University of Leicester
Browse
IGARSS_2018_PAPER.pdf (501.83 kB)

DETECTING VEGETATION RESPONSE TO OIL POLLUTION USING HYPERSPECTRAL INDICES

Download (501.83 kB)
conference contribution
posted on 2020-05-04, 14:55 authored by NN Onyia, H Balzter, JC Berrio
Vegetation reflectance from polluted and control transects was extracted from hyperspectral image acquired on 23rd November 2015. Band depths of absorption maxima for chlorophyll (Chl), anthocyanins (AnC) and carotenoids (CaR) were determined using the continuum removal. Results from polluted transects depict increased reflectance at the Chl absorption features (around 445 and between 650-700 nm) and decreased at the CaR and AnC absorption features, both commonly associated with stressed vegetation. Strong relationship was observed between total petroleum hydrocarbons (TPH) levels in soil and field measured Chl data, suggesting that oil pollution altered pigment content in vegetation growing on impacted transects. This study appears to be a first of its kind in that the variables tested were measured in the 'real' world, thus asserting the ecological validity of the research. In addition, integrating remote sensing data with field measurements provide a novel alternative to field based post-impact assessment studies.

History

Citation

N. N. Onyia, H. Balzter and J. C. Berrio, "Detecting Vegetation Response to Oil Pollution Using Hyperspectral Indices," IGARSS 2018 - 2018 IEEE International Geoscience and Remote Sensing Symposium, Valencia, 2018, pp. 3963-3966.

Source

38th IEEE International Geoscience and Remote Sensing Symposium (IGARSS)

Published in

IGARSS 2018 - 2018 IEEE INTERNATIONAL GEOSCIENCE AND REMOTE SENSING SYMPOSIUM

Volume

2018-July

Pagination

3963 - 3966 (4)

Publisher

IEEE

issn

2153-6996

isbn

9781538671504

Acceptance date

2018-07-02

Copyright date

2018

Available date

2018-11-05

Publisher version

https://ieeexplore.ieee.org/abstract/document/8519398

Spatial coverage

Valencia, SPAIN

Temporal coverage: start date

2018-07-22

Temporal coverage: end date

2018-07-27

Language

en

Usage metrics

    University of Leicester Publications

    Categories

    No categories selected

    Exports

    RefWorks
    BibTeX
    Ref. manager
    Endnote
    DataCite
    NLM
    DC