Timing and extent of glaciation in the tropical Andes.

摘要

We used surface exposure dating with cosmogenic radionuclides (CRNs; beryllium-10 and aluminum-26) to develop an absolute glacial chronology for the tropical Andes that spans multiple glacial cycle and includes exposure ages greater than 1 million years before present (1 Ma). We dated boulders on moraines in valleys bordering the Junin Plain (11 S, 76 W) in the central Peruvian Andes and in Milluni and Zongo Valleys in the Cordillera Real, Bolivia (16 S, 68 W). Our glacial chronology provides evidence that the local last glacial maximum (LLGM) in the tropical Andes was both earlier than the global last glacial maximum (LGM) and less extensive than several previous glaciations. Interpreting the chronology of moraines for pre-LLGM glaciations in the Junin region requires consideration of boulder erosion, surface uplift, boulder exhumation, and inheritance of CRNs. The presence of boulder surfaces with zero-erosion/zero-uplift exposure ages greater than 1 Ma indicates boulder erosion rates are approximately 0.3--0.5 meters per million years, which is comparable to some rates for Antarctica. After recalculating exposure ages with boulder erosion and surface uplift, we tentatively distinguish four periods of moraine deposition in Junin valleys. The depositional periods include every glacial marine isotope stages (MIS) from MIS 4 through MIS 16 except for MIS 12. The global ice volume record suggests that MIS 12 was a major glaciation, but MIS 12 may have been relatively minor in the tropical Andes, as was MIS 2. Published estimates of LGM snowline depression in the tropical Andes range from low (e.g., 200--300 m, Junin region) to high (e.g., 1100--1350 m, Cordillera Oriental, Peru). Although abundant data on paleosnowlines exists for the tropical Andes, absolute dating is lacking for most sites. Our estimates of LLGM snowline depression (approximately 300--600 m) reflect about half the temperature change of widely cited figures, which helps resolve the discrepancy between estimates of terrestrial and marine temperature depression during the last glacial cycle. Our chronology serves to emphasize both the challenges involved in interpreting pre-LGM surface exposure ages and the rewards of having chronological data, even with uncertainties, when trying to decipher the paleoclimatic history of a region.