[Realtime-feed-users] Duplicate Fiji earthquakes last night

[Fee, Jeremy]

Hello,

This was due to an error in one of the messages from the National Tsunami
Warning Center, which prevented the automatic association of events.  They
are aware of the problem and have taken steps to prevent it in the future.


Here is the partial timeline:

2016-05-28T05:38:51.430Z origin time

2016-05-28T05:44:56.000Z PT solution
Event added, PT solution is most preferred.

latitude = -21.961
longitude = -177.862
magnitude = 6.6 Mi
time = 2016-05-28T05:38:49.000Z


2016-05-28T05:49:30.999Z AT solution with rounding problem
Event updated, AT solution is most preferred (neither AT and PT are
authoritative, and AT contributed most recently)

latitude = -22
longitude = -177
magnitude = 6.6 Mi
time = 2016-05-28T05:38:50.000Z


2016-05-28T05:56:54.040Z US solution
Event added, US solution is most preferred (US and AT solutions outside
automatic association window of 16s and 100km)

latitude = -22.0213
longitude = -178.1606
magnitude = 6.6 Mwp
time = 2016-05-28T05:38:51.430Z


2016-05-28T06:27:36.324Z AT solution deleted
PT and US solutions automatically merged, US solution is most preferred.


Thanks,

Jeremy


On Sat, May 28, 2016 at 9:20 AM, James Richards <james at artisanglobal.com>
wrote:

> Hi Jeremy,
>
> Friday night at around 10:45 - 11:00 pm Pacific Time, I observed that two
> earthquakes were appearing separately in the real time data feed located SE
> of Fiji. The associated quake IDs were pt16149050 and us20005yqp. A short
> time later the quake with ID pt16149050 disappeared from the feed and was
> instead listed as an alternate quake ID in the ids property for us20005yqp.
>
> Can you tell us anything about what caused this, and what the proper
> terminology is for such occurrences? For example, is this what you would
> call a “ghost” event?
>
> Also, is there an errata associated with the real time feeds where we can
> verify any mistaken events that may have been reported and withdrawn? I
> couldn’t find anything on earthquake.usgs.gov about that particular quake
> pt16149050 being in error.
>
> Thanks in advance!
>
> Kind regards,
> James
>
> James Richards
> Artisan Global LLC
> james at artisanglobal.com
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://geohazards.usgs.gov/pipermail/realtime-feed-users/attachments/20160601/4fab1c0d/attachment.html>