Good day! On USGS web-site I can get co-called PGA/PGV (Peak Ground Acceleration/Velocity) values in the region of earthquake. It is possible to get the map of velocities with respect to time (to see how earthquake developed in time)?
Thank you.
Add a comment
|
1 Answer
$\begingroup$
$\endgroup$
2
No. However, you can look at backprojection results which show where the energy is coming from. It gives you some idea about how the earthquake evolved in time, e.g., see here (click play to see the video).
The only other way to get what you want is to use observations from dense arrays. E.g., see here, but this is only possible in certain regions.
EDIT (answer to your comment below): Codes like SPECFEM3D are good for simulating low frequency ground motion (e.g., 0.1-1Hz). High frequency (1-10Hz) ground motion, that affects smaller structures like 1-2 story houses or even mid-rise buildings is very difficult to simulate (you need a very accurate, very high resolution, velocity model with the right soil/geotechnical layer etc., a highly accurate source model with the 'right' slip distribution, rise time, rupture velocity etc., and finally, a very large supercomputer). You're better off using GMPE's like USGS but that will not tell you how things evolve in time.
Bottomline is that statistics of high frequency ground motion are rather well understood (scales with distance) but not for lower frequencies where effects like rupture directivity/velocity, path/basin effect etc. become even more important.
Finally, Okada's elastostatic solutions are for calculating the final displacement field and doesn't have time component, i.e., you can simulate the final slip but not how it evolves with time which is an elastodynamic problem. Also, the final displacement field alone doesn't tell you anything about PGA/PGV.
-
$\begingroup$ Hi! Thanks for direction to see! It is great feature of backprojections, but as I understood I cannot get information about amplitudes at the sources of energy release, but only positions. Is there any models that can be used to simulate vertical motion near the epicenter? For example, can I model vertical motion near the fault by calculating synthetic seismograms from CMT solution using MINEOS or SPECFEM3D GLOBE software? Or they are good only for far positions modeling? Can I use Okada model somehow for this purpose? Any suggestion will be very valuable! Thanks! $\endgroup$– NikkeyDJan 2, 2018 at 19:45
-
$\begingroup$ Thank you! Feel more comfortable now. Just to clarify my last question from the comment above. SPECFEM3D/MINOES simulates the propagation of body/surface waves, but can they with good accuracy model the uplifting (I am actually interested only in vertical component and periods of waves of minutes (mHz)) of surface right at the rupture zone (meaning in a territory of 0-200km around the rupture)? Or it is good only for relatively far fields from the epicenter? My question is because such models don't take into account rupture structure and consider source as point source. $\endgroup$– NikkeyDJan 3, 2018 at 16:57