Peer-Reviewed Articles
237 The SRTM Data Finishing Process and Products
James A. Slater, Graham Garvey, Carolyn Johnston, Jeffrey
Haase, Barry Heady, George Kroenung, and James Little
Abstract Download
Full Article
The Shuttle Radar Topography Mission (SRTM) successfully
acquired terrain elevation data for 80 percent of the Earth’s
landmass in February 2000. The radar system and data
collection scheme designed by NASA’s Jet Propulsion Laboratory (JPL) met the global requirements of the U.S. Department
of Defense for Level 2 Digital Terrain Elevation Data (DTED®).
JPL processed the raw data into unfinished DTED® 2 and
other products that were delivered to two contractors of the
National Geospatial-Intelligence Agency. The contractors
edited the unfinished DTED® 2, updated the associated
products, and generated finished products for distribution.
Automated processes were developed by each contractor to
identify, delineate and set heights for lakes, rivers, and
ocean coastlines in conformance with an extensive set of
editing rules created to maintain consistency and uniformity
in the final products. The finished DTED® is significantly
better than the 16 m vertical accuracy required by the
original specification.
249 A Global Assessment of the SRTM Performance
Ernesto Rodríguez, Charles S. Morris, and J. Eric Belz
Abstract Download
Full Article
The NASA/NGA Shuttle Radar Topography Mission (SRTM)
collected interferometric radar data which has been used by
the Jet Propulsion Laboratory to generate a near-global
topography data product for latitudes smaller than 60°. One
of the primary goals of the mission was to produce a data
set that was globally consistent and with quantified errors.
To achieve this goal, an extensive global ground campaign
was conducted by NGA and NASA to collect ground truth that
would allow for the global validation of this unique data set.
This paper documents the results of this SRTM validation
effort using this global data set. The table shown below
summarizes our results (all quantities represent 90 percent
errors in meters).
| Africa | Australia | Eurasia | Islands | N. America | S. America | |
| Absolute Geolocation Error | 11.9 | 7.2 | 8.8 | 9.0 | 12.6 | 9.0 |
| Absolute Height Error | 5.6 | 6.0 | 6.2 | 8.0 | 9.0 | 6.2 |
| Relative Height Error | 9.8 | 4.7 | 8.7 | 6.2 | 7.0 | 5.5 |
In the paper, we present a detailed description of how the results in this table were obtained. We also present detailed characterizations of the height and planimetric components of the error, their magnitudes, geographical distribution, and spatial structure.
261 How Complementary are SRTM-X and -C Band Digital
Elevation Models?
Jörn Hoffmann and Diana Walter
Abstract Download
Full Article
Two different digital elevation models (DEM) were derived
during the 2000 Shuttle Radar Topography Mission from
C- and X-band interferometric radar data. While these two
DEMs share several of their properties, they were processed
independently. Here, we investigate what can be gained by
merging the two DEMs into a single composite DEM for four
different test areas. Based on an analysis of the relative
differences and the deviations from an absolute reference in
one test area, we propose an algorithm for combining the
two DEMs optimally. We then compare the composite DEM
with both individual DEMs and with a reference of a large
number of precise GPS profiles in one test area in southern
Germany. We find that in our test areas, the area of missing
values is reduced significantly in the composite DEM. Even
compared with the more complete C-band DEM, the number
of void pixels can be reduced by 22 percent to 53 percent.
Also, outlier values resulting from errors in the interferometric phase unwrapping can often be identified and removed
in the merging. The deviations of both C- and X-band DEMs
from the GPS reference are very similar and well within the
accuracy specifications of the global data set. The standard
deviation of the difference between the composite DEM and
the reference is about 14 percent below that of the original
values. Depending on the requirements for completeness and
accuracy, merging the two SRTM elevation data sets may
provide an important improvement above either of the
original DEMs.
269 Geomorphometry from SRTM: Comparison to NED
Peter L. Guth
Abstract Download
Full Article
The Shuttle Radar Topography Mission (SRTM) produced
near-global 1" and 3" DEMs. The cartographically-derived
National Elevation Dataset (NED) provides a mechanism to
assess SRTM quality. We compared 12 geomorphometric
parameters from SRTM to NED for about 500,000 sample
areas over the continental United States. For basic parameters like average elevation or relief, the two data sets correlate very highly. For more derived measures, such as
curvature and higher moments (skewness and kurtosis), the
correlations are much lower, with some parameters essentially uncorrelated between the two DEMs. Correlations
improve after restricting analysis to region with average
slopes greater than 5 percent, and the SRTM data set compares more closely to simulated 2" NED than to 1" NED. SRTM
has too much noise in flat areas, increasing average slope,
while in high relief areas SRTM over smoothes topography
and lowers average slopes. The true resolution of 1" SRTM
DEMs proves to be no better than 2".
279 Validation of SRTM Elevations Over Vegetated and
Non-vegetated Terrain Using Medium-Footprint Lidar
Michelle Hofton, Ralph Dubayah, J Bryan Blair, and
David Rabine
Abstract Download
Full Article
The Shuttle Radar Topography Mission (SRTM) generated one
of the most complete high-resolution digital topographic
data sets of the world to date. The elevations generated by
the on-board C-band sensor represent surface elevations in "bare earth" regions, and the elevations of various scatterers
such as leaves and branches in other regions. Elevations
generated by a medium-footprint (>10m diameter) laser
altimeter (lidar) system known as NASA's Laser Vegetation
Imaging Sensor (LVIS) were used to assess the accuracy of
SRTM elevations at study sites of variable relief, and land-cover. Five study sites in Maine, Massachusetts, Maryland,
New Hampshire, and Costa Rica were chosen where coincident LVIS and SRTM data occur. Both ground and canopy top
lidar elevations were compared to the SRTM elevations. In"bare earth" regions, the mean vertical offset between the
SRTM elevations and LVIS ground elevations varied with
study site and was approximately 0.0m, 0.5m, 3.0m,
4.0m, and 4.5m at the Maine, Maryland, Massachusetts,
New Hampshire, and Costa Rica study sites, respectively.
In vegetated regions, the mean vertical offset increased,
implying the phase center fell above the ground, and the
offset varied by region. The SRTM elevations fell on average
approximately 14m below the LVIS canopy top elevations,
except in Costa Rica where they were approximately 8 m
below the canopy top. At all five study sites, SRTM elevations
increased with increasing vertical extent (i.e., the difference
between the LVIS canopy top and ground elevations and
analogous to canopy height in vegetated regions). A linear
relationship was found sufficient to describe the relationship
between the SRTM-LVIS elevation difference and canopy
vertical extent.
287 SRTM C-band and ICESat Laser Altimetry Elevation
Comparisons as a Function of Tree Cover and Relief
Claudia C. Carabajal and David J. Harding
Abstract Download
Full Article
The Geoscience Laser Altimeter System (GLAS) instrument
onboard the Ice, Cloud, and land Elevation Satellite (ICESat)
provides a globally distributed elevation data set that is
well-suited to independently evaluate the accuracy of digital
elevation models (DEMs), such as those produced by the
Shuttle Radar Topography Mission (SRTM). We document
elevation differences between SRTM C-band 1 and 3 arcsecond resolution DEMs and ICESat 1064 nm altimeter channel
elevation data acquired in an areas of variable topography
and vegetation cover in the South American Amazon Basin,
Asian Tibetan Plateau – Himalayan Mountains, East Africa,
western Australia, and the western United States. GLAS
received waveforms enable the estimation of SRTM radar
phase center elevation biases and variability with respect to
the highest (canopy top where vegetated), centroid (distance-weighted average), and lowest (ground) elevations detected
within ICESat laser footprints. Distributions of ICESat minus
SRTM elevation differences are quantified as a function of
waveform extent (a measure of within-footprint relief), SRTM
roughness (standard deviation of a 3 × 3 array of elevation
posts), and percent tree cover as reported in the Vegetation
Continuous Field product derived from Moderate Resolution
Imaging Spectrometer (MODIS) data. SRTM roughness is
linearly correlated with waveform extent for areas where
percent tree cover is low. The SRTM phase center elevation
is usually located between the ICESat highest and lowest
elevations, and on average is closely correlated with the
ICESat centroid. In areas of low relief and sparse tree cover,
the mean of ICESat centroid minus SRTM phase center
elevation differences for the five regions examined vary
between -3.9 and 1.0m, and the corresponding standard
deviations are between 3.0 and 3.7 m. With increasing SRTM
roughness and/or tree cover, the SRTM elevation remains
essentially unbiased with respect to the ICESat centroid
but the standard deviations of the differences increase to
between 20 and 34m, depending on the region. For the
Australia, Amazon, Africa, United States, and Asia regions,
including all tree cover and roughness conditions, 90 percent
of the SRTM elevations are within 6.9, 11.5, 12.1, 16.8, and
37.1m of the ICESat centroid, respectively. In vegetated
areas, the SRTM elevation on average is located approximately
40 percent of the distance from the canopy top to the
ground. The variability of this result increases significantly
with increasing SRTM roughness. The results are generally
consistent for the five regions examined, providing a method
to estimate for any location the correspondence between
SRTM elevations and highest, average, and lowest elevations
using the globally-available MODIS-derived estimate of tree
cover and the measure of SRTM roughness.
299 Mapping Height and Biomass of Mangrove Forests in
the Everglades National Park with SRTM Elevation Data
Marc Simard, Keqi Zhang, Victor H. Rivera-Monroy, Michael
S. Ross, Pablo L. Ruiz, Edward Castañeda-Moya, Robert R.
Twilley , and Ernesto Rodriguez
Abstract Download
Full Article
We produced a landscape scale map of mean tree height in
mangrove forests in Everglades National Park (ENP) using the
elevation data from the Shuttle Radar Topography Mission
(SRTM). The SRTM data was calibrated using airborne lidar
data and a high resolution USGS digital elevation model
(DEM). The resulting mangrove height map has a mean tree
height error of 2.0 m (RMSE) over a pixel of 30 m. In addition,
we used field data to derive a relationship between mean
forest stand height and biomass in order to map the spatial
distribution of standing biomass of mangroves for the entire
National Park. The estimation showed that most of the
mangrove standing biomass in the ENP resides in intermediate-height mangrove stands around 8 m. We estimated the
total mangrove standing biomass in ENP to be 5.6 × 109 kg.
313 Capability of SRTM C and X Band DEM Data to
Measure Water Elevations in Ohio and the Amazon
Brian Kiel, Doug Alsdorf, and Gina LeFavour
Abstract Download
Full Article
We analyze Shuttle Radar Topography Mission (SRTM) water
surface elevation data to assess the capacity of interferometric
radar for future surface water missions. Elevations from three
Ohio reservoirs and several Amazon floodplain lakes have
standard deviations, interpreted as errors, that are smaller in
C-band compared to X-band and are smaller in Ohio than in
the Amazon. These trends are also evident when comparing
water surface elevations from the Muskingum River in Ohio
with those of the Amazon River. Differences are attributed to
increased averaging in C-band compared to X-band, greater
sensitivity to surface water motion in X-band, and generally
larger off-nadir look angles in X-band. Absolute water surface
elevations are greater in the C-band DEM for much of the two
study areas and yield expected slope values. Height and slope
differences are attributed to differing usage of geoids and
ellipsoids. These SRTM measurements suggest the great
possibility for space-based, laterally-spatial (2D) measurements of water surface elevations.
321 Detection of Ancient Settlement Mounds – Archaeological
Survey Based on the SRTM Terrain Model
B.H. Menze, J.A. Ur, and A.G. Sherratt
Abstract Download
Full Article
In the present study we demonstrate the value of the SRTM
three arcsecond terrain model for a virtual survey of archaeological sites: the detection and mapping of ancient settlement mounds in the Near East. These so-called "tells" are
the result of millennia of occupation within the period from
8000–1000 BC, and form visible landmarks of the world's
first farming and urban communities. The SRTM model
provides for the first time an opportunity to scan areas not
yet surveyed archaeologically on a supra-regional scale and
to pinpoint probable tell sites. In order to map these historic
monuments for the purpose of settlement-study and conservation, we develop a machine learning classifier which
identifies probable tell sites from the terrain model. In a test,
point-like elevations of a characteristic tell shape, standing
out for more than 5 to 6 m in the DEM were successfully
detected (85/133 tells). False positives (327/(600*1200)
pixels) were primarily due to natural elevations, resembling
tells in height and size.