Faecal stanols offer an exciting opportunity to determine population change in the past but the controls of their concentrations and ratios within lake sediments are not well understood. We present the variability in stanol concentrations and ratios from lakes across environmental gradients, both between and in lakes across climatic and land-use gradients in Mexico, Guatemala, and Belize in order to determine the factors controlling preservation and degradation in lacustrine sediments. We also measured physical properties of lakes such as dissolved oxygen, pH, and water column temperature and estimated the approximate human population in each catchment, with the goal of producing a semi-quantitative calibration of human population to coprostanol+epi-coprostanol as a ratio to cholestanol, a more widely produced bacterial stanol. In particular we explore the hypothesis that a dominant control on concentrations and ratios is proximity to a human settlement. We evaluate this hypothesis in two lakes (L. Peten-Itza and L. Izabal) where we collected samples at varying distances from major population centres. This will have implications for the targeting of lake cores in studies where determining population change is the goal. In addition to this work we will share three intriguing preliminary palaeo-records of stanol concentrations from Guatemala (Laguna Itzan, Laguna Peten-Itza, and Lago Izabal). These records imply highly dynamic millennial scale changes in human populations, and we apply the modern sediment data to better constrain the interpretation of these data. Our work shows that faecal stanols have a strong potential as proxies for changes in human population and land-use change through time, and can be used to complement archaeological datasets to link human populations with palaeoenvironmental change.

Continued global warming is expected to result in drying of Central America, with projections suggesting a decrease in precipitation. Poor hindcasting of precipitation, however, due in part to spatial and temporal limitations in instrumental data, subjects these projections to considerable uncertainty. Paleoclimate proxy data are therefore critical for understanding regional climate responses during times of global climate reorganization. Here we present two lake-sediment based records of precipitation variability in Guatemala along with a synthesis of Central American hydroclimate records spanning the last millennium (800-2000 CE). The synthesis reveals that regional climate responses have been strikingly heterogeneous, even over relatively short distances. Our analysis further suggests that shifts in the mean position of the Intertropical Convergence Zone, which have been invoked by numerous studies to explain variability in Central American and circum-Caribbean proxy records, cannot alone explain the observed pattern of hydroclimate variability. Instead, interactions between several ocean-atmosphere processes and their disparate influences across variable topography have resulted in complex precipitation responses. These complexities highlight the difficulty of reconstructing past precipitation changes across Central America and point to the need for additional paleo-record development and analysis before the relationships between external forcing and hydroclimate change can be robustly determined. Such efforts should help anchor model-based predictions of future responses to continued global warming.