Inversion of airborne electromagnetic data in 2.5D
In this work, we implement an inversion algorithm for airborne electromagnetic (AEM) data in the frequency domain by using 2D conductivity models. First, we discretize the 2.5D Maxwell's equations on a staggered grid and test the numerical accuracy of the forward solution. The inverse problem i...
| Main Author: | |
|---|---|
| Language: | English |
| Published: |
University of British Columbia
2012
|
| Online Access: | http://hdl.handle.net/2429/43416 |
| Summary: | In this work, we implement an inversion algorithm for airborne electromagnetic (AEM) data in the frequency domain by using 2D conductivity models. First, we discretize the 2.5D Maxwell's equations on a staggered grid and test the numerical accuracy of the forward solution. The inverse problem is then solved by regularized minimization approach using the limited memory BFGS variant of the quasi-Newton method. Next, EM responses from a synthetic 2D conductivity model are inverted to validate the algorithm. Finally, we use the algorithm on an AEM field dataset from a RESOLVE survey and compare the inversion results to those obtained from a well-established 1D implementation. |
|---|