Please cite this article in press as: Somers, B., et al., Spectral mixture analysis to monitor defoliation in mixed-aged Eucalyptus glob-
ulus Labill plantations in southern Australia using Landsat 5-TM and EO-1 Hyperion data. Int. J. Appl. Earth Observ. Geoinf. (2010),
doi:10.1016/j.jag.2010.03.005
ARTICLE IN PRESS
G Model
JAG-328; No. of Pages 8
International Journal of Applied Earth Observation and Geoinformation xxx (2010) xxx–xxx
Contents lists available at ScienceDirect
International Journal of Applied Earth Observation and
Geoinformation
journal homepage: www.elsevier.com/locate/jag
Spectral mixture analysis to monitor defoliation in mixed-aged Eucalyptus
globulus Labill plantations in southern Australia using Landsat 5-TM and EO-1
Hyperion data
B. Somers
a,∗
, J. Verbesselt
b
, E.M. Ampe
a
, N. Sims
b
, W.W. Verstraeten
a
, P. Coppin
a
a
Dept. of Biosystems, M3-BIORES, Katholieke Universiteit Leuven, W. de Croylaan 34, BE-3001 Leuven, Belgium
b
CSIRO Sustainable Ecosystems, Private Bag 10, Clayton South, VIC 3169, Australia
article info
Article history:
Received 19 January 2010
Accepted 29 March 2010
Keywords:
Defoliation
Unmixing
Hyperspectral
Multi-spectral
MESMA
Weighted spectral mixture analysis
Forest
LANDSAT
Hyperion
Leaf area index
abstract
Defoliation is a key parameter of forest health and is associated with reduced productivity and tree mor-
tality. Assessing the health of forests requires regular observations over large areas. Satellite remote
sensing provides a cost-effective alternative to traditional ground-based assessment of forest health, but
assessing defoliation can be difficult due to mixed pixels where vegetation cover is low or fragmented. In
this study we apply a novel spectral unmixing technique, referred to as weighted Multiple Endmember
Spectral Mixture Analysis (wMESMA), to Landsat 5-TM and EO-1 Hyperion data acquired over a Eucalyptus
globulus (Labill.) plantation in southern Australia. This technique combines an iterative mixture analysis
cycle allowing endmembers to vary on a per pixel basis (MESMA) and a weighting algorithm that pri-
oritizes wavebands based on their robustness against endmember variability. Spectral mixture analysis
provides an estimate of the physically interpretable canopy cover, which is not necessarily correlated
with defoliation in mixed-aged plantations due to natural variation in canopy cover as stands age. There
is considerable variability in the degree of defoliation as well as in stand age among sites and in this
study we found that results were significantly improved by the inclusion of an age correction algorithm
for both the multi-spectral (R
2
no age correction
= 0.55 vs R
2
age correction
= 0.73 for Landsat) and hyperspectral
(R
2
no age correction
= 0.12 vs R
2
age correction
= 0.50 for Hyperion) image data. The improved accuracy obtained
from Landsat compared to the Hyperion data illustrates the potential of applying SMA techniques for
analysis of multi-spectral datasets such as MODIS and SPOT-VEGETATION.
© 2010 Elsevier B.V. All rights reserved.
1. Introduction
The amount of foliage is one of the primary physiological
controls of plant functioning, which ultimately influences plant
survival and growth. Repeated severe defoliation events have been
linked to reduced growth rates and tree mortality in softwood plan-
tations (Kurz et al., 2008; Verbesselt et al., 2009) and hardwood
forests (Stone and Coops, 2004) throughout the world. The most
commonly deployed methods of assessing defoliation are forest
health surveys (FHS) including aerial surveillance, drive-through
surveys and ground inspections (Carnegie et al., 2008; Johnson and
Wittwer, 2008). However, FHS requires skilled staff for on-ground
and airborne surveys, diagnostics, analysis and support, and the
∗
Corresponding author. Tel.: +32 16329749; fax: +32 16329760.
E-mail addresses: ben.somers@biw.kuleuven.be (B. Somers),
jan.verbesselt@csiro.au (J. Verbesselt), eva.ampe@student.kuleuven.be
(E.M. Ampe), neil.sims@csiro.au (N. Sims), willem.verstraeten@biw.kuleuven.be
(W.W. Verstraeten), pol.coppin@biw.kuleuven.be (P. Coppin).
accuracy of assessments is dependent upon the skill of the surveyor
(Stone and Coops, 2004). FHS assessments are therefore usually
limited in frequency to once or twice a year in most areas. Incorpo-
rating remote sensing technologies to assist FHS has the potential to
reduce the time and cost of assessment, and provide regular infor-
mation over large areas (van Aardt and Norris-Rogers, 2008; Stone
et al., 2008; Coops et al., 2009; Eklundh et al., 2009).
A number of recent studies have demonstrated the potential
of measuring defoliation from remotely sensed observations of
Eucalyptus crowns (Barry et al., 2008; Pietrzykowski et al., 2008).
These studies use linear regression modeling between field assess-
ments of symptom levels and vegetation indices calculated from
the images to identify the level of crown damage from a range of
damaging agents including fungal infections and insect predation.
Good correlations were found between the expression of damage
symptoms in tree crowns and vegetation index values at a range
of scales from individual crowns to entire estates (Verbesselt et al.,
2009).
Remote sensing methods using vegetation indices are limited,
however, by their dependence on the visibility of leaves in image
0303-2434/$ – see front matter © 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.jag.2010.03.005