Search for: All records

Abstract The termination of the last glacial period is marked by the northward migration of the ITCZ and the weakening of the South American Summer Monsoon (SASM). The transition between the wetter glacial period and the more arid Holocene period across the South American continent is punctuated by several abrupt millennial-scale tropical hydroclimatic events. While the Northern Hemisphere temperature forcing of these millennial-scale events is generally accepted, recently, equatorial forcing mechanisms have been put forward. In particular, the dipole between northeastern Brazil and the western Andes of Peru is absent during Heinrich 1, with wet conditions recorded in both regions. To explain this anomalous atmospheric behavior, researchers have suggested changes in the ENSO and Walker circulation over South America and questioned whether the ‘amount effect’ relationship between δ18O and precipitation persists through time. To better resolve tropical hydroclimate changes over the last glacial termination, more robust paleoclimate proxies are needed. Here, we present a new paleo-precipitation reconstruction based on trace metal (Mg/Ca, Sr/Ca, and Ba/Ca) and isotope (δ18O and δ13C) speleothem records from Antipayarguna cave in the Peruvian Andes (3800 masl). Our records date from 2,600 to 4,700 and 7,700 to 19,000 years BP, with an average age resolution of 44 years. These records overlap the previously published speleothem records from nearby Pacupahuain and Huagapo caves. The Antipayarguna δ18O data are highly correlated with southern hemisphere summer insolation and the Huascaran ice core δ18O record. The Antipayarguna trace metal ratios and δ18O isotope values correlate well over most of the record, suggesting that the δ18O at our site reflects the amount of local precipitation. However, at the end of the Younger Dryas (11.5-10.3 ka) and Heinrich Stadial 1 (16.4-14.9 ka), there is a decoupling of these proxies. These anomalies may be due to changes in δ18O caused by shifts in moisture source region or precipitation condensation factors (e.g. convergence level or subcloud evaporation). Alternatively, this could be due to a change in trace metal sources. We explore potential causes for these brief decoupling events through comparison with other paleoclimate records. 

For the past few decades, many researchers have sought to understand how tropical hydroclimate responds to climate change via lakes, marine sediments, and speleothems records. Speleothem δ18O records throughout South America have shown that regional rainfall responds to Northern Hemisphere forcing on the millennial scale. Areas under the influence of the South Atlantic Convergence Zone (SACZ) have also shown a close relationship with local insolation on longer timescales. However, apart from the Cruz et al. (2007) record in Southern Brazil, long-term speleothem records throughout the continent have relied primarily on stable oxygen isotopes and are therefore limited to describing large-scale regional variability in rainfall. As such, many areas in South America still lack long-term records of local hydroclimate, which is critical to understanding how different components of the monsoon system respond to orbital and millennial-scale climate change. One proxy that has gained more attention in recent years is trace metal-to-calcium ratios (TM/Ca). Sr, Mg, and Ba to Ca ratios in speleothems are known in certain situations to respond to the degree of Prior Calcite Precipitation (PCP) above a drip site, a phenomenon directly tied to local aridity. In this study, we have obtained high-resolution TM/Ca measurements to pair with stable isotopes from samples spanning 23 to 66 ka from Huagapo Cave in the Peruvian Andes (11.27°S; 75.79°W). TM/Ca ratios in these samples are strongly correlated (R2>0.89), making them suitable for use as PCP proxies. We see that decreases in δ18O during Heinrich events are accompanied by a drop in TM/Ca. The period defined by the MIS 4/3 transition is accompanied by a simultaneous increase in TM/Ca and δ18O. TM/Ca and δ18O negatively correlate with local insolation for the entire record. Interestingly, the Paraíso Cave record from the Amazon Basin shows no correlation between regional or local hydroclimate and insolation during the last glacial period. The discrepancy between the two records and the close relationship between TM/Ca, δ18O, and local insolation in Huagapo samples, may call for a revised interpretation of Andes speleothem δ18O variability, which was originally thought to reflect rainout over the Amazon Basin.