The southern Central Andes (SCA, 29°S-39°S) are characterized by the subduction of the oceanic Nazca Plate beneath the continental South American Plate. One striking feature of this area is the change of the subduction angle of the Nazca Plate between 33°S and 35°S from the Chilean-Pampean flat-slab zone (< 5° dip) in the north to a steeper sector in the south (~30° dip). Subduction geometry, tectonic deformation, and seismicity at this plate boundary are closely related to the lithospheric strength in the upper plate. Despite recent research focused on the compositional and thermal characteristics of the SCA lithosphere, the lithospheric strength distribution remains largely unknown. Here we calculated the long-term lithospheric strength on the basis of an existing 3D model describing the variation of thickness, density and temperature of geological units forming the lithosphere of the SCA.

The model consists of a continental plate with sediments, a two-layer crust and the lithospheric mantle being subducted by an oceanic plate. The model extension covers an area of 700 km x 1100 km, including the orogen (i.e. magmatic arc, main orogenic wedge), the forearc and the foreland, and it extents down to 200 km depth.

To compute the lithospheric strength distribution in the SCA, we used the geometries and densities of the units forming the 3D lithospheric scale model of Rodriguez Piceda et al. (2020a,b). The units considered for the rheological calculations are (1) oceanic and continental sediments; (3) upper continental crystalline crust; (4) lower continental crystalline crust; (5) continental lithospheric mantle (6) shallow oceanic crust; (7) deep oceanic crust; (8) oceanic lithospheric mantle; and (9) oceanic sub-lithospheric mantle.

The thermal field was derived from a temperature model of the SCA (Rodriguez Piceda et al. under review) covering the same region as the structural model of Rodriguez Piceda et al. (2020a). To calculate the temperature distribution in the SCA, the model volume was split into two domains: (1) a shallow domain, including the crust and uppermost mantle to a depth of ~50 km below mean sea level (bmsl), where the steady-state conductive thermal field was calculated using as input the 3D structural and density model of the area of Rodriguez Piceda et al. (2020b, a) and the finite element method implemented in GOLEM (Cacace and Jacquey 2017); (2) a deep domain between a depth of ~50 and 200 km bmsl, where temperatures were converted from S wave seismic velocities using the approach by Goes et al. (2000) as implemented in the python tool VelocityConversion (Meeßen 2017). Velocities from two alternative seismic tomography models were converted to temperatures (Assumpção et al. 2013; Gao et al. 2021). A detailed description of the method can be found in Rodriguez Piceda et al. (under review).

The yield strength of the lithosphere (i.e. maximum differential stress prior to permanent deformation) was calculated using the approach by Cacace and Scheck-Wenderoth (2016). We assumed brittle-like deformation as decribed by Byerlee’s law (Byerlee 1968) and steady state creep as the dominant form of viscous deformation. Low-temperature plasticity (Peierls creep) at differential stresses greater than 200 MPa was also included (Goetze et al. 1978; Katayama and Karato 2008). In addition, effective viscosities were computed from a thermally activated power-law (Burov 2011)

We assigned rheological properties to each unit of the model on the basis of laboratory measurements (Goetze and Evans 1979; Ranalli and Murphy 1987; Wilks and Carter 1990; Gleason and Tullis 1995; Hirth and Kohlstedt 1996; Afonso and Ranalli 2004). These properties were chosen, in turn, based on the dominant lithology of each layer derived from seismic velocities and gravity-constrained densities. More methodological details and a table with the rheological properties are depicted in Rodriguez Piceda et al. (under review). The rheological results using the thermal model derived from the seismic tomography of Assumpção et al. (2013) and Gao et al. (2021) can be found in Rodriguez Piceda et al. (under review, under review), respectively

Two comma-separated files can be found with the calculated lithospheric temperature, strength and effective viscosity for all the points in the model (2,274,757). These points are located at the top surface of each model unit. Therefore, the vertical resolution of the model is variable and depends on the thickness and refinement of the structural modelled units. SCA_RheologicalModel_V01.csv corresponds to the results using the mantle thermal field from the tomography by Assumpção et al. (2013) and presented in Rodriguez Piceda et al. (under review).

SCA_RheologicalModel_V02.csv includes the results using the mantle thermal field of Gao et al. (2021) and presented in Rodriguez Piceda et al. (under review). Each of these files contains the following columns:
- Northing as " X COORD (m [UTM Zone 19S]) "
- Easting as " Y COORD (m [UTM Zone 19S]) "
- Depth to the top surface as " Z COORD (m.a.s.l.)"
- Temperature in degree Celsius as " TEMP (deg. C) "
- Yield strength in MPa as “STRENGTH (MPa)”
- Effective viscosity in base-10 logarithm of Pas as “EFF VISCOSITY (log10(Pas))”

The dimensions of the model is 700 km x 1100 km x 200 km. The horizontal resolution is 5 km, while the vertical resolution depends on the thickness of the structural units.