bug report - plot vs decompose

[sjaye013]

Plotting the DC component of the moment tensor via plot function outputs a different beach ball, compared to running the decompose function and plotting the DC matrix. (The former results to an all-red explosion beach ball, while the latter plots the correct DC beach ball.)

Figures attached. Script below:

% mopad plot 713668000000000,12428000000000,-726096000000000,-654200000000000,332200000000000,8175000000000 -P dc -f sampledc.png

% mopad decompose 713668000000000,12428000000000,-726096000000000,-654200000000000,332200000000000,8175000000000 -p dc

Double Couple part in NED-coordinates:

/ 0.78 -0.59 0.49
| -0.59 0.22 0.18 | x 435092335422952.562500
\ 0.49 0.18 -1.00 /

% mopad plot 0.78,0.22,-1,-0.59,0.49,0.18 -f sampledc2.png

[megies]

Not sure if you'll get answers here, mopad is essentially not maintained by the original authors, as far as I know.

If you come across the same problems when using obspy (which has mopad plugged in for beachball plotting) you could let us know at obspy/obspy.

[emolch]

A new and much shorter code (hopefully also with fewer errors :-) to plot beachballs is available in pyrocko.beachball (examples). It can plot full, deviatoric and dc beachballs.

from matplotlib import pyplot as plt
from pyrocko import beachball, moment_tensor as pmt, plot

fig = plt.figure(figsize=(4., 2.))
fig.subplots_adjust(left=0., right=1., bottom=0., top=1.)
axes = fig.add_subplot(1, 1, 1)
axes.set_xlim(0., 4.)
axes.set_ylim(0., 2.)
axes.set_axis_off()

mt = pmt.as_mt((713668000000000, 12428000000000, -726096000000000,
                -654200000000000, 332200000000000, 8175000000000))

for i, beachball_type in enumerate(['full', 'deviatoric', 'dc']):
    beachball.plot_beachball_mpl(
        mt,
        axes,
        beachball_type=beachball_type,
        size=60.,
        position=(i+1, 1),
        color_t=plot.mpl_color('scarletred2'),
        linewidth=1.0)

fig.savefig('beachball-example03.png')
plt.show()

print mt.standard_decomposition()

See also: MomentTensor.standard_decomposition()