Thread: FAP vs EVALRESP in SAC-101.4

George,

First off, I got your message twice, the one to which I am replying came
about 40 minutes later. I did not get a copy from [sac-dev] to my first
message to Sheila, but I looked in the archives and it was there. I think
I have gotten copies of subsequent messages sent to the listserv.

Next: the SAC distributed through IRIS is not SAC2000, and I am not sure
when you refer to SAC2000 which program is being referenced. The FAP got
dropped in the transition to IRIS/SAC I think because the way it was
written it was part of a bigger package that Someone decided to drop.

As of the current version, EVALRESP produces FAP files. Earlier versions
had FA and FP files, and this one combined them. We used the EVALRESP
convention rather than the old one to which you refer for compatibility.
Besides, it is a simpler format.

EVALRESP does not apply FAP files to do instrument corrections, it just
produces the file. As discussed in the current TRANSFER help file, an
advantage of either using a FAP file produced by EVALRESP or by using
EVALRESP within the TRANSFER command is that the response includes all the
stages documented in the RESP file, whereas a POLEZERO subcommand includes
only the first stage. However, running EVALRESP separately allows one to
select the stages to include and provides flexibility about the frequency
band, so a FAP file produced by EVALRESP can refine the instrument
correction more than simply running TRANSFER in SAC with the EVALRESP
subcommand.

Sheila's research uses FAP files, and as of version 101.4 she can now run
these within SAC. Maybe not totally to her liking, but before this
version it was not an option in SAC. As you probably know, 30 years ago I
used FAP files to deal with instrument corrections for broadband
seismomenters recorded on magnetic tape. Also, Martin Chapman and I
developed a procedure to do instrument corrections of network data
transmitted over phone lines and we used FAP files.

Let me know if I have not dealt adequately with your questions/comments.

I am sending this at 16:53 EDT.

Arthur

On Wed, 29 Jun 2011, George Helffrich wrote:

Dear All -

(I resending this to the sac-dev list after de- and re-registering).
This is a separate topic to the FAP behavior that Sheila Peacock was
requesting, hence a different message thread. It was prompted by the
exchange on that topic, so while appearing similar is a different
question.

Here's the question: If EVALRESP produces and handles FAP responses,
what is the added functionality of TRANSFER SUBTYPE FAP? It seems that
there are two mechanisms to achieve the same response correction, one
especially built into SAC and the other built into EVALRESP (and
implicitly used by SAC).

I could rationalize this if the FAP file format was different. I
implemented FAP in MacSAC based on the documented specification of a
pIDC FAP file. However SAC2000 seems to have abandoned this format (or
at least stopped documenting it), so has lost this particular
justification.

Hence the question about added functionality. Any enlightenment?

George Helffrich
george.helffrich<at>bris.ac.uk