But%20We%20Have%20A%20Standard.

1
But We Have A Standard.

• Can SEED Support Synthetics?
• Linus Kamb
• IRIS DMC

2
SEED Format

• Volume Control Headers
• Abbreviation Control Headers
• Station Control Headers
• Time Span Control Headers
• Data Records

SEED Volume
3
Mini-SEED

• Data-only volumes
• Data records
• Fixed Section of Data Header
• Blockette 1000
• Other data record headers

4
What about synthetics?

• Standard Time Series data
• Additional synthetic metadata
• Simulation codes, extents, Earth models, boundary
  conditions, source description
• May correspond to a Station-Channel
• Volumetric (4D) data

5
Some Possibilities

• Shoe-horn into existing SEED blockettes
• Blockette overloading
• Define new blockettes
• Specific to synthetics, or
• General-purpose blockettes
• Use blockette 2000
• XML SEED

6
Blockette Overloading

• Abbreviation Control Headers
• Comment blockettes
• Field overloading
• Eg. Station Start effective time maps to
  simulation run time

7
Blockette Overloading

• Abbreviation Control Headers
• Comment blockettes
• Field overloading
• Eg. Station Start effective time maps to
  simulation run time
• This is a Bad Idea.

8
Add Blockettes

• Specific blockettes for synthetic metadata
• Modeling software, Earth model,
• Generic blockettes
• Name-value pairs
• XML Document blockette
• FDSN Standards process
• Changes required to SEED software

9
Blockette 2000

• Variable Length Opaque Data
• Data Record Header
• Can be included in MiniSEED volume
• Opaque header fields
• SEED readers dump out data as is
• Could be Synthetic Metadata Document
• Use external reader

10
In Conclusion

• SEED can contain synthetic data.
• People are using it.
• Metadata support is not there.
• Community needs to decide how to proceed.

11
The (93) Manual Says
We also did not design SEED for the exchange of
processed (e.g., filtered) data or synthetic data
(i.e., created by computer modeling). While such
use is possible, we will not support it. To be
removed in the current manual, perhaps.