Towards quantitative environmental reconstructions from ancient non-analogue microfossil assemblages: Ecological preferences of Paleocene - Eocene dinoflagellates

We explore a novel approach towards quantification of paleo-ecological signals from non-analogue microfossil assemblages by quantifying relations between assemblages and high-quality geochemical and sedimentological data from sedimentary archives. We test this approach using one group of microfossils, the organic cysts of dinoflagellates (dinocysts), which are widely used in shelf and open marine settings for biostratigraphy and reconstructions of past environments and environmental change. Similar to other microfossil groups, dinocysts can be used to reconstruct environments with relatively high confidence for recent time periods, as species affinities can be derived from observational and instrumental data. In the absence of such data, the ecology of extinct species is much less certain and as a result reconstructions in deep time are often limited to qualitative interpretations. To explore if quantitative empirical relations between extinct dinocysts and high-quality geochemical data can be established, we study a major event of climate change, the Paleocene-Eocene Thermal Maximum (PETM; similar to 56 Ma ago). The PETM is a particularly suitable period for this exercise as there is a multitude of large environmental perturbations associated with the transient global warming, such as deoxygenation, sea level rise and an accelerated hydrological cycle. The synthesized published dataset exhibits better spatial and temporal coverage compared to any other period in deep time. We extract empirical relations for the abundance of previously proposed paleoecological groups as a function of independent environmental proxies for example, sea surface temperature and terrestrial organic matter input. The results unequivocally illustrate that many dinocysts show relations to several of the reconstructed environmental variables. Notably, we show that one genus (Apectodinium) and an ecogroup (epicystal Goniodomidae) required sea surface temperatures in excess of 20 degrees C, and 25 degrees C, respectively, while one species (Florentinia reichartii) was only abundant between 30 and 35 degrees C. Other groups apparently favored either a limited (Spiniferites) or high (Senegalinium) terrestrial input to the study site, relating to salinity, nutrient levels or suspended sediment load (i.e. murkiness). Crucially, our work shows that the validation and quantification of ecological signals by independent environmental proxy data provides the opportunity to extract more quantitative information from a wide range of (non-analogue) microfossil assemblages. While this approach is not limited to any specific group of microfossils (or macrofossils), we stress that proper calibration datasets, high-quality sedimentological and geochemical proxy records, are needed and should ideally have decent geographical coverage and include one or more environmental perturbations. Using this approach such empirical relations can be established for a wide range of microfossil groups that have highly complementary geological records, which increases the value of existing data and ensures future application of microfossil-based paleoecology.