Utilisation of stochastic {MT} inversion results to constrain gravity inversion

Jeremie Giraud and H. Seillé and G. Visser and V. Ogarko and Mark Lindsay and Mark Jessell. ( 2021 )
in: 82nd {EAGE} {Annual} {Conference} \& {Exhibition}, pages 1--5, European Association of Geoscientists \& Engineers

Abstract

We introduce a methodology developed with the objective of exploiting complementary information between 1D magnetotelluric (MT) and gravity inversion. To maintain flexibility, we propose a cooperative workflow leveraging standalone inversions. We first perform 1D probabilistic MT inversion to obtain ensembles of models representative of the measurements. We then use the probabilities of presence of the different rock units derived from these ensembles to divide the studied area into domains characterized by positive probabilities to observe the different rock units. Thirdly, these domains are used to constrain the inversion of gravity data by restricting density values accordingly with the rock units of each domain obtained from MT-derived probabilities. We perform the synthetic proof-of-concept using a realistic model based on the framework of a region in the Mansfield area (Victoria, Australia). Results reveal that our methodology can improve subsurface imaging and can be applied to field data.

Download / Links

BibTeX Reference

@INPROCEEDINGS{Giraud20218EACE,
    author = { Giraud, Jeremie and Seillé, H. and Visser, G. and Ogarko, V. and Lindsay, Mark and Jessell, Mark },
     title = { Utilisation of stochastic {MT} inversion results to constrain gravity inversion },
 booktitle = { 82nd {EAGE} {Annual} {Conference} \& {Exhibition} },
      year = { 2021 },
     pages = { 1--5 },
 publisher = { European Association of Geoscientists \& Engineers },
   address = { Amsterdam, The Netherlands, },
       url = { https://www.earthdoc.org/content/papers/10.3997/2214-4609.202113105 },
       doi = { 10.3997/2214-4609.202113105 },
  abstract = { We introduce a methodology developed with the objective of exploiting complementary information between 1D magnetotelluric (MT) and gravity inversion. To maintain flexibility, we propose a cooperative workflow leveraging standalone inversions. We first perform 1D probabilistic MT inversion to obtain ensembles of models representative of the measurements. We then use the probabilities of presence of the different rock units derived from these ensembles to divide the studied area into domains characterized by positive probabilities to observe the different rock units. Thirdly, these domains are used to constrain the inversion of gravity data by restricting density values accordingly with the rock units of each domain obtained from MT-derived probabilities. We perform the synthetic proof-of-concept using a realistic model based on the framework of a region in the Mansfield area (Victoria, Australia). Results reveal that our methodology can improve subsurface imaging and can be applied to field data. }
}