Published September 4, 2023 | Version v1
Conference paper Open

Estimating noise in AEM data

Creators

  • 1. CSIRO, 26 Dick Perry Ave, WA, AUSTRALIA, 6151, Aaron.davis@csiro.au

Description

In this paper, I discuss a method to obtain reliable noise estimates for airborne electromagnetic (AEM) surveys based on the reversible-jump Markov chain Monte Carlo method. In addition to estimating electrical conductivity and thickness using 1D layered-earth models, the method provides estimates of the additive error required to make all measurements of a repeat line agree. The noise estimates can also be obtained from a single line where repeat line information is unavailable. The resulting additive noise estimates then can be used in a general deterministic inversion. Analysis of inversions shows that model regularisation has little effect at depths where the data is informative. This improves the reliability of the inverted models, since it is the noise-adjusted data which is informing the model.

Other

Open-Access Online Publication: October 30, 2023

Files

AEM2023_ID012.pdf

Files (777.9 kB)

Name Size Download all
md5:28c41327f4966c7cbbe9979f2f3adb36
777.9 kB Preview Download