Join: Merge sections (horizontally) from the Mask or Cluster step
Join: Input files
Join input file: Mask/Cluster/Join report
You can give any number of Mask or Cluster reports as inputs for the Join step. You can also give Join report files, to join sections that were themselves made by joining other sections. See List Input Files for ways to list multiple files.
Note
SEISMIC-RNA will not double-count any of your original sections, even if they appear in multiple report files you are joining. It will just log a warning if it finds any sections given multiple times.
Mask, Cluster, and Join reports will be joined only if they share all of
the top-level output directory, i.e.
--out-dir(-o)the sample
the reference
whether they are clustered (i.e. sections from Mask will not be joined with sections from Cluster)
For each combination of these attributes, SEISMIC-RNA will produce a joined section from all Mask/Cluster/Join reports with those attributes. The original Mask/Cluster/Join report files will not be deleted or modified; you will merely get a new Join report file for each joined section.
For example, if you ran the command
seismic join -J {joined} {out}/{sample}/cluster/ref-1/sect-1 {out}/{sample}/cluster/ref-1/sect-2 {out}/{sample}/mask/ref-1/sect-1 {out}/{sample}/mask/ref-1/sect-2 {out}/{sample}/mask/ref-2/sect-1 {out}/{sample}/mask/ref-2/sect-2
where {out} is the path of your output directory and {joined} is the
name you want to give to each joined section, then you would get three new Join
reports representing the joined sections:
{out}/{sample}/cluster/ref-1/{joined}/cluster-report.json: made from{out}/{sample}/cluster/ref-1/sect-1/cluster-report.jsonand{out}/{sample}/cluster/ref-1/sect-2/cluster-report.json{out}/{sample}/mask/ref-1/{joined}/mask-report.json: made from{out}/{sample}/mask/ref-1/sect-1/mask-report.jsonand{out}/{sample}/mask/ref-1/sect-2/mask-report.json{out}/{sample}/mask/ref-2/{joined}/mask-report.json: made from{out}/{sample}/mask/ref-2/sect-1/mask-report.jsonand{out}/{sample}/mask/ref-2/sect-2/mask-report.json
To join all valid combinations of Mask/Cluster/Join reports in {out} into
sections named {joined}, you can use the command:
seismic join -J {joined} {out}
Join: Settings
Join setting: Name of joined section
You can choose the name of your joined section(s) using --joined (-J).
If you omit this option in seismic join, then it will default to joined.
If you omit this option in seismic wf, then the Join step will not run.
Join setting: Clusters to join
When joining cluster reports specifically, you may specify which cluster of each section should be joined. For example, if you are joining three sections, each of which you have clustered up to order 2, then you may want cluster 1 of the joined section to comprise
cluster 2 of section 1
cluster 1 of section 2
cluster 2 of section 3
and conversely, cluster 2 of the joined section to comprise
cluster 1 of section 1
cluster 2 of section 2
cluster 1 of section 3
You can specify the clusters to join using --join-clusts (-j) followed
by a file of the clusters to join.
See Metadata for Joined Clusters for more information on this file.
Join: Output files
All output files go into the directories {out}/{sample}/mask/{ref}/{joined}
and/or {out}/{sample}/cluster/{ref}/{joined}, where {out} is the output
directory, {sample} is the sample name, {ref} is the reference name,
and {joined} is the joined section name.
Join output file: Join report
SEISMIC-RNA writes a Join report file called either mask-report.json or
cluster-report.json, depending on whether the joined sections came from Mask
or Cluster reports.
See Join Report for more information.
The file is named mask-report.json or cluster-report.json not because
its contents are identical to those of a Mask/Cluster report file (they aren’t)
but because SEISMIC-RNA can more easily use them interchangably when they have
the same file names.
You can give Join reports to the Table step just as you would Mask or Cluster
reports.
Join: Troubleshoot and optimize
Joined section … got duplicate sections …
This warning means that an original (unjoined) section appeared more than once in the report files you are joining.
For example, suppose that you join section-A and section-B:
seismic join -J section-AB out/sample/mask/ref/section-A out/sample/mask/ref/section-B
Then you try to join section-A with the joined section section-AB:
seismic join -J section-AAB out/sample/mask/ref/section-A out/sample/mask/ref/section-AB
This second command will warn that section-A is duplicated because it
appears in both the report files for sample-A and pool-1.
If you get this warning, then you should check your Pool report file to ensure it contains all the samples you want and none that you don’t.
Overwriting … would cause data loss
This error means that you attempted to create a joined section with the same
name as an existing non-joined section while using --force, e.g.
seismic join --force -J section-A out
if section-A already exists.
Doing so would overwrite the Mask/Cluster report for the original, non-joined
section, making it unusable.
To prevent data loss, the Join step refuses to overwrite Mask/Cluster reports,
even with --force.