Re: [seisan] [EXTERNAL] Seisan FAP file format to StationXML?

Fri, 09 Jul 2021 11:17:44 -0700

You're welcome! I interpreted it as an accelerometer because I've seen A-code used for that in our system, and because I think the response plot for the GSE file gave the same output compared to Seisan FAP when I input it as accelerometern but not for seismometer... but I may have easily made a typo that caused that.

Cheers,
Felix

--
Felix Halpaap | Postdoctoral Research Fellow | Dep. of Earth Science | University of Bergen
Allégaten 41, 5007 Bergen, Norway | [email protected] | +47 55 58 34 11  |  Office 2B18c

Sendt: 9. juli 2021 20.12
Emne: Re: [seisan] [EXTERNAL] Seisan FAP file format to StationXML?

Thank you Felix for the comprehensive answer. Very useful.

 

Yes, the MBRY station is a seismometer and the ILI resampled the data at the observatory at a weird rate of 75.19 Hz. I’ll draft a script to automate this after hand testing some examples. The LOC and CHAN codes are odd for this dataset. I have a Python code to fix those.

 

Cheers,

 

Glenn

 

From: [email protected] <[email protected]> On Behalf Of Felix Halpaap
Sent: Friday, July 9, 2021 1:49 PM
To: [email protected]
Subject: Re: [seisan] [EXTERNAL] Seisan FAP file format to StationXML?

 

Hi,

I think there's no fully automatic tool to convert from Seisan FAP to any other format yet, unfortunately. Here's the general workflow that I used for converting from Seisan responses to stationXML is as follows:

1. For Seisan FAP files: use program RESP and manually enter the parameters of the seismometer described in the Seisan file to create a GSE2-file. (For Seisan PAZ files, use program GSERESP to convert from Seisan to GSE).

For your example in MBRY_A_JN.1996-10-01-0000, the commands would be:

  • RESP
  • 4 (GSE2 PAZ)
  • 3 (Accelerometer)
  • *** see below: step not applicable here, only needed if above choice is "2" - seismometer -- see *** below
  • 500 (loaded generator constant, 3rd value in 3rd line)
  • LA10 (seismometer type,  from 2nd line)
  • 1000000 (recording media gain, 5th value in 3rd line)
  • 50 (sampling rate - this is a guess, it's not in Seisan resp-file and needs to be checked from waveform database)
  • 2430ILI (digitizer model, from 2nd line)
  • 1.0 (or ENTER; amplifier gain 4th value from 3rd line if value is not Zero)
  • 1 (number of filters: count number of non-zero value pairs in 3rd+4th line that come after the first 6 values, i.e.,
    ...1.000  -2.000   0.000   0.000
       0.000   0.000   0.000   0.000   0.000   0.000   0.000   0.000   0.000   0.000 means 1 filter
  • 1 -2 (poles and zeros)
  • ENTER (no file given for Poles and Zeros)
  • ENTER (no FIR filters)
  • ENTER (no file for measured values)
  • ENTER
  • Y (GSE response file)
  • MBRY (station code)
  • A  N (component code; check if you'd like to adjust this to modernized FDSN-compliant values)
  • 19961001 (start date, from file name)
  • ENTER

 

*** if instrument is seismometer and not accelerometer, these 2 steps are required as 3rd + 4th answer in list above:

  • 1.0 (seismometer natural period, 1st value in 3rd line of file MBRY_A_JN.1996-10-01-0000)
  • 0.7 (damping ratio, 2nd value in 3rd line)

2. Check the plot that now opens, and compare it to a plot of the response for the Seisan FAP-file (use PRESP to plot, with input file MBRY_A_JN.1996-10-01-0000). They should look exactly the same.

3. Manually enter response end time (e.g. 1997-10-01) into the newly created GSE file (MBRY_A__N.1996-10-01-0000_GSE), so that it looks like:
CAL2 MBRY  A N      LA10     0.70E+00       1.   50.00000 1996/10/01 00:00 1997/10/01 00:00

3. Once you have converted all Seisan files for a station, use program GSERESP2SEED to convert from GSE2 to dataless SEED. It will do some checks and enter some metadata from Seisan's STATION0.HYP file (e.g., latitude, longitude, elevation).

4. Read SEED file straight into Obspy and write to stationXML, or use  IRIS's stationXML-seed-converter as a standalone tool to convert to stationXML.

 

The steps under 1) can probably be scripted in a shell to read the values from FAP file and send as inputs to the program. But if you want to include the correct information on the sampling rate you may have to query the waveform database, and for the correct seismometer and digitizer models you need to check that all comment lines are written in the same format (has free formatting in Seisan FAP files). So depending on the number of files that you want to convert doing it manually gives you some extra QC.

Cheers,
Felix

 

On 09/07/2021 16.25, Glenn Thompson wrote:

Hi Andy,

 

I have many of these FAP files from the Montserrat analog and digital networks from 1996-2004. An example of each attached.

 

Glenn

 

 

From: Lockhart, Andrew B <[email protected]>
Date: Friday, July 9, 2021 at 10:14 AM
To: Glenn Thompson <[email protected]>, [email protected] <[email protected]>
Subject: Re: [EXTERNAL] [seisan] Seisan FAP file format to StationXML?

Hello Glenn -

 

I'm struggling with getting analog responses into StationXML right now (like, today). Can you send me a file or two to look at? When do these date from? 

 

-Andy

 

From: [email protected] <[email protected]> on behalf of Glenn Thompson <[email protected]>
Sent: Friday, July 9, 2021 6:48 AM
To: [email protected] <[email protected]>
Subject: [EXTERNAL] [seisan] Seisan FAP file format to StationXML?

 

 

 This email has been received from outside of DOI - Use caution before clicking on links, opening attachments, or responding.  

 

Hi Seisan community,

 

I have many Seisan CAL files in what I believe is Seisan FAP format, which is described in Appendix C of the Seisan manual.

 

What is the best way to convert these Seisan FAP files into StationXML? I want to construct an ObsPy workflow to correct Seisan WAV files from counts to m/s using the full instrument response from these Seisan FAP files.

 

For comparison, I do also have many RESP files in my Seisan CAL directory for later years. I can process these to StationXML like this:

 

from obspy.io.xseed.core import _read_resp

this_inv = _read_resp(respfile)

this_inv.write(xmlfile , format="STATIONXML")

 

but I have not find a way to read the Seisan FAP files with ObsPy. I would guess people have solved this problem before though.

 

Cheers,

Glenn

 

 

[EXTERNAL EMAIL] DO NOT CLICK links or attachments unless you recognize the sender and know the content is safe.



_______________________________________________
seisan mailing list
[email protected]
https://mailman.uib.no/listinfo/seisan
-- 
Felix Halpaap| Postdoctoral Research Fellow | Dep. of Earth Science | University of Bergen
Allégaten 41, 5007 Bergen, Norway |  Office 2B18  | [email protected]  | +47 55 58 34 11

[EXTERNAL EMAIL] DO NOT CLICK links or attachments unless you recognize the sender and know the content is safe.

_______________________________________________
seisan mailing list
[email protected]
https://mailman.uib.no/listinfo/seisan

Reply via email to