Globally consistent land ice mass balance estimates from GRACE/GRACE-FO: Sensitivity to methodological choices

Satellite gravimetry as realized by the GRACE and GRACE-FO missions (both referred to as GRACE in the following) is a major observation for ice mass balance estimates. Thereby, GRACE-based studies were often conducted separately for either the Antarctic Ice Sheet, the Greenland Ice Sheet, or the glo...

Full description

Bibliographic Details
Main Authors: Döhne, T., Horwath, M., Tripathi, V.
Format: Conference Object
Language:English
Published: 2023
Subjects:
Online Access:https://gfzpublic.gfz-potsdam.de/pubman/item/item_5021142
Description
Summary:Satellite gravimetry as realized by the GRACE and GRACE-FO missions (both referred to as GRACE in the following) is a major observation for ice mass balance estimates. Thereby, GRACE-based studies were often conducted separately for either the Antarctic Ice Sheet, the Greenland Ice Sheet, or the global glacier domain or selected glacier regions. The start of the Glacier Mass Balance Intercomparison Exercise (GLAMBIE) in addition to the Ice-sheet Mass Balance Intercomparison Exercise (IMBIE) gives new impetus to questions about the consistency between GRACE-based estimates for different land ice domains and about the sensitivity of the estimates to methodological choices. We report investigations on a consistent land ice mass balance estimate based on our tailored sensitivity kernel approach previously applied to the Antarctic Ice Sheet and the Greenland Ice Sheet. In the tailored sensitivity kernel approach the sensitivity kernel of the mass change estimator is directly optimized in a formal minimization of propagated GRACE solution errors and leakage errors. The focus on specific regions is realized by 'tailoring' the method to the signal of interest based on available signal covariance information of mass changes, without violating global consistency of the analysis. We show via a sensitivity analysis how different parameters of the estimation method (such as weighting of GRACE error covariance information, degree of external information employed for the signal covariance information, and accounting for sea-level fingerprints) influence the resulting mass change estimates.