PHOTOGRAMMETRIC ENGINEERING & REMOTE SENSING
January 2014
43
PHOTOGRAMMETRIC ENGINEERING & REMOTE SENSING
January 2014
43
Abstract
The objective of this study is to investigate the feasibility of
using dual polarization
ENVISAT ASAR
and
ALOS PALSAR
data to
capture temporal and spatial variations of pasture at paddock
scale in Western Australia.
SAR
backscatter was correlated to
NDVI
,
NDWI
, and soil moisture index (
M.I
) to demonstrate the
capability of
SAR
to measure biomass, plant water content,
and soil moisture, respectively. The results proved that
ENVI-
SAT ASAR
and
ALOS PALSAR
are able to monitor pasture, and the
ability varies with sensor parameters (wavelength, polariza-
tion, and incidence angle) and field properties (soil moisture,
vegetation type, biomass level). Paddock biomass and human
activities can be detected; C-band is suitable for grass biomass
and L-band detects water content of grass in drying stage.
HV
(
VH
) has advantage in detecting plant water content, while
VV
is sensitive to soil moisture. Ground measurements of pasture
biomass successfully validated part of the correlation results.
Introduction
Pasture is the primary source of food for grazing animals,
such as cattle and sheep, particularly in regions where pasture
land is unsuitable for any other agricultural production. It
is an important biotic resource on the Earth. Efficient use of
feed resources in the livestock industries is a major factor
determining farm profitability and environment sustainability.
This is very important for growing season, since pasture that
is not utilized before the end of the growing season dries up
with a greatly reduced nutritive value. In the growing season,
typically, grass is rotationally grazed with a paddock shift to
maintain high milk and pasture production. Paddock is small
field of pasture, normally fenced and separated by farmers for
keeping sheep or cattle. Paddocks with higher biomass may
be used to lengthen the grazing interval while paddocks with
lower biomass indicate an impending feed deficit. Herds’
pasture demands are then reduced to allow these deficit pad-
docks more time to grow to the desired mass. Therefore, the
farmers need to know which paddock’s biomass is higher or
in better condition so that they can move herds to that pad-
dock, leaving lower biomass paddocks to recover. Timely
estimates of available feed quantity allow better decisions on
feed budgeting and grazing rotations. However, the conven-
tional technique based on visual scoring and rising plate
Xin Wang, Linlin Ge, and Xiaojing, Li are with the School of
Civil and Environmental Engineering, the University of New
South Wales, Sydney, NSW 2052, Australia (
).
Stephen Gherardi is with the Department of Agriculture and
Food Western Australia, Australia.
meter is labor-intensive, costly, time consuming, requires
some skill, and is limited to single transect (McNeill
et al
.,
2010).
Remote sensing technology can provide temporal and
spatial information on pasture allowing producers to more
effectively manage their enterprise. From the perspective of
optical remote sensing, Normalized Difference Vegetation
Index (
NDVI
) has been used for estimation of grass biomass at
various scales (Tucker
et al
., 1985; Diallo
et al
., 1991; Serrano
et al
., 2000; Schino
et al
., 2003; Flynn, 2006; Xu
et al
., 2008).
NDVI
is particularly useful for retrieval of quantitative proper-
ties of agriculture with highly seasonal agro-ecosystems where
growth commences and continues without limitation until
a point where temperature, moisture, or phenology causes it
to cease (Yang
et al
., 1998; Hill and Donald, 2003). A distinct
growing season is observed in Western Australia, and the
“Pastures from Space” project over there can provide near
real-time biomass information from
MODIS NDVI
but only at
coarser scales (Hill
et al
., 2004; Donald et al., 2010). The finer
scale of biomass monitoring from Landsat
TM
and
SPOT
can
only be reached at very low temporal resolution (Edirisinghe
et al
., 2011).
MODIS
visits frequently (twice per day), but it is
difficult to provide details of paddocks due to coarse resolu-
tion of 250 m (Sakamoto
et al
., 2009; Heller et al., 2012; Liu
et al
., 2012). Landsat
TM
, with spatial resolution of 30 m,
is capable of describing pasture at paddock scale but low
revisit frequency (16 days per cycle) and cloudiness limits
its application. Therefore, satellite
SAR
, advantages of cloud
penetration and higher spatial resolution, can be used with
TM
for dense observation of pasture at paddock scale.
NDVI
is able to represent for biomass, but cannot describe
plant water content accurately. The Normalized Difference
Water Index (
NDWI
) is commonly used and accepted as an
accurate estimate of plant water content by using short wave
infrared (
SWIR
) and near infrared (
NIR
) bands (Gao, 1996).
The combination of the
NIR
with the
SWIR
removes variations
induced by leaf internal structure and leaf dry matter content,
improving the accuracy in retrieving the vegetation water
content (Ceccato
et al
., 2001). This spectral index has been
correlated with ground-based measurements of plant water
content at both the leaf and canopy scales (Penuelas
et al
.,
1993). Based on that correlation, it has been applied to remote
detection of plant water content for crops (Chen
et al
., 2005;
The Feasibility of Using ENVISAT ASAR
and ALOS PALSAR to Monitor Pastures in
Western Australia
Xin Wang, Linlin Ge, Xiaojing, Li, Stephen Gherardi
Photogrammetric Engineering & Remote Sensing
Vol. 80, No. 1, January 2014, pp. 43–57.
0099-1112/14/8001–43/$3.00/0
© 2014 American Society for Photogrammetry
and Remote Sensing
doi: 10.14358/PERS.80.1.43
