Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

So since I live near epicenter, there's one thing that changed geographically very recently. Nearby round valley reservoir was being worked on for years, last fall they got the authorization to pump it back to full and it was told it might take years. Rainy asf winter though so it's pumped back almost full from already 66% to 92% now. I'm wondering if billions of gallons of water getting added nearby affected pressure on the fault. Here's today's operations data (we just had 3.5in of rain past couple days too): https://www.njwsa.org/uploads/1/0/8/0/108064771/data.pdf

EDIT: just had another aftershock right after posting this. Felt the 2.0 a couple hours ago too

EDIT2: yup usgs finally added it https://earthquake.usgs.gov/earthquakes/eventpage/us7000ma95...



It's possible. Seismicity has been associated with reservoirs[1] (the most well known is probably an earthquake in Egypt following the construction of the Aswan High Dam) and even rainfall[2]. There are both potential triggers from the increase in the stress on the fault, as you state, due to the weight of the water; and (probably more important) the increase in pore fluid pressure along the fault.

It's a bit complicated, because the increase in stress on the fault from the weight of the reservoir can actually be primarily an increase in normal stress, i.e. an increase in the clamping force preventing the fault from slipping (i.e. the crust on either side of the fault sliding, i.e. the earthquake), so the weight of the reservoir can actually impede, rather than promote, seismicity. However if the geometry of the system is just right such that the increase in shear stress from the weight of the reservoir can increase on the fault, then the reservoir could really trigger the earthquake. Typically, though, the increase in pore fluid pressure is the most likely trigger as it's not geometry dependent--pumping a fracture with fluids will always promote failure. This is what happens with the seismicity associated with natural gas fracking and subsequent wastewater fluid injection.

The thing that is hard to grapple with all of this is that changes in stress and fluid pressure on the fault are really, really small compared with the ambient stress and fluid pressure. These are tiny changes, but there is good reason to believe that many faults, especially those in seismically active areas, are very close to frictional failure (i.e., slipping and producing an earthquake) all the time. So therefore little things can perturb the system. However, it's worth keeping in mind that most of these small changes slightly increase or decrease the time to failure (as the primary loading that is causing the stress on the fault is probably continually happening at very low rates), rather than being the ultimate cause of the earthquake.

[1]: https://scholar.google.com/scholar?hl=en&as_sdt=0%2C48&q=dam...

[2]: https://agupubs.onlinelibrary.wiley.com/doi/full/10.1029/200...


Good shit, this is the kind of answer I hoped for posting on HN. I don't know too much about earthquake seismology, but I am curious and know where to pull random data.

So I thought, if the ground is moving the aquifer levels might change (no clue whether they would go up or down). I just pulled the ground water levels for Readington Township's well (in which whitehouse station is a place). Sure enough can clearly see when the earthquake hit: https://waterdata.usgs.gov/monitoring-location/4035170744525...

So is this expected? Is this a thing that normally happens with earthquakes?

EDIT: Japan paper was cool https://sci-hub.ru/https://www.sciencedirect.com/science/art... apparently they measure increased seismicity at times of high inflow. Also just had a big aftershock approx 22:00UTC. sounded like a bomb (I am in whitehouse station now)


30% is 16.5 Billion gallons.

OP is saying that much rain got into a reservoir that sits on top of an ancient volcanic caldera and has no rivers flowing into? Unexpectedly? and they didn’t think to stop pumping water in from the South Branch or release some water?

The math ain’t mathing.


It's not at full capacity yet, they are pumping into it daily (around 200 million gallons) until it is. Only spruce run reservoir upstream of it fills naturally (and is spilling currently).

Operations data for the reservoir updates every morning during workdays, so we'll know Monday if they paused pumping into it after the quakes, I don't have any better than daily numbers for it. Letting the water down isn't an option, most of the downstream gauges on the south branch were just at or near flood stage. River gauges available here: https://water.weather.gov/ahps2/index.php?wfo=phi


Just to follow up on this, there was another way to tell they maybe shut the pumps off for a few hours after the quake (and I linked it in the same comment). The downstream gauge: https://water.weather.gov/ahps2/hydrograph.php?wfo=phi&gage=...


Does anyone know if there's a real-time feed of Earthquakes somewhere? The USGS website doesn't post the Earthquakes until 10-15 minutes after it's over - which nullifies any type of automated warning for our data centers.


There is in Japan. I don't know how the realtime data feed is set up, but here is one public webpage that shows the degree of shaking as it happens:

https://typhoon.yahoo.co.jp/weather/jp/earthquake/kyoshin/

I felt a moderate earthquake here in Yokohama a few days ago. I had my phone with me, so I clicked my bookmark for that page and, before the shaking stopped, could see that it was a magnitude 6 with epicenter offshore from Fukushima--nothing to worry about. A minute or so later the permanent record of the quake was online:

https://typhoon.yahoo.co.jp/weather/jp/earthquake/2024040412...


Curious: Why might data centers need earthquake warnings, and how might they prepare if given a few minutes or seconds of heads up?


I can think of some ideas: parking the heads on hard drives is one.

Magnetic hard drives are sensitive to vibration. You can shout at hard drives and measure the effects (video: https://www.youtube.com/watch?v=tDacjrSCeq4).

One of the worst-case scenarios is a head crash. A head crash will damage the media and may result in data loss. https://en.wikipedia.org/wiki/Head_crash

My guess is that earthquakes powerful enough to cause a head crash are powerful enough for widespread destruction anyway, but I’m no expert. I did some quick searches for hard drives damaged by earthquake, and the only results I got were scenarios where the hard drives or the whole rack got knocked over by the earthquake and hit the floor.


Some other thoughts:

- Personnel-level warning to immediately rerack servers, close racks, and get off ladders and away from fall hazards

- Proactively spin up generators to reduce failover in the more-likely event of a power disruption

- Potentially temporarily shut off very large circulation fans so that blades don't collide with the housings

- Potentially stop and carefully restart cooling water loops, in case there's a rupture in the system somewhere


Good list.

Add to this: initiate failover or zone transfer of distributed servers / services to other DCs outside the likely impacted area.


Personally, I’d rather have the generators off when the earthquake hits!


Need realtime seismographs, my friend's dad has one he built as a hobby up in quebec city that picked it up: https://alainmichaud.ca/


Another great site with a ton of citizen stations (some are placed next to USGS seismographs too): https://shakenet.raspberryshake.org




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: