Naming Standards
Pyatoa enforces a few standards for book keeping which are listed here.
Receiver codes
Receivers codes define station names and components for a given station. Wildcards are typically acceptable when specifying station codes in Pyatoa.
CODE: Standard code are expected in format NN.SSS.LL.CCC (where N=Network, S=Station, L=Location, C=Channel); e.g. NZ.BFZ.10.HHZ
NET (network): Two character network code NN (e.g. NZ)
STA (station): Multi character station code SSS (e.g.BFZ)
LOC (location): Two digit location code (e.g., 10).
CHA (channel): Three character channel code BIO (where B=Band code, I=Instrument code, O= Orientation code); e.g.HHZ
COMP (component): One character channel component, a.k.a the orientation code in CHANNEL
Note
When saving new data, band code is enforced via the data sampling rate, following the format of the SEED convention.
Event ID
Unique event ID’s are critical for book keeping. The following standard is enforced throughout Pyatoa:
Event IDs are generated from:
Obspy.Event.resource_id.id
Event IDs must be strings
All Event IDs must be unique for a given inversion
ASDFDataSets must correspond to a single event, saved as:
{event_id}.h5
Iteration
Iteration number is used to keep track of the number of full iterations in an inversion.
Iteration must be an integer [1, inf).
iteration tag
: string formatted iteration. E.g., i01 for iteration 1.
Step Count
During a line search, multiple forward simulations may take place. Step Count is used to track each forward simulation in a line search.
Step count must be an integer [0, inf).
step tag
: string formatted step count. E.g., s00 for step count 0.
Evaluation
An evaluation is a single step within a single inversion.
Evaluation is a string comprised of
iteration_tag
andstep_tag
The first evaluation of an inversion is: i01s00
Whenever iterations are incremented, step counts reset to 0
Observation Waveforms
Related to the Data Discovery page, Observation waveforms stored locally (on disk) should be saved in a specific format:
Default Directory Template:
path/to/observed/YYYY/NN/SSS/CCC/
Default File ID Template:
NN.SSS.LL.CCC.YYYY.DDD
Where:
YYYY: Year with century (e.g., 1987)
NN: Network code (e.g.NZ)
SSS: Station code (e.g.BFZ)
LL: Location code (e.g.10)
CCC: Channel code (e.g.HHZ.D)
DDD: Julian day of the year (January 1 is 001)
Example directory for station NZ.BFZ, for the day 2018-02-18: path/to/observed/2018/NZ/BFZ/HHZ/NZ.BFZ.10.HHZ.D.2018.049
Station Response
Related to the Data Discovery page, station metadata or response information stored locally (on disk) should be saved in a specific format:
Default Directory Template: path/to/responses/SSS.NN
Default File ID Template: RESP.NN.SSS.LL.CCC
Where:
NN: Network code (e.g. NZ)
SSS: Station code (e.g. BFZ)
LL: Location code (e.g. 10)
CCC: Channel code (e.g. HHZ.D)
Example directory for station NZ.BFZ: path/to/response/BFZ.NZ/RESP.NZ.BFZ.10.HHZ
Synthetic Waveforms
Synthetic waveforms therefore must follow the naming convention dictated by SPECFEM.
NN.SSS.BIO.sem**U**
Where:
N = Network
S = Station
B = Band code
I = Instrument code (Must always be **X for synthetics**)
O = Orientation code
U = Unit code
Unit code U is dictated by the chosen output units in SPECFEM, e.g.:
d = displacement
v = velocity
a = acceleration
Example for displacement synthetic waveforms for the vertical component of New Zealand station BFZ: NZ.BFZ.BXZ.sem*