DispersionPhaseShift(
*args ,
phase_velocities: Any ,
approx_resolution: Any = None,
approx_freq: Any = None,
)-> None
Compute dispersion images using the phase-shift method.
The patch must have time and distance dimensions (see notes).
| Parameter | Description |
|---|---|
| phase_velocities |
NumPy array of positive velocities, monotonically increasing, for which the dispersion will be computed. |
| approx_resolution |
Approximated frequency (Hz) resolution for the output. If left empty, the frequency resolution is dictated by the number of samples. |
| approx_freq |
Minimum and maximum frequency to compute dispersion for, in Hz If left empty, minimum is 0 Hz, and maximum is Nyquist |
See also Park, Miller, and Xia (1998)
Inspired by https://geophydog.cool/post/masw_phase_shift/.
Dims/Units of the output are forced to be ‘frequency’ (‘Hz’) and ‘velocity’ (‘m/s’).
Each channel’s distance is read as its position along the wave’s path, so data are effectively mapped along a 2-D line; the coordinate need not be sorted.
The image depends only on distances relative to each other, so any origin works for a one-sided gather, but the wave must travel toward increasing distance. For a gather whose wave travels toward lower distance, negate the coordinate; for a two-sided gather, use the offset from the source (abs(distance - source_distance)). The new values carry no units, so convert to metres first: p = patch.convert_units(distance="m") then p.update_coords(distance=-p.get_array("distance")). Reversing the patch with flip changes nothing, since it reverses the data and the coordinate together.
import dascore as dc
import numpy as np
# Example 1 - Right-sided wavefield
patch = (
dc.get_example_patch('dispersion_event')
)
disp_patch = patch.dispersion_phase_shift(np.arange(100,1500,1),
approx_resolution=0.1,approx_freq=[5,70])
ax = disp_patch.viz.waterfall(show=False, cbar=False)
ax.set_xlim(5, 70)
ax.set_ylim(1500, 100)
disp_patch.viz.waterfall(show=True, ax=ax)
```
| Name | Description |
|---|---|
| check | Refuse a patch which does not carry what the operation needs. |
| get_metadata | Return the image’s metadata, and the frequencies the kernel keeps. |
| model_copy | Copy the model, dropping cached values the update invalidates. |
| new | Create new instance with some attributed updated. |
| numpy_kernel | Return the normalized phase-shift stack of each velocity and frequency. |
| dispersion_phase_shift | Compute dispersion images using the phase-shift method. |
| reconcile | Return metadata or a patch holding the final data; default as is. |
| run | Run the operation: check, get_metadata, kernel, reconcile, record. |