Hello there,
Anyone experienced with this? The main question I have is, how do I determine Werner solution parameters ( Min./Max window length, window expansion increment, window shift increment, cut-offs). I know it requires iteration. I am mainly looking to understand the basement structure on a regional scale, not small anomalies. I have a 400km*400km grid. Based on this, any idea what can be some good values to start with.
I have started with Min. window length, Max window length, window expansion increment, window shift increment with 2000m, 100000m, 1000m, 1000m respectively. Played with parameters too by changing those but not getting good clusters.
P.S: I have read Ku and Sharp (1983), but was not able to make decisions on these values based on that.
Have a look at https://www.spgindia.org/conference/5thconf_hyd04/5475.PDF
https://www.geosoft.com/media/uploads/resources/technical-notes/montaj_depthtobasement_technote.pdf
https://www.tandfonline.com/doi/abs/10.1080/14400958908527954?needAccess=true&journalCode=taje20
Dear Md. Nabidul Hasan:
Please, let me share a couple of suggestions about your question related with Werner deconvolution for basement depth estimation. I will put emphasis in two topics that you focus:
1- You are trying to illuminate the basement at regional scale,
2- You have a 400 km by 400 km grid area.
The first idea, before you try to adjust the Werner´s solution parameters, I recommend:
You are talking about a regional-scale, perhaps, is possible that your data come from an aeromagnetic / aerogravity investigation. Therefore, you must check the survey design and acquisition techniques due that the most aerogravity and aeromagnetic surveys require that measurements be taken on a horizontal plane, but factors such as changes in aircraft, wind-speed and direction, a draped or mean-terrain clearance survey, may result in flight height and significant variation in the Werner final solution. A very good paper to understand this phenomenon that I recommend to read is:
“Werner deconvolution for variable altitude aeromagnetic data” by Ostrowski et al., Geophysics 58 (10), 1481-1490, September 1993.
The second idea is, before any kind of quantitative solution response, that you take a little time of study to inspection the main shapes of the grid, such as width, curvature, azimuths, amplitude, wavelength ... to try to predict as a qualitative answer without any quantitative approach. This is very useful if you are working with magnetic data, instead gravity ones.
Then you will apply a quantitative evaluation of the basement by automatic processes and techniques, such as Werner, or Euler, or SPI.
But, it is more interesting to explain that in my own study-cases experiences, before that, you will have a very robust response using the radial average power spectrum (RAPS) techniques, even with the limitations that professor and authority such as prof. Alan Reid explain in his website. You will have an idea, maybe with a little more time if you have, to a quickly inspection about the depths and wavelength of your whole area. But, your area is huge. Also, if your inspection of the main shapes push you that the idea / hypothesis that you have two, three, four geological-scenarios, with a little more time you should get the RAPS answer of each-ones, and compare the amplitude, wavelength, etc, etc. to get a better quantitative interpretation. You will have a very important value from this exercice: the potential cutoff that you will use in SPI, for example.
After that, due that you are looking for the basement depth estimation, it means: the thickness of the sedimentary column, it is a good hypothesis to start to imaging that the basement in your vast area (160,000 sqkm!!! = Uruguay !!! Tunez" at all!!!) are controlled at regional-scale by individual / system faults. Also, the first inspection of the gravity and magnetic anomaly patterns, in a fault-block structural-scenario, will give you an idea not only of your Werner´s parameters, but also the azimuth of your profile, a critical issue to understand the choice of the constraints. It means: profiles should be straight, lie perpendicular to strike – fault azimuth, profiles along the sides of anomalies do not tell you a real answer, etc.
On the other hand: you have a grid as input! but you use just a 2D solution? If it is your case, you must need external constraints to choose the best model. You must use wells, outcrops, seismic… Do you have this information?
Therefore, I recommend trying to use your whole data, such as 3D Euler deconvolution / Source Parameter Imaging SPI methods to estimate and characterize the depths of the basement from gridded data. These techniques use first-order X, Y, and Z derivatives to determine location and depth of targets such as sphere, cylinder, thin dike, contact, characterized by an explicit structural index. For example, the Euler deconvolution solution will allow you to delineate basement depth and lineaments or faults, and do not require information about the magnetization vector. In the SPI, any solution output can be cutoff by a depth value, also, any solutions deeper than this cutoff level will be rejected. And you have an idea about this value, because you run before the RAPS! In both cases, the solutions are obtained from new databases in which the reference values can be gridded, and the final answer is not a profile, it is a grid, also, the imaging of the "magnetic and/or gravity basement" is more confidence.
I recommend to read the following two papers about 3D solutions:
D. T. Thompson, “EULDPH: a new technique for making computer-assisted depth estimates from magnetic data,” Geophysics, vol. 47, no. 1, pp. 31–37, 1982).
A. B. Reid, J. M. Allsop, H. Granser, A. J. Millett, and I. W. Somerton, “Magnetic interpretation in three dimensions using Euler deconvolution,” Geophysics, vol. 55, pp. 80–90, 1990.
Finally, perhaps the best solution for your regional study-case are a cloud-based 3D inversion / 3D modeling technology, an iterative surface inversion that calculate the gravity and / or magnetic effect of all the blocks, a real voxel solution, an advances exploration tool for improved geologic understanding.
I know that my answer do not response your question about the Werner´s parameters, but I try to help you to compared the whole scenario before use one or another method.
My best regards,
Mario E. Sigismondi
No, it gives non unique solutions. I then used tilt depth method which worked for my purpose.
- Badges
- Science topic
- Similar topics
- Geoscience
- Cartography
- Mapping
- Maps