High-Resolution Temporal Geoid Modeling in Alaska for a New Geopotential Datum

In 2022, the US National Geodetic Survey will introduce a geoid model to serve as the basis for a new geopotential datum. Because the geoid changes in time, this model must account for geoid change signals of more than 1 cm on the timescale of a decade to maintain its accuracy. The GRACE mission provided monthly geopotential solutions that may be converted to geoid trends at ~300 km resolution, which is adequate for capturing most geopotential change in North America. However, highly localized ice mass loss from mountain glaciers in Alaska produces geoid change signals exceeding 2 mm yr^-1 at higher spatial resolution than GRACE can capture, with truncation effects attenuating peak geoid change amplitudes by a factor of 2. Additionally, the terrestrial gravity and leveling datasets used to construct and verify the static geoid model were taken over many decades and contain signals of uplift and geoid change at the decimeter level. To predict present-day geoid change, we combine Goddard Space Flight Center mascon solutions with high-resolution ice mass loss rates developed with data from ICESat, ICESat-2, and airborne lidar missions to generate high-resolution predictions of geoid change. These models are verified against dense GNSS bedrock velocity measurements and tide gauge records. We examine the resolution improvement offered by the Goddard L1B regression trend mascon solutions. Finally, we use comparisons of modern altimetry with aerial photogrammetry from the mid-20th century to develop models of past geoid change and verify these predictions against tide gauge records, leveling, and terrestrial gravity observations.