Attention:The NSF Public Access Repository (PAR) system and access will be unavailable from 11:00 PM ET on Thursday, October 15th until 12:00 AM ET on Friday, October 16th due to maintenance. We apologize for the inconvenience.

Explore Research Products in the PAR
It may take a few hours for recently added research products to appear in PAR search results.


Title: PhenoVision: A framework for automating and delivering research‐ready plant phenology data from field images
Abstract Plant phenology plays a fundamental role in shaping ecosystems, and global change‐induced shifts in phenology have cascading impacts on species interactions and ecosystem structure and function. Detailed, high‐quality observations of when plants undergo seasonal transitions such as leaf‐out, flowering and fruiting are critical for tracking causes and consequences of phenology shifts, but these data are often sparse and biased globally. These data gaps limit broader generalizations and forecasting improvements in the face of continuing disturbance. One solution to closing such gaps is to document phenology on field images taken by public participants. iNaturalist, in particular, provides global‐scale research‐grade data and is expanding rapidly.Here we utilize over 53 million field images of plants and millions of human annotations from iNaturalist—data spanning all angiosperms and drawn from across the globe—to train a computer vision model (PhenoVision) to detect the presence of fruits and flowers. PhenoVision utilizes a vision transformer architecture pretrained with a masked autoencoder to improve classification success, and it achieves high accuracy on held‐out test images for flower (98.5%) and fruit presence (95%), as well as a high level of agreement with an expert annotator (98.6% for flowers and 90.4% for fruits).Key to producing research‐ready phenology data is post‐calibration tuning and validation focused on reducing noise inherent in field photographs, and maximizing the true positive rate. We also develop a standardized set of quality metrics and metadata so that results can be used effectively by the community. Finally, we showcase how this effort vastly increases phenology data coverage, including regions of the globe where data have been limited before.Our end products are tuned models, new data resources and an application streamlining discovery and use of those data for the broader research and management community. We close by discussing next steps, including automating phenology annotations, adding new phenology targets, for example leaf phenology, and further integration with other resources to form a global central database integrating all in situ plant phenology resources.  more » « less
Award ID(s):
2514928 2223509 2419155 2223512
PAR ID:
10628300
Author(s) / Creator(s):
; ; ; ; ; ; ; ;
Publisher / Repository:
Wiley
Date Published:
Journal Name:
Methods in Ecology and Evolution
Volume:
16
Issue:
8
ISSN:
2041-210X
Page Range / eLocation ID:
1763 to 1780
Format(s):
Medium: X
Sponsoring Org:
National Science Foundation
More Like this
  1. ; ; ; ; ; ; ; ; ; ; et al (, New Phytologist)
    Summary Plant phenology, the timing of recurrent biological events, shows key and complex response to climate warming, with consequences for ecosystem functions and services. A key challenge for predicting plant phenology under future climates is to determine whether the phenological changes will persist with more intensive and long‐term warming.Here, we conducted a meta‐analysis of 103 experimental warming studies around the globe to investigate the responses of four phenophases – leaf‐out, first flowering, last flowering, and leaf coloring.We showed that warming advanced leaf‐out and flowering but delayed leaf coloring across herbaceous and woody plants. As the magnitude of warming increased, the response of most plant phenophases gradually leveled off for herbaceous plants, while phenology responded in proportion to warming in woody plants. We also found that the experimental effects of warming on plant phenology diminished over time across all phenophases. Specifically, the rate of changes in first flowering for herbaceous species, as well as leaf‐out and leaf coloring for woody species, decreased as the experimental duration extended.Together, these results suggest that the real‐world impact of global warming on plant phenology will diminish over time as temperatures continue to increase. 
    more » « less
  2. ; ; ; ; ; ; (, Applications in Plant Sciences)
    Abstract PremisePlant phenology dictates many aspects of community function and ecosystem dynamics. Yet, global phenology data are still limited, especially in areas lacking monitoring programs. Here we present a new data resource, PhenoVision–Leaf, which extends a computer vision pipeline utilizing iNaturalist digital image vouchers to produce global‐scale leaf phenophase data for deciduous woody genera. MethodsWe first discuss our implementation of a new human annotation framework for leaf phenology on iNaturalist, aligning with phenophase definitions used by the larger phenology community. We then showcase the use of 165,988 crowdsourced annotated records to train a Vision Transformer model with a two‐stage regime to maximize precision across single‐ and multi‐image records. This approach extends PhenoVision from scoring individual images to aggregating at the iNaturalist record level, better aligning with human annotation processes. ResultsPost‐hoc validation showed high performance for detecting present green and colored leaves (>98% precision) and reasonable precision for breaking leaf buds (>87% precision). Applying PhenoVision–Leaf to over six million iNaturalist records yielded 5.6 million record‐level phenology observations across 6500 species and 57 families, filling geographic and taxonomic gaps. DiscussionThese data, now accessible through the Phenobase web application, establish a foundation for near real‐time monitoring of leaf phenology, supporting global‐scale synthesis analyses. 
    more » « less
  3. (, Functional Ecology)
    Abstract Plant functional strategies change considerably as plants develop, driven by intraindividual variability in anatomical, morphological, physiological and architectural traits.Developmental trait variation arises through the complex interplay among genetically regulated phase change (i.e. ontogeny), increases in plant age and size, and phenotypic plasticity to changing environmental conditions. Although spatial drivers of intraspecific trait variation have received extensive research attention, developmentally driven intraspecific trait variation is largely overlooked, despite widespread occurrence.Ontogenetic trait variation is genetically regulated, leads to dramatic changes in plant phenotypes and evolves in response to predictable changes in environmental conditions as plants develop.Evidence has accumulated to support a general shift from fast to slow relative growth rates and from shade to sun leaves as plants develop from the highly competitive but shady juvenile niche to the stressful adult niche in the systems studied to date.Nonetheless, there are major gaps in our knowledge due to examination of only a few environmental factors selecting for the evolution of ontogenetic trajectories, variability in how ontogeny is assigned, biogeographic sampling biases on trees in temperate biomes, dependencies on a few broadly sampled leaf morphological traits and a lack of longitudinal studies that track ontogeny within individuals. Filling these gaps will enhance our understanding of plant functional ecology and provide a framework for predicting the effects of global change threats that target specific ontogenetic stages. Read the freePlain Language Summaryfor this article on the Journal blog. 
    more » « less
  4. ; (, Functional Ecology)
    Abstract Despite the importance of flowers to plant fitness, it remains unclear whether flowers display ecophysiological strategies and how floral traits are associated with leaf economic traits.Using a comparative analysis of flowers and leaves, we evaluated associations between seven homologous traits related to carbon and water economics in 245 plant species sampled at four locations spanning more than 100 degrees of latitude: Arctic tundra, tropicalcampos rupestres, Patagonian steppe and a botanical garden in the Mediterranean‐type climate of California. We used correlation analysis, standard major axes, principal component analysis and trait networks to evaluate the associations among flower and leaf traits.We found flower trait combinations ranging from large‐sized flowers with thick petals and lengthy water turnover times (τ) to small‐sized flowers with thin petaloid structures and high residual conductance (gres). There was positive scaling among flower and leaf traits such asgres,τ, petal and leaf thickness, and area. Network analysis showed no modular structure of flower and leaf traits when all species were evaluated together.Our results showcase that flower phenotypes are distributed along a continuum of trait variation and are functionally linked to leaves through water‐ and size‐related traits. This cross‐organ correlation indicates that habitat filtering based on the ecophysiological strategies of one organ would influence the phenotypes of the other organ. Read the freePlain Language Summaryfor this article on the Journal blog. 
    more » « less
  5. ; ; ; ; (, New Phytologist)
    Summary The scope of plant control over its microbiome is a central question in evolutionary biology and agriculture. Leaf traits are known to shape pathogen colonization and disease development, but their impact on the broader community of largely non‐pathogenic fungi that colonize plant leaves remains an open question.We used reciprocal common gardens of the model tree,Populus trichocarpa(black cottonwood), to examine relationships between leaf traits and the leaf mycobiome in two strongly contrasting environments. We measured six leaf traits (stomatal length, stomatal density, carbon‐to‐nitrogen ratio, leaf thickness, leaf dry matter content, and specific leaf area) and used fungal marker gene sequencing to characterize leaf fungal communities for 57 tree genotypes replicated in one mesic and one xeric common garden (809 trees).Several leaf traits covaried with the leaf mycobiome, yet one relationship was paramount: plant genotypes with longer, sparser leaf stomata hosted a greater richness and diversity of more similar fungal species compared to plant genotypes with shorter, denser leaf stomata.These relationships, while modulated by the environment plants were sourced from and grown in, suggest that stomatal traits may be a general mechanism through which plants and the leaf mycobiome influence one another. 
    more » « less
  • Citation Formats
  • MLA
  • APA
  • Chicago
  • BibTeX
  • Text Encoded Conversion
  • XML
  • Save / Share this Record
  • Send to Email