Enhanced geolocation of spaceborne laser altimeter surface returns: parameter calibration from the simultaneous reduction of altimeter range and navigation tracking data
The accurate geolocation of a laser altimeter's surface return, the spot from which the laser energy reflects on the Earth's surface, is a critical issue in the scientific application of these data. Pointing, ranging, timing and orbit errors must be compensated to accurately geolocate these data. Detailed laser altimeter measurement models have been developed and implemented within precision orbit determination software providing the capability to simultaneously estimate the orbit and geolocation parameters from a combined reduction of altimeter range and spacecraft tracking data. In preparation for NASA's future dedicated Earth observing spaceborne laser altimeter missions, the Vegetation Canopy Lidar (VCL) and the Ice, Cloud and land Elevation Satellite (ICESat), data from two Shuttle Laser Altimeter (SLA) missions have been reprocessed to test and refine these algorithms and to develop the analysis methodologies for the production and verification of enhanced geolocation products. Both direct altimetry and dynamic crossover data have been reduced in combination with navigation tracking data to obtain significant improvement in SLA geolocation accuracy. Residual and overlap precision tests indicate a factor of two improvement over the previously released SLA Standard Data Products, showing 40-m RMS horizontal and 26-cm RMS elevation geolocation precision for the long SLA-01 arcs. Accuracy estimates by comparing SLA profiles to Digital Elevation Models show horizontal positioning accuracy at the 60-m (1 σ) level. Vertical accuracies, on the order of 1 m (1 σ) for low slope surfaces are now dominated by the ±75-cm one-way range resolution of the instrument. Comparable relative improvements are also observed in the analysis of the SLA-02 data. The analyses show that complex temporal variations in parameters (i.e., pointing) can be recovered and not just simple biases. The methodology and results obtained from the detailed analysis are discussed in this paper, along with their applicability to VCL and ICESat.