Re: Filtering gravity data

If you have a profile you can use "filter1d".
If you have a grid you can use "grdfilter" or "grdfft".

Filtering of gravity is a deep and rich subject and there isn't a simple answer to your problem. Classical gravity interpretation methods, such as in a book like Applied Geophysics by W M Telford et al., will have formulas for the anomaly due to a tilted slab. You can use these and the depth range of the slab that you expect to come up with a typical width of the anomaly, and then use a filter that will reject wavelengths much shorter than that width.

Another approach uses correlation techniques. You first try to remove the gravity effects that can be correlated with the topography (notice this isn't quite the same as the free-air or Bouguer corrections), and then see what is left.

Unfortunately gravity suffers a non-uniqueness, and it may not be possible to infer the slab depth and tilt from your data. I think looking at the formula for a tilted slab (which may be in Telford if I remember) will give you a sense of what you should be expecting in the gravity data.

Good luck.

Walter


On Mar 10, 2014, at 9:25 PM, Mr Abhash Kumar <[email protected]> wrote:

> Hi All,
> I recently determined the Nazca slab geometry between 10 degree to 18 degree south in Peru using local earthquake data. I found some evidence of slab steepening between 11 degree to 14 degree south at nearly 100km depth. This result is different than all earlier results of expected flat slab at 100km depth in that region (Cahill and Isacks, 1992; Hayes et al., 2012). I was wondering if the slab is steep as opposed to flat then it should be noticeable in the bouguer gravity profile. I have the bouguer anomaly grid file from the Department of Mining, Peru. In order to do a course forward modelling of the observed bouguer anomaly, I need to filter the data. I would be extremely grateful if anyone can advice me about the frequency range to filter the data to see  the  long  wavelength effects of the slab. Is there any GMT command to do the filtering of gravity data. 
> 
> With regards
> Abhash kumar
> PHD candidate
> University of North Carolina at Chapel Hill
> 

To unsubscribe, send the message "signoff gmt-help" to [email protected]