Description
- Abstract:
- The objective of this thesis is to use first order geochemical and geophysical observations to understand melt generation and migration processes beneath mid-ocean ridge spreading centers. First-order observations are building blocks of the current understanding of the Earth. However, individual observation is a finite perspective. Conclusions drawn from one set of observations may not be fully respected by other theories motivated by different observations. This thesis believes that the power of two observations is greater than one, the more the better, and we can deepen the understanding of a geological system by synthesizing existing observations that are not easily combined. Such a philosophy will be expressed in the context of mid-ocean ridges (MOR) because a wide range of observations are available. In following paragraphs, examples will be outlined for how models and interpretation methods are improved to extract insights from a combination of existing observations. In Chapter 1, I examine an important geochemical problem to mantle melting beneath mid-ocean ridges: if chemical disequilibrium is needed to explain the observed rare earth elements (REE) in clinopyroxene (cpx) in abyssal peridotites. I read three pieces of information from concentrations of REE in cpx in abyssal peridotites. They are: the degree of melting, the melt porosity, and the extent of chemical disequilibrium. Observations of concentrations of REE in clinopyroxene (cpx) in abyssal peridotites has been used to constrain the first two parameters since the seminal work of Johnson et al. (1990). The importance of disequilibrium melting has also been recognized shortly after (Qin, 1992; Iwamori, 1993; Van Orman, 2002). The difficulty in modeling the effect of chemical disequilibrium is the coupling of grain scaler diffusion to tectonic scale melt migration. For that reason, the data of REE concentration in cpx has not been fully utilized to test the effect of chemical disequilibrium. In the paper published in Geochimica et Cosmochimica Acta 203:216-234, I developed a simple model that can predict concentrations of REE given the set of these three parameters. I implemented an inversion method, called Markov chain Monte Carlo method, to directly acquire the most probable estimate and uncertainty intervals of these three parameters. The main conclusion is a robust correlation between the degree of melting and the extent of disequilibrium. This correlation can be used to deduce the grain size of cpx and the potential temperature once the temperature-dependence of the diffusivity of REE and the melt productivity during decompression melting are known. In Chapter 2, I examine an important but previously unknown question on magma genesis beneath mid-ocean ridges: how the size and shape of mantle heterogeneity affect the correlation of isotope ratios in mid-ocean ridge basalts (MORB) and abyssal peridotites. The size of chemical heterogeneities has been inferred from both geochemical and geophysical perspectives to be a wide spectrum of scales, from centimeters (Warren and Shimizu, 2010) to thousands of kilometers (Hart, 1984; Garnero and McNamara, 2008). In a paper published in Science Advances 3 (11), I have put some constraints on the spectrum of scales by studying the effect of the size of chemical heterogeneities on these two sets of geochemical observations. Results show that the observed difference in 143Nd/144Nd in MORB and abyssal peridotites cannot be produced if most chemical heterogeneities are smaller than 1 km, and the observed range of 143Nd/144Nd in MORB cannot be produced if most chemical heterogeneities are larger than 60 km. These two constraints can be combined with the estimate of the enrichment of Nd in heterogeneities to put tighter constraints on the size of chemical heterogeneities. The purpose of Chapter 3 is to build a platform for assimilating as many geochemical and geophysical observations as possible. Any mid-ocean ridge model should be able to explain these first order observations. These include fine scale petrological features (Kelemen et al., 2000; Holtzman et al., 2003), crustal thickness, compositions of MORB and residues (Elliott and Spiegelman, 2003; Lundstrom, 2003; Warren, 2016; White and Klein, 2014), and geophysical interpretations of melt distributions (Baba et al., 2006; Yang et al., 2007; Key et al., 2013). By applying the approach of “double-porosity model”, I am able to capture fine scale petrological features and make internally-consistent predictions of some other observations such as the crustal thickness, the REE pattern in the pooled melt and abyssal peridotites, excesses of (230Th/234U) and (226Ra/230Th) in MORB, the depth of high porosity region beneath the ridge axis. These predictions are compared with real observations to obtain constraints on processes beneath mid-ocean ridges. Practices as such have improved the understanding of rock fabric-facilitated melt focusing and produced a preliminary constraint on the depth of the beginning of melt extraction.
- Notes:
- Thesis (Ph. D.)--Brown University, 2018
Access Conditions
- Rights
- In Copyright
- Restrictions on Use
- Collection is open for research.
Citation
Liu, Boda,
"MELT MIGRATION DYNAMICS AND TRACE ELEMENT FRACTIONATION BENEATH MID-OCEAN RIDGES"
(2018).
Earth, Environmental and Planetary Sciences Theses and Dissertations.
Brown Digital Repository. Brown University Library.
https://doi.org/10.26300/khy8-rv76
Relations
Collection:
-
Earth, Environmental and Planetary Sciences Theses and Dissertations
Theses and Dissertations for the Earth, Environmental and Planetary Sciences department....