On orbit calibration and validation activities for the Ice, Cloud, and Land Elevation Satellite – 2 (ICESat-2) Mission

Abstract:

The Ice, Cloud, and Land Elevation Satellite – 2 (ICESat-2) Mission began collecting data on 14 October 2018. The mission uses green laser light emitted by the Advanced Topographic Laser Altimetry System (ATLAS) to detect individual photons that are reflected by the Earth’s surface and returned to ATLAS. The time of flight of these photons, when combined with information on the pointing direction, and position of the observatory in space, provide a geolocation and elevation for every telemetered photon event spanning the globe from 88 degrees north latitude to 88 degrees south. The ATL03 Global Geolocated Photon data product provides these geodetic coordinates for each photon event telemetered to Earth for each of ATLAS’s six beams. This product also discriminates photons between high, medium, and low signal confidence levels and those events associated with background photons. The higher level, along-track products each use different strategies for photon aggregation to optimize the precision and accuracy of the surface retrievals over specific surface types. These types include land ice, sea ice, vegetation/land, ocean, and inland water. Calibration efforts use well designed on-orbit maneuvers to identify both pointing and range biases attributed to orbital variations on the satellite. Once corrected, the science-quality data products were released to the public in May 2019 and are currently Release 003.

We present our ongoing work to evaluate and validate the geolocation and elevation accuracy and surface measurement precision for data from the full ICESat-2 mission. The approaches are diverse in both location and methodology to ensure that we have a comprehensive assessment of the ATLAS performance variations throughout the mission. These strategies include comparisons with ground-based and airborne elevation measurements over the ice sheets (such as those from Operation IceBridge), detailed analysis of returns from well-surveyed corner cube retro-reflectors, comparison of sea ice freeboard measured by airborne lidars, evaluation of global-scale ocean elevation through comparison with radar altimeters, and comparison of vegetation canopy height metrics measured by airborne lidar. Our work to date demonstrates that individual photon elevations are accurate to approximately 30 cm vertically, and 6 m radially. Aggregating many photons together reduces the elevation uncertainty to less than 5 cm for relatively flat and smooth ice sheet interiors and allows for a precision of better than 2 cm over sea ice leads.