Multichronometer thermochronologic modeling of migrating spreading ridge subduction in southern Patagonia

Abstract

In Stevens Goddard and Fosdick (2019), we use thermal history modeling of a composite thermochronological data set to identify northward-migrating accelerated rock cooling (tcool) from 20 to 5 Ma in the southern Patagonian Andes (47°–54°S) that systematically predates subduction of the Chile Ridge by ~2–5 m.y. We propose that this signal reflects northward migration of the Chile Ridge and associated cooling caused by lithospheric thickening along the leading edge of the obliquely subducting ridge that drives topographic uplift and exhumation—herein the crustal welt model (CWM; Furlong and Govers, 1999). Husson et al. (2019) raise two points for consideration: (1) that north-south crustal shortening required to explain the magnitudes of rock exhumation inferred from thermochronology are not evident in the field geology of Patagonia, and (2) that an alternative mechanism, transient dynamic topography—hereafter, the dynamic uplift model (DTM)—better fits the spatiotemporal distribution of tcool in southern Patagonia. We agree that the CWM requires crustal thickening to drive exhumation, but disagree with Husson et al. about the magnitude of crustal shortening necessary to initiate the observed rock cooling. Husson et al. state that up to ~6.5 km of lithospheric thickening (and a corresponding 20%–25% crustal shortening) is needed to generate ~3 km of exhumation and reset apatite fission-track (AFT) ages. Here we re-emphasize that the migrating record of accelerated cooling, tcool, is not evident by individual thermochronometric dates in any single system (zircon U-Th)/He, AFT, and apatite (U-Th)/He) alone. Instead, thermal history modeling of multichronometer data sets resolves the onset of tcool along the Patagonian Andes at temperatures as low as 50 °C in some cases (Stevens Goddard and Fosdick, our figure DR2), indicative of <3 km exhumation. Residual elevated heat flow (Ávila and Dávila, 2018), both from inherited thinned lithospheric structure and passage of older slab windows, further reduces the magnitude of crustal shortening required to generate subsequent erosion and rock cooling. Furthermore, the bulk closure temperature of a single system can be a poor constraint for calculating the magnitude of required exhumation—and predicted shortening—as closure temperature is strongly dependent on the rock cooling rate, which, in this case, varies across samples (although all record accelerated cooling). Indeed, there is limited documented field evidence for north-south– verging structures in the Patagonian Andes. However, Georgieva et al. (2016) report northwest-southeast structures near the present-day Chile triple junction based on both field evidence and thermochronometric cooling patterns. Additionally, north-south and northeast-southwest– striking reverse faults have been mapped surrounding the nearby Liquiñe Ofqui Fault Zone (G. De Pascale, 2019, personal commun.), consistent with north-south shortening. Moreover, the CWM invokes coupling between the viscous mantle beneath the cooler overriding crust, which may concentrate distributed crustal thickening and shortening in the lower crust (Furlong and Govers, 1999). Husson et al. interpret orogen-scale cooling using the DTM in which the maximum amplitude of surface uplift is centered above the subducting ridge (Guillaume et al., 2009). In Patagonia, the DTM predicts the simultaneous onset of uplift along broad segments of the margin (400+ km) and migrates from ~54°S to ~45°S between ca. 14 and 9 Ma (Guillaume et al., 2009, 2013), as shown in figure 1C of Husson et al. Although we agree that dynamic topography associated with the Chile Ridge slab window is an important contributor to the surface response in Patagonia, we find the DTM predictions of uplift to be a poor fit with the spatiotemporal patterns of rock cooling along the orogen (Fig. 1). Figure 1. Chile Ridge segments SCR-1, SCR-2, and Esmerelda transform fault zone are plotted by latitude based on reconstructions by Breitsprecher and Thorkelson (2009). Dates of tcool, across all tectonic domains (modified from Stevens Goddard and Fosdick, 2019) poorly fit the timing and rate of predictions made by the dynamic uplift model (DTM) shown by the red line (Husson et al., 2019), but are consistent with the timing of terrace deformation (Guillaume et al., 2009).