Hi, Uday,
First let me clarify some of our sloppy wording and that's the use of the word "Event" - it can become ambiguous.
The messages are kicked off upon the receipt of the first Event Mode message. Each Event Mode message contains the tsunami trigger time (note so far it has not been a tsunami that triggers - it's is either the seismic signals or a command-initiated trigger). What is unique is the combination of station id and trigger time (since time on DART buoy is resolved to 15 seconds - it can't be cut any finer).
For our own internal purposes we assign these events an id like 52404_20080429054415 for last night's. We could add that to each message that would be no problem and probably embed it within some tags (welcome suggestions), perhaps:
<Event_ID>52404_20080429054415</Event_ID>

Would that help? We would be open to developing in conjunction with you and NGDC any needed xml tags that could be added to the messages.

Now we don't have it yet, but at some point we wanted to also provide a message when a station resumes normal (or Standard Mode) operations and then give a summary of the Event Mode data (if you look at the drop-down list for a station event it gives a  range of values of the Event Mode data (tt: 2008 Apr 29 05:44:15, ts: 2008 Apr 29 04:00:00 to 2008 Apr 29 08:30:00", where tt is the trigger time and the ts (time stamp of the data) is the range of Event Mode data.

What Kelly would then want is an easy way to retrieve data from - 12 hours to +12 hours around that trigger time.

The attachment is a paper we did last year - has some info on triggering mechanisms, but it's Tables I and II show how the DARTs are employed. You'll note that several stations had more than on trigger - each is unique in that regards. We only one of three types of characterizations of the triggers - Geophysical, Manual, or False. We don't have the expertise to distinguish between the different types of Geophysical Triggers - at some point it perhaps the interface you provide Kelly looks like those tables with Column of "Triggering Mechanisms" being "Preliminary" and then allow PTWC or PMEL to add a final characterization to the table.

Hope this wasn't too much at once. Bottom line if there are existing data that only need some labels placed in the messages that could be done fairly easily.

Uday Kari wrote:

Thanks, Richard.  Indeed, I want to be on that list, if it is allowed.  Further I am thinking that this email notification is probably "dependable enough", performance-wise, for NGDC acrchiving purposes (probably not for warning purposes!).   If our ESB is able to periodically scan an email inbox that we setup exclusively for receiving inbound information, I think we will have one of the possible options of the required "automation" accomplished.  This would facilitate the "PUSH" path from NDBC to NGDC and is indeed scalable to any source that is able to send email.  The converse "PULL" option talked about at the meeting would involved periodically scanning the http://www.ndbc.noaa.gov/dart.shtml page for the last 24-hour information box.  Experience will tell us which one is faster and either way they should be complementary, that is we should have enough sophistication in our database schema to tell us whether we are looking at A) a new event B) an update to the new event C) a duplicate event.  Is there any NDBC generated unique identifier embedded within each message released  that ensures that we are able to tell from A) B) or C)...if not that would be *VERY, VERY* useful to take care of at the source (NDBC) rather than within programming logic in the ESB.  If this is not possible within the development timeframe, we need to probably look at some sort of "key" that may be derived from station ID, timestamp etc that can distinguish between A), B) and C). Your thoughts...

Napravnik, Lee J. wrote:

Thanks Richard.  It would be useful to see these messages if you could sign me up for the mailing list.  It will give me that much more info to work with.

 

-lee

 

---

From: Richard Bouchard [mailto:Richard.Bouchard@noaa.gov]
Sent: Tuesday, April 29, 2008 8:29 AM
To: Uday Kari; Napravnik, Lee J.
Subject: [Fwd: Re: ALERT! DART Event at 52403]

 

Uday & Lee,
This is the follow-up in which the analyst provides an assessment of the character of the event.

If you want, I can add you and whoever to the mailing list for these.

-------- Original Message --------

Subject:	Re: ALERT! DART Event at 52403
Date:	Tue, 29 Apr 2008 01:04:28 -0500
From:	Kelly LaRue <Kelly.LaRue@noaa.gov>
To:	NDBC.DARTAlert@noaa.gov
References:	<200804290554.m3T5sWxh013218@web01.ndbc.noaa.gov>

 

Reason: Suspected false trigger.

The NDBC WEB Data Feed wrote:

> DART Buoy 52403 went into event reporting mode at 05:44:15 UTC on 04/29/2008.

> Event data are available at www.ndbc.noaa.gov/dart.shtml.

> 

>   

DART Buoy 52403 went into event reporting mode at 05:44:15 UTC on 04/29/2008.
Event data are available at www.ndbc.noaa.gov/dart.shtml.