Reply on RC2

Abstract

The South American Monsoon System (SAMS) is the main driver of regional hydroclimate variability across tropical and subtropical South America. It is best recorded on paleoclimatic timescales by stable oxygen isotope proxies, which are more spatially representative of regional hydroclimate than proxies for local precipitation alone. Network studies of proxies that can isolate regional influences lend particular insight into various environmental characteristics that modulate hydroclimate, such as atmospheric circulation variability and changes in the regional energy budget as well as understanding the climate system sensitivity to external forcings. To extract the coherent modes of variability of the SAMS over the Last Millennium (LM), we use a Monte Carlo Empirical Orthogonal Function (MCEOF) decomposition of 14 δ18O proxy records and compare them with modes extracted from a similar decomposition using isotope-enabled climate models. The two leading modes reflect the isotopic expression of the upper-tropospheric monsoon circulation (Bolivian High – Nordeste Low waveguide) and the latitudinal displacement of the South Atlantic Convergence Zone (SACZ), respectively. The spatial characteristics of these modes appear to be robust features of the LM hydroclimate over South America and are reproduced both in the proxy data and in isotope-enabled climate models, regardless of the nature of the model-imposed external forcing. Model analyses suggests that the local isotopic composition is primarily a reflection of an upstream rainout processes. The proxy data document that the SAMS was characterized by considerable temporal variability throughout the LM, with significant departures from the mean state during both the Medieval Climate Anomaly (MCA) and the Little Ice Age (LIA). The monsoon was intensified during the LIA over the central and western parts of tropical South America and the South Atlantic Convergence Zone (SACZ) was displaced to the southwest. These centennial-scale changes in monsoon intensity over the LM are underestimated in climate models, complicating the attribution of changes on these timescales to specific forcings and pointing toward areas of important model development.