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.

Early Last Interglacial ocean warming drove substantial ice mass loss from Antarctica.

The future response of the Antarctic ice sheet to rising temperatures remains highly uncertain. A useful period for assessing the sensitivity of Antarctica to warming is the Last Interglacial (LIG) (129 to 116 ky), which experienced warmer polar temperatures and higher global mean sea level (GMSL) (+6 to 9 m) relative to present day. LIG sea level cannot be fully explained by Greenland Ice Sheet melt (∼2 m), ocean thermal expansion, and melting mountain glaciers (∼1 m), suggesting substantial Antarctic mass loss was initiated by warming of Southern Ocean waters, resulting from a weakening Atlantic meridional overturning circulation in response to North Atlantic surface freshening. Here, we report a blue-ice record of ice sheet and environmental change from the Weddell Sea Embayment at the periphery of the marine-based West Antarctic Ice Sheet (WAIS), which is underlain by major methane hydrate reserves. Constrained by a widespread volcanic horizon and supported by ancient microbial DNA analyses, we provide evidence for substantial mass loss across the Weddell Sea Embayment during the LIG, most likely driven by ocean warming and associated with destabilization of subglacial hydrates. Ice sheet modeling supports this interpretation and suggests that millennial-scale warming of the Southern Ocean could have triggered a multimeter rise in global sea levels. Our data indicate that Antarctica is highly vulnerable to projected increases in ocean temperatures and may drive ice-climate feedbacks that further amplify warming.

In-situ sequencing reveals the effect of storage on lacustrine sediment microbiome demographics and functionality.

The sediment microbiome is a demographically diverse and functionally active biosphere. Ensuring that data acquired from sediment is truly representative of the microbiome is critical to achieving robust analyses. Sample storage and the processing and timing of nucleic acid purification after environmental sample extraction may fundamentally affect the detectable microbial community and thereby significantly alter resultant data. Direct sequencing of environmental samples is increasingly commonplace due to the advent of the portable Oxford Nanopore MinION sequencing device. Here we demonstrate that storing sediment subsamples at - 20 °C or storing the cores at 4 °C for 10 weeks prior to analysis, has a significant effect on the sediment microbiome analysed using sedimentary DNA (sedDNA), especially for Alpha-, Beta- and Deltaproteobacteria species. Furthermore, these significant differences are observed regardless of sediment type. We show that the taxa which are predominantly affected by storage are Proteobacteria, and therefore recommend on-site purifications are performed to ensure an accurate representation of these taxa are observed in the microbiome. Comparisons of sedimentary RNA (sedRNA) analyses, revealed substantial differences between samples purified and sequenced immediately on-site, samples that were frozen before transportation, and cores that were stored at 4 °C prior to analysis. Our data therefore suggest that a more accurate representation of the sediment microbiome demography and functionality may be achieved by environmental sequencing as rapidly as possible to minimise confounding effects of storage.

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.

A 700-year paleoecological record of boreal ecosystem responses to climatic variation from Alaska.

Recent observations and model simulations have highlighted the sensitivity of the forest-tundra ecotone to climatic forcing. In contrast, paleoecological studies have not provided evidence of tree-line fluctuations in response to Holocene climatic changes in Alaska, suggesting that the forest-tundra boundary in certain areas may be relatively stable at multicentennial to millennial time scales. We conducted a multiproxy study of sediment cores from an Alaskan lake near the altitudinal limits of key boreal-forest species. Paleoecological data were compared with independent climatic reconstructions to assess ecosystem responses of the forest tundra boundary to Little Ice Age (LIA) climatic fluctuations. Pollen, diatom, charcoal, macrofossil, and magnetic analyses provide the first continuous record of vegetation fire-climate interactions at decadal to centennial time scales during the past 700 years from southern Alaska. Boreal-forest diebacks characterized by declines of Picea mariana, P. glauca, and tree Betula occurred during the LIA (AD 1500-1800), whereas shrubs (Alnus viridis, Betula glandulosa/nana) and herbaceous taxa (Epilobium, Aconitum) expanded. Marked increases in charcoal abundance and changes in magnetic properties suggest increases in fire importance and soil erosion during the same period. In addition, the conspicuous reduction or disappearance of certain aquatic (e.g., Isoetes, Nuphar, Pediastrum) and wetland (Sphagnum) plants and major shifts in diatom assemblages suggest pronounced lake-level fluctuations and rapid ecosystem reorganization in response to LIA climatic deterioration. Our results imply that temperature shifts of 1-2 degrees C, when accompanied by major changes in moisture balance, can greatly alter high-altitudinal terrestrial, wetland, and aquatic ecosystems, including conversion between boreal-forest tree line and tundra. The climatic and ecosystem variations in our study area appear to be coherent with changes in solar irradiance, suggesting that changes in solar activity contributed to the environmental instability of the past 700 years.

Global Peak in Atmospheric Radiocarbon Provides a Potential Definition for the Onset of the Anthropocene Epoch in 1965.

Anthropogenic activity is now recognised as having profoundly and permanently altered the Earth system, suggesting we have entered a human-dominated geological epoch, the 'Anthropocene'. To formally define the onset of the Anthropocene, a synchronous global signature within geological-forming materials is required. Here we report a series of precisely-dated tree-ring records from Campbell Island (Southern Ocean) that capture peak atmospheric radiocarbon (14C) resulting from Northern Hemisphere-dominated thermonuclear bomb tests during the 1950s and 1960s. The only alien tree on the island, a Sitka spruce (Picea sitchensis), allows us to seasonally-resolve Southern Hemisphere atmospheric 14C, demonstrating the 'bomb peak' in this remote and pristine location occurred in the last-quarter of 1965 (October-December), coincident with the broader changes associated with the post-World War II 'Great Acceleration' in industrial capacity and consumption. Our findings provide a precisely-resolved potential Global Stratotype Section and Point (GSSP) or 'golden spike', marking the onset of the Anthropocene Epoch.

Rapid global ocean-atmosphere response to Southern Ocean freshening during the last glacial.

Contrasting Greenland and Antarctic temperatures during the last glacial period (115,000 to 11,650 years ago) are thought to have been driven by imbalances in the rates of formation of North Atlantic and Antarctic Deep Water (the 'bipolar seesaw'). Here we exploit a bidecadally resolved 14C data set obtained from New Zealand kauri (Agathis australis) to undertake high-precision alignment of key climate data sets spanning iceberg-rafted debris event Heinrich 3 and Greenland Interstadial (GI) 5.1 in the North Atlantic (~30,400 to 28,400 years ago). We observe no divergence between the kauri and Atlantic marine sediment 14C data sets, implying limited changes in deep water formation. However, a Southern Ocean (Atlantic-sector) iceberg rafted debris event appears to have occurred synchronously with GI-5.1 warming and decreased precipitation over the western equatorial Pacific and Atlantic. An ensemble of transient meltwater simulations shows that Antarctic-sourced salinity anomalies can generate climate changes that are propagated globally via an atmospheric Rossby wave train.A challenge for testing mechanisms of past climate change is the precise correlation of palaeoclimate records. Here, through climate modelling and the alignment of terrestrial, ice and marine 14C and 10Be records, the authors show that Southern Ocean freshwater hosing can trigger global change.

Use of Radioimmunoassay to Determine the Nature, Quantity and Source of Allergenic Contamination of Sunflower Butter.

A girl known to be allergic to peanuts experienced a mild allergic reaction after eating sunflower butter, a peanut butter facsimile prepared from sunflower nuts. After learning that the same manufacturing plant also produced peanut butter, we examined several sunflower butter preparations for peanut butter contamination by a solid-phase radioimmunoassay which used naturally-occurring immunoglobulin E (IgE) antibodies from known peanut-sensitive individuals. Peanut butter contamination, ranging from 0.3 to 3.3%, was found in six of eight sunflower butter samples, including both creamy and chunky varieties, and including lots prepared with either peanut oil or palm oil. We concluded that inadequate cleaning of food processing machines following peanut butter production permitted the contamination by peanut butter of subsequent jars of sunflower butter in amounts which could pose risks to peanut-sensitive individuals.