A Robust Test of the Milankovitch Cycle Hypothesis Predictions for Influencing Abyssal Hill Morphology: A Pilot Study

点击翻译按钮获取中文摘要

Vast regions of the ocean floor are covered by features called abyssal hills, highly linear ridges that are formed at the volcanic mid-ocean ridges, and then rafted away as the tectonic plates spread apart. A new hypothesis contends that the topography of abyssal hills is closely linked to changes in sea level, and therefore to changes in Milankovitch cycles of the orbit of the Earth. If true, then the implications are potentially enormous: mapping the topography of the seafloor could be used to investigate Earth history, and thereby greatly enrich our understanding of Earth's varying climate associated with the changes in the orbit of the Earth. The hypothesis to be investigated suggests that the rising and falling global sea level, controlled primarily by Milankovitch-scale variability in Earth's orbit, modulates the intensity of volcanic eruptions at mid-ocean ridges and, in turn, the formation of abyssal hills on the seafloor. However, a connection between abyssal hills and Milankovitch cycles has been disputed by other researchers. There are two fundamental objections: (1) decades of observational evidence have shown that abyssal hills are dominantly fault-generated structures, rather than volcanic constructs, and (2) the Milankovitch Cycle hypothesis predicts that the width of abyssal hills will increase with spreading rate, contrary to observations indicating the opposite is true. Nevertheless, modulation of mid-ocean ridge volcanism by changes in sea level is a plausible and, indeed, exciting prospect, and recent work on abyssal sediments provides independent evidence in support of the hypothesis. The central problem is that, if there is a topographic signal of sea level-modulated volcanism in the morphology of abyssal hills, measuring this effect will require decoupling it from the dominant fault-generated component. This project will develop a methodology for measuring, if it exists, an age-dependent component of abyssal hill topography. The central hypothesis is that fault-generated topography is random with respect to crustal age, and can thus be treated as a correlated random noise process superimposed on any age-dependent (possibly volcanic) signal; this hypothesis is consistent with prior studies that model fault scaling parameters with random distributions governed by scaling laws. The proposed methodology will stack, or average, large quantities of data as a function of crustal age so that the random component (noise) destructively reduces while non-random component (signal) constructively sums. Characterization of the abyssal hill covariance structure is necessary to determine the expected variance of the stacked signal, and thus to rigorously test the null hypothesis that no age-dependent topographic signal exists; the null hypothesis can be rejected if the stacked profile exceeds the bounds for a given confidence level. The methodology will be applied to three large regions along the flanks of the East Pacific Rise that provide the most comprehensive bathymetric coverage available along the ridge flanks.