SRTM Radar Penetration
Why were international partners included in this mission?
Space exploration is becoming an international effort and it is to every nation's advantage to collaborate on missions to study Earth, the solar systems and the universe. The international partnership with the German and Italian space agencies resulted in the addition of the X-band radar system to the mission. Flying X-SAR added high-resolution data that would not have been collected otherwise.
How many orbits a day did the space shuttle make?
The space shuttle orbited Earth 16 times each day. During the 11-day mission, Space Shuttle Endeavour, carrying the SRTM payload, completed 176 orbits of Earth, flying tail forward at 7.5 km/sec (17,000 mph).
How did the development of SRTM relate to Mission to Planet Earth and the long-term plans of NASA for environmental monitoring missions?
The Shuttle Radar Topography Mission provides important information for NASA's Earth Sciences Enterprise, which is dedicated to understanding the total Earth system and the effects of human activity on the global environment. Data will give scientists a better understanding of natural systems and a more reliable method for predicting changes in Earth's atmosphere, land and sea that are being brought on by natural events and
Topographic data are critical to the accuracy of these computer models and global changes in climate, land formations, sea-surface heights and atmospheric changes. That is because Earth's shape determines the flow of air, water and ice, and the spatial pattern of all life, including people. Topographic data also give scientists clues about the underlying structures of Earth, including its tectonic activity.
The data have a variety of uses in scientific disciplines ranging from hydrology, geology and archaeology to ecology and studies of urban development and its impact on the environment.
Some of the radar mission's civilian and commercial uses benefit the transportation industry, as well as the communications and information technologies markets. In telecommunications, wireless service providers and operators are particularly interested in this digital elevation data. Topographic data can be used for building better transceiver stations, identifying the best geographic locations for cellular
telephone towers and having the best terrain data for planning and construction. In fact, just about any industry that requires accurate digital elevation and topographic data stands to benefit from this mission.
Can you describe some of the field work done to support these orbital observations? How many scientists were involved in field work? What types of activities did they carry out?
Geodetic surveying was the main type of field work. Scientists used a method called Kinematic Global Positioning System surveying. This method facilitates the very rapid long lines of precise positions from a moving vehicle. The actual survey work was conducted by several entities, including private contractors, NGA geodesists and JPL scientists. In all, about 70,000 kilometers of survey lines were collected in support of this mission. The data were used to model long-wavelength error sources in the data.
In addition, JPL deployed corner reflectors during the mission. These are highly reflective structures that appear as a bright point in the radar image. These reflectors deployed with precisely measured coordinates, served as control points in the Shuttle Radar Topography Mission data.
Why not use commercial satellites to collect the data?
Optical satellites have problems collecting two clear images of many sites because of cloud cover and darkness. Currently operating radar satellites were not designed to collect this type of data. Critical information such as their exact location is not available. Ground control points can be used to get around some problems, but
the SRTM required no ground control, which was an advantage in inaccessible regions. The use of the new technology of single-pass radar interferometry was experimental and involved some risk because it tested technologies never used before to measure the mast and shuttle attitude. But part of NASA's charter is to experiment and test flight technologies in space for future use on
commercial satellites or transfer to the commercial sector.
How did this mission differ from Space Radar Laboratory, which flew in 1994? What's new about SRTM?
The heart of the SRTM radar was the SIR-C/X-SAR radar, which flew twice on the space shuttle in 1994. Several modifications were made, which gave the SRTM system new capabilities compared with the SIR-C/X-SAR. The major changes were the addition of C-band and X-band antennas at the end of the 60-meter (200-foot) mast. These secondary, or "outboard" antennas, allowed the radar to use a technique called interferometry to map the elevation of the terrain in a single pass, which was not possible with SIR-C/X-SAR. Interferometry can be likened to a person dropping two pebbles into a puddle of water and watching the ripples, or concentric circles of water emanating outward from the splash, meet and interfere with each other. Those interference patterns caused by the rippling water from the two pebbles are measured by the radar systems onboard the shuttle to acquire topographic data. The main antenna on the shuttle and the outboard antenna on the tip of the mast bounced radar signals off Earth simultaneously, and then retrieved "backscattered" radar data as the signals from both antennas scattered and began to interfere with each other.
The design of the SRTM mission was also different from SIR-C/X-SAR. Instead of focusing on a limited number of "supersite" targets for repeated viewing, as was done with SIR-C/X-SAR, SRTM was designed to map as much of the land surface as possible. SRTM covered all of the land surface between 60 degrees north and 56 degrees south latitudes. The SIR-C/X-SAR covered less than 30 percent of the Earth's land area.
All this talk about
interferometry and metrology (the science of weights and measures) on SRTM sounds great. What other applications are there?
The techniques of interferometry and metrology are used in a wide variety of applications in industry and the sciences. Some particularly useful applications of long baseline interferometry in the space sciences include the use of very-long baseline radio and laser interferometry to study black holes and other objects of astrophysical interest. Some spaceborne examples of this include
LISA. And if SRTM's goal of mapping most of the Earth isn't ambitious enough, there are many upcoming JPL projects in
Optical and Infrared Interferometry. The
ultimate goal is to someday use huge spaceborne infrared
interferometers to image Earth-like planets in other solar systems.
How much did the SRTM payload weigh?
The SRTM payload weighed 13,600 kg (29,920 lbs). For comparison, an
adult male African elephant can weigh more than 6,000 kilograms (13,200
pounds), so the payload weighed about as much as 2.3 elephants.
How big was the mast?
The SRTM inboard and outboard antennas were separated by a mast which
was 60 meters (200 feet) long. From the tip of the shuttle's opposite wing
to the edge of the outboard antenna, the shuttle and the SRTM instrument
together measured 83 meters (272 feet). It is the largest rigid structure
ever flown in space.
How was the mast stowed?
So that it could be stowed inside a canister for take-off and landing,
the SRTM mast collapsed from 60 meters to 3 meters (200 feet to 10 feet),
a 20-to-1 compression ratio. That would be like shrinking Shaquille
O'Neal, who is 7 feet, 1-inch tall, down to about 4.25 inches. To
accomplish this, the structure unlatched and spiraled into the canister
along with all the wires running to the outboard antennas.
Why did you need to measure the length and orientation of the mast?
Wasn't it a rigid structure?
It's true the mast was surprisingly rigid given its length. In fact, once
it was deployed, we didn't expect the mast tip to move by more than 15
centimeters (5.9 inches) for the rest of the mission. Unfortunately, that
much motion is very significant for the SRTM height measurements. For example,
an error in our knowledge of the mast tip position in the "worst" (baseline
roll) axis of only 3 millimeters (1/8 inch) would result in a height
error of about 9 meters (29.7 feet). The reason the mast moves is due to
many factors, including shuttle thruster firings, astronaut activity and
thermal distortions as the Space Shuttle moves around the Earth in orbit,
going in and out of sunlight.
What if the mast didn't deploy correctly?
A number of
corrective strategies were planned to ensure that the mast deployed
properly, including a space walk to manually crank the mast in or out. If
these strategies had failed, the mast would have been retracted and stowed
for return to Earth. If it couldn't be stowed properly, it would have been
Can C-band and X-band radar operate at the same time? What are the
advantages or disadvantages of this?
The C-band and X-band radar can operate simultaneously or
independently. For the most part during the mission, they operated
together. The only disadvantage was that joint operations consume
more power, but since the coverage of sites with both frequencies was
desirable, the basic plan took this power usage into account.
How many hours of total coverage did the radar take?
The radar systems collected data during 159 orbits. The systems were
turned on and off as the data plan dictated.
Why was the L-band antenna not used for SRTM?
The payload weighed too much; so the L-band panels had to be removed.
Why were C-band X-band used? Why didn't you use only one? I
realize that two radar images have to be combined, but those two images
are from the outboard and inboard antennas, I thought. Please elaborate on
The X-band system was supplied by the German and Italian space agencies
and was flown in 1994 aboard SIR-C/X-SAR. As a continuation of this
international cooperation, NASA invited them to contribute a second
antenna, to form an X-band single-pass interferometer. Unlike the C-band
system, the X-band couldn't steer its beam, so it couldn't operate in
ScanSAR mode and therefore couldn't get full coverage of the earth. Its 50
kilometer swath offered nearly complete coverage at high latitudes,
though. Because it doesn't scan, the X-band system also had a little
better vertical error- probably around 5 meters.
What was the radiation level from SRTM at ground level?
The radars produced about 10,000 watts, about the same as 15 microwave
ovens. This power was spread out over an oval area about 6 X 20 kilometers
(120 square kilometers or 30 square miles). The power measured on the
ground was about one 50,000th the level of a radio station and about one
ten millionth the level of a cell phone. In addition, SRTM only
illuminated a given area for about 1/10 of a second.
Were any special precautions taken to ensure the safety of the
The microwaves were emitted in a very narrow beam, which never came
close to the orbiter. Stray microwave levels were far below levels posing
a danger to people or equipment in the orbiter. No special precautions
were required during the flight.
I read that the SRTM mapped over all land surfaces between 60
degrees North latitude and 56 degrees South latitude. What about the
oceans? Was the radar transmitted over the oceans?
This was a topography mission, so the oceans weren't included. The
instrument was turned on over the ocean from time to time near the coasts,
but only for reference purposes.
The SRTM used "single-pass interferometry" for the first time. After
one orbit, was the mission completed? If not, how many times did the
shuttle orbit the earth for the information?
The radar cut a swath along the ground as the shuttle flew. It took 159
orbits to map the earth.
What about the data from the mission? Were they stored on board
the shuttle or sent back to ground control?
The data were stored on board the shuttle. However, we sent a small amount of
data to the ground during the flight to allow monitoring of the end-to-end
system. Some of these data were released during the flight.
What determined whether data were downlinked live or recorded on the
All data was recorded onboard the shuttle. Appropriate communications
links were available about once per day to downlink the data. This
once-a-day opportunity was shared by both the X-band and C-band, and data
was played back in real time if it was collected at that time, or
tape-recorded data was played back. The C-band and X-band shared the
link time according to mutually agreed on priorities established during
mission planning. The downlinked data was used to verify the performance
of the sensor.
How long did it take to playback recorded data?
It took four times as long to play back C-band data as it took to
record it. X-band data were played back at half the rate they were recorded.
Why couldn't all the data be downlinked live?
The shuttle was able to transmit at 50 million bits per second
through NASA's Tracking and Data Relay Satellites (TDRS) to the White
Sands, New Mexico, station. The radars produced data six times
faster. Relay link time was scheduled and limited by satellite
position and the priorities of other customers.
Did the C-band always record four channels of data? What was the
purpose of this?
Two channels were required to obtain the full 225-kilometer C-band
swath from each of the two antennas, making a total of four channels. Two
channels, one for each of the X-band antennas, was used for X-band.
How did the X-SAR data compare to the C-band data?
The X-SAR data are slightly higher resolution than the C-band's, but
there are gaps in the coverage. The size of the X-SAR swath was 1/4
that of C-band.
Are X-band data processed the same way as C-band data?
X-band and C-band data are processed in a similar way from a
mathematics point of view. The basic computer processes are the same.
There are differences in the type of processing equipment and location of
the processing equipment. They each produce similarly formatted data
Why was the absolute horizonal accuracy for the Level-2 posts only +/-20 meters?
Doesn't the Global Positioning System provide better accuracy? Is it related to
The horizontal accuracy of SRTM posts is only weakly related to the
accuracy of GPS. Many other factors are involved, including the accuracy
of our measurements of the mast length and orientation (the biggest error
source), timing errors, multipath, phase measurement errors, and thermal
noise in the radar system. We pushed the radar hardware we "inherited"
from SIR-C/X-SAR to its limits in order to get the best horizontal and
vertical errors possible.
Are the data sets provided in the standard USGS format?
The USGS data set are reformatted to be consistent with the other Digital Elevation Model
data sets they have.
How big (MBytes) is one image, and what is the availability of radar
each scattering coefficient data?
The data products from the SRTM are in the form of mosaics of image strips rather
than individual image frames. The US Geological Survey Eros Data Center
distribute the data, but we haven't decided yet how to segment
the mosaics for distribution. Probably it will be something like 5 deg
latitude x 5 deg longtitude. The current plan is to produce a publicly available Digital
Elevation Model at 3 arc seconds (about 90-meters) resolution and 2 image mosaics, possibly at
the full 30 m resolution. The image mosaics would represent ascending
passes and descending passes and would therefore have illumination from
opposite sides. We are not planning on rigorously calibrating the image
data, but we will try to characterize it during processing. In addition,
the individual strip image data may be made available, but we have not
decided how to do that.
Will NASA (EDC?) produce the 5-degree Terrain Height Data Sets for
We're still working out the details of who does what part of the
processing, but it's probable that JPL will produce the final 5 degree
tiles for NGA.
The SRTM data files have names like "N34W119.hgt". What do the
letters and numbers refer to, and what is ".hgt" format?
Each data file covers a one-degree-of-latitude by one-degree-of-longitude
block of Earth's surface. The first seven characters indicate the southwest
corner of the block, with N, S, E, and W referring to north, south, east,
and west. Thus, the "N34W119.hgt" file covers latitudes 34 to 35 North and
longitudes 118-119 West (this file includes downtown Los Angeles,
California). The filename extension ".hgt" simply stands for the word
"height", meaning elevation. It is NOT a format type. These files are
in "raw" format (no headers and not compressed), 16-bit signed integers,
elevation measured in meters above sea level, in a "geographic" (latitude
and longitude array) projection, with data voids indicated by -32768.
International 3-arc-second files have 1201 columns and 1201 rows of data,
with a total filesize of 2,884,802 bytes ( = 1201 x 1201 x 2). United
States 1-arc-second files have 3601 columns and 3601 rows of data, with a
total filesize of 25,934,402 bytes ( = 3601 x 3601 x 2). For more
information read the text file "SRTM_Topo.txt" at
Did the radar sample the tops of trees or the ground level?
The radar does not "see" through thick vegetation canopies. It probably
penetrated a little way into some canopies, but in general
it followed near the top of the canopy.
Did the radar signal bounce off treetops, or topography, or some
combination of both that will provide separate data sets (geodesists like
myself care about topography, whereas scientists more interested in
forestry care about the height of the canopy).
Unfortunately, the wavelength used, 5.6
centimeters, didn't penetrate vegetation very well. That means, for
moderate-heavy vegetation, we mapped near the canopy top. We did penetrate
a little, as some studies comparing our technique with laser altimeters
showed, but not to the ground. If the vegetation was sparse, or had no
leaves, we might get a return from the ground. The Vegetation Canopy
Lidar, scheduled to fly as part of the Earth Observing System, will have
this capability, which may provide some interesting data-set
SRTM viewed Earth from orbit, flying above the clouds. How was
SRTM able to get a clear view of Earth's surface even in cloudy tropical
SRTM used imaging radar technology to view Earth's surface. The behavior
of non-visible radiation like radar is complex and unfamiliar to most
people, but in short, the answer to this question is "radar looks through
A longer answer is as follows. Radar instruments transmit and
receive microwave radiation, which is part of the electromagnetic
spectrum, as is natural visible light. Electromagnetic energy travels in
waves and its "spectrum" refers to its range of wavelengths. Visible
light has a wavelength range of about 0.4 to 0.7 micrometers
(millionths of a meter) as it varies from blue, through cyan, green, and
yellow, to red. Ultraviolet has shorter wavelengths and is invisible to
our eyes. Infrared has longer wavelengths and is also invisible to our
eyes. Microwaves have much longer wavelengths that are far beyond the
range of visible wavelengths. SRTM used radar with a wavelength of 5.6
centimeters (about 2.3 inches). Given that background, here now is the
key point. In general, particles smaller than 1/4 to 1/2 of the
wavelength of radiation do not reflect that radiation but instead allow
the radiation to pass by. Cloud droplets are far smaller than the 5.6
centimeter wavelength of the SRTM radar signal (almost always less than
0.5 centimeters, even for falling raindrops), so the SRTM radar signal
simply "looked through" clouds and rain. (Note: Weather radars, often
seen on television weather reports, are designed to map rainfall, so
(unlike SRTM) they use radar wavelengths short enough to reflect off