SEED how everything comes together Part 1

1
SEED - how everything comes together (Part 1)
IRIS/PASSCAL DATA/SOFTWARE GROUP

• Information compiled and re-edited from DMC web
  site http//www.iris.edu/data/

2
PURPOSE

• Review basic concepts related to data delivery
• Why mseed?
• What is the difference between seed/miniseed and
  dataless.
• Understand the relation between these concepts
• and the data flow for data delivery to the DMC
• Why everything should agreed?

3
Any questions?

• Why Seed
• Why fix the headers?
• What if the headers do not match the batch_file?
• What if the batch_file does not match the
  dataless and the data submitted?
• What are the implications of
• mismatching,
• wrong/discrepancy on information,
• lack of information
• in data headers, batch - database and/or dataless
• How/when I will notice these problems?

4
What is the SEED format and why do we use it?

• The Standard for the Exchange of Earthquake Data
  (SEED). It consists of Volume Control Headers,
  Abbreviation Control Headers, Station Control
  Headers, Time Span Control Headers and finally
  Data Records.
• It was designed for use by the earthquake
  research community, primarily for the exchange
  between institutions of unprocessed earth motion
  data.
• From http//www.iris.edu/manuals/SEED_chpt1.ht
  m

5
So what is miniseed?

• The term Data Only SEED Volumes (Mini-SEED) has
  come to be used to identify SEED data records
  without any of the associated control header
  information.

And a dataless SEED volume?

• A dataless SEED volume contains the metadata for
  a request only, including instrument responses,
  instrument coordinates, compression type, etc..
• A dataless, by definition, contains no "data", in
  the sense that no waveform data are included,
  only headers.
• http//www.iris.edu/data/datalessDef.htm

6
Summary

• SEED - Headers and Data records
• mseed - data records without any of the
  associated control header information.
• Dataless - only headers - metadata for request
  only

Time span control headers
Volume Control header
SEED
Dataless
mseed
Abbreviation CH metadata
Headers data
7
How do we obtain an agreement between our data,
header and dataless? and what for?
8
Why Fix the headers?

• Unify information for the stations in the network
  according to SEED format.
• Corrections (band code, network code, station
  name)
• Add information (Network code)
• Edit errors.
• Used SEED format to describe more precisely the
  data in the experiment.

9
Format Definitions for mseed headers

• MiniSEED (or MSEED/mseed) format is strictly
  defined by the Reference Manual, SEED format
  version 2.4. (http//www.iris.edu/manuals/SEEDManu
  al_V2.4.pdf)
• Network code ltnetgt Two (2) alphabetical
  characters assigned by DMC at start of
  deployment EXAMPLE XN,YD, PI
• Station name ltstagt Up to five (5) alphanumeric
  characters no special characters allowed (STAT1,
  cali1, luca5, etc)
• Channel name ltchngt MSEED channel names are
  defined by the sample rate, instrument (i.e.
  sensor) type, and channel orientation. Please
  refer to the seed manual V2.4 - Appendix A for
  more detail.

10

• Band Code
• The first letter specifies the general sampling
  rate and the response band of the instrument.
  (The A code is reserved for administrative
  functions such as miscellaneous state of health.)
  

11

• The second character specifies the family to
  which the sensor
• belongs. The third letter specifies the physical
  configuration of the
• members of a multiple axis instrument package or
  other parameters
• as specified for each instrument.
• Instrument Code
• H High Gain Seismometer
• L Low Gain Seismometer
• G Gravimeter
• M Mass Position Seismometer
• N Accelerometer
• Third character Orientation

12
Building the Database

• dbbuild_batch (antelope)
• The batch mode of dbbuild reads input from batch
  file and adds to the database db accordingly.
  Primarily, the input specifies settings for the
  various entries on the interactive window.

13
The famous BATCH_FILE
sta NM27 34.3857 -106.524 1.870 B1 time
01/19/2005 000000 net XK ristra1.5 datalogger
dm24 S0270 sensor cmg3t 0 3C7100 axis Z 0 0 - 1
1 axis N 0 90 - 2 1 axis E 90 90 - 3 1 samplerate
25sps channel Z BHZ 30 channel N BHN 31 channel E
BHE 32 samplerate 1sps channel Z LHZ 00 channel N
LHN 01 channel E LHE 02 add sta UT53 37.346
-110.3308 1.291 B3 time 06/09/2004
000000 datalogger dm24 S0269 sensor cmg3t 0
3C2200 axis Z 0 0 - 1 1 axis N 0 90 - 2 1 axis E
90 90 - 3 1 samplerate 25sps channel Z BHZ
30 channel N BHN 31 channel E BHE 32 samplerate
1sps channel Z LHZ 00 channel N LHN 01 channel E
LHE 02 Add Close NM27 12/31/2005 235959 Close
UT53 12/31/2005 235959

• IT IS a record of the history of your network,
  changes in data and configurations done during
  the experiment.
• The information in the batch file is used
  indirectly to generate the dataless (when use
  mk_dataless_seed)
• The close time for each station corresponds to
  the date your DMC network code expires.

14
Adding waveforms miniseed2days/miniseed2db

• Entries for batch file and entries from waveforms
  will be included in the database.
• Disagreement between database (seed and database)
  and dataless (headers) will be reflected in the
  availability of data at the DMC.

15
So now you can ask these questions

• What if the headers do not match the batch_file?
• What if the batch_file does not match the
  dataless and the data submitted?
• What are the implications of
• mismatching,
• wrong/discrepancy on information,
• lack of information
• How/when I will notice these problems?

16
CONCLUSIONS

• Field notes are very important! You knew that
  right?
• Well, just a little reminder
• The batch file should agree with
• The data headers
• Your database - description on batch and
  waveforms
• Dataless
• The batch file can be considered as the HISTORY
  of your experiment. Every change in
  configuration/data should be registered on the
  batch file
• The dataless is created based on the information
  in the batch file and your data

17
To keep in mind

• Content agreement between the information in the
  batch file, the headers , your database and the
  dataless will be reflected by the availability of
  data at the DMC

100 Data available at the DMC
Batch_file/ DB
Field Notes
WAVEFORMS
GOOD DATALESS
INTEGRITY COMPLETNES