Bug #1229

psmeca color spilling over + modifying text string

Added by Januka 6 months ago. Updated 6 months ago.

Status:FeedbackStart date:2018-03-28
Priority:NormalDue date:
Assignee:-% Done:


Target version:-
Affected version:5.4.2 Platform:


I am trying to plot a moment tensor with the compression quadrant colored black. However, the color spills over the focal sphere. Note that my moment tensor has a large isotropic component, but this is intentional. I am using Version 5.4.2 (r18461) and here's my code. How do I fix this? Also, how do you change the text string attributes (position/size/font). There are no notes on this in the manual. Thank you!


gmt psbasemap -J$proj -R$range -Bxa10 -Bya10 -BSWNE -V -K -P > $outps

gmt psmeca -J -R -Sm0.25i -W0.3p -G$C_col -E$T_col -C0.3p \
-T0 -Fa0.025i/cc -Fe$T_col -Fg$C_col -N -V -P -O >> $outps << END
146.203 -38.2828 10.2 -0.0000 -2.2350 -0.5587 -1.2374 -0.3892 -0.5304 24 151.203 -37.2828 xyz

1996-Seismological Research Letters.pdf (2.23 MB) Paul, 2018-04-05 09:55


#1 Updated by Paul 6 months ago

Apparently the large isotropic moment is a problem. The original developers of psmeca are not very involved anymore but wrote:

Here is the snippet from utilmeca.c

f = - v[1] / v[d];
    iso = vi / v[d];
    jp_flag = 0;
    djp = -1;
    mjp = -1;

    /* Cliff Frohlich, Seismological Research letters,
      * Vol 7, Number 1, January-February, 1996
      * Unless the isotropic parameter lies in the range
      * between -1 and 1 - f there will be no nodes whatsoever */

Frolich, C. (1996), Cliff's Nodes Concerning Plotting Nodal Lines for P, Sh and Sv, Seismological Research Letters, 67, 16-24. http://dx.doi.org/10.1785/gssrl.67.1.16

I attach the paper and perhaps you may have suggestions?

#2 Updated by Paul 6 months ago

As for text etc, the usage says for* -S*:

       Optionally add /fontsize[/offset][u] [Default values are /9/3.000000p]
       fontsize < 0 : no label written; offset is from the limit of the beach ball.

#3 Updated by Januka 6 months ago

Thanks Paul. That was a good read. HybridMT package used for MT inversions has a plotting routine that does not have this problem. Perhaps it might be a way forward if a modification is to be made to psmeca in the future.


Also available in: Atom PDF