Please use this identifier to cite or link to this item:
|Scopus||Web of Science®||Altmetric|
|Title:||Bootstrap resampling as a tool for uncertainty analysis in 2-D magnetotelluric inversion modelling|
|Citation:||Geophysical Journal International, 2015; 203(1):92-106|
|Publisher:||Oxford University Press|
|Sebastian Schnaidt, Graham Heinson|
|Abstract:||Uncertainty estimation is a vital part of geophysical numerical modelling. There exist a variety of methods aimed at uncertainty estimation, which are often complicated and difficult to implement. We present an inversion technique that produces multiple solutions, based on bootstrap resampling, to create a qualitative uncertainty measure for 2-D magnetotelluric inversion models. The approach is easy to implement, can be used with almost any inversion code, and does not require access to the inversion software's source code. It is capable of detecting the effect of data uncertainties on the model result rather than just analysing the effect of model variations on the model response. To obtain uncertainty estimates for an inversion model, the original data set is resampled repeatedly and alternate data set realizations are created and inverted. This ensemble of solutions is then statistically analysed to determine the variability between the different solutions. The process yields interpretable uncertainty maps for the inversion model and we demonstrate its effectiveness to qualitatively quantify uncertainty in synthetic model tests and a case study.|
|Keywords:||Instability analysis; inverse theory; magnetotellurics|
|Rights:||© The Authors 2015. Published by Oxford University Press on behalf of The Royal Astronomical Society.|
|Appears in Collections:||Geology & Geophysics publications|
Files in This Item:
There are no files associated with this item.
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.