Re: Detemining gradient at individual points

Hi Abhash

One way you can do that is to calculate the gradient of the whole grid and then use grdtrack to get the information for only the points you are interested in.

Cheers
Joey

----- Original Message -----
From: "Mr Abhash Kumar" <[email protected]>
To: [email protected]
Sent: Wednesday, February 12, 2014 9:10:48 PM
Subject: Re: [GMT-HELP] Detemining gradient at individual points


Hi, 
I do have slab grid file in terms of latitude, longitude and depth in km. I don't want to determine gradient at every single latitude and longitude, which I have in the slab grid file. My slab grid file (slabsurf2.grd) has 345639 data points in it and extended between 10-18 degree south and 66-78 degree west. I want to determine gradient at some user defined latitude and longitude (example: 74.567 degree west, 12.345 degree south), which falls with in the range of extension of the slab but doesn't coincide exactly with any specific point present in the slab grid file. Is there any way to determine gradient at a user specified location. 

Thanks 
Abhash kumar 
University of North Carolina at Chapel Hill 




From: World-Wide GMT Usage and Help Mailing List <[email protected]> on behalf of Paul Wessel <[email protected]> 
Sent: Wednesday, February 12, 2014 9:50 PM 
To: [email protected] 
Subject: Re: [GMT-HELP] Detemining gradient at individual points 

If the grid itself is in in dimensions of lat and lon then you need to supply -fg or -M, depending on GMT version, to convert degrees to meters. Also if the depth to the slab is in km then you need to account for that before taking atan since your slope would be in km/m. So perhaps something like 




grdgradient slabsurf2.grd -Dn -Sgradient.grd -V 

grdmath gradient.grd 1000 MUL ATAN = dips.grd 



On Feb 12, 2014, at 4:28 PM, Mr Abhash Kumar < [email protected] > wrote: 





Hi, 
I am grateful for your concerned reply Dr. Wessel. I actaully have the slab grid file and latitude and longitude of few points. I need to find the dip of the slab at exactly those points. I am using the grdgradient command in the following manner: 


grdgradient slabsurf2.grd -Dn -Sgradient.grd -V 


I don't know where in the above command line, I need to put the latitude and longitudes of those points, where I actually want to calculate the dip of the slab. Should I make a separate text file having coordinates of those points and use that text file in the above command somewhere. I would be extremely grateful for any suggestions. 



Thanks 
Abhash kumar 
University of North Carolina at Chapel Hill 



From: World-Wide GMT Usage and Help Mailing List < [email protected] > on behalf of Paul Wessel < [email protected] > 
Sent: Wednesday, February 12, 2014 8:55 PM 
To: [email protected] 
Subject: Re: [GMT-HELP] Detemining gradient at individual points 

Hi Abhash- 


That is a good question. However, grdgradient with -D and -S can also return the direction of the gradient or the strike (-Dn), as well as the absolute value of the gradient. If your units are correct (all meters or geographic with -fg (or -M in GMT4)) then take atan of the gradient and that should give you dip, no? 


-p 



On Feb 12, 2014, at 3:37 PM, Mr Abhash Kumar < [email protected] > wrote: 






Hi, 
I have a 2-D grid file for the subducting slab surface. I need to find dip of the slab at a particular latitude and longitude. I looked at the man page of grdgradient and it can only calculate the gradient along a particular azimuth. I would be extremely grateful for any suggestions in this regards. 

Thanks 
Abhash kumar 
PHD candidate 
University of North Carolina at Chapel Hill 
To unsubscribe, send the message "signoff gmt-help" to [email protected] 
To unsubscribe, send the message "signoff gmt-help" to [email protected] To unsubscribe, send the message "signoff gmt-help" to [email protected] 
To unsubscribe, send the message "signoff gmt-help" to [email protected] To unsubscribe, send the message "signoff gmt-help" to [email protected] 

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