Plant phenology – the timing of plant life-cycle events such as leafing out, flowering, and fruiting – plays a fundamental role in the functioning of natural ecosystems and human agricultural and economic systems. Plant phenology is sensitive to environmental change and changes in plant phenology can dramatically impact ecosystem function and services. Therefore, plant phenology is at the center of understanding how accelerating environmental changes (e.g., climate and land-use changes) are reshaping biosphere. However, urgent phenology-related scientific questions remain difficult to answer due to both a lack of sufficient data and substantial challenges with integrating those data that have been collected. To address this challenge, we will build upon our substantial preliminary work over the last several years to 1) develop a global, standardized knowledge base (Phenobase) of phenological data by integrating different phenology observation networks around the world; 2) expand this knowledge base by using computer vision (CV) techniques to generate new, high-quality phenological data from the rapidly growing collection of community-submitted plant photographs on iNaturalist and Budburst; 3) add critical historical phenological data to the knowledge base by using similar CV techniques on herbarium specimens available through iDigBio and GBIF; 4) develop tools for data query, access, and visualization delivered via the Web and as software packages; and 5) foster compelling, community-driven use cases showcasing the use of Phenobase for new research and for public good. In short, Phenobase will support community needs for generating and delivering high-precision, harmonized and semantically integrated plant phenological data at unprecedented taxonomic, geographic, and temporal scales, along with new tools to help scientists and the public engage with these data.
Creating an accessible, integrated knowledge base of global plant phenological data with broad taxonomic, spatial, and temporal coverage (and tools to aid data use) will have far-reaching basic and applied scientific impact. Phenobase has two major intellectual merits. First, by closing taxonomic, spatial and temporal knowledge-gaps, it will enable finer-grain and broader extent analyses, dramatically improving understanding of global plant phenology pattern and process, relation to ecosystem function, and responses to climate and landscape change. This will, in turn, provide the basis for better predictive models that can be used for responding to the conservation, agricultural, and economic consequences of phenology change. A second area of merit is use of new tools to automate assembly and integration of vast and growing phenological data sources. At the core of this proposed work is a well-tested ontology- based data integration system that supports machine reasoning to dramatically enhance discovery of research-ready phenology data. This core is strongly augmented by links to cutting edge computer vision approaches that provide a means to keep pace with the exponential increases in plant photographs and borne-digital specimen vouchers and integrate them with in situ human observation data.
Plant phenology data and related products are well suited for outreach initiatives given their obvious and intuitive connections to a variety of topics of general interest, such as food supply and agriculture, gardening, conservation, and climate change. Phenobase leverages the innate societal interest in phenology in developing three major community, educational and public outreach components. First, we will host annual hybrid working groups built around Phenobase to catalyze community involvement and new collaborations within the phenology research community. Second, we will provide hands-on, multidisciplinary training connecting the science of phenology with software development for undergraduate and graduate researchers who will be directly involved in the development and testing of our data access software packages. We will develop curriculum to improve the data science literacy of undergraduate biology students through the exploration of computer vision approaches and phenology data via a new data carpentry module. Finally, we will work with citizen science projects such as Notes from Nature, iNaturalist, and Budburst to empower people around the world to contribute to phenology research.
Daijiang is an Assistant Professor in the Department of Biological Sciences and the Center for Computation & Technology at the Louisiana State University. Daijiang’s research focuses on understanding and forecasting the effects of global environmental change on biodiversity, phenology, and species interactions by integrating community ecology, statistics, and ecoinformatics.
John Deck is the founder of Biocode, LLC, as well as a staff programmer and research associate at the University of California at Berkeley, Berkeley Natural History Museums. For the last 20 years, John has worked on biodiversity informatics and connecting field biological samples to tissue samples, DNA sequences, and publications. He has built the Genomic Observatory MetaDatabase (GEOME) and projects annotating traits with ontology terms and building pipelines for assembling disparate data sources. John also owns and operates Deck Family Farm in Junction City, Oregon.
Ellen Denny is the Monitoring Design & Data Coordinator for the USA National Network (USA-NPN), which is based at the University of Arizona in Tucson. She has coordinated the development of the USA-NPN protocols for the collection of standardized ground-based plant and animal phenology observations across the nation. She serves as a data manager for the organization, and as a global liaison, helping to advise developing national phenology networks around the world, and working towards global integration of phenology data. She has a Master of Forest Science from the Yale School of Forestry & Environmental Studies. Ellen lives in Kittery, Maine.
Russell Dinnage is a Research Assistant Professor at Florida International University and a machine learning consultant, focusing on biological applications. He is interested in ecological and evolutionary processes at broad scales, and how advanced computational tools can help to understand them, ultimately in the service of developing strategies for the coexistence of human society and biological diversity. He has a particular interest in developing tools for organismal biology based on cutting edge artificial intelligence breakthroughs, and exploring their promise for generating novel biological insights.
Erin Grady is a botany master’s student at the University of Florida. Her work focuses on the dynamics of iNaturalist user activity and their role in filling spatial data gaps. She is passionate about connecting individuals to their local natural environments and is pursuing an Environmental Education and Communication graduate certificate alongside her degree. Other interests include urban forestry and equitable access to recreation and green space. She has a Bachelor of Science in Biology from California Polytechnic State University, San Luis Obispo.
Rob Guralnick is a UF Research Foundation Professor and Curator of Biodiversity Informatics at the University of Florida and Florida Museum of Natural History. His work focuses on understanding global environmental change, and he uses community science approaches along with machine learning to generate and assemble data and models about specie distributions and traits. These approaches are all in service to better understand which species might win or lose in the face of global change. He has special interest in phenology and phenological responses to changing environments.
Taran is the Community Science Programs Manager at Chicago Botanic Garden. Taran largely works on the climage change focused project Budburst, bringing together researchers, educators, and community scientists. Generally interested in how humans are a part of nature, Taran loves opportunities to show people the wonders of plants they might overlook.
Nevyn Neal is a second-year graduate student in the Ecology and Evolutionary Biology department at the University of Arizona. His research focuses on optimizing the ways in which Community Ecology-sourced datasets are analyzed and integrated, work which includes developing AI Computer-vision models at large scales. Before moving to UA, Nevyn studied at Louisiana State University, and received his Biology and Computer Science bachelor degrees at Williams College.
Carrie works for iNaturalist, a platform connecting millions of people to nature and advancing understanding of biodiversity. As the Head of Engagement, Carrie works to expand the impact of iNaturalist through collaborations and capacity building. Prior to joining the iNaturalist team in 2018, she was a Science & Technology Policy Fellow with the American Association for the Advancement of Science at the National Science Foundation where she worked on open data policy. She has a PhD from the University of Illinois at Chicago where she studied seed dispersal ecology. Carrie lives in Washington, DC.
Ramona L. Walls, Ph.D. is a consulting ontologist for Phenobase. She is Executive Director of Data Science at the Critical Path Institute (C-Path). She oversees multiple efforts including the development of C-Path’s Data and Analytics Platform, expansion and modernization of C-Path’s data integration pipeline to encompass new data types, and development of a rare disease knowledge graph. Walls joined C-Path in December 2020 and in 2021 she was promoted to Associate Director of Data Science. Walls retains an appointment in the Bio5 Institute of the University of Arizona. She received a bachelor’s degree in Environmental Resource Management and Horticulture at Penn State and a Ph.D. in Ecology and Evolution from Stony Brook University and did a post-doc at the New York Botanical Garden. Google scholar profile