Petrochronological Close-Up on the Thermal Structure of a Paleo-Subduction Zone (W. Alps)

Document Type


Publication Date




We present the results from a study combining zoned geochronology, mineral trace element geochemistry and phase equilibria modeling of a mafic lithology from the Zermatt Saas Ophiolite in order to elucidate the tectonic rates and thermal structure of the Western Alps paleo-subduction interface. The Zermatt-Saas Ophiolite represents dismembered slices derived from an eclogitized, 60-km wide coherent fragment of Tethyan oceanic lithosphere. Two cm-sized garnet crystals from a pyrite-rich chloritetalcschist (representative of sub-seafloor hydrothermally-altered basalts) from the Servette mine locality (St. Marcel Valley, Italy) were microdrilled to separate core and rim growth generations and dated using Sm-Nd geochronology to determine the overall duration of garnet growth. Garnet cores were dated to 46.9 ± 1.6 Ma, signifying the timing of the initiation of garnet growth. Garnet rims were dated to 43.5 ± 1.3 Ma, signifying the timing of burial to peak depths (constrained to ∼75 km). Major and trace element zoning in garnet suggest two distinct generations of garnet growth. Garnet crystal cores display evidence for rapid nucleation and growth, while garnet crystal rims suggest growth at relatively slow (tectonic) rates. The duration of garnet growth (3.4 ± 2.1 Ma), when coupled to constraints from phase equilibria modeling and Zr-in-rutile thermometry, provides estimates on the burial and heating rate of 4.7 km/Myr and 13.2 °C/Myr, respectively. Using the depth and temperature conditions recorded by laterally equivalent meta-ophiolites from across the Western Alps, a ‘snapshot’ of the geothermal gradient at the Alpine subduction interface is presented. This reveals a pressure-temperature trajectory highlighted by a significant decrease of the thermal gradient from ∼10 °C/km down to ∼3 °C/km at ∼45 Ma. We further present geometric calculations to identify the tectonic parameters that best fit this dual-segmented thermal gradient. Using convergence rates from paleogeographic reconstructions during Eocene times (on the order of 1.5-2 cm/yr), our calculations require a paleo-slab angle of 15- 20°, in line with dip estimates from most modern subduction zones. This study highlights how a detailed petrochronology-based approach enables assessing the thermal structure and geometry of a paleo-subduction zone, with implications on seismicity, stress distribution, and devolatilization potential of deeper portions of the subduction interface.