Prof. Diego Melgar
@diegosismologo.bsky.social
660 followers 160 following 88 posts
Director of the Cascadia Region Earthquake Science Center (@cascadiaeqs.bsky.social) and Associate Professor of geophysics at University of Oregon. I research big earthquakes and tsunamis.
Posts Media Videos Starter Packs
diegosismologo.bsky.social
That's a good way to put it, they need a day job. I do think putting things like this into assimilation approaches like 4d-Var used in weather prediction has potential
diegosismologo.bsky.social
What do you think of this @rocangel.bsky.social? The Boussinessq solver in geoclaw is still depth averaged so, perhaps? My money is on the source we used lacking short-wavelength detail. I've run some tests and with a "rougher" initial condition you quickly get a lot of dispersive ripples
diegosismologo.bsky.social
The data is not real-time, there is latency, but that aside and assuming we solved it, I could see a few of these being useful for real-time data assimilation. The issue is that I'm not sure you could justify the cost of those missions on tsunami hazards alone since big events are so infrequent!
diegosismologo.bsky.social
Some more:
• There is clear evidence that dispersion matters behind the main front, shallow water equations alone don't cut it. This is very visible in the SWOT pass
•.While we were able to model dispersive waves, we're still missing some energy. Updated source models need to account for this!
diegosismologo.bsky.social
Some highlights:
• First high-resolution spaceborne track of a great subduction-zone tsunami
• DART-driven inversion points to ~400 km rupture with peak uplift ~4 m (longer and farther downdip than early slip models)
• A blended source best matches both DART and SWOT observations
diegosismologo.bsky.social
On July 29, 2025 a Mw 8.8 earthquake off Kamchatka launched a Pacific-wide tsunami, right as the SWOT satellite flew south-to-north taking altimetry measurements of the wave train. In our new preprint, we pair SWOT with nearby DART buoys to capture the tsunami’s propagation and its source
diegosismologo.bsky.social
Discussing tsunamis at @omsiedu.bsky.social last night was great fun. Definitely the biggest screen I've ever presented on!
Reposted by Prof. Diego Melgar
rocangel.bsky.social
We are entering a new era for tsunami observations and model validation-- The SWOT era! Thanks a lot @baptistegomb.bsky.social for reaching out, this was a fun collaboration. Also great working with @diegosismologo.bsky.social and @valah.bsky.social. @haskoliislands.bsky.social
baptistegomb.bsky.social
Superb observation by the SWOT altimetry satellite of the July 30 #tsunami. Data acquired ~1h after the M8.8 Kamchatka earthquake show the waves propagating across the Pacific with a great agreement with @rocangel.bsky.social's model!
[more here](www.aviso.altimetry.fr/fr/missions/...)
#CNES #CLS
diegosismologo.bsky.social
I hate we can't trust anything anymore more
diegosismologo.bsky.social
Lol, as the parent of two dogs, this was of big concern for me too!
diegosismologo.bsky.social
Right, to be clear––the screen grab comes from an email sent to the ITIC tsunami mailing list which is the "official" mailing list for tsunamis scientists. 99.99% odds the vid is legit
diegosismologo.bsky.social
A sprawling collaboration with colleagues in Myanmar, Thailand, Singapore, China, and the US. Yes, undoubtedly super-shear, and unique because a set of trenches had fortuitously been opened at the Sagaing fault where the event nucleated a few years before the 2025 earthquake
weareseismica.bsky.social
🌏A Supershear Earthquake Strikes Myanmar: Fast, Long, and Repeating
Melgar et al. reveal a fast 450 km rupture from March 28 M7.8 quake, ties to past Sagaing Fault quakes, highlighting seismic repeatability & the urgent need for resilience.
Read more: seismica.library.mcgill.ca/article/view...
diegosismologo.bsky.social
It's real. See my recent post.
diegosismologo.bsky.social
Video is almost certainly real. From colleagues in Russia:
diegosismologo.bsky.social
FWIW most (all?) my "tsunami friends" think it's real. Not proof by any means, but no one sees physical incongruities
diegosismologo.bsky.social
The ambiance looks really similar to the other video does it not? +1 for the dog video being real.
diegosismologo.bsky.social
Having lived in Oregon for long enough now this also doesn't look quite like your regular sneaker wave. It could be something more niche like a meteotsunami but...
diegosismologo.bsky.social
Does it? It's hard to tell where z=0 is pre- and post-, but I agree this is the dumbest timeline where we can't trust anything...
diegosismologo.bsky.social
The period looks hella short yes but (i) tide gauges sample at 1-6 min so you wouldn't see shorter periods and the vid is not long enough to see if the intial surge goes on for longer. I'm on the fence but calling this one 50-50 still
diegosismologo.bsky.social
Could be! Fun game of geo-guesser
diegosismologo.bsky.social
This looks consistent with (official) reports of 10-20 m run-ups in Kamchatka/Okhotsk. With complex topo/bathy short wavelength features like the waves seen here can develop and superimpose on the long period tsunami. We've seen this in other places. Time will tell if this can be verified
newseye.bsky.social
NEW: Extraordinary footage has emerged of the huge tsunami that hit Russia’s remote Kamchatka peninsula after the 8.8 magnitude earthquake on July 30th.

(🎥 Doni Nikz)
diegosismologo.bsky.social
A report was shared on the IOC UNESCO mailing list showing ~17 m at Paramushir island in the Sea of Okhotsk, so this doesn't seem too far off
diegosismologo.bsky.social
In 2004, it took many, many days to produce detailed tsunami models after the devastating Sumatra earthquake. Last night? Slip models were being produced and shared within hours. That’s federally funded science in action.
diegosismologo.bsky.social
Last night, after the M8.7 in Kamchatka, warnings went out to Hawaii, Japan, and the U.S. West Coast within minutes. Wave forecasts were updated as new info came in. That’s a huge leap from where we were even 20 years ago.