7.4 earthquake in Taiwan, 34km depth | Hacker News Reader

7.4 earthquake in Taiwan, 34km depth

earthquake.usgs.gov

487 pointsthrowaway598176 comments

This was felt pretty strongly in Taipei, some reports of electricity outage too. All my friends and I seem ok though.

Is there a stream of the big dampening device in Taipei 101? I cannot find it via usual searches but wonder if it is available in another language search.

It happened too early 8:03 AM. Probably no visitor or even staff in Taipei 101. Last time videos are all from visitor.

That’s too bad. They should live stream it so they can allow performance after earthquakes, a testament to successful engineering.

I can't imagine how terrifying it will be if I was on Taipei 101 during this quake...

There's no building I'd rather be in during an earthquake in all of Taiwan.

Scratch that - a TSMC fabrication facility is probably safer.

You shouldn't be.

Tall buildings (> 40 floors) are one of the safer places to be in an earthquake.

1) They're engineered.

2) They're already designed for side loads from wind.

3) Their fundamental frequency is well lower than the peak of the earthquake spectrum, so they don't amplify the motion. Good rule of thumb is that period = floors/10, so fo a 50 story building you're looking at a 5 second period. Earthquakes have most of their energy in ~1 second period waves. So effectively, the base wiggles and the top just floats along.

The place to be terrified is in a 6-10 story unreinforced masonry apartment building.

> 1) They're engineered.

Famous last words.

They're more engineered than a 6 story apartment building.

(civil engineer by training, though never got a PE because startups)

I’m also in Taipei. It’s the worst one ever I’ve felt in my life. Staying on the 7th floor with noway to escape in time is probably the worst feeling

If it's any consolation, there's not really time to escape a collapsing building from any level, since the bottom floors usually are crushed first and near instantaneously. If you look at pictures from partially collapsed buildings in Hualian you can see that sometimes it's just the bottom floor that's destroyed and the remaining floors are fine for sometimes days or even until the building is manually taken down.

The pictures of collapsed buildings after earthquakes are scary but the chance of an individual building collapsing are quite low, you're at far greater risk if you're outside near buildings because broken glass or similar could fall on you.

The standard advice is good advice for a reason: stay inside, under sturdy furniture, on your elbows and knees, covering your head. You can learn more here: https://sf-fire.org/disaster-preparedness-information-resour...

I appreciate the context you're providing, but "Hey you'd be dead regardless of where you are in the building at the time if it collapses" isn't as reassuring as you'd think.

Read his comment again

Any data on survivability of occupants of upper floors in cases where the bottom floor is destroyed and the rest of the building settles on top of the rubble instead of pancaking? This is exactly what happened with some apartment blocks in Japan earlier this year, too.

I guess it depends a lot on whether the building stays upright or falls over, as there are plenty of objects that can crush a person in any space that suddenly turns sideways.

Which is why the advice has always been to hide under something sturdy, like a table or bed. For the most part, it should break the impact of anything on top falling onto it.

Yes, the usual advice is optimized for buildings that stay upright. But this [1] happens, too, surprisingly often with reinforced concrete buildings. A table turned sideways isn't going to stop the fridge that's hurtling toward you.

Ships roll all the time in heavy seas but don't have the problem of loose fridges, because everything is anchored much more strongly than on land. Perhaps that's the kind of standard we should aim for, at least for concrete buildings in earthquake zones with a certain width to height ratio.

[1] First photo in https://www.japantimes.co.jp/news/2024/01/02/japan/noto-peni...

Table or bed? What they teach here locally is to get under a door frame or a pillar. I guess that recommendation has something to do with construction codes. When your walls are made out of light wood a sturdy table might protect you from fractured pieces I guess.

If I remember from Introduction to Natural Disasters in college, the door frame example is mainly for masonry buildings, and is a bit of a historical artifact.

My understanding is that advice today strongly suggests getting under something sturdy in the attempt to end up in a void during a collapse event.

> My understanding is that advice today strongly suggests getting under something sturdy in the attempt to end up in a void during a collapse event.

A table is sturdy?

The doorframes here are made out of solid steel. It's literally the most sturdy place to be next to pillar. On top of that if the solid steel door gets stuck you will not leave the apartment until emergency services with steel drills or someone with welding equipment frees you from there. I'm a bit puzzled at what you were taught there.

Contrary to the US, Taiwan has had proper earthquake compliant building codes that for the most part have been enforced(and reinforced) since the last disaster stuck the island over 20 years ago. For most places in the world outside of Japan, Taiwan, and a few other places an earthquake of this magnitude would be a mass casualty event.

One is that in an open floor plan, you may not have time to find the nearest actual doorway.

I would also imagine that the strength of any given doorway would depend on if the wall it was in is load bearing. A non load bearing wall is going to get crushed to bits once an actual load falls on it.

I do think the advice is geared towards traditional furniture and not today’s IKEA particleboard crap.

As a Taiwanese, I got used to it.

Are earthquakes in Taiwan as frequent as in Japan?

The recommendation that I was taught when I took Chinese lessons btw is to stand either below a pillar or a door frame, and to open the doors so that once the initial shaking is over you can escape(for example the door frame might get stuck and then you cannot even escape in case of an emergency).

Most of the collapsed buildings however seem to have not collapsed directly, but rather the foundation seems to have given in, so that if you were above the second floor you would probably have been able to be rescued.

What I found most terrifying is that there are construction sites where material has flown to neighbouring buildings and destroyed parts of them. Imagine your building being up to code and then killing just because your apartment happened to be facing a highrise construction site.

I have a feeling that the injury and death toll will probably take a while to be assessed. This felt genuinely terrifying.

It's quite impressive that this red building Hualien County is just leaning and didn't collapse. Even most of the glass panes are still in place and mirror the ground just fine. Comparing this to earthquakes in places like Turkey when the buildings just collapse into a pile of rubble, I doubt the builders in Hualien could have done much better expect for foundation piles going far deeper than code dictates.

Yeah I agree. A friend of mine in Hongkong told me that his windows (in a new construction) would fall out during smaller seismic events. What I meant though were things outside of the control of the building construction.

In 1999 Taiwan had a major earthquake where some 2400 people died. 3 months after the disaster the government released a change in building construction codes, to some extent, what you see here is the result of those policies being enforced right away. It's quite a feat of engineering, but I don't know to what extent this extends to sewage, gas, and/or other piping systems not directly related to the building construction.

> and to open the doors so that once the initial shaking is over you can escape

Just to make sure to stand far away from the doors you leave open as they can swing and hit you.

I'm in Taichung, it shook hard as hell but I don't think we took any damage comparing to Taipei or Hualien.

I'm also living in central Taipei, this one woke me up, we felt really strong shaking in our flat on the 4th floor. Some tiles fell off the buildings on my street but otherwise everyone is fine.

Right after the quake saw police and fire dept patrolling around.

Japan’s meteorological agency forecast a tsunami of up to 3 meters (9.8 feet) for those curious

https://www.washingtonpost.com/world/2024/04/02/taiwan-earth...

3 m sound not too bad?

It can be a very long wave. If you have a coastal area that is mostly at or very close to sea level then that entire area could now be 3m underwater which is very bad.

Also, even under normal conditions the type of wave you would see at the beach with a height of 3m is insanely powerful. I think the highest I have body surfed in was maybe 2-2.5m and if you get caught in the break of the wave you are literally powerless and get dragged under water and spun around for quite some distance. It is a bizarre experience and doesn't really match the basic intuition of how powerful it should be.

Yep, where I live (rent) a 3m wave would fill my house with 1m of water. Even if it was one quick wave, that washed over without staying long enough to completely flood the inside of my house, I'd probably at least have to deal with a bunch of water damage.

I would be more worried about the force of the water hitting things and about objects it drags with it than about water damage.

One meter of water weighs 1000kg/m². If a wave a few meters long slams into a wall of your house, that can easily be more force than that of a decent sized car doing the same.

My crude understanding and associated mental imagery is that tsunami waves are more like tide than "usual" waves: your typical 3m wave is mostly "surface-ish" in comparison, while 3m tsunami has both greater wavelength and "depth", which is what makes the "usual" wave break quickly when it reaches shore, whereas tsunami waves continue pushing real hard for real long because of all the length * undersurface water you don't see.

Gently shaking a glass of water sideways or blowing upon it won't make much water spill out, but jolt it vertically and most of it goes out real quick.

This is all probably very incorrect physics but it helps drive my intuitive understanding that 3m wave and 3m tsunami are quite different, the former being scary but the latter being muuuuch scarier.

I know 3 m waves are powerful, I was within them.

But they are normal on most coast and that they are bad on a flat beach I can imagine. But Taiwan is quite steep, so aside from some damaged boathouses, I cannot see the big danger (compared to the consequences of the earth quake). But apparently no Tsunami came.

Our local offshore break that we grew up with as kids was, and still is, a total monster in storm (a long way back and out to sea) conditions:

https://www.youtube.com/watch?v=xjHaFOGBPzk&t=138s

https://www.surfertoday.com/surfing/the-right-the-monster-we...

It's a little bigger than 3m at the face and is a "slab" .. not a thin wave but one backed by a mass of water.

Somewhere in that video you might see a triple lip .. three curling breaks, one within another and another again.

Think of it as a 3m tall bulldozer the width of the country with effectively unlimited pushing power. Watch some videos of the Japan tsunami, it's basically an unstoppable force that gobbles up everything in it's path. Very uncool.

It's generally very bad, but it also depends on the shoreline. If your city is on the coast at sea level with no real barrier, it's devastating. If there's a big cliff and everything is at higher elevation, you can ignore it.

Unfortunately, people tend to build settlements at sea level.

Consider for instance this place

https://commons.wikimedia.org/wiki/Category:Diablo_Canyon_Po...

which is on top of a high bluff thus much more secure than the Fukushima NPP.

Unfortunately, that power plant is very close to two fault lines, and they didn't even know when they built it, so it's questionable how well it'll withstand an earthquake.

https://en.wikipedia.org/wiki/Diablo_Canyon_Power_Plant

They increased the anticipated seismic loads and also reinforced the structure accordingly. I don’t think any NPP has been scrutinized to the extent that Diablo has been.

We should set up some stone monuments to delineate the high water mark, for future generations.

Same but different, there's an old town near a lake that suffered a devastating flood way back in its history. The downtown area has markers on the buildings that show the high water mark for the flood.

The Japanese have those. A lot of the monuments are hidden in the forest in the hills above the coastal villages...

That's what parent poster was referring to. They have high tide markers showing where tsunamis have reached, and then in the past 50-100 years people started saying "Waves Of Unusual Size? I don't think they exist" and building closer to the shoreline.

a 3-meter high wall of water is not really good either, due to kinetic energies, if no significant tsunami coastal protections, that water flow can reach far inland

i'm always amazed how little appreciation people have for massive amount of water. a 3m wave will more than ruin your day. trust me, it's plenty

I lived in a coastal region where, every storm season, people would get swept out to sea and drown because they were watching storm waves from the beach. Sometimes they'd even go in the water, out on jetties or out on piers, without any appreciation for the sheer forces at work that push the water around. With enough force, even ankle to knee deep water can sweep you off your feet and suck you out to sea.

Sadly parents let their kids do this and some have lost their lives.

We had two tourists just get killed last week during storms. Both swept off harbor walls wanting to look at the waves. There are literally warning signs everywhere.

Yeah, it's usually tourists or people vacationing from their beach houses. It's sad when more than one person drowns this way, because it usually means the additional people that lost their lives were trying to rescue the others. Recently happened to a parent and their kid here.

Well, I was in the ocean saving a person getting caught by such a big wave on the beach, so I do appreciate the power of the water here. But I think 3 m waves are quite normal on most coast lines in rough sea.

So my question was, why would a Tsunami of the same height be that much worse. And in comparison to the destruction of the earthquake in general here.

A 3 m tsunami isn't like a 3 m wave from wind action. It is like a 3 m tide or storm surge added on top of whatever normal tide level you are at. So if it hits at low tide it might be somewhat negated, but if it hits at high tide it may overwhelm sea walls and other coastal infrastructure.

When people talked above about the long wavelength of a tsunami, they mean it doesn't crest and possibly break in seconds like a surface wave. It can keep building for many minutes and continue to flow inland after the crest.

Look for some tourist video of the 2004 tsunami that flowed over islands in Thailand. It looks more like a river or levee breach flooding over the whole field of view. In steeper coastal terrain, the wave can sometimes be funneled or focused into a narrower port or valley and then reach much higher elevations than you would expect.

Imagine a 10 foot wall of water flooding the coast that just doesn't stop. Smaller tsunamis have devastated coastal regions before. People, beach houses, etc are swept away under that force even though the height is rather low.

3m is about 10 ft which is huge, though there are huger waves.

How tall are you?? Imagine standing at the shoreline and the water level suddenly being 3m higher...

3 meters is significant if you live in an inundation zone. You can see the effect on a seaside community here: https://www.floodmap.net

This comment author was not correct, but this question prompted a lot of great answers and improved the quality of the discussion. Per HN rules, this comment should be upvoted, not downvoted.

It’s not really a wave, it is more of a tide that comes in quickly. So everything gets flooded to 3m above the normal water level, and stays that way for a while.

The tsunami reports I've heard here on the north coast are that it ended up being less than 1 metre tall, so luckily nothing to worry about.

I’m on the southwestern coast of Ishigaki island (one of the closest to Taiwan), and the effects really seem to have been minimal.

The harbor area here was dead silent for a while, empty streets. But things are back to normal now that the tsunami warning has been lifted.

The Circle line has dislodged https://www.setn.com/News.aspx?NewsID=1448651

Live stream of news about the incident: https://www.youtube.com/watch?v=SydScnOrmf8

There's been pretty big aftershocks like clockwork after, it's gonna be a hard day. My dog is freaked out, my fish tanks are half empty, we're not sure if we can open the restaurant since the MRT is closed and our chef lives quite far, and that's just life in Taipei, I can't imagine what people in Hualian are going through.

Edit: the aftershock just now was as strong as some of the scarier earthquakes I've ever felt in my life. Such a rough day for Hualien. I envision going down to the southeast to support but it might actually be impossible right now, due to landslides and the train lines going down. Even if I took my motorcycle, I doubt I could get through.

Edit: pov from a car on an elevated road https://twitter.com/jimmy_su/status/1775347726697587167?s=20 most highways in Taiwan are elevated maybe 100m or so.

Edit: hw9 near Hualien lost a bridge https://udn.com/news/story/7314/7874283 that is one of the most beautiful, well maintained, fun, and safest roads I've ever motorcycled in my life. I hope they are able to repair it.

Edit: I'm furious with YouTube for forcing this to be a "short" with bad-on-purpose UX but here's a video of odd water behavior in a port https://youtube.com/shorts/91Rmacvw4ho?si=d78KAUvDTeg_ovbh

What kind of restaurant do you have?

Southern fried chicken biscuit. So far as we know, first of its kind in Taiwan. https://www.thejispot.com/

Great name!

> Edit: I'm furious with YouTube for forcing this to be a "short" with bad-on-purpose UX but here's a video of odd water behavior in a port https://youtube.com/shorts/91Rmacvw4ho?si=d78KAUvDTeg_ovbh

https://www.youtube.com/watch?v=91Rmacvw4ho

https://youtube.com/v/91Rmacvw4ho

I think my first thought would be to go to the highest point I could find instead of filming the water. Scary!

To use legacy UX you can put in the watch. https://youtube.com/shorts/91Rmacvw4ho to https://www.youtube.com/watch?v=91Rmacvw4ho

> Edit: I'm furious with YouTube for forcing this to be a "short" with bad-on-purpose UX but here's a video of odd water behavior in a port https://youtube.com/shorts/91Rmacvw4ho?si=d78KAUvDTeg_ovbh

Is that water receding as a form of pre-tsunami?

As of 40 minutes after the event, someone has already created a Wikipedia article about it:

https://en.wikipedia.org/wiki/2024_Taiwan_earthquake

And with a timeline of events, assessment of impact, and all 21 references, damn.

While it's cool that Wikipedia could be used for breaking news, "Recentism"[0] is generally not seen favorably on Wikipedia

[0] https://en.wikipedia.org/wiki/Wikipedia:Recentism

Wikipedia itself isn't ideal for this, but I keep waiting for some major news institution to migrate away from the ancient newspaper article format and towards something more wiki-like for covering stories.

I think you could productively fit any kind of news coverage into a wiki article plus a stream of recent updates.

Having a newspaper article giving context and then having updates at the bottom isn't really something that doesn't exist. It happens all the time.

That's just love coverage CNN and major news do nowadays. Find the live event page and there's minute by minute updates, background information / quick recap, other sources and links, images, etc all updated in real time.

That seems to nail the "news" aspect, but missing the vital "wiki" aspect which is all about collaboration and user-edits, which I think the parent was hinting at.

What impress me the most is not that there is a Wikipedia article. But it's translated in 19 languages already !

Anyone have a link to a video of that sweet Tuned Mass Damper (https://www.atlasobscura.com/places/tuned-mass-damper-of-tai...) in action this time around?

Not a live one, but this vid is from 2022 when there was a 6.8 magnitude earthquake

https://www.youtube.com/watch?v=Tkz6b7Q3dRk

It's one of the few tuned mass dampers the public can go see.

I see earthquake reports (like this one) describe earthquake depth. What is the significance of it? Is it somehow actionable?

https://en.wikipedia.org/wiki/Depth_of_focus_(tectonics)

Thanks! I must admit it’s hard to find an answer myself without the vocabulary.

This is where LLMs really shine.

That's a very good point, I was trying to google how buttons on a calculator communicate with the chips underneath, basically how contacts work, and Googling that kept giving me stuff about calculators not electronics, but ChatGPT gave me the perfect bullet point breakdown because it understood the context of my more general question. People often communicate/think in analogies to the underlying concepts.

As somebody who has been through more than 10 thousand quakes, it's as significant as the force itself.

7.4 @ 300km deep you just yawn and roll over. @10km deep you pray your concrete roof tiles don't come through the ceiling and kill you in your bed... you basically just lie on top of your wife, give thanks for the time you did have, and hope for the best.

> As somebody who has been through more than 10 thousand quakes

What's this now?

There are a LOT of very small/deep/etc quakes in some places.

Sure, I see Riverside CA had over 6000 last year, though I doubt if someone who moved there two years ago would say they've "been through" over 10,000 quakes since, depending on their sensitivity, they might not have felt any of them, or maybe a dozen of them or something. I'm interested in specifics!

How many drops is this for you, Lieutenant?

Thirty eight... simulated.

<https://www.cactushugs.com/wp-content/uploads/2016/06/aliens...>

(For the uninitiated.)

Christchurch quake had 10's of thousands of aftershocks. A few people I know were rattled to the point they were shell shocked and just had to leave town - every tremor triggering off "could this be it" all over again.

For a long while, every damn heavy truck driving past the office got a few people going really silent and pensive.

It's a bullshit number - ignore it. The same location could have had 1, 10, 1000 or 1000000 quakes depending on the threshold chosen. Or if just talking about ones you can feel, then the count varies a lot depending on exact location (and your sensitivity).

And the time and strength distribution of aftershocks is not linear - so the number gives you no feeling for what it means.

Well, imagine it like the distance you want to be away from a big explosion. The further away, the better.

And the deeper it is, the further it is away from the ground, where humans are.

A shallow 6 is very different from a deeper 7.

My city Christchurch was fucked by a 6.2 or 6.3 - but it was both (a) shallow (5km) and (b) close to dense residential areas.

Christchurch was near to the 7.1 Canterbury earthquake but that did much less damage as it was 10km deep and Christchurch wasn't on top of it (it mostly affected less dense rural properties in the region near Christchurch).

Be careful not to think of earthquakes as a point - talking about epicentres and precise depth is very misleading. Earthquakes are a fault-sheet or wrinkles. Surface fault lines only show where the 3D rips through the solid earth meet the surface. Fault lines matter a bit especially since they indicate local shallowness, but I suspect we get distracted by them because they make spectacular news reports. Reality is complex.

My parents house was right above the fault plane of the Canterbury fault - and their house came though mostly okay. I know lots of people very near the Christchurch quake and many of their houses were unrecoverably damaged.

The Nisqually quake in Seattle was a 6.8 but it was 57km (35 miles) deep. It was also on the other side of Tacoma sort of in between Tacoma and Olympia.

The depth of that quake made it a lot less severe than it could have been, only killed one person, but estimated it cost $1b to $4b in damages.

I remember how it felt, I thought I was sick, because I felt this rocking motion. I thought it was just me, then I stood up and saw a plant shaking, so I knew it was everything else that was shaking too. The amplitude slowly increased, reached a maximum, then decreased over the course of what seemed like a minute. It was a quite slow frequency, possibly due to the depth. This was from Portland.

I was on the 8th floor of a brick building in Seattle. If it had been a shallow quake I would be dead. There was a whole lotta motion. Can't remember what it damaged, though, other than some glass bottles on top of the refrigerator that fell off and shattered everywhere in the kitchen.

There's actually some interesting visuals that illustrate this showing the difference in ground velocity between the two earthquakes :

2010 Quake: https://www.youtube.com/watch?v=eA5m5PZ5p8E

2011 Quake: https://www.youtube.com/watch?v=KJW7tkTwqXo

And a good explanation vid: https://www.youtube.com/watch?v=aaA1nxYVi6k

IIRC another aspect was that the fault-line that did Feb 22 was underneath the Port Hills, and the volcanic rock deflected the energy towards the city.

Follow-up question. I understand how Richter Scale + depth could be useful, academically. But when communicating with the public, wouldn't it be easier to have a single number - rather than requiring the public to understand the relationship between two different quantities? Maybe: surface-equivalent Richter scale?

Of course transitioning to a new standard would be even more confusing for a while.

Edit: apparently there are many different scales in wide use already. https://en.wikipedia.org/wiki/Seismic_magnitude_scales

In Japan for domestic quakes they always report the Shindo scale (based on surface peak ground acceleration) instead of Magnitude. It’s way more useful a measure IMHO and whenever I see western news report use M it tells me nothing about how bad the quake actually was.

https://en.wikipedia.org/wiki/Japan_Meteorological_Agency_se...

We have a similar measure (MMI) in the US, but it's harder to report a local measure in a single article.

This is why the Shindo scale is favoured in Japan, as it addresses the effects on people and buildings in a particular area, so you can tell just by looking at what is shaking and how much to get an idea of the strength. The worst part of experiencing an earthquake, for me, is realising that anything can happen in the next moments, otherwise it's weirdly normal.

Also could take into account distance to the point of interest. So, in city X you have a number and city Y another.

That's exactly how the shindo scale is reported in Japan. They assign a number to every major city on a map, just like temperatures in a weather forecast.

For example, January's big quake was 7 in Ishikawa, but only 2-3 in the Tokyo area.

I put together a little display of the earthquake action of the first 4-5 hours after the large event using data from the USGS site. [0] You can follow the action from event to event over the course of the first 22 events (about 4-5 hours I think).

[0] https://postimg.cc/HckCRn1m

It started near Hualien City with the magnitude 7.4 and then if you folow the blue lines you can see how it rattled NE-SW mostly along the coast for the first hour (6 events) then following the pink line you can see it popped at the NE end along the coast before a series of events that curved out into the sea to the E-SE for the next hour (5 events). That was followed during the third hour by 4 events oriented roughly NE-SW in green with all but one offshore to the NE. The fourth hour saw 4 more events almost complete the square in yellow. Things were quieting by then and the fifth hour only saw two events, both offshore in light blue with the series ending red with a single event offshore.

I have a graph of activity showing the magnitude and depth versus the time of the event. You can see that the later events trend deeper and lower magnitude making them unlikely to be felt as much as the initial event and less likely to cause as much damage. [1]

[1]https://postimg.cc/zyfdWhNT

To understand the timing of the events I have displayed them in an elapsed time versus magnitude display based on the initial event timing. [2]

[2] https://postimg.cc/68w81VqG

You can see how the magnitude of the events tails off and they become less frequent with time. In the map view you can get a feel for the region's relaxation after the main quake and the other graph allows you to see how the event magnitudes decreased while the later events tended deeper.

not easy to find good devs in taipei, so shameless plug: for folks living in taipei, who want to find a good english speaking dev community: https://join.taipeidev.com

No embedded software as an option?

Glad to find this gem

Haha ok fellow coffee coworker, if we're plugging then I'm plugging the co-op https://508.dev

Are fabs knocked offline?

This was my first thought also. I know they have seismic isolation tech, but how good is it?

I wish this wasn't my first thought too, but it was. I can't imagine all those wafers are happy.

It's really good, best in the world, specifically on the leading edge.

Issue is now probably bridges/ roads and other interlinking logistics - if this stuff holds up supply chain/ stops people getting to and from work that's an issue.

I don't see any major TSMC fabs[0] near the epicenter in Hualien. The Taichung fabs are closer but on the western side, while the Tainan fabs are pretty far away from the quake.

Edit: There will be some disruption as TSMC has evacuated employees[1]. No structural damage reported, though.

[0] https://en.wikipedia.org/wiki/List_of_semiconductor_fabricat...

[1] https://www.bloomberg.com/news/articles/2024-04-03/tsmc-evac...

I'm hopeful for the 3nm Tainan fabs, but I expect disruption at the Taichung and Hsinchu fabs since the quake was felt in Taipei. It's common for the equipment to be recalibrated and tested, if you can feel an event. I suspect this was felt in the west. Hopefully, not down for a week, if everything is good.

Hopefully, the towns and people on the east coast are OK, but this one is big.

edit- Since there are reports of shaking in Kaohsiung (farther south than Tainan), I'm less hopeful for the southern fabs.

The standard operation is to stop and check if any machine was out of calibration. So yes

TSMC is extremely prepared for earthquakes. The YouTuber Asionmetery has a good video on it: https://www.youtube.com/watch?v=MNKF21B01NQ

Its incredible what they do in terms of accuracy in a place with so much seismic activity.

From what I understand (I used to live in the main city for that), even the vibration from local trains are a factor

wow that is crazy attention to detail it must be so negligible but significant enough for the production

TSMC still evacuated. https://finance.yahoo.com/news/tsmc-evacuates-production-lin...

when the entire building shakes you really can't prepare for that

What’s the significance of an earthquake’s depth?

Mostly just significant in terms of distance. If it's deeper, it's farther away and you'll feel it less.

https://pressbooks.cuny.edu/gorokhovich/chapter/causes-of-ea....

https://news.ycombinator.com/item?id=39912520

Shallower earthquakes cause much more damage (on the surface), esp. if near densely populated areas; also depends greatly on how earthquake-proof structures and buildings are. Today's one was 34km deep/21 miles.

1989 SF (Loma Prieta) was magnitude 6.9, epicenter 56 mi south of SF, depth 11 miles. (Typical California earthquake focal depths are 4 to 6 miles).

1906 huge SF earthquake was magnitude 7.9, depth 5 miles.

2008 Sichuan, China earthquake (87,587 killed) was magnitude 8.0, depth 19 km.

2011 Fukushima, Japan earthquake(+ tsunami) was magnitude 9.0, depth of 18.6 mi/30km at sea.

This CNN article hs a list of previous Taiwan earthquakes, magnitude, location, deaths and injuries, but not depth: https://www.cnn.com/2024/04/02/asia/taiwan-earthquake-tsunam...

Report from the Taiwan agency:

https://www.cwa.gov.tw/V8/E/E/EQ/EQ113019-0403-075809.html

And a list of all recent earthquakes, showing the aftershocks:

https://www.cwa.gov.tw/V8/E/E/index.html

Whenever I look at those reports, I keep wondering how many automated systems have to be there in place to generate all that. All the waveform records, the intensity maps, etc..., they should be all auto-generated, and likely verified by humans afterwards? Would be super curious of the IT setup and deployment of such things.

This also likely feeds into the automatic warning systems (sent to mobile phones to warn of an incming earthquake, tsunami, or something else), which is likely going to be discussed afterwards, as loads of people didn't get a warning. (As opposted to recent Chinese satellite launch where _everyone_ got the overly scary rocket alert.)

Edit: now they are saying their calculation has to project a minimum "peak ground acceleration" (PGA) of 25 (what units?) to have an alert, and a lot of the places didn't hit that, in part due to underestimating the intensity at the epicentre. I guess they will be revising this criteria, as this was overly conservative on the "less noise" side, while people are likely more forgiving in reverse (getting an alert when they didn't need one).

18 aftershocks/additional earthquakes across the island in 4.7-6.4 range as of this comment. I’m not sure if that is a normal amount, but seems high.

https://earthquake.usgs.gov/earthquakes/map/?extent=-22.9179...

One must note that epicenter is in Hualien. The small city near the Taroko National Park. Which is know to be beautiful and subject to rock falls. I think it will be close tomorrow.

There is also a power plant there but I guess it will be fine.>

Edit: A lot of the victims are from rock falls. The others part are from falling building. It's really sad.

another 6.5 earthquake after 10 minutes:

https://earthquake.usgs.gov/earthquakes/eventpage/us7000m9gc...

If you're in the region -- please stay safe.

I'm in Kaohsiung (southwest Taiwan) and this woke me up around 8am. I'm on the 15th floor and definitely felt the building shaking. It was more excitement than I wanted that early in the morning.

Last time this happened in Taiwan buildings collapsed and peopled died because corrupt local construction companies had put too much sand in their concrete to save money.

Impressive building codes there after looking at some of the video. I bet a plane could hit them and they'd still stand.

That was a big one here this morning 50 minutes ago. Earthquake alert and later a rare tsunami alert. Very shaky.

TSMC evacuated their facilities, hopefully there is no damage there and no serious injuries or deaths across the country. Taiwan has said there are a number of Chinese warships that have entered their waters as well, supposedly willing to lend aid...though not sure if that's a good deal to take.

Hoping to start a thread talking about this. Anyone read Richter 10 by Arthur C Clarke. Where California broke away from the mainland.

Think anything in our lifetime will ever match the prediction tech from the book?

I know nothing about earthquakes. Hoping for a HN education!

Maybe not the right place to ask, but what’s the tech scene like in Taiwan? Other than TSMC, are there any interesting things going on? I’m in Kaohsiung but haven’t investigated the scene.

If you've never visited South East Asia, then Taipei is probably the best first city to visit over there. Nice people, amazing food, many English speakers, easy to navigate. I'm sure there's a tech scene there, but I would visit even if there wasn't.

Taiwan is not South East Asia, they are not even an Asean member.

You're in for a treat - the tech scene here is phenomenal.

First of all, obviously we have some of the best hardware engineers on the planet, but they all work at TSMC / nvidia and live in Hsinchu so those of us in Taipei never really get to hang out with them.

If you're into it, there's a shitton of cryptocurrency companies and devs out here.

TONS of gamedev companies, especially in the mobile market.

The market is all over the place, we have some really god-tier engineers and then hordes of people building the worst websites on earth in Java / Spring or PHP.

My favorite thing about tech here is the newly formed Ministry of Digital Affairs, headed by https://news.ycombinator.com/user?id=audreyt , who is quite famous in the FOSS world. You can just go to their office during open office hours and have a discussion with them if you want to talk about digital issues, as long as you agree to the conversation being published for anyone to watch.

In line with that is the supremely awesome g0v: https://g0v.tw/intl/en/ Their regular hackathons and slack server constitute an incredible community of tech-associated activists building all sorts of solutions to problems such as fake news (which is a huge problem in Taiwan due to CPC spies / saboteurs), illegal factories, civil defense, etc. If you're in gaoxiong, I highly recommend coming up to Taipei for the next hackathon!

In taipei there's also lots of groups doing regular coworking meetups as well as self-study nights, in English. As for mandarin, I mean, take your pick, every university is chock full of clubs, not to mention meetup.com.

I've been building a software engineering co-op to connect the really skilled english-speaking engineers here with gigs that are befitting their skill level, we're at https://508.dev .

My friend said his building was shaking. I was there last week, didn’t feel it in Tokyo. Looks like some tsunami alerts have been sent. Seing some bad videos from hualien

My Taiwanese cab driver today in Los Angeles said the CCP made the earthquake

Where can I read/watch more about the unfalsifiable theory on an otherwise natural process

I went through a 7.6 quake one time. It was wild.

I hope everyone stays safe.

this guy is right again at predicting these earth quakes: https://twitter.com/ssgeos/status/1775146068079960148

Would love to see the deleted predictions lol

he's been doing this for years with youtube videos accompanying his tweets i dont think hes here to sell anything

this is the first time i've ever seen this person. how is he received in non youtube audiences? is this fringe level stuff, or is this accepted science? i have a hard time wrapping my head around a planetary alignment with Neptune affecting Earth. I'm not dismissing it, but my knee jerk is this can't be real, yeah?

his track record indicates this is not a random reinforcement but something worth looking into. I also didn't know what this planetary alignment is but there have been papers around it.

When I was a kid some of my neighbor's trees occasionally had papers around them too. Might even have been the same kind of paper.

Without independent statistical analysis, he has no "track record".

All earthquake prediction is fringe science.

(FWIW, predicting a M6+ earthquake isn't that hard--an M6+ earthquake happens on Earth about every 3 days.)

Looking at his feed, his big prediction was for a M7 earthquake around 3/30, so his prediction is actually 3 days off.

Before that, he was looking at a M8 earthquake around 3/13.

And before that, strong tremors around the end of 10/23/2023, seeming later updated to more like 10/25/2023 or 10/28/2023.

None of this panned out.

In general, I'm seeing a lot of misses and no real hits.

so he's off by a few days? did you go back to his past predictions that also came true last year and the year before? he's better than any predictive processes we had so far

His prediction is wrong. He buttonholed it to a few specific days, and missed that window.

As for prior predictions, I went back 6 months. None of his predictions panned out.

Although, admittedly, it's a bit difficult for me to figure out what actually constitutes a prediction since most of them tend to be the kind of vague generality that's hard to grade: "The Mars-Mercury-Jupiter conjunction on 22 September can result in a strong shake later on 23 or 24 September."--what's a "strong shake"? "Stronger seismic activity is likely to occur in the coming days"? These are the kinds of predictions that are worded vaguely enough to be dismissed if they don't come true, and can be highlighted as a miracle of prediction if something does happen.

(I haven't even attempted to grade location because I can't even tell what's supposed to be suggested as possible locations. As noted elsewhere, even guessing that a M6+ earthquake is likely in the next few days isn't a difficult guess, because they'll occur more or less weekly: https://earthquake.usgs.gov/earthquakes/browse/m6-world.php?...).

you went only back 6 months and he was off by a few days was to conclude your own bias to begin with...

I went back 6 months to see his history of predictions to see if he had any merit. If you want to be charitable and call this a hit, that's 1 for 5 in predictions. Looking at the history of M7+ earthquakes in the past 3 months, he had no prediction of occurrence for 2 of the 3 earthquakes that have happened (and this pattern of failing to predict M7+ earthquakes continues further back in 2023, I just don't have the stats in front of me right now).

You have provided, to wit, zero evidence of his track record of successful predictions. Not even a (second) example of successful prediction, much less any attempt to actually quantify over any time scale how successful prediction is. And you claim that someone attempting to do the latter is only to conclude my own bias...

Some more data: if you go back to the Wayback Machine archive of his webpage (see https://web.archive.org/web/20231001000000*/ssgeos.org), you can see that there is persistently an earthquake forecast of over 10%, and on frequent basis even 40% or 50%, chance of magnitude 8.5+ earthquakes (right now, it's a 30% chance). The last such earthquake is 2012. Simply put, even on a probabilistic basis, it's an absolute, abject failure of a prediction.

FWIW, I should also point that when I tried to drill down to the scientific basis of the predictions from the source, I arrived at pure numerology: https://ssgeos.org/articles/geometry-dynamics-solar-system.h.... Not a great look!

https://twitter.com/rrichcord reminds me of this guy

Alright. I'm a bit of a geophysicist (oil industry though, not seismology) so I figured I would take a look at this to see whether it holds up to any serious scrutiny. I'm also a huge sceptic and a cynic, only moderately prone to sarcasm but I'm gonna do my best here.

It took a bit of work to dig thru the claims he makes. His claims involve planetary alignments serving as a trigger for earthquake activity so that a careful monitoring of those alignments could potentially be useful as a prediction tool. Venus especially when she falls into alignment with outer planets [0], seems to play a large role according to his work.[1]

[0] https://ssgeos.org/articles/planetary-conjunctions-major-ear...

[1] https://ssgeos.org/articles/venus-and-major-earthquakes.htm

It's interesting that he doesn't blame it all on an alignment of the masses of the planets (gravitational trigger) as much as he blames it on electromagnetic forces which he says are stronger and affect the atmosphere during the conjunction. Alrightee.

The twitter post and the youtube video one shouldn't need to waste their time watching are available in the parent comment and in those videos you can find a general prediction. His predictions tend to be slippery. I understand why. The reason is that you need a lot of wiggle room for an earthquake prediction because you will be wrong a lot more than you will ever be close to being right. The key is to flood the market with plausible predictions that statistically have an elevated likelihood of becoming close enough to gain more eyeballs.

The twitter/youtube prediction of note can be found in the nice red box which SnagIt has helped me grab for your edification. [2] It says there will be a series of right angles (conjunctions in his way of looking at the universe) that will set up opportunities for "major seismic activity, potentially peaking mid 7 magnitude around the 30th".

[2] https://postimg.cc/hhCgfdFR

In fact, he updates his predictions as conditions change as you can see in his March 30 update where he gives a 60% likelihood of a 7.0-7.4 quake occurring. [3] Standard procedure is to update when your prediction date passes since by reason and statistics, that means that you're another day closer to it finally coming true. Betting on the inevitable always works if you can spin the calendar far enough ahead without running out of pages.

[3] https://postimg.cc/QF3gBL8W

In earthquake prediction we have to let close be good enough so obviously this is a winner. Just kidding. I decided to take a look at the solar system to see if I could see what he thought he hoped he was seeing. It took me more than one try to find an app that would allow me to unwind the clock back to his setup dates so that I could figure whether any of it made sense.

I found an app called Solar System Scope [4] that allowed me to wind the clock back and forth to watch how it all shakes out up there in the sky and so I SnagIted some screenshots to bracket the relevant days for analysis. I used his prediction to set the bounds as March 27th to April 2nd. Since the quake happened 3 days after his target date I used the 27th as the start and his target as the center date. I hope that makes sense. The next SnagIts show the planetary positions for those three dates and I gave it a good country try to set the time of the SnagIt to the time of the earthquake on each date.

[4] https://postimg.cc/gallery/W36Gxh4

You can see in the Snags that the inner planets Mercury and Venus are rolling around the Sun getting closer to an actual alignment each day that passes and Mars is headed toward that a similar geometry though it is well out of line during those dates. The nearest outer planets Saturn and Jupiter are well out of alignment while Neptune, way the hell out in center field will almost be aligned though it's relative motion will not matter much over such a short time period. Poor Pluto is out of sight to the right and I found Uranus way out in left field in case you still can't find it. Just trying to help.

The most interesting part of this is that all planets except Mercury will be on the far side of the Sun for a few weeks. Mercury, as you can see, will be swinging around the front of the Sun such that by the time we have the total eclipse next week he will be visible to those who have appropriate eye protection in front of the Sun as the Moon eclipses the Sun on the 8th. I SnagIted a shot of that for anyone interested. [5]

[5] https://postimg.cc/vDDK0ytF

Unfortunately I did not get that shot from the exact same perspective as the others since I managed to click somewhere inside that active window and the whole shitteroo shifted on me. I think a reasonable observer can get the idea though that everything is swinging into alignment over the next few weeks and Mercury, like the Moon, will pass directly between the Sun and Earth just like it does regularly, with high predictability.

To me, if his ideas have any merit this might be a better time to predict an earthquake since all this is falling into alignment anyway. The Moon will be passing over Texas and then on up the hill to good ol' New Madrid, Misery, site of one of the most destructive quakes to ever hit North America. It will darken their day for a full two minutes as it passes. With the Moon close by and Mercury's phat mass between us and the Sun while Venus swings to the backside, all we need is a nice coronal mass ejection to provide the electromagnetic pulse needed to kick start the shaker.

Remember though that this method of earthquake prediction doesn't work any better than a rolling, random guess that gets an update every time the previous time window passes without anything of note happening. No one, in my opinion, will predict earthquakes using planetary alignments and end up being right often enough for it to be attributed to skill as opposed to pure shithouse luck. Even a blind squirrel gets a nut once in a while applies here to this method all year long, every year.

Notice that his latest update after the Taiwan quake [6] suggests the possibility of magnitude 6-7 quakes in a few days, but those days don't include eclipse day even though the chart on that same page highlights eclipse day alignments. For some unknowable reason though the chart doesn't mention Mercury in the April 8th alignment group. Probably an oversight or maybe he just lost it in the glare of the Sun. [7]

[6] https://postimg.cc/9whCS4DL

[7] https://postimg.cc/zHCKZR0v

Anyway, while I was here and dinking with it anyway I noticed that we were gonna get a much better mass alignment in a few weeks that might be an even better place to predict something so I thought I would "test" my own skills by winding the clock till I made a nice alignment in the skies. I tried to put all the inner planets as close as possible to a kinda straight line and make the moon be close to centered over the Ring of Fire so that region would feel the strongest tidal forces during this alignment and maybe something would shake loose. This is what I came up with. [8] This date has Earth, Mercury, the Sun, and Venus roughly in a kinda-sorta line while Mars and Saturn line up on the hypotenuse of a really long triangle. Jupiter is still on third base and I know where Uranus is too in case you still haven't found it.

[8] https://postimg.cc/hJ80R6vc

Please don't take anything that I have predicted seriously folks. I am not a seismologist. I am a geophysicist and believe it or not, that doesn't automatically mean I am qualified or competent enough to make predictions about anything.

Personally, this is all bullshit of the finest quality. I wonder why I have wasted so much of my life assembling this tonight. I should've been sleeping.

Since one of his most prominent hedges is the electromagnetic hedge (according to his report it isn't alignments that trigger these quakes, it is electromagnetic energy). He states that we can't measure that electromagnetic energy and this is why it is so hard to predict even knowing all about these alignments.

Of course he is ignoring that we have had satellites measuring coronal activity for decades and there are numerous sites on land where we track solar emissions and monitor auroras.

Just visit Spaceweather.com or Spaceweatherlive.com and you can get your fill of useful data about solar activity that could trigger the atmospheric events that he needs for his predictions to work.

Looking back over the period of his prediction there was a geomagnetic storm anticipated in the days following a March 30 sunspot eruption but that storm watch was cancelled. The current watch suggests a disturbance around April 4 which likely will only affect the polar latitudes if it happens. [9]

[9] https://postimg.cc/gallery/mP2hcHD

Anyway. I can't help but conclude that anyone making predictions, myself included, using the methods that this guy claims to employ probably is the blind squirrel finding the nut when one of those predictions gets close to a real event.

> I wonder why I have wasted so much of my life assembling this tonight. I should've been sleeping.

I know that feel :)

If it makes you feel better, I read through the whole thing and gave you an upvote. Peace :)

Peace to you too pardner. I appreciate you sticking with me all the way through.

Tsunami warnings in Japan. US is evaluating whether to issue warnings.

Tsunami warnings in Okinawa. Not in the main islands.

Probably not much to worry about in the US. The Fukushima earthquake was much more powerful, but the resulting tsunami did almost nothing to the US west coast. Some boats in Crescent City were wrecked, but that’s due to local geography funneling the wave a certain way. I’m on an island in the SF bay, we saw nothing.

I visited Hualien a few years ago. Very sad to see this.

Multiple buildings have collapsed. There will be many fatalities:

https://twitter.com/Huberton/status/1775320722115539349

https://twitter.com/dom_lucre/status/1775325642386481216

Man in that second video they look so calm standing under a massive potential landslide that could sweep them to their death.

That first one with the leaning building...have they not seen footage from 9/11 or any disaster movie about how far the debris field can travel in an uncontrolled building collapse? That's a definite nope from me.

Most of the pictures are a little misleading for distance etc, as well as time. When people are close to the building is before the fire department arrived, and people are trying to get other people out of the building. It sounds like they're relatively situationally aware, warning eachother about live cables, gas, etc. Obviously there's a risk of death but there's also a risk of your old neighbor getting squashed if you don't get her out, so, I guess it's up to each person how they react.

Within about 20 minutes the fire department had completed its first evaluation of damage and identified the 3 partially collapsed buildings.

If you're a lookie-loo with your phone camera out, you're not very concerned about your neighbor. You're shooting disaster porn for the likes

I've lived here about 18 months and this is honestly the first time I've seen Taiwanese people looking worried about anything ^^ They're a stalwart bunch.

The caption to the first video is misleading. That building is in Hualien, not in Taipei.

https://maps.app.goo.gl/8cErBHJXBa5ynyVy9

Reply on news.ycombinator.com