Bayesian integration of satellite geodetic data with models to separate land hydrology and surface deformation signals
Reliable quantification of water mass changes (or redistribution) within the different compartments of the water cycle is important for understanding processes and feedback loops within the Earth's climate system. This information is also essential in geodesy because it changes the Earth's...
| Main Author: | |
|---|---|
| Format: | Thesis |
| Language: | English |
| Published: |
2021
|
| Subjects: | |
| Online Access: | https://orca.cardiff.ac.uk/id/eprint/145871/ https://orca.cardiff.ac.uk/id/eprint/145871/13/Mehrnegar_PhDthesis.pdf https://orca.cardiff.ac.uk/id/eprint/145871/2/mehrnegarn.pdf |
| Summary: | Reliable quantification of water mass changes (or redistribution) within the different compartments of the water cycle is important for understanding processes and feedback loops within the Earth's climate system. This information is also essential in geodesy because it changes the Earth's orientation (importance for defining reference frames) and the Earth's gravity field, which is the physical shape of the Earth and is used for defining reference datum. The Gravity Recovery and Climate Experiment (GRACE) and its Follow-On mission (GRACE-FO) provide time-variable Earth's gravity fields that contain signals related to different processes such as non-steric sea level changes, Terrestrial Water Storage Changes (TWSC), ice sheet melting, and Post Glacial Rebound (PGR). Although GRACE(-FO) data represent an accurate superposition of these anomalies, separating this integrated signal into its contributors is desirable for many hydro-climatic and geophysical applications. In this thesis, three novel Bayesian data-model fusion frameworks are developed to separate land hydrology (surface and sub-surface) and surface deformation (due to PGR) from GRACE(-FO) data. The three main frameworks of this thesis include: 1- the Dynamic Model Data Averaging (DMDA), that is formulated to merge multi-model data with GRACE(-FO) data; 2- Markov Chain Monte Carlo-Data Assimilation (MCMC-DA), as an extension of DMDA, to recursively estimate components of the TWSC, while accounting for temporal dependencies between the storage compartments; and 3- the Constrained Bayesian-Data Assimilation (ConBay-DA) to use multi-sensor data for GRACE(-FO) signal separation. DMDA is used to compare several global hydrological models and merge them with GRACE data. The groundwater and soil water storage changes are extracted within the Conterminous United States (CONUS) by implementing the MCMC-DA approach. ConBay-DA is applied, based on the hierarchical MCMC optimization, to use GRACE data and the surface uplift rates from the Global Navigation ... |
|---|