Joint inversion of Magnetotelluric, Seismic and Gravity data

In this thesis joint inversions of magnetotelluric and seismological data from the north-western section of USArray together with satellite gravity data are performed to investigate howsatellite data interact with the ground–based observations. Specifically the question as to whether satellite data can improve models that have been created using data sets with large spatial gaps is addressed. Additionally, more technical questions considering how the various data sets complement each other given their different resolution capabilities and sensitivities, as well as the influence of the cross–gradient coupling constraint on the joint inversion models are investigated.

Firstly, all three observables are inverted separately to assess their respective sensitivities. Using the resulting models as a baseline forcomparisons, both ground–based data sets are inverted together with the gravity observations, respectively. The USArray data represents a best–case scenario for ground–based data with a regular station coverage. This can also be used to simulate data sets with an irregular station coverage and large gaps between stations.

In both joint inversions the physical properties related to the ground–based data do not change significantly. Even when removing up to half of the MT data the addition of data does not make a large difference. The density model, on the other hand, changes drastically in both cases. It is much better constrained vertically, but the final density models look very different and indicate the noneuniqueness of the solutions.

There are two reasons for this behavior: the lack of vertical resolution of the gravity data and the limitations of the cross–gradient constraint. While the resolution capability of the gravity data can not be changed, other options for the coupling mechanism exist (e.g. a mutual information coupling) which present promising options to better constrain models from ground based data using gravity observations.