Installation

1. Create a conda environment with: pandas and obspy

2. Clone the repository:

   $ git clone https://github.com/lvanderlaat/lakiy_utils.git

3. Install the package in the conda environment:

   (env) $ cd lakiy_utils (env) $ pip install -e .

Exporting waveforms for detection training (4 s samples)

From Google Docs database to txt

Copy the daily records of the Google Docs file in a txt file with format name %Y-%m-%d.txt. Clean, format:

Label_1- %H:%M, %H:%M,
Label_2- %H:%M, %H:%M,
...

Txt to Swarm csv

$ txt-to-swarm %Y-%m-%d.txt

If recent data (in FDSN memory)

Download daily MSEED file

$ download-waves %Y-%m-%d_Swarm.csv

Repick in Swarm

Move the pick to the onset and add an end label.

Check the file

$ check-swarm %Y-%m-%d_Swarm.csv

Download and write segments

You must have a csv file with stations information (network,station,channel columns)

$ download-segments %Y-%m-%d_segments.csv stations.csv outpath

If not in FDSN memory

Download daily MSEED files

$ download-retro year julday stations.csv outpath

Repick in Swarm

Move the pick to the onset and add an end label.

Check the file

$ check-swarm %Y-%m-%d_Swarm.csv

Trim segments

trim-segments %Y-%m-%d_segments.csv path/to/mseed path/to/out

Statistics of the database

$ db-stats path/to/wfs

Exporting waveforms for classification (complete event)

Volcano-tectonic VT events

Get catalog from Antelope

db2TOMODD extracts a PHA (hypoDD format) catalog from the Antelope database:

db2TOMODD database output [julianstartdate julianenddate] [latmin latmax lonmin lonmax][Mmin][Nass] [EIDadd]

Example:

db2TOMODD ../locate/vulca lakiy-VT 2014-001 2021-001  9.955  10.066 -83.809 -83.682 2 0

Extract waves and catalog

Use the PHA catalog to extract waveforms (.msd) and event information (xml).

(env) $ VT-extract PHA_FILE STATIONS.CSV OUTPUT

Add TC.CVTR waveforms

For events occured before CVTR inclusion we add the correspondant waveforms:

(env) $ waves-add-CVTR -i path/to/wfs_in -o path/to/wfs_out -w path/to/raw

Pick coda

Copy the folder containing the extracted data to your computer.

Once the coda is picked, we copy the files to a new folder to discriminate picked events.

(env) $ VT-pick -i path/to/wfs_in -o path/to/wfs_out -e eventid

Convert to Lakiy format