Geography Reference
In-Depth Information
and L-band (SEASAT-1, JERS-1, ALOS PALSAR, SIR-A/
B/C/X-SAR), their potential for deriving flood parameters
in complex and small-scaled scenarios is clearly limited.
Since 2007, the successful launch of polar orbiting
platforms including TerraSAR-X, RADARSAT-2 (fine
beam), a constellation of four COSMO-SkyMed satellites
and more recently TanDEM-X marks a new generation
of fine resolution SAR systems suitable for flood
monitoring purposes. These satellites provide data up to
1 m pixel spacing, thus permitting information retrieval
about detailed hydrological parameters from space in a
near-real time mode. These recent efforts are evidence
for the ever growing consensus among space agencies to
strengthen support for such high resolution SAR imagery.
Before reporting notable case studies on flood mon-
itoring with SAR imagery and integration with flood
inundation models, the next section provides a brief
outline of several studies that have employed SAR inter-
ferometry, which involves mapping change from coherent
SAR image pairs, to investigate water level variations on
river floodplains and wetlands over space and time.
(DEMs), which can be in the order of (many) metres (see
SRTM DEM validation studies by e.g. Farr et al. (2007),
Rodriguez et al. (2005), Sun et al. (2003) and Bamler et al.
(2003)). Furthermore, any unwanted variation in delay
of the returned signal (e.g. caused by temporal and spa-
tial variations in atmospheric water vapour content) can
lead to significant height and displacement errors (Smith,
2002). For the global DEM produced from elevation data
acquired during the Shuttle Radar Topography Mission
(SRTM-DEM) at 3 arc second (
∼
90 m) spatial resolution,
average global height accuracies vary between 5 and 9 m
(Farr et al., 2007). This vertical error has been shown to
be correlated with topographic relief with large errors and
data voids over high-relief terrain while in the low-relief
sites errors are smaller but still affected by hilly terrain
(Falorni et al., 2005).
Despite these limitations, topography has been mapped
successfully with InSAR technology. For instance, an
InSAR DSM of the UK was produced using repeat pass
InSAR techniques applied to ERS-2 satellite data in the
LandMap project (Muller, 2000). This has a height stan-
dard deviation of
11m and a spatial resolution of 25 m.
The main airborne InSAR is the InterMap STAR-3i.
This is a single-pass across-track X-band SAR interfer-
ometer on board a Learjet 36, with the two antennae
separated by a baseline of 1 m. In the NextMap Britain
project in 2002/3, an accurate high resolution DSM of the
whole of Britain was built up containing over 8 billion ele-
vation points (Mason et al., 2011). This meant that for the
first time there was a national height database with height
accuracies better than
±
6.2.3 SARinterferometry
Interferometry, of which Bamler et al. (1998) provide a
thorough technical review, requires two SAR images from
slightly different viewing geometries. Co-registration of
the two images to a sub-pixel accuracy and subtrac-
tion of the complex phase (time delay) and amplitude
(intensity) for each SAR image pixel allows changes in
surface topography or displacements to be mapped. The
value of the resulting interferometric phase at each pixel
varies between -
1m and spatial resolutions of
5 m (10 m) in urban(rural) areas (www.intermap.com).
Using in-house software, Intermap is able to filter the
digital surface model (DSM) to strip away features such
as trees and buildings to generate a bare-earth digital
terrain model (DTM).
For changes in water level retrieval with InSAR tech-
nology, the specular reflection of smooth open water that
causes most of the return signal to be reflected away from
the antenna and the roughening of the surface (by e.g.
wind or wavelength properties) result in complete loss
of temporal coherence between SAR images acquired at
different times, rendering interferometric retrieval diffi-
cult if not impossible (Alsdorf et al., 2002). However, for
inundated floodplains where there is emergent vegeta-
tion Alsdorf et al. (2000, 2001) show that it is possible
to obtain reliable interferometric phase signatures of
water stage changes (at centimetre scale) from the double
±
and is primarily a function
of the distance between the radar antenna positions (or
baseline) during acquisition, topographic relief, surface
displacement, and the degree of correlation between the
individual scattering elements that comprise each pixel
location, i.e. coherence (Alsdorf et al., 2002). Moreover, it
is worth noting that displacements are determined solely
from interferometric phase (measured to half a wave-
length) but require a reliable reference to a dataset where
no motion has occurred (i.e. reference to 'permanent scat-
terers' (Perissin, 2006). Measurement of surface elevation
is also controlled by baseline effects, which are very dif-
ficult to determine accurately. Thus, large uncertainties
and inaccuracies may be introduced to interferomet-
ric SAR (or InSAR) derived digital elevation models
π
and
+π
Search WWH ::
Custom Search