Geography Reference
In-Depth Information
wide swath systems could not be relied on for operational
monitoring. For finer resolution SAR systems the same
issue occurs, but here revisit times can be up to 35 days
and so one can only be guaranteed to capture flooding
imagery in the very largest river basins such as the Amazon
which have a monomodal annual flood pulse that lasts for
several months. In the majority of river basins therefore
the chances of imaging a flood with a high resolution SAR
system becomes vanishingly small.
Moreover, for many applications river reaches tend to
be studied at a much smaller scale however and there-
fore do actually require much finer spatial resolutions.
Very fine resolution (
over different spatial and temporal scales, might include
discharge, flow velocity and direction, water volume and
level, flooded area and flood edge. Remote sensing can
provide information about most of these with varying
degrees of accuracy, however discharge and flow veloci-
ties can only be obtained indirectly through integration
with a hydrodynamic model or gauging networks. Also,
water volume and in some cases water level estimation
requires the use of a topographic data set.
Using the microwave region (1 mm-1 m or 300 GHz-
300 MHz) of the electromagnetic spectrum has some
advantages over the visible and infrared. Commonly used
radar bands (Table 6.1) for monitoring flood inunda-
tion processes include L, C, X and Ka with C being the
most widespread and L being generally preferred to map
flooding beneath vegetation given its longer wavelength.
The information that is obtained with microwave systems
is complementary to that retrieved with systems used
in other regions of the spectrum as microwave interac-
tions are governed by different physical parameters. The
amount of microwave energy of a particular wavelength
scattered off an object or feature is affected by its size,
shape, texture and water content (Woodhouse, 2006).
Further advantages of microwaves are that they can pen-
etrate clouds and measurements can be recorded at any
time without relying on background illumination. The
following sections outline the use of passive radiometry,
active radar systems (SAR) and radar interferometry for
5 m) imagery and other types of
remotely sensed data become a prerequisite when moni-
toring and modelling urban areas where most assets at risk
of flooding are located and where the width of individual
streets often determines the ability to model or monitor
flood inundation patterns accurately. Here constellations
of multiple fine resolution SAR systems present a possible
solution. COSMO-SkyMed for instance can get a 3 m
image sequence with a time from request to acquisition
of the first image of 26-50 hours, then subsequent images
at 12-hour intervals. With a system in place such as the
ESA GPOD FAIRE system (http://gpod.eo.esa.int) that
allows SAR images to be available to the user three hours
after acquisition, rapid delivery of fine resolution image
and information is technically feasible and might be a
common form of dissemination in the near future.
This chapter aims to discuss the potential for, and
uncertainties of, SAR imagery to monitor floods and
support hydraulic modelling over a variety of different
scales. The chapter first describes the basic principles of
microwave remote sensing with a view to flood inunda-
tion studies. Then, the potential of SAR image data to
monitor and map flood inundation is outlined and dis-
cussed. Finally, the support these data can offer to flood
inundation modelling is reported via a number of rural
and urban case studies. In all these sections, the uncer-
tainty of the data is discussed and the need to move from
deterministic binary information (e.g. wet/dry maps) to
fuzzy data types (e.g. fusing different interpretations of a
single image or multiple flood images) is highlighted.
<
Table 6.1
Most commonly used parts of the microwave
region of the electromagnetic spectrum (after IEEE).
Wavelength
Bands (Standard Radar
Frequency
(cm)
Nomenclature)
Range (GHz)
30-100
P (UHF*)
0.3-1
15-30
L
1-2
8-15
S
2-4
4-8
C
4-8
2.5-4
X
8-12
1.7-2.5
Ku**
12-18
1.1-1.7
K
18-27
0.75-1.1
Ka**
27-40
0.4-0.75
V
40-75
6.2 Microwave imaging of water
and flooded land surfaces
0.27-0.4
W
75-110
<
0.27 mm 110-300
∗
P band used to be called UHF (ultra high frequency) band
∗∗
Ku and Ka stand for K 'under' (from the German word
'
unter
') and K 'above', respectively
The variables which both scientists and practitioners
involved with flood risk management would like to mea-
sure or estimate during a flood event, and hopefully
Search WWH ::
Custom Search