The response information gives the gain of a channel in counts/m and to get the correct ground displacement, the count values must be divided by the response values. In the current SEISAN system, only the programs MULPLT, WAVETOOL and SPEC use the response information when doing spectral analysis, generating Wood Anderson or ground motion traces. The programs will look first in the CAL (or alternative) directory for a valid response file and if not found there use the header information in the waveform file. A message will be given if the file header information is used.
If waveform files are generated on the SEISAN system from raw field station files or other input files without response information, the conversion programs (e.g. QNXSEI from a SEISLOG QNX system) will look in the CAL (or alternative) directory to find the response information to include with the Seisan waveform file. The response will be only put into the SEISAN waveform file, if the response is stored in SEISAN format. The response files are generated with RESP, see below and Appendix C.
The instrument response can be defined for each channel of digital data in either SEISAN, GSE, SAC or SEED (ascii version) response format. There are three places in the system where it can be stored. Often the instrument response is part of each channel header in the digital waveform file in SEISAN waveform format (see the Appendix B for format description). However, the instrument response is often not available at the moment the data arrives, or it is later discovered that the response given in the waveform file is wrong. There is therefore by default a directory CAL that contains one response file for each channel and for each date from which it is valid. Since the filenames contain the date from which a change in the response was made and the channel code and component code, a directory listing of CAL will give the history in chronological order of the response of a given channel. This is the most common way to use the response information in SEISAN.
Response information can also be kept in any other directory specified with the environmental variable LOCAL_CAL. The variable must be set with the full path to the directory e.g. /home/seismo/WOR/test/, or on PC, C:\seismo\new\cal\. On Sun it can be set in the SEISAN.csh file and under Windows by using the setting of environmental variables Control panel/system/advances/environmental variables. The variable can also be set from the keyboard (Linux/Sun: 'setenv LOCAL_CAL directory', PC, 'set LOCAL_CAL=directory'. This is a useful option when testing response files.
The file name of the SEISAN, GSE and SAC response files defines the start time of which they are valid. If the response of a station is changed, a new response file must be made with a new time stamp.
The SEED response files are named e.g. RESP.IU.TRIS.10.BHZ (see rdseed below), where the network, station, location and channel codes are part of the filename. When SEISAN looks for a SEED response files it will search for a file with a name matching the network, station, location and channel code of the data in memory.
The RESP filename must match the waveform data, so if the waveform data does not include network code and location code, then filename must be e.g.: RESP..BER..BHZ
So if the location code in the waveform data is '-' then the RESP filename must include the '-', like e.g.: RESP.NO.BER.-.BHZ
And if the location code in the waveform data is empty ' ' then the RESP filename must be like: RESP.DK.BSD..BHZ
The SEED response files can include multiple changes of response and must have valid Start and End dates for the time of the data that are processed. For the current response the "End date" must be set to "No Ending Time".
SEISAN will use this scheme until it finds a valid response file:
In the very unlikely case where a selected data window covers two different response files, the response file belonging to the first part of the data is used.
The RESP program (section 41) can be used to generate the response files. The filenames for the response files are STATTCOMP.YYYY-MM-DD-hhmm_FOR where STATT is station code, COMP is component, YYYY is year, MM is month, DD is day, hh is hour, mm is minute and FOR is the format indicator which can be SEI or GSE. If FOR is not given, the format is SEISAN. An example is BER__S__Z.1999-05-05-1244.. You should take a backup of the response files before you run the program (see chapter 3).
The response files can be located in CAL, or, if many files are available optionally also in a subdirectory structure. This optional structure simply consist of a subdirectory for each station and the subdirectory name must have 5 letters so base BER would have the name BER__. The system automatically locates the response files whether all are in CAL or in the subdirectory structure.
The response file can store the response in different ways:
NOTE: When rotating signals, it is in SEISAN assumed that the response is the same on all 3 channels. !!!!
Response files can be plotted from MULPLT showing the actual response information that is used with a given trace. Response files can also be plotted directly with program PRESP, see below.
All or a subset of the response files can be printed out in a table with program PR_RESP. The program must be executed from the directory with the response files. Make a listing (file filenr.lis) of files to print out with DIRF and run the program. It will produce an output file ready for printing.
A response file can be plotted with the program PRESP. The program is started with commandpresp filename, where filename is the response file name. If no file name is given, the program asks for a filename or number. If a DIRF has been made and the list of files in filenr.lis is available, a response file can then be selected with a number. The program produces a PostScript output file with name presp.eps.