Investigate Seismic Wave Speed From an Alaska Earthquake

[MaleenKidiwela]

Title
Using ocean bottom data to investigate seismic wave speed from an Earthquake

Premise
Choose one earthquake from the Alaska Earthquake Catalog to investigate how P- and S-waves from that earthquake travel through the earth as recorded by an ocean bottom seismometer network. These observations could provide us with insights in to what controls wave-speeds on earth. Scientifically try to draw conclusions from what the data presents (Eg:- Are these velocities azimuth-dependent? if so why would it have different P and S wave velocities in different directions)
Use the dataset of data from one earthquake and P and S wave arrival time of each station to produce a GIF/video of P- and S-wave arrivals across the seismometer network.

Result would be a much complex version of the following GIF, and you could use all the resources of bathymetry plotting, point plotting we taught you in the tutorial.

Code
Plot distance vs. arrival time plots for both P- and S-waves, fit with curves to estimate velocities
Create GIF/video of P- and S-wave arrivals across the seismometer network

Personnel
Maleen Kidiwela, Zoe Krauss

[zoekrauss]

The earthquake catalog for the extensive Alaskan ocean bottom seismometer network, a community experiment, is available here: https://scholarworks.alaska.edu/handle/11122/11967

[abkfenris]

Is it available through IRIS? There are some nice libraries (that I'm currently forgetting the names of) that can retrieve and process data through a whole bunch of observatories.

[MaleenKidiwela]

You could retrieve and process cataloged data from Obspy. we could help you retrieve data from the alaska catalog!

[zoekrauss]

The offshore catalog I linked has more detail (labelled P- and S-wave arrivals) than is available through IRIS, but I do know that the raw waveform data that was used to make the earthquake catalog is available through IRIS! The obspy library is incredible for retrieving waveform, earthquake catalog, and seismic station data through a remote client.

[bartonnoah]

Perhaps it would be possible to compare actual time of arrival at various OBS to expected (mean travel speed) time of arrival? From my limited understanding of the topic, this may allow us to make statements about the density of the Earth along the wave's path and potentially indicate the presence of features such as magma chambers. We could then compare this against known features in the region to see if we correctly predicted the features.

[gmanuch]

That's an awesome project -- if you're undecided, please have a look at it!

[bartonnoah]

https://github.com/bartonnoah/ohw22_NW_git/blob/08fcb883c11e7a988a4409f5abd3d7a26bcaf0f6/just%20playin%20around.ipynb

Will make this much cleaner tomorrow morning but as a status update I played around with Maleen's code and now understand how to handle .quakeml file formats. Looking forward to discussing tomorrow!

[MaleenKidiwela]

Great start! Now what you now need is the

1. location of all these individual stations -> try plotting stations
2. Look up how the speed varies with azimuth
3. Then work on how to make interactive plots of travel time
4. Work on how to make GIFs of travel time passing (some sort of a moving contour maybe... be creative)
5. Perhaps interpolate to get travel times for any point
6. Try exploring whether velocity varies with different earthquakes events
7. try to plot how the seismic waveform looks like for stations with P and S wave times
8. plot different events on the catalog

I guess part of this is to do your own exploration, of dataset, see whether there are interesting things

For science portion,

1. Think what causes velocity variations on earth, try to come up with a interpretation, there are many things other than magma that slows down waves, and also sometimes they speed up
2. Think about implications of being able to determine the wave-speed of an earthquake
3. Think about how fast these are compared to a tsunami wave propergating through water from the origin (more reading)

[zoekrauss]

Regarding gathering the metadata of station location, here is a notebook example I have:
https://github.com/zoekrauss/marine-seismo-intro/blob/main/explore_data.ipynb

Some of it is redundant from the MGG tutorials already available but there will be some new helpful stuff in there as well!