Geoscience Reference
In-Depth Information
Based on vegetation indices, biophysical parameters of the vegetation canopy can
be estimated from optical/near-infrared remote sensing. The leaf-area index (LAI)
of a vegetation canopy is defi ned as the ratio of the area of all leaves to that of the
ground surface. LAI can be estimated from remote sensing by applying radiative
transfer modelling techniques. Another biophysical parameter that can be derived
from Earth Observation is the fraction of absorbed photosynthetically active radia-
tion (fAPAR), which describes the amount of radiation absorbed by plants during
photosynthesis.
Synthetic Aperture Radar (SAR)
SAR is an active remote sensing technique in the microwave domain. Electromag-
netic energy pulses are transmitted from the satellite to the surface. The pulse is
then scattered by the target and a certain fraction of it is scattered back in the direc-
tion of the satellite and received by the sensor. The backscatter intensity depends
both on the viewing geometry and target characteristics, particularly the water
content and the structure (dielectric constant, orientation distribution and number
density of the scattering elements, e.g., branches and leaves). SAR signals can be
transmitted and received at defi ned polarisations, either horizontal (H) or vertical
(V). In the case of a fully polarimetric SAR, the instrument can thus record hori-
zontally transmitted and horizontally received (HH) backscatter as well as VV, HV
and VH backscatter. Since the backscatter intensity at these different polarisations
varies as a function of target structure and viewing geometry a fully polarimetric
system can be used to estimate fractions of backscatter caused by fundamental scat-
tering mechanisms: rough surface scattering from the ground, double-bounce scat-
tering (e.g., trunk/ground interactions) and volume scattering (e.g., from multiple
scattering in a tree crown). The contributions of these three basic scattering mecha-
nisms provide information on the properties of the imaged target, e.g., how rough
the ground surface is, whether the microwaves penetrate the canopy and are scat-
tered from the stems or whether most of the radiation is scattered from a dense
canopy. The wavelength of the SAR plays an important role in determining which
scattering elements contribute most to the signal. At longer wavelengths like L- or
P-band branches and tree trunks contribute more to the backscatter, while at shorter
wavelengths like X- or C-band, the leaves and needles are important scattering ele-
ments (fi gure 19.1). This means that multi-wavelength observations can give impor-
tant structural information about the target.
If two SAR acquisitions were taken at a suitable spatial baseline, the complex
correlation coeffi cient between the two SAR acquisitions can be calculated pixel by
pixel. This technique is called SAR interferometry because the resulting interfero-
gram image shows characteristic fringe patterns caused by the interference of the
electromagnetic waves from the two acquisitions. The interferometric coherence is
the magnitude of the complex correlation coeffi cient between the two SAR images
and indicates how similar the SAR signals are. The interferometric phase is an
angular measurement and is related to the vertical height of the scattering phase
centre, the location from which most of the signal originates. The scattering phase
centre is the integral of the returns from a large ensemble of scatterers, which include
ground, stems, branches, and leaves or needles. Which types of scatterers interact
most strongly with the radar wave depends on the wavelength, polarization, inci-
dence angle and vegetation density. Interferometric techniques can be used for esti-
