5.1 Magnitude Earthquake Near San Jose, CA
earthquake.usgs.gov
earthquake.usgs.gov
Even if you didn't feel but live in the Bay Area please fill it out, they need that data too!
But expected intensity information isn't included because: 1) it's still a very imprecise estimate at that time, 2) actual intensity of ground motion (what you care about) will vary dramatically by location, and 3) people don't deal well with that kind of information in an alert that is really meant to be "hey, take cover".
I'm thinking the more info you provide, the longer it takes the brain to process, and split seconds matter here.
I suppose the people getting these notifications really just want to know if they should scramble for safety immediately.
probably true, but when you live in an earthquake zone you pick it up fairly quick.
for me it is sort of like
5 lots of shakes.
6 damage starts
7 lots of damage
8+ god help us all
If you send notifications for minor earthquakes like this, an estimate of the intensity is mandatory. At least in countries like Chile (and Japan?) where earthquakes are common. Otherwise people will learn to ignore them as false alarms.
I agree with your point that the message should have way less info which could buy me additional time.
Something like: 4.5-5.9: “Earthquake cat 1” 6.0-6.3: “Earthquake cat 2”
Etc. something to tell me, get ready, or something to tell me this is going to be huge, but not precise numbers or even better:
“Earthquake, 3.87 seconds” with a count down would be ideal. If I have 7 seconds I can get outside. 2 seconds and I duck and cover.
Remember, for this to be usable you need to start sending it out BEFORE the actual earthquake ends at the place when it is first detected.
Followed by a couple of tiny wobbles and then business as usual.
(In this case it seems like the app worked.)
Earthquakes move through the ground at the speed of sound. Once earthquake detectors on the ground near the epicenter detect it they send the information out at the speed of light, so even factoring in a little bit of delay for calculations, advance warning on the order of several seconds is possible depending on how far away the recipient is.
https://earthquake.usgs.gov/earthquakes/eventpage/nc73799091...
Santa Clara had 2 seconds of advance warning, and Sunnyvale had 4 seconds. San Francisco got 18 seconds of warning, though was outside the "ground motion" zone.
Though I wish it would have overridden silent mode. 10s of advanced warning isn't all that helpful if I only find out about the notification 10 minutes later.
Apparently my phone supports notification history, but it was also turned off by default... Bad defaults strikes twice in a row.
https://earthquake.usgs.gov/earthquakes/eventpage/nc73799091...
Santa Clara had 2 seconds of advance warning, and Sunnyvale had 4 seconds. San Francisco got 18 seconds of warning, though was outside the "ground motion" zone.
Definitely worth checking the earthquake kit though!
Probably the latter since I think I did grant location privileges to the app.
> The developers quite correctly spent their time..
We don't know the truth of those assumptions. I've seen reports of MyShake failing to report an earthquake at all which indicates the opposite.
If your bar is "it must work perfectly" you probably won't be satisfied with any earthquake notification or prediction tools for many years to come.
That said, this app provided value to many people during this quake, and likely has access to a much larger number of sensors than any other providers that the HN ecosystem seems to know about, so it's unlikely there's a magical additional service out there that's astoundingly better than this one.
But you're certainly free to uninstall the app and dis the company for not being perfect, that's your choice to make.
I'm happy they are doing the work they are doing and looking forward to them continuing to get better, as innovative technology products frequently do.
For the first few minutes a Google search for "earthquake" initially had it located some distance from "San Hose". Clearly someone was typing that it in manually.
Any app is going to make use of the fact that internet messages travel faster than earthquakes, so the farther you are from the epicenter the more warning you'll get. Close to the epicenter is going to be no warning regardless of whose app you use.
The details are messier because groundwater flow very much does affect earthquakes (poroelasticity), but by and large, seasons don't affect groundwater flow significantly at the depths involved. E.g. wet vs dry weather doesn't matter. But on the other hand, a lake filling up or draining can, especially in terms of triggering very small earthquakes.
https://www.accuweather.com/en/winter-weather/what-is-a-fros...
Many years ago ago I looked quite deeply at stick-slip friction models, the mathematics, dynamics and signals.
I'd wager, in the time since 2005, there's a lot we could do with machine learning, predictive signal analysis and good historical data.
Even 5 minutes warning could save a lot of lives.
Any ML + geo-scientists on the case?
I heard this on Audacy from the Philadelphia area.
Not to take away from ShakeAlert in any way though! Just clarifying that in terms of another comment I made about earthquake precursor signals.
ShakeAlert and similar aren't about precursors, but instead are about "the earthquake has happened, let's tell folks before the damaging surface waves reach them".
Here's the short version:
We're good at predicting where. We're still not good at when.
Understanding what segments of a fault zone have built up the most elastic strain is something we're really quite good at these days. Given historical earthquake data, GPS measurements of deformation, and knowledge of the geometry and makeup of the subsurface, we can reasonably accurately model what segments of a fault are most likely to fail next. No ML needed or wanted.
The issue with the "when" is what the precursor signals are.
The precursor signals we're most likely to be able to reliably measure involve drilling down and putting sensors on both sides of the fault zone and measuring the sonic velocity across the fault zone repeatedly. As rocks or even a fault gouge begin to fail, they have to dilate slightly so that grains can rotate into position to slip (bad explanation, sorry). This dilation causes a drop in sonic velocity. Stresses on the rock mass as a whole also have velocity signals. Lab experiments have observed this fairly reliably. We've actually set things up to measure this in the field as well -- look up SAFOD. There were some likely precursor signals (a bit contentious, but it's fair to say there were interesting signals in the data). However, this type of instrumentation just isn't practical to deploy at scale.
Other possible precursor signals involve hydrologic changes (e.g. well waters rising / changes in pore pressure) or Radon spikes, both due to the same dilation. Again, though, these are local effects and difficult to deploy a system to measure. (e.g. you need a deep borehole for the hydrologic changes, and the radon spikes are most reliably observed at depth as well)
There is one more controversial category that is easily observable: ionospheric perturbations. GPS data is sensitive to the ionosphere, and one of the things you need to do in geodetic GPS station operations is correct for changes in the ionosphere. We tend to have tons of geodetic GPS stations deployed along fault zones. Folks have noticed some very interesting ionospheric perturbations before very large earthquakes. There have been a lot of papers published on this, but it's still not too widely accepted, last I checked. The proposed mechanism (piezoelectric effect and/or dislocation glide/creep building up charge) and the observed magnitude of change don't match. It also has only been documented as a precursor around very major earthquakes (which are rare) and only in retrospect.
At any rate, tons of data don't help if it's all noise. The data that we think should have the most signal is _really_ hard to get. Things are still in the "promising, but a long ways off" state when it comes to predicting "when", and seismologists are _very_ careful to avoid doing work that could be taken as predictions due to past PR SNAFUs.
I suppose another issue that then arises, even if we did have decent prediction, is how to present and use predictive info. A "traffic light" system leading up to a countdown? How not to scare people, destroy economies with false positives and off-timing?
A strange conclusion from my study (for sound) is that where eathquakes are concerned the next best thing to none at all is regular earthquakes.
1. Prediction actually is getting quite good, just not on the time scales that we're interested in, which would probably be somewhere between weeks and minutes.
2. I don't think that at the moment we have access to input variables with sufficient power to be predictive on those time scales. Right now, I think one of the best predictors of large EQ's on short time scales is a sudden increase in variance of seismic activity. Which is sort of obvious when you think about it... The best predictor of a big slip is when it's already started slipping.
3. There have been good gains on minute time scale warnings simply by using sensors and communication networks that can outpace the speed of the seismic waves. This can actually be incredibly helpful for seismic sensitive systems!
I am surprised California and other wobbly places don't use the same system, with the addition of quake and forest fire alerts. Everybody here knows the alert sound instinctively now, so even if you're in public and hear someone else's phone go off, we all know what it is.
It's funny to me that one of the world's tech capitals has taken so long to get this rolling. To be fair, California doesn't have earthquakes as regularly as Japan (at least not of the same magnitude). I guess this is another symptom of our crumbling institutions in the US, we only get services quickly if there's a profit to be made. In the popular imagination, innovation is only the domain of private companies.
Anyways, the wireless emergency alert system has like 5-30s of latency. I suspect that might be enough to make push notifications a better solution. There's a # of apps taking that approach (full disclosure, my employer makes one of them), I think they all ingest data from a private USGS Kafka server
One downside with WEA is that it doesn't give a ton of granularity, so in theory they only issue them in areas where the alert will potentially save lives (despite latency concerns). One thing that the Android alerts do which I think is smart is to send out lower-priority alerts for non-dangerous earthquakes that still have enough intensity to be felt. Having that lower prio alert helps build confidence and trust in the system. Turns out people actually like advance warnings for smaller earthquakes, and unlike a real "oh shit duck and cover" alert those ones don't bypass Do Not Disturb mode.
[1]: https://www.shakealert.org/education-outreach/event-review-f...
To me, this kills its entire purpose. The boy who called wolf, and all...
I haven't had one in a really long time now, so maybe they've fixed that.
I lived in lake tahoe at the time and was a freshman in HS and was at football practivce at 5:04 PM when it happened, and my dad was waiting in the car and it car shook so much he got out because he thought us kids were pranking him by jumping on his bumper....
This 5.1 is a good thing, as it has relieved geo-stress... (I hope!)
but as a california native, I think of earthquakes like I do california poppies (which I just planted a bunch of) they are magnificent, just dont pick any...
EDIT: The 1989 earthquake was also in october.... October 11 I think... but I wonder if the seasonal shift in temperatures affect earthquakes. As October is generally the best month in the bay area weather-wise...
(No, not looking forward to another New Madrid quake.)
This one was 4 miles deep. That would be surprising.
That earthquake shook me deeply. It's not a coincidence that I now live on a whole different tectonic plate.
For those not in earthquake country, this was a moderate quake, not big enough to cause damage other than maybe knocking something over right near the center. The epicenter was in Joseph Grant Park to the east of San Jose, near Mount Hamilton.
So, I guess it traveled along the East Bay fault lines all the way to Alameda and Berkeley. I am just surprised it was felt so far for a 5.1
the modeling for an aftershock in the coming week predicts:
64% chance of a 3+
16% chance of a 4+
2% chance of another 5+
and today's quake being a foreshock is also possible, with the model's predictions being
0.2% (1 in 500) chance of a 6+
0.02% (1 in 5000) chance of a 7+
The GSM lineage (including 4G, LTE, 5G) is implemented using cell broadcast which is a protocol data unit that solves for the one to many problem. Extensions added geo-fencing & geo-targeting on the client devices (device compares its idea of location to the incoming message), and also cell based routing optimizations network side. The cell broadcast part is technologically available ~everywhere as a result of broad GSM deployment, but whether it is integrated into an alert system is a jurisdiction issue. Sadly few countries (in the global scope) have the kind of power of the US or the EU to gain access to things like extended on device features.
The app based systems are largely based on some form of location reporting and push messages which are far less reliable - one to one messaging systems (like Apple/Google push notifications) are physically impossible to scale for real time sub-second alerting in dense populations, given the bandwidth constraints of radio.
CB is expensive at the radio tower, and can disrupt normal communications (potentially including emergency services communications) as such it's use is more limited than other solutions, for example in todays event it was only used in an area of about a 25km radius. To put this in some context, many folks commenting here, including myself still in the "felt it" areas were within a ~90km radius (SF is about 85km away from the center).
I wasn’t in the area, but I believe WEA would have been sent in the inner ~20km based on the system report, which is sjc and the surrounding area just up to the southern tip of the bay.
Today is actually the 33 year anniversary of the Loma Prieta quake -- weird.
There's a bit more information available here: https://news.ycombinator.com/newsfaq.html
Or are you asking: "how does this earthquake in the place where much tech is centered and a large number of HN users live pertain to HN?"
The answer turns out to be the same: it's relevant to a lot of people here and they upvoted it.
Or at-least depend on companies located here.
You want to worry about something that's actually broken, then worry about whether the US MNT will get it together to call up leading goalscorer Jordan Pefok by 13 Nov deadline for the World Cup. Now that's worth losing sleep over. Or worry about Cuba, Haiti, Puerto Rico reconstruction from Hurricane Ian; Cuba got it bad.
USGS.gov has a page up near-instantaneously with the details including crowdsourced reports of perceived intensity by location, as they always do: https://earthquake.usgs.gov/earthquakes/eventpage/nc73799091...
PPS: in that area, Mt Hamilton (4265') and Lick Mill Observatory are worth a visit. And over west in Menlo Park visitors can visit the USGS station for free, that's a very fun and educational thing to do.
Today's non-event was actually a Good Thing in that it also gives us yet another occasion to calibrate earthquake early-warning sensors, to stay prepared for any actual serious earthquake, which this wasn't.
I also said nothing whatsoever about "excitement" or "disappointed", so do not misrepresent me. I did imply that media is increasingly sensational (which it is) and lazily skewed towards merely collating cellphone "eyewitness reports" from places with 4G coverage, without making any semi-objective comparison to the rest of the planet; and I correctly pointed out that zero people were affected here in SJ today whereas 11 million Cubans were affected by Hurricane Ian [0], to pick one obvious example. I also know some friends-of-friends whose house was severely damaged by the 11/2020 Marikina, Manila floods/Typhoon Vamco/Ulysses [1] (98 deaths, 19 missing, 360K rescued, ~43000 houses damaged).
Also, a magnitude 6.2-6.5 struck rural northern Luzon, Philippines earlier today [2], no reported damage or injuries [3], but noone's mentioning that.
[0]: https://www.state.gov/u-s-support-for-hurricane-ian-recovery...
[1]: https://en.wikipedia.org/wiki/Typhoon_Vamco#Philippines_2
[2]: https://earthquake.usgs.gov/earthquakes/eventpage/us6000iwds...
[3]: https://newsinfo.inquirer.net/1684866/magnitude-6-7-rocks-ab...
“Big yawn” is a loaded word: It was not in the least terrifying and actually a big yawn; no known damage or injuries; …
Dismissing earthquakes: nothing like all those San Andreas movies. …
Minimizing risk: in case there's a big aftershock, which there almost never is.
Comparing earthquakes to sporting events: You want to worry about something that's actually broken, then worry about whether the US MNT will get it together to call up leading goalscorer Jordan Pefok by 13 Nov deadline for the World Cup
Experts believe a major quake in the Bay Area is “likely” (>=6.7) by 2032 per the usgs
The topic here was never "the next big one" or "major future event on the Hayward and San Andreas Faults for the next 30 years". If you want to start a separate HN thread discussing either of those, you're free to. This ain't it.
I actually dug up the USGS report to cite the epicenter, depth, intensity to teach people here basic literacy to make them immune from sensationalist reports without context. I showed what zipcode was affected and correctly stated how its population is almost zero and has no tall structures. Hence, "big yawn" is ̶n̶o̶t̶ ̶l̶o̶a̶d̶e̶d̶, it's a 120% accurate characterization of an earthquake miles up the foothills inside a county park. I'm not going to misrepresent those facts. I can't see why anyone else would.
̶M̶i̶n̶i̶m̶i̶z̶i̶n̶g̶ ̶r̶i̶s̶k̶ Correctly stating that the risk of aftershock on today's event was minimal, and even in the tiny likelihood there had been one, given the original 5.1 caused zero reported damage, a (say) 4.6 aftershock would likely cause... zero damage. Zipcode 95140 is not downtown SJ or SF or LA.
To do otherwise would be to distort the facts on today's non-event into some emotionally-driven narrative that minimizes other actually serious events which actually resulted in deaths, injuries, destroyed houses, power outages; why should we distort the facts like that? It's actively harmful. Sure, people can be needlessly frightened by sensationalist media reports, while simultaneously ignoring other natural disasters which flattened countries. Surely that's a trend we should fight against, yes; not contribute to?
> Experts believe a major quake in the Bay Area is “likely” (>=6.7) by 2032 per the USGS
That was never the topic of this thread, also many of us who already all knew that a decade ago. So why try to hijack this thread with that? By all means start a separate thread to discuss it, and I'll contribute constructively over there.
̶C̶o̶m̶p̶a̶r̶i̶n̶g̶ ̶e̶a̶r̶t̶h̶q̶u̶a̶k̶e̶s̶ ̶t̶o̶ ̶s̶p̶o̶r̶t̶i̶n̶g̶ ̶e̶v̶e̶n̶t̶s̶ Comparing distorted sensationalism around non-events, to other contemporary events, including other natural disasters: "Or worry about Cuba, Haiti, Puerto Rico reconstruction from Hurricane Ian; Cuba got it bad." Even failing to accurately identify the epicenter, intensity and depth (facts which are typically available within a few minutes on USGS.gov) actively pander to that sensationalism; I posted them; you haven't discussed them once.
To summarize: today's non-event in 95140 caused zero deaths, zero injuries and zero damage; compared to others that actually impacted people and continue to over a month later, like Hurricane Ian. The earthquake early-warning systems worked as intended and that experience makes us stronger for the future. (By comparison, PG&E-started fires and electricity outages have caused far more disruption this year than earthquakes (:zero), and again that's a major political issue currently in California, not this.) Almost any other minor media story is more worthy than this one, this in particular on a day with multiple huge stories (foreign and domestic) worthy of coverage. Pick any other story worth being covered more, to replace the US soccer one.
We’ll this is terminologically correct due to that fact that if a stronger one follows then the initial one is a foreshock and the stronger one they prepared for is not an aftershock but the main event.
Anyway, I gave unsolicited feedback. Reject it or embrace it, no matter.
> You want to worry about something that's actually broken, then worry about whether the US MNT... _Or worry about Cuba, Haiti, Puerto Rico reconstruction from Hurricane Ian; Cuba got it bad._
So, we already established within a handful of minutes that zero Californians were affected, but 12 million Cubans, or many thousands of Haitians or Puerto Ricans. Yet we're wasting time debating the California non-event instead of any other thing that actually matters. Or even, at absolute minimum, teaching people the technial literacy to quickly understand that the CA earthquake had no impact, and thus to resist sensationalism. It was extremely clear from the original that that was my point.
To state the same point another way: even (say) tonight's Fetterman-Oz debate (PA Senate race) will have more permanent impact on people's lives. Or, Hurricane Ian. Like I said.