Advancing understanding of past environmental dynamics: Reproducible analytical workflows with large-scale fossil pollen compilations

The advent of large, open-access databases such as Neotoma has revolutionized the field of paleoecology, providing unprecedented opportunities to conduct large-scale analyses of past environmental change. These databases allow for the integration of thousands of fossil pollen records, enabling a more comprehensive understanding of spatial and temporal variability across ecosystems. By combining these data with advanced numerical methods and/or other proxies, we can refine our understanding of how past climatic changes influenced biodiversity. This integrated approach holds the potential to push paleoecology into exciting new directions, with implications for forecasting future climate and biodiversity changes.

Our work explores innovative uses of fossil pollen datasets, particularly large-scale compilations of Late Quaternary records, to investigate long-term vegetation dynamics and climate change. We apply novel spatio-temporal techniques to gain new insights into biodiversity change. This approach has enabled us to uncover global patterns of vegetation change and deepen our understanding of climate-vegetation interactions (Mottl et al. 2021a). By quantifying rates of ecological change (Mottl et al. 2021b), we demonstrated that vegetation rates of change began accelerating globally between three to four thousand years ago, and that recent rates of change now are even higher than those associated with the end of the last ice age. Our follow-up comparative study comparing our results with other proxies across the Amazon, provided a much-needed interdisciplinary framework to examine past environmental conditions in this region (Albert et al. 2023), showing that rates of change in both geological and paleoecological records are exceptionally high over recent geological times.

When handling such large, heterogeneous datasets (e.g., fossil pollen compilations) for advancing paleoecological research, reproducibility is essential. The integration of open-access databases like Neotoma into research workflows must be accompanied by rigorous, transparent procedures for data sourcing, cleaning, filtering, and analysis. The establishment of reproducible workflows ensures that the entire process, from dataset compilation to final analysis, is transparent, reliable, and accessible for future researchers. In all our work, we emphasize the importance of standardized data preparation and validation steps, using our newly developed FOSSILPOL workflow (Flantua et al. 2023; FOSSILPOL website). This not only facilitates the synthesis of complex datasets but also fosters interdisciplinary collaboration. By ensuring that the analysis of paleoecological data is fully reproducible, we can reduce biases, improve the quality of results, and build a robust foundation for further interdisciplinary climate and biodiversity studies.

REFERENCES

Albert, J. S. et al. (2023). Human impacts outpace natural processes in the Amazon. Science, 379(6630), eabo5003. 

Flantua et al. (2023). A guide to the processing and standardization of global palaeoecological data for large‐scale syntheses using fossil pollen. Global Ecology and Biogeography, 32(8), 1377–1394.

Fossilpol website: https://hope-uib-bio.github.io/FOSSILPOL-website/index.html

Mottl et al. (2021a). Global acceleration in rates of vegetation change over the past 18,000 years. Science, 372(6544), 860–864.

Mottl et al (2021b). Rate-of-change analysis in paleoecology revisited: A new approach. Review of Palaeobotany and Palynology, 293, 104483.