research_code

My scripts, files, etc. from my research, and for my GEOS694 final project. Current subfolder is das.

What, Why, and How:

das contains all of my DAS (Distributed Acoustic Sensing) research. This is, in other words, seismology conducted using fiber-optic cables as arrays of seismometers. As this folder is currently in use it is a bit messy, but here are the main features:

• das_coords_bathymetry
  This is where the .xcyz bathymetry files for the cable coordinates are stored.

• das_freq_plots
  This contains early frequency-filtering trials of the DAS data. The code to generate these figures exists, in some form, in minor_das_scripts/Uberfigure.ipynb.

• minor_das_scripts
  This folder is where various scripts not currently in use are stored. There is often a figure or two in this folder as well.

  • Within minor_das_scripts are the majority of scripts under this project.

  • DAS_Bathymetry.ipynb generates local bathymetry to das_seafloor_bathymetry.png, and reads the files in das_coords_bathymetry/ to render a 3D plot of the seafloor with both the KKFL-S and TERRA cable routes mapped onto it.

  • ETI2.0 Slide Maker.ipynb is a quickly-made script to create introductory figures for a conference presentation. It loads a single event file to create figures of a few waveforms over a small distance.

  • DAS_Raw_Noise_Files.ipynb plots absolute median strain across channels in both cables for a set of low-magnitude events, loading the final 45 seconds of each 2-minute long event and plotting their noise profiles.

  • event_metadata_cache.json is a JSON cache of seismic events spanning June–December 2023, with the USGS event ID serving as the key, and storing latitude, longitude, depth, and origin time. This JSON is used in various scripts throughout the project to avoid repeated API calls.

  • das_seafloor_bathymetry.png is the output of DAS_Bathymetry.ipynb, and shows the two cables (KKFL-S in Red, TERRA in Blue) overlaid on a 3D seafloor bathymetry map of the lower Cook Inlet region of Alaska. Properly, this file would be in das_figures/ but I haven't rerouted that script's output.

  • Mapview_TERRA_events.ipynb plots the TERRA cable and all cached seismic events onto a regional Albers Equal Area map using Cartopy, made for a conference presentation.

  • DAS_TauP_Spaghetti_Maps.ipynb contains two related scripts, the first uses TauP to compute P and S-wave travel times between the first and last cable channels for each event with the ak135 velocity model and outputs a summary table. The second generates plots of predicted arrival times across all cable channels for every event simultaneously, and these outputs look a bit like spaghetti thrown at a wall, hence the filename.

  • Noise_Analysis.ipynb is the primary noise characterization script, and has become a 'working file' similar to MS-DAS.ipynb. As it is in active development, this description may not be fully up to date. For each event, this script extracts median absolute strain amplitude over the first 5 seconds and last 20 seconds for comparison purposes, then displays noise amplitude readings for each event, sorted by event magnitude, which are fetched live from the USGS API. Contains multiple plotting variants.

  • DAS_URL_table.ipynb Queries the USGS FDSN event service for M3.2+ earthquakes within 150 km of the cable center during 2023 and prints a formatted table of origin time, magnitude, coordinates, and depth. Used for manually identifying events to download from the DAS data portal, which does not support automated crawling.

  • DAS_USGS_File_Renaming.ipynb renames raw DAS files from the website to USGS event ID format (e.g. ak0237eejw69_TERRA.h5) to match event_metadata_cache.json. This one might not be terribly useful as it was only for renaming from my initial scheme.

  • Uberfigure.ipynb produces a series of wiggle plots of the complete DAS cable waveforms, with zoomed-in subplots. This gave the primary outputs during early data exploration days. Resulting figures are saved to das_freq_plots/.

• MS-DAS.ipynb
  Named to convey the proper pronunciation of DAS by likening it to the operating system MS-DOS, this is the main active research environment, containing multiple scripts within its Jupyter Notebook code blocks. Over time, inactive or mostly finalized scripts are broken out of MS-DAS.ipynb and into separate notebook files in das/minor_das_scripts/, then reworked into functioning in their new location. The current MS-DAS.ipynb file contains a few minor scripts along with the main functionality script focusing on plotting calculated arrival-time windows for multiple events over observed arrival-time data.