Deductive automated pollen classification in environmental samples via exploratory deep learning and imaging flow cytometry.

Pollen and tracheophyte spores are ubiquitous environmental indicators at local and global scales. Palynology is typically performed manually by microscopic analysis; a specialised and time-consuming task limited in taxonomical precision and sampling frequency, therefore restricting data quality used to inform climate change and pollen forecasting models. We build on the growing work using AI (artificial intelligence) for automated pollen classification to design a flexible network that can deal with the uncertainty of broad-scale environmental applications. We combined imaging flow cytometry with Guided Deep Learning to identify and accurately categorise pollen in environmental samples; here, pollen grains captured within c. 5500 Cal yr BP old lake sediments. Our network discriminates not only pollen included in training libraries to the species level but, depending on the sample, can classify previously unseen pollen to the likely phylogenetic order, family and even genus. Our approach offers valuable insights into the development of a widely transferable, rapid and accurate exploratory tool for pollen classification in 'real-world' environmental samples with improved accuracy over pure deep learning techniques. This work has the potential to revolutionise many aspects of palynology, allowing a more detailed spatial and temporal understanding of pollen in the environment with improved taxonomical resolution.

Nonlinear landscape and cultural response to sea-level rise.

Rising sea levels have been associated with human migration and behavioral shifts throughout prehistory, often with an emphasis on landscape submergence and consequent societal collapse. However, the assumption that future sea-level rise will drive similar adaptive responses is overly simplistic. While the change from land to sea represents a dramatic and permanent shift for preexisting human populations, the process of change is driven by a complex set of physical and cultural processes with long transitional phases of landscape and socioeconomic change. Here, we use reconstructions of prehistoric sea-level rise, paleogeographies, terrestrial landscape change, and human population dynamics to show how the gradual inundation of an island archipelago resulted in decidedly nonlinear landscape and cultural responses to rising sea levels. Interpretation of past and future responses to sea-level change requires a better understanding of local physical and societal contexts to assess plausible human response patterns in the future.

Palaeogenomics of the Hydrocarbon Producing Microalga Botryococcus braunii.

Botryococcus braunii is a colonial microalga that appears early in the fossil record and is a sensitive proxy of environmental and hydroclimatic conditions. Palaeozoic Botryococcus fossils which contribute up to 90% of oil shales and approximately 1% of crude oil, co-localise with diagnostic geolipids from the degradation of source-signature hydrocarbons. However more recent Holocene sediments demonstrate no such association. Consequently, Botryococcus are identified in younger sediments by morphology alone, where potential misclassifications could lead to inaccurate paleoenvironmental reconstructions. Here we show that a combination of flow cytometry and ancient DNA (aDNA) sequencing can unambiguously identify Botryococcus microfossils in Holocene sediments with hitherto unparalleled accuracy and rapidity. The application of aDNA sequencing to microfossils offers a far-reaching opportunity for understanding environmental change in the recent geological record. When allied with other high-resolution palaeoenvironmental information such as aDNA sequencing of humans and megafauna, aDNA from microfossils may allow a deeper and more precise understanding of past environments, ecologies and migrations.