Geography Reference
In-Depth Information
some additional applications, and identifying future
challenges and research needs.
limitations were introduced (i.e., where are the weak
links in the image chain)'. This approach thus provides
a convenient, insightful framework for examining the
various radiative transfer mechanisms operating in fluvial
environments, the measurement of reflected solar energy
by remote detectors, and the conversion of image data
into information on channel attributes. We thus begin
our discussion by introducing some physical quantities
used to characterise electromagnetic energy. We then use
these quantities to describe the chain of processes by
which sunlight interacts with stream channels to produce
images of rivers.
3.2 An overview of radiative transfer in
shallow stream channels
In essence, passive optical remote sensing of rivers
involves measuring energy -
electromagnetic radiation
that originated from the sun, propagated through the
Earth's atmosphere, and interacted with the fluvial envi-
ronment in various ways. Some proportion of the solar
energy incident upon a river channel is reflected upward
and travels through the atmosphere to an air- or space-
borne sensor, where a light-sensitive detector records
the amount of energy received. Multispectral or hyper-
spectral instruments partition this energy into several,
distinct ranges of wavelengths, called bands, and thus
provide information on the distribution of energy across
the spectrum. For remote sensing of rivers, optical data
spanning the visible and near-infrared wavelengths from
400 nm to approximately 1100 nm (1 nm
3.2.1 Quantifyingthe lightfield
Electromagnetic radiation is one of the dominant forms
by which energy is transferred through the environment;
measurement of this radiation is the basis for remote
sensing. Elementary particles of light called
photons
travel
as waves, each moving at the speed of light and having
aspecific
wavelength
(distance between two consecutive
peaks or troughs) and corresponding
frequency
(number
of waves traveling past a fixed point in one second).
Wavelength (
10
−
9
m)
are most useful; longer wavelengths are strongly absorbed
by purewater. In any case, measurements of reflected solar
energy are the basic data from which river information is
derived. It is important to note, however, that although
remote sensors record the total amount of energy incident
upon them, they cannot distinguish whence that energy
came - the river channel of interest, adjacent terrestrial
features, or the atmosphere. Consequently, only a portion
of the at-sensor signal is directly related to the channel
characteristics of interest, and some understanding of the
various ways in which electromagnetic energy interacts
with riverine environments is needed to decompose and
interpret this composite signal.
A useful concept in this context is that of the
image
chain
, articulated by Schott (1997). The image chain
approach involves viewing the remote sensing process as
a series of interdependent, related events or steps that
lead to a final output - an image of a river or some data
product derived therefrom. Importantly, because each
link along the image chain results from previous links
and influences subsequent links, this approach provides a
means of identifying potential limiting factors. As stated
by Schott (1997, p. 14), 'if we study and understand the
chain of events associated with a particular output image
or product, we will be better able to understand what
we have (i.e., what the product tells us or means), what
we don't have (i.e., the limitations or confidence levels
associated with the output product), and where those
=
1
×
,ins
−
1
or Hertz), and
λ
, in m), frequency (
ν
the speed of light
c
are related as
c
= λν
(3.1)
10
8
m/s in a vacuum.
The amount of
radiant energy
associated with a single
photon, denoted by
q
and expressed inunits of joules (J), is
directly proportional to the frequency at which the photon
oscillates and inversely proportional to wavelength:
where
c
has a value of 2
.
9979
×
h
c
λ
q
=
h
ν =
(3.2)
where
h
=
6
.
6256
×
10
−
34
J
·
s is Planck's constant. The
important implication of these expressions is that
shorter-wavelength photons, such as those comprising
blue light, have a higher frequency and thus a greater
amount of energy than longer wavelength, lower-
frequency photons in the near-infrared. The total energy
Q
encompassed by a beam of light is a function of the
number of photons present and their wavelengths, or,
equivalently, frequencies:
q
i
=
n
i
h
c
Q
=
λ
i
=
n
i
h
ν
i
(3.3)
i
=
1
i
=
1
where the summation is over all the wavelengths of light
present in the beam and
n
i
is the number of photons of
each wavelength.
Search WWH ::
Custom Search