Precise geolocation of the laser surface bounce point is of crucial importance in utilizing laser altimetry data in scientific applications. The laser altimetry system measures the time of flight of the laser pulse. This time of flight determines the range to the surface 'bounce point', which in conjunction with the knowledge of the position and pointing of the laser altimeter, enables the geolocation of the surface return. Geolocation of SLA-02 data was done applying the same techniques previously used to georeference the SLA-01 altimetry data. Much of the SLA-02 operations were conducted during intervals spanning astronaut sleep periods, corresponding to 'quiet' periods with no orbital maneuvers or attitude mode changes. However, significant maneuvers were performed during some data acquisition periods. The times and attitude modes for every acquisition period are given in the SLA data distribution matrix. Each continuous period of SLA data collection is referred to here as an observation, of which there were 23 during the flight of SLA-02. Times and attitude modes for all geolocated observations to date are given in the SLA data distribution table, which can be accessed from "http://denali.gsfc.nasa.gov:8001/".

Before proceeding with the geolocation process, time tag data were analyzed to ensure consistent assignment of time to the range data. SLA time is calculated by synchronizing the SLA internal time keeping mechanism with the Shuttle's clock. SLA time provides an incremental time measurement, and the Shuttle's clock provides an absolute time by giving time with respect to the Mission's Reference start time. The Shuttle time is received by SLA as a serial time reference message from the master timing unit, a 4.608-MHz stable crystal-controlled timing source for the orbiter, which provides synchronization for instrumentation payloads and other systems (http://www.ksc.nasa.gov/shuttle/technology/sts-newsref/sts-inst.html#sts-mtu). Shuttle time is read and recorded once per minute by the SLA flight software. The SLA internal time is kept by means of an inexpensive oscillator (standard for the flight computer), with a 1.193 MHz frequency. This oscillator acts as a high resolution 16 bit counter which counts down from 65536 to 0, and sends an interrupt to the computer every time it rolls over. In the flight code, for each laser fire the value of the counter and the incremented interrupts are stored. SLA time is then computed using the following equation:

System time=[(ticks * 65536.0d0) + (65535.0D0 - hirez)]*838.09580D-9;

For every 1 minute pulse from the Shuttle, the value for hirez, ticks and the Shuttle's 1 pulse per minute time are read and stored. This information is combined to produce the laser shot's time-tag.

The offset between the Shuttle and SLA time is calculated, and the oscillator's drift is accounted for by fitting the best quadratic function that models the residual of the two time series. The altimeter range time-tags were corrected using this best-fit function. On occasion, the flight software misinterpreted the Shuttle time while unpacking the bytes that contained the minute tag information. This resulted in jumps in the time line that when easily identified were accounted for and fixed. Duplicate time tags that resulted from a buffering problem during data recording were also identified, and eliminated before geolocation. For several observations (5, 6, 14 and 14a) time-tag inconsistencies have not been resolved, and the data have therefore not been geolocated.

During the observational periods processed, the Shuttle was in a -ZLV (Shuttle -Z, cargo bay normal axis, pointed to local vertical) YVV (Shuttle-Y, wing axis, pointed in velocity vector direction) Local Vertical Local Horizontal (LVLH) attitude mode. The orientation of the spacecraft remained constant within some attitude control 'dead-band'. Twenty-one of the observation periods processed were mostly in a 1 degree attitude dead-band, where roll, pitch, and yaw maximum deviation from the aforementioned attitude was +/- 1 degree. Two of the observation periods processed (18 and 20) were in a 0.1 degree attitude dead-band. Observation periods 13, 17, 18 and 19 show deviations from the 1 degree dead-band that can range from 2 to 5 degrees for roll and pitch. Observations 11 and 20 show significant yaw maneuvers. The recovery of pointing biases during these observation periods was considerably more difficult and sometimes hindered, as for observation 17, for which no correction was applied.

Johnson Space Center's Mission Evaluation Workstation System (MEWS) provided the Shuttle body attitude in the form of quaternions. Shuttle attitude data provided by the MEWS is derived from the Inertial Measurement Unit (IMU) observations. The Shuttle's IMUs were calibrated before, entering the -ZLV Earth viewing mode used for SLA observations by means of star cameras. The MEWS attitude data are used for two purposes: 1) orienting the offset between the antenna used for TDRSS tracking and the Shuttle center-of-gravity (cg) necessary for measurement modeling corrections, and 2) obtaining a first approximation of the SLA pointing vector, initially considered to be coincident with the Shuttle body -Z axis. The quaternions were made continuous and interpolated at observation times. The analysis of the attitude mode history was taken into account for the arc selection process for precision orbit determination.

The sequential steps used in the geolocation of SLA data are shown in Figure 1. The altimetry geolocation process is extensively treated in the literature cited [2,4,6,7], and it is also documented in the SLA-01 data products site from the Laser Altimeter Processing Facility (LAPF) (http://denali.gsfc.nasa.gov/lapf/sla/srowton/summary.html). We will present a quick overview of the process, referring the reader to the previously mentioned documentation for further details on the subject. In support of SLA-02, meter level Root-Mean Square (RMS) Shuttle radial orbit accuracy has been achieved from Tracking and Data Relay Satellite System (TDRSS) Doppler observations. Traditionally, the Tracking and Data Relay Satellite (TDRS) orbits themselves have been the dominant source of error in Shuttle orbit determination during quiescent attitude periods. The technique utilizing TOPEX/Poseidon's (T/P) precise orbit knowledge, plus the TDRSS-T/P Doppler tracking in conjunction with BRTS and TT&C range data were used to precisely position the TDRS [3]. Furthermore, a special T/P-TDRSS tracking scenario was devised and implemented in support of the STS-85 mission. This tracking scenario optimized the sampling of the TDRS orbits with the best possible tracking data and was not employed for STS-72. The significant improvement in TDRS-4 orbit precision gained from this tracking scenario can be seen in Table 1 when compared to the TDRS orbit precisions obtained in support of STS-72. The TDRS-1 orbit precision is significantly worse than the other TDRS due to that fact that T/P was not tracked by this TDRS. However, nearly all of the STS-85 tracking data was acquired with TDRS-4 & 5.

This approach, along with improved modeling and parameterization, has allowed the computation of precise Shuttle orbits from TDRSS-Shuttle Doppler tracking data [2]. Table 2 presents a comparison of Shuttle-TDRS 2-way range rate residual RMS averaged over all arcs. The data shows an improved fit for the STS-85 case. This was mainly due to more relaxed constraints employed for STS-85 and significantly shorter arcs on average. However, it should be noted that this does not indicate improved orbit accuracies.

In support of SLA-01, an extensive STS-72 orbit precision and accuracy study was performed [2]. This study showed the shuttle orbits to be accurate to within 1.5 m radial RMS and 8 m total position RMS. From the STS-72 study results, the TDRS orbit precisions and shuttle tracking data presented above and some limited orbit accuracy analysis, the STS-85 orbits are considered to be accurate within 10 m total position RMS and a few meters radial RMS. Ocean comparisons for the first 4 observation periods showed ~2 meter radial orbit accuracy for the middle of the arcs, when no tails were included. The STS-85 orbit accuracies are considered not to be as good as those that were obtained for STS-72 due to shorter arc lengths and significantly more attitude and orbit maneuvers. The data is currently being further analyzed in an attempt to improve the STS-85 orbit accuracies.

Once precise Shuttle orbits are obtained, SLA range data (corrected for a constant range bias and tropospheric effects) are combined with and Shuttle attitude data in solving for the bounce point location using GEODYN [1]. GEODYN is a state-of-the-art precision orbit determination and geodetic parameter estimation software suite developed at Goddard Space Flight Center. This software suite has extensively been modified to include a rigorous laser altimeter range measurement model and new dynamic cross-over analysis algorithms. The range used in the geolocation process is the range to the first backscatter signal above the detection threshold. The resulting elevations thus correspond to the highest detected surface within the 100 meters diameter laser footprint. For cloud-free paths to land targets this could be the upper-most canopy where vegetation is present, the tops of buildings or structures, or the highest ground where vegetation, buildings and structures are absent.

With the excellent shuttle orbit accuracies achieved from the above described POD analysis, the remaining significant factor driving vertical and horizontal geolocation accuracy is the laser pointing knowledge. The pointing knowledge is significantly affected by laser and spacecraft body systematic misalignments due to mounting errors, IMU misalignment, and Shuttle body flexure. An attempt is therefore made to extract pointing biases from the data. Errors in the a priori Shuttle body attitude contribute to the resulting height errors, which are significantly larger during 1 degree dead-band modes than during 0.1 degree modes. However, it is considerably easier to both observe and separate the roll and pitch errors during 1 degree dead-band than during 0.1 degree dead-band, even though the increase in attitude hold thrusting required, impacts the orbit determination process by increasing unmodeled dynamical effects. Roll and pitch biases are modeled and corrected before obtaining the final geolocation information. Roll and pitch biases can each be established because Shuttle attitude changes in roll and pitch are significantly out of phase [4]. A first order approach in recovering pointing and range biases is done using a direct altimetry range residual analysis, combining spacecraft attitude information with ocean range residuals [4]. The direct altimetry ocean data are compared with OSU (Ohio State University) 1995 Mean Sea Surface Model [5], plus the effects of tides from the Ray Ocean Altimeter Pathfinder Tide model giving the height error in the signal. Surface elevations of the open ocean are known, through measurements and modeling, to the 12cm (1 sigma) level, providing a global reference surface to compare to the altimeter range measurements throughout the mission. During the 0.1 degree dead-band modes the height errors are considerably smaller, but large horizontal positioning errors may still exist. SLA pointing corrections to the roll and pitch were computed for each of the SLA-02 observation periods, iterating to solve for single constant biases as the ocean residuals become smaller and convergence was reached. This first comparison neglects only the smaller contributions from barotropic pressure (<10 cm), earth tides (<20 cm) and the time dependent part of dynamic sea surface topography (<50 cm). SLA-ocean surface height differences on the order of 30 meters were observed (before pointing bias estimation), with larger residuals present in some of the arcs where the Shuttle attitude exhibited a significant number of maneuvers. After establishing the attitude corrections, the geolocation was performed using T/P consistent reference frames, precise shuttle orbits described above, a SLA optical center to Shuttle cg offset correction, a -5.6 meters altimeter range bias, and the Marini Murray tropospheric refraction correction. Average values for the constant attitude biases obtained from the first four observation periods (periods without significant maneuvers) were applied to the data when attitude biases were difficult to extract from the residuals. Data corresponding to observation period 17 were geolocated applying no bias corrections for roll and pitch.

A leveling correction is applied to the resulting elevation data for each laser bounce point to correct the effects of long wavelength orbit errors. The ocean range residuals time series is smoothed by a fixed time-length, rather than fixed number-of-points, sliding boxcar filter with a 120 seconds length, equivalent to 1200 laser shots. The minimum number of points allowed within the window for the point to be included was 50, and a 3 sigma editing is performed. Care was taken during the interpolation process at the edges of an ocean pass when extrapolating this correction across land. The geolocation process yields elevations referenced to the T/P ellipsoid (see SLA-01 Geolocation document). Orthometric elevations were computed by subtracting the geoid height at each laser bounce-point defined by the Earth Geoid Model 96 (EGM96) [7]. The leveling correction is provided in the SLA02.BP.SURFACE_3 parameter, which is a measure of the elevation error due to long-wavelength orbit errors. This level of processing constitutes the SLA-02 Standard Data Product Version 2 (SDP v2).

An assessment of horizontal accuracy was performed by differencing bounce-point elevations for selected segments with high-resolution topography from 90-meter resolution Digital Terrain Elevation Data (DTED), as shown in Figure 2. The SLA track was shifted horizontally with respect to the DTED searching for the best-fitting profile that minimized the RMS differences between SLA and the high-resolution topography profiles. The orthometric heights were corrected by the width of the transmit pulse. Last return profiles were computed by subtracting the return pulse widths derived from the waveforms yielding the lowest surface intercepted within the footprint. First return differences with respect to DTED with a mean of 9.59 meters and a standard deviation of 12.47 meters were found for SLA-02 data collected across Maryland during the second observation period. A mean of -7.86 meters and a standard deviation of 15.55 meters were found when computing last return differences. Both first and last returns best fits with respect to DTED yielded a horizontal shift of 185 meters to the South.

[1] Rowlands, D.D., J.A. Marshall, J.J. McCarthy, S.C.Rowton, D. Moore, D.E.Pavlis, S.B. Luthcke, and L.S. Tsaoussi, 1993, GEODYN II System Description, Hughes STX Contractor Report, Greenbelt, Maryland.

[2] Rowlands, D.D., S.B. Luthcke, J.A. Marshall, C.M. Cox, R.G. Williamson, and S.C.Rowton, 1997, Space Shuttle Precision Orbit Determination in Support of SLA-1 Using TDRSS and GPS Tracking Data, The Journal of the Astronautical Sciences, 45(1): 113-129.

[3] Luthcke, S.B., J.A. Marshall, S.C. Rowton, K.E. Rachlin, C.M. Cox, and R.G. Williamson, 1997, Enhanced Radiative Force Modeling of the Tracking and Data Relay Satellites, The Journal of the Astronautical Sciences, 45(3): 349-370.

[4] Luthcke, S.B., D.D. Rowlands, J.J. McCarthy, E.Stoneking, D.E. Pavlis, 1999, Spaceborn laser altimeter pointing bias calibration from range residual analysis, submitted to The Journal of Spacecraft and Rockets.

[5] Yi, Yuchan "Determination of Gridded Mean Sea Surface from TOPEX, ERS-1 and

GEOSAT Altimeter Data," 1995, Ohio State University, Department of Geodetic Science and Surveying, Report No. 434.

[6] Carabajal, C.C., D.S. Harding, S. B. Luthcke, W. Fong, S.C. Rowton and J.J. Frawley, 1999, Improvements in Shuttle Laser Altimeter Data Geolocation of Shuttle Laser Altimetry, abstract submitted to the ISPRS Workshop in "Mapping Surface Structure and Topography by Airborne and Spaceborne Lasers," La Jolla, CA.

[7] Luthcke, S.B., C.C. Carabajal, D.D. Rowlands, D.E. Pavlis, 1999, Improvements in Shuttle Laser Altimeter Data Geolocation of Shuttle Laser Altimetry, in preparation.

[8] Lemoine, F.G., S.C. Kenyon, J.K. Factor, R.G. Trimmer, N.K. Pavlis, D.S. Chin, C.M. Cox, S.M. Klosko, S.B. Luthcke, M.H. Torrence, Y.M. Wang, R.G. Williamson, E.C. Pavlis, R.H. Rapp, and T.R. Olson, 1998, The Development of the Joint NASA GSFC and the National Imagery and Mapping Agency (NIMA) Geopotential Model EGM96, NASA/TP-1998-206861.

Responsible NASA official: David Harding

Technical Contact: Scott B. Luthcke/Claudia C. Carabajal

Webpage Developer: Claudia Carabajal

Email sluthcke@magus.stx.com, or claudia@stokes.gsfc.nasa.gov with questions, comments or suggestions.

Last modified November 5, 1999.