THE DURATION OF PEAK METAMORPHISM IN THE CONNECTICUT VALLEY SYNCLINORIUM FROM AN ANALYSIS OF DIFFUSION ZONING IN GARNET

Diffusional zoning of Mn along the rim of garnets from samples collected from the Connecticut Valley Synclinorium (CVS) in Vermont and Massachusetts have been modeled using numerical diffusion models to estimate the duration over which the zoning formed. Typical core-rim Rayleigh depletion zoning of Mn in garnets from the garnet to the staurolite-kyanite zones is decorated by symmetrical increases of Mn up to several mole percent spessartine over length scales of several tens to a few hundred µm. These zoning profiles are interpreted to have formed from resorption of the garnet rim and diffusion of Mn back into the garnet. The diffusion model incorporates a moving garnet boundary of changing composition with time. Critical inputs are the initial Mn zoning profile (before resorption), the change in garnet rim composition and position with temperature, and the temperature-time history over which the diffusion occurs — the most critical input being the temperature–time history.

Excellent fits between models and observed zoning are possible with a range of assumed garnet boundary conditions. However, uncertainty in the maximum temperature of around 25 ˚C propagates to an uncertainty of around 2 in the total time, so estimates of the peak temperature are essential. Fe-Mg exchange thermometry between garnet and biotite yields temperatures that are lower than those inferred from petrogenetic grids and pseudosections. Quartz in garnet (QuiG) barometry has been used to infer pressure of garnet growth. These are combined with temperatures estimated from thermobarometry by extrapolating resorbed garnet rims to provide estimates of peak temperature that are more consistent with phase equilibria.

All zoned garnets analyzed give time scales for the peak metamorphic episode that range from a few hundred thousand to a few million years — results that agree well with timescales determined from Ti diffusion in quartz. These timescales are very short compared with times inferred from 1-D thermal models of crustal thickening of several tens of millions of years. It is unlikely that they are the result of thermal pulses from intrusions. Rather, the cause is postulated to be rapid tectonic burial and exhumation of either individual tectonic slices or a continuum of slices at plate tectonic velocities of at least 5 cm/year.