About

Surangi W Punyasena is an Associate Professor of Plant Biology at the University of Illinois Urbana-Champaign, with affiliations in the Center for Latin American and Caribbean Studies, the Carl R. Woese Institute for Genomic Biology, Geography and Geographic Information Science, and Earth Science and Environmental Change. She holds a BA from Yale University and both an SM and PhD from The University of Chicago. Her research focuses on palynology, microscopy, machine learning, and evolution, with a particular emphasis on understanding the influence of climate on the composition, structure, and long-term evolution of lowland Neotropical plant communities. Using the fossil pollen record, her lab documents plant responses to past climate variability, leveraging the widespread presence of pollen and spores in terrestrial sediment records to study long-term trends in plant ecology and evolution. Professor Punyasena's work aims to re-imagine paleoecology by expanding the range of ecological and evolutionary hypotheses that can be addressed through increasing the throughput, reproducibility, and taxonomic resolution of microfossil data. Her current research focuses on developing advanced microscopy and computer automation methods, including image analysis and machine learning, to improve the quantity and quality of pollen and spore counts. These new tools enable the creation of larger and more comprehensive datasets, thereby broadening the scope of paleoecological research. Her long-term goal is to transform the paleoecological analysis workflow from imaging to classification to interpretation, enhancing the ability to study plant evolution and environmental adaptations over time.

Research topics

• Biology
• Artificial Intelligence
• Computer Science
• Political Science
• Ecology
• Geography
• Zoology
• Optics
• Physics
• Environmental science
• Botany
• Environmental resource management
• Environmental planning
• Evolutionary biology

Selected publications

• Herbivory in Pennsylvanian Peat Forests: A Systematic Framework for Arthropod Coprolite Morphotypes

  SSRN Electronic Journal · 2026-01-01

  preprintOpen accessSenior author
  PublisherDOI

• Digitizing microscope slide-based natural history collections: A protocol using slide scanner technology

  PLoS ONE · 2026-04-24 · 1 citations

  articleOpen access

  Natural history collections contain millions of microscope slides documenting global microscopic biodiversity, yet these materials remain largely undigitized and are vulnerable to deterioration and loss. Recent advances in slide scanner technology, originally developed for medical pathology, offer new opportunities for comprehensive digitization of slide-based collections. Here we present an optimized protocol for digitizing diverse microscope slide specimens, using the Hamamatsu NanoZoomer S20 slide scanner, developed while imaging slides at the Smithsonian National Museum of Natural History. We provide specimen-specific recommendations for scanning parameters, including scan area, focal points, Z-stack configuration, and file management workflows. Scanning times range from 41 seconds for small invertebrates to 18 minutes for palynological samples, with final compressed file sizes of 0.15-28 GB. High-resolution images (0.23 μm/pixel) captured diagnostic morphological features across all specimen types, including pollen, diatoms, radiolarians, plant and fungi tissues, and invertebrates. Using this method, we estimated that just the NMNH's paleo-palynology slide collection contains approximately 4.3 billion individual specimens, 30 times more than the current estimated size of the entire NMNH collection. Slide scanning enables 3D data capture, facilitates remote collaboration, improves reproducibility of taxonomic identifications, and creates permanent digital records that mitigate risks of physical deterioration. This protocol provides practical guidance for institutions looking to digitize slide-based collections to preserve and unlock their full research potential.

  PublisherDOI

• Detrital Processing in Pennsylvanian Peat Forests: A Systematic Framework for Arthropod Coprolite Morphotypes 

  SSRN Electronic Journal · 2026-01-01

  preprintOpen accessSenior author
  PublisherDOI

• Pollen morphology, deep learning, phylogenetics, and the evolution of environmental adaptations in <i>Podocarpus</i>

  bioRxiv (Cold Spring Harbor Laboratory) · 2025-01-08 · 1 citations

  preprintOpen accessSenior authorCorresponding

  Summary Podocarpus pollen morphology is shaped by both phylogenetic history and the environment. We analyzed the relationship between pollen traits quantified using deep learning and environmental factors within a comparative phylogenetic framework. We investigated the influence of mean annual temperature, annual precipitation, altitude, and solar radiation in driving morphological change. We used trait-environment regression models to infer the temperature tolerances of 31 Neotropical Podocarpidites fossils. Ancestral state reconstructions were applied to the Podocarpus phylogeny with and without the inclusion of fossils. Our results show that temperature and solar radiation influence pollen morphology, with thermal stress driving an increase in pollen size and higher UV-B radiation selecting for thicker corpus walls. Fossil temperature tolerances inferred from trait-environment models aligned with paleotemperature estimates from global paleoclimate models. Incorporating fossils into ancestral state reconstructions revealed that early ancestral Podocarpus lineages were likely adapted to warm climates, with cool-temperature tolerance evolving independently in high-latitude and high-altitude species. Our results highlight the importance of deep learning-derived features in advancing our understanding of plant environmental adaptations over evolutionary timescales. Deep learning allows us to quantify subtle interspecific differences in pollen morphology and link these traits to environmental preferences through statistical and phylogenetic analyses.

  PublisherOA PDFDOI

• Addressing the open world: detecting and segmenting pollen on palynological slides with deep learning

  bioRxiv (Cold Spring Harbor Laboratory) · 2025-01-05 · 1 citations

  preprintOpen accessSenior authorCorresponding

  In the open world, categorical classes are imbalanced, test classes are not known a priori, and test data are captured across different domains. Paleontological data can be described as open-world, as specimens may include new, unknown taxa, and the data collected, such as measurements or images, may not be standardized across different studies. Fossil pollen analysis is one example of an open-world problem in paleontology. Pollen samples capture large numbers of specimens, including not only common types but also rare and even novel taxa. Pollen is diverse morphologically and features can be altered during fossilization. Additionally, there is little standardization in the methods used to capture and catalog pollen images and most collections are mounted on microscope slides. Therefore, generalized workflows for automated pollen analysis require techniques that are robust to these differences and can work with microscope images. We focus on a critical first step, the detection of pollen specimens on a palynological slide and review how existing methods can be employed to build robust and generalizable analysis pipelines. First, we demonstrate how a mixture-of-experts approach -- the fusion of a general pollen detector with an expert model trained on minority classes -- can be used to address taxonomic biases in detections, particularly the missed detections of rarer pollen types. Second, we demonstrate the efficiency of domain fine-tuning in addressing domain gaps -- differences in image magnification and resolution across microscopes, and of taxa across different sample sources. Third, we demonstrate the importance of continual learning workflows, which integrate expert feedback, in training detection models from incomplete data. Finally, we demonstrate how cutting-edge segmentation models can be used to refine and clean detections for downstream deep learning classification models.

  PublisherDOI

• Reconstructing Terrestrial Arthropod Feeding Dynamics in Pennsylvanian Peat Forest Using Coprolites

  Abstracts with programs - Geological Society of America · 2025-01-01

  articleSenior author
  PublisherDOI

• Global Coal Ball Stable Isotope Analysis to Define Permineralizing Environments

  Abstracts with programs - Geological Society of America · 2025-01-01

  article
  PublisherDOI

• Digitizing collections to unlock the full potential of palynology: A case study with the Smithsonian palynology collection

  Plants People Planet · 2025-08-01 · 4 citations

  articleOpen access

  Societal Impact Statement Large palynological collections have been built over decades and contain vital information. However, they are often difficult to access and use effectively. What is the point of having such collections if they are not fully utilizable? To solve this problem, we digitized the Smithsonian palynological collection using both light and confocal microscopy. We digitized the pollen of 12,000 species and took 40 million photos. Our image library will support a wide array of applications, including environmental monitoring, public health, biodiversity studies, paleoclimate, and the analysis of landscape changes across spatial and temporal scales. It will also aid in geological correlations used in water exploration and in hydrocarbon storage/production. Summary Palynology is a century‐old practice, contributing data to various fields, from geology to medicine and forensics. Palynological analyses are highly time‐consuming and involve visually finding, identifying, and counting thousands of palynomorph grains on microscope slides. These analyses are especially challenging in high‐diversity tropical settings. Fortunately, the development of deep learning and the capability to digitize entire microscope slides are allowing palynology to enter a new era. Foundational to this transformation is building solid digital collections that can be achieved by digitizing botanical collections. We are digitizing the Smithsonian palynological collections, which contain ~18,000 species, most of which are Neotropical taxa in the Graham Pollen Collection . This digital product consists of high‐resolution images of different types—transmitted light, differential interference contrast, and optical superresolution (Airyscan)—which will be freely available and lay the groundwork for training deep‐learning models and applying novel image analysis to palynomorph morphology. Image quality matters, so we outline the best practices we have developed throughout the years of imaging and experimentation. High‐resolution imaging of palynological collections holds the key to unraveling the full potential that the study of pollen and spores can offer.

  PublisherOA PDFDOI

• Deep learning approaches to the phylogenetic placement and classification of fossil grass silica short cell phytoliths

  Abstracts with programs - Geological Society of America · 2025-01-01

  article
  PublisherDOI

• Identifying the Big Questions in paleontology: a community-driven project

  Paleobiology · 2025-08-01 · 4 citations

  articleOpen access

  Abstract Paleontology provides insights into the history of the planet, from the origins of life billions of years ago to the biotic changes of the Recent. The scope of paleontological research is as vast as it is varied, and the field is constantly evolving. In an effort to identify “Big Questions” in paleontology, experts from around the world came together to build a list of priority questions the field can address in the years ahead. The 89 questions presented herein (grouped within 11 themes) represent contributions from nearly 200 international scientists. These questions touch on common themes including biodiversity drivers and patterns, integrating data types across spatiotemporal scales, applying paleontological data to contemporary biodiversity and climate issues, and effectively utilizing innovative methods and technology for new paleontological insights. In addition to these theoretical questions, discussions touch upon structural concerns within the field, advocating for an increased valuation of specimen-based research, protection of natural heritage sites, and the importance of collections infrastructure, along with a stronger emphasis on human diversity, equity, and inclusion. These questions offer a starting point—an initial nucleus of consensus that paleontologists can expand on—for engaging in discussions, securing funding, advocating for museums, and fostering continued growth in shared research directions.

  PublisherDOI

Recent grants

• Collaborative Research: ABI Innovation: Breaking through the taxonomic barrier of the fossil pollen record using bioimage informatics

  NSF · $514k · 2013–2017

• Automated Analysis of the Reproductive Response of Two Panamanian Forests to ENSO Climate Variation - A Machine Learning Pilot Study

  NSF · $237k · 2012–2015

• Collaborative Research: Biological Shape Spaces, Transforming Shape into Knowledge

  NSF · $297k · 2010–2014

Frequent coauthors

• Luke Mander

  The Open University

  24 shared

• Michael A. Urban

  19 shared

• Charless C. Fowlkes

  18 shared

• Shu Kong

  18 shared

• Carlos Jaramillo

  17 shared

• David Tcheng

  13 shared

• Ingrid Romero

  10 shared

• Derek S. Haselhorst

  Urbana University

  10 shared

Labs

• Punyasena Paleoecology LabPI

  palynology, microscopy, machine learning, evolution