Machine Learning Undercounts Reproductive Organs on Herbarium Specimens but Accurately Derives Their Quantitative Phenological Status: A Case Study of <i>Streptanthus tortuosus</i>
Machine learning (ML) can accelerate the extraction of phenological data from herbarium specimens; however, no studies have assessed whether ML-derived phenological data can be used reliably to evaluate ecological patterns. In this study, 709 herbarium specimens representing a widespread annual herb...
Main Authors: | , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2021-11-01
|
Series: | Plants |
Subjects: | |
Online Access: | https://www.mdpi.com/2223-7747/10/11/2471 |
_version_ | 1797508658437816320 |
---|---|
author | Natalie L. R. Love Pierre Bonnet Hervé Goëau Alexis Joly Susan J. Mazer |
author_facet | Natalie L. R. Love Pierre Bonnet Hervé Goëau Alexis Joly Susan J. Mazer |
author_sort | Natalie L. R. Love |
collection | DOAJ |
description | Machine learning (ML) can accelerate the extraction of phenological data from herbarium specimens; however, no studies have assessed whether ML-derived phenological data can be used reliably to evaluate ecological patterns. In this study, 709 herbarium specimens representing a widespread annual herb, <i>Streptanthus tortuosus,</i> were scored both manually by human observers and by a mask R-CNN object detection model to (1) evaluate the concordance between ML and manually-derived phenological data and (2) determine whether ML-derived data can be used to reliably assess phenological patterns. The ML model generally underestimated the number of reproductive structures present on each specimen; however, when these counts were used to provide a quantitative estimate of the phenological stage of plants on a given sheet (i.e., the phenological index or PI), the ML and manually-derived PI’s were highly concordant. Moreover, herbarium specimen age had no effect on the estimated PI of a given sheet. Finally, including ML-derived PIs as predictor variables in phenological models produced estimates of the phenological sensitivity of this species to climate, temporal shifts in flowering time, and the rate of phenological progression that are indistinguishable from those produced by models based on data provided by human observers. This study demonstrates that phenological data extracted using machine learning can be used reliably to estimate the phenological stage of herbarium specimens and to detect phenological patterns. |
first_indexed | 2024-03-10T05:07:48Z |
format | Article |
id | doaj.art-74bc584fbc4140d7a554f9f0788e9d00 |
institution | Directory Open Access Journal |
issn | 2223-7747 |
language | English |
last_indexed | 2024-03-10T05:07:48Z |
publishDate | 2021-11-01 |
publisher | MDPI AG |
record_format | Article |
series | Plants |
spelling | doaj.art-74bc584fbc4140d7a554f9f0788e9d002023-11-23T01:06:20ZengMDPI AGPlants2223-77472021-11-011011247110.3390/plants10112471Machine Learning Undercounts Reproductive Organs on Herbarium Specimens but Accurately Derives Their Quantitative Phenological Status: A Case Study of <i>Streptanthus tortuosus</i>Natalie L. R. Love0Pierre Bonnet1Hervé Goëau2Alexis Joly3Susan J. Mazer4Department of Ecology, Evolution, and Marine Biology, University of California, Santa Barbara, CA 93106, USABotany and Modeling of Plant Architecture and Vegetation (AMAP), French Agricultural Research Centre for International Development (CIRAD), French National Centre for Scientific Research (CNRS), French National Institute for Agriculture, Food and Environment (INRAE), Research Institute for Development (IRD), University of Montpellier, 34398 Montpellier, FranceBotany and Modeling of Plant Architecture and Vegetation (AMAP), French Agricultural Research Centre for International Development (CIRAD), French National Centre for Scientific Research (CNRS), French National Institute for Agriculture, Food and Environment (INRAE), Research Institute for Development (IRD), University of Montpellier, 34398 Montpellier, FranceZENITH Team, Laboratory of Informatics, Robotics and Microelectronics-Joint Research Unit, Institut National de Recherche en Informatique et en Automatique (INRIA) Sophia-Antipolis, CEDEX 5, 34095 Montpellier, FranceDepartment of Ecology, Evolution, and Marine Biology, University of California, Santa Barbara, CA 93106, USAMachine learning (ML) can accelerate the extraction of phenological data from herbarium specimens; however, no studies have assessed whether ML-derived phenological data can be used reliably to evaluate ecological patterns. In this study, 709 herbarium specimens representing a widespread annual herb, <i>Streptanthus tortuosus,</i> were scored both manually by human observers and by a mask R-CNN object detection model to (1) evaluate the concordance between ML and manually-derived phenological data and (2) determine whether ML-derived data can be used to reliably assess phenological patterns. The ML model generally underestimated the number of reproductive structures present on each specimen; however, when these counts were used to provide a quantitative estimate of the phenological stage of plants on a given sheet (i.e., the phenological index or PI), the ML and manually-derived PI’s were highly concordant. Moreover, herbarium specimen age had no effect on the estimated PI of a given sheet. Finally, including ML-derived PIs as predictor variables in phenological models produced estimates of the phenological sensitivity of this species to climate, temporal shifts in flowering time, and the rate of phenological progression that are indistinguishable from those produced by models based on data provided by human observers. This study demonstrates that phenological data extracted using machine learning can be used reliably to estimate the phenological stage of herbarium specimens and to detect phenological patterns.https://www.mdpi.com/2223-7747/10/11/2471regional convolutional neural networkobject detectiondeep learningvisual data classificationclimate changenatural history collections |
spellingShingle | Natalie L. R. Love Pierre Bonnet Hervé Goëau Alexis Joly Susan J. Mazer Machine Learning Undercounts Reproductive Organs on Herbarium Specimens but Accurately Derives Their Quantitative Phenological Status: A Case Study of <i>Streptanthus tortuosus</i> Plants regional convolutional neural network object detection deep learning visual data classification climate change natural history collections |
title | Machine Learning Undercounts Reproductive Organs on Herbarium Specimens but Accurately Derives Their Quantitative Phenological Status: A Case Study of <i>Streptanthus tortuosus</i> |
title_full | Machine Learning Undercounts Reproductive Organs on Herbarium Specimens but Accurately Derives Their Quantitative Phenological Status: A Case Study of <i>Streptanthus tortuosus</i> |
title_fullStr | Machine Learning Undercounts Reproductive Organs on Herbarium Specimens but Accurately Derives Their Quantitative Phenological Status: A Case Study of <i>Streptanthus tortuosus</i> |
title_full_unstemmed | Machine Learning Undercounts Reproductive Organs on Herbarium Specimens but Accurately Derives Their Quantitative Phenological Status: A Case Study of <i>Streptanthus tortuosus</i> |
title_short | Machine Learning Undercounts Reproductive Organs on Herbarium Specimens but Accurately Derives Their Quantitative Phenological Status: A Case Study of <i>Streptanthus tortuosus</i> |
title_sort | machine learning undercounts reproductive organs on herbarium specimens but accurately derives their quantitative phenological status a case study of i streptanthus tortuosus i |
topic | regional convolutional neural network object detection deep learning visual data classification climate change natural history collections |
url | https://www.mdpi.com/2223-7747/10/11/2471 |
work_keys_str_mv | AT natalielrlove machinelearningundercountsreproductiveorgansonherbariumspecimensbutaccuratelyderivestheirquantitativephenologicalstatusacasestudyofistreptanthustortuosusi AT pierrebonnet machinelearningundercountsreproductiveorgansonherbariumspecimensbutaccuratelyderivestheirquantitativephenologicalstatusacasestudyofistreptanthustortuosusi AT hervegoeau machinelearningundercountsreproductiveorgansonherbariumspecimensbutaccuratelyderivestheirquantitativephenologicalstatusacasestudyofistreptanthustortuosusi AT alexisjoly machinelearningundercountsreproductiveorgansonherbariumspecimensbutaccuratelyderivestheirquantitativephenologicalstatusacasestudyofistreptanthustortuosusi AT susanjmazer machinelearningundercountsreproductiveorgansonherbariumspecimensbutaccuratelyderivestheirquantitativephenologicalstatusacasestudyofistreptanthustortuosusi |