PlantTraitNet: An Uncertainty-Aware Multimodal Framework for Global-Scale Plant Trait Inference from Citizen Science Data

Global plant maps of plant traits, such as leaf nitrogen or plant height, are essential for understanding ecosystem processes, including the carbon and energy cycles of the Earth system. However, existing trait maps remain limited by the high cost and sparse geographic coverage of field-based measurements. Citizen science initiatives offer a largely untapped resource to overcome these limitations, with over 50 million geotagged plant photographs worldwide capturing valuable visual information on plant morphology and physiology. In this study, we introduce PlantTraitNet, a multi-modal, multi-task uncertainty-aware deep learning framework that predictsfour key plant traits (plant height, leaf area, specific leaf area, and nitrogen content) from citizen science photos using weak supervision. By aggregating individual trait predictions across space, we generate global maps of trait distributions. We validate these maps against independent vegetation survey data (sPlotOpen) and benchmark them against leading global trait products. Our results show that PlantTraitNet consistently outperforms existing trait maps across all evaluated traits, demonstrating that citizen science imagery, when integrated with computer vision and geospatial AI, enables not only scalable but also more accurate global trait mapping. This approach offers a powerful new pathway for ecological research and Earth system modeling.

• Publication year

  2025

• Venue

  arXiv.org

• Publication date

  2025-11-10

• Fields of study

  Biology, Computer Science, Environmental Science

• Identifiers
  DOI 10.48550/arXiv.2511.06943arXiv 2511.06943
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar

• No claims are published for this paper.

• No concepts are published for this paper.

• From smartphones to satellites: Uniting crowdsourced biodiversity monitoring and Earth observation to fill the gaps in global plant trait mapping

  2025influential reference
• Intercomparison of global foliar trait maps reveals fundamental differences and limitations of upscaling approaches

  2024cited by this paper
• SatCLIP: Global, General-Purpose Location Embeddings with Satellite Imagery

  2023cited by this paper
• GeoCLIP: Clip-Inspired Alignment between Locations and Images for Effective Worldwide Geo-localization

  2023cited by this paper
• Uncertainty Voting Ensemble for Imbalanced Deep Regression

  2023cited by this paper
• DINOv2: Learning Robust Visual Features without Supervision

  2023cited by this paper
• From spectra to plant functional traits: Transferable multi-trait models from heterogeneous and sparse data

  2023cited by this paper
• BioCLIP: A Vision Foundation Model for the Tree of Life

  2023cited by this paper
• Robustness via Cross-Domain Ensembles

  2021cited by this paper
• Climatic and soil factors explain the two-dimensional spectrum of global plant trait variation

  2021cited by this paper
• Pl@ntNet-300K: a plant image dataset with high label ambiguity and a long-tailed distribution

  2021cited by this paper
• Observing the Observers: How Participants Contribute Data to iNaturalist and Implications for Biodiversity Science

  2021cited by this paper
• Deep learning and citizen science enable automated plant trait predictions from photographs

  2021influential reference
• The Semi-Supervised iNaturalist Challenge at the FGVC8 Workshop

  2021cited by this paper
• Emerging Properties in Self-Supervised Vision Transformers

  2021cited by this paper
• Understanding deep learning (still) requires rethinking generalization

  2021influential reference
• Components of leaf-trait variation along environmental gradients.

  2020cited by this paper
• Assessing the reliability of predicted plant trait distributions at the global scale

  2020cited by this paper
• sPlotOpen – An environmentally balanced, open‐access, global dataset of vegetation plots

  2020cited by this paper
• Quantifying the Carbon Emissions of Machine Learning

  2019influential reference
• TRY plant trait database - enhanced coverage and open access.

  2019influential reference
• A methodology to derive global maps of leaf traits using remote sensing and climate data

  2018cited by this paper
• ape 5.0: an environment for modern phylogenetics and evolutionary analyses in R

  2018cited by this paper
• Future global productivity will be affected by plant trait response to climate

  2018cited by this paper
• Global trait–environment relationships of plant communities

  2018cited by this paper
• Mapping local and global variability in plant trait distributions

  2017influential reference
• Global climatic drivers of leaf size

  2017cited by this paper
• The iNaturalist Species Classification and Detection Dataset

  2017cited by this paper
• Climatologies at high resolution for the earth’s land surface areas

  2016cited by this paper
• The global spectrum of plant form and function

  2015cited by this paper
• A fully traits-based approach to modeling global vegetation distribution

  2014cited by this paper
• A Large and Persistent Carbon Sink in the World’s Forests

  2011cited by this paper
• TRY – a global database of plant traits

  2011influential reference
• TESTING FOR PHYLOGENETIC SIGNAL IN COMPARATIVE DATA: BEHAVIORAL TRAITS ARE MORE LABILE

  2003cited by this paper
• A fast and elitist multiobjective genetic algorithm: NSGA-II

  2002cited by this paper
• Primary production of the biosphere: integrating terrestrial and oceanic components

  1998cited by this paper
• Biogeochemistry: An Analysis of Global Change

  1991cited by this paper

• No citing papers are available for this paper.