Deep learning models map rapid plant species changes from citizen science and remote sensing data

Significance Worldwide, plant biodiversity is changing rapidly due to habitat destruction and a warming climate. However, we lack methods at high enough spatial and temporal resolution to detect these changes for individual species. Here, we develop a deep learning-based approach trained with citizen science data that detects thousands of plant species from satellite or aerial imagery. We show how this approach can detect individual species at meter-resolution in California and can detect rapid changes in the makeup of plant communities across both space and time. Our approach provides an efficient way to map plant biodiversity from above that is easily scalable to a global system for monitoring plant biodiversity.

• Publication year

  2024

• Venue

  Proceedings of the National Academy of Sciences of the United States of America

• Publication date

  2024-09-05

• Fields of study

  Biology, Medicine, Computer Science, Environmental Science

• Identifiers
  DOI 10.1073/pnas.2318296121PMID 39236239PMCID 11406280
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar, PubMed

• No claims are published for this paper.

• No concepts are published for this paper.

• Deep learning models map rapid plant species changes from citizen science and remote sensing data

  2024cited by this paper
• A global biodiversity observing system to unite monitoring and guide action

  2023cited by this paper
• Capturing long‐tailed individual tree diversity using an airborne imaging and a multi‐temporal hierarchical model

  2023cited by this paper
• Macroclimate data overestimate range shifts of plants in response to climate change

  2023cited by this paper
• The undetectability of global biodiversity trends using local species richness

  2023cited by this paper
• Highly accurate protein structure prediction with AlphaFold

  2021cited by this paper
• Convolutional neural networks improve species distribution modelling by capturing the spatial structure of the environment

  2021cited by this paper
• Deforestation reduces rainfall and agricultural revenues in the Brazilian Amazon

  2021cited by this paper
• Getting the big picture of biodiversity

  2021cited by this paper
• Predictive performance of presence‐only species distribution models: a benchmark study with reproducible code

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

  2021cited by this paper
• How deregulation, drought and increasing fire impact Amazonian biodiversity

  2021cited by this paper
• Monitoring biodiversity in the Anthropocene using remote sensing in species distribution models

  2020cited by this paper
• Bias in presence-only niche models related to sampling effort and species niches: Lessons for background point selection

  2020cited by this paper
• Image-based classification of plant genus and family for trained and untrained plant species

  2019cited by this paper
• Soil carbon sequestration accelerated by restoration of grassland biodiversity

  2019cited by this paper
• Predicting geographic location from genetic variation with deep neural networks

  2019cited by this paper
• Ongoing accumulation of plant diversity through habitat connectivity in an 18-year experiment

  2019cited by this paper
• WorldClim 2: new 1‐km spatial resolution climate surfaces for global land areas

  2017cited by this paper
• Positive biodiversity-productivity relationship predominant in global forests

  2016cited by this paper
• Plant diversity patterns in neotropical dry forests and their conservation implications

  2016cited by this paper
• Unprecedented remote sensing data over King and Rim megafires in the Sierra Nevada Mountains of California.

  2016cited by this paper
• Has land use pushed terrestrial biodiversity beyond the planetary boundary? A global assessment

  2016cited by this paper
• Structure and Composition of Old-Growth and Unmanaged Second-Growth Riparian Forests at Redwood National Park, USA

  2014cited by this paper
• High-Resolution Global Maps of 21st-Century Forest Cover Change

  2013cited by this paper
• Correlation and process in species distribution models: bridging a dichotomy

  2012cited by this paper
• The role of biotic interactions in shaping distributions and realised assemblages of species: implications for species distribution modelling

  2012cited by this paper
• The fate of Amazonian forest fragments: A 32-year investigation

  2011cited by this paper
• Mapping understory vegetation using phenological characteristics derived from remotely sensed data

  2010cited by this paper
• Paying the Extinction Debt in Southern Wisconsin Forest Understories

  2009cited by this paper
• A Significant Upward Shift in Plant Species Optimum Elevation During the 20th Century

  2008cited by this paper
• Genetic effects of chronic habitat fragmentation in a wind-pollinated tree.

  2006cited by this paper
• A globally coherent fingerprint of climate change impacts across natural systems

  2003cited by this paper
• Ecological Meltdown in Predator-Free Forest Fragments

  2001cited by this paper
• A role for ecotones in generating rainforest biodiversity

  1997cited by this paper

• Continental-scale computer vision models reveal generalizable patterns and pitfalls for urban tree inventories with street-view images

  2026cites this paper
• A calibration framework to improve mechanistic forecasts with hybrid dynamic models

  2025cites this paper
• Multi-scale species richness estimation with deep learning

  2025cites this paper
• The pipelines of deep learning-based plant image processing

  2025cites this paper
• Trait mediation explains decadal distributional shifts for a wide range of insect taxa

  2025cites this paper
• Assessing the role of citizen science in the validation of remote sensing habitat classifications

  2025cites this paper
• Interpretable and scalable deep learning for urban NO2 prediction via multisource data

  2025cites this paper
• Robust variable selection for spatial point processes observed with noise

  2025cites this paper
• Automated mask generation in citizen science smartphone photos and their value for mapping plant species in drone imagery

  2025cites this paper
• A Meta-Review on New Technologies for Marine Biodiversity Monitoring and Assessment

  2025cites this paper
• Linking to images and AI‐based identification tools—The only way for Flora projects to survive

  2025cites this paper
• Heterogeneous graph neural networks for species distribution modeling

  2025cites this paper
• MaskSDM with Shapley values to improve flexibility, robustness and explainability in species distribution modelling

  2025cites this paper
• Mapping Biodiversity at Very-High Resolution in Europe

  2025cites this paper
• Multimodal Spatio-Temporal Data Visualization Technologies for Contemporary Urban Landscape Architecture: A Review and Prospect in the Context of Smart Cities

  2025cites this paper
• GlobalGeoTree: A Multi-Granular Vision-Language Dataset for Global Tree Species Classification

  2025cites this paper
• Forest maturation and its drivers on the Qinghai-Xizang Plateau

  2025cites this paper
• The Performance and Potential of Deep Learning for Predicting Species Distributions

  2024cites this paper
• MALPOLON: A Framework for Deep Species Distribution Modeling

  2024cites this paper
• Contrastive ground-level image and remote sensing pre-training improves representation learning for natural world imagery

  2024influential citation
• Deep learning sharpens vistas on biodiversity mapping

  2024cites this paper
• Hybrid Spatial Representations for Species Distribution Modeling

  2024cites this paper
• DivShift: Exploring Domain-Specific Distribution Shift in Volunteer-Collected Biodiversity Datasets

  2024influential citation
• Applying the maximum entropy principle to neural networks enhances multi-species distribution models

  2024cites this paper
• Deep learning models map rapid plant species changes from citizen science and remote sensing data

  2024cites this paper
• S PATIALLY I NFORMED A UTOENCODERS FOR I NTER - PRETABLE V ISUAL R EPRESENTATION L EARNING

  year unknowncites this paper