Thread: MUSTANG timing_quality metric, bug fix

Started: 2021-07-06 09:32:26
Last activity: 2021-10-05 17:32:53
Topics: MUSTANG QA
gillian@iris.washington.edu
2021-07-06 09:32:26
Dear MUSTANG Users,

It has come to our attention that the MUSTANG timing_quality metric has been calculated with a slightly different algorithm than the intended method in its description. The timing_quality metric should r eport the daily average of timing quality values stored in SEED Blockette 1001. Instead, it has been reporting the sum of the Blockette 1001 values divided by the number of miniSEED records, under the assumption that the number of records is equal to the number of Blockette 1001 values. While this is often a good assumption, some digitizers used by the IU network prior to 2015 did not record Blockette 1001 values for the first and last record of the day, resulting in slightly lower timing_quality values than intended. This issue may also affect other networks and other times.

We have corrected the code for the timing_quality metric so that it will correctly calculate the average of the Blockette 1001 timing quality values per day as described in [ http://service.iris.edu/mustang/metrics/docs/1/desc/timing_quality/ | http://service.iris.edu/mustang/metrics/docs/1/desc/timing_quality/ ] . All new calculations of the metric going forward will report the correct values and we will be re-calculating the metric back through the archive for days that recorded values greater than 0 or less than 100. This criteria should cover all instances where the number of records may be different from the number Blockette 1001 values. The number of days potentially affected by this issue represents ~10% of our archive and it may take several weeks to recalculate.

Best Regards,
Gillian Sharer
IRIS DMC


  • gillian@iris.washington.edu
    2021-10-05 17:32:53
    Dear MUSTANG Users,

    We have completed our recalculation of the timing_quality metric across our archive.

    Best Regards,
    Gillian

    From: "Gillian Sharer (via IRIS)" <mustang-qa-bounce<at>lists.ds.iris.edu>
    To: "mustang-qa" <mustang-qa<at>lists.ds.iris.edu>
    Sent: Tuesday, July 6, 2021 9:33:20 AM
    Subject: [IRIS][mustang-qa] MUSTANG timing_quality metric, bug fix

    Dear MUSTANG Users,

    It has come to our attention that the MUSTANG timing_quality metric has been
    calculated with a slightly different algorithm than the intended method in its
    description. The timing_quality metric should r eport the daily average of
    timing quality values stored in SEED Blockette 1001. Instead, it has been
    reporting the sum of the Blockette 1001 values divided by the number of
    miniSEED records, under the assumption that the number of records is equal to
    the number of Blockette 1001 values. While this is often a good assumption,
    some digitizers used by the IU network prior to 2015 did not record Blockette
    1001 values for the first and last record of the day, resulting in slightly
    lower timing_quality values than intended. This issue may also affect other
    networks and other times.
    We have corrected the code for the timing_quality metric so that it will
    correctly calculate the average of the Blockette 1001 timing quality values per
    day as described in [
    http://service.iris.edu/mustang/metrics/docs/1/desc/timing_quality/ |
    http://service.iris.edu/mustang/metrics/docs/1/desc/timing_quality/ ] . All new
    calculations of the metric going forward will report the correct values and we
    will be re-calculating the metric back through the archive for days that
    recorded values greater than 0 or less than 100. This criteria should cover all
    instances where the number of records may be different from the number
    Blockette 1001 values. The number of days potentially affected by this issue
    represents ~10% of our archive and it may take several weeks to recalculate.

    Best Regards,
    Gillian Sharer
    IRIS DMC

    ----------------------
    MUSTANG QA
    Topic home: http://ds.iris.edu/message-center/topic/mustang-qa/ | Unsubscribe:
    mustang-qa-unsubscribe<at>lists.ds.iris.edu

    Sent from the IRIS Message Center (http://ds.iris.edu/message-center/)
    Update subscription preferences at http://ds.iris.edu/account/profile/

08:46:50 v.01697673