Stable Isotope Analysis of Lake Sediments from Laguna Santa Elena and Laguna Azul, Costa Rica

Lake sediments are increasingly important archives of human-environment interactions and paleoclimate in the neotropics. In Costa Rica, Anchukaitis and Horn (Palaeogeography, Palaeoclimatology, Palaeoecology 221: 35–54, 2005) established a land-use history for Laguna Santa Elena (8.9306 N, 82.9275 W, 1055 m elevation), a small lake in the Diquís archaeological region, based on pollen and charcoal analyses of a 7-meter sediment core. I carried out stable carbon and nitrogen isotope and loss-on-ignition analyses at higher resolution to extend the existing 2000-year record. The new geochemical data parallel major trends in botanical proxies but also reveal aspects of human and environmental dynamics not apparent in the prior analysis. Inferred changes in land use in the watershed are consistent with archaeological evidence. Geochemical trends strongly suggest a population collapse at the site around the time of the Terminal Classic Drought of the Mayan region. The generally close correspondence between microfossil assemblages and geochemistry in the Santa Elena core demonstrates the usefulness of stable isotope analysis as a first line of investigation in paleoenvironmental research.

Sediment samples for carbon isotope analysis need to be acidified to remove carbonates that can affect isotope measurements, and debate exists over whether nitrogen isotope analysis can use these acidified samples or require non-acidified samples. My thesis research tested the effects of pre-analysis acidification of sediment and soil samples from Laguna Santa Elena and a second lake in Costa Rica, Laguna Azul (9.9558 N, 83.6519 W, 630 m elevation) in the Central Highlands-Atlantic Watershed archaeological region. Results show that acidification may cause statistically significant differences in nitrogen isotope values. These differences appear to be random and unpredictable, and can manifest as either positive or negative shifts that have the potential to alter or even reverse relative trends in nitrogen isotope signals in lake sediment profiles. More tests are needed, but the results of this analysis suggest that researchers should avoid dual-mode analysis, in which data for both stable carbon and nitrogen isotopes are obtained from a single acidified sample, and should continue analyzing an additional non-acidified sample to obtain nitrogen isotope values.