GenericMappingTools
meca: Possible bugs or improvements
Sample data file meca.dat:
112 32 25 30 90 0 4 Strike-slip
115 34 15 30 60 90 5 Reverse
118 32 45 30 60 -90 6 Normal
-S+l may not work as expected
The documentation says:
scale is the size for magnitude = 5 (i.e. scalar seismic moment M0 = 4.0E23 dynes-cm). However, if +l is used then radius will be proportional to the seismic moment instead.
gmt begin
gmt basemap -R110/120/30/35 -JM10c -Baf
gmt meca meca.dat -Sa1c+l
gmt end show
As I understand it, an Mw=5 earthquake should still have a radius of 1 cm, and an Mw=4 earthquake should have a much smaller radius because now the radius is proportional to the seismic moment. However, the above script doesn't plot any beachballs.
Sometimes the nodal plane lines are not drawn correctly:
gmt begin beachball5 png
gmt basemap -R110/120/30/35 -JM10c -Baf
gmt meca meca.dat -Sa1c -T0
gmt end show
pen doesn't work if pen thickness is not given
-T0/red doesn't work but -T0/0.2p,red works:
gmt begin beachball5 png
gmt basemap -R110/120/30/35 -JM10c -Baf
gmt meca meca.dat -Sa1c -T0/red
gmt end show
-Lblue doesn't work, but -L0p,blue works:
gmt begin beachball5 png
gmt basemap -R110/120/30/35 -JM10c -Baf
gmt meca meca.dat -Sa1c -Lblue
gmt end show
-A+pred also doesn't work, but -A+p0p,red works.
gmt meca meca.dat -Sa1c+l
I don't think that +l should be allowed in the Aki-Richards convention. It makes no sense
With -Sm it works.
julia> meca(mat2ds([239.384 34.556 33 -.27 -2.13 2.40 -.07 -1.32 -.79 24 240.0 35], ["BlaBla"]), moment_tensor="0.1c+l", show=1, Vd=1)
psmeca -JX12cd/0d -Baf -BWSen -R227/252/32.8/36.4 -Sm0.1c+l -P -K > C:\TEMP\GMTjl_tmp.ps
gmt meca meca.dat -Sa1c+l
I don't think that +l should be allowed in the Aki-Richards convention. It makes no sense
With -Sm it works.
julia> meca(mat2ds([239.384 34.556 33 -.27 -2.13 2.40 -.07 -1.32 -.79 24 240.0 35], ["BlaBla"]), moment_tensor="0.1c+l", show=1, Vd=1) psmeca -JX12cd/0d -Baf -BWSen -R227/252/32.8/36.4 -Sm0.1c+l -P -K > C:\TEMP\GMTjl_tmp.ps
I don't understand why you think it makes no sense for -Sa.
Because the Aki & Richards uses the magnitude directly. What is the sense of making the symbol depending on the seismic moment if that information is not in the parameters file? (and is not even calculable because we don't know the fault dimensions)
Seismic moment can be calculated from magnitude directly. The +l modified was added in response to this feature request (https://forum.generic-mapping-tools.org/t/linear-scaling-of-moment-tensor-symbols-for-psmeca/1978).
Given that M= mu * Area * slip and mag is log(M), how can we compute moment from magnitude alone?
mag is not log(M).
Mw = (2/3)*(log M0 - 16.1).
in which Mw is moment magnitude and M0 is seismic moment.
That's what I (loosely) said magnitude depends on log(M). But yes, we can compute the moment without knowing Area * slip (that was my Tsunami branch thinking).