![](https://csdnimg.cn/release/download_crawler_static/10932105/bg13.jpg)
t = temporal information, i.e., when, how long, and how
often the data were acquired;
θ = set of angles that describe the geometric relationships
between the radiation source (e.g., the Sun), the terrain target
of interest (e.g., a corn field), and the remote sensing system;
P = polarization of back-scattered energy recorded by the
sensor; and
Ω = radiometric resolution (precision) at which the data
(e.g., reflected, emitted, or back-scattered radiation) are
recorded by the remote sensing system.
It is useful to briefly review characteristics of the parameters
associated with Equation 1 and how they influence the nature
of the remote sensing data collected.
Spectral Information and Resolution
Most remote sensing investigations are based on developing
a deterministic relationship (i.e., a model) between the
amount of electromagnetic energy reflected, emitted, or
back-scattered in specific bands or frequencies and the
chemical, biological, and physical characteristics of the phe-
nomena under investigation (e.g., a corn field canopy). Spec-
tral resolution is the number and dimension (size) of specific
wavelength intervals (referred to as bands or channels) in
the electromagnetic spectrum to which a remote sensing
instrument is sensitive.
Multispectral remote sensing systems record energy in mul-
tiple bands of the electromagnetic spectrum. For example,
in the 1970s and early 1980s, the Landsat Multispectral
Scan- ners (MSS) recorded remotely sensed data of much of
the Earth that is still of significant value for change detection
studies. The bandwidths of the four MSS bands are dis-
played in Figure 6a (band 1 = 500 – 600 nm; band 2 = 600
– 700 nm; band 3 = 700 – 800 nm; and band 4 = 800 – 1,
100 nm). The nominal size of a band may be large (i.e.,
coarse), as with the Landsat MSS near-infrared band 4 (800
– 1,100 nm) or relatively smaller (i.e., finer), as with the
Landsat MSS band 3 (700 – 800 nm). Thus, Landsat MSS
band 4 detectors recorded a relatively large range of
reflected near- infrared radiant flux (300 nm wide) while the
MSS band 3 detectors recorded a much reduced range of
near-infrared radiant flux (100 nm wide).
The four multispectral bandwidths associated with the Posi-
tive Systems ADAR 5500 digital frame camera are shown
for comparative purposes (Figure 6a, c, and d). The cam-
era’s bandwidths were refined to record information in
more specific regions of the spectrum (band 1 = 450 – 515
nm; band 2 = 525 – 605 nm; band 3 = 640 – 690 nm; and
band 4 = 750 – 900 nm). There are gaps between the
spectral sensi- tivities of the detectors. Note that this digital
camera system is also sensitive to reflected blue wavelength
energy.
The aforementioned terminology is typically used to de-
scribe a sensor’s nominal spectral resolution. However, it is
difficult to create a detector that has extremely sharp band-
pass boundaries such as those shown in Figure 6a. Rather,
the more precise method of stating bandwidth is to look at
the typical Gaussian shape of the detector sensitivity, such
as the example shown in Figure 6b. The analyst then deter-
mines the Full Width at Half Maximum (FWHM). In this
hypothetical example, the Landsat MSS near-infrared band
3 under investigation is sensitive to energy between 700
and 800 nm.
A hyperspectral remote sensing instrument typically
acquires data in hundreds of spectral bands (Goetz, 2002).
For example, the Airborne Visible and Infrared Imaging
Spectrometer (AVIRIS) has 224 bands in the region from 400
to 2,500 nm spaced just 10 nm apart based on the FWHM cri-
teria (Clark, 1999; NASA, 2006). An AVIRIS hyperspectral
datacube of a portion of the Savannah River Site near Aiken, SC,
is shown in Figure 7. Ultraspectral remote sensing involves
data collection in many hundreds of bands.
Certain regions or spectral bands of the electromagnetic
spectrum are optimal for obtaining information on biophysi-
cal parameters. The bands are normally selected to maxi-
mize the contrast between the object of interest and its
background (i.e., object-to-background contrast). Careful
selection of the spectral bands might improve the probability
that the desired information will be extracted from the
remote sensor data.
Spatial Information and Resolution
Most remote sensing studies record the spatial attributes of
objects on the terrain. For example, each silver halide crystal
in an analog aerial photograph and each picture element in a
digital remote sensor image is located at a specific location
in the image and associated with specific
x,y coordinates on
the ground. Once rectified to a standard map projection, the
spatial information associated with each silver halide crystal
or pixel is of significant value because it allows the remote
sensing–derived information to be used with other spatial
Remote Sensing of the Environment
14