How long should we wait for an overdue earthquake on the San Andreas? | Hacker News Reader

How long should we wait for an overdue earthquake on the San Andreas?

rocksandwater.net

72 pointscossatot90 comments

I lived in SoCal and worked in SFO for some decades and witnessed both the Northridge and the Oakland quakes first-hand, narrowly avoiding personal injury or even death in both cases (big things falling around me), and the conclusion I made after these events, which I eventually acted on, was simply put thus: GTFO!

I really didn't want to be there when the Big One happened, nor even some of the minor small ones, so I've moved to Central Europe where such things have a lower chance of happening .. in a way, I have a perverse desire to see The Big One happen from so far away, but I always dread the mind-game of trying to conceive what SoCal will be like, afterwards. I truly dread the idea, and I honestly can't understand how any society with scientific awareness of its surroundings can nevertheless commit to building so much expensive civilization and infrastructure in a known catastrophe zone. I guess its just a matter of momentum, but I sure wish the citizenry of the PNW and SoCal would prepare themselves for this event a bit better. It seems so defeatist to continue to build things in that area, knowing it could fall into the sea in a moment, and there is nothing we could do about it. California should just be a massive nature park stripped of all human involvement, just in case, but of course this is never going to happen, so if The Big One does happen in my lifetime, its just going to be a massive tragedy no matter what we humans could have done about it with all our science and stuff. Argh!

I guess if I lived through two significant earthquakes in that way, I'd probably have had a thought along the lines of "This is a sign to move" as well.

>It seems so defeatist to continue to build things in that area

Fatalistic is probably the better word. I can buy an old missile depot in the Midwest and stock it up with supplies and prepare myself for the zombie apocalypse or whatever.

Or I can live somewhere I enjoy, and has good opportunities, etc. and not worry too much about events I don't have a lot of control over or a lot of insight into their probabilities.

Of course, there are situations between those two extremes and it's not unreasonable IMO to include "the Big One might hit" into a calculus for evaluating moving to or staying on the West coast. (As with living in areas with a lot of potential for floods, hurricanes, and tornadoes.)

Actually turns out Las Vegas, for example, scores pretty well on non-disaster metrics. One reason a lot of datacenters are there--in addition to evaporative cooling.

Its not like I didn't have a great time living in California, because I truly did, and in many ways its a delightful civilization in which to participate.

Perhaps its just a cultural comfort I seek, in which the citizenry at least seem aware and prepared for total and utter disaster, any minute now, and live accordingly. I guess its exhausting to live that way, so if it happens, it happens; maybe that liberal and - lets just get out and say it, "Zen-"like mindset truly is just a part of the California love ..

Honestly, though, I'm not sure anyplace in the US is different. I lived in New Orleans and I was well aware of the hurricane scenarios as well as some of the failure scenarios of Corps of Engineers structures up the Mississippi. (Had a good friend who worked for the Corps.)

So it seems a pretty typical attitude in the US. Perhaps it's different elsewhere. Abstractly understand the worst case scenarios, plan to mitigate them (or not), but go on living life.

>> I honestly can't understand how any society with scientific awareness of its surroundings can nevertheless commit to building so much expensive civilization and infrastructure in a known catastrophe zone.

The same thing can be said about many other places on earth where people decided to settle and build a society. Near volcano's, below sea level, between rivers that could flood the whole area, in arid regions that barely support agriculture, very high up in the mountains where it's cold and dangerous, in regions infested by parasites and disease-carrying insects, etc.

I think you should not try to approach these things from a rational/calculating perspective, that's not how societies work. Why societies cluster at certain locations is determined by so many different factors that building a highly advanced, densily populated, and expensive to (re-) build society close to a faultline should IMO be seen as a 'historical accident that, in the case of SF, had a high probability of happening'.

People often assume the 1989 earthquake was particularly dangerous. While it caused quite a bit of property damage (about $6 billion according to Wikipedia), there were only 63 deaths and 3,757 injuries. The previous major earthquake in the region was in 1906, and there haven't been any since 1989.

Given that, it is extremely unlikely to die or be injured by earthquakes in the Bay Area. On a list of most likely ways to die, it would be near the bottom.

edit: I should mention I lived in the Bay Area in 1989 and experienced it first hand. I still live here and of all the reasons I might move, earthquakes isnt even on the list.

Question: what happens to land rights after a quake? Since the quake moves land around, stretching in parts and destroying land into cracks in others. Do property lines get redrawn? Who owns any newly created land space? Is it fair game for anyone to claim?

It doest really work like that, it's not like new land is created by pushing out into the sea, but even if it was the property owners boundary is likely the sea, therefore the land is incorporated within their area.

But what if you have a fence that gets shifted around, and no longer matches the boundaries in the property maps?

Say there's a crack or fissure that opens up, creating new "space". Someone then fills that crack with soil making new "land". Or say the land under a fence moves. Since land ownership diagrams contain specific measurements, are they redrawn? Replaced? Accepted? Measurements ignored and features used instead (apple tree to 7th fence post)?

If my property deed says my yard is 20 metres wide, but post earthquake it's 21 because a 1 metre wide split occured in it what happens when I sell? Is accuracy thrown out the window?

My best guess would be that the state would claim original ownership of the new land. If you take a small coastal lot and then start building landfill out into the ocean you are going to have a hard time convincing the state that you own the property without some pre-existing contract with the appropriate regulatory bodies stating this when you started construction.

It very much does work like that though... The San Andreas fault is a "strike-slip" fault. The plates are (effectively) moving past each other, in opposite directions.

Imagine you owned a rectangular piece of land, with the fault running right up through the middle of it. After an earthquake, the western half of your property is now shifted a non-trivial distance to the north relative to the eastern half. What happens to the property line? Does it now have a 'kink' in it? Does it remain in the same place relative to the 'ideal' geodesic plane (meaning various landmark like fenceposts, etc, are now in a different place relative to the property line).

It's certainly an interesting question...

You have ownership of a piece of land, not a spot on a map. If the land moves then you still own the land even if it is non-contiguous. I am not sure what happens in the case where some event creates new land adjacent to yours such as the case where a lava flow extends land out past yours into what was once water or if an uplift fault raises seabed to become contiguous with yours, but I would assume that the state would claim ownership over the new land.

What happens to "newly created land" depends on local legislation. Where I live (Finland) this is a common question, because it happens all the time. We're still recovering from the depression of ground, caused by latest Ice Age. In Western Finland and Eastern Sweden, the land rises at about 1 cm per year, thus claiming area from the sea.

1 cm per year may not sound like a lot, but in shallow waters, it means that entire coastal towns become inland towns over the course of 100 years. Harbours have to be relocated, etc. Kilometres of new land is claimed from the sea over the course of centuries (and within a couple of millenia, the Gulf of Bothnia will dry up in places and create a land connection across, assuming water level rise caused by global climate change does not slow it down too much).

Over here, the law works so that the newly created land at the shore becomes property of the owner of the water area (fishing) rights which is a "commonwealth" of local land owners, but the owner of the particular piece of land formerly adjacent to the sea or lake has the pre-emptive right to purchase the area and attach it to his/her property, at a "fair" price i.e. price comparable to the price of his/her own property. So in effect you have to remember to buy regularly (every few decades) the new land between your plot and the water.

But that was Finland. How does this work regarding land created or lost in earthquakes in California? Let's assume that the fault line of tectonic plates happens to run through your plot, the plates push at each other and then in an earthquake run over each other, and the corner points of your plot move closer to each other. Is that just your loss? I.e. your plot became smaller? And likewise, if the borders of your plot move away from each other in an earthquake, your property expands, possibly impacting e.g. how large buildings can be constructed according to zoning regulations?

>You have ownership of a piece of land, not a spot on a map.

That's sort of silly. How do you define a 'piece of land'? What if I dig out my piece of land so it's a 100ft hole and then there is a sheering motion from the plates so half of my hole slides under someone else's land? Do I still own both halves? If so, do I no longer own the land at the bottom of the top half? If not, then it would not seem that I do not own a 'piece of land'.

Land ownership is not really absolute; it is always subject to the definitions of the local law and, ultimately, the whims of the government.

You don't have complete power over what you do to your land, because the government has the last word. On the other hand, without government, there is no land ownership because the definition and enforcement of land ownership is one of the most ultimate functions of any government, since the creation of first city-states (e.g. ancient Babylon).

So if the government permits you to dig that hole in the first place, then what happens in an earthquake depends on local legislation. I don't really know how it works e.g. in the U.S. but I would assume that the boundary of your piece of land is defined in relation to physical markers that are recorded with their geocoordinates, but if the lay of the land changes so that the markers move, then the boundary of your plot will also move. And this could mean that the size of your plot changes.

Where I live, for instance, there are old boundary stones (which would in an earthquake naturally move along with the land on which they are laid) and newer marker bolts attached to bedrock that serve as reference points for geodesic measurement purposes. Also, there are height markers; for instance, when I started to build my house, the city would send a guy with equipment and start from one of these markers and then tell me if the foundations of the house are at right elevation. Should there be a major earthquake (not really a possibility here, ours is just about the most stable bedrock on Earth) then the changed positions (coordinates) would have to be recorded in land registry.

You may find learning more about the foundations of common-law to be interesting (since it seems you are from a Civil Law perspective).

http://faculty.msb.edu/hasnasj/GTWebSite/TortWeb.html

http://papers.ssrn.com/sol3/papers.cfm?abstract_id=586941

Thanks, these were interesting reads, although I would still appreciate a hint how common law specifically would handle emergence of new land differently from civil law systems?

A piece of land is typically defined in terms of geographic coordinates, though. Which do not move when the physical solids constituting your land do.

But not always - my parents own land in Australia and one of their borders is defined as meeting a river (since the river is publicly-owned so anyone can get water from it). When the river changes, they gain or lose land.

I imagine if you have boundary markers, such as a fence line, that would remain your boundary line. Without a physically marked boundary line then you would be stuck with the shape as shown on maps.

Trade an easement for $1 with the neighbors to return the parcel to a usable shape.

Question: what happens to your IP when it ends up on a torrent and the world copies it all over?

Human institutions do not win against nature/reality. Not that we shouldn't try, but we have to remember that our institutions are limited ultimately by the laws of physics and mathematics/statistics.

You should ask the residents of Hollister.

http://www.bldgblog.com/2015/04/the-town-that-creep-built/

California has the Cullen Earthquake Act to deal with that.[1] That's a consolidation law, so that all the claims in an affected area can be resolved in one court proceeding.

[1] http://www.leginfo.ca.gov/cgi-bin/displaycode?section=ccp&gr...

I was under the impression that the bigger danger was the expected eventual earthquake of the Cascadia Subduction Zone, potentially leading to tens of thousands killed or injured and millions more displaced.[1][2]

[1] http://www.newyorker.com/magazine/2015/07/20/the-really-big-...

[2] https://en.wikipedia.org/wiki/Cascadia_subduction_zone#Earth...

The damage from a Cascadia rupture is going to be more widespread but the ground shaking in the metro areas may be less than for the San Andreas because the San Andreas hits the surface quite close to populated areas (LA also has more people that WA and OR combined). The intensity of ground shaking decreases with distance due to geometric spreading and energy absorption by the intervening rock mass. Cascading ruptures are also every 500 years (or so) vs every 100 years, so the hazard per year is less. But yes, both are major concerns.

Also, speaking of other faults, WA also has the smaller, but much more shallow, Seattle fault. Because it's so much closer to the surface, and right under the city, it can potentially do more localized damage than Cascadia.

(There are other dangerous faults like that in Puget Sound region, but Seattle fault gets special attention because of how many densely populated Seattle metro areas sit literally right on top of it.)

There's a ton of data on the very regularly occurring (hourly) small earthquakes around the world available from USGS: http://earthquake.usgs.gov/earthquakes/ I'd like to see whether this analysis works on this rich data, not on the very sparse data set of San Andreas earthquakes.

Some fault lines seem to accumulate energy, so the longer the waiting period between two earthquakes, the stronger it gets. Is it the same for San Andreas, so that more frequent earthquakes would actually be beneficial as they would be easier to handle structurally?

I’m not an expert and I have the same understanding. Futuristic earthquake mitigation approaches talk about inducing small, non-destructive earthquakes in a controlled way to reduce the magnitude. The goal is to completely manage earthquakes.

That's insanely cool!

Oh sure... now you want to frack. ;-)

Maybe Oklahoma and Arkansas are the best places to go to avoid "The Big One"?

Not as good an idea as it sounds. The problem is that the largest earthquakes release massive amounts of energy. Each point on the Richter scale corresponds to about a factor of 30 in energy. So to release the energy of one magnitude-7 quake would take some 900 magnitude-5 quakes. And a 5 in a populated area can still do a little damage. Can you imagine how people would feel about having one of these a week for 20 years?

Why not 810,000 Magnitude 3 quakes?

On the other hand, a daily (smallish) earthquake could easily replace your alarm clock.

"Sorry I'm late... I slept through the tremor this morning."

Time your sex well and you can have (or fake) an earth-shaking orgasm.

What I can't understand is that after the 1906 San Francisco earthquake, where "Over 80% of the city was destroyed by the earthquake and fire" ( https://en.wikipedia.org/wiki/1906_San_Francisco_earthquake ), why on earth was it rebuilt? Surely that's a sign to move the new city elsewhere?

You're only looking at the downside risk. There are huge upsides to the area - notably the San Francisco Bay is a natural harbor, which makes it a really good place to put a port. The promise of wealth from trade meant people would live them pretty much regardless of the downsides.

Fine, but what's likely to be cheaper, building an artificial harbour away from the fault line or rebuilding a damaged city every 100 years or so?

The fault line is exactly on the coast and 800 miles long. Rebuilding affected cities every 100 years or so is cheaper than re-routing all ship traffic affecting a quarter of the US.

You'd have to go a long way from the San Francisco Bay to be a safe distance away from the fault line, which would have a dramatic impact on the costs of operating the port - it'd add hundreds of miles to the distance you'd need to move the goods off the ships to where they were needed. Factoring that additional cost in over the decades between earthquakes (SF burned down quite regularly prior to 1906), I wouldn't be surprised if it was actually cheaper to rebuild the city occasionally.

> "You'd have to go a long way from the San Francisco Bay to be a safe distance away from the fault line"

Is there an equation that's useful in working out how the strength of an earthquake changes as you move further away from a fault line? I tried to find one, but haven't had any luck yet.

> What's cheaper, rebuilding a damaged city every 100 years or so?

I wonder whether this is actually expensive, compared to the job gravitational opportunity. Surely landlords of the skyscrapers downtown have included the risk in their financial plans. Home owners will support the greatest cost, but once again, they would have much less if they didn't live in SF. Health costs may already be included in your insurances. The federal government may help a bit. The rest of the people may go bankrupt, but there are many homeless people who were former researchers already, so it's renewing the workforce (the wrong way). It's extremely cynical, but city planners aren't the most compelled to act.

I personally wouldn't go live in Vancouver for that reason, I don't go to conferences in USA because of the TSA, and I'll never do tourism in Japan because their conviction rate is 98%. One needs to include the risks in their life calculations.

And the plane you take could crash. Or someone could run into your car. Or you could choke on that piece of chicken.

I dunno. I'm unlikely to take a tourist trip to certain countries these days or to do certain sports activities that for me the risks exceed the rewards.

But I'm also not going through life worrying about what might happen in, say, Japan of all places. But maybe we're saying the same thing and my risk calculus is just way different. I have done things like fairly high altitude climbing for example.

I'm not afraid of visiting the SF bay area, because I know my risk for a few days there is really really low. But I refuse to live there, because the total risk over the remainder of my life is higher than I'm comfortable with.

You don't have to rebuild the city every 100 years. From what I recall, most of the damage to the city in the 1906 earthquake was due to the fires which couldn't be controlled, which was because the earthquake damaged the water lines too badly and they didn't have any working backups. But not only do we have better infrastructure now, we also have multiple auxiliary water systems put in place specifically for emergencies. If a big earthquake hit tomorrow there certainly would be a lot of damage, but there would not be widespread fire and the city would not be completely destroyed.

Fun fact: One of those auxiliary systems is 177 cisterns containing between 75,000 and 200,000 gallons of water buried all over the city. If you've ever noticed red brick circles embedded in streets, those mark the cisterns. They're maintained specifically for dealing with fires.

> why on earth was it rebuilt?

They probably thought: Great, now we have 100 years until the next one. ;)

Do what makes economic sense today. Then it will be someone else's problem.

Thankfully, we never make decisions like that with respect to software :-)

And Tokyo in 1923. But really the entire Rim of Fire.

We build things in areas that are relatively dangerous geologically because there are other advantages to those areas. San Francisco has one of the best harbors on the West Coast and access to a lot of inland resources.

But just another reason not to concentrate a good chunk of the US software industry in the Bay area I guess.

Fun fact: why did they decide to build a huge nuclear power station close to the fault, and 6m above sea level so that a tsunami will have Fukushima like effects?

That's what will bring Silicon Valley down, not direct quake causes.

What does an historic earthquake look like in the layers of rock? How do they know when they occurred?

The science of finding out the (pre-)historic earthquake history is called 'paleoseismology'. You can read about it from a variety of sources.

In most instances the earthquakes are dated by finding carbon-bearing material in deposits that look like they were created during the earthquake, and doing radiocarbon dating of them. The ages may be bracketed by radiocarbon dating of sediments that were clearly broken by the earthquake (and are therefore older) and ones that overlie the broken rocks (and are therefore younger).

In a few other instances, earthquakes can be dated by tree-ring dating; for example during big subduction zone earthquakes, the coast often subsides by a meter or two, which will put some forested beaches under sea level (or at least high tide level) and kill all of the trees that can't deal with salt water.

In others, the dates of sediment layers broken or caused by earthquakes can be dated through other means, or correlated with climate events of known time.

Some example of deformed and undeformed sediments in the Wrightwood trench site on the San Andreas is here[0,1]. [2] is from the nearby San Jacinto fault. [3] has a lot of images. [4] is a good introduction to the science with good pictures.

[0]: http://www.bssaonline.org/content/97/4/1054/F14.large.jpg [1]: http://www.bssaonline.org/content/92/7/2726/F11.large.jpg [2]: http://lithosphere.gsapubs.org/content/7/3/211/F6.large.jpg [3]: http://earthquake.usgs.gov/research/external/reports/04HQGR0... [4]: http://isites.harvard.edu/fs/docs/icb.topic988563.files/EPS2...

So, we aren't only overdue but a quake seems imminent? The upper bound of time between quakes was very slim.

This is a stupid and selfish question, but what does a quake on the San Andres fault mean for San Francisco? My knowledge of all of this is based on 4th grade science.

Bad news. Whilst building codes may be better than in the past, due to population density the risks of a high number of deaths is significant.

I would imagine a very significant portion if the Marina district just turning into rubble unfortunately. I'm hoping this doesn't happen until buildings have a chance to improve even more.

No chance for improvement. New buildings follow better building codes, such as the tall ones that may sway like Japanese skyscrapers, old buildings have no mandate to improve, and San Francisco was mostly built after the 1906 earthquake, with many of the walkups that litter the view being in the 1920s - 1940s.

From my understanding, any hope you might have is neglecting a reality that an overdue earthquake will cause devastation largely reserved for the third world.

San Francisco can't even mandate building codes on sidewalks, which involves the most marginalized and disenfranchised inhabitants, you really think they could come to a conclusion that involves landlords?

This is fear mongering. The tectonic structure of the SA fault means that the upper bound for magnitude is limited, and building codes are really good.

Up in Portland and Seattle, the risk for high numbers of causalties is far greater due to the subduction fault rather than the transverse fault.

Perhaps, but I have to say the view of San Francisco from Alcatraz makes it look like one good shake could send the whole thing tumbling into the bay.

I'm not a seismologist, but the city really doesn't look like it should be built the way it is in an area prone to earthquakes.

Note that this article is specifically about the south end of the San Andreas near LA.

Hopefully San Andreas will wait another 10 years, because by that time there will be no Nuclear Reactors operating in California. We are down to only ONE, and it will shut down within the next 6 years. PG&E is getting out of Nukes.

If it doesn't happen this year, just wait longer.

winning by induction

Subduction actually.

People always say Hayward Fault is more troubling for California because so much is built directly ON the fault.

Possibly a silly question, but...

I hear the term 'overdue' applied particularly to earthquakes, rather than any other naturally occurring event. Does this mean 'overdue' in the same sense that rolling a six is 'overdue' if you've not seen one after six rolls?

No, It means that the tension between tectonic plates steadily builds up, so an earthquake is more likely.

I see - the article though seems to talk about 'surface-breaking' earthquakes occurring every 100 years in that region, and that we're now at 159 years without one.

This idea of a time expectation doesn't seem to match your explanation (which seems more intuitive to me).

They're the same, really -- those quakes happen every 100 years because that's how long it takes for that much pressure to build up. 159 years without one means there's already going to be quite a lot of pressure built up that's going to have to give way sooner or later.

We know that fluid injections for fracking can cause earthquakes.

How about injecting some extremely viscous fluid deep underground within the plate boundaries?

So it can release the shear force over months or years rather than seconds. Something like pitch (https://en.wikipedia.org/wiki/Pitch_drop_experiment)

Of course, the major problem with trying quake mitigation strategies might be the legal responsibility if you cause a quake.

Aside from the very real liability issue, the idea of lubricating faults is interesting. From a practical standpoint it would be hard to inject something that remains viscous enough at the relevant temperatures. Earthquakes nucleate at about 10 km depth in the San Andreas, which is about 350 degrees C. So aside from the difficulty of drilling in these conditions, you would have to have a fluid that maintains its viscosity from surface temperatures to 350 C.

Given fracking injections are a lot shallower than 10km, I'm not sure it would have to be injected that deep. But it's a good point, that temperature will be a factor.

Getting a high-melting-point fluid down there might be done by dissolving or suspending it in water that is allowed to boil off, or having a heated injection pipeline.

Silicone and silicate lubricants are viable at those temperatures. Might not be viscous enough, though.

A low melting phosphate glass might work.

My understanding is that the probability and expected magnitude of an earthquake actually increases over time as tectonic tension builds. So no, not like dice.

I meant in the sense of probability, not literally like dice :)

Well yes, pretty much any naturally occurring process involves some probability...

It essentially comes down to whether you assume that the probability of an earthquake every subsequent year is independent of the result in previous years. The notion of tectonic pressure build-up suggests that that assumption isn't true.

The author has a reasonable idea in applying survival analysis to estimate the probability without making any assumptions about the underlying model.

I think it was clear what you meant.

If you roll a dice 10 times and don't get a six, the probability of rolling a six is still 1/6 for the next roll. However, as the time since the last earthquake increases, not only does the probability of one increase (unlike dice which has a constant probability) but also the likely magnitude of it increases as well.

The equivalent with dice would be you roll a non-six 10 times in a row, and instead of 1/6 probability of rolling a six, you now have a 1/3 probability of rolling a six and a 1/100 probability of setting off a nuclear explosion.

Thanks for the helpful response. I still don't think I got quite the question across that I meant though.

It wasn't so much the probability model (independent or otherwise) I intended to question (hence my very poor choice of the die rolling analogy), but the choice of describing an event as 'overdue'.

At the risk of introducing another down-voted analogy, presumably hurricanes have a probability of occurring based on things like ocean temperature, time of year etc., but I don't think I've ever heard the phrase 'the east coast of the US is overdue a hurricane'.

The term 'overdue' is used because, as earlier posters explained, the probability of an earthquake actually increases the longer that it doesn't happen. But it comes to hurricanes, I suppose the probability of them occurring does not increase if they don't happen (or we just don't know), and that's why we (generally) don't use the term overdue.

Is there a way to milk tectonic tension?

Well yes, you can with more earthquakes. Theoretically assuming the build up of tension energy is constant, you could trade many smaller earthquakes for a big one. Ex 30 mag 6's instead of a mag 7 to get the same amount of energy released. Whether that is a good idea or not is debateble.

Practically we have no control over earthquakes (except maybe with fracking) and no way to really forecast them.

It does seem like tectonic pressure could be a great source of clean, renewable energy! :-) Maybe some day in the future there will be no earthquakes because we'll have a way of pulling out the energy gradually and putting it to use.

Another idea I wish would work is getting energy from lightning strikes. I guess it's not practical, but if it could work, "powered by lightning" sure would sound awesome. :-)

It's like knowing that 1/2 dice rolls yields an even number, and after rolling twice in a row and getting odds making a big deal of the supposed statistical anomaly, because we got two odd numbers in a row we're really overdue for an even number, don't you think?

When volcanos/earthquakes etc. are said to occur every N years, that's averaged over geological periods. An event that occurs on average every 100 years over tens or hundreds of thousands of years could be occurring once every 10 years some millennium, or once every 250, 500 or 1000 years for some time span.

Please read up on binomial and Gaussian distributions.

Avar has a good point in general, although I'm not sure it's relevant to the recent San Andreas earthquake history. Over geological time, the actual probability distributions are unstable, because the boundary conditions causing the events change as plate configuration, climate, etc. changes.

With dice, each roll is independent of the others. With earthquakes, the probability of an earthquake occurring in any given year is (almost always) dependent on how long it has been since the last event. That is represented by the hazard function in the post (lambda t). I say 'almost always' because for a few distributions the hazard rate is constant with time.

An interesting thing about the hazard function with earthquakes is that it isn't always monotonically increasing; I've done this analysis on some other faults that show a very high hazard in the years to decades after an earthquake, and then the hazard drops down and doesn't recover until the far right tail of the distribution. This reflects temporal clustering of earthquakes, which is something we're gradually becoming aware of.

I would compare it to filling water balloons and timing when they pop. Our current water balloon should have popped a while ago, but the tension/pressure is still building.

Reply on news.ycombinator.com