|
»
gmtmath TCRIT
|
View:
New views
18 Messages
—
Rating Filter:
Alert me
|
 |
gmtmath TCRIT
by Paul Krusic
::
Rate this Message:
I was wondering if anyone could explain the usage of gmtmath, particularly how to pass more than one argument to a function like TCRIT?
I have a file of 3 columns, and 6 rows tst.xyz and I would like to know the 90 or 95% critical level of z? I have tried a few times and each time I get this same error message, e.g., gmtmath -C3 800_xyz TCRIT .90, 6 = tst Operation "TCRIT" requires 2 operands What am I doing wrong? Thank you, pjk Bert Bolin Centre for Climate Research Department of Physical Geography and Quaternary Geology Stockholm UniversitySE-106 91 Stockholm +46 816-4713 http://www.bbcc.su.se |
||
|
|
Re: gmtmath TCRIT
by Peter Schmidt-3
::
Rate this Message:
Hi Paul
I'm not really familiar with the use of TCRIT, but a few pointers. 1. Your file according to your mail is tst.xyz (and not 800_xyz as in your command) 3. You probably want to drop the comma sign after TCRIT .90 2. The syntax when using polish notation, a single operator needing two operands (e.g. A+B) would be: A B + hence in your case you would have something like: gmtmath -C3 800_xyz .90 6 TCRIT = tst Hope this helps enough to get you runing regP Paul Krusic wrote: > I was wondering if anyone could explain the usage of gmtmath, > particularly how to pass more than one argument to a function like TCRIT? > I have a file of 3 columns, and 6 rows tst.xyz and I would like to > know the 90 or 95% critical level of z? I have tried a few times and > each time I > get this same error message, e.g., > > gmtmath -C3 800_xyz TCRIT .90, 6 = tst > / > Operation "TCRIT" requires 2 operands > > What am I doing wrong? > > Thank you, > pjk > > Paul J. Krusic/ > / > / > / > TRL-LDEO > Rt.9W Palisades, NY > pjk@... <mailto:pjk@...> > http://www.ldeo.columbia.edu/trl > > / > /Bert Bolin Centre for Climate Research/ > /Department of Physical Geography and Quaternary Geology/ > /Stockholm University/ > /SE-106 91 Stockholm/ > /+46 816-4713/ > / paul.krusic@... <mailto:paul.krusic@...>/ > http://www.bbcc.su.se > */ > > /* > > > > > To unsubscribe, send the message "signoff gmt-help" to > listserv@... -- ******************************************************************************** Peter Schmidt Tel: +46-18-4717104 PhD Student Mobile: +46-73-3190975 Dept. of Earth Sciences e-mail: peter.schmidt@... Geophysics http://www.geofys.uu.se/?q=view/home/75 Uppsala University Villavagen 16 SE-75236 Uppsala ******************************************************************************** To unsubscribe, send the message "signoff gmt-help" to listserv@... |
||
|
|
Re: gmtmath TCRIT
by Paul Krusic
::
Rate this Message:
Thank you Peter, Yes, I did misrepresent the file names in my message. Using the correct name this new suggestion I get a new warning/error message. File <tst> is created but it only contains the contents of file 800_003_xyz. What does the warning regrading operands mean? Sincerely, pjk gmtmath -C3 800_003_xyz .90 6 TCRIT = tst gmtmath: Warning: 1 more operands left on the stack! Paul J. Krusic Bert Bolin Centre for Climate Research Department of Physical Geography and Quaternary Geology Stockholm UniversitySE-106 91 Stockholm +46 816-4713 http://www.bbcc.su.seOn Oct 22, 2009, at 7:52 AM, Peter Schmidt wrote: gmtmath -C3 800_xyz .90 6 TCRIT = tst To unsubscribe, send the message "signoff gmt-help" to listserv@... |
||
|
|
Re: gmtmath TCRIT
by Peter Schmidt-3
::
Rate this Message:
Hi again
I'm back to talk about things I really do not have a clue about so... Anyway the opreands are your input variables to your operator, TCRIT apparantelly takes two (alpha and n according to the man page) the command you are using now has three operands: 1. the file 800_003_xyz 2. .90 3. 6 one of them has to go, but to cite my statement above, not knowing how TCRIT works (what is alpha and what is n?).... regP Paul Krusic wrote: > Thank you Peter, > Yes, I did misrepresent the file names in my message. Using the > correct name this new suggestion > I get a new warning/error message. File <tst> is created but it only > contains the contents of file > 800_003_xyz. > What does the warning regrading operands mean? > > Sincerely, > pjk > > gmtmath -C3 800_003_xyz .90 6 TCRIT = tst > gmtmath: Warning: 1 more operands left on the stack! > > / > Paul J. Krusic/ > / > / > / > TRL-LDEO > Rt.9W Palisades, NY > pjk@... <mailto:pjk@...> > http://www.ldeo.columbia.edu/trl > > / > /Bert Bolin Centre for Climate Research/ > /Department of Physical Geography and Quaternary Geology/ > /Stockholm University/ > /SE-106 91 Stockholm/ > /+46 816-4713/ > / paul.krusic@... <mailto:paul.krusic@...>/ > http://www.bbcc.su.se > */ > > /* > > > > > On Oct 22, 2009, at 7:52 AM, Peter Schmidt wrote: > >> gmtmath -C3 800_xyz .90 6 TCRIT = tst > > To unsubscribe, send the message "signoff gmt-help" to > listserv@... -- ******************************************************************************** Peter Schmidt Tel: +46-18-4717104 PhD Student Mobile: +46-73-3190975 Dept. of Earth Sciences e-mail: peter.schmidt@... Geophysics http://www.geofys.uu.se/?q=view/home/75 Uppsala University Villavagen 16 SE-75236 Uppsala ******************************************************************************** To unsubscribe, send the message "signoff gmt-help" to listserv@... |
||
|
|
Re: gmtmath TCRIT
by Paul Wessel-2
::
Rate this Message:
Paul-
The TCRIT function accepts the alpha level and degrees of freedom. I assume that is the 0.9 and 6. So what do you expect the data file argument to achieve (currently it is just sitting there with the result of TCRIT hence 2 items on the stack and the = operator will save the one on top. -p On Oct 21, 2009, at 8:55 PM, Paul Krusic wrote:
|
||
|
|
Re: gmtmath TCRIT
by Peter Schmidt-3
::
Rate this Message:
Paul Wessel wrote:
> Paul- > > The TCRIT function accepts the alpha level and degrees of freedom. I > assume that is the 0.9 and 6. So what do you expect the data file > argument to achieve (currently it is just sitting there with the > result of TCRIT hence 2 items on the stack and the = operator will > save the one on top. Suggestion, try gmtmath -C3 .90 800_003_xyz TCRIT = tst \p > -p > > On Oct 21, 2009, at 8:55 PM, Paul Krusic wrote: > >> Thank you Peter, >> Yes, I did misrepresent the file names in my message. Using the >> correct name this new suggestion >> I get a new warning/error message. File <tst> is created but it only >> contains the contents of file >> 800_003_xyz. >> What does the warning regrading operands mean? >> >> Sincerely, >> pjk >> >> gmtmath -C3 800_003_xyz .90 6 TCRIT = tst >> gmtmath: Warning: 1 more operands left on the stack! >> >> / >> Paul J. Krusic/ >> / >> / >> / >> TRL-LDEO >> Rt.9W Palisades, NY >> pjk@... <mailto:pjk@...> >> http://www.ldeo.columbia.edu/trl >> >> / >> /Bert Bolin Centre for Climate Research/ >> /Department of Physical Geography and Quaternary Geology/ >> /Stockholm University/ >> /SE-106 91 Stockholm/ >> /+46 816-4713/ >> / paul.krusic@... <mailto:paul.krusic@...>/ >> http://www.bbcc.su.se >> */ >> >> /* >> >> >> >> >> On Oct 22, 2009, at 7:52 AM, Peter Schmidt wrote: >> >>> gmtmath -C3 800_xyz .90 6 TCRIT = tst >> >> To unsubscribe, send the message "signoff gmt-help" to >> listserv@... <mailto:listserv@...> > > To unsubscribe, send the message "signoff gmt-help" to > listserv@... -- ******************************************************************************** Peter Schmidt Tel: +46-18-4717104 PhD Student Mobile: +46-73-3190975 Dept. of Earth Sciences e-mail: peter.schmidt@... Geophysics http://www.geofys.uu.se/?q=view/home/75 Uppsala University Villavagen 16 SE-75236 Uppsala ******************************************************************************** To unsubscribe, send the message "signoff gmt-help" to listserv@... |
||
|
|
Interpolating irregular data on a different heights
by yacob sen
::
Rate this Message:
To unsubscribe, send the message "signoff gmt-help" to listserv@... |
||
|
|
Re: Interpolating irregular data on a different heights
by Ali Dehghani
::
Rate this Message:
nearneighbor is another possibility.
Best regards Ali Dehghani yacob sen schrieb:
-- ****************************************************** Dr. G.Ali Dehghani Institute of Geophysics Centre for Marine and Atmospheric Sciences (ZMAW) Faculty of Mathematics, Informatics and Natural Sciences University of Hamburg Bundesstr. 55 D-20146 Hamburg, Germany Tel.: +49 40 42838 2978 Fax : +49 40 42838-2978 or -5441 e-mail: ali.dehghani@... http://www.geophysics.zmaw.de/index.php?id=103 ******************************************************To unsubscribe, send the message "signoff gmt-help" to listserv@... [ali_dehghani.vcf] begin:vcard fn:Dr. G. Ali Dehghani n:Dehghani;Dr. G. Ali email;internet:ali.dehghani@... tel;work:040-428 38 2978 tel;fax:040-428 38 1978 version:2.1 end:vcard |
||
|
|
Re: gmtmath TCRIT
by Paul Krusic
::
Rate this Message:
Dear P,
Yes, I have no idea what to expect, because I do understand the gmtmath argument syntax, hence my writing the forum. Now, I still have no more understanding on the use of this procedure than I did in the beginning so in less time then it took to find out I wrote a little fortran program to calculate the probabilities I was after and moved on. Fortunately, GMT still makes lovely plots, thank you for that. But, while I have your attention, how do you like to see GMT cited in a publication, particularly the nearneighbor gridding program? I know you provide citations for some of the spline routines, I was just wondering about the more general elements of the collection. Sincerely, pjk Paul J. Krusic Bert Bolin Centre for Climate Research Department of Physical Geography and Quaternary Geology Stockholm UniversitySE-106 91 Stockholm +46 816-4713 http://www.bbcc.su.seOn Oct 22, 2009, at 10:53 AM, Paul Wessel wrote:
|
||
|
|
Re: Interpolating irregular data on a different heights
by Joaquim Luis-2
::
Rate this Message:
Yacob,
This matter is a bit more complicated than just playing with the merits of different interpolation algorithms. If your gravity data was acquired at different altitudes, you must first reduce your measures to a common datum before you can do any analysis. (Free Air, Bougeur simple-or-complete corrections, etc). Only after this has been done is that you worry with the grid interpolations.
Joaquim Luis
To unsubscribe, send the message "signoff gmt-help" to listserv@...
On Thu, Oct 22, 2009 at 11:46 AM, yacob sen <yacob_123@...> wrote:
|
||
|
|
Re: Interpolating irregular data on a different heights
by yacob sen
::
Rate this Message:
To unsubscribe, send the message "signoff gmt-help" to listserv@... |
||
|
|
Re: Interpolating irregular data on a different heights
by Joaquim Luis-2
::
Rate this Message:
So the data no longer represent measures at different heights. And yes, use surface but whatever method you use there are no miracles and you must check that no unreasonable values are obtained by the interpolator far from the location of the data points.
Nearneighbor is not a good choice for this and it would risk to fill nodes with NaNs, which you cannot in grids that are to be FFTed.Â
To unsubscribe, send the message "signoff gmt-help" to listserv@...
On Thu, Oct 22, 2009 at 3:25 PM, yacob sen <yacob_123@...> wrote:
|
||
|
|
Re: gmtmath TCRIT
by Paul Wessel-2
::
Rate this Message:
Hi Paul-
Typically, people will just cite the GMT EOS articles unless a specific publication is given in the man pages (e.g., surface, greenspline). For nearneighbor you could say things like "produced by nearest-neighbor gridding using the nearneighbor application in the Generic Mapping Tolos [Wessel & Smith, 1998]" . Cheers, Paul On Oct 22, 2009, at 1:05 AM, Paul Krusic wrote:
|
||
|
|
Re: Interpolating irregular data on a different heights
by Walter H. F. Smith
::
Rate this Message:
A subtle detail is the tension value to use in surface. Using a
tension of zero will minimize the high-wavenumber power, equivalent to pushing the sources of the field as low as possible. This will lead to the most stable downward continuations. w On Oct 22, 2009, at 11:22 AM, J Luis wrote: > So the data no longer represent measures at different heights. And > yes, use surface but whatever method you use there are no miracles > and you must check that no unreasonable values are obtained by the > interpolator far from the location of the data points. > Nearneighbor is not a good choice for this and it would risk to > fill nodes with NaNs, which you cannot in grids that are to be FFTed. > > > On Thu, Oct 22, 2009 at 3:25 PM, yacob sen <yacob_123@...> > wrote: > Dear Joaquim > > Thank you for your reply. Yes, I have done all sorts of complicated > terrain correction at each gravity station. I have also subtracted > the long and intermediate wavelengths from satellite gravity > models. What I have left with is a very tiny residual gravity > disturbance not anomaly in my particular case. > > Now I want to do a Fast Fourier Transform (FFT) on the data set to > continue upward/downward uisng some type of FFT which basically > needs data on a grid. To feed my FFT program I need a gridded > values of the tiny gravity disturbance and hence the question of > which type(s) of interpolation schems comes to GMT forum. > To unsubscribe, send the message "signoff gmt-help" to > listserv@... To unsubscribe, send the message "signoff gmt-help" to listserv@... |
||
|
|
Re: Interpolating irregular data on a different heights
by yacob sen
::
Rate this Message:
To unsubscribe, send the message "signoff gmt-help" to listserv@... |
||
|
|
Re: Interpolating irregular data on a different heights
by Paul Wessel-2
::
Rate this Message:
That was Walter. Anyway, I assume the difference reflects the fact that grdfft uses the exact solution (exp(-k*h)) whereas your Taylor stops at order 2 in an infinite series.
-p On Oct 22, 2009, at 1:51 PM, yacob sen wrote:
|
||
|
|
Re: Interpolating irregular data on a different heights
by yacob sen
::
Rate this Message:
To unsubscribe, send the message "signoff gmt-help" to listserv@... |
||
|
|
Re: Interpolating irregular data on a different heights
by Walter H. F. Smith
::
Rate this Message:
Yes, the gravity field is a function of x and y so the Taylor
expansion properly consists of partial derivatives in x and y. For the second order term, there are also cross-derivatives d2u/dxdy). w On Oct 23, 2009, at 05:29, yacob sen wrote: > Dear Walter, Paul > > Sorry that I addressed my reply to Paul. Having said that, I have > checked the contributions of the third order and it is indeed very > small. I do not think the difference with FFT is reflected by the > third order term of the Taylor expansions. I was thinking that the > difference might come from only taking the radial derivatives, > perhaps only perhaps that I might need latitudinal/longitudinal > derivatives in the taylor expansions. But calculating those > derivatives are another matter. > > Regards > Yacob > > > --- On Fri, 10/23/09, Paul Wessel <p.wessel@...> wrote: > > From: Paul Wessel <p.wessel@...> > Subject: Re: [GMT-HELP] Interpolating irregular data on a different > heights > To: GMT-HELP@... > Date: Friday, October 23, 2009, 8:25 AM > > That was Walter. Anyway, I assume the difference reflects the fact > that grdfft uses the exact solution (exp(-k*h)) whereas your Taylor > stops at order 2 in an infinite series. > > -p > > On Oct 22, 2009, at 1:51 PM, yacob sen wrote: > >> Dear Paul >> >> Thank you for your insight. If I may ask, let me pose the >> following question some how related to my previous question but >> loosely. I have upward continued the residual gravity with out >> including long and intermidiate wavelengths through the FFT >> routine and also did the same thing a taylor expansion up to the >> second order only in the radial direction: >> >> dg(h)= dg(h=0) + dg/dh(0) *(h) -0.5*h*h*dg''/dh''(0) >> >> The comparion between the taylor expansion to some height h and >> FFT to the same height h gives different result. The FFT upward >> continuation method gave the correct result. Do you have a reason >> why this is the case ? >> >> Regards >> yacob >> >> >> --- On Fri, 10/23/09, Walter H. F. Smith >> <Walter.HF.Smith@...> wrote: >> >> From: Walter H. F. Smith <Walter.HF.Smith@...> >> Subject: Re: [GMT-HELP] Interpolating irregular data on a >> different heights >> To: GMT-HELP@... >> Date: Friday, October 23, 2009, 12:21 AM >> >> A subtle detail is the tension value to use in surface. Using a >> tension of zero will minimize the high-wavenumber power, >> equivalent to pushing the sources of the field as low as >> possible. This will lead to the most stable downward continuations. >> >> w >> >> On Oct 22, 2009, at 11:22 AM, J Luis wrote: >> >> > So the data no longer represent measures at different heights. >> And yes, use surface but whatever method you use there are no >> miracles and you must check that no unreasonable values are >> obtained by the interpolator far from the location of the data >> points. >> > Nearneighbor is not a good choice for this and it would risk to >> fill nodes with NaNs, which you cannot in grids that are to be FFTed. >> > >> > >> > On Thu, Oct 22, 2009 at 3:25 PM, yacob sen <yacob_123@...> >> wrote: >> > Dear Joaquim >> > >> > Thank you for your reply. Yes, I have done all sorts of >> complicated terrain correction at each gravity station. I have >> also subtracted the long and intermediate wavelengths from >> satellite gravity models. What I have left with is a very tiny >> residual gravity disturbance not anomaly in my particular case. >> > >> > Now I want to do a Fast Fourier Transform (FFT) on the data set >> to continue upward/downward uisng some type of FFT which >> basically needs data on a grid. To feed my FFT program I need a >> gridded values of the tiny gravity disturbance and hence the >> question of which type(s) of interpolation schems comes to GMT forum. >> > To unsubscribe, send the message "signoff gmt-help" to >> listserv@... >> >> To unsubscribe, send the message "signoff gmt-help" to >> listserv@... >> >> To unsubscribe, send the message "signoff gmt-help" to >> listserv@... > > To unsubscribe, send the message "signoff gmt-help" to > listserv@... > > To unsubscribe, send the message "signoff gmt-help" to > listserv@... To unsubscribe, send the message "signoff gmt-help" to listserv@... |
| Free embeddable forum powered by Nabble | Forum Help |
