Abstract

The prepuna of northern Chile, a transitional zone between the hyperarid Atacama Desert to the arid ‘Puna’ and Altiplano found between 2000-3200 masl, suffered widespread abandonment by indigenous Aymara communities beginning around the 1850’s. Towns were abandoned and local populations often fell by orders of magnitude. Causes cited for this abandonment mostly rely on sociocultural factors, e.g. the start of nitrate mining and coastal urban development. The lack of historical records of past climate change has hampered efforts to relate this abandonment to other factors such as drought. Likewise, isolated populations of columnar cacti with no active recruitment or regeneration have also been described along the lower limits of the prepuna. Such “relict populations” have been attributed to the prevalence of sustained drought but such explanations have not been tested. We use plant macrofossil and pollen analyses from 15 Abrocoma cinerea (Rodentia: Abrocomidae) middens collected at Quebrada La Higuera (18°S, 3400 masl, prepuna of Arica). The middens have been AMS14C-dated to the last 650 cal yrs BP and provide adequate temporal coverage. We use several key indicators to infer past changes in rainfall, including plant macrofossil composition, the presence of Parastephia (a shrub found today at higher elevations) pollen and midden pellet size diameters (a proxy for body size and productivity). Plant macrofossils reveal diverse plant communities with a strong affinity towards prepuna assemblages. Parastrephia pollen is absent in all of our middens until 350 years ago (1645 AD), increases abruptly at 1760 AD and peaks at 1805 AD. Our pellet diameter record shows that the largest animals appeared in the last 200 years. Overall, our results agree with other regional proxies that point to arid conditions in the central Andes during the Little Ice Age, between 1300-1750 AD. Precipitation then increased dramatically between 1760-1805 AD, which is recognized as a bonanza epoch in the historical and ethonographical records, followed by a major decline since 1850 AD that continues till the present. Our results imply that the many columnar cacti are relict populations that likely flourished during a wetter early 1800s. Aymara populations also flourished in the late 1700s to early 1800s. This surge was coeval with the pluvial event described here, as is the subsequent population collapse with increasing and sustained drought implying that regional climate change may play a major role in explaining indigenous human population trends in the Andes.