search

SPECFEM / specfem3d

SPECFEM3D_Cartesian simulates acoustic (fluid), elastic (solid), coupled acoustic/elastic, poroelastic or seismic wave propagation in any type of conforming mesh of hexahedra (structured or not).
GNU General Public License v3.0
409 stars 228 forks source link

PGV vs PVS #1500

Closed eikmeier-cn closed 2 years ago

eikmeier-cn commented 2 years ago

Hello everybody,

I would like to confirm that my understanding of what is stated in the manual is correct. But before that, let's see if our understanding of the terminology is the same. I understand PGV (peak ground velocity) as, in a time interval, the highest velocity value among the three components of a sensor, and PVS (peak vector sum) as, in that same time interval, the value of the resulting velocity vector sum of the three components.

On page 75 of the manual, version 3.0, we have, for the Par_file: MOVIE_TYPE: if set to 1, the horizontal peak-ground values of displacement/velocity/acceleration are output. If set to 2, the maximum length of the particle displacement/velocity/acceleration vector is output. Please be aware that these peak-ground values can differ from each other.

So, the following doubts arise, considering velocity: 1) Is option 1 the maximum velocity value among the 2 horizontal components (PGV-horizontal), the 3 components (PGV) or is it the PVS? 2) Would option 2 be the maximum velocity value among the 3 components (PGV) or is it the PVS?

Thanks,

danielpeter commented 2 years ago

for option 1, it outputs the peak value in the Y-component (which is N-component if UTM setting is on) of the velocity field. for option 2, it outputs the norm of the particle velocity vector.

any suggestion on modifying this output to make it more convenient for you is welcome.

eikmeier-cn commented 2 years ago

Thanks for the answer @danielpeter. So, I understood option 1 as a N-component PGV (or PPV - Peak Particle Velocity) and option 2 as a PVS.