MULTI-PROXY LAKE SEDIMENT RECORDS AND WATERSHED MODEL DEVELOPMENT: AN EXAMPLE FROM THE SOUTHERN MAYA LOWLANDS OF PETEN, GUATEMALA

Variations in the oxygen isotopic composition of biogenic carbonate from lake sediment cores have been used to infer past environmental changes in the southern Maya lowlands of the Yucatan Peninsula. At Lake Salpeten, Peten, Guatemala, oxygen isotopic records indicate that pronounced changes in watershed hydrologic balance were caused by human-induced deforestation and/or natural climate change. Strictly climatic interpretation of the oxygen isotopic records suggests higher precipitation during the period of Maya settlement expansion between 2400 and 1800 years ago. Alternatively, this period of minimum oxygen isotopic values may have been a consequence of increased surface runoff and groundwater inflow to the lake related to watershed deforestation by the Maya. When Maya population declined beginning ~1100 years ago, oxygen isotopic values increased as a consequence of reduced hydrologic input to the lake caused by decreased precipitation or forest recovery. Paleolimnological results from Lake Salpeten suggest that human-induced changes in watershed vegetation cover can alter lake hydrologic budgets, thereby confounding paleoclimatic inferences based on the oxygen isotopic composition of biogenic carbonate.

The ambiguity of oxygen isotopic records from the southern Maya lowlands highlights the need for complementary sediment proxies and catchment models that can be used to test climatic inferences. In this study, a hydrologic and isotopic mass balance model of Lake Salpeten provides the first quantitative assessment of late Holocene environmental change in lowland Peten, Guatemala. Precipitation inputs nearly 30 percent above modern mean values were required to produce the low oxygen isotopic values and inferred lake level increase observed in Lake Salpeten during the late Holocene. In contrasting model experiments, the observed shift in isotopic composition was simulated by a 75 percent decrease in watershed forest cover and consequent reductions in evapotranspiration and increases in surface runoff. The modeled reduction in vegetation cover is consistent with palynological evidence for forest loss from Peten watersheds.