CuticleTrace: A toolkit for capturing cell outlines from leaf cuticle with implications for paleoecology and paleoclimatology.

Premise: Leaf epidermal cell morphology is closely tied to the evolutionary history of plants and their growth environments and is therefore of interest to many plant biologists. However, cell measurement can be time consuming and restrictive with current methods. CuticleTrace is a suite of Fiji and R-based functions that streamlines and automates the segmentation and measurement of epidermal pavement cells across a wide range of cell morphologies and image qualities. Methods and Results: We evaluated CuticleTrace-generated measurements against those from alternate automated methods and expert and undergraduate hand tracings across a taxonomically diverse 50-image data set of variable image qualities. We observed ~93% statistical agreement between CuticleTrace and expert hand-traced measurements, outperforming alternate methods. Conclusions: CuticleTrace is a broadly applicable, modular, and customizable tool that integrates data visualization and cell shape measurement with image segmentation, lowering the barrier to high-throughput studies of epidermal morphology by vastly decreasing the labor investment required to generate high-quality cell shape data sets.

Deep learning approaches to the phylogenetic placement of extinct pollen morphotypes.

The phylogenetic interpretation of pollen morphology is limited by our inability to recognize the evolutionary history embedded in pollen features. Deep learning offers tools for connecting morphology to phylogeny. Using neural networks, we developed an explicitly phylogenetic toolkit for analyzing the overall shape, internal structure, and texture of a pollen grain. Our analysis pipeline determines whether testing specimens are from known species based on uncertainty estimates. Features from specimens with uncertain taxonomy are passed to a multilayer perceptron network trained to transform these features into predicted phylogenetic distances from known taxa. We used these predicted distances to place specimens in a phylogeny using Bayesian inference. We trained and evaluated our models using optical superresolution micrographs of 30 extant Podocarpus species. We then used trained models to place nine fossil Podocarpidites specimens within the phylogeny. In doing so, we demonstrate that the phylogenetic history encoded in pollen morphology can be recognized by neural networks and that deep-learned features can be used in phylogenetic placement. Our approach makes extinction and speciation events that would otherwise be masked by the limited taxonomic resolution of the fossil pollen record visible to palynological analysis.

Review of the cellulose acetate peel method and the physical and digital curation of coal balls.

Coal balls, in which fossil plants are preserved in permineralized peat deposits, have widely been described from coal deposits representing the tropical forest of the Carboniferous. Coal ball preparation techniques have evolved over the past century, with the cellulose acetate peel method becoming the standard in the 1950s. While coal ball research is not as active as it has been in the past, large collections of coal balls and their respective peels still form a large part of many museum and university collections. This contribution aims to review coal ball preparation methods, curation, and the digital archiving of peels to create a cohesive guide for researchers working with coal balls and other petrified plant material. The physical and digital curation of cellulose acetate peels and other types of coal ball specimens is critical for long-term preservation and accessibility. Physical curation involves embedding coal balls in media to slow pyrite deterioration. Digital curation creates high-resolution scans of peels, which can be shared and accessed online. Cellulose acetate peels and their digital curation are a valuable and accessible technique for the analysis of coal balls, and physical and digital curation ensures long-term preservation.

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.

Cuticle and subsurface ornamentation of intact plant leaf epidermis under confocal and superresolution microscopy.

Plant cuticle micromorphology is an invaluable tool in modern ecology and paleoecology. It has expanded our knowledge of systematic relationships among diverse plant groups and can be used to identify fossil plants. Furthermore, fossil plant leaf micromorphology is used for reconstructing past environments, most notably for estimating atmospheric CO2 concentration. Here we outline a new protocol for imaging plant cuticle for archival and paleoecological applications. Traditionally, both modern reference and fossil samples undergo maceration with subsequent imaging via environmental SEM, widefield fluorescence, or light microscopy. In this paper, we demonstrate the capabilities of alternative preparation and imaging methods using confocal and superresolution microscopy with intact leaf samples. This method produces detailed three-dimensional images of surficial and subsurface structures of the intact leaf. Multiple layers are captured simultaneously, which previously required independent maceration and microtome steps. We compared clearing agents (chloral hydrate, KOH, and Visikol); mounting media (Eukitt and Hoyer's); fluorescent stains (periodic acid Schiff, propidium iodide); and confocal vs. superresolution microscopes. We conclude that Eukitt is the best medium for long-term preservation and imaging. Because of nontoxicity and ease of procurement, Visikol made for the best clearing agent. Staining improves contrast and under most circumstances PAS provided the clearest images. Supperresolution produced higher clarity images than traditional confocal, but the information gained was minimal. This new protocol provides the botanical and paleobotanical community an alternative to traditional techniques. Our proposed workflow has the net benefit of being more efficient than traditional methods, which only capture the surface of the plant epidermis. Microsc. Res. Tech. 81:129-140, 2018. © 2016 Wiley Periodicals, Inc.

Comparative performance of airyscan and structured illumination superresolution microscopy in the study of the surface texture and 3D shape of pollen.

The visualization of taxonomically diagnostic features of individual pollen grains can be a challenge for many ecologically and phylogenetically important pollen types. The resolution of traditional optical microscopy is limited by the diffraction of light (250 nm), while high resolution tools such as electron microscopy are limited by laborious preparation and imaging workflows. Airyscan confocal superresolution and structured illumination superresolution (SR-SIM) microscopy are powerful new tools for the study of nanoscale pollen morphology and three-dimensional structure that can overcome these basic limitations. This study demonstrates their utility in capturing morphological details below the diffraction limit of light. Using three distinct pollen morphotypes (Croton hirtus, Dactylis glomerata, and Helianthus sp.) and contrast-enhancing fluorescent staining, we were able to assess the effectiveness of the Airyscan and SR-SIM. We further demonstrate that these new superresolution methods can be easily applied to the study of fossil pollen material.

A morphometric analysis of vegetation patterns in dryland ecosystems.

Vegetation in dryland ecosystems often forms remarkable spatial patterns. These range from regular bands of vegetation alternating with bare ground, to vegetated spots and labyrinths, to regular gaps of bare ground within an otherwise continuous expanse of vegetation. It has been suggested that spotted vegetation patterns could indicate that collapse into a bare ground state is imminent, and the morphology of spatial vegetation patterns, therefore, represents a potentially valuable source of information on the proximity of regime shifts in dryland ecosystems. In this paper, we have developed quantitative methods to characterize the morphology of spatial patterns in dryland vegetation. Our approach is based on algorithmic techniques that have been used to classify pollen grains on the basis of textural patterning, and involves constructing feature vectors to quantify the shapes formed by vegetation patterns. We have analysed images of patterned vegetation produced by a computational model and a small set of satellite images from South Kordofan (South Sudan), which illustrates that our methods are applicable to both simulated and real-world data. Our approach provides a means of quantifying patterns that are frequently described using qualitative terminology, and could be used to classify vegetation patterns in large-scale satellite surveys of dryland ecosystems.

Visual Recognition Software for Binary Classification and Its Application to Spruce Pollen Identification.

Discriminating between black and white spruce (Picea mariana and Picea glauca) is a difficult palynological classification problem that, if solved, would provide valuable data for paleoclimate reconstructions. We developed an open-source visual recognition software (ARLO, Automated Recognition with Layered Optimization) capable of differentiating between these two species at an accuracy on par with human experts. The system applies pattern recognition and machine learning to the analysis of pollen images and discovers general-purpose image features, defined by simple features of lines and grids of pixels taken at different dimensions, size, spacing, and resolution. It adapts to a given problem by searching for the most effective combination of both feature representation and learning strategy. This results in a powerful and flexible framework for image classification. We worked with images acquired using an automated slide scanner. We first applied a hash-based "pollen spotting" model to segment pollen grains from the slide background. We next tested ARLO's ability to reconstruct black to white spruce pollen ratios using artificially constructed slides of known ratios. We then developed a more scalable hash-based method of image analysis that was able to distinguish between the pollen of black and white spruce with an estimated accuracy of 83.61%, comparable to human expert performance. Our results demonstrate the capability of machine learning systems to automate challenging taxonomic classifications in pollen analysis, and our success with simple image representations suggests that our approach is generalizable to many other object recognition problems.

A neotropical Miocene pollen database employing image-based search and semantic modeling.

PREMISE OF THE STUDY: Digital microscopic pollen images are being generated with increasing speed and volume, producing opportunities to develop new computational methods that increase the consistency and efficiency of pollen analysis and provide the palynological community a computational framework for information sharing and knowledge transfer. • METHODS: Mathematical methods were used to assign trait semantics (abstract morphological representations) of the images of neotropical Miocene pollen and spores. Advanced database-indexing structures were built to compare and retrieve similar images based on their visual content. A Web-based system was developed to provide novel tools for automatic trait semantic annotation and image retrieval by trait semantics and visual content. • RESULTS: Mathematical models that map visual features to trait semantics can be used to annotate images with morphology semantics and to search image databases with improved reliability and productivity. Images can also be searched by visual content, providing users with customized emphases on traits such as color, shape, and texture. • DISCUSSION: Content- and semantic-based image searches provide a powerful computational platform for pollen and spore identification. The infrastructure outlined provides a framework for building a community-wide palynological resource, streamlining the process of manual identification, analysis, and species discovery.

Accuracy and consistency of grass pollen identification by human analysts using electron micrographs of surface ornamentation.

PREMISE OF THE STUDY: Humans frequently identify pollen grains at a taxonomic rank above species. Grass pollen is a classic case of this situation, which has led to the development of computational methods for identifying grass pollen species. This paper aims to provide context for these computational methods by quantifying the accuracy and consistency of human identification. • METHODS: We measured the ability of nine human analysts to identify 12 species of grass pollen using scanning electron microscopy images. These are the same images that were used in computational identifications. We have measured the coverage, accuracy, and consistency of each analyst, and investigated their ability to recognize duplicate images. • RESULTS: Coverage ranged from 87.5% to 100%. Mean identification accuracy ranged from 46.67% to 87.5%. The identification consistency of each analyst ranged from 32.5% to 87.5%, and each of the nine analysts produced considerably different identification schemes. The proportion of duplicate image pairs that were missed ranged from 6.25% to 58.33%. • DISCUSSION: The identification errors made by each analyst, which result in a decline in accuracy and consistency, are likely related to psychological factors such as the limited capacity of human memory, fatigue and boredom, recency effects, and positivity bias.

Classification of grass pollen through the quantitative analysis of surface ornamentation and texture.

Taxonomic identification of pollen and spores uses inherently qualitative descriptions of morphology. Consequently, identifications are restricted to categories that can be reliably classified by multiple analysts, resulting in the coarse taxonomic resolution of the pollen and spore record. Grass pollen represents an archetypal example; it is not routinely identified below family level. To address this issue, we developed quantitative morphometric methods to characterize surface ornamentation and classify grass pollen grains. This produces a means of quantifying morphological features that are traditionally described qualitatively. We used scanning electron microscopy to image 240 specimens of pollen from 12 species within the grass family (Poaceae). We classified these species by developing algorithmic features that quantify the size and density of sculptural elements on the pollen surface, and measure the complexity of the ornamentation they form. These features yielded a classification accuracy of 77.5%. In comparison, a texture descriptor based on modelling the statistical distribution of brightness values in image patches yielded a classification accuracy of 85.8%, and seven human subjects achieved accuracies between 68.33 and 81.67%. The algorithmic features we developed directly relate to biologically meaningful features of grass pollen morphology, and could facilitate direct interpretation of unsupervised classification results from fossil material.

Novel application of confocal laser scanning microscopy and 3D volume rendering toward improving the resolution of the fossil record of charcoal.

Variations in the abundance of fossil charcoals between rocks and sediments are assumed to reflect changes in fire activity in Earth's past. These variations in fire activity are often considered to be in response to environmental, ecological or climatic changes. The role that fire plays in feedbacks to such changes is becoming increasingly important to understand and highlights the need to create robust estimates of variations in fossil charcoal abundance. The majority of charcoal based fire reconstructions quantify the abundance of charcoal particles and do not consider the changes in the morphology of the individual particles that may have occurred due to fragmentation as part of their transport history. We have developed a novel application of confocal laser scanning microscopy coupled to image processing that enables the 3-dimensional reconstruction of individual charcoal particles. This method is able to measure the volume of both microfossil and mesofossil charcoal particles and allows the abundance of charcoal in a sample to be expressed as total volume of charcoal. The method further measures particle surface area and shape allowing both relationships between different size and shape metrics to be analysed and full consideration of variations in particle size and size sorting between different samples to be studied. We believe application of this new imaging approach could allow significant improvement in our ability to estimate variations in past fire activity using fossil charcoals.

Variability within the 10-year pollen rain of a seasonal neotropical forest and its implications for paleoenvironmental and phenological research.

Tropical paleoecologists use a combination of mud-water interface and modern pollen rain samples (local samples of airborne pollen) to interpret compositional changes within fossil pollen records. Taxonomic similarities between the composition of modern assemblages and fossil samples are the basis of reconstructing paleoclimates and paleoenvironments. Surface sediment samples reflect a time-averaged accumulation of pollen spanning several years or more. Due to experimental constraints, modern pollen rain samples are generally collected over shorter timeframes (1-3 years) and are therefore less likely to capture the full range of natural variability in pollen rain composition and abundance. This potentially biases paleoenvironmental interpretations based on modern pollen rain transfer functions. To determine the degree to which short-term environmental change affects the composition of the aerial pollen flux of Neotropical forests, we sampled ten years of the seasonal pollen rain from Barro Colorado Island, Panama and compared it to climatic and environmental data over the same ten-year span. We establish that the pollen rain effectively captured the strong seasonality and stratification of pollen flow within the forest canopy and that individual taxa had variable sensitivity to seasonal and annual changes in environmental conditions, manifested as changes in pollen productivity. We conclude that modern pollen rain samples capture the reproductive response of moist tropical plants to short-term environmental change, but that consequently, pollen rain-based calibrations need to include longer sampling periods (≥7 years) to reflect the full range of natural variability in the pollen output of a forest and simulate the time-averaging present in sediment samples. Our results also demonstrate that over the long-term, pollen traps placed in the forest understory are representative samples of the pollen output of both canopy and understory vegetation. Aerial pollen traps, therefore, also represent an underutilized means of monitoring the pollen productivity and reproductive behavior of moist tropical forests.

Tracking taphonomic regimes using chemical and mechanical damage of pollen and spores: an example from the Triassic-Jurassic mass extinction.

The interpretation of biotic changes in the geological past relies on the assumption that samples from different time intervals represent an equivalent suite of natural sampling conditions. As a result, detailed investigations of taphonomic regimes during intervals of major biotic upheaval, such as mass extinctions, are crucial. In this paper, we have used variations in the frequency of chemical and mechanical sporomorph (pollen and spore) damage as a guide to taphonomic regimes across the Triassic-Jurassic mass extinction (Tr-J; ∼201.3 Ma) at a boundary section at Astartekløft, East Greenland. We find that the frequency of sporomorph damage is extremely variable in samples from this locality. This likely reflects a combination of taxon-specific susceptibility to damage and the mixing of sporomorphs from a mosaic of environments and taphonomic regimes. The stratigraphic interval containing evidence of plant extinction and compositional change in the source vegetation at Astartekløft is not marked by a consistent rise or fall in the frequency of sporomorph damage. This indicates that natural taphonomic regimes did not shift radically during this critical interval. We find no evidence of a consistent relationship between the taxonomic richness of sporomorph assemblages and the frequency of damage among sporomorphs at Astartekløft. This indicates that previously reported patterns of sporomorph richness across the Tr-J at this locality are likely to be robust. Taken together, our results suggest that the patterns of vegetation change at Astartekløft represent a real biological response to environmental change at the Tr-J.

Classifying black and white spruce pollen using layered machine learning.

Pollen is among the most ubiquitous of terrestrial fossils, preserving an extended record of vegetation change. However, this temporal continuity comes with a taxonomic tradeoff. Analytical methods that improve the taxonomic precision of pollen identifications would expand the research questions that could be addressed by pollen, in fields such as paleoecology, paleoclimatology, biostratigraphy, melissopalynology, and forensics. We developed a supervised, layered, instance-based machine-learning classification system that uses leave-one-out bias optimization and discriminates among small variations in pollen shape, size, and texture. We tested our system on black and white spruce, two paleoclimatically significant taxa in the North American Quaternary. We achieved > 93% grain-to-grain classification accuracies in a series of experiments with both fossil and reference material. More significantly, when applied to Quaternary samples, the learning system was able to replicate the count proportions of a human expert (R(2) = 0.78, P = 0.007), with one key difference - the machine achieved these ratios by including larger numbers of grains with low-confidence identifications. Our results demonstrate the capability of machine-learning systems to solve the most challenging palynological classification problem, the discrimination of congeneric species, extending the capabilities of the pollen analyst and improving the taxonomic resolution of the palynological record.

Capturing the surface texture and shape of pollen: a comparison of microscopy techniques.

Research on the comparative morphology of pollen grains depends crucially on the application of appropriate microscopy techniques. Information on the performance of microscopy techniques can be used to inform that choice. We compared the ability of several microscopy techniques to provide information on the shape and surface texture of three pollen types with differing morphologies. These techniques are: widefield, apotome, confocal and two-photon microscopy (reflected light techniques), and brightfield and differential interference contrast microscopy (DIC) (transmitted light techniques). We also provide a first view of pollen using super-resolution microscopy. The three pollen types used to contrast the performance of each technique are: Croton hirtus (Euphorbiaceae), Mabea occidentalis (Euphorbiaceae) and Agropyron repens (Poaceae). No single microscopy technique provided an adequate picture of both the shape and surface texture of any of the three pollen types investigated here. The wavelength of incident light, photon-collection ability of the optical technique, signal-to-noise ratio, and the thickness and light absorption characteristics of the exine profoundly affect the recovery of morphological information by a given optical microscopy technique. Reflected light techniques, particularly confocal and two-photon microscopy, best capture pollen shape but provide limited information on very fine surface texture. In contrast, transmitted light techniques, particularly differential interference contrast microscopy, can resolve very fine surface texture but provide limited information on shape. Texture comprising sculptural elements that are spaced near the diffraction limit of light (~250 nm; NDL) presents an acute challenge to optical microscopy. Super-resolution structured illumination microscopy provides data on the NDL texture of A. repens that is more comparable to textural data from scanning electron microscopy than any other optical microscopy technique investigated here. Maximizing the recovery of morphological information from pollen grains should lead to more robust classifications, and an increase in the taxonomic precision with which ancient vegetation can be reconstructed.

Comment on "The response of vegetation on the Andean flank in western Amazonia to Pleistocene climate change".

Cárdenas et al. (Reports, 25 February 2011, p. 1055) used the presence of Podocarpus pollen and wood to infer ≥5°C cooling of Andean forests during Quaternary glacial periods. We show that (i) Podocarpus has a wide elevation range in the Neotropics, and (ii) edaphic factors cannot be discounted as a factor governing its distribution. Paleoecologists should therefore reevaluate Podocarpus as a cool-temperature proxy.

Mass extinction events and the plant fossil record.

Five mass extinction events have punctuated the geological record of marine invertebrate life. They are characterized by faunal extinction rates and magnitudes that far exceed those observed elsewhere in the geological record. Despite compelling evidence that these extinction events were probably driven by dramatic global environmental change, they were originally thought to have little macroecological or evolutionary consequence for terrestrial plants. New high-resolution regional palaeoecological studies are beginning to challenge this orthodoxy, providing evidence for extensive ecological upheaval, high species-level turnover and recovery intervals lasting millions of years. The challenge ahead is to establish the geographical extent of the ecological upheaval, because reconstructing the vegetation dynamics associated with these events will elucidate the role of floral change in faunal mass extinction and provide a better understanding of how plants have historically responded to global environmental change similar to that anticipated for our future.