HVSR Progress summary

by Hongyu(Jeff) Xiao

@August 16, 2022

1. IRIS-HVSR

1.1 Project Source

GitHub - iris-edu/HVSR: A set of Python scripts to compute and plot HVSR curves using MUSTANG PDF-PSD web service from IRIS

A set of Python scripts to compute and plot HVSR curves using MUSTANG PDF-PSD web service from IRIS - GitHub - iris-edu/HVSR: A set of Python scripts to compute and plot HVSR curves using MUSTANG PDF-PSD web service from IRIS

https://github.com/iris-edu/HVSR

1.2 Anaconda and Jupyter Notebook Setup

• Anaconda
  • Install

  Anaconda | Anaconda Distribution

  Anaconda's open-source Distribution is the easiest way to perform Python/R data science and machine learning on a single machine.

  https://www.anaconda.com/products/distribution

  

  • Conda Cheat Sheet

    https://s3-us-west-2.amazonaws.com/secure.notion-static.com/189d0299-c515-4062-8d09-2f585f762d7d/conda-cheatsheet.pdf

  • Jupyter Notebook setup

    0 Setting_Up_Environment&Install_Packages

    Here is a very good information page on installing anaconda Without digging into what Anaconda is, we just need to go to Anaconda Installation Document page, and install anaconda following the steps there. Please be aware that you will need to download corresponding file based on your system (MacOs/Windows/Linux) After your installation, you could verify your installation.

    https://hy-x.github.io/Jupter_Notebooks/0%20Setting_Up_Environment%26Install_Packages.html

• Environment Name: HVSR

• Display interface: Jupyter Notebook

• Python Version: 3.8

• Package content:

  CHANGES.txt
     - a text file containing the history of changes to this bundle
  
  INSTALL.txt
     - installation notes
  
  README.md
     - this file
  
  bin/
     - scripts directory containing:
          + computeStationChannelBaseline.py (described above)
          + computeHVSR.py (described above)
  
  param/
     - parameters directory containing:
          + getStationChannelBaseline_param.py - the configuration parameter file for the computeStationChannelBaseline.py script above
          + computeHVSR_param.py - the configuration parameter file for the computeHVSR.py script above
  
  lib/
     - bundle library files:
          + fileLib.py - a collection of functions to work with files and directories
          + msgLib.py - a collection of functions to print messages

• Example:

  computeHVSR.py - uses IRIS DMC’s MUSTANG noise-psd and noise-pdf web services to compute HVSR for a given station based on the MUSTANG hourly PSDs.

  computeStationChannelBaseline.py uses IRIS DMC’s MUSTANG noise-pdf web service to compute channel-specific noise-baseline for a given station

  • IRIS-MUSTANG：

  IRIS DMC Web Services

  These services may be used under IRIS Data Services Terms of Service in accordance with our Usage Guidelines. Usage of the services and data in publications should cite the services according to our citation instructions and data by network. Welcome to the MUSTANG data quality metrics web service home page.

  http://services.iris.edu/mustang/

  

  getStationChannelBaseline.py net=IU sta=ANMO loc=00 chan=BHZ start=2002-11-20 end=2008-11-20 plot=1 plotnnm=1 verbose=1 percentlow=10 percenthigh=90

  computeHVSR.py net=TA sta=TCOL loc=-- chan=BHZ,BHN,BHE start=2013-01-01 end=2013-01-01 plot=1 plotbad=0 plotpsd=0 plotpdf=1 verbose=1 ymax=5 xtype=frequency n=1 removeoutliers=0 method=4

• Report example

  

  • this could be further processed with more analysis
  • report comply with SESAME HVSR check standard

    https://s3-us-west-2.amazonaws.com/secure.notion-static.com/aedcfb1b-cfe7-4535-91a3-cb08b774f865/SESAME-HV-User-Guidelines.doc

• Image example
  • baseline example

    

  • HVSR example

    

1.3 Modefication (refer to the modified py file)

• starttime and end modification
  • the original setting:
    • if the start date == the end date, it will calculate hvsr over 24 hours of that day
  • modefied setting:
    • for the purpose of the ISGS work, the field deployment time is normally 20 minutes (up to 1 hour)
    • modefied the length requirement to be 1 hour for the station
    • this could be furthuer modified to be an hourly estimate
    • default length 30/60/90 days in the summer for each station
    • calculation based on 1 hour interval
  • location: modified py file under bin folder

• report txt file seperation and save
  • saved as txt file with timedate.station.network…txt file
    • original setting:
      • original setting did not include timedate in the report file
      • original setting will only keep one report file and the later runs will replace the existing file
    • modefied setting:
      • include the timedate into the report file
      • file is still in txt format
      • files could be further processed by Bash or other scripting language for statistial analysis
  • report with SESAME HVSR check for HVSR peak

    https://s3-us-west-2.amazonaws.com/secure.notion-static.com/aedcfb1b-cfe7-4535-91a3-cb08b774f865/SESAME-HV-User-Guidelines.doc

  • location: data or report folder

• final result file image save
  • saved as jpeg file with timedate.station.network…HVSR jpeg/png file
  • original setting:
    • with timestamp but overwrite jpegs
  • modefied setting:
    • include timedate into the file name for scripting
    • in jpeg/png format in image folder
  • each image containes 3 subplots (H,V, HVSR with peak marked)
  • location: image folder

Raspberry Shake

Station View:

RS StationView

https://stationview.raspberryshake.org/#/?lat=38.00613&lon=-89.83940&zoom=7.000&net=AM&sta=RC345

Data request example URL

https://data.raspberryshake.org/fdsnws/dataselect/1/query?starttime=2022-08-15T20:42:10&endtime=2022-08-16T20:42:10&network=AM&station=RC345

URL Breakdown:

• red colors are contents defined by users

  https://data.raspberryshake.org/fdsnws/dataselect/1/query?starttime=

  2022-08-15T20:42:10

  &endtime=

  2022-08-16T20:42:10

  &network=

  AM

  &station=

  RC345

Jupyter Notebook Notes:

2.1 Base modefication (refer to Jupyter file)

1. Time, Station,network information setup

2. Automatic URL requesting

3. Automatic MSEED file downloading

4. Automatic MSEED seismic preprocessing (Taper, demean, detrend etc..).

2.2 MSEED to HVSR Bridging Modefication (refer to Jupyter file)

• Bridging the RasShake3D and IRIS-HVSR package
  • Probabilistic Power Spectral Densities calculation
    • this is the input required by IRIS-HVSR package
    • in IRIS-HVSR package, PSD was provided by IRIS MUSTANG service
    • Replace the MUSTANG cloud input with MSEED calculated PSD from the pacakge
  • PPSD Example based on Obspy

  

  

Reference:

obspy.signal.spectral_estimation.PPSD - ObsPy 1.3.0 documentation

Attributes Public Methods Private Methods Helper routine to get a list of 2-tuples with gapless portions of processed PPSD time ranges. This means that PSD time history is split whenever to adjacent PSD timestamps are not separated by exactly . Reuse a previously created figure returned by plot(show=False) and plot the current histogram stack (pre-computed using ) into the figure.

https://docs.obspy.org/packages/autogen/obspy.signal.spectral_estimation.PPSD.html

Visualizing Probabilistic Power Spectral Densities - ObsPy 1.3.0 documentation

The following code example shows how to use the PPSD class defined inobspy.signal . The routine is useful for interpretation of e.g. noise measurements for site quality control checks. For more information on the topic see[McNamara2004] . Read data and select a trace with the desired station/channel combination: Metadata can be provided as an Inventory (e.g.

https://docs.obspy.org/tutorial/code_snippets/probabilistic_power_spectral_density.html

https://s3-us-west-2.amazonaws.com/secure.notion-static.com/acaaf755-8a6f-4023-bb4e-201ba1400a9e/McNamaraBuland_BSSA.pdf

2.3 Follow-work

• Port-in the aquired PSD into IRIS-HVSR setting

• Could be reformating involved

• Processed PSDs are in XML format

Trimino TRC file decode

Moho Company Email contact:

• Claiming codec to be private and refused to provide codec

TRC file

• Binary

• Non-standared codec

• decoded files follow SAF (SESAME ASCII) format

Hints on the TRC codec decoding

• Binary analysis tools

  Binary analysis tools

  Binary analysis is a specialization that requires technical knowledge, patience, and especially the right tools. It is also known as reverse engineering and continues to be in demand by security firms. Reverse engineering is definitely a valuable skill to have or obtain. Typically it is used to investigate malware and during digital forensics.

  https://linuxsecurity.expert/security-tools/binary-analysis-tools

  

• Python Decode exmaples

  codecs.decode() in Python - GeeksforGeeks

  With the help of codecs.decode() method, we can decode the binary string into normal form by using codecs.decode() method. Syntax : codecs.decode(b_string) Return : Return the decoded string. Example #1 : In this example we can see that by using codecs.decode() method, we are able to get the decoded string which can be in binary form by using this method.

  https://www.geeksforgeeks.org/codecs-decode-in-python/

  

• Tried Codecs
  • EBCDIC
  • ISO 8859-1 Western Europe
  • ISO 8859-2 Western and Central Europe
  • ISO 8859-3 Western Europe and South European (Turkish, Maltese plus Esperanto)
  • ISO 8859-7 Greek
  • ISO 8859-9 Western Europe with amended Turkish character set
  • ISO 8859-10 Western Europe with rationalised character set for Nordic languages, including complete Icelandic set
  • ISO 8859-13 Baltic languages plus Polish
  • ISO 8859-14 Celtic languages (Irish Gaelic, Scottish, Welsh)
  • ISO 8859-15 Added the Euro sign and other rationalisations to ISO 8859-1
  • ISO 8859-16 Central, Eastern and Southern European languages (Albanian, Bosnian, Croatian, Hungarian, Polish, Romanian, Serbian and Slovenian, but also French, German, Italian and Irish Gaelic)
  • CP437, CP720, CP737, CP850, CP852, CP855, CP857, CP858, CP860, CP861, CP862, CP863, CP865, CP866, CP869, CP872
  • UTF-8
  • UTF-16