Éditeur(s) : HAL CCSD Elsevier
Résumé : International audience
Laboratory models have been conducted to improve our understanding of the role that the resistance of the slab to bending and its coupling to the ambient mantle play in subduction dynamics over geological time scales. Our models are set up with a viscous plate of silicone (lithosphere) subducting under negative buoyancy in a viscous layer of glucose syrup (mantle). For our study, the lithosphere/upper mantle viscosity contrast has been systematically varied, from similar to 10 to similar to 10(5) in order to explore the parameter space between weak and strong slab dynamics. We found that subduction is characterized by a retreating mode for viscosity ratios > 10(4), by the coexistence of a retreating mode and an advancing mode for viscosity ratios between similar to 10(4) and similar to 10(2), and quasi-stationary, Rayleigh-Taylor like behaviour for ratios < 10(2). By combining our experimental results and kinematic data from current subduction zones in four reference frames which differ in the amount of net rotation, we infer that a lithosphere/upper mantle viscosity contrast of 150-500 is necessary to obtain realistic trench/subducting plate velocity ratios as well as the variability of subduction styles observed in nature.
ISSN: 0012-821X

hal-00412008
https://hal.archives-ouvertes.fr/hal-00412008
DOI : 10.1016/j.epsl.2008.04.006

Éditeur(s) : HAL CCSD Oxford University Press (OUP)
Résumé : International audience
Ash flow tuffs, or ignimbrites have been recently proposed to be a good material for palaeointensity determination. In this paper, we present a multidisciplinary study, combining geochronology, petrology, rock magnetism and palaeomagnetism, carried out on Oligocene to Early Miocene ignimbrites and related flows from the southern part of the Sierra Madre Occidental. Two new 40Ar/39Ar ages were determined for ignimbrites; 20.4 ± 0.2 Ma (the youngest age obtained so far in this area) and 29.2 ± 0.5 Ma. Density measurements, as a proxy for welding, proved to be extremely useful to estimate the emplacement temperature and the origin of the magnetizations carried by the ignimbrites. After alternating field and thermal demagnetizations, the mean palaeomagnetic pole (Lat = 66.8°N; Long = 180.5°E; Kappa = 142; A95 = 6.3°), calculated for the period 28-31 Ma, is in close agreement with our only Miocene determination. Comparison with the North America Synthetic Apparent Polar Path indicates a net counter-clockwise vertical axis rotation of about 10 ± 4° compared to stable North America, which occurred likely during the last extensional episode in the Late Miocene (ca. 12-9 Ma). Palaeointensity estimates, obtained with the Thellier-Coe method, are mainly questionable and should not be used for global interpretation. Therefore, these ignimbrites are not a viable material for reliable palaeointensity determinations.
ISSN: 0956-540X

hal-00854457
https://hal.archives-ouvertes.fr/hal-00854457
DOI : 10.1093/gji/ggt079