WEBVTT 00:00:01.000 --> 00:00:11.000 Hello, everyone! Welcome back! It is time for the final, final, final batch of Thunder Talks of the workshop. 00:00:11.000 --> 00:00:16.000 Does everybody feel the thunder? 00:00:16.000 --> 00:00:23.000 So, let's give it over to our amazing moderator, Grace Parker, to lead this round. 00:00:23.000 --> 00:00:33.000 Thanks, Sarah, and thanks, everyone, for tuning in to the last Thundertalk session. You probably know how this goes at this point, but if you don't, each talk will be 3 minutes. 00:00:33.000 --> 00:00:38.000 And there will be no breaks in between. They'll segue straight into the next speaker. 00:00:38.000 --> 00:00:44.000 We have, uh, 7 talks in this session, and so we're expecting to have 00:00:44.000 --> 00:00:57.000 Um, a decent amount of time for live Q&A afterwards, so of course, you can continue to use the chat to ask questions, but I would encourage you to maybe hold your questions till the end of the talks and ask them live. 00:00:57.000 --> 00:00:59.000 Uh, and so… 00:00:59.000 --> 00:01:02.000 With that, I don't see… 00:01:02.000 --> 00:01:25.000 There we go. Uh, I'll hand it over to Danielle to get us started. 00:01:25.000 --> 00:01:35.000 is Danielle in the audience? 00:01:35.000 --> 00:01:37.000 Danielle is not here! Is Rolla heal? 00:01:37.000 --> 00:01:38.000 Yeah. 00:01:38.000 --> 00:01:41.000 I don't… this is Roland. 00:01:41.000 --> 00:01:46.000 I can quickly explain, even though I haven't, um… 00:01:46.000 --> 00:01:48.000 prepared at all. Do you want me to do that? 00:01:48.000 --> 00:01:53.000 That would be fantastic. Thank you so much, Roland. 00:01:53.000 --> 00:01:57.000 Okay, so really quickly, um, Danny conducted this study 00:01:57.000 --> 00:02:04.000 of slow-moving landslides, uh, in the Northern California Mendocene Triple Junction area. 00:02:04.000 --> 00:02:12.000 in the aftermath of the 2022 Ferndale earthquake. Next slide. 00:02:12.000 --> 00:02:20.000 And so what she did is, using ALOS 2 INSA data, conduct a comprehensive analysis of 00:02:20.000 --> 00:02:24.000 really hundreds of landslides in the area, 00:02:24.000 --> 00:02:27.000 And consider changes in… 00:02:27.000 --> 00:02:33.000 landsliding raid as a function of time, particularly interested in 00:02:33.000 --> 00:02:44.000 how temporal changes due to wet seasons, as well as the occurrence of earthquakes, came into play. Next slide. 00:02:44.000 --> 00:02:53.000 And so, looking at that, she looked at, uh, landslides, creeping them into landslides that responded, um, 00:02:53.000 --> 00:02:58.000 in a much more traumatic fashion, uh, to these time-dependent… 00:02:58.000 --> 00:03:09.000 factors, so, uh, rainy weather and earthquake occurrence, and so she developed this classification to be able to see 00:03:09.000 --> 00:03:12.000 which groups of landslides showed, uh, 00:03:12.000 --> 00:03:15.000 what could be a significant response. 00:03:15.000 --> 00:03:22.000 Uh, to those environmental changes. Next slide. 00:03:22.000 --> 00:03:29.000 And so, it does appear that, um, events, uh, that were affected by, uh, 00:03:29.000 --> 00:03:37.000 higher shaking did show a more significant response, but this is challenged by rainfall 00:03:37.000 --> 00:03:44.000 Um, that also could bring the same winter season, you know, being temporarily highly correlated. 00:03:44.000 --> 00:03:54.000 So it's really by having hundreds of slides and being able to make kind of a statistical argument that there's some evidence that some, not at all, 00:03:54.000 --> 00:04:01.000 all of them, of these landslides did, uh, produce a response. 00:04:01.000 --> 00:04:04.000 Next slide. 00:04:04.000 --> 00:04:13.000 And so, there does seem to be an effect of this, and with additional analysis and with NICER 00:04:13.000 --> 00:04:19.000 coming along with more regular and frequent L-band acquisitions. 00:04:19.000 --> 00:04:26.000 Uh, there's probably a lot more of this kind of research to do, and I know that Danny, in a new position, 00:04:26.000 --> 00:04:29.000 at GNS, um… 00:04:29.000 --> 00:04:36.000 has a project that is targeting that very problem in New Zealand, and maybe comparing it with the results 00:04:36.000 --> 00:04:49.000 that she developed here in California. Thank you. 00:04:49.000 --> 00:04:55.000 Uh, thanks everyone. My name is Greg Crawford, I'm a prof at Ontario Tech University up here in Canada. 00:04:55.000 --> 00:05:02.000 Uh, used to work at Cal Poly Humboldt, Humboldt State University, and have continued to do, uh, collaborative work 00:05:02.000 --> 00:05:08.000 Uh, on, uh, some of the fluid flows, tsunami flows in Humboldt and Crescent City area. 00:05:08.000 --> 00:05:14.000 Uh, today I'm just going to talk about estimating tsunami currents by proxy at Humble Bay and Crescent City. Humboldt Bay, 00:05:14.000 --> 00:05:19.000 Uh, who cares, uh, Humboldt Bay is important because it's the only deep draft, uh, 00:05:19.000 --> 00:05:28.000 bay between San Francisco and Coos Bay. That's about 300 nautical miles, so it's an important place to hide out in weather. 00:05:28.000 --> 00:05:35.000 Uh, and Kristen City, as many of you know, get hammered by tsunamis fairly regularly, so there's interest in what goes on in these places. 00:05:35.000 --> 00:05:43.000 Notably, of course, they have… both places have tide gauge, that's where most of the historical observations have been made. 00:05:43.000 --> 00:05:45.000 Currents also turn out to be important. 00:05:45.000 --> 00:05:51.000 Um, and so, uh, here's the problem at these two locations. This is for the Kamchatka. 00:05:51.000 --> 00:05:55.000 a tsunami from last summer. 00:05:55.000 --> 00:06:00.000 There were no direct current measurements available, per se, because there's no monitoring of currents at Crescent City. 00:06:00.000 --> 00:06:06.000 In Humboldt Bay, they have a ports program. They're supposed to have current profilers. One was not working. 00:06:06.000 --> 00:06:19.000 And the other one, uh, there was a fuel barge in the way, and this happens actually fairly routinely. You can't just go grab data off the web and assume that it makes sense that the position of this fuel barge over here is problematic. 00:06:19.000 --> 00:06:23.000 Uh, when it's in the way, and of course it was there when the tsunami hit. 00:06:23.000 --> 00:06:29.000 So, um, uh, well, I've done some work with colleagues looking at the relationship between tsunami currents, 00:06:29.000 --> 00:06:37.000 And tsunami water level, that is removing the tides, just looking at the tsunami wave and the current associated with that wave. 00:06:37.000 --> 00:06:45.000 And in Humboldt Bay, this is data, the scatter plot shows data from 3 different tsunamis, all, uh, composited. 00:06:45.000 --> 00:07:00.000 Here, and there's a pretty good linear correlation. So, even though we couldn't get direct measurements of tsunami currents in Humboldt Bay, uh, we've worked out a little algorithm to calculate tsunami currents based on the tsunami water levels that we measured. 00:07:00.000 --> 00:07:06.000 And so that's a way that we have, we think now, and we think it's pretty robust, uh, 00:07:06.000 --> 00:07:15.000 Chris and Sidi, I'm going to say briefly, uh, is more problematic. It's not as simple a relationship between tsunami currents and tsunami water levels. 00:07:15.000 --> 00:07:22.000 Uh, we also have very little data from that. There was a three and a half hour, uh, collection of video footage from the Tohoku tsunami. 00:07:22.000 --> 00:07:31.000 Amanda admired it and analyzed that for her master's thesis way back when. And there was up to 5 meters a second currents from that video footage. 00:07:31.000 --> 00:07:44.000 Citizens Dock, which is where the tide gauge is. We put, uh, ADCP in for about 2 years. Um, we detected one small tsunami, so we have a bit of data. Not particularly great, but nevertheless, we've run out some, uh, 00:07:44.000 --> 00:07:51.000 Estimates of tsunami currents at the entrance to the small boat basin, and also at Citizens Dock. 00:07:51.000 --> 00:07:56.000 Um, and so we see on the order of 1.5% to 1.9, uh, 00:07:56.000 --> 00:08:02.000 maximum amplitude currents of the tsunami part of the wave, we including the tides right here. 00:08:02.000 --> 00:08:07.000 Um, uh, we've got more work that we're going to be doing here, uh, so just to wrap up, um, 00:08:07.000 --> 00:08:17.000 Uh, the relationships in Humboldt and at Crescent City, we're going to be correlating that with some tsunami models. Barack Uslu's done work both in both places before. He's dusted off his old models. 00:08:17.000 --> 00:08:22.000 We're rerunning for this as, uh, these old, uh, tsunamis as well as Kamchatka. 00:08:22.000 --> 00:08:29.000 Uh, we've also got a, uh, kind of like what Amanda had before, but much longer set of data. There's actually a security camera footage 00:08:29.000 --> 00:08:38.000 from, um, from the Harbormaster here, we got, uh, several days' worth of data, so we're looking at trying to analyze the security camera footage for it to estimate currents. 00:08:38.000 --> 00:08:49.000 Uh, in the entrance to the small boat basin, uh, and we're working on that, uh, kind of as we speak. The issue here, of course, is that the tsunami here arrived at about 1 o'clock in the morning. 00:08:49.000 --> 00:08:55.000 I can't see the currents from the camera when it's not daylight, so there are limitations. Nevertheless, it's data that 00:08:55.000 --> 00:09:14.000 that otherwise wouldn't be available. So, um, that's it from me. Thank you. 00:09:14.000 --> 00:09:22.000 Hi, everyone. My name is Elizabeth Grant, and I'm with the California Department of Water Resources Project Geology Group. 00:09:22.000 --> 00:09:28.000 And at Project Geology, we are interested in trying to characterize seismic hazard within the vicinity of dams. 00:09:28.000 --> 00:09:32.000 So in this case, I'll be focusing on Delvaldam, which is in Livermore, California. 00:09:32.000 --> 00:09:38.000 And the first step in characterizing seismic hazard is really to understand the bedrock geology and the structures in the region. 00:09:38.000 --> 00:09:44.000 So I'm gonna take us back to the basics a little bit. 00:09:44.000 --> 00:09:56.000 Delville Dam is about 5 miles south of the Livermore City Center, and just for some regional context, it sits between the Calaveras and Greenville, right lateral faults, highlighted in red on the insect map. 00:09:56.000 --> 00:10:02.000 And the local vaults in yellow include the Verona Williams Fault and the Las Pasitas Left Lateral Fault. 00:10:02.000 --> 00:10:07.000 The map on the right is the Stockton Quad 100K map from CGS. 00:10:07.000 --> 00:10:10.000 And as you can see, this is a relatively small-scale map, 00:10:10.000 --> 00:10:17.000 Which includes the major faults, but which cannot accommodate the more granular details of a large-scale map. 00:10:17.000 --> 00:10:23.000 So, on the left is a clip from the Huey 1948 geologic map. 00:10:23.000 --> 00:10:29.000 Um, sorry, I lost my swap. And this is a 62.5K map. 00:10:29.000 --> 00:10:33.000 So, uh, this is also a somewhat small-scale map. 00:10:33.000 --> 00:10:38.000 On the right is a basic geology map that DWR completed in 1963. 00:10:38.000 --> 00:10:44.000 And you can start to see more detailed mapping of units, contacts, and, um, some queried faults. 00:10:44.000 --> 00:10:48.000 And this is essentially what we had when we began dam excavation. 00:10:48.000 --> 00:10:58.000 The lower left is a photo of the dam site pre-excavation, and the lower right is a photo of the dam site during excavation, and they removed all the fluvial and alluvial deposits. 00:10:58.000 --> 00:11:01.000 and excavated down to the bedrock. 00:11:01.000 --> 00:11:10.000 And at this point, DWR mapped the footprint of the dam in great detail, but it didn't really extend beyond the footprint. 00:11:10.000 --> 00:11:15.000 So this is a 1965 geologic map. Um, this is post… 00:11:15.000 --> 00:11:23.000 uh… post-excavation, pre-construction, and I've juxtaposed two map sheets for the regional view. 00:11:23.000 --> 00:11:33.000 Um, and you can see, um, I think my slides are going wild here. Um, I don't know what's happening. I'm sorry. Um, you can see a number of shears indicated on these maps, and these are largely indicated through drilling. 00:11:33.000 --> 00:11:38.000 Um, and you can see in the inset, there are some shears that are actually projected shear traces. 00:11:38.000 --> 00:11:46.000 Um, and this is really our best available data. We know there are structures, but we don't really know the nature of the structure or the age. 00:11:46.000 --> 00:11:52.000 And there isn't a lot of obvious, uh, geomorphic evidence of these shears, because they're either at depth or they're under the dam. 00:11:52.000 --> 00:11:59.000 So, this leaves us with the question of how can we accurately characterize seismic hazard if we don't have a complete characterization? 00:11:59.000 --> 00:12:02.000 of the structures in the region. 00:12:02.000 --> 00:12:04.000 And just to highlight some of the challenges, 00:12:04.000 --> 00:12:12.000 Um, we've run into, uh, land access, the terrain is fairly steep and rugged, um, we have… 00:12:12.000 --> 00:12:17.000 the challenge of interpreting historical documents, there's institutional limitations. 00:12:17.000 --> 00:12:20.000 But going forward, um, 00:12:20.000 --> 00:12:25.000 I think that we have… 00:12:25.000 --> 00:12:33.000 Um, going forward, we've done some field mapping and trenching already, but I think a great future direction would be reflection and refraction studies to both ID and confirm faults. 00:12:33.000 --> 00:12:38.000 And this could help us focus our trenching and field mapping to, like, specific areas. 00:12:38.000 --> 00:12:50.000 And potentially even date some samples, we could determine if faults are quaternary active. And we are somewhat limited in what we can do, but we do have some ability to collaborate with others, as long as it's sort of within the scope of our program goals. 00:12:50.000 --> 00:12:58.000 Um, I'd be happy to chat with folks about possible collaborations, and also my colleague Don higher up, I think, would be the point person on that. 00:12:58.000 --> 00:13:18.000 So, thank you so much! 00:13:18.000 --> 00:13:19.000 You're muted. 00:13:19.000 --> 00:13:28.000 Sorry. Thanks for giving us the opportunity to provide you with updates to our SCAC community fault model. 00:13:28.000 --> 00:13:36.000 Um, that's a product we generate a support group. Next slide. 00:13:36.000 --> 00:13:42.000 But, uh, critically depends… no. Okay. 00:13:42.000 --> 00:13:57.000 Currently, it depends on many contributors and evaluators. As well as infrastructure support by SCACRC and web teams, and SISMOLab Caltech, for example, we have a new CFM homepage on Skekt.org. 00:13:57.000 --> 00:14:08.000 Which, uh, it looks great and has details and background information on the model. Go check it out. Next slide. 00:14:08.000 --> 00:14:14.000 Last year, we released 7 CFM version 7.0, the first statewide model. 00:14:14.000 --> 00:14:19.000 And now completed a peer review of the model in Central and Northern California. 00:14:19.000 --> 00:14:23.000 by a large team of more than 40 members. 00:14:23.000 --> 00:14:32.000 were involved, so big thanks to everybody who was involved. Many of you are here or were here in the last few days in particular, to the CFM Aviation. 00:14:32.000 --> 00:14:41.000 Group leaders, which organized most of that evaluation, and without them would not have been possible at all, so big thanks. 00:14:41.000 --> 00:14:50.000 To everybody, as a result, we received feedback for 132 faults in the model, here shown on the map, the colored traces. 00:14:50.000 --> 00:15:03.000 And, uh, specific suggestions. for more than 90-folds. Also, we received recommendations to add 45 more important faults to the model. 00:15:03.000 --> 00:15:12.000 So this will give you a basis to work in progress, and we do that in coordination with the Crescent CFM. 00:15:12.000 --> 00:15:26.000 group, for example, we already… shared the evaluation results for the Northern California portion in the overlap area with Crescent. Next slide. 00:15:26.000 --> 00:15:38.000 With Crescent, we also collaborated on. Defining and reassessing in a much more detailed border crossing falls. 00:15:38.000 --> 00:15:51.000 So this is a perspective view looking from Northern California into southern Oregon. In red, you can see 10 new 3D fault representations we developed based on all available data in the. 00:15:51.000 --> 00:15:57.000 area, so we shared that with the Crescent group, and so this is now available. 00:15:57.000 --> 00:16:07.000 in a report and data files, and so will be included in the next CFM release. Next slide. 00:16:07.000 --> 00:16:14.000 Finally, let me point you to refreshes of our online tools and resources. 00:16:14.000 --> 00:16:25.000 the CFM web explorer now lets you. plot recent and near-real-time seismicity along. 00:16:25.000 --> 00:16:30.000 with the faults in MapFuse and 3D views as well. 00:16:30.000 --> 00:16:38.000 And an earthquake Fault Association service, fully automated, now uses the latest CFM geometries. 00:16:38.000 --> 00:16:43.000 And naming has in the notifications you will receive after, directly after, and. 00:16:43.000 --> 00:16:52.000 event, large magnitude free. So in the equation, I have direct links to the CFM Explorer. 00:16:52.000 --> 00:17:00.000 Giving you immediate access to the. Uh, event and the associated fault for context. 00:17:00.000 --> 00:17:08.000 Finally, we have a new publication coming out in BSSA next month, very soon use that, and also with. 00:17:08.000 --> 00:17:16.000 Synodo DUIs to refer to the CFM. Uh, finally. 00:17:16.000 --> 00:17:37.000 real life HIG announcement coming out later this year, thank you. 00:17:37.000 --> 00:17:47.000 Okay, thank you. Sorry, I guess I thought I was one after this. My name is Evan Reese, and I'm a structural engineer, and I am vice president of SafeHub. 00:17:47.000 --> 00:17:52.000 We presented a couple years ago about what we were up to. 00:17:52.000 --> 00:17:58.000 at the Thunder Talk, and so I'm happy to kind of give you an update on what we're doing about how we are 00:17:58.000 --> 00:18:06.000 complementing, uh, the USGS earthquake mapping capabilities with our network of, uh, 00:18:06.000 --> 00:18:09.000 distributed wireless sensors. 00:18:09.000 --> 00:18:16.000 slide… I don't know how to go to the next slide, though, so maybe somebody can go to the next slide. Thank you. 00:18:16.000 --> 00:18:21.000 So, Safe Hub has, uh, developed a set of 00:18:21.000 --> 00:18:27.000 small, low-cost, easy-to-install sensors. They're about the size of an iPhone, a little thicker. 00:18:27.000 --> 00:18:35.000 Uh, and they can typically be installed in about 10 minutes at a building, wherever you have cellular access and power. 00:18:35.000 --> 00:18:41.000 They are run, uh, by a regular power connection into the wall with, uh, 00:18:41.000 --> 00:18:46.000 a backup battery that lasts about a week, and they provide… they use cellular connectivity. 00:18:46.000 --> 00:18:54.000 to connect to the outside world and transmit all of the data. Uh, they run off of a MEMS accelerometer, 00:18:54.000 --> 00:18:58.000 Uh, which we have tested against other, uh, 00:18:58.000 --> 00:19:05.000 more traditional wired MEMS accelerometers and, uh, force balance sensors that are used in the ANSS network. 00:19:05.000 --> 00:19:13.000 And under the periods of interest that we are concerned about, mainly for the performance of buildings, 00:19:13.000 --> 00:19:19.000 Uh, they perform exceedingly well, virtually indistinguishable, according to the test, uh, the test results. 00:19:19.000 --> 00:19:23.000 Uh, and so, um, next slide, please. 00:19:23.000 --> 00:19:36.000 We provide these sensors to clients for two purposes. The first is to assess damage to buildings directly with sensors in the buildings, and the second is for earthquake insurance. 00:19:36.000 --> 00:19:43.000 Um, this is an example of how, even in California, which has got to be the most censored place in the world, right? 00:19:43.000 --> 00:19:50.000 The addition of a network of easy-to-install sensors can improve 00:19:50.000 --> 00:19:57.000 the shaking estimates. On the left is a shake map, uh, developed for the Highland Park earthquake in 2024. 00:19:57.000 --> 00:20:05.000 The circles are, uh, ANSS stations that were accessed by USGS to create their shake map. 00:20:05.000 --> 00:20:08.000 On the right is… 00:20:08.000 --> 00:20:19.000 a shake net, that is our tool that essentially uses a similar process as ShakeMap, but also includes all of our sensors that are in the neighborhood. 00:20:19.000 --> 00:20:31.000 And as you can see, especially in the area near Commerce, California, we had about a dozen sensors that recorded significantly different ground shaking from that which was anticipated 00:20:31.000 --> 00:20:41.000 by, um, the, uh, shake map. And so you can see a revised map, if you will, that's created with much more granularity in that location. 00:20:41.000 --> 00:20:49.000 When you have facilities that have a very high value and, uh, you know, assessing damage is critical, 00:20:49.000 --> 00:20:56.000 These sensors are an important part of being able to quickly assess damage. 00:20:56.000 --> 00:20:59.000 Next slide. 00:20:59.000 --> 00:21:07.000 The other use case that we have for these sensors is with something called parametric insurance. It's a relatively new style of insurance, 00:21:07.000 --> 00:21:09.000 Where a payout 00:21:09.000 --> 00:21:17.000 After an earthquake is based on the size or the intensity of shaking at the site, and not a traditional 00:21:17.000 --> 00:21:28.000 claims adjustment that can go on for months or even years, and so the advantage of this is the money is accessible if you get a payout within a couple of weeks after an earthquake. 00:21:28.000 --> 00:21:33.000 Uh, and so the University of California, um, uh, 00:21:33.000 --> 00:21:41.000 bought a parametric policy using our sensors. We've installed 200 sensors across the UC footprint. 00:21:41.000 --> 00:21:46.000 And these sensors are used in their high-value facilities to 00:21:46.000 --> 00:21:56.000 trigger payouts for their insurance policy, and this… the insurance money can be used to repair the buildings, but it can also be used for anything else, like paying overtime for 00:21:56.000 --> 00:22:01.000 Emergency staff, or sending students home if they need to close a building. 00:22:01.000 --> 00:22:10.000 But the idea here is they are able to use these sensors not only for their post-earthquake damage assessment, but also as part of their insurance program. 00:22:10.000 --> 00:22:16.000 And because they were easy to install, these 200 sensors were installed in about 7 months. 00:22:16.000 --> 00:22:23.000 Uh, they… it makes it very possible to quickly get up to speed emergency response programs, 00:22:23.000 --> 00:22:44.000 And insurance programs. So, that's the exciting stuff that we've been doing over the past 2 years since I presented last, so thank you very much. 00:22:44.000 --> 00:22:48.000 I am Michael Burrell, and I will be talking about the Event Sequence product. 00:22:48.000 --> 00:22:51.000 The event pages on the USGS website 00:22:51.000 --> 00:22:54.000 display a lot of information about each earthquake. 00:22:54.000 --> 00:23:00.000 But they display earthquakes one at a time, whereas we know that earthquakes can be grouped into sequences. 00:23:00.000 --> 00:23:06.000 So, our goal is to display information about earthquake sequences on the event page. 00:23:06.000 --> 00:23:12.000 To achieve this goal, we have formed a collaboration between the Operational Aftershock Forecast Team, 00:23:12.000 --> 00:23:15.000 or OAF, at the Earthquake Science Center. 00:23:15.000 --> 00:23:17.000 And the Hazards Development Team, or HASDEV, 00:23:17.000 --> 00:23:20.000 the Geologic Hazard Science Center. 00:23:20.000 --> 00:23:25.000 OEF is involved because aftershock forecasts already deal with sequences. 00:23:25.000 --> 00:23:31.000 The OAF system automatically identifies and tracks earthquake sequences as they occur. 00:23:31.000 --> 00:23:37.000 A sequence is defined by a circular region on the Earth's surface, a start time, and an end time. 00:23:37.000 --> 00:23:43.000 Every earthquake that occurs in the region between the two times is considered to be part of the sequence. 00:23:43.000 --> 00:23:50.000 Also, the OIF team has analysts who can examine and, if needed, adjust the sequence definition. 00:23:50.000 --> 00:23:53.000 Today, there is a so-called alpha… 00:23:53.000 --> 00:23:56.000 version that displays limited sequence information. 00:23:56.000 --> 00:24:01.000 Here's how that works. When the OIF system generates an aftershock forecast, 00:24:01.000 --> 00:24:03.000 It sends an OAF product to ComCat. 00:24:03.000 --> 00:24:08.000 A product is a package of earthquake information that can be sent across the internet. 00:24:08.000 --> 00:24:12.000 An OAF product contains an Aftershock forecast. 00:24:12.000 --> 00:24:16.000 And ComCat is a big database that stores all the products. 00:24:16.000 --> 00:24:19.000 The event page is constructed from the data in Comcast. 00:24:19.000 --> 00:24:25.000 With this architecture, the event page displays limited information about sequences. 00:24:25.000 --> 00:24:28.000 Specifically, the event page for a main shock… 00:24:28.000 --> 00:24:31.000 Displays a notification that it is the main shock of a sequence, 00:24:31.000 --> 00:24:34.000 And a map of the sequence. But… 00:24:34.000 --> 00:24:38.000 This is limited to main shocks that have an active aftershock forecast. 00:24:38.000 --> 00:24:44.000 Which generally means main shocks that occurred in the U.S. within the last 2 years. 00:24:44.000 --> 00:24:48.000 To do better, we are introducing the Event Sequence product. 00:24:48.000 --> 00:24:51.000 Which contains just the sequence definition. 00:24:51.000 --> 00:24:54.000 The circular region, start time, and end time. 00:24:54.000 --> 00:24:57.000 When the OIF system generates an aftershock forecast, 00:24:57.000 --> 00:25:01.000 It will send both the OAF product and the event sequence product. 00:25:01.000 --> 00:25:04.000 Making event sequence into a separate product, 00:25:04.000 --> 00:25:10.000 means that sequence information is no longer limited to main shocks that have an active aftershock forecast. 00:25:10.000 --> 00:25:16.000 The event sequence product, unlike the OIF product, will remain in Comcast permanently. 00:25:16.000 --> 00:25:21.000 Also, it becomes possible to display much more information about sequences. 00:25:21.000 --> 00:25:25.000 For example, the event page for every Aftershock and Forshock… 00:25:25.000 --> 00:25:27.000 We'll have a pointer to the event page of the main shock. 00:25:27.000 --> 00:25:32.000 Also, it will be possible to display sequence-oriented information 00:25:32.000 --> 00:25:47.000 such as a magnitude time plot of the sequence. Thank you. 00:25:47.000 --> 00:25:57.000 Hello, everyone. Uh, I'm Dave Croker. I'm the. project lead for the Northern California Seismic Network. 00:25:57.000 --> 00:26:08.000 Which is the, uh… operational, uh, USGS seismic network that covers the Parkfield region, and today I'm. 00:26:08.000 --> 00:26:14.000 Uh, not giving you information, but asking for your input about, uh. 00:26:14.000 --> 00:26:20.000 your interest in Parkfield. Next, please. 00:26:20.000 --> 00:26:30.000 Uh, so, for those of you that have… in the seismological community that have been under a rock for a few decades, Park Field is, um. 00:26:30.000 --> 00:26:38.000 There's a sleepy little town. about halfway between San Francisco and Los Angeles. 00:26:38.000 --> 00:26:49.000 Um, and next, please. It's in the coast, uh, ranges, and it's known because it sits on this big, thick. 00:26:49.000 --> 00:26:52.000 red line, which is known as the San Andreas Fault. 00:26:52.000 --> 00:26:59.000 Next, please. And, uh, it's also known because it has a long history of. 00:26:59.000 --> 00:27:05.000 Magnitude 6… earthquakes, uh, in the region. You can see the. 00:27:05.000 --> 00:27:14.000 the list there of, uh… historic earthquakes, um… And… next, please. 00:27:14.000 --> 00:27:21.000 Um, I can tell the interest is, uh, growing, uh, because the indoor. 00:27:21.000 --> 00:27:32.000 seismic gap is, uh, starting to close now. Um… We are coming… closer to that windows. 00:27:32.000 --> 00:27:47.000 Uh, so to speak. Um, next… Um, so Parkfield, uh, has also been a, uh, earth laboratory for, uh, decades. 00:27:47.000 --> 00:27:54.000 The first, uh, NCSN station was installed in the late 60s. 00:27:54.000 --> 00:28:00.000 And there's been… Uh, there's, uh, undoubtedly monitoring before that. 00:28:00.000 --> 00:28:09.000 Uh, this map is showing what the. current semi or… or a permanent or semi-permanent, uh, monitoring. 00:28:09.000 --> 00:28:15.000 looks like in the region. Parkfield, uh, the town of is, uh. 00:28:15.000 --> 00:28:19.000 close to the center of the, um… the map there. 00:28:19.000 --> 00:28:26.000 Um, of course, through the years, there's been much research with temporary, uh. 00:28:26.000 --> 00:28:31.000 Um, uh, monitoring… Uh, coming and going. 00:28:31.000 --> 00:28:43.000 Uh, but this is what, uh, remains today. Um, one thing I'm not showing on this map is, uh, um, there's also an array of about 40. 00:28:43.000 --> 00:28:51.000 Or so, um, temporary. or campaign-style GPS monuments at USGS also monitors. 00:28:51.000 --> 00:28:57.000 Uh, from time to time. Next, please. 00:28:57.000 --> 00:29:07.000 Uh, and one reason why I'm… talking today is that the seismic Monitoring Network, um, is showing its age. 00:29:07.000 --> 00:29:20.000 And, uh, in need of upgrade, and. There isn't a clear… path forward, uh, for funding. 00:29:20.000 --> 00:29:24.000 Or, uh, well, mainly funding, uh, but also time. 00:29:24.000 --> 00:29:32.000 Um, through the years, uh, stations have, um… Uh, like I said, aged. 00:29:32.000 --> 00:29:40.000 Um, they're not a lot of… the, uh, over half the USGS instrumentation is still old analog. 00:29:40.000 --> 00:29:48.000 Um, instrumentation, which, uh, we're… slowly not supporting any longer. 00:29:48.000 --> 00:29:53.000 Haven't necessarily had adequate maintenance because of the, um. 00:29:53.000 --> 00:29:59.000 the old instrumentation. Uh, and, um, yeah, there. 00:29:59.000 --> 00:30:06.000 things are, uh, starting to not work. Uh, next please. 00:30:06.000 --> 00:30:13.000 So, um, that's why I'm here, and that's why there's been a lot more interest in the last couple years. 00:30:13.000 --> 00:30:23.000 I've already attended a couple informal gatherings. Um, but we're asking for, uh, everyone's input on, uh. 00:30:23.000 --> 00:30:30.000 Um, how Parkfield monitoring is important to you and your studies and your goals. 00:30:30.000 --> 00:30:37.000 Um, of course, um, you've all been to the cafe, uh, that claims it's the earthquake capital of the world. 00:30:37.000 --> 00:30:48.000 Um… Uh, that's the, uh, that's the view outside the cafe. Uh, next, please. 00:30:48.000 --> 00:30:56.000 Um, yeah, so… As I say, there's been a lot more interest recently. 00:30:56.000 --> 00:31:05.000 in, uh… Parkfield monitoring, um, we're looking for your, your input. There has, uh. 00:31:05.000 --> 00:31:11.000 been, uh, some small victories and improvements, uh, through the years. 00:31:11.000 --> 00:31:21.000 For instance, UC Berkeley's Seismological laboratory. did get a small, uh, bump in funding to, uh. 00:31:21.000 --> 00:31:32.000 bring their high-resolution seismic network, borehole network, um, at least back operational again as more of a band-aid fix than anything. 00:31:32.000 --> 00:31:39.000 Um, the shake alert, uh, project has infused money to upgrade at least. 00:31:39.000 --> 00:31:43.000 a handful of… stations in the area. 00:31:43.000 --> 00:32:00.000 Um, but certainly a lot are still. The USGSGNSS network that's operational there that they took over from UNAVCO, now Earthscope, um… is more or less running well. 00:32:00.000 --> 00:32:06.000 But they also don't have a full staff, and um. 00:32:06.000 --> 00:32:19.000 facing funding. Uh, issues as well. Uh, a big victory, uh, recently is California CGS, the Strong Motion Instrumentation Program, has. 00:32:19.000 --> 00:32:30.000 Um… gotten funding to upgrade a lot of their strong motion stations, about half of their stations now have been upgraded, so, um, that's excellent, but. 00:32:30.000 --> 00:32:41.000 There's a long way to go there still. the SAFOD, uh, deep… drilling project is, um… off and running again, they're installing a new fiber. 00:32:41.000 --> 00:32:47.000 Uh, in the hole for, uh, gas array. Um, which is exciting. 00:32:47.000 --> 00:32:55.000 Recently, Sue Huff has convened an informal group of folks to try to. 00:32:55.000 --> 00:33:08.000 Uh, this is not so much, um… monitoring, but she's, uh, working to try and, um… correct the current Comcast catalog for the main shocks. 00:33:08.000 --> 00:33:12.000 So, way to go, Sue, on that. That'll be exciting. 00:33:12.000 --> 00:33:24.000 Uh, next, please. And thank… thankfully, uh, Liz Hearn from the Earthquake Hazards Program Office has, uh, stepped in and is, um. 00:33:24.000 --> 00:33:30.000 Looking to help, uh… consolidate, um… Uh, people's input. 00:33:30.000 --> 00:33:38.000 And, uh, see what… We can do monitoring moving forward, so… Uh, please give us your feedback. 00:33:38.000 --> 00:33:46.000 Let us hear from you, and um… Also, don't forget, there's a SCAP conference being organized later this year. 00:33:46.000 --> 00:33:50.000 And, uh, I think I saw in the chat earlier that, um. 00:33:50.000 --> 00:33:55.000 Julian, uh, has a Parkfield class. I'm interested in that. I need to. 00:33:55.000 --> 00:34:04.000 Learn more about that. Anyway… That's all from me. Thanks. 00:34:04.000 --> 00:34:08.000 Thanks, Steve, and thanks to all of our speakers in this session. 00:34:08.000 --> 00:34:11.000 So now we have, um, 10 minutes for a live… 00:34:11.000 --> 00:34:15.000 Q&A, and I think maybe what we could do is first take… 00:34:15.000 --> 00:34:22.000 general questions to any of the speakers, and then maybe transition into a short 00:34:22.000 --> 00:34:24.000 discussion about Parkfield, if that sounds… 00:34:24.000 --> 00:34:27.000 And we want to use our time that way. 00:34:27.000 --> 00:34:33.000 Um, and there was lots of questions in the chat, or… 00:34:33.000 --> 00:34:38.000 Evan, um, and maybe I will… 00:34:38.000 --> 00:34:40.000 start off by asking, 00:34:40.000 --> 00:34:49.000 Um, or echoing a question that Ruth had about where your sensors are installed in buildings and, um, 00:34:49.000 --> 00:34:55.000 How do you think about, kind of, separating ground response from building response? 00:34:55.000 --> 00:35:04.000 Yeah, great question. So, our sensors are located in buildings. We do have weatherproof enclosures that can sit outside, but all of our sensors so far are in buildings. 00:35:04.000 --> 00:35:15.000 Um, the ones that we use to create maps are at the ground floor or the basement of the buildings. We do have sensors in higher floors to look at the building response, you know, directly to the floor response. 00:35:15.000 --> 00:35:22.000 Um, when it comes to, say, soil structure interaction or the impact of the building itself on the reading, 00:35:22.000 --> 00:35:27.000 Our sensors are at the ground level or at the basement level. Most of the sensors 00:35:27.000 --> 00:35:30.000 are in low-rise buildings. 00:35:30.000 --> 00:35:44.000 Um, most of them also are on stiff soil, so we believe that the soil structure interaction is generally minimal. If we have buildings that are in really… or sensors in really unique conditions, we typically won't use them to create our maps, but obviously, 00:35:44.000 --> 00:35:54.000 If we want to understand building performance, and that's really the main point, then we want the soil structure interaction. We want whatever's going to happen to happen, and we want to get the actual reading from the 00:35:54.000 --> 00:35:57.000 this sensor in the building. 00:35:57.000 --> 00:36:03.000 Neat. And just a reminder to the audience that you can raise your hand to ask your questions. 00:36:03.000 --> 00:36:05.000 Um… 00:36:05.000 --> 00:36:09.000 And I have a follow-up question. So, you showed the shake map? 00:36:09.000 --> 00:36:10.000 Sure. Right. 00:36:10.000 --> 00:36:12.000 The two shake maps, um, for 0.3 second. 00:36:12.000 --> 00:36:18.000 PSA or response spectra. So did you kind of, um, assume that the recordings 00:36:18.000 --> 00:36:20.000 from your instruments in… 00:36:20.000 --> 00:36:23.000 the ground level of buildings, 00:36:23.000 --> 00:36:29.000 represented that 0.3 second response vector, or did you calculate that from the acceleration? 00:36:29.000 --> 00:36:30.000 Okay. 00:36:30.000 --> 00:36:34.000 We calculated that. We did the transform and calculate the full response spectra, yeah. 00:36:34.000 --> 00:36:39.000 Thanks. And I saw in the chat that, um… 00:36:39.000 --> 00:36:41.000 Danielle is? 00:36:41.000 --> 00:36:47.000 on the call now, and if she doesn't mind me putting her on the spot, I have a question. 00:36:47.000 --> 00:36:48.000 For her. Yeah, thanks for joining us. Um… 00:36:48.000 --> 00:36:54.000 Go for it. 00:36:54.000 --> 00:36:56.000 Sorry about me missing out, I… Got the time around. It's fine. 00:36:56.000 --> 00:37:02.000 No problem, I know that time zones are a challenge. Um, thanks for being here now. 00:37:02.000 --> 00:37:10.000 Um… I had a question about the spatial distribution of the landslides that you observed after Ferndale. 00:37:10.000 --> 00:37:17.000 Um, there was a line of text on one of the slides saying that, um… 00:37:17.000 --> 00:37:19.000 I wrote it down, it was like the… 00:37:19.000 --> 00:37:24.000 50 kilometers away from the source was the… 00:37:24.000 --> 00:37:28.000 the… where you saw the most amount of landslides. I was wondering if you could… 00:37:28.000 --> 00:37:32.000 elaborate on that a little bit, and… 00:37:32.000 --> 00:37:36.000 Yeah, um, so I think it has to do with the, like. 00:37:36.000 --> 00:37:42.000 directivity of the second earthquake compared to the first one, so we had the different kind of source mechanisms, where you had, like, the. 00:37:42.000 --> 00:37:56.000 two faults going versus, like, one classic, like, rupturing northward. And so, I think where there was, like, the peak shaking is where we had the most, like, accelerated landslides. So although the source started further away, the actual lake. 00:37:56.000 --> 00:38:08.000 high-frequency shaking was, like, at a further distance, and so, um, there was, like, a cluster of landslides that were all kind of accelerated together, all on that kind of peak PTA area. 00:38:08.000 --> 00:38:24.000 And so I think that, like, um, it's kind of like a useful thing to remember in the immediate aftermath after a big earthquake, not to just look, like, closest to the source. Like, there you expect the landslides to fail catastrophically, but when you get out of, like, the high… like, the closest ones, you can accelerate these ones. 00:38:24.000 --> 00:38:30.000 Like, it gives it a push, but not enough to fail catastrophically, so there's this, like, I think some kind of distance. 00:38:30.000 --> 00:38:32.000 Shaking… thing to look into. 00:38:32.000 --> 00:38:44.000 Neat, thanks, and I'm glad I asked, because my intuition was going to be maybe it wasn't the peak amplitude of shaking, but it had, you know, more cycles at some farther distances, but you think maybe it really was kind of the… 00:38:44.000 --> 00:38:46.000 the peak, um… 00:38:46.000 --> 00:39:00.000 I haven't looked at it into, like. really in-depth, it was just something that I'd kind of noticed. Like, it was like… opposite to my intuition, which was like, oh, it should be closest to the source, we have the most, like, triggering, but actually, maybe there's, like, a thresholding where. 00:39:00.000 --> 00:39:12.000 If the shaking is above a certain threshold, you'll just catastrophically fail, but if it's, like, below a certain threshold, then you're… you can give it a kick, but it's not enough to catastrophically fail. 00:39:12.000 --> 00:39:13.000 Thanks. Uh, and I see Jeff. 00:39:13.000 --> 00:39:15.000 Something like that, maybe. 00:39:15.000 --> 00:39:24.000 asked Evan in the chat, will the data you collect from major earthquakes end up publicly available? 00:39:24.000 --> 00:39:26.000 And you're muted, sorry. 00:39:26.000 --> 00:39:32.000 Um, not at this point, you know, our data typically is for clients that… 00:39:32.000 --> 00:39:38.000 want to keep that data confidential. We use it for our own research purposes, and we do have some agreements with 00:39:38.000 --> 00:39:45.000 some researchers to access the data and use it, but it's not a publicly available service at this point. 00:39:45.000 --> 00:39:48.000 Although we are going to be producing the maps, 00:39:48.000 --> 00:39:55.000 Uh, publicly at some point. So you won't see an individual sensor recording 00:39:55.000 --> 00:40:03.000 piece of data, but the maps will include all the data that the sensors did aggregate. 00:40:03.000 --> 00:40:05.000 Thanks. 00:40:05.000 --> 00:40:12.000 Okay, does anyone in the audience have a question? 00:40:12.000 --> 00:40:15.000 Uh, if not, maybe we could… 00:40:15.000 --> 00:40:18.000 Ask if anyone in the audience has, um… 00:40:18.000 --> 00:40:24.000 comments, questions, discussion about, um, Dave's talk on Parkfield? 00:40:24.000 --> 00:40:26.000 And I thought maybe it would be fun. 00:40:26.000 --> 00:40:27.000 to ask? 00:40:27.000 --> 00:40:32.000 I saw a question. Uh, sorry, Grace, I saw a question from Rufus I can just answer. 00:40:32.000 --> 00:40:34.000 Sure. 00:40:34.000 --> 00:40:42.000 Here, um… So, they're… Um, speaking for the NCSN. 00:40:42.000 --> 00:40:47.000 There are 16, uh, analog stations still in operation. 00:40:47.000 --> 00:40:53.000 Um, officially, active? Actually, no, sorry, um… It's 19, I think. 00:40:53.000 --> 00:41:01.000 Um, in quick review with, uh, colleagues at Berkeley, about 3 of those are kind of, um. 00:41:01.000 --> 00:41:14.000 uh, duplicating efforts of, uh, where Berkeley's. Um, network covers, so if we… only upgrade the 16 remaining there. 00:41:14.000 --> 00:41:34.000 Uh, it would be about $640,000 to, um. upgrade what's there. Plus, uh, probably some additional… telemetry infrastructure, um… That would be required to make it all. 00:41:34.000 --> 00:41:38.000 And then, um, on top of that, um, I know. 00:41:38.000 --> 00:41:53.000 Berkeley's network, um… Uh, really gets used, uh… by a lot of folks, um… You know, it's one of the premier examples of a borehole, um, weak motion network anywhere. 00:41:53.000 --> 00:42:00.000 And, um… Again, the funding that they got was just enough to really basically get it. 00:42:00.000 --> 00:42:11.000 running, again, new batteries, and that's about it, so… Uh, there's… there's a lot of deferred maintenance that, uh, really. 00:42:11.000 --> 00:42:12.000 Thanks. 00:42:12.000 --> 00:42:14.000 should happen with their network as well. 00:42:14.000 --> 00:42:23.000 Well, I was gonna maybe do something fun and ask the audience to raise their hand if they were involved in any of 00:42:23.000 --> 00:42:35.000 putting out any of the instrumentation at Parkfield, or I've ever used data that that instrumentation has collected in their… collected in their work in any capacity. I know I have, and I'm imagining 00:42:35.000 --> 00:42:42.000 a lot of people have. I think this, um… 00:42:42.000 --> 00:42:47.000 you know, the observatory there has been really… 00:42:47.000 --> 00:42:49.000 impactful in a lot of people's work. 00:42:49.000 --> 00:43:00.000 And I am trying to figure out how to use their reactions, but my hand is raised. 00:43:00.000 --> 00:43:06.000 How do we count the number of hands up? 00:43:06.000 --> 00:43:07.000 Uh, exactly 37. 00:43:07.000 --> 00:43:10.000 John, is there a way to do that? 00:43:10.000 --> 00:43:11.000 How's that? Was that good? 00:43:11.000 --> 00:43:20.000 Okay. 00:43:20.000 --> 00:43:23.000 Yeah, I think it's a lot. 00:43:23.000 --> 00:43:29.000 The answer to Sarah is yes. 00:43:29.000 --> 00:43:30.000 Okay, well maybe I'll turn… 00:43:30.000 --> 00:43:33.000 Samoa's question was, are we counting geodetic data as well as seismic data? 00:43:33.000 --> 00:43:39.000 Yes, yeah, yes, of course. 00:43:39.000 --> 00:43:40.000 Okay. 00:43:40.000 --> 00:43:43.000 Well, we are just about at 2 o'clock. Do we want to wrap things up, Grace? 00:43:43.000 --> 00:43:46.000 Yes, I think we're in. 00:43:46.000 --> 00:43:54.000 All right, awesome! Well, let's thank all of our wonderful Thundertalk speakers with special spontaneous guest appearance from Roland Borgman. 00:43:54.000 --> 00:43:59.000 And our wonderful moderator, Grace Porkle. Thank you so much, this has been fantastic! 00:43:59.000 --> 00:44:05.000 Everybody, please enjoy a brief 15-minute break, and we will be back. 00:44:05.000 --> 00:44:09.000 2.15 PM Pacific Time for a very special 00:44:09.000 --> 00:44:11.000 Celebrity Roundtable Discussion! 00:44:11.000 --> 00:44:13.000 So, um, if you are not, uh, 00:44:13.000 --> 00:44:16.000 On the Roundtable, please take 15 minutes. If you are on the roundtable, please stick around so we can say hi and get you set up. 00:44:16.000 --> 00:44:23.000 Thank you