Physics
Scientific paper
Apr 2003
adsabs.harvard.edu/cgi-bin/nph-data_query?bibcode=2003eaeja.....7856e&link_type=abstract
EGS - AGU - EUG Joint Assembly, Abstracts from the meeting held in Nice, France, 6 - 11 April 2003, abstract #7856
Physics
Scientific paper
Early Mars is assumed to have melted completely during accretion. The goals of this study are to create a simple model for Martian magma ocean crystallization and assess the outcome of overturn due to density instability. We explore the possibility that this can explain aspects of the Martian magnetic field and can reproduce the compositions of Martian meteorite source regions. The bulk composition of the Martian mantle is assumed to be that of Bertka and Fei (1997) renormalized without sodium. A 2000 km-deep magma ocean is modeled as fractionally crystallizing in half percent increments, with each increment crystallizing at the solidus (from Longhi et al., 1992), where its temperature remains. The phases crystallizing at each pressure are specified a priori: from 24 to 14 GPa, 50% majorite and 50% spinel crystallize. From 14 to 3 GPa, 10% garnet, 40% pyroxene, and 50% olivine crystallize, above which 40% pyroxene and 60% olivine crystallize. We present two models. The first, called the simple fractional crystallization model, is the result of fractionally crystallizing the magma ocean in the mineralogy described above. The second model, called the garnet segregation model, is a product of considering the density inversion at about 7.5 GPa. Below this inversion, olivine and pyroxene float in their coexisting silicate liquid, and garnet sinks. In both models, iron is increasingly enriched in the cumulates as crystallization proceeds, leading to the higher density of shallower cumulates. In the garnet segregation model, the garnet-rich layer is the densest and most aluminum-rich layer. The cumulate stratigraphy is gravitationally unstable in both models. Crystallization of the magma ocean produces a differentiated mantle, with enriched and depleted reservoirs. The constraint that the reservoirs are created by 10 to 100 Ma after accretion (Shih et al., 1999; Blichert-Toft et al., 1999) is consistent with rapid magma ocean crystallization. The mid-mantle is depleted in alumina and enriched in iron in the garnet segregation model; melting of this material may be consistent with SNC major element compositions. During cumulate overturn, hot material rising from depth may cross its solidus, producing melt through depressurization. When incomplete melt segregation are taken into consideration, the melt produced in the garnet segregation model is sufficient to create less than 100 km of crust over the planet surface. Overturn provides a possible mechanism for creating high core-mantle heat flow, and an attendant short-lived magnetic field. During overturn cold cumulates fall to the core-mantle boundary. Depending on the model, overturn creates a temperature difference across the core-mantle boundary of between 300 and 900 degrees C. Conductive heating of the lowermost mantle could sustain a heat flux out of the core for times on the order of a few hundred Myr. While the simple models presented here do not include all relevant physical processes, they are able to describe to first order the ages of differentiation and composition of mantle reservoirs, the moment of inertia factor of the planet, and they also provide a mechanism for magnetic field generation and cessation.
Elkins Tanton Linda T.
Hess Paul C.
Parmentier Marc E.
No associations
LandOfFree
A model for Martian magma ocean crystallization and overturn does not yet have a rating. At this time, there are no reviews or comments for this scientific paper.
If you have personal experience with A model for Martian magma ocean crystallization and overturn, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and A model for Martian magma ocean crystallization and overturn will most certainly appreciate the feedback.
Profile ID: LFWR-SCP-O-843962