Fine scale plant community assessment in coastal meadows using UAV based multispectral data

Miguel Villoslada, Thaisa Fernandes Bergamo, Raymond Ward, Niall Burnside, Chris Joyce, Robert Bunce, Kalev Sepp

Research output: Contribution to journalArticle

Abstract

Coastal meadows worldwide are subjected to habitat degradation due to abandonment, intensification and the impacts of global change. In order to protect and restore these habitats and ensure the supply of valuable ecosystem services, it is necessary to know the extent and location of plant communities in coastal meadows. In this study, five plant communities were mapped at very high resolution in three different study sites in West Estonia. A fixed wing UAV was used to obtain multispectral images and derive a set of vegetation indices. Two different image classification techniques were used to cluster the vegetation indices maps and produce plant community distribution maps. The highest classification accuracy was obtained using a Random Forest classifier and 13 vegetation indices. Additionally, the spectral characteristics of the training samples were correlated with aboveground biomass and species diversity. Both biomass and species diversity were positively correlated with the spectral diversity of training samples and are thus likely to have an effect on the classification accuracy. The results of this study highlight the need to utilize a wide array of vegetation indices and assess the spectral characteristics of training samples in order to obtain high classification accuracies and understand the nature of misclassification errors. The resulting maps provide a solid foundation for global change impact assessment and habitat management and restoration in coastal meadows.
Original languageEnglish
Article number105979
JournalEcological Indicators
Volume111
DOIs
Publication statusPublished - 14 Jan 2020

Fingerprint

vegetation index
meadow
plant community
global change
species diversity
habitat restoration
habitat management
multispectral image
habitat
image classification
aboveground biomass
ecosystem service
biomass

Keywords

  • Coastal plant communities
  • UAV
  • vegetation indices
  • Random Forests
  • unsupervised classification
  • Vegetation indices
  • Unsupervised classification
  • Random forests

Cite this

@article{5105d6669ae84440b68e088a94aa0cf6,
title = "Fine scale plant community assessment in coastal meadows using UAV based multispectral data",
abstract = "Coastal meadows worldwide are subjected to habitat degradation due to abandonment, intensification and the impacts of global change. In order to protect and restore these habitats and ensure the supply of valuable ecosystem services, it is necessary to know the extent and location of plant communities in coastal meadows. In this study, five plant communities were mapped at very high resolution in three different study sites in West Estonia. A fixed wing UAV was used to obtain multispectral images and derive a set of vegetation indices. Two different image classification techniques were used to cluster the vegetation indices maps and produce plant community distribution maps. The highest classification accuracy was obtained using a Random Forest classifier and 13 vegetation indices. Additionally, the spectral characteristics of the training samples were correlated with aboveground biomass and species diversity. Both biomass and species diversity were positively correlated with the spectral diversity of training samples and are thus likely to have an effect on the classification accuracy. The results of this study highlight the need to utilize a wide array of vegetation indices and assess the spectral characteristics of training samples in order to obtain high classification accuracies and understand the nature of misclassification errors. The resulting maps provide a solid foundation for global change impact assessment and habitat management and restoration in coastal meadows.",
keywords = "Coastal plant communities, UAV, vegetation indices, Random Forests, unsupervised classification, Vegetation indices, Unsupervised classification, Random forests",
author = "Miguel Villoslada and Bergamo, {Thaisa Fernandes} and Raymond Ward and Niall Burnside and Chris Joyce and Robert Bunce and Kalev Sepp",
year = "2020",
month = "1",
day = "14",
doi = "10.1016/j.ecolind.2019.105979",
language = "English",
volume = "111",
journal = "Ecological Indicators",
issn = "1470-160X",

}

Fine scale plant community assessment in coastal meadows using UAV based multispectral data. / Villoslada, Miguel; Bergamo, Thaisa Fernandes; Ward, Raymond; Burnside, Niall; Joyce, Chris; Bunce, Robert; Sepp, Kalev.

In: Ecological Indicators, Vol. 111, 105979, 14.01.2020.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Fine scale plant community assessment in coastal meadows using UAV based multispectral data

AU - Villoslada, Miguel

AU - Bergamo, Thaisa Fernandes

AU - Ward, Raymond

AU - Burnside, Niall

AU - Joyce, Chris

AU - Bunce, Robert

AU - Sepp, Kalev

PY - 2020/1/14

Y1 - 2020/1/14

N2 - Coastal meadows worldwide are subjected to habitat degradation due to abandonment, intensification and the impacts of global change. In order to protect and restore these habitats and ensure the supply of valuable ecosystem services, it is necessary to know the extent and location of plant communities in coastal meadows. In this study, five plant communities were mapped at very high resolution in three different study sites in West Estonia. A fixed wing UAV was used to obtain multispectral images and derive a set of vegetation indices. Two different image classification techniques were used to cluster the vegetation indices maps and produce plant community distribution maps. The highest classification accuracy was obtained using a Random Forest classifier and 13 vegetation indices. Additionally, the spectral characteristics of the training samples were correlated with aboveground biomass and species diversity. Both biomass and species diversity were positively correlated with the spectral diversity of training samples and are thus likely to have an effect on the classification accuracy. The results of this study highlight the need to utilize a wide array of vegetation indices and assess the spectral characteristics of training samples in order to obtain high classification accuracies and understand the nature of misclassification errors. The resulting maps provide a solid foundation for global change impact assessment and habitat management and restoration in coastal meadows.

AB - Coastal meadows worldwide are subjected to habitat degradation due to abandonment, intensification and the impacts of global change. In order to protect and restore these habitats and ensure the supply of valuable ecosystem services, it is necessary to know the extent and location of plant communities in coastal meadows. In this study, five plant communities were mapped at very high resolution in three different study sites in West Estonia. A fixed wing UAV was used to obtain multispectral images and derive a set of vegetation indices. Two different image classification techniques were used to cluster the vegetation indices maps and produce plant community distribution maps. The highest classification accuracy was obtained using a Random Forest classifier and 13 vegetation indices. Additionally, the spectral characteristics of the training samples were correlated with aboveground biomass and species diversity. Both biomass and species diversity were positively correlated with the spectral diversity of training samples and are thus likely to have an effect on the classification accuracy. The results of this study highlight the need to utilize a wide array of vegetation indices and assess the spectral characteristics of training samples in order to obtain high classification accuracies and understand the nature of misclassification errors. The resulting maps provide a solid foundation for global change impact assessment and habitat management and restoration in coastal meadows.

KW - Coastal plant communities

KW - UAV

KW - vegetation indices

KW - Random Forests

KW - unsupervised classification

KW - Vegetation indices

KW - Unsupervised classification

KW - Random forests

UR - http://www.scopus.com/inward/record.url?scp=85077769097&partnerID=8YFLogxK

U2 - 10.1016/j.ecolind.2019.105979

DO - 10.1016/j.ecolind.2019.105979

M3 - Article

VL - 111

JO - Ecological Indicators

JF - Ecological Indicators

SN - 1470-160X

M1 - 105979

ER -