Improving the taxonomy of fossil pollen using convolutional neural networks and superresolution microscopy.

Taxonomic resolution is a major challenge in palynology, largely limiting the ecological and evolutionary interpretations possible with deep-time fossil pollen data. We present an approach for fossil pollen analysis that uses optical superresolution microscopy and machine learning to create a quantitative and higher throughput workflow for producing palynological identifications and hypotheses of biological affinity. We developed three convolutional neural network (CNN) classification models: maximum projection (MPM), multislice (MSM), and fused (FM). We trained the models on the pollen of 16 genera of the legume tribe Amherstieae, and then used these models to constrain the biological classifications of 48 fossil Striatopollis specimens from the Paleocene, Eocene, and Miocene of western Africa and northern South America. All models achieved average accuracies of 83 to 90% in the classification of the extant genera, and the majority of fossil identifications (86%) showed consensus among at least two of the three models. Our fossil identifications support the paleobiogeographic hypothesis that Amherstieae originated in Paleocene Africa and dispersed to South America during the Paleocene-Eocene Thermal Maximum (56 Ma). They also raise the possibility that at least three Amherstieae genera (Crudia, Berlinia, and Anthonotha) may have diverged earlier in the Cenozoic than predicted by molecular phylogenies.

Summer-Wet Hydrologic Cycle during the Middle Miocene of the United States: New Evidence from Fossil Fungi.

Hydrologic reconstructions from North America are largely unknown for the Middle Miocene. Examination of fungal palynomorph assemblages coupled with traditional plant-based palynology permits delineation of local, as opposed to regional, climate signals and provides a baseline for study of ancient fungas. Here, the Fungi in a Warmer World project presents paleoecology and paleoclimatology of 351 fungal morphotypes from 3 sites in the United States: the Clarkia Konservat-Lagerstätte site (Idaho), the Alum Bluff site (Florida), and the Bouie River site (Mississippi). Of these, 83 fungi are identified as extant taxa and 41 are newly reported from the Miocene. Combining new plant-based paleoclimatic reconstructions with funga-based paleoclimate reconstructions, we demonstrate cooling and hydrologic changes from the Miocene climate optimum to the Serravallian. In the southeastern United States, this is comparable to that reconstructed with pollen and paleobotany alone. In the northwestern United States, cooling is greater than indicated by other reconstructions and hydrology shifts seasonally, from no dry season to a dry summer season. Our results demonstrate the utility of fossil fungi as paleoecologic and paleoclimatic proxies and that warmer than modern geological time intervals do not match the "wet gets wetter, dry gets drier" paradigm. Instead, both plants and fungi show an invigorated hydrological cycle across mid-latitude North America.