![[SIR-CED LOGO]](../../LOGO.GIF){kind=logo}
Radar stands for RAdio Detection and Ranging. A radar transmits microwave signals and measures the strength and time delay of the returned energy. The time delay, t, of the echo can be used to determine the range or distance ,R, to an object, since microwaves travel at the speed of light, c, and R = ct.
Imagine you are standing in 'echo valley' and you shout the word 'radar' at the top of your voice. Then you hear an echo ('radar'). How loud is it? How long did it take for the echo to come back to you? These are the properties a radar measures: the strength of an electromagnetic wave can be compared with the loudness of a sound wave; the time delay is the same idea, except that sound waves travel at the speed of sound, not the speed of light.
Pulse Duration
A radar transmits a parcel or packet of photons known as a pulse. The photons
all have the same wavelength (or frequency), which will be in the range 1cm to
1m for radar. The pulse duration (length) is
which is typically around 10 micro-seconds. The transmitted pulse travels at
the speed of light for the medium (3x108 m/s in a vacuum). The power
level of the transmitted pulse is typically around 1 kilo-watt.
In a simple radar system, the radar transmitter generates a high power radar pulse, which is fed to a switch known as a circulator, which directs the transmitted pulse to the radar antenna. Radar antennas are constructed to transmit and receive pulses at particular radar wavelengths. On transmission, the antenna directs the transmitted pulse towards the target area. The energy reflected off the earth's surface comes back as a radar echo, which is then received by the antenna. When receiving, the circulator switch directs the returned echoes into the radar receiver, which converts the echoes to digital numbers. The radar data is then passed to the data recorder, which stores the data on disk or on tape. In imaging radars, the recorded data have to be processed first before they can be displayed as an image. This takes a lot of computations and thus can take a long time.
Note that the radar does not transmit and receive at the same time - the circulator switch is switched rapidly (say 1000 times a second) between the transmit position and the receive position. The switch also serves to protect the receiver from the high-power transmitted pulses, which could blow the receiver.
Much like human vision, a radar system has a Source, a Scatterer, and an Observer. In a radar system, the radar antenna acts as both the Source and the Observer. Look at the diagram of the radar system. The first step in collecting radar images is for the antenna to emit a parcel or packet of photons known as a pulse. When the antenna is emitting a pulse of energy, it is referred to as a transmitter. Energy from this radar pulse that is reflected back, known as backscatter, is collected by the antenna. When the antenna is collecting backscatter, it is referred to as a receiver. The circulator switch alternates the antenna between transmitter mode and receiver mode. The transmitter and the receiver can not operate at the same time because the energies being emitted and collected would interfere with each other.After the energy has gone from Source (Transmitter) to Scatterer (planet surface), and back to the Observer (Receiver), the receiver then converts this energy into digital information and sends it to the data recorder and display. This data is then processed into a radar image of the target area.
some of the energy in the radar pulse is reflected back towards the radar. This is what the radar measures. It is known as radar backscatter. The radar can not measure energy reflected in other directions.
Use a flash camera in your classroom, together with the diagram of a radar system, to show your students how imaging radar is just like a flash camera, operating using radio waves.
As the radar moves along its flight track, it images a swath along the ground. The schematic diagram above shows a radar in the Space Shuttle looking out to the right and down as it flies along. The radar beam footprint is shown as a shaded ellipse. This footprint moves as the radar moves. As it moves along an image is built up of the area in the swath. Each pixel (picture element) in the radar image represents the radar backscatter for that area on the ground. Darker areas in the image represent low backscatter, while brighter areas in represent high backscatter. The area shown is Los Angeles, with the Palos Verdes peninsula near the top of the picture, in the center, and Los Angeles International airport near the top right-hand corner.
Some types of imaging radar are often referred to as Synthetic Aperture Radar's or SAR's for short. SIR-C, for example, is a Synthetic Aperture Radar. This refers to a technique used to synthesize a very long antenna by combining signals (echoes) received by the radar as it moves along a track. A long antenna is desirable since, the longer the antenna, the finer the detail the radar can resolve, and the smaller the objects the radar can 'see'. Aperture refers to the opening used to collect the reflected waves which are used to form an image. In the case of a camera, this would be the shutter opening; for radar it means the antenna. A synthetic aperture can be constructed by moving a real aperture or antenna to different positions. At each position a pulse is transmitted, then the return echoes pass through the receiver and are recorded in an 'echo store'.
Using the flashlight analogy, the bigger the reflecting bowl of the
flashlight, the narrower the beam of light generated. For a flashlight the
aperture is the reflecting bowl. Similarly, larger radar antennas will have
finer beams. The beam width for an antenna (or aperture) of size D is
approximately
/D
in radians, where
is the wavelength at which the antenna operates.
For a 'real' antenna of length D, and a radar of wavelength
,
imaging an object at range R, the smallest resolvable object is of size ~
R/D.
For a 10m antenna, operating at 800 km range at wavelength 25 cm, this gives a
minimum object size of 20 km. With a synthetic aperture of length 20km, a
minimum object size of 5m can be resolved, since the resolution of a synthetic
aperture is half the size of the actual antenna used, i.e. D/2.
Synthetic aperture radar is a technique used to generate radar images in which fine detail can be resolved. To understand what a photograph means, we do not need to understand the process use to grind the lens; in the same way, it is not necessary to understand the synthetic aperture radar technique to understand what a radar image means.
Radar scientists measure backscatter (also known as radar cross section or
RCS) in units of area (i.e. square meters). The radar cross section of an
object is often related to its actual size, as seen by the radar. For example,
the RCS of a sphere is
r2,
where r is the radius of the sphere. Calculating the backscatter for more
complicated geometrical shapes is less straightforward. In fact, many radar
engineers have spent their careers constructing mathematical models which can
be used to predict radar cross section of an object. These models are used to
design aircraft or missiles with the lowest possible radar cross section, so
that they can not be detected by 'enemy' radars.
For looking at the earth, radar scientists often use a measure called
normalized radar cross section,
,
which is given by:
= 10log10 (RCS/A) where A is the area of the image pixel
The units for normalized radar cross section are decibel (dB), which is a kind
of log scale.
has the advantage for radar scientists that it is independent of the
pixel size in the image data. Thus,
measured for a corn field using an image with 100mx100m pixels would be
the same as for an image with 10mx10m pixels.
Example: for an object with
= -10dB, and an image pixel size of 10mx10m, the measured RCS would be
10m2. If the normalized radar cross section,
,
is constant at -10dB over a large area and we look at it with an image pixel
size of 100mx100m, the RCS would be 1000m2.
Typical values of
for natural surfaces are from + 5dB to - 40dB. + 5dB would correspond to a very
bright feature in a radar image, and would mean that a very large fraction of
the radar pulse energy is reflected back towards the radar. - 40dB would be a
very dark feature in a radar image, and would mean that very little of the
energy in the radar pulse is reflected back towards the radar. Those students
who cannot grasp the log derivation of
,
can still learn to use the dB values to separate features.
With the capability to use different polarizations on transmit and receive, we can measure the normalized radar cross section for combinations of polarizations:
- Horizontal transmit, Horizontal receive
- Vertical transmit, Vertical receive
- Horizontal transmit, Vertical receive
Total power (TP) - defined as 0.25(
+ 2
+
)
- Phase difference between the HH and VV
measurements
- Correlation between HH and VV measurements
When we transmit a radar pulse and look at the reflected echo, we can measure
the phase of the return as well as its amplitude. The phase is related to the
time it takes for the pulse to go out and back. Every time the pulse travels a
distance of one wavelength, the phase changes by 2
(look at a sine wave to see this for yourself). In radar, the phase difference
between the HH and VV returns is often significant, as will be seen later. The
phase difference can be thought of as a difference in round-trip time for the
HH and VV returns. The phase difference,
,
is measured in degrees (i.e. not dB) and is therefore a number between -180 and
+180 degrees.
Another significant parameter is the correlation coefficient,
,
between the HH and VV returns. It is a measure of how alike the HH and VV
scatterers are, and is given as a number between 0 and 1, or as a percentage (0
to 100%).
= 0 means the HH and VV scatterers are completely different.
= 1 means the HH and VV scatterers are the same. Values in between 0 and 1 mean
that there is a mixture of identical and different HH and VV scatterers.
Images of each of the different polarization measurements can be displayed using the program Sigma0. This can be demonstrated with any of the radar image data (Filename.stk files) provided. After selecting an AIRSAR or SIR-C image for display, the various polarizations come up on a menu for selection. Once a selection has been made, and an image displayed, select an area with the mouse. Then execute the Mean and Std Dev option under the Statistics menu to display the mean values for the polarization measurements listed above for the selected area.
Backscatter can change if we use a different polarization, wavelength, time of observation, or observation angle. Different polarizations can often be used to distinguish between different types of scattering, as will be seen later. For example, smooth surfaces, rough surfaces and vegetation layers all give different signatures if the various polarizations are examined. Using different wavelengths can often give sensitivity to different sizes of structure. For example, when looking at a forest, the radar cross section at shorter wavelengths, e.g. C-band (6cm), would be sensitive to the leaves at the top of the canopy. Longer wavelengths, e.g. L-band (24cm), would be sensitive to the branches, and very long wavelengths, e.g. P-band (68cm), would be sensitive to the trunks of the trees.
Because radar sensors can always acquire an image, independent of cloud cover or whether it is day or night, they are very useful in monitoring change over time. When looking at the earth's surface, radar images are sensitive to changes in structure and in water content. For example, changes in ocean wave direction, crop growth, harvesting, deforestation, leaf state (on or off), are all structural changes that have been noted in radar images. Urban growth should also be very easy to detect in radar images. Frozen versus thawed trees, dry versus wet vegetation, dry versus wet snow and dry versus wet soil are all examples of moisture content changes that have been observed in radar images.
i) Track Angle
Man-made features, such as urban areas and fences, terrain with significant height variation, crops with strong row directions, ocean waves all demonstrate strong dependence on track angle in their radar backscatter behavior. Natural vegetation in relatively flat areas does not, since it tends to grow equally in all directions.
The second angle is the incidence angle (angle of radar wave at earth's
surface, compared with local vertical). Most types of scatterers show
significant variation in backscatter versus incidence angle. If the radar is
pointing in a direction at right angles to the surface (e.g. straight down for
a flat surface, with
i
= 0), the backscatter will be very strong. As the incidence angle
increases, generally the backscatter will decrease.
ii) Incidence Angle
Note that by Pythagoras cos(
)
= h / R, where h is the height of the spacecraft and R is the distance from the
radar to the object (the range ) .
Another problem which occurs commonly in radar images is layover, which causes mountains to appear as if they have 'fallen over'. This phenomenon occurs because imaging radar sensors measure how far away a scatterer is from the radar, then map the results onto some nominal ground plane. Thus, the top of a mountain may be closer to the radar than the bottom, and will appear nearer to the radar in the final image. Often the side of the mountain facing the radar will give very high backscatter, because of the low incidence angle. This is also illustrated in the graphic above. Radar shadow and layover are visible in many radar images: an example is the SEASAT radar image of Los Angeles seen in LASEASAT.pic from Module 2. With a little practice, it becomes easy to recognize mountainous areas in radar images from the characteristic bright leading edge of the layover and dark shadows.
Imaging radars operate by transmitting a series of radar pulses, then measuring the strength of the returned echo. They measure radar backscatter, which can be thought of as the amount of pulse energy reflected back towards the radar by an object. A radar image, then, is a two-dimensional representation of the radar backscatter of the surface being imaged. The question is, how to interpret the measurements?
Radar backscatter is primarily a function of the physical structure of the surface being imaged. This means that backscatter changes as the surface topography (relief) and roughness vary across the target area. Radar backscatter is also sensitive to the electrical properties of the target area.
If a radar is pointing almost straight down towards the target surface then it will reflect strongly because more surface area will be "EXPOSED" to the radar beam and the backscatter will be high. Man-made objects, such a bridges, buildings and fences can appear very bright in a radar image when viewed from the correct angle.
Radar waves will penetrate very dry surfaces, such as the sands of the Sahara Desert, to give us a glimpse of the underlying rocks or buried river channels
Here is a summary, indicating potential explanations for high and low backscatter in a radar image:
| Backscatter range | Possible Explanation |
|---|---|
| Very high backscatter (above -5dB) |
|
| High backscatter (-10 to 0dB) |
|
| Moderate backscatter (-20 to -10dB) |
|
| Low backscatter (below -20dB) |
|
These are, of course, just rules-of-thumb, to which many exceptions are possible and have been found by scientists analyzing radar image data. There is often considerable overlap between the backscatter ranges, so it might be difficult to differentiate between forest and shrubs, using backscatter alone.
| Teacher's Guide - Table of Contents |
Converted to the IBM-PC by Al Wong, sirced03@southport.jpl.nasa.gov
Jet Propulsion Laboratory
4800 Oak Grove Drive
Pasadena, CA 91109