Spatio-temporal changes in forest composition inferred from fossil pollen records in the Upper Midwestern USA

Background/Question/Methods

Understanding compositional changes in vegetation over the common era provides insight about ecosystem dynamics in response to changing environments. Past vegetation reconstructions rely predominantly on fossil pollen data from sedimentary lake cores, which acts as a proxy record for the surrounding vegetation. Stratigraphic changes in these pollen records allow us to infer changes in composition and species distributions. Pollen records collected from a network of sites allow us to make inference about the spatio-temporal changes in vegetation over thousands of years. To reconstruct vegetation composition, we build a Bayesian model for the Upper Midwestern USA that links vegetation composition to fossil pollen data via a dispersal model. We estimate the relationship between vegetation and pollen for the settlement era using United States Public Land Survey data and a network of pollen records. Using parameter estimates from the settlement era, we use the fossil pollen proxy records to estimate species distributions and relative abundances over the last 2000 years.

Results/Conclusions

Results from the settlement era calibration model show that we can estimate pollen assemblages at each lake using the estimated parameter values for differential production and dispersal. By accounting for differential production and dispersal of pollen, we improve upon the assumption that proximal vegetation abundance is representative of deposited pollen. This improvement is most noticeable for taxa that are low-producers or are under-dispersed, such as maple, as well as taxa that are over-produce or over-disperse, such as pine. These process parameter estimates allowed us to use the prediction model to estimate spatial maps of relative abundance. Preliminary results indicate that forests in the Upper Midwest have not been at steady-state over the last 2000 years. For example, spatial maps indicate changes in the relative abundance of oak in southern Minnesota, as well as pockets of increasing or decreasing relative abundance for many other key taxa including pine, birch, maple, and spruce. Challenging the assumption of forest stationarity prior to settlement has important consequences for ecosystem models, whose predictions are often based on this steady-state assumption. Our spatio-temporal composition and abundance estimates will be used to improve the forecasting capabilities of such models.