WEBVTT 00:00:01.000 --> 00:00:18.000 So our first plan; our first item on the agenda was to have us all go around and introduce ourselves. But seeing as they're still 131 people logged in, we probably shouldn't do that if we want to be able to leave today. 00:00:18.000 --> 00:00:27.000 Forget the first item on the agenda. So new rule, the first time you say something, just introduce yourself who you are, where you come from. 00:00:27.000 --> 00:00:30.000 Hi! I'm Sarah Minson. I'm from the U.S. Geological 00:00:30.000 --> 00:00:36.000 Survey in Moffett Field/Menlo Park; wherever we are currently in the move. And your moderators for this evening are Keith and Greg. 00:00:36.000 --> 00:00:43.000 Say, "hi" Keith and Greg. 00:00:43.000 --> 00:00:44.000 Sarah and Greg. 00:00:45.000 --> 00:00:49.000 Okay. 00:00:49.000 --> 00:00:51.000 You want an introduction? 00:00:51.000 --> 00:00:53.000 Oh, just say Hi! 00:00:53.000 --> 00:00:56.000 Okay. So I'm Greg Beroza. I'm a professor at Stanford, currently at an Air Force tactical meeting in Florida. 00:00:56.000 --> 00:01:05.000 I'm also co-director of SCEC. 00:01:05.000 --> 00:01:08.000 And I'm gonna help Sarah out. I'm Keith Knudsen. 00:01:09.000 --> 00:01:15.000 I'm the emeritus of the Northern California Coordinator for the Earthquake Hazards Program. 00:01:15.000 --> 00:01:27.000 Thank you, Keith. And so what brings us together today is the expansion of SCEC from Southern California into Northern California and beyond, and this is the Northern California Earthquake Hazards Workshop. 00:01:27.000 --> 00:01:32.000 So let's talk community activities in Northern California and the SCEC expansion. 00:01:32.000 --> 00:01:36.000 So let's start by talking about the SCEC expansion. 00:01:36.000 --> 00:01:40.000 Yuhuda, would you like to talk about that? 00:01:40.000 --> 00:01:50.000 Yeah, thank you very much, Sarah. And can I show screen? 00:01:50.000 --> 00:01:55.000 You see my screen? 00:01:55.000 --> 00:01:58.000 Almost. Yes, we see a fish. 00:01:58.000 --> 00:02:02.000 Just a second. I don't. I don't see it. 00:02:02.000 --> 00:02:03.000 Okay, let me make it make me let me expand it. 00:02:03.000 --> 00:02:11.000 Okay, yeah, so, so thank you. I will present very briefly, rather rapidly, three slides. 00:02:11.000 --> 00:02:21.000 The first two slides, and to provide context for the planned expansion of SCEC to statewide coverage, and then the third slide will actually begin to move into next 00:02:21.000 --> 00:02:31.000 items on the agenda. The third slide 00:02:31.000 --> 00:02:42.000 we list examples of key topics that can lead to productive collaborations amongst statewide researchers. 00:02:42.000 --> 00:02:46.000 So on April 22, SCEC submitted an NSF 00:02:46.000 --> 00:03:00.000 proposal to develop the next earthquake center with the statewide coverage. The proposal had five key questions that begin with, listed on the left here. It begins with, 00:03:00.000 --> 00:03:10.000 "How do geodynamic and other processes load the San Andreas Fault System across various locations and space-time scale?" 00:03:10.000 --> 00:03:16.000 Then the next question is, "How do fault zones in the bulk of the crust respond to the loading at different locations, depths, and times?" 00:03:16.000 --> 00:03:24.000 Question 3 is, "How does the active fault network in California evolve on different timescales?" 00:03:24.000 --> 00:03:31.000 And then "How do the evolving properties in turn exert control on the continuous deformation and ground motion? 00:03:31.000 --> 00:03:40.000 Question four is, "How can data analysis and simulations improve the predictability of strong ground motion? 00:03:40.000 --> 00:03:48.000 And the last question is, "How can system level modeling and observation improve large earthquake forecasting?" 00:03:48.000 --> 00:03:54.000 To address these questions we propose four research thrust 00:03:54.000 --> 00:04:07.000 with 12 subtopics in total. The first thrust focus is on observation, and in particular, closing critical data gaps. 00:04:07.000 --> 00:04:15.000 And here there is a focus on new fault studies, the dynamics of the shallow crust 00:04:15.000 --> 00:04:21.000 using machine learning AI techniques of the type we heard earlier today 00:04:21.000 --> 00:04:27.000 and then integrating all of these results to multi-scale community models. 00:04:27.000 --> 00:04:39.000 The second thrust is developing rheologies that bridge the multitude of scales and condition both for fault zones and the crust as a whole. 00:04:39.000 --> 00:04:45.000 Then the third trust is developing at one's modeling frameworks 00:04:45.000 --> 00:04:53.000 the queues, the constituent laws, and constraints based on the community models to perform simulations. 00:04:53.000 --> 00:04:58.000 Simulations for integrated tectonic modeling, coupled evolution of earthquakes and faults, 00:04:58.000 --> 00:05:08.000 and then an improved estimates of seismic hazard in California. And finally, the last thrust is on improving 00:05:08.000 --> 00:05:17.000 predictive analysis of seismicity with various subtopics. 00:05:17.000 --> 00:05:24.000 So the fault system in southern California was the focus of SCEC for decades, but with notable exceptions, statewide exceptions that include the community fault model, 00:05:24.000 --> 00:05:38.000 UCERF, focusing activities such as REALM and CSEP, and also see CEO activities among others. 00:05:38.000 --> 00:05:42.000 These statewide exceptions that have been going on for a long time now 00:05:42.000 --> 00:05:56.000 actually, they provide examples of models. How can we develop statewide activities in other topics? 00:05:56.000 --> 00:06:11.000 The Statewide extent of the research provides a considerable more and better opportunities, and for the [indiscernible] by just studying portions of the system separately. 00:06:11.000 --> 00:06:32.000 Very briefly we know that the San Andreas System as a whole exhibit significant diversity of properties, both geometrical properties of faults, geological properties of faults, starting with mixed affinity, oxide styles changing to continental affinity, and then changing again to 00:06:32.000 --> 00:06:52.000 oceanic affinity in the north, and also a diversity of behaviors such as great earthquakes, such as with creeping, such as with [indiscernible] and transitions between all of these sections, so the diversity of these properties and behaviors in the entire system improves the 00:06:52.000 --> 00:07:04.000 possibilities to which deeper and broader understanding of earthquake behavior in all parts of the system. Some key hazard related questions 00:07:04.000 --> 00:07:19.000 we would like to focus on, together with, collaborators from the USGS and other institutions in northern California, how? What are the probabilities of large earthquakes and damaging strong ground motion in different parts of the plate boundary? 00:07:19.000 --> 00:07:33.000 Second general question is, what evolving conditions, and processes of multiple space-time scales allow large earthquakes to occur? And how these evolving conditions are manifested by diverse observable data. 00:07:33.000 --> 00:07:39.000 This is obvious relation to focusing. The second question is, what fraction 00:07:39.000 --> 00:07:49.000 of the cumulative elastic time energy is released seismically in different areas? We know that there's more creeping in the north, 00:07:49.000 --> 00:08:04.000 but there's also something in the south. There's also awful deformation, and we would like to quantify all of those issues better in the next center, together again, with a statewide collaborators. 00:08:04.000 --> 00:08:18.000 [white screen, technical issue] So in aspiring, in considering statewide extent, we aspire to give out the understanding of the earthquake system in the plate boundary in California, together with statewide researchers from the USGS 00:08:18.000 --> 00:08:32.000 and other institutions. Example of our productive collaborative goals include our first developing improved statewide AI based desktop and focal mechanism catalogs. 00:08:32.000 --> 00:08:40.000 We have two great talks about this today from Greg Beroza just now and earlier, [indiscernible] focal mechanisms. 00:08:40.000 --> 00:08:46.000 There are some in addition to just the work, there's some important outstanding questions we should consider together as a community. 00:08:46.000 --> 00:08:50.000 It can actually open up the questions, how do we test, compare, and rank the different catalogs? 00:08:50.000 --> 00:08:54.000 It's not obvious. There's no solution for this at present. 00:08:54.000 --> 00:09:00.000 The next topic is extending the following, the example, with the CFM. 00:09:00.000 --> 00:09:04.000 Which would be developed further, extending also the same CVM. 00:09:04.000 --> 00:09:10.000 CGM to statewide to enable uniform physics based seismic hazard studies. 00:09:10.000 --> 00:09:14.000 So here they are like in all other topics. I won't get into too many details, but there are different categories of challenges. 00:09:14.000 --> 00:09:21.000 One category is we need to agree on information technology. 00:09:21.000 --> 00:09:30.000 That is modern, and flexible, and allow accessibility and visualization. 00:09:30.000 --> 00:09:42.000 So there are many issues here, and we also should figure out ways that we can benefit from all existing velocity models and try to develop manage multi-scale models. 00:09:42.000 --> 00:09:52.000 This topics will be, I'll say a little bit more about this in the workshop on February 14 for Team's, so I'll leave it at that for now. 00:09:52.000 --> 00:09:59.000 Then we want to develop other statewide community models as feasible. 00:09:59.000 --> 00:10:08.000 So this includes community stress model, community temperature model, community rheology model, and community paleo-seismic models. 00:10:08.000 --> 00:10:16.000 I will develop to different extent in different regions, and that would be a great topics for joint collaborative work. 00:10:16.000 --> 00:10:17.000 Now, next, if we have these community models, the next challenge would be very important. 00:10:17.000 --> 00:10:43.000 Would be to use community models in different regions to get Digital Twins for regions that we have well developed models and then use the Digital Twins for simulations of both tectonic deformations, ruptures and ground motion. This is very important because some models inherently, are 00:10:43.000 --> 00:10:49.000 going to remain incomplete because they're dealing with data that are unobserved or very difficult to observe 00:10:49.000 --> 00:11:10.000 such as space and temperature. And if we have validated simulation with Digital Twins, we can, of course, use those results to update a community model, community displacement or temperature models then creatively engage in productive positive feedback. 00:11:10.000 --> 00:11:18.000 Next, we want to pursue physics and AI best earthquake focusing using improved catalogues that aim to perform 00:11:18.000 --> 00:11:21.000 ETAS and this of course would be accompanied by CSEP testing. 00:11:21.000 --> 00:11:38.000 We would like to develop, there's a whole big topic that will require collaborations with everyone who is capable and willing. We want to develop, import physics-based earthquake simulators. 00:11:38.000 --> 00:11:42.000 So there are two parts here; one is, it will be important to extend the ask 00:11:42.000 --> 00:11:49.000 to become statewide. This is already to some extent in progress, and is doing well. 00:11:49.000 --> 00:11:54.000 The next simulator that we want to develop is next generation asks simulators 00:11:54.000 --> 00:11:58.000 that accounts for coupled evolution of earthquakes and faults. 00:11:58.000 --> 00:12:11.000 That is the fault properties post geometry and material, are evolving as part of the deformation, and we studied the whole system as one dynamical, 00:12:11.000 --> 00:12:15.000 the whole behavior in evolution is one dynamical system. 00:12:15.000 --> 00:12:16.000 We want to develop also improved physics based probabilities, seismic 00:12:16.000 --> 00:12:25.000 hazards, using the updated CyberShake and updated multi-scale community velocity models. 00:12:25.000 --> 00:12:30.000 And the last topic I list here before I open it up for discussion or 00:12:30.000 --> 00:12:41.000 other topics is we want to work together on community workshops, meetings, and trainings; and good examples that we have to continue our 00:12:41.000 --> 00:12:47.000 workshop from dynamic rupture that were very helpful in benchmarking calls. 00:12:47.000 --> 00:12:58.000 The recent workshop that we had tag actually is that we had on strip strength of validation and workshop and training on ground motion modeling. 00:12:58.000 --> 00:13:06.000 So I have more, but I'll stop here and open it up for bringing it back to the moderator. 00:13:06.000 --> 00:13:10.000 Bring it up for discussions. 00:13:10.000 --> 00:13:11.000 Okay. 00:13:11.000 --> 00:13:16.000 Excellent. Thank you so much, Yehuda, and I will just say, don't worry to the audience, don't 00:13:16.000 --> 00:13:30.000 worry quite yet about understanding all the CXMs. Up next, we have Noah and Scott to talk about them in more detail, but perhaps they can now ask Greg to entertain questions on everything else. 00:13:30.000 --> 00:13:43.000 Sure. So questions from the group about the proposal. 00:13:43.000 --> 00:13:52.000 Looking in the chat. There's specific questions, but nothing really focus on the proposal I'm finding yet. 00:13:52.000 --> 00:13:56.000 Oh, I'll ask a quick question. This is um, I'll 00:13:56.000 --> 00:13:57.000 turn on my video I'm Jonathan Dolphin, a system 00:13:57.000 --> 00:14:06.000 professor Purdue University. It look like you guys are really focusing on the transform deformation. 00:14:06.000 --> 00:14:14.000 Are there plans to kind of leave the subduction related area out of this proposal? 00:14:14.000 --> 00:14:20.000 We plan to collaborate with the parallel initiative that was developed to study the subduction zone. So we are not. 00:14:20.000 --> 00:14:29.000 We don't intend to leave it altogether. 00:14:29.000 --> 00:14:41.000 No, The answer is, no, but we do plan to focus on the transform plate boundaries and in the region which overlaps with the subduction to do research. 00:14:41.000 --> 00:14:48.000 Most in our center, as well as to collaborate with the other centers. 00:14:48.000 --> 00:14:49.000 Got it. Thank you. Go ahead. 00:14:49.000 --> 00:14:54.000 Yeah, the boundaries are as as mentioned, as Jay mentions in the chat. 00:14:54.000 --> 00:15:03.000 The boundaries are not black and white, so it would be a natural point of collaboration between people studying seduction zones. 00:15:03.000 --> 00:15:13.000 And this more transform focused center to collaborate because the systems should interact. 00:15:13.000 --> 00:15:23.000 Yeah, and I also want to add to this. So we have always had global activities actually, it was never limited to the focus nature [indiscernible] 00:15:23.000 --> 00:15:30.000 but there are clear differences, the tectonic differences, you know, between the transform and the subduction. 00:15:30.000 --> 00:15:40.000 So we will study this as well. We study that as well, but it will be focused for another center 00:15:40.000 --> 00:15:42.000 Probably. 00:15:42.000 --> 00:15:44.000 Okay, perfect. Thank you. 00:15:44.000 --> 00:15:50.000 Thank you. So other questions or comments. 00:15:50.000 --> 00:15:58.000 So there's one from Alex Hatem. How will the seismic hazard stress overlap with USGS efforts? Collaborative? 00:15:58.000 --> 00:16:04.000 Or with SCEC? Or will SCEC lead a more research oriented effort. 00:16:04.000 --> 00:16:06.000 Hear that. You want to speak to that? 00:16:06.000 --> 00:16:11.000 Yeah, so, I will speak to this briefly. So the short answer is, Alex, we don't know. 00:16:11.000 --> 00:16:15.000 We have not worked out the details as we would like to collaborate. 00:16:15.000 --> 00:16:28.000 Of course, and working on all of the topics that I listed will have to be worked out by multiple [indiscernible] working on that. 00:16:28.000 --> 00:16:37.000 We will decide together what is the best approach to address any or any and all of those topics. 00:16:37.000 --> 00:16:43.000 So I'm not giving you a direct answer, because there is no clear plan, it is 00:16:43.000 --> 00:16:53.000 all flexible and we will need to work on all of this together. 00:16:53.000 --> 00:16:56.000 Cool. Thank you. 00:16:56.000 --> 00:17:02.000 So Mike Oskin points out in the chat, Mike, you want to unmute and make your point. 00:17:02.000 --> 00:17:13.000 Sure. I just wanted to add on to Yuhuda's point about collaborating at the boundary with the subduction system, that the Triple Junction region is an important area for understanding how the transform plate boundary evolves. 00:17:13.000 --> 00:17:16.000 It's essentially the template for what's happening further to 00:17:16.000 --> 00:17:21.000 the southeast. 00:17:21.000 --> 00:17:22.000 Yeah, right. 00:17:22.000 --> 00:17:27.000 Thanks. 00:17:27.000 --> 00:17:31.000 There was a question in the chat if the proposal was already submitted. 00:17:31.000 --> 00:17:37.000 The answer to this is yes. It was submitted in April 2022, 00:17:37.000 --> 00:17:41.000 This is Zia Zafir which I had a question. So already submitted, 00:17:41.000 --> 00:17:49.000 which means you already have submitted what you are planning to do, so, you know what you're gonna do, are you going to lead, you're gonna collaborate, 00:17:49.000 --> 00:17:53.000 So all of those things are already submitted. 00:17:53.000 --> 00:17:58.000 So I think you have a good answer right now. Right? You can answer those. 00:17:58.000 --> 00:18:04.000 No, we submitted for propose research, proposed studies, but we did not, 00:18:04.000 --> 00:18:11.000 and we have been collaborative from the beginning, and we plan to continue to be collaborative. 00:18:11.000 --> 00:18:17.000 But we did not work out all the details of the collaboration ahead of time, so... 00:18:17.000 --> 00:18:18.000 And let me just add that we, you know there are limitations in the USGS 00:18:18.000 --> 00:18:29.000 ability to be involved in proposal development for another Federal agency. 00:18:29.000 --> 00:18:34.000 So we had to work around those. 00:18:34.000 --> 00:18:38.000 Okay. 00:18:38.000 --> 00:18:52.000 But we certainly wish to collaborate simply from the fact that the problems are very difficult, very deep, and we need, you know, the best available data, the best available people. 00:18:52.000 --> 00:19:05.000 Oh, precisely, we will collaborate remains to be worked out and we will be open to suggestion of productive mode of collaborations. 00:19:05.000 --> 00:19:16.000 Right, and I would add that people from the USGS have been deeply, and including people from northern California who are on this call have been deeply involved in SCEC leadership, and we imagine that will continue. 00:19:16.000 --> 00:19:19.000 We hope it will continue. 00:19:19.000 --> 00:19:26.000 Well, I personally think these questions about you know how should this collaboration work, and what should the plan be going forward 00:19:26.000 --> 00:19:36.000 is the single, most important question, and it is on our agenda as the last item on the agenda, so perhaps you should move along so that we have the proper time to answer that important question. 00:19:36.000 --> 00:19:37.000 Fair enough. 00:19:37.000 --> 00:19:44.000 So to that end can I bring up Laurent Montesi and Scott Marshall to talk a little bit about the SCEC community models. 00:19:44.000 --> 00:19:57.000 Sure. Let me share my screen and get stuff going if I'm allowed, I am allowed. Yay. 00:19:57.000 --> 00:20:00.000 Okay, that's not the first side. That's the first side. 00:20:00.000 --> 00:20:06.000 Okay, so what everybody can hear and see, right? Yeah. Okay, good. 00:20:06.000 --> 00:20:07.000 So what Lauren and I are gonna do is we're going to cover, 00:20:07.000 --> 00:20:14.000 Give you an overview of what the community models have currently and a little bit on some of the methodologies. 00:20:14.000 --> 00:20:28.000 But keep in mind that as co-leaders of the community modeling group, our task is basically to assist these different developers with, you know, disseminating their models and getting them out to the public, so we don't actually build all of these ourselves. So first thing, 00:20:28.000 --> 00:20:29.000 you should know is all the community models fall under the CXM. 00:20:29.000 --> 00:20:36.000 umbrella where X means anything, community fill in the blank model, right? 00:20:36.000 --> 00:20:46.000 So these are all available at the URL, right there. The SCEC.org/research/CXM, and you can click on them and browse and look through all this stuff. 00:20:46.000 --> 00:20:58.000 I mentioned the umbrella. Here's the literal incantation of that, if you will. Under the CXM umbrella, we have a fault model, a geodetic model, a rheology, model, a paleoseismic database which breaks the pattern but 00:20:58.000 --> 00:21:02.000 that's okay. A stress model, a thermal model, a velocity model, 00:21:02.000 --> 00:21:05.000 And then a software platform that integrates lots of velocity models. 00:21:05.000 --> 00:21:10.000 So we have a lot of different community models. All of them are at various stages of development. 00:21:10.000 --> 00:21:15.000 Some of them have been around for almost 20 years. Almost? 00:21:15.000 --> 00:21:18.000 They've been around almost 35 years. 00:21:18.000 --> 00:21:23.000 So, I'm gonna start with the community fault model first, and I'm gonna talk about some of these, and then Laurent will take over and he'll discuss some of them. 00:21:23.000 --> 00:21:25.000 Just so we're on the same page. The community fault model is different from UCERF, 00:21:25.000 --> 00:21:34.000 it's different from the USGS Q faults, and not plotted here. 00:21:34.000 --> 00:21:36.000 It's different from the national seismic hazard 00:21:36.000 --> 00:21:40.000 update the 2023 model. It's different from all of those things. 00:21:40.000 --> 00:21:47.000 The community fault model is fully 3 dimensional. It's as complex as the geologic and geophysical data dictate. 00:21:47.000 --> 00:21:53.000 So when you see a simple model in the community fault model, you know that that fault has very little data constraining it. 00:21:53.000 --> 00:22:02.000 But these models are all different, but in some ways they all are sort of based on each other, and are built off of each other. 00:22:02.000 --> 00:22:06.000 So default model right now is in version 5.3, although version 6.0 is finished. 00:22:06.000 --> 00:22:09.000 We're just working on some of the IT stuff to get it released. 00:22:09.000 --> 00:22:17.000 That is the result of a community evaluation for something like 30 SCEC participants evaluated the model and the highest ranked 00:22:17.000 --> 00:22:21.000 faults became CFM 6. This is the figure at 00:22:21.000 --> 00:22:22.000 the right over here shows you what the CFM looks like in 3D. 00:22:22.000 --> 00:22:32.000 However, you have to have specialized software to see this. So what we did is we developed a set of web-based tools that allows the user to browse a map-based interface 00:22:32.000 --> 00:22:38.000 that's interactive sort of like a Google map and you can browse and download the model individually, 00:22:38.000 --> 00:22:48.000 you can see seismicity, or you can actually click on individuals fault and hit the plot option and actually look at the faults in 3D without having any specialized software. 00:22:48.000 --> 00:22:49.000 So that's pretty cool. I would encourage you to go see the CFM 00:22:49.000 --> 00:22:57.000 homepage and then there is a link to the the download and viewer tools. 00:22:57.000 --> 00:23:00.000 I think one of the topics we wanted to talk about is how these models are built. 00:23:00.000 --> 00:23:02.000 So I figured I would talk about the CFM. 00:23:02.000 --> 00:23:07.000 The CFM is not built using a uniform methodology because the data is not uniform. 00:23:07.000 --> 00:23:11.000 The data that we have for constructing fault models is extremely diverse. 00:23:11.000 --> 00:23:16.000 So, for example, this is the Ventura fault model, Pitas Point fault from Judith Hubbard and others. 00:23:16.000 --> 00:23:17.000 paper that fault geometry is based on several cross sections of seismic reflection data, oil industry, 00:23:17.000 --> 00:23:29.000 well data, stratigraphy, surface trace locations, field mapping, everything! 00:23:29.000 --> 00:23:30.000 And we put this into three-dimensional CAD software. 00:23:30.000 --> 00:23:36.000 And we basically fit a three-dimensional surface to the data. 00:23:36.000 --> 00:23:39.000 So this kind of method is how most of the faults in the CFM 00:23:39.000 --> 00:23:50.000 are built. Unfortunately, when you build faults by hand like this, it's not possible to automate, and you can't put an error bar on these things because a lot of these things are interpretational. 00:23:50.000 --> 00:23:56.000 But I think this is still a good way to build fault services, because it integrates the most different kinds of data. 00:23:56.000 --> 00:24:09.000 There are cases where we can automate this at least semi-automate it Reisner et al. wrote a paper in 2017, where they showed, if you have a system that has a lot of seismicity and focal mechanisms, there is a semiautomated method for 00:24:09.000 --> 00:24:20.000 producing faults. So the Ridgecrest sequence was used to develop the community fault model representation of the Ridgecrest fault system, which is quite complicated, and this is objective 00:24:20.000 --> 00:24:24.000 and it's based on a semi-automated method. 00:24:24.000 --> 00:24:28.000 However, to use this method, you need a lot of seismicity and a lot of focal mechanisms. 00:24:28.000 --> 00:24:35.000 So this kind of method really is only possible in areas with a lot of seismicity. 00:24:35.000 --> 00:24:40.000 We also have Community Velocity Models, I think the official Community Velocity Models are the CVM-H, 00:24:40.000 --> 00:24:44.000 for Harvard and the CVM-S4.26. 00:24:44.000 --> 00:24:48.000 But there are actually a lot of really smaller scale models with high resolution 00:24:48.000 --> 00:24:55.000 if you see the basins over here, and here's sort of a map figure showing the different CVM and where they are spatially. 00:24:55.000 --> 00:25:02.000 So what the UCVM platform does is it allows a user to integrate all of these models into a single velocity model, using various smoothing platforms. 00:25:02.000 --> 00:25:13.000 And so the velocity models are available through the SCEC web pages. And so is the UCVM software platform, which is very powerful. 00:25:13.000 --> 00:25:14.000 We have a new Paleoseismic database coming this year. 00:25:14.000 --> 00:25:22.000 It's getting close to being done. The source data for it is basically the National Seismic Hazard Map update in 2023. 00:25:22.000 --> 00:25:23.000 So Alex, you heard asking some questions. So it's myself, Akciz, 00:25:23.000 --> 00:25:33.000 and Alex Hatem actually sorry I said his name wrong, Alex and we're putting this together. 00:25:33.000 --> 00:25:37.000 And the idea is basically to have as new studies come out, we want a place for this to live. 00:25:37.000 --> 00:25:40.000 So it's interactive and new studies can be added to it easily. 00:25:40.000 --> 00:25:46.000 So this will contain slip rate estimates and event chronologies. 00:25:46.000 --> 00:25:47.000 We also have a geodetic model. Here's a box showing basically what the geodetic model covers. 00:25:47.000 --> 00:25:56.000 Although Mike Floyd tells me that for GNSS data, I think basically all of California is not covered. 00:25:56.000 --> 00:26:04.000 That's mostly automated. But the cool thing about the geodetic model is the integration of InSAR for version 2 that's planned to be released this year 00:26:04.000 --> 00:26:16.000 I think. The group basically took a bunch of different InSAR tracks from sentinel data and used all kinds of different methods and then sort of average them. 00:26:16.000 --> 00:26:20.000 And what they found is this dramatically decreased the errors and made the data much more robust. 00:26:20.000 --> 00:26:22.000 So CGM-2 is really quite the thing. And I think when it comes out everybody will be happy to use it. 00:26:22.000 --> 00:26:34.000 The GNSS data, though, there's several different processing centers that where you can get GNSS data for basically all of California. 00:26:34.000 --> 00:26:41.000 CGM, has a web page on that web page we try to do with most of our SCEC products is package them in a DOI, 00:26:41.000 --> 00:26:42.000 So it's citable. So we're typically using Zenodo. 00:26:42.000 --> 00:26:46.000 So version 1 of CGM is available through Zenodo and so is the Community Fault Model. 00:26:46.000 --> 00:26:52.000 And so is the Rheology Model. So if you go to these home pages you'll see them there 00:26:52.000 --> 00:26:53.000 please reference them, it's our way to sort of give people credit for putting the time into the developing 00:26:53.000 --> 00:27:04.000 these models. Okay, Laurent's gonna take over and you can share your screen, or I can just switch the slide when you say next. If you want. 00:27:04.000 --> 00:27:07.000 I just ask you to switch the slide. I think it's gonna be faster that way. 00:27:07.000 --> 00:27:08.000 Okay. 00:27:08.000 --> 00:27:17.000 Yeah. I'm Laurent Montessi from the University of Maryland, and I will be talking to you a little bit about the various components of the CXM that have maybe a little bit further from data, and now more like interpretation and kind of models in that sense. 00:27:17.000 --> 00:27:24.000 And the first one probably is the most complex of models we have, which is the CSM, that's stands for the Community Stress 00:27:24.000 --> 00:27:46.000 Model. And this one is actually a collection of various estimates of the stress field or the stressing rate field in California, in southern California, you see, we have 6 stress models, 7 stress same rate models. 00:27:46.000 --> 00:27:51.000 All of them are based on different data sets and different assumptions, on different theories, etc. 00:27:51.000 --> 00:27:55.000 They are just been put together on a consistent format and grid, and are all available individually on the SCEC websites that have been mentioned earlier. 00:27:55.000 --> 00:28:06.000 What I'm showing there essentially is an average of those various models of auto replaces 00:28:06.000 --> 00:28:13.000 if you want to have that kind of one vision of those various components, 00:28:13.000 --> 00:28:18.000 all of them are available through the SCEC website. Next, please. 00:28:18.000 --> 00:28:21.000 Then, we're gonna move towards what we call the Community of Rheology Model, and that one is essentially built on the top of several models. 00:28:21.000 --> 00:28:25.000 Like, I said, we're going further and further in the abstraction level. 00:28:25.000 --> 00:28:35.000 One of the things that the CM is built on is what we call a geological framework. 00:28:35.000 --> 00:28:36.000 Sometimes it's like GFM, the one that doesn't actually fit the actual templates. 00:28:36.000 --> 00:28:53.000 As you can see, the way it is done there is that it's a structural representation of the structure of southern California, where they have been 23 lethal tectonic provinces that have been defined over the region. For each one of them 00:28:53.000 --> 00:28:57.000 we have a column stratographic column through the entire lithosphere, 00:28:57.000 --> 00:29:12.000 so it's a fairly natural, you know, very broad scope type of story we're not looking at the details of the stratigraphy, as has been mentioned earlier today, but in our case we're looking at the broad, the strategies at various level of the crust and in the 00:29:12.000 --> 00:29:29.000 mantle. You can see there to the kind of columns for some of those models, some of those blocks and the colors correspond to certain rock types or the lithology which are defined just by assemblage of minerals. Next please. The next element 00:29:29.000 --> 00:29:39.000 there is the Community Thermal Model which has been developed largely by Wayne Thatcher and Vic Chapman, which is very similar to this geological framework, 00:29:39.000 --> 00:29:51.000 but thinking about temperature structures. So they define 14 regions of constant heat flow, and for each one of them they have developed a geotherm that is kind of a representation of that place. 00:29:51.000 --> 00:30:04.000 Most of them are at steady state when a few of cases there are some transient thermal processes that are included. We actually distributed an alternative version that was developed by Billy Shinevar, which is more an interpolation is not really block based, 00:30:04.000 --> 00:30:10.000 like the CTM is, but, on the other hand, only includes thirty-state profiles, 00:30:10.000 --> 00:30:15.000 so it may be less realistic in some locations, but smoother throughout the entire region. 00:30:15.000 --> 00:30:20.000 Next, please. And on top of those two you can see... the next slide 00:30:20.000 --> 00:30:25.000 please. 00:30:25.000 --> 00:30:32.000 So on top of those two we have what we call the CRM, the Community of Rheology Model. 00:30:32.000 --> 00:30:37.000 At this stage, it's essentially providing an estimates of the ductile flow, 00:30:37.000 --> 00:30:55.000 the ductile rheology at each place over the study area. The ductile rheology is obtained by mixing mineral specific lab-derived flows, for various minerals, and they are assembled to make the rock types that are forming the geological framework and if you take the temperature from the 00:30:55.000 --> 00:30:57.000 CTM and a strain rate or stress estimate 00:30:57.000 --> 00:31:07.000 then you could calculate viscosities. So we have there, for example, views of the rheologies, temperatures, and discussed here are different 00:31:07.000 --> 00:31:09.000 depths 2424 meters. 00:31:09.000 --> 00:31:12.000 The parts on the right kind of shows blocks that have similar properties. 00:31:12.000 --> 00:31:16.000 If you start to think about share localization and things like that. 00:31:16.000 --> 00:31:21.000 Next, please. 00:31:21.000 --> 00:31:22.000 [noise] I think there's a delay. 00:31:22.000 --> 00:31:34.000 Now some of the plans that we are some of the current developments are include reconciling some of the definitions we've heard, for example, earlier on northern California, how they have some different instance of comparison about velocity and a structural models. 00:31:34.000 --> 00:31:40.000 We are also looking here other various blocks and definitions. 00:31:40.000 --> 00:31:43.000 CTM and so GFM can we become by them. 05:50:42.000 --> 05:51:01.00 We can start to use some of this rheology to compare the predictions in viscosity from those rheology models with information such as processing deformations that the 4 panels you see on the right. And kind of use this to develop a new stress model that could be added or 00:32:02.000 --> 00:32:05.000 combined, or evaluated against the CSM 00:32:05.000 --> 00:32:18.000 that I've mentioned earlier. Next. I think this is my last slide, which kind of emphasizes something that Scott already mentioned earlier, is that most of those models are available, not just the data set itself, but also to some viewers you mentioned. 00:32:18.000 --> 00:32:32.000 The CFM one default models is our snapshots of 2 other viewers on the left for that geological framework that I mentioned. 00:32:32.000 --> 00:32:37.000 There is a blog that is highlighted here, and you also have like two little markers and of course you cannot see it. 00:32:37.000 --> 00:32:39.000 But at the bottom of the screen cannot read it, but at the bottom of the screen you have, which rock type is there? What temperature? 00:32:39.000 --> 00:32:47.000 Web velocity, etc. On the right we have a snapshot of the community 00:32:47.000 --> 00:32:50.000 the modern velocity viewers which can be using 3D. 00:32:50.000 --> 00:32:53.000 You can make cross-sections, and all of those things. 00:32:53.000 --> 00:32:56.000 So we've actually worked on trying to make the model easily accessible. 00:32:56.000 --> 00:32:57.000 And today can be acquired also through those interfaces, so that they can be reaching out to the [indiscernible] 00:32:57.000 --> 00:33:10.000 populations, in range of scientists as possible. I think that's what the last slide. 00:33:10.000 --> 00:33:13.000 Thank you. 00:33:13.000 --> 00:33:14.000 Thanks, Laurent. 00:33:14.000 --> 00:33:20.000 Thank you. So I think we should go straight on into Keith with the help of Roland, was pretty neat 00:33:20.000 --> 00:33:29.000 the big picture how all the models across the State look together! So maybe we should just go straight to that, Keith, if you want to take it away. 00:33:29.000 --> 00:33:43.000 Okay, I'm not great at sharing screen in zoom, but I'm gonna do that right here. 00:33:43.000 --> 00:33:48.000 So you should see a table. Is that correct? Okay, great. Okay. 00:33:48.000 --> 00:33:53.000 So we're heading into the less formal part of this. 00:33:53.000 --> 00:34:02.000 And hopefully this, I put together with a number of others, will help trigger discussion on the theme that was really in the chat. 00:34:02.000 --> 00:34:12.000 As the SCEC representatives were speaking, and that question is, well, how are we going to integrate or unify or work together 00:34:12.000 --> 00:34:26.000 essentially with the existing models in the North, and the existing miles in the South, and so 24 hours ago, I started this table and shared it with a number of others who have collaborated on trying to populate it. 00:34:26.000 --> 00:34:36.000 And it seems like we need something like this to understand what's going on in the opposite part of the state from each of us. 00:34:36.000 --> 00:34:42.000 So, we could do a variety of things 00:34:42.000 --> 00:34:50.000 now we could spend time just trying to populate this. I think that might be a little bit awkward, although people could certainly put things in the chat. 00:34:50.000 --> 00:34:58.000 The first question on the agenda, after my little slot on the agenda was about continuity of methods between North-South Community models. 00:34:58.000 --> 00:35:10.000 So if we wanted to focus on a discussion topic, we could try to populate that column in this table. 00:35:10.000 --> 00:35:26.000 I'm willing to serve as a scribe, and I'm sure others who have access to this Google sheet would be willing to do the same if people want to put things into the chat about suggestions for what goes in different boxes and again, what's in this 00:35:26.000 --> 00:35:35.000 is very fresh, and certainly needs lots of eyes looking at it, and and the corrections made. 00:35:35.000 --> 00:35:43.000 But I thought this would be a start to the, to the framework of the question, well, how are we gonna have integrated models and can we have integrated models? 00:35:43.000 --> 00:35:46.000 And what happens at the boundaries? And that sort of thing, and in fact, 00:35:46.000 --> 00:35:52.000 what do we have in the North and in the South? So got a variety of different, 00:35:52.000 --> 00:36:02.000 we start off with models, but I realized, as we were proceeding through this, over the little bit, we started including databases, too. 00:36:02.000 --> 00:36:07.000 So there are 3D velocity models or faults. In the North, 00:36:07.000 --> 00:36:13.000 I think we have done a lot to try to characterize creeping faults, and we have data around those. 00:36:13.000 --> 00:36:15.000 There are a variety of different seismicity databases essentially. 00:36:15.000 --> 00:36:24.000 There are, I think, multiple databases of repeating earthquakes in the North. 00:36:24.000 --> 00:36:39.000 We just heard a little bit about the the paleoseismic and slip database it was developed for the National Hazard Model, and sounds like that's already being turned into a SCEC CXM model. 00:36:39.000 --> 00:36:44.000 We heard about the rheology model, which I don't think we have an equivalent to in the North. 00:36:44.000 --> 00:36:50.000 I think we both have geology models. In the South, 00:36:50.000 --> 00:37:01.000 there are rapid fault association being made. I think there are a variety of different crustal velocity and geodetic models in the North and the South. 00:37:01.000 --> 00:37:06.000 Ones I tend to think about are the ones that the national side of the hazard model relies on for 00:37:06.000 --> 00:37:12.000 It's work in both areas. We have basin sets. 00:37:12.000 --> 00:37:13.000 We just heard about the stress model 00:37:13.000 --> 00:37:23.000 in southern California. I'm not aware of an equivalent model in the North, not aware of a thermal model in the North. 00:37:23.000 --> 00:37:35.000 I think that SCEC has with all these computer science and IT professionals has developed great sets of tools for visualization and querying. 00:37:35.000 --> 00:37:42.000 And we don't have that really in the North. The geologic slip rate, compilations. I think 00:37:42.000 --> 00:37:48.000 Alex added this in and I'm not sure who added the geodetic deformation models. 00:37:48.000 --> 00:37:54.000 There are a range of them, but I think we need to operationalize them. 00:37:54.000 --> 00:38:00.000 And then one area where southern Californians are really ahead of us is with hats. 00:38:00.000 --> 00:38:03.000 Well, I think we need to come up to speed on that 00:38:03.000 --> 00:38:09.000 category. So, 00:38:09.000 --> 00:38:14.000 well, it really depends on what people want to do. 00:38:14.000 --> 00:38:18.000 We could try to correct errors in this, we could populate it. 00:38:18.000 --> 00:38:21.000 We could talk about the continuity question. 00:38:21.000 --> 00:38:27.000 This is my contribution to this meeting. 00:38:27.000 --> 00:38:33.000 So, Keith can you clarify the continuity you said continuity of methods? 00:38:33.000 --> 00:38:40.000 Is this really what you mean? Continuity of methods is this what you really mean? or continuities of data, sets of models? 00:38:40.000 --> 00:38:43.000 That's a great question, I guess it's probably both, Yahuda. 00:38:43.000 --> 00:38:47.000 I think, yeah, I think it's both. 00:38:47.000 --> 00:38:49.000 So. 00:38:49.000 --> 00:38:56.000 But, the obvious question is, there are boundary faults between the models in the north and the models in the south 00:38:56.000 --> 00:39:01.000 in most cases, and is that due to different methodologies? 00:39:01.000 --> 00:39:06.000 Is it due to different platforms? Is it simply resolved by just shifting 00:39:06.000 --> 00:39:11.000 geologic boundaries on a geologic map, for example. 00:39:11.000 --> 00:39:13.000 So I'll just say, I see Davis is in. 00:39:13.000 --> 00:39:25.000 I'll just make one quick comment. I won't solve the problem right now, but it's also in part result with the results of different data coverage actually, which would produce these natural boundaries. 00:39:25.000 --> 00:39:30.000 And I think one of the topics we would like to work on together. 00:39:30.000 --> 00:39:50.000 We should work on is first improving locally at each place the existing models, led by the people working on this before, so this should continue, and then we should address the continuity problem by developing methodology to merge models. 00:39:50.000 --> 00:39:57.000 That is, eliminate the boundaries. That's what I mean by merging, but this is quite complicated, and probably will require machine learning techniques. 00:39:57.000 --> 00:40:10.000 So I will not continue. But it's a great research topic, actually. 00:40:10.000 --> 00:40:18.000 And, then I want to say that all of the models in different regions and the merge models must be validated. 00:40:18.000 --> 00:40:22.000 So we need to validate. We need to develop this iterative loop of developing improving models. 00:40:22.000 --> 00:40:29.000 Locally validating, locally merging and validating, you know, etc. 00:40:29.000 --> 00:40:32.000 It's a spiral. Hopefully. 00:40:32.000 --> 00:40:37.000 Yeah, merging models ain't that easy, right? 00:40:37.000 --> 00:40:38.000 Even just with 3D velocity models, 00:40:38.000 --> 00:40:52.000 There are atomographic models, and their are geology-based models. 00:40:53.000 --> 00:40:55.000 3D-geology based models, and even in northern California we have parallel sets of those two kinds of models, and getting those merged is a challenge, so merging what is primarily a 3D 00:40:55.000 --> 00:41:05.000 geology-based model in the North with what is primarily a tomographic base model in the South is... 00:41:05.000 --> 00:41:06.000 Yeah, I'm not sure you can just merge those. 00:41:06.000 --> 00:41:15.000 So we need to develop techniques to link, to test, and rank models based on their performance. 00:41:15.000 --> 00:41:19.000 All models are developed with best intentions, 00:41:19.000 --> 00:41:24.000 but how do we know that they actually perform? We need to validate. 00:41:24.000 --> 00:41:29.000 Yeah, and test, it also would apply to the catalogs. 00:41:29.000 --> 00:41:33.000 There are many, many catalogues now for the [indiscernible], for example, 00:41:33.000 --> 00:41:36.000 Greg talked a little bit about this. How do we know, so we don't know. 00:41:36.000 --> 00:41:42.000 We don't. These are our open questions, but I think it should be part of the research platform to test, validate. 00:41:42.000 --> 00:41:53.000 compare, rank and then make decisions. You know what to keep and and how to merge. 00:41:53.000 --> 00:41:55.000 Yeah, it's a good, 00:41:55.000 --> 00:41:58.000 it's a rich question, it's an important question, not just for for SCEC, not just for California, but globally 00:41:58.000 --> 00:42:04.000 there are people working on merging models. So we we have a couple of hands up. 00:42:04.000 --> 00:42:08.000 I'd like to call on Dave Jackson. 00:42:08.000 --> 00:42:09.000 Yes, thanks. Can you hear me? 00:42:09.000 --> 00:42:12.000 Yes, we can hear you. 00:42:12.000 --> 00:42:29.000 Okay. Good. I think it's true that by, I think all of our agreement that the real target for all of this work is a scientific approach to estimating seismic hazard and risk. 00:42:29.000 --> 00:42:44.000 And so turning all these things into a kind of earthquake forecast, it's on different time scale, and you know some different geography are really important, and I think it's well worth looking at 00:42:44.000 --> 00:43:01.000 what happens in the UCERF3, and you know what's now becoming an extension of that, because they made a very concerted effort with a lot of talent 00:43:01.000 --> 00:43:10.000 to accomplish this job of turning, you know the kind of information we have about earthquakes and stress and everything else 00:43:10.000 --> 00:43:31.000 into earthquake forecast, and they've done a lot to try to bridge the boundaries between northern and southern California. So, I think it's kind of a review of how it worked and what have been the real successes and what have been some of the problems. 00:43:31.000 --> 00:43:40.000 And I think some of the problems are complexity things that took a lot of duct tape to paste together. 00:43:40.000 --> 00:43:50.000 But nevertheless it's, I think, a really monumental prospect, and certainly, you know, we've heard the UCERF3 00:43:50.000 --> 00:43:54.000 results quoted many, many times already in the last two days. 00:43:54.000 --> 00:44:00.000 So that's my recommendation to spend some time looking at how that worked. 00:44:00.000 --> 00:44:11.000 Like I can comment on that. The UCERF3, the fault model used in UCERF3 is lacking in northern California. 00:44:11.000 --> 00:44:19.000 It started with the CFM, and that was migrated into a state FM, 00:44:19.000 --> 00:44:25.000 and then northern California, or I've talked to Andreas and others about this at nauseam 00:44:25.000 --> 00:44:26.000 really. The same level of attention needs to be devoted to the fault model 00:44:26.000 --> 00:44:33.000 in northern California. 00:44:33.000 --> 00:44:39.000 Yeah, so, in the chat Russ Graymer says the point that we should use the information, emphasize the information that's most reliable, 00:44:39.000 --> 00:45:00.000 where it's most reliable. So you know, if we had the model and their uncertainties, we can sort of combine information in the Bayesian framework, or something like that, either formally or informally. Scott I think you've had your hand up for a while. 00:45:00.000 --> 00:45:07.000 Yeah, thanks. So, SCEC is an interesting organization, right? Because it's investigator driven. 06:04:06.000 --> 06:04:17.00s So a lot of these things that are on the spreadsheet in front of us that are, you know, termed southern California, may actually have proposals in for all of California. 00:45:18.000 --> 00:45:24.000 I know, for example, I have a proposal in that involves expanding one of the community models to all of California. 00:45:24.000 --> 00:45:28.000 Actually two, but those are all pending right now, so we don't know. 00:45:28.000 --> 00:45:29.000 But with the last year, the bridge funding at SCEC, we had some flexibility to look to the future and expand. 00:45:29.000 --> 00:45:43.000 So at least a few of the CXM models in California have pending proposals to expand statewide. And Keith, 00:45:43.000 --> 00:45:49.000 we do have a statewide Community Fault Model, that is included in our proposal to actually formalize it. 00:45:49.000 --> 00:46:04.000 And the the tricky part of the Community Fault Model is, once you build the fault, then you have to have all of the data structure stuff like all the metadata, all the everything has to be tidy and neat, or else none of the tools work so the model itself is actually already built through 00:46:04.000 --> 00:46:12.000 collaboration with this group and others, but formalizing it and getting into a product that's publicly released is gonna take the effort 00:46:12.000 --> 00:46:15.000 and that's what the proposal focuses on. 00:46:15.000 --> 00:46:26.000 Well good luck with your proposals, Scott. If you need help identifying people to collaborate with in northern California, I'm sure there are many people in this meeting that would jump at the chance. 00:46:26.000 --> 00:46:31.000 Yeah, we have routine, the Community Fault Model group has probably a couple of times a year, 00:46:31.000 --> 00:46:40.000 we meet with the UCERF group, people that we think are concerned, and some people have reached out to us to encourage meetings, and we've done that. 00:46:40.000 --> 00:46:45.000 So we know that there will be, that whatever we release for the version, one of northern California will just be a starting point, you know that the first thing about a Community Model is just get it out there 00:46:45.000 --> 00:46:55.000 So people can improve it. No Community Model is ever finished. 00:46:55.000 --> 00:47:17.000 Yeah, so that's an important point. And another important point you brought up was the investigator 00:47:18.000 --> 00:47:19.000 driven nature of the SCEC enterprise. And so all these things that we're contemplating are gonna require champions like Scott and Laurent for the Community Models to do the work and carry them forward. Laurent. 00:47:19.000 --> 00:47:25.000 Like in the chat, for example, Hal was mentioning an attenuation model, why don't we have one? 00:47:25.000 --> 00:47:31.000 Maybe because nobody wanted or proposed to do one at this stage, and there's something that can be added. 00:47:31.000 --> 00:47:35.000 Also, I want to point out that a lot of those models we described have alternative versions embedded into them. 00:47:35.000 --> 00:47:43.000 For example, you know the velocity models have some of them of the basins, some of them don't. 00:47:43.000 --> 00:47:50.000 There's several variations. And I think that's important, because I think you mentioned that [indiscernible] spaces 00:47:50.000 --> 00:47:53.000 you will have better constraint for one approach compared to another approach. 00:47:53.000 --> 00:48:00.000 Whatever model you're looking at, and that's exactly why we have an alternative realization of a lot of those models. 00:48:00.000 --> 00:48:02.000 And so that's brings up the idea that we try to be flexible. 00:48:02.000 --> 00:48:10.000 And when we say Community Model, we don't try to get like one answer for everybody, at least less and less 00:48:10.000 --> 00:48:19.000 we try to get to that one. And so for everybody, but more like the collection of answers that should be helpful for everybody. 00:48:19.000 --> 00:48:22.000 Yeah. And then that probably better represents a true state of knowledge. 00:48:22.000 --> 00:48:27.000 So we have. Ian Pierce. 00:48:27.000 --> 00:48:28.000 Yup! 00:48:28.000 --> 00:48:30.000 Hey? Can you hear me? Okay, it's so. My name is Ian Pierce. 00:48:30.000 --> 00:48:33.000 I'm currently a postdoc in Oxford, in the UK. 00:48:33.000 --> 00:48:36.000 I'm in Santa Cruz right now, actually. 00:48:36.000 --> 00:48:53.000 I wanted to say, I think this is a really good opportunity with SCEC expanding to northern California, to kind and this is based off some conversations with other from the paleoseismology and an earthquake geology community and building off of what Chris Madrugo brought up 00:48:53.000 --> 00:49:10.000 in his talk yesterday about, you know there's some some data priorities for specific faults, but I think one of the things from that paleoseismology perspective would be to try to organize maybe a workshop early on in this new SCEC to identify exactly what priorities we have 00:49:10.000 --> 00:49:19.000 for either, you know, updating work that's been done with new methods, or where we have big data holes and seeing what those priorities would be. 00:49:19.000 --> 00:49:24.000 Because, you know, looking at the impressive, really impressive amount of work that's been done all over the State. 00:49:24.000 --> 00:49:35.000 But in southern California we know a lot more about a lot of faults than we do in northern California, and I think that's a good opportunity to expand that. 00:49:35.000 --> 00:49:37.000 Yeah, that's a good point, I think, 00:49:37.000 --> 00:49:39.000 and Yuhuda can correct me if I'm wrong. 00:49:39.000 --> 00:49:48.000 I think there's an explicit effort in the proposal to fill observational data gaps as part of SCEC's strategy. 00:49:48.000 --> 00:49:50.000 So that would be one type of gap. 00:49:50.000 --> 00:49:51.000 Yeah. 00:49:51.000 --> 00:49:53.000 Yeah, I think that gets into the thrust objectives. 00:49:53.000 --> 00:49:57.000 So, yes. And let me just say that this table that we see the spreadsheet that we see, I mean, I fully agree with you. 00:49:57.000 --> 00:50:06.000 Yeah, we will need sets of workshops... 19:31:06.000 --> 20:07:41.000 Note: due to technical difficulties, there are no closed captions for the remainder of this session.