Separation of signal components in global gravity models
- 1Chair for Astronomical and Physical Geodesy, Technical University of Munich, Munich, Germany (betty.heller@tum.de)
- 2Professorship Big Geospatial Data Management, Technical University of Munich, Munich, Germany
Vertical movements of the Earth’s surface represent mass displacements, which cause a temporal gravity signal that can be measured by dedicated satellite gravity missions such as the GRACE or GRACE-FO missions. Especially observations of vertical movements that are caused by mantle dynamic processes would enable to constrain numerical mantle convection models using geodetic data sets, thereby improving our understanding about the physical behavior of the Earth’s interior.
Using satellite gravity data to observe the above-mentioned vertical movements poses two main challenges:
First, the small amplitudes of the geoid trend signals induced by mantle dynamic signals require data accuracies and record lengths that will only be met by future satellite gravity missions. Indeed, it is known from previous simulation studies that temporal gravity signals produced by mantle convection will be detectible in future double-pair satellite gravity missions such as the planned Mass Change and Geoscience International Constellation (MAGIC).
The second challenge to make use of gravity data sets for constraining geophysical mantle models is the extraction of the signal of interest from the total gravity signal. While temporal gravity data sets include the cumulative mass displacement signal, the problem of how to separate the superimposed signals produced by phenomena in the hydrosphere, cryosphere, atmosphere, oceans and solid Earth is still unsolved.
In this contribution, using the gravity signals given by the updated ESA Earth System Model, we address the task of signal separation in temporal gravity data and present two approaches for it. To this end, the knowledge of the spatial and temporal characteristics of the individual signal components is exploited by applying principal component analysis as well as a machine learning approach.
How to cite: Heller-Kaikov, B., Pail, R., and Werner, M.: Separation of signal components in global gravity models, EGU General Assembly 2023, Vienna, Austria, 24–28 Apr 2023, EGU23-1257, https://doi.org/10.5194/egusphere-egu23-1257, 2023.