In arid and semi-arid regions of South America, historical evidence for climate and vegetation change is scarce despite its importance for determining reference conditions and rates of natural variability in areas susceptible to modern desertification. Normal lines of evidence, such as pollen stratigraphies from lakes, are either rare or unobtainable in deserts; studies of late Quaternary vegetation history are few and generally inconclusive. This gap in knowledge may be corrected with discovery and development of fossil rodent middens in rocky environments throughout arid South America. These middens, mostly the work of Lagidium, Phyllotis, Abrocoma and Octodontomys, are rich in readily identifiable plant macrofossils, cuticles and pollen, as well as vertebrate and insect remains. In the North American deserts, more than 2,500 woodrat (Neotoma) middens analyzed since 1960 have yielded a detailed history of environmental change during the past 40,000 years. Preliminary work in the pre-puna, Monte and Patagonian Deserts of western Argentina, the Atacama Desert of northern Chile/southern Peru, the Mediterranean matorral of central Chile, and the Puna of the Andean altiplano suggest a similar potential for rodent middens in South America. Here we borrow from the North American experience to synthesize methodologies and approaches, summarize preliminary work, and explore the potential of rodent midden research in South America.
