Date of Award

Spring 1-1-2019

Document Type


Degree Name

Doctor of Philosophy (PhD)


Aerospace Engineering Sciences

First Advisor

R. Steven Nerem

Second Advisor

Delores Knipp

Third Advisor

Tomoko Matsuo

Fourth Advisor

Michael Willis

Fifth Advisor

David N. Wiese


This work presents methods for combining time-variable gravimetry, altimetry, and GNSS bedrock motion data to distinguish sources of mass change in Antarctica with enhanced spatial resolution. GRACE gravimetry provided direct measurements of mass variation of the Antarctic Ice Sheet at monthly timescales between 2002 and 2017, but was limited to a spatial resolution of ~300 km. Laser altimetry from ICESat (2003-2009) provided elevation change measurements at fine spatial resolution, better resolving the spatially concentrated sources of Antarctic mass loss, but with sparse temporal sampling. The processes of glacial isostatic adjustment (GIA), ice dynamics, surface mass balance, and firn compaction affect geodetic measurements with distinct magnitudes, timescales, and spatial scales, which means they may be separated through combination of different data sources. This work explores separation of GIA and ice sheet processes by combining data from ICESat and GRACE. This is accompanied by separation of GIA and ice sheet processes using GNSS vertical crustal motion estimates and GRACE. Monthly solutions for combined ice sheet mass variation are developed using monthly GRACE solutions combined with high-resolution ICESat elevation rates. High-resolution monthly solutions for Antarctic mass variation are developed by combining ICESat elevation rates statistical information from models of the processes of interest with monthly GRACE data. This work also examines of the impact of atmospheric modeling errors on estimates of Antarctic mass loss, finding that errors in models used to remove atmospheric signals from GRACE solutions obscure additional acceleration in total Antarctic mass loss. Finally, time-variable GRACE and ICESat data are combined directly to produce 17 time-variable mass solutions at high spatial resolution between 2003 and 2009 with an updated GIA model. The solutions indicate 2003--2009 average mass loss of 116+10-51 Gt yr-1 and a total GIA mass rate of 110+60-24 Gt yr-1. With GRACE Follow-On and ICESat-2 now concurrently in orbit, the methods developed in this work pave the way toward simultaneous assimilation of their respective gravity and elevation data into a monthly, high-resolution solution for Antarctic mass change.