Options

Blog - week 318

I'm starting the second article of This Week's Finds about glacial cycles and Milankovitch cycles, here:

I'll post another comment here when I've got enough to be worth looking at!

Comments

  • 1.
    edited August 2011

    Nice blog post! It is very clear on what we know and what we don't know about the relation between the Milankovic cysles and the climate.

    There is one point about these power spectra - but maybe that's just me, since it's a point that is usually not stressed. Suppose that you had a dynamical system with approximate periods of say 41 ka. Then, small pertubations could easily bring the periods out of phase, so that these periods would not show up in the Fourier series. But this does not happen, It's not just the length of the cycle that stays constant, it's phase doesn't seem to change either. So even before comparison to the astronomical data, the existence of these peaks in the power spectrum tells us that there is a fairly accurate clock working behind the scene, one that does not get lagged over many periods. It's hard to think of one except for the Milankovic cycles.

    Another thing, there must exist more recent and more precise data on than those used in 1976 by Hays et al. Is there an updated version of that paper, computing the power spectrum more accurately and over a longer stretch of time?

    Comment Source:Nice blog post! It is very clear on what we know and what we don't know about the relation between the Milankovic cysles and the climate. There is one point about these power spectra - but maybe that's just me, since it's a point that is usually not stressed. Suppose that you had a dynamical system with _approximate_ periods of say 41 ka. Then, small pertubations could easily bring the periods out of phase, so that these periods would not show up in the Fourier series. But this does not happen, It's not just the length of the cycle that stays constant, it's phase doesn't seem to change either. So even before comparison to the astronomical data, the existence of these peaks in the power spectrum tells us that there is a fairly accurate clock working behind the scene, one that does not get lagged over many periods. It's hard to think of one except for the Milankovic cycles. Another thing, there must exist more recent and more precise data on than those used in 1976 by Hays et al. Is there an updated version of that paper, computing the power spectrum more accurately and over a longer stretch of time?
  • 2.
    edited August 2011

    Okay, I'm almost done with "week318". If anyone has suggestions to make, please make them now!

    Marcel wrote:

    Then, small pertubations could easily bring the periods out of phase, so that these periods would not show up in the Fourier series.

    If they slowly went out of phase, I thinki they would still show up as peaks in the Fourier transform of a finite-length time series (i.e., the sort of Fourier transform we actually can do).

    So even before comparison to the astronomical data, the existence of these peaks in the power spectrum tells us that there is a fairly accurate clock working behind the scene, one that does not get lagged over many periods. It's hard to think of one except for the Milankovic cycles.

    I agree with that!

    Another thing, there must exist more recent and more precise data on than those used in 1976 by Hays et al. Is there an updated version of that paper, computing the power spectrum more accurately and over a longer stretch of time?

    There must be, but I don't know it yet.

    Comment Source:Okay, I'm almost done with "week318". If anyone has suggestions to make, please make them now! Marcel wrote: > Then, small pertubations could easily bring the periods out of phase, so that these periods would not show up in the Fourier series. If they slowly went out of phase, I thinki they would still show up as peaks in the Fourier transform of a finite-length time series (i.e., the sort of Fourier transform we actually can do). > So even before comparison to the astronomical data, the existence of these peaks in the power spectrum tells us that there is a fairly accurate clock working behind the scene, one that does not get lagged over many periods. It's hard to think of one except for the Milankovic cycles. I agree with that! > Another thing, there must exist more recent and more precise data on than those used in 1976 by Hays et al. Is there an updated version of that paper, computing the power spectrum more accurately and over a longer stretch of time? There must be, but I don't know it yet.
  • 3.
    edited August 2011

    back around 1875. He decided that what really matters is the amount of sunlight hitting the far northern latitudes in winter. (...)

    in the 1920s, Milankovitch made the opposite claim: what really matters is the amount of sunlight hitting the far northern latitudes in winter.

    ??? :-)

    Comment Source:> back around 1875. He decided that what really matters is the amount of sunlight hitting the far northern latitudes in winter. > (...) > in the 1920s, Milankovitch made the opposite claim: what really matters is the amount of sunlight hitting the far northern latitudes in winter. ??? :-)
  • 4.
    edited August 2011

    I would think if there were an updated version of the Hays paper between 1976 and 2001 it would have been cited here (you've probably seen this one cited on wikipedia):

    I didn't see it there, nor could I find a more recent review. I changed what I think were a few minor typos on the wiki. Hope that's ok.

    Comment Source:I would think if there were an updated version of the Hays paper between 1976 and 2001 it would have been cited here (you've probably seen this one cited on wikipedia): * J. Zachos, M. Pagani, L. Sloan, E. Thomas, and K. Billups, [Trends, rhythms, and aberrations in global climate 65 Ma to present](http://rivernet.ncsu.edu/courselocker/PaleoClimate/Zachos%20et%20al%202001%20Cenozoic%20Climate.pdf)., [Science](http://dx.doi.org/10.1126/science.1059412), vol. 292, no. 5517, pp. 686-93, Apr. 2001. I didn't see it there, nor could I find a more recent review. I changed what I think were a few minor typos on the wiki. Hope that's ok.
  • 5.

    A small suggestion - maybe a Wikipedia link for greybody?

    Naively, we can treat the Earth as a greybody.

    (Or a reference to Tim's blog post where it is defined)

    Comment Source:A small suggestion - maybe a Wikipedia link for greybody? > Naively, we can treat the Earth as a greybody. (Or a reference to [Tim's blog post](http://johncarlosbaez.wordpress.com/2011/06/19/putting-the-earth-in-a-box/) where it is defined)
  • 6.
    edited August 2011

    Thanks a million, Martin Cameron and Frederik.

    [James Croll] decided that what really matters is the amount of sunlight hitting the far northern latitudes in winter.

    in the 1920s, Milankovitch made the opposite claim: what really matters is the amount of sunlight hitting the far northern latitudes in winter.

    Cut-and-paste error! Milankovitch realized it's the summer sun in high latitudes that matters!

    I will add a link to the term "greybody".

    Comment Source:Thanks a million, Martin Cameron and Frederik. >[James Croll] decided that what really matters is the amount of sunlight hitting the far northern latitudes in winter. > in the 1920s, Milankovitch made the opposite claim: what really matters is the amount of sunlight hitting the far northern latitudes in winter. Cut-and-paste error! Milankovitch realized it's the _summer_ sun in high latitudes that matters! I will add a link to the term "greybody".
Sign In or Register to comment.