5.2 Implications for the timing of Carnegie Ridge subduction
The subduction of bathymetric highs has been suggested to trigger widespread deformation and uplift in non-collisional orogens (e.g., Lallemand et al., 1992; von Huene et al., 1997; Dominguez et al., 1998; Pubellier et al., 1999; Ramos, 2005; Martinod et al., 2013; Spikings and Simpson, 2014; Georgieva et al., 2016) and island arcs (Taylor et al., 1987; Mann et al., 1998; Rosenbaum and Mo, 2011; Freisleben et al., 2021). In addition, the subduction of oceanic aseismic ridges also appears to cause changes in the geochemical composition of magmatism with a shift to a more juvenile isotope signal (Chiaradia et al., 2020). Our thermochronological data from the Western Cordillera in Ecuador clearly document the onset of a regional uplift and exhumation phase that started at ~6-5 Ma (Fig. 4). Our data suggest that the Ponce Enríquez and Montalvo reverse faults are major structures controlling rock uplift and exhumation in the Western Cordillera. Importantly, these faults appear to be associated with transpression and the extrusion of the North Andean Sliver along the dextral Puna-Pallatanga-Cosanga strike-slip fault. Although the North Andean Sliver may have initiated as early as 15 Ma due to oblique subduction (e.g., Alvarado et al., 2016), the acceleration of northward motion of this block suggests increased coupling at the plate interface after the interception of the Carnegie Ridge (Witt et al., 2006). Importantly, earlier cooling and exhumation phases identified in the Eastern Cordillera and the Coastal Cordillera between 15 and 7 Ma (Spikings et al., 2005, 2010; Brichau et al., 2021) that have been associated with subduction of the Carnegie Ridge (i.e., Spikings et al., 2010), were synchronous with deformation and uplift recorded at 17-15 Ma, 13 Ma, and 9-8 Ma in the sub-Andes of northern Peru (e.g., Mégard, 1984; Sébrier et al., 1988; Moreno et al., 2020). These earlier cooling phases, however, are not recorded in the Western Cordillera, suggesting that earlier uplift and exhumation in other parts of Ecuador were independent from aseismic ridge subduction processes.
Initial Carnegie Ridge subduction at ~6-5 Ma is consistent with geodynamic reconstructions and with the results of studies of the offshore morphology and subsidence history of the Gulf of Guayaquil that propose an onset of ridge subduction at 5 Ma and document ongoing ridge subduction by ~1.8 Ma (e.g., Witt et al., 2006; Collot et al., 2019). Onset of ridge subduction at ~6-5 Ma is also compatible with the timing of regional compressional tectonic inversion of the forearc (Daly, 1989). Moreover, an eastward shift of volcanism at 6-5 Ma in northern Ecuador (Barberi et al., 1988) and a subsequent change in magma chemistry also support the notion of onset of Carnegie Ridge subduction at this time (e.g., Bourdon et al., 2003; Samadiego et al., 2010).
Taken together, our thermochronological and geochronological data suggest that the subduction of the Carnegie Ridge triggered shortening and exhumation in the Ecuadorian Andes starting at ~6-5 Ma (Fig. 5). However, a delay between the onset of ridge subduction and exhumation processes in the high Andes is possible. Thus, we cannot exclude an onset of Carnegie Ridge subduction at 9-8 Ma as previously suggested by Daly (1989) and Gutscher et al. (1999) based on plate kinematic reconstruction. Finally, Hafnium isotopic compositions of zircons in Ecuador become more juvenile at 8-6 Ma (George et al., 2020) and could also reflect the arrival of the Carnegie Ridge at the trench and ensuing changes in volcanic arc geochemistry at that time.