"Why do Great Continental Earthquakes Nucleate on Branch Faults?" by Dr. Dr. Ross Stein

Summary

Dr. Ross Stein's lecture at the University of Utah delves into the complex interactions between large continental earthquakes and their nucleation on branch faults rather than the main transform faults. Despite the established view of main faults being the primary sites for seismic activity, the lecture highlights how major earthquakes since 2000 have originated on these smaller faults, posing new questions for seismologists and the public alike. By dissecting these anomalies, Dr. Stein sheds light on the hidden dynamics of the Earth's crust, suggesting that our understanding of earthquake initiation might need a paradigm shift towards examining these minor branch faults more closely.

Highlights

• Dr. Ross Stein is an expert in tectal physics and seismology, with a focus on earthquake hazards. 🎓
• Branch faults, not main transforms, are often the starting point for major earthquakes. 📍
• The 7.8 magnitude Karaman Marsh earthquake was a pivotal event that redirected focus to branch faults. ⚡
• Older seismic events show similar patterns, pointing to a systemic issue in fault understanding. 📜
• Branch faults might help reveal why great earthquakes occur where they do, suggesting a pivotal study area for the future. 🔬

Key Takeaways

• Great continental earthquakes often start on branch faults, challenging previous seismological assumptions. 🌍
• The interaction between branch and main faults may unlock secrets about earthquake triggers. 🔑
• These insights could alter future earthquake monitoring and prediction strategies. 🔭
• Branch faults may act as on-ramps to initiate large earthquakes on main faults. 🚦
• Understanding this could help in better assessing risks in earthquake-prone areas. 🏠

Overview

Dr. Ross Stein from the University of Utah presented an intriguing take on the origin of large continental earthquakes in his lecture titled 'Why do Great Continental Earthquakes Nucleate on Branch Faults?'. Shifting the focus from the primary transform faults to the branch faults, Dr. Stein underscores how these often-overlooked geological features might hold the key to understanding seismic activity better. By analyzing recent and historic earthquake data, he offers a compelling argument that challenges the conventional wisdom of earthquake nucleation.

One of the core revelations from Dr. Stein's lecture is the role of branch faults as possible precursors or triggers for larger seismic events on major transform faults. The evidence, he suggests, points towards a scenario where branch faults act like on-ramps, preparing the conditions for a bigger release of tectonic energy. This insight not only reshapes the geological understanding but also has significant implications for earthquake preparedness and hazard assessment worldwide.

Dr. Stein's discussion goes beyond just identifying branch faults as potential harbingers of major earthquakes; he dives into the mechanics of how these faults function, comparing them to 'transform superhighways'. Through this lens, he posits a new perspective on how we need to interpret seismic data and improve our predictive models. The lecture ends on a note urging further exploration and monitoring of these understudied faults to better predict when and where the next big earthquake might strike.

Chapters

• 00:00 - 00:30: Introduction The chapter introduces a distinguished lecture event at DLS, where the speaker, R Stein, is welcomed by the host. The host, likely named Plin, also acknowledges Bob Smith, who is perhaps another notable figure present at the event. The host mentions a longstanding professional relationship with R Stein, setting the stage for a lecture that draws from their collaborative experiences.
• 00:30 - 01:30: Collaborations with Bob Smith In this chapter, the narrator recounts a memorable experience with Bob Smith during their time at Cambridge University in 1985. They visited the Bay of Corinth, where they investigated a significant seismic event, a magnitude 6.8 earthquake that was caused by a known fault line. This exploration was part of their studies on normal faulting and seismic activity. The trip is highlighted as a significant joint endeavor, illustrating both the academic collaboration and the personal memories shared with Bob Smith.
• 01:30 - 03:30: Ross Stein's Professional Background In this chapter, the author describes a professional experience of Ross Stein, where he examined the hanging wall of a normal fall that was submerged under the ocean surface. Despite being a challenging and somewhat unorthodox task, Stein and his team took it upon themselves to swim out and observe the site closely. However, their actions attracted the attention of local authorities, who scolded them for their unconventional approach. The chapter illustrates Stein's commitment to his fieldwork and his willingness to take risks to gather valuable geological data.
• 03:30 - 05:30: Awards and Recognitions The chapter details a story where the speaker describes an encounter with law enforcement concerning antiquity. The narrative includes a quick escape upon realizing the presence of a policeman. It also mentions Ross, a tectal physicist and seismologist, indicating his work on earthquake interaction and hazards.
• 05:30 - 06:30: Introduction to Branch Faults in Earthquakes The chapter provides an introduction to branch faults in the context of earthquakes, featuring insights from a seismic hazard expert. The speaker is the co-founder and president of a seismic hazard study company called TBL. TBL offers a unique service where users can input their home address to receive a seismic hazard determination, demonstrating a practical application of seismic studies.
• 06:30 - 08:00: Tribute to Paul Tapponnier The chapter titled 'Tribute to Paul Tapponnier' acknowledges the significant contributions of Paul Tapponnier. It highlights his recognition as a fellow of the American association for his extensive work and acknowledges his roles as a former editor of 'JGR' (Journal of Geophysical Research) and president of AGU (American Geophysical Union) Tech.
• 08:00 - 10:00: Nucleation of Earthquakes on Branch Faults This chapter discusses the nucleation of earthquakes on branch faults, highlighting key achievements and awards in the field of physics relevant to this topic. The chapter mentions multiple prestigious awards such as the hu Gilbert award for natural hazards, the white award, PA silver award for outstanding science, and the two maker award for distinguished achievement. These awards are associated with AGU (American Geophysical Union) and GSA (Geological Society of America), illustrating the recognition of significant contributions to the understanding of earthquakes and related natural hazards.
• 10:00 - 12:00: Examples of Earthquakes on Branch Faults The chapter delves into the phenomenon of significant continental earthquakes. It focuses on the discussion of how these earthquakes begin or 'nucleate' on specific types of faults known as 'branch faults.' The terminology and specifics of branch faults are introduced. The text hints at an ongoing exploration or analysis by a character referred to as 'Our Kind friend boss,' who seems to be pivotal in the forthcoming discussion or presentation about this subject matter.
• 12:00 - 15:00: Investigation of Historical Earthquakes The chapter begins with a speaker addressing an audience, expressing interest in the outcomes of a talk. The speaker has recently returned from a conference in Paris honoring Paul Tapener, a figure significant to their scientific field.
• 15:00 - 18:30: Challenges with Smaller Earthquakes The chapter titled 'Challenges with Smaller Earthquakes' discusses the significant contribution of work initiated by a figure alongside Peter Molar. Their research focused on understanding the geological impact of the 15-million-year-long collision between India and Asia. This event, known as the concept of an indenter, led India to push into Tibet, resulting in the creation of complex fault and marginal basin structures across Asia. This foundational work highlights the intricate challenges posed by smaller earthquakes in this region, which are influenced by this historical tectonic activity.
• 18:30 - 22:30: Transform Faults and Branch Faults Relationship This chapter discusses the geological phenomena of transform faults and branch faults, focusing particularly on their role in shaping regions such as the South China Sea, the Philippines, Southeast Asia, Indonesia, and Taiwan. The narrative draws a vivid comparison to a slow-motion car crash to underscore the prolonged, 50-million-year impact of tectonic collisions on these areas.
• 22:30 - 25:30: Characteristics of Transform Faults The chapter discusses the distinctive features of transform faults, emphasizing how these geological structures can lead to significant changes in the earth's landscape. An example is given, where a collision causes the uplift of the earth's surface in Tibet, creating a large fault system. This event is contrasted with areas like Southern India, where the effects of such tectonic activities are not felt. The narrative credits Paul and Peter Moldar for first applying the principles of plate tectonics to understand these phenomena.
• 25:30 - 29:30: Role of Branch Faults in Initiating Earthquakes The chapter discusses the role of branch faults in initiating earthquakes. It references recent research work that extends several thousand kilometers into continents, which had not been done before. The chapter poses a question about why continental earthquakes nucleate on branch faults. This question is explored through collaborative work with Peter Bird from UCLA. The conclusion drawn is that branch fault onramps are crucial for affecting transform super faults.
• 29:30 - 32:30: Super Shear Ruptures The chapter delves into the topic of 'Super Shear Ruptures', focusing on how significant continental transform earthquakes actually rupture through branch faults rather than the main transforms themselves. The speaker aims to convince the reader of this reality and explore its implications for our understanding of major earthquakes. This discussion is prompted by a notable event that is familiar to the audience, emphasizing the importance of understanding why these ruptures occur in the manner they do and what new insights this knowledge can provide into earthquake mechanisms.
• 32:30 - 36:30: Public Awareness and Alerts This chapter discusses the unexpected rupture of a magnitude 7.8 earthquake that occurred a few years ago on the East Anatolian fault, one of the world's major transform faults. However, the earthquake did not rupture on the main fault itself but on an adjacent lesser-known fault called the Gnlrl Fault. The Gnlrl Fault, described as having a minimal slip rate and characteristics dissimilar to the main transform fault, was an unexpected site for such a significant geological event.
• 36:30 - 39:30: Northridge Earthquake and Mapping Challenges The chapter begins by dispelling the notion that certain earthquake phenomena are mere aberrations. It references the Denali earthquake, a magnitude 7.9 event in 2002, which ruptured on the Citna Glacier thrust fault parallel to the transform and Konon fault. This section discusses various faults, including the one discovered by Molar and Tapon, and how they relate to notable seismic events like the Kocil earthquake, which ruptured on the Tyang Lake fault.
• 39:30 - 45:00: Discussion on Fault Behaviors The chapter discusses fault behaviors, specifically focusing on the occurrences and characteristics of earthquakes since the year 2000. The text highlights how improved data has allowed for better differentiation between nucleation on a branch versus the main fault. The Wenchuan earthquake, which resulted in 70,000 fatalities, is presented as an example of an earthquake nucleating on a blind section of the fault at depth. Additionally, the chapter touches on the unusual characteristics of the Kikura Kakura earthquake, noting that it deviated from typical patterns or rules.
• 45:00 - 50:00: Conclusion The conclusion discusses the occurrence of significant earthquakes along the Kakura rupture, highlighting that these events took place on branch faults. Despite Kakura having multiple branch faults, it stresses that all major earthquakes since 2000 that measured 7.8 or larger happened on these branch faults. The segment suggests viewers consider this information in understanding seismic activities related to branch faults.

"Why do Great Continental Earthquakes Nucleate on Branch Faults?" by Dr. Dr. Ross Stein Transcription

• 00:00 - 00:30 right uh welcome to DLS um it's my plin and Bob Smith may not have met Bob either but Bob is thank you guys for all coming for the distinguish lecture of Our Guest R Stein who I have worked with throughout my entire
• 00:30 - 01:00 rare in good places and bad and I will just give you a little quip boss and I were at Cambridge University in 1985 we went to the Bay of Corinth magnitude 68 a big falsear and we were walking in P this SC because we were studying normal
• 01:00 - 01:30 fa and the hanging wall being under the ocean surface about one or two meters because it was a hanging wall of a normal fall and we took it upon ourselves to swim out and look at all these [Music] buried as we were getting out police would pulled in and really gave us help said uh doing is
• 01:30 - 02:00 illegal and he says but I'm a policeman I'm going to get the Antiquity policeman stay right here turned around the corner one second we were out of there that's one of the things we've done together so Ross is a tectal physicist seismologist and he works on earthquake interaction and earthquake hazard he's
• 02:00 - 02:30 the co-founder and president of tar a company that does seismic Hazard study globally and I recommend that you take a look at it because in in [Music] TBL you can actually put in your own home address and it will create a seis haard determination for you so it's it's a hard but very very
• 02:30 - 03:00 useful tool um boss is a fellow of American this and of the of America for his long record Appliance former editor of jgr which many of us ha to and president importantly of the agu tech
• 03:00 - 03:30 physics uh he's won multiple Awards and I'm not going to give the mod I'll just give a couple here the hu Gilbert award for natural hazards the white award PA silver award for outstanding science two maker award for distinguish Achi these are agu Awards and the GSA
• 03:30 - 04:00 uh distinguish and this goes on and onward as well so I want to introduce Our Kind friend boss who's going to talk about why do great Continental earthquakes nucleate on normal faults Branch faults Branch faults what branch faults branching Falls the question is we're on a continental fall here in the WW
• 04:00 - 04:30 and so the results of your talk will be of great interest to us thank you Bob boss go ahead before I begin I just came from Paris for a conference in honor of Paul tapon and many of you may not know who Paul tapener is and because he's so important to our science I thought I would just take a moment for you to
• 04:30 - 05:00 understand what his contribution was and it was the work that he did initially with Peter molar uh that that understood that the 15 million yearlong Collision of India into Asia is responsible for the entire fault structure and marginal Basin structure of Asia so this concept of an indenter which is what India did is it pushed into Tibet produced all these f
• 05:00 - 05:30 which you see on the right and created the South China Sea and the configuration of the Philippines of Southeast Asia of Indonesia of Taiwan this enormous impact and just to give you a sense of this without being too modelin this is a 50 million yearlong Collision if this was a national Transportation board video of a car crash what has happened is that this crash is moving so slowly
• 05:30 - 06:00 but so decidedly that the people in the front seat the engine of the car is in now in the front seat but the people in the back seat don't even know the crash is occurring so in Southern India you don't feel a thing but the lifting up of the roof of the earth through Tibet and the creation of this giant fault system is all a product of this remarkable Collision we owe that to Paul who with Paul uh Peter moldar first took the principles of play tectonics and
• 06:00 - 06:30 and extended them several thousand kilometers into the continents which before they did this work had never been done okay so now I've posed this question why the Continental earthquakes nucleon Branch vaults and this is work with Peter bird from UCLA okay here's our answer the branch vaa onramps are needed to hit the transform Super
• 06:30 - 07:00 Highway okay but first I have to convince you that this is a thing that in fact the largest continental transform earthquakes have ruptured through Branch vaults and not on those transforms themselves and to the extent that I can convince you that this is true then it becomes important why would this be so and what does this unlock about our understanding of how these great earthquakes occur so this work was triggered by an event that I think all of us saw just a
• 07:00 - 07:30 few years ago the magnitude 7.8 caraman Marsh earthquake on the East Anatolian fault a major one of the world's major transform faults and oddly it didn't rupture on the East Anatolian fault it ruptured on the gnarly fault this little wannabe fault with almost no known slip rate and it doesn't have a sense of slip anything like the transform itself so we thought oh well well this has got to be
• 07:30 - 08:00 an aberration but when we began to look we saw no this is not the case you probably some of you remember the Denali earthquake magnitude 19 7.9 in 2002 it ruptured on the citna glacier thrust fault that was more or less parallel to the transform and the kunon fault one of the faults discovered by molar and tapon uh produced the kocil earthquake but that rupture on the tyang lake fall fall
• 08:00 - 08:30 another spay feature and if you look at the other two events that have occurred since the year 2000 since the data was good enough really to be able to distinguish nucleation on a branch versus the main fault you can see the wwan earthquake which killed 70,000 people ruptured or nucleated on some kind of blind part of the fault at depth and finally kikura kakura broke all rules you might say well wait a minute
• 08:30 - 09:00 almost the entire kakura rupture is a bunch of Branch faults that's that's true but it ruptured on another one of these Branch faults so even though kyura has lots of Branch faults that's where it ruptured okay so if we look at these together what we see is that the five 7.8 and larger earthquakes that we've had since the year 2000 all ruptured on Branch Falls now you may want to equival
• 09:00 - 09:30 with that you may want to say well wien's a little different because it's blind kakur is a little different because there are a lot of branches but let me flip it over and say we are looking at, 1500 kilometers of straight transform faults in these images collectively and it none of these ruptures occurred or began on any of those straight simple parts of the Fall that's got to be important so if it is
• 09:30 - 10:00 then is this consistent with older events even though the evidence is never going to be as good for older events we'd like to know that this it is we're not just playing with the statistics of small numbers so here's probably one of the most important transform events that occurred this Century the aanan earthquake which started the spectacular falling Domino series of earthquakes along the north Anatolian fault where in 60 years a thousand kilometers of that fault r cured one after another but I
• 10:00 - 10:30 want you to see here is that the epicenter the blue star and its uncertainty place it off the fault on a mapped Branch at a bend in the north Anatolian fault so it looks like a good candidate we'll never be certain but looks like a good candidate okay what about the San Andreas well its big earthquakes are too long ago ever to be able to say exactly where they ruptured but here's what we can say where these two earthqu Quakes probably ruptured there were plenty of
• 10:30 - 11:00 Branch faults available doesn't mean it ruptured on the branch faults but we can't eliminate the possibility that both of these earthquakes could have nucleated on Branch faults that's about the strongest statement that we would be able to make in these cases because these are also 7.8 7.9 earthquakes as best we know okay but there's a problem here which is when we just drop down to 7.7 which is about 30% smaller by moment
• 11:00 - 11:30 it's not necessarily so so here are two events on which they're 7.7 and they did not rupture on Branch faults and here are two other events where the evidence is debatable but it certainly isn't compelling so why would things be different for a 7.7 than a 7.8 and larger and that's the question that this counter evidence or falsification evidence raises now I can think of a couple of
• 11:30 - 12:00 explanations or perhaps three one as the earthquakes get smaller the distance between the epicenter whether or not the we epicenter is off or on the transform becomes harder to know because of a location uncertainty okay or it could be mechanical if you don't rupture one of these things on a branch fault the rupture Fizzles out sooner and you end up with a smaller earthquake or it could be that we are being fooled by the statistic of small numbers and selection
• 12:00 - 12:30 bias confirmation bias is around every corner in science so we have to be alert to the fact that we could be fooled by a happenstance set of circumstances but let's take a closer look at one of these well-studied earthquakes the the magnitude 7.6 isit earthquake in Turkey which killed 30,000 people and here it's peculiar so one interpretation of the result of
• 12:30 - 13:00 the mapping is that this earthquake ruptured at a Bend or break in the fault that involved normal faulting and another one would be that since the epicenter is off the fault that there is a branch but it's unmapped because it's under the ismet bay and we don't know which is right but you might say well now wait a minute couldn't in general these earthquakes rupture it bends and breaks and faults that's an idea that was first proposed by Jeff King who Bob just mentioned and John
• 13:00 - 13:30 nabalik and while that is possible there's a problem with it which is that if the if it ruptured all the way to the failure stress on the fault then locally you'd have a huge stress drop about 80 or so bar megap pascals and we never see that in other words big transform earthquakes have very low stress drops they don't ever seem to get any were near failure
• 13:30 - 14:00 and so if it's going to rupture on that fault we would expect to see at least locally a giant spike in the stress drop and we never see it all right now let's ask another falsification question if Branch faults are so important whenever we have a let's say a magnitude six or larger earthquake within let's say 15 kilometers of a transform does that transform rupture and the answer is no something like a
• 14:00 - 14:30 quarter to a half of those events do though so the ones that don't are shown in yellow in these pictures so this is this is evidence in the eye of the beholder you can either say well look a lot of these Branch faults don't trigger transform so that makes this concept suspicious or you could say wow 25 to 50% that's a very high ratio of potential triggering here so why wouldn't all of them do the job
• 14:30 - 15:00 well we can think of some reasons maybe the transform had just had a rupture recently so it's unlikely to rupture again even if you could fire off a magnitude 6 earthquake maybe the geometry was unfavorable between the two earthquakes or maybe there's some other element about the magnitude sixes that seems to be important I want to show you evidence that these transforms have unique character istics in that they're
• 15:00 - 15:30 never close to failure they all rupture prematurely you give me a magnitude 7.8 transform earthquake you're looking at a prei let me show you what I mean here is the San Andreas that were ruptured in 1857 and 1906 and the 1872 Owens Valley earthquake and all three repor zones are seismicity holes today this is magnitude six since the turn of Century I could show you just
• 15:30 - 16:00 like magnitude 3 since the in the last 10 years you'd see the same thing they holes okay that's a riddle and what does it mean my interpretation of what it means is that they're not close to failure if they were close to failure we'd see a lot of earthquakes occur right we would expect that with time the stress would rebuild on the fault and we'd come back to the configuration that there were lots of earthquakes somewhere along the way and
• 16:00 - 16:30 eventually it would rupture but instead they're holes and as far as we can tell there holes in the years before they rupture too we can see that with the kunon fault and uh the Kora Vault and the other ones that we looked at so what might that mean it it might tell us why Branch vaults are important so let's think really simply just in terms of GE geometry the thing that you
• 16:30 - 17:00 need to know is that there's practically no Continental transform on Earth which is what we would call optimally oriented in other words it's aligned properly for the regional compression to rupture so here's a right lateral transform and the principal compression axis and this is almost exactly what we what we see for the San Andreas in other words the principal compression is almost perpendicular to the fall okay so let me give you an analogy I just ate a watermelon okay and so I take that
• 17:00 - 17:30 watermelon sle uh seed which is very slippery and we've all done this you put this on the table and you push your thumb down your thumb can almost go straight down the thing goes boom so we're not optimally aligned we're not pushing it to go sideways if it's slippery enough if the fault has a low enough coefficient of friction then even if the principle compression is almost perpendicular to the fall we can get it to rupture but if we leave that watermelon seed out overnight and repeat
• 17:30 - 18:00 the experiment and push down your thumb it's going nowhere because now it has high friction and the fault won't move so let's say we have this situation what it's telling us is it's very hard to get the sandreas or this transform to move and to do so it would have to have an extremely low friction coefficient it has to behave like a watermelon seed but what if we introduce a branch fall now this Branch fall happens to be a reverse fault and it's perpendicular to the
• 18:00 - 18:30 compression axis so now the branch vault is optimally oriented for the stress it's more likely to rupture okay here's another one here's a left lateral Branch fault again optimally oriented at high angle of the compressional stress that Kai can go even if the transform can't so if you imagine that basically the gods sprinkle all kinds of mangy Branch vaults around a transform some of them will happen to be well oriented for the
• 18:30 - 19:00 regional stress and they're more likely to go that's the first thing I want you to see just on the basis of geometry we can imagine scenarios that we fire one of these things off okay and so now let's go one more step and ask the question okay what are the unique characteristics of transforms due to their huge amounts hundreds of kilometers of cumulative slip that make them behave differently than all other
• 19:00 - 19:30 faults and there's been a lot of research on this both observational and theoretical and the the long and the short of it is they have to have sliding friction that's extremely low they have to behave like watermelon seeds and so that they produce damage zones zones of several hundred kilometers uh several hundred meters wide with with combinated trashed rocks that are often fluid filled what's important about that is that trap seismic waves reverberating
• 19:30 - 20:00 in this Zone can trigger a lot of shaking and a lot of intense stresses and if you can Flash Heat this water the water expands and the fall surfaces can float off each other and suddenly you can have an enormous frictional drop that's the idea that's been promoted through a large number of studies that suggests that transform faults never get
• 20:00 - 20:30 to reach their failure stress they rupture prematurely at very low stress somehow they managed to do that now how does the branch fault come into the picture the problem with this picture this view that that Jim Rice has been particularly important in advancing this hypothesis of the unique ability of these faults to suddenly have a drop in in frictional sliding uh behavior is how do you get it started
• 20:30 - 21:00 how do you flash heat that water when you're just sitting there because it's got to get going that those those effects happen because the fault's moving at um you know 35 miles 3500 miles an hour all of a sudden produces all that heat but how do you get that heat to happen so the the canundrum of this strategy is it doesn't provide a way for the rupture to begin so introduce the branch Vault and if you
• 21:00 - 21:30 can slam that Branch vault which does reach its full frictional value if you can slam it into the transform then what happens is you can start that flash Heating and then you can get a rupture to go okay so this is why we say that Branch vaults are the on-ramps to the transform Super Highway and maybe this is the only G way you get one of those bees to fire off a large
• 21:30 - 22:00 earthquake now there might be another element in this that could be important that could give some Branch ruptures extreme power to trigger a transform and that's called super sheer rupture so super sheer rupture is what happens when the rupture basically breaks the sound barrier in the Earth's crust and when a plane flies overhead and breaks a sound layer you hear a boom and that boom is called a mock cone in the air and that's
• 22:00 - 22:30 exactly what happens in the earth if the rupture can speed the speed of the unzipping of the fault moves faster than the seismic waves and it produces this mck cone so this is a a rupture moving this way and it produces this big cone which has suddenly strong uh stresses and if this is the transform what happens is you not only hit the transform in one spot but along this
• 22:30 - 23:00 entire cone you could hit the transform at once so a super sheer rupture is capable of hitting the transform much harder than a normal rupture okay so is this a rosettastone not clear the evidence is inconclusive there are a couple of studies that show that that gnarly fault rupture in the caraman marsh earthquake was super sheer but there's a couple of settings that show that it is isn't and in all the other cases I showed you the evidence
• 23:00 - 23:30 either is lacking or ambiguous so it could be important I think we can say that but we don't know that that's the requirement okay so I think you've seen there is evidence it's not overwhelming evidence but there's evidence that Branch fault ruptures play an important role in transforms and there's evidence that this could solve a problem in terms of how we get these big
• 23:30 - 24:00 ruptures to start but we don't I I think it's fair to say okay the jury's out but the problem is this has already been in the public eye twice so we're going to need to figure out what's happening here and I want to show you these two cases so in August of last year there was a magnitude um 7.1 earthquake on the edge of what the Japanese think to be the future nankai rupture that man itude 7.1 earthquake was believed to be a high angle if you
• 24:00 - 24:30 like Branch rupture off the mega thrust and the Japanese declared a mega Quake advisory for one week and that had huge societal consequences so basically they were playing out the scenario that we're talking about in this room but before the entire public and of course you have a very big problem here because what are you going to do at the end of the week that the you know the Earth doesn't have an all clear signal so when do you
• 24:30 - 25:00 decide that you no longer have an issue you know if you're going to do this probably uh the next Tuesday you're going to have the big earthquake and you called it off nevertheless they did this and a little while earlier in 2016 the California governor's office of emergency services declared an earthquake advisory following a swarm in the saltan sea also for a week and you can think of the Broly seismic Zone on which that swarm occurred as a branch off the southern end of the San Andreas
• 25:00 - 25:30 kind of like what we saw in the konon earthquake so they played with fire too so like it or not this is out there we need to figure it out and most recently in December there was a magnitude 7 on the mesino fall and I think you could easily call right or wrong the mesino fault a branch off the northern tip of the San Andreas with aftershocks leaking on apparently to the
• 25:30 - 26:00 1906 rupture so amazingly they didn't call it so you can see it's not even being applied very consistently but these issues are out there and we need to understand them so here's where I want to leave you Branch fall rupture maybe how great Continental earthquakes are born and if so and I'm first to admit it's a big if we've been looking in all wrong places or where these great
• 26:00 - 26:30 earthquakes begin thank you Northridge so as you may remember if we think back to this San Andreas um and
• 26:30 - 27:00 I'll just use this picture for that purpose this is the Garlock F so the San Andreas is Right lateral and here's a left lateral fault that comes into it so you think I showed you an example of a branch fault that is a left lateral Branch off a right lateral fault maybe it's maybe it's such a feature and a Northridge occurred in here Northridge is some kind of Branch off maybe one of the bigger thrust faults in Southern California you know unfortunately most of our efforts in Fault mapping have
• 27:00 - 27:30 gone to the major faults uh these Branch faults get no love and so we don't know very much about them and we I think it's fair to say that imagine two scenarios there's a magnitude six that occurs on the San Andreas versus there's a magnitude six occurs I don't know uh near Kinga 15 kilometers off the San Andreas wh for which one is the USG is going to freak out we know which one but it could be
• 27:30 - 28:00 backwards they could have it just wrong it's hard to get that six to do anything if it's actually nucleating on the on the San Andreas I think yes say say it again so do normal faults and thrust faults have this same unique behavior of
• 28:00 - 28:30 this rice like damage Zone it's not clear I would say the best evidence is this is something unique to transforms because they alone have hundreds of kilometers of cumulative slip in general normal faults and thrust faults don't now the exception of course is a mega thrust but Mega thrusts are a different kind of Beast so I'm not sure to what it to what extent it applies to them um as you may know there was a a what we would
• 28:30 - 29:00 call retrospectively a fores shock to the tohoku earthquake uh a couple days beforehand I'm not sure if it was on the mega thrust surface or below or above it's very very difficult to tell because you need very accurate depths so I'm not making that argument at least but maybe it applies in some cases Jim let's
• 29:00 - 29:30 what criteria possible okay that's right it's such a great question so I went to the earthquake early warning
• 29:30 - 30:00 people the various gurus and said okay um what about a feature in earthquake early warning which we have on the West Coast United States I know you don't have it here yet in which if a magnitude 6 occurs on a branch Vault within let's say 15 kilometers of San Andreas the warning you get is um you're likely to experience the shaking for magnitude 6 but there's a 25% chance that this will be followed by a magnitude eight
• 30:00 - 30:30 okay so um would that be helpful and would that be accurate I actually am getting a very mixed reading um everybody doesn't agree that that's even appropriate and and that's because little earthquakes can be overachievers even a six can produce stronger shaking than an eight at some small likelihood so it may not be that different but I think that one of the ways that this could be harness would be to try to build it into earthquake early
• 30:30 - 31:00 warning and to recognize that some that all sixes are not alike and if if a six is in a certain location with respect to a major transform then yes there's a possibility of a much larger earthquake than we would otherwise expect yes
• 31:00 - 31:30 okay so the our view is that the branch faults themselves don't have the low Co coefficient of friction so in other words they they are what what Peter would call berly faults they they have a high value of friction and they go all the way up to their failure stress but the transform never gets there and this is interesting this means that if if these faults here's one of the implications of this
• 31:30 - 32:00 if the transform actually came up to failure had an earthquake came up to failure then you could imagine it would have some kind of regular recurrence regular is right okay the San Andreas maybe 250 years plus or minus 50 or 100 years but if instead it never gets to failure and whether or not it ruptures is a function of just this random walk of of Branch vaults you wouldn't expect it to be regular at all it's kind of just a roll of the dice yes you wouldn't
• 32:00 - 32:30 have two eights in a row but essentially you would expect the recurrence behavior of big transforms to be rather poor and with the exception of the Alpine fault in New Zealand that's exactly what we see we make every attempt to make them look regular but they're not yes
• 32:30 - 33:00 I wouldn't imagine that's kind of what I think is the most likely scenario that that the that the branch fault comes slamming in to the transform Heats things up and bang it goes but okay there are a lot of processes and there's even theoretical basis for delays between one and the
• 33:00 - 33:30 other so you could make an argument which is basically what JMA in Japan and the governor's office of emergency services did and said well okay we we're having a branch event but we could get a delayed repture on the big one it's hard for me to see I because you'd think okay all that Heat's going to dissipate whatever you're doing it's going to be over so I see it as a onew deal
• 33:30 - 34:00 Chris yeah that's true so you know if I were on the other side of this argument I'd say okay let's wait for the next one of these 7.8 if it happens on a branch fault we're all in if not you're toast that would be
• 34:00 - 34:30 fair science is supposed to be predictive and we always run the risk that we are reading too much into the Tea Leaves of the records that we have and I think that would be a fair criticism of this hypothesis but I think it solves a problem regardless I think it's otherwise almost impossible to get these these big guys to go yes
• 34:30 - 35:00 probably it is I'm not an expert at it you know the idea you have to have water in a confined space in The Fault Zone that can't as it heats up leak out so only then can can you basically boil it and cause an expansion and suddenly have the drop in um friction so that you go from a uh old watermelon to a slippery
• 35:00 - 35:30 watermelon seed right away uh so there are lots of possibilities for these processes and Flash Heating and melting of the surface the rupture surface the couple of millimeters of rock along the rupture surface is another one there's an enormous amount of heat that's presumably produced by the rupture itself I think what's important is the evidence that none of these transforms are optimally oriented none of them in
• 35:30 - 36:00 in finite uh element models that try to get all the plates to move and the faults to move you never can get these things to move unless you bring the friction down to this ex extremely low value that is not what we see for other kinds of faults yes that's a good question I haven't looked it's a really good question
• 36:00 - 36:30 yes regular yeah that's a good point right because once
• 36:30 - 37:00 that fault ruptures it drops the stress along there and a thwart there and then slowly the stress is going to rebuild and so yes the transform is still the Big Driver of the variation in stress through the earthquake cycle that's right and so you can imagine therefore that since the stress drops if this is a transform and we're looking across it the stress drops from its earthquake and then it slowly re building so you'd imagine that Branch vaults getting
• 37:00 - 37:30 closer and closer to the San Andreas would start to get up to failure particularly if they were properly oriented so yes good idea yes it's a great great question and I I profess my bias you know coming from
• 37:30 - 38:00 California we're just in love with the San Andreas it's like the velvetine rabbit everything but its eyes have popped out um so what do I know about Oceanic Transformers very little would fit on the head of a pin you know I I would say the big difference is oceanic transforms are very slender and they're even more slender as you get to the ridges on either end of them so they are they don't they're aren't able to produce as big earthquakes in the first
• 38:00 - 38:30 place uh but in every other respect you're probably right that they could they have huge amount of cumulative slip I think that's what you're pointing out and therefore why don't they have the same behavior could it could be J
• 38:30 - 39:00 yeah be a great idea no we have I think that would be really useful probably you'd have this problem of
• 39:00 - 39:30 small numbers but it would be worth looking and the other question you might say is well do you see any buildup of seismicity on the branch vaults before they go off they indicate they're close to failure you know you're feeding us this line that they're close to failure but not the transform prove it you could be saying and I'd have a hard time proving it because you know virtually no Faults start to become seismic before they uh knock off you know you can think of Two Worlds the world we inhabit in which
• 39:30 - 40:00 aftershocks happen after the main shock and another world in which for shocks happened before the main shock that that second world we'd all be predicting earthquakes so effectively that we wouldn't even have these meetings uh but unfortunately we live in the other world where the earthquakes generally occur afterwards so I I I doubt there's a buildup but we haven't looked okay I really am glad to have had this conversation with you and I
• 40:00 - 40:30 remember Paul tapier and um and I thank my co-author Peter bird and my introducer Bob Smith