This page includes all of the information I have managed to collect on the seismic instrumentation used in the SKIPPY project. Thanks to Rob van der Hilst, John Grant, Eddie Spassov, and Daisuke Suetsugu for helping with this.
The table below indicates which instrument was used at each station during the different deployments of the experiment. Anchors on 6 of the stations of the SK4 deployment indicate important information concerning a temporary change in sample rate. The instrument types given in parentheses next to the serial number correspond to the transfer functions given in the section on Instrument Response.
| Instrument | Bandwidth | SK1 | SK2 | SK3 | SK4 | SK5 | SK6 | BASS |
|---|---|---|---|---|---|---|---|---|
| T365(ESP) | 30s-50Hz | SA07 | SB03 | N/U | N/U | SE10 | SF03 | N/U |
| T366(ESP) | 30s-50Hz | SA08 | SB04 | SC05 | SD04 | N/U | N/U | YB08 |
| T367(ESP) | 30s-50Hz | SA03 | SB02 | N/U | SD01 | SE04 | N/U | N/U |
| T368(ESP) | 30s-50Hz | SA05 | SB09 | SC09 | N/U | N/U | N/U | YB09 |
| T3136(ESP) | 30s-50Hz | SA02 | SB08 | SC06 | SD03 | SE06 | SF06 | N/U |
| T3137(ESP) | 30s-50Hz | SA06 | SB07 | SC04 | SD08 | SE09 | SF04 | N/U |
| T3138(ESP) | 30s-50Hz | SA01 | SB05 | SC08 | SD09 | SE08 | SF08 | N/U |
| T3139(ESP) | 30s-50Hz | SA04 | SB06 | SC01 | SD06 | SE07 | SF05 | N/U |
| T3205(ESP+) | 60s-50Hz | N/U | SB01 | SC03 | SD07 | SE02 | SF10 | N/U |
| T3206(ESP+) | 60s-50Hz | N/U | ZB11 | SC02 | SD05 | SE05 | SF07 | YB02 |
| T3207(ESP+) | 60s-50Hz | N/U | ZB12 | SC07 | SD02 | SE03 | SF09 | YB07 |
| T3208(ESP+) | 60s-50Hz | N/U | SB10 | SC10 | SD10 | SE01 | SF01 | YB03 |
| T4139(40T) | 30s-50Hz | N/U | N/U | N/U | N/U | N/U | N/U | YB05 |
| T4148(40T) | 30s-50Hz | N/U | N/U | N/U | N/U | N/U | N/U | YB04 |
In the following table are listed the (velocity) sensor sensitivities for which information is available. N/A indicates that the data were not available, and in this case the sensitivity should be close to 2000 V/(M/s). These are straight from the information supplied by Guralp.
| Instrument | Sensor Sensitivity in Volts/(meter/sec) | |||
|---|---|---|---|---|
| Serial No. | Model | Vertical | North/South | East/West |
| T365 | CMG-3 ESP | 2173.0 | 2179.8 | 2163.8 |
| T366 | CMG-3 ESP | 2132.0 | 2207.0 | 2205.8 |
| T367 | CMG-3 ESP | 2104 | 2078 | 2112 |
| T368 | CMG-3 ESP | 2190 | 2178 | 2126 |
| T3136 | CMG-3 ESP | 1990.6 | 2005.6 | 2001.2 |
| T3137 | CMG-3 ESP | N/A | N/A | N/A |
| T3138 | CMG-3 ESP | 1986.8 | 2004.0 | 2001.4 |
| T3139 | CMG-3 ESP | 1985.0 | 2002.2 | 2012.4 |
| T3205 | CMG-3 ESP | 1969.22 | 1999.68 | 1984.58 |
| T3206 | CMG-3 ESP | 1969.98 | 1993.9 | 1980.44 |
| T3207 | CMG-3 ESP | 1980.28 | 1996.9 | 1998.74 |
| T3208 | CMG-3 ESP | 1977.06 | 2002.10 | 1982.66 |
| T4139 | CMG-40T | 792 | 806 | 808 |
| T4148 | CMG-40T | 796 | 804 | 792 |
Reftek recording units were used throughout, but these consisted of two different types:
| Recording Medium | Gain | Dynamic Range | LSB in microVolts |
|---|---|---|---|
| DAT | 32 | 24 bits | 1.19 |
| Disk | 8 | 16 bits | 152.60 |
To calculate the instrument responses one begins by calculating the normalized transfer function, which is specified by the following tables of complex poles and zeroes, which were taken from some of the SEED volumes produced for the SKIPPY data:
| Instrument | Normalization (A0) | Poles | Zeroes |
|---|---|---|---|
| ESP | 574656 | (-1.47900e-01, 1.4790e-01) | (0.0,0.0) |
| (-1.47900e-01,-1.4790e-01) | (0.0,0.0) | ||
| (-3.11020e+02, 0.0) | |||
| (-1.84726e+03, 0.0) | |||
| ESP+ | 574655 | (-7.39500e-02, 7.39500e-02) | (0.0,0.0) |
| (-7.39500e-02,-7.39500e-02) | (0.0,0.0) | ||
| (-3.11020e+02, 0.0) | |||
| (-1.84726e+03, 0.0) | |||
| 40T | Same as for ESP above, only sensor sensitivity is different | ||
Once the normalized transfer function is calculated, the actual transfer function is obtained by multiplying it by 3 factors: (a) the sensor sensitivity, (b) the gain of the recording unit, and (c) the inverse of the LSB. See Sensor Sensitivities to determine (a), and Recording Units to determine (b) and (c). CAVEAT: The sensitivity written into the SKIPPY seed volumes (blockette B058F04) is the sensor sensitivity (a) only. In, e.g., IRIS seed volumes, this sensitivity includes everything which is multiplied with the normalized transfer function to obtain the complete instrument response, which in the case of SKIPPY would be (a)*(b)*(c). Thus, programs for evaluating the instrument response based on the information in the seed volume should be used with care. Recent experience indicates that the program 'evalresp' available from IRIS will not work at all, but
Almost all SKIPPY data were sampled at a rate of 25 samples per second. However, during SK4 the failure rate for the DAT tape drives was particularly high, so the sample rates for some stations were samples per second in the hope that the drives would last longer. The stations and dates of the sample rate changes are:
| Station | Changed 25->20 samp/sec on |
|---|---|
| SD01 | 26 May (Day 146) 1995 |
| SD03 | 28 May (Day 148) 1995 |
| SD07 | 30 May (Day 150) 1995 |
| SD08 | 31 May (Day 151) 1995 |
| SD04 | 1 June (Day 152) 1995 |
| SD10 | 3 June (Day 154) 1995 |
Geoff Clitheroe has written a simple Perl script which interpolates 20s/sec SKIPPY data in SAC format to 25s/sec. Click here to download it. Please read the comment about the "grep" command used in the script, and also note that it needs to run the command rh_sac_delta in Geoff's bin directory, ~geoffc/bin.