Abstract
Archaeological evidence plays a key role in longitudinal studies of humans and climate. Climate proxy data from Peruvian archaeological sites provide a case study through insight into the history of the "flavors" or varieties of El Niño (EN) events after ∼11 ka: eastern Pacific EN, La Niña, coastal EN (COA), and central Pacific or Modoki EN (CP). Archaeological proxies are important to the coastal Peruvian case because more commonly used paleoclimate proxies are unavailable or equivocal. Previously, multiproxy evidence from the Peruvian coast and elsewhere suggested that EN frequency varied over the Holocene: 1) present in the Early Holocene; 2) absent or very low frequency during the Middle Holocene (∼9 to 6 ka); 3) low after ∼6 ka; and 4) rapidly increasing frequency after 3 ka. Despite skepticism about the reliability of archaeological proxies, nonarchaeological proxies seemed to confirm this archaeological EN reconstruction. Although there is consensus that EN frequency varied over this period, some nonarchaeological and archaeological proxies call parts of this reconstruction into question. Here we review Holocene EN frequency reconstructions for the Peruvian coast, point to complexities introduced by apparent contradictions in a range of proxy records, consider the impact of CP and COA phenomena, and assess the merits of archaeological proxies in EN reconstructions. Reconciling Peruvian coastal paleoclimate data is critical for testing models of future EN behavior under climate variability.