TIMING CORRECTIONS

The log files had to be edited in order to be used properly by the PASSCAL program, refrate, in order to make the appropriate timing corrections. Because every instrument had relatively old Reftek firmware, the required timing corrections were written incorrectly to the log files. Refrate reads the log files and pairs statements regarding timing events, such as `unlocked-phase error' and `phase error-DSP clock set.' (`Unlocked-phase error' defines the time interval between the last time the Reftek clock was unlocked or not synchronized to an external transmitter, and the report of the phase difference between the signal from the external clock and the Reftek's internal clock after the internal clock was synchronized to an external transmitter. `Phase error-DSP clock set' defines the time interval between the report of the phase difference after locking, and the resetting of the digital signal processing (DSP) chip clock when the phase error between external and internal clocks exceeds limits of -5 and +11 milliseconds.) These event pairs formed intervals in which time corrections were interpolated and used by clockcor to correct the absolute start times in the header of each event Some corrections were simple and straightforward; however, we will now address what was done for the more problematic cases.

In many cases, the log files had to be edited by hand to report correct event pairs in order to make the correct time change to the header. All of the log files have names of the form I.** where ** is the DAS number of the Reftek at the sites. Specific lines in the log files containing `phase error' or `DSP clock set' event reports should have contained all essential timing correction information. Normally, the corrections contain a seconds and a milliseconds portion of time offset. However, for almost all of our cases, the milliseconds portion was not recorded in the log file. Moreover, if the offset was positive, a whole second was erroneously added. These errors were corrected by manually editing the log files so that they reported the correct phase error. The corrected phase errors were then used by refrate and clockcor to correct the start time in the event headers, as before. The modified and original log files have been stored together with the data set.

There still remain some problems with the absolute timing in the seismograms which have not been modified but should be easy to observe in the data. The one-second (leap second) and 10-second (time pulse) errors were not removed during the data processing from the stations with Omega timing receivers. See Table 1 for the stations where these timing errors can be expected.

Table of Contents