The G-GG method: Improving inversion of Gravity Gradiometer Data
TaronishPithawala
Posts: 230 mod
in Oasis montaj
G-GG (gravity referenced gravity gradiometer) is a method of combining regional gravity with gravity gradiometer data to model discreet targets within a regional setting. While gravity gradiometer surveys yield smaller spatial wavelength anomalies and gravimeter surveys yield longer spatial wavelength anomalies, G-GG models both. Even if all you have is gravity gradient data, regional gravity data is readily available through Geosoft Seeker, or public domain websites . Using Geosoft VOXI Earth Modelling, G-GG is easy to apply and requires no additional scripting of inversion code or conditioning of the input data. It better resolves discreet targets within a regional context compared to conventional gravity gradient trend removal methods (e.g. mean, linear, quadratic trend removal). Joint inversion of gravity and gravity gradient data is possible but can be problematic when dealing with different sampling densities in the data and levels of error. For this reason we think G-GG is far more efficient and effective in practice.
An illustrative example starts with a synthetic model to simulate the gravity and FALCON TM AGG responses (GNE and GUV):
The Gz survey responds to the bulk density and the data is dominated by the crystalline/sediment contrast. The GNE data on the other hand is insensitive to the N-S striking cover, while the GUV data are very sensitive to it. Conventional approaches to removing a consistent, simple linear or quadratic trend from all channels of the gravity gradiometer data would prove difficult as the trend in the GUV data does not manifest in the GNE data.
The G-GG method first solves the inverse problem for the Gz data to create a regional model. This model is then used as a parameter reference model for the gravity gradiometer inversion. The parameter reference model serves as a guide for the gradiometer inversion to trend towards in the absence of influence from the short wavelength GNE and GUV data. The effect is illustrated in the images below:
Read More
This work was submitted to the Airborne Gravity Workshop at the ASEG-PESA 2016 conference in Adelaide, Australia. The submitted work included the application of the G-GG method to the R.J Smith Test site in Kauring, Australia. The authors are Ellis, R.E; Pouliquen, G.; Pithawala, T.; and MacLeod, I - the abstract is attached.
An illustrative example starts with a synthetic model to simulate the gravity and FALCON TM AGG responses (GNE and GUV):
The Gz survey responds to the bulk density and the data is dominated by the crystalline/sediment contrast. The GNE data on the other hand is insensitive to the N-S striking cover, while the GUV data are very sensitive to it. Conventional approaches to removing a consistent, simple linear or quadratic trend from all channels of the gravity gradiometer data would prove difficult as the trend in the GUV data does not manifest in the GNE data.
The G-GG method first solves the inverse problem for the Gz data to create a regional model. This model is then used as a parameter reference model for the gravity gradiometer inversion. The parameter reference model serves as a guide for the gradiometer inversion to trend towards in the absence of influence from the short wavelength GNE and GUV data. The effect is illustrated in the images below:
Read More
This work was submitted to the Airborne Gravity Workshop at the ASEG-PESA 2016 conference in Adelaide, Australia. The submitted work included the application of the G-GG method to the R.J Smith Test site in Kauring, Australia. The authors are Ellis, R.E; Pouliquen, G.; Pithawala, T.; and MacLeod, I - the abstract is attached.
Customer Success Manager - Geophysical Modelling
2
Comments
-
Excellent example!0
-
Thanks @NickWilliams1!
Have you had a chance to try it out?Customer Success Manager - Geophysical Modelling0
This discussion has been closed.