Machine Learning Using Digitized Herbarium Specimens to Advance Phenological Research

Katelin D. Pearson (1) , Gil Nelson (1) , Myla Aronson (2) , Pierre Bonnet (3, 4) , Laura Brenskelle (5) , Charles C. Davis (6) , Ellen Denny (7) , Elizabeth R. Ellwood (8) , Hervé Goëau (3) , J Mason Heberling (9) , Alexis Joly (10) , Titouan Lorieul (10) , Susan Mazer (11) , Emily Meineke (12) , Brian Stucky (13) , Patrick W. Sweeney (14) , Alexander White (15) , Pamela S. Soltis (13)

1 FSU - Florida State University [Tallahassee]
2 WINLAB Rutgers University
3 UMR AMAP - Botanique et Modélisation de l'Architecture des Plantes et des Végétations
4 Cirad-BIOS - Département Systèmes Biologiques
5 UF - University of Florida [Gainesville]
6 Harvard University
7 University of Arizona
8 Natural History Museum of Los Angeles County
9 Carnegie Museum of Natural History [Pittsburgh]
10 ZENITH - Scientific Data Management
11 UC Santa Barbara - University of California [Santa Barbara]
12 UC Davis - University of California [Davis]
13 Florida Museum of Natural History [Gainesville]
14 Peabody Museum of Natural History
15 Smithsonian Institution

Machine learning (ML) has great potential to drive scientific discovery by harvesting data from images of herbarium specimens—preserved plant material curated in natural history collections—but ML techniques have only recently been applied to this rich resource. ML has particularly strong prospects for the study of plant phenological events such as growth and reproduction. As a major indicator of climate change, driver of ecological processes, and critical determinant of plant fitness, plant phenology is an important frontier for the application of ML techniques for science and society. In the present article, we describe a generalized, modular ML workflow for extracting phenological data from images of herbarium specimens, and we discuss the advantages, limitations, and potential future improvements of this workflow. Strategic research and investment in specimen-based ML methods, along with the aggregation of herbarium specimen data, may give rise to a better understanding of life on Earth.

Mots clés: Phenology; Machine learning; Biodiversity; Climate change; Deep learning

Domaines: Systématique, phylogénie et taxonomie; Ecosystèmes; Botanique; Biodiversité et Ecologie

Katelin D. Pearson, Gil Nelson, Myla Aronson, Pierre Bonnet, Laura Brenskelle, et al.. Machine Learning Using Digitized Herbarium Specimens to Advance Phenological Research. Bioscience, 2020, 70 (7), pp.610-620. ⟨10.1093/biosci/biaa044⟩. ⟨hal-02573627⟩