Glacial cycles drive variations in the production of oceanic crust

See allHide authors and affiliations

Science  13 Mar 2015:
Vol. 347, Issue 6227, pp. 1237-1240
DOI: 10.1126/science.1261508
  • Fig. 1 Simulated bathymetric relief driven by Plio-Pleistocene sea-level variation.

    (A) Imposed sea-level variation [black, from (14)] and predicted bathymetric relief (color) for the past 1.25 million years from simulations at three half-spreading rates U0 and three permeability levels K0. Isostatic compensation is assumed to scale the amplitude of crustal thickness variation by 6/23 to give bathymetric relief. Permeability in the simulations is computed by applying K(x,z) = K0(ϕ/ϕ0)3 m2 to the porosity field ϕ(x,z), where ϕ0 = 0.01 is a reference porosity. Light blue, dark blue, and red lines correspond to log10K0 = –(13, 12.5, 12), respectively. (B) Power spectral density estimates for each time series, made by using the multitaper method with seven tapers. Axes are logarithmic.

  • Fig. 2 Crustal thickness admittance, computed for a sinusoidal variation in sea level with period TSL.

    (A and B) Admittance curves derived from (A) numerical simulations and (B) the reduced model (13). (C) A plot of depth z versus the integrand from the reduced model of magma production owing to sea-level variation, Embedded Image (Eq. 1 and text following). The model is evaluated for U0 = 4 cm/year, K0 = 10−13 m2, and three values of sea-level oscillation period TSL.

  • Fig. 3 Bathymetry at a section of the Australian-Antarctic Ridge.

    A region of consistent bathymetry is indicated between the black lines (top right) and is shown in profile (bottom left, blue) after converting off-axis distance to an estimate of time. Time is zero at the approximate ridge center. Also shown is bathymetry after filtering frequencies outside of 1/150 ky and 1/10 ky (green), and simulated bathymetry (black, for U0 = 3.3 cm/year and K0 = 10−13 m2). Spectral estimates (bottom right) are shown for the unfiltered bathymetry (blue) and model results (black), where the latter are offset upward by an order of magnitude for visual clarity. Data availability is uneven across the ridge, and spectral estimates are for the longer, southern flank. Unlike in Fig. 1B, spectral estimates are prewhitened in order to improve the detectability of spectral peaks (supplementary materials). Vertical dashed lines indicate frequencies associated with 100-ky late-Pleistocene ice ages, obliquity, and precession. Axes are logarithmic. Statistical significance is indicated by the black bar at the top right; spectral peaks rising further than the distance between the mean background continuum (corresponding to the black dot) and 95th percentile (top of black bar) are significant.

Supplementary Materials

  • Glacial cycles drive variations in the production of oceanic crust

    John W. Crowley, Richard F. Katz, Peter Huybers, Charles H. Langmuir, Sung-Hyun Park

    Materials/Methods, Supplementary Text, Tables, Figures, and/or References

    Download Supplement
    • Figs. S1 to S3
    • Table S1
    • Full Reference List

    Additional Data

    Data File KR1
    Data File KR2

Navigate This Article