precession cycle (although
a periodic repetition of B cycles is less clear in the data). But
in no instance the warming phase seems to be perturbed by the
astronomical cycle.
....Nor CO2
concentration either the astronomical cycle change the way in
which the warming phase takes place. The coincidence in this
phase is strong among all the characterised cycles; also, we
have been able to recognise the presence of a similar warming
phase in the early stages of the transition from glacial to
interglacial age.
So it would seem, as is often reported in the literature, that we don't know yet what causes the brief but strong releases of energy that cause the warmings, we only know that CO2 was not the agent provocateur for the most recent 13 warmings prior to the Holocene. and those that have studied in depth the transitions in the Vostok, Dome Concordia and Law Dome cores know that CO2 did not play a role in any of the ice age/interglacial transitions dating at least back to near the Mid Pleistocene Transition.
Sole et al conclude:
"Some other sources of global variability, specially those ones hich could release sudden pulses of energy able to explain the steep warming phase in global oscillations, must be researched."
http://solarcycle24com.proboards.com/index.cgi?board=globalwarming&action=display&thread=908