High-relief glacial valleys shape the modern topography of the Southern Patagonian Andes, but their formation remains poorly understood. Two Miocene plutonic complexes in the Andean retroarc, the Fitz Roy (49°S) and Torres del Paine (51°S) massifs, were emplaced between 16.9–16.4 Ma and 12.6–12.4 Ma, respectively. Subduction of oceanic ridge segments initiated ca. 16 Ma at 54°S, leading to northward opening of a slab window with associated mantle upwelling. The onset of major glaciations caused drastic topographic changes since ca. 7 Ma. To constrain the respective contributions of tectonic-mantle dynamics and fluvio-glacial erosion to rock exhumation and landscape evolution, we perform inverse thermal modeling of a new dataset of zircon and apatite (U-Th)/He from the two massifs, complemented by apatite 4He/3He data for Torres del Paine. Our results show rapid rock exhumation recorded only in the Fitz Roy massif between 10 and 8 Ma, which we ascribe to local mantle upwelling forcing surface uplift and intensified erosion around 49°S. Both massifs record a pulse of rock exhumation between 7 and 4 Ma, which we interpret as enhanced erosion during the beginning of Patagonian glaciations. After a period of erosional and tectonic quiescence in the Pliocene, increased rock exhumation since 3-2 Ma is interpreted as the result of alpine glacial valley carving promoted by reinforced glacial-interglacial cycles. This study highlights that glacial erosion was the main driver to rock exhumation in the Patagonian retroarc since 7 Ma, but that mantle upwelling might be a driving force to rock exhumation as well.

Optically stimulated luminescence (OSL) thermochronometry is a tool for constraining cooling histories in low-temperature domains (several tens of degree Celsius) during the past 104–105 years [1][2][3]. This method is currently applied only to rapidly denuded regions (~5mm/yr when a general geothermal gradient in is assumed to be ~0.03℃/m) because luminescence signals in slowly denuded regions saturate before the rocks are exhumated to the surface. However, cooling histories in slowly denuded regions may be constrained if unsaturated samples are obtained from deep boreholes. In addition, using deep borehole core enable to compare the results between samples at multiple depths, which is useful to isolate the denudation history from other events, such as faulting or hydrothermal activity. We applied multi-OSL-thermochronometry [2] to the deep borehole core drilled at the Rokko Mountains, Japan, where slow denudation rates (0.1-1.0 mm/yr) are expected from previous studies [4][5][6]. We used the Kabutoyama core collected by National Research Institute for Earth Science and Disaster Resilience [5][7]. The total length of Kabutoyama core is 1,313 m and we collected the samples at 408, 642, 818 and 1048 m for OSL-thermochronometry. Our results showed that useful thermal information can be extracted from the infrared stimulated luminescence signals of samples collected at depths ≥408 m. We found that the sample temperatures remained around the present ambient temperature at each depth for the last 0.1 Myr, indicating that the Rokko Mountains is topographically stable, which was consistent with previous findings. Thus, the thermal denudation history of slowly denuded regions may be constrained by multi-OSL-thermochronometry using samples from deep borehole cores. However, the denudation rates in the Rokko Mountains were too low and could not be determined by this method. Further research is required to quantify the denudation rate. This study was funded by the Ministry of Economy, Trade and Industry (METI), Japan as part of its R&D supporting program titled “Establishment of Advanced Technology for Evaluating the Long-term Geosphere Stability on Geological Disposal Project of Radioactive Waste (Fiscal Years 2019-2021)”. References: [1] Herman et al. (2010). Earth and Planetary Science Letters, 297, 183-189; [2] King et al. (2016). Quaternary Geochronology, 33, 76-87; [3] Herman and King (2018). Elements, 14, 33-38; [4] Huzita (1968). The Quaternary Research, 7, 248-260; [5] Sueoka et al. (2010). Journal of Geography, 119, 84-101; [6] Matsuhi et al. (2014). Transactions, Japanese Geomorphological Union, 35, 165-185; [7] Yamada et al. (2012). Technical Note of the National Research Institute for Earth Science and Disaster Prevention, 371, 27p.