International Archives of Photogrammetry and Remote Sensing, Vol. 32, Part 3W14, La Jolla, CA, 9-11 Nov. 1999
PROCESSING OF SHUTTLE LASER ALTIMETER
RANGE AND RETURN PULSE DATA IN SUPPORT OF SLA-02
Claudia C. Carabajal', David J. Harding?, Scott B. Luthcke?, Waipang Fong”, Shelley C. Rowton’, and J. J. Frawley®
INVI, Inc., (0 NASA/Goddard Space Flight Center (Code 926), claudia@stokes.gsfc.nasa.gov
*NASA/Goddard Space Flight Center (Code 921), harding@denali.gsfc.nasa.gov
?Raytheon ITSS, (à NASA/Goddard Space Flight Center (Code 926), sluthcke@ geodesy2.gsfc.nasa.gov
‘Raytheon ITSS, Greenbelt, MD 20770, USA, waipang@magus.stx.com
Raytheon ITSS, Greenbelt, MD 20770, USA, srowton@magus.stx.com
® Herring Bay Geophysics, @ NASA/Goddard Space Flight Center, hbgjjf@ltpmail.gsfc.nasa.gov
KEY WORDS: space-borne laser altimetry, remote sensing, laser bounce-point geolocation, laser backscatter modeling.
ABSTRACT
The second flight of the Shuttle Laser Altimeter (SLA) flew on board the space Shuttle Discovery, during August 1997 during the
STS-85 Mission.
The nearly 3 million laser shots transmitted during the course of the 11 day SLA-02 mission yielded
approximately 590,000 geolocated returns from land and more than 1,500,000 from ocean surfaces. These data were analyzed to
produce a data set that provides laser altimetry elevations of high vertical accuracy that can be used for scientific purposes.
Processing of the data included the geolocation of surface returns, involving precision TDRSS-tracking based Shuttle orbit
determination and pointing bias calibration, ellipsoid to geoid reference frame transformations, conversion of engineering
parameters to physical units, application of scaling factors to obtain a consistent measure of the backscatter energy, and
classification of the returns based on comparisons with reference elevation data (TerrainBase Digital Elevation Model (DEM) and
mean sea level). Additionally, the digitized laser returns were analyzed and modeled using constrained non-linear least-squares
optimization techniques. The elevation data were compared to both high-resolution DEMs and a reference ocean surface to
assess data accuracy. Ancillary data, such as NDVI (Normalized Digital Vegetation Index) and Land Cover classification data,
were also included in the distributed data set. Key aspects of the data analysis are discussed. Further documentation concerning
SLA-02 data processing procedures, problems evidenced in the data, and its distribution format is provided in the SLA-02 site
(http://denali.gsfc.nasa.gov:8001/).
1. INTRODUCTION
The Shuttle Laser Altimeter (SLA) was designed as a pathfinder
experiment to evaluate engineering and algorithm techniques to
aid the transition of the airborne laser altimeter and lidar
technology developed at Goddard Space Flight Center to low
Earth orbit operational space-borne systems (Garvin et al.,
1996). Two flights of SLA have provided high-resolution,
orbital laser altimeter observations of terrestrial surfaces that
constitute scientific data sets of value in addressing global Earth
System science issues. SLA also serves as a test-bed for
upcoming orbital laser altimeters, such as the Multi-Beam Laser
Altimeter (MBLA) (Bufton et al.,, 1999) and the Geoscience
Laser Altimeter System (GLAS), that will be launched aboard
the Vegetation Canopy Lidar (VCL) mission in 2000 and the
Ice, Cloud and land Elevation Satellite (ICESat) in 2001,
respectively. The equatorial observations provided by the first
flight of SLA (SLA-01) were extended to 57 degrees by SLA-
02, characterizing ocean, land, and cloud top elevations in 100
meter diameter footprints spaced every 700 meters utilizing a
laser transmitter firing at a rate of 10 pulses per second (Figure
1). The SLA instrument provides the round-trip travel distance
of short duration (1064nm wavelength) laser pulses to the first
encountered surface, either a cloud top, vegetation canopy top,
bare ground, or water, with a 0.75 m precision (Bufton et al.,
1995). As for SLA-01, ranging was augmented by digitizing
the time-varying return pulse energy from surfaces distributed
vertically within the laser footprint, enabling a measurement of
within-footprint relief introduced by vegetation cover and
topographic slope and roughness. Combining the laser ranging
data with shuttle position and pointing knowledge yielded
highly accurate surface elevation data. The processing
procedures involved in producing the SLA-02 data set are
summarized here. The processing procedures and data
distribution format are fully specified in the documentation
accompanying the data set, available for downloading at
http://denali.gsfc.nasa.gov:8001/. The data were acquired and
processed as ‘observations’, which represent a continuous
period of instrument operation.