A continental reconstruction of hydroclimatic variability in South America during the past 2000 years

Paleoclimatological field reconstructions are valuable for understanding past hydroclimatic variability, which is crucial for assessing potential future hydroclimate changes. Despite being as impactful on societies as temperature variability, hydroclimatic variability - particularly beyond the instrumental record - has received less attention. The reconstruction of globally complete fields of climate variables lacks adequate proxy data from tropical regions like South America, limiting our understanding of past hydroclimatic changes in these areas. This study addresses this gap using low-resolution climate archives, including speleothems, previously omitted from reconstructions. Speleothems record climate variations on decadal to centennial timescales and provide a rich dataset for the otherwise proxy-data-scarce region of tropical South America. By employing a multi-timescale paleoclimate data assimilation approach, we synthesize climate proxy records and climate model simulations capable of simulating water isotopologs in the atmosphere to reconstruct 2000 years of South American climate. This includes surface air temperature, precipitation amount, drought index, isotopic composition of precipitation amount and the intensity of the South American Summer Monsoon. The reconstruction reveals anomalous climate periods: a wetter and colder phase during the Little Ice Age (similar to 1500-1850 CE) and a drier, warmer period corresponding to the early Medieval Climate Anomaly (similar to 600-900 CE). However, these patterns are not uniform across the continent, with climate trends in northeastern Brazil and the Southern Cone not following the patterns of the rest of the continent, indicating regional variability. The anomalies are more pronounced than in previous reconstructions but match trends found in local proxy record studies, thus highlighting the importance of including speleothem proxies. The multi-timescale approach is essential for reconstructing multi-decadal and centennial climate variability. Despite methodological uncertainties regarding climate model biases and proxy record interpretations, this study marks a crucial first step in incorporating low-resolution proxy records such as speleothems into climate field reconstructions using a multi-timescale approach. Adequately extracting and using the information from speleothems potentially enhances insights into past hydroclimatic variability and hydroclimate projections.