Investigation of Long-Term Drought in Mesoamerica Using Lacustrine Proxy Records, Instrumental Data, and Model Output

Paleoclimate research in Mesoamerica suggests that multidecadal droughts in the late Holocene were a factor in the demise of many pre-Columbian sites. However, our understanding of the spatial patterns of these droughts, and the causal mechanisms underlying them, remains poor. Our research is motivated by two main questions: First, we ask whether there is evidence that late Holocene droughts in Mesoamerica were spatially coherent across broad spatial scales. Second, we ask what mechanisms may be responsible for the patterns of paleoclimate change observed in proxy records. To address our first question, we present a record of late Holocene paleoclimatic change from the maar lake Aljojuca, in the eastern Trans-Mexican Volcanic Belt. A chronology established via radiocarbon dating shows that the core spans 6,200 cal. years B.P. We use geochemical proxies, in particular oxygen isotopes from authigenic carbonates, to reconstruct changes in lake level. We also draw on previously published lacustrine paleoclimatic records from across Mesoamerica. Using resampling techniques, we evaluate the impact of age uncertainty on estimates of the timing of late Holocene droughts over the past 2000 years. Our initial results suggest evidence of coherent drought in proxy records from the Yucatan Peninsula to highland Mexico between 1300 and 1000 cal yr. B.P., although there are subsequent intervals when proxy records show diverging trends. To address our second question, we use multivariate statistical techniques to explore coupled patterns of variability between Mesoamerican rainfall and remote sea surface temperatures in instrumental data and climate model control simulations (i.e. CCSM4.0's pre-industrial control simulation). Initial results suggest that low-frequency changes in rainfall may be forced by changes in SSTs in the Pacific or Atlantic basins that alter patterns of moisture transport.