List of TIDs for future work
vc1492a opened this issue · 24 comments
We need to put together a list of tsunami events which resulted in clear TID signatures. It would also be good if we could go ahead and get the data for those (in the same format as previous work) (12 days before the day of the earthquake, day of earthquake, and two days after if possible).
Labels are important. We need estimated start times, but also end times specified for each location / event, for each satellite.
Hi @MichelaRavanelli do you have any updates you can share? Thanks!
@MichelaRavanelli do you happen to know the second of day (SoD) start and finish times for the TID in the Chiillean data from 2015 on day of year (DoY) 259? Before we have all the other data and labels available, we could start a basic form of the experiment with the Chillean data serving as out of sample data (while the model is trained on the 2012 Hawaii event). This would be a big help in the meantime. Thanks!
Hi @MichelaRavanelli can you provide an update for the above? As a reminder, we can get started on experiments that mirror the broader experiment we want to accomplish with just a handful of labels for the 2015 Chile earthquake and tsunami event. The labels for the Hawaii data look like this:
{
"302": {
"G04": {
"start": 31400,
"finish": 33200
},
"G07": {
"start": 31160,
"finish": 32960
},
"G08": {
"start": 31900,
"finish": 33700
},
"G10": {
"start": 29900,
"finish": 31700
},
"G20": {
"start": 31150,
"finish": 32950
}
}
}You can provide it in the above format or simply the name of the satellite with the day of year, start and finish times (in second of day) of the TID (e.g. G20, 259, 29000, 32540).
Hi Val,
sorry for the late answer: it was a busy week.
These are the satellites in which the signature is stronger.
G12: 83059 - 84600
G24: 83179.0-84469
G25: 83209-84889
I want always to highlight that these start-end times (sod) are really dependent by the positions of the IPPs.: I am considering the closest IPPs tracks to the epicenter, where we want to start the detection.
In this case, we are looking at a different wave into the ionosphere: the AGWepi and not the IGW tsuna.
Thank you @MichelaRavanelli! Thanks also for noting that the start times are dependent on the positions of the IPPs. Unrelated to the above discussion, but is it possible to quantitatively describe the relationship between the start times and the positions of the IPPs? I'd like to get a better understanding for my own knowledge.
Do we know which other events we want to capture in our total of 10 events (our target / goal)? The 11 March 2011 Tohuku event is one - others?
For now, I will suggest
- 2011 Tohoku-Oki earthquake and tsunami and foreshocks (Mw7.4 Sanriku-oki) - Japan
- 2014 Iquique earthquake and tsunami (Chile)
- 2020 Alaska Peninsula (Mw 7.8)
For these events, I can for sure retrieve data easily.
I'm doing other researches for other events, such as in New Zealand and other Pacific zones.
I will send them to you in the next days
We can also add
- 06/04/10 Banyaks 7.8 earthquake
- 25/10/10 Mentawai 7.8 earthquake
- 2009 Fiordland earthquake (New Zealand)
- 2010 Maule earthquake and tsunami
@MichelaRavanelli really excited to see this list!! 😄 to clarify, the earthquake events above generated tsunami waves which generated signatures we can detect in the ionosphere?
In detail:
- The 2020 Alaska Peninsula (Mw 7.8) generated a relatively small tsunami that arrived approximately 20 minutes later at the closest DART 46403, located approximately 295 km away from the epicenter.
- The Banyaks earthquake generated a much weaker tsunami that did not cause any damage
- 2009 Fiordland earthquake also caused a small tsunami
The other events generated very strong tsunami waves.
I think that analysing small tsunami events can be also useful to improve the response of warning systems, that always triggered warning, even though they didn't cause any damages.
@MichelaRavanelli wanted to ask if you thought there was one more event we could identify and add to the list to gather data for (so we have an even 10). Any thoughts?
Related, made a visual to show the locations of the events. Useful for when we are explaining our work.
@MichelaRavanelli wanted to ask if you thought there was one more event we could identify and add to the list to gather data for (so we have an even 10). Any thoughts?
Hi Val, let me think about it, but I could definetely find another one!
Related, made a visual to show the locations of the events. Useful for when we are explaining our work.
Super cool!
It gives a perfect idea of the tsunamigenic potential of the Pacific zone!
I would plot also the Haida Gwaii event that is in Hawaii, so in the middle of Pacific! ;)
In any case, we could try if we see something with the 2013 Solomon Islands earthquake and tsunami.
Hey @MichelaRavanelli wanted to get an update on how Chile / Japan data prep is going, thanks!
Hi Val,
I finally finished processing the Chilean dataset that I sent you: it took me more time than I thought.
For what concerns Japan, I'm requesting the data since my access was expired.
Thanks for your patience!
An update about Japan data: they have just informed that the application may take up to 10 days.
I will keep you posted and thanks for your patience!
Thanks for the update @MichelaRavanelli!
I would plot also the Haida Gwaii event that is in Hawaii, so in the middle of Pacific! ;)
Well, I would but this plot shows earthquake epicenters, not locations where we see tsunami detections from TIDs.
New plot!
Cool!
@MichelaRavanelli any updates you can share about Tohoku and the other events we have identified? We need the modeling data soon or we will miss or objectives for the project this fiscal year. Thanks!