Do Trenches Advance?
David C. Mathews, Lin Zheng, and Richard G. Gordon
Department of Earth Science, Rice University, Houston, TX 77005
In recent years, the work of W. Schellart and colleagues has shown that, in an “absolute” frame of reference, some trenches at convergent plate boundaries advance, that is, they move toward the overriding plate instead of retreating seaward. Here we update and improve on prior analyses through use of recently estimated angular velocities and the propagation of errors. We apply the MORVEL global set of relative plate angular velocities and a new deep mantle reference frame derived from seismic anisotropy inferred from shear wave splitting data to estimate absolute trench velocities and uncertainties.
In contrast to prior studies, we find that trench retreat occurs along the entire South America plate subduction boundary. Consistent with prior studies, we find that the fastest trench advance occurs at locations along the Marianas- Izu-Bonin trench (43 ± 8 mm/yr (95% confidence limit)), especially along the Izu- Bonin segment, which appears to be the most clear-cut example of trench advance. We also find that the southern Kermadec trench (30 ± 8 mm/yr) advances significantly, but tomographic imaging and depth of seismicity at the southern Kermadec trench suggest that only a short length of slab has been subducted there, so that not much—if any—slab is advancing there. Some indications of trench advance depend on poorly constrained estimates of the angular velocities of microplates overriding subducting lithosphere, such as the Kermadec microplate, and our work highlights the need for new data to better constrain such estimates.
AAPG Search and Discovery Article #90182©2013 AAPG/SEG Student Expo, Houston, Texas, September 16-17, 2013