WEBVTT 00:00:05.000 --> 00:00:11.000 Welcome back everyone. You will notice that this session and a bunch of other sessions on our agenda 00:00:11.000 --> 00:00:20.000 have titles that begin, "Where in the world is Northern California Earthquake Hazards Workshop Today? These sessions were inspired by an idea from one of our organizers, Josie Nevitt. 00:00:20.000 --> 00:00:25.000 And she had a really great idea that we should look at some, 00:00:25.000 --> 00:00:45.000 we should pick some particular areas around Northern California and then look at them in the most multidisciplinary way possible, to bring in as many different aspects of laboratory work, paleoseismology, geology, seismology, geodesy, theoretical physics, everything we can to understand the science, and 00:00:45.000 --> 00:01:00.000 the hazards in that particular region. Now, obviously, we're not going to be able to bring every facet to bear every location, but that's the theme for these sessions that begin "Where in the World is the Workshop Today?" and for our first little trip to a part of the Bay 00:01:00.000 --> 00:01:14.000 Area/Northern California we are going to look at the excerpts of the East Bay, which is an increasingly populous region, and where we have a delta that can do all sorts of interesting things, and some potentially understudy faults like the Greenville fault and so to take us on our 00:01:14.000 --> 00:01:18.000 journey to the East Bay today we have Don and Chad as our moderators. 00:01:18.000 --> 00:01:19.000 Please take it away. Don and Chad. 00:01:19.000 --> 00:01:25.000 Hey! Good afternoon, everybody. Thanks for having us. My name is Don Hoirup. 00:01:25.000 --> 00:01:26.000 [noise] 00:01:26.000 --> 00:01:45.000 I'm with the California Department of Water Resources. I'm an engineering geologist there. I'm going to be sharing the moderator duties with Chad Trexler with USGS. Chad's going to introduce himself and then introduce our first presentation. Take it away, Chad. 00:01:45.000 --> 00:01:54.000 Thanks, Don, and thank you all. We're gonna get started today with a trip to the Sacramento-San Joaquin. 00:01:54.000 --> 00:01:55.000 Delta. [noise] 00:03:32.000 --> 00:03:34.000 My name is Chris Madugo. I am a geologist with the Geosciences Department at PG&E. Today, 00:03:34.000 --> 00:03:45.000 I will be talking about the Concord-Green Valley Fault Zone, providing an overview in industry perspective. The photo on the right shows en 00:03:45.000 --> 00:03:51.000 echelon, stepping crack pattern in the city of Concord 00:03:51.000 --> 00:03:58.000 along the map trace of the fault, which is a creeping fault. 00:03:58.000 --> 00:04:14.000 The Concord-Green Valley Fault Zone is not as well known as other faults in the Bay Area, such as the San Andreas fault and Hayward fault, both of which have had large destructive earthquakes in the historic period. The Concord-Green Valley Fault Zone has 00:04:14.000 --> 00:04:18.000 not had a large earthquake, but studies over the last 25 years show that it's capable of earthquakes 00:04:18.000 --> 00:04:35.000 equal to, or greater than magnitude 6. The fault crosses population centers, lifelines, and distributed infrastructure and paleoseismic studies show that it's about 200 years past 00:04:35.000 --> 00:04:38.000 this average occurrence interval. In this talk, I will be talking about fault characteristics, modeled earthquakes, 00:04:38.000 --> 00:04:52.000 magnitudes, and rates, and the earthquake record industry supported research and consulting studies on the Concord 00:04:52.000 --> 00:04:58.000 fault and research needs. 00:04:58.000 --> 00:05:20.000 The map on the right shows the distributed San Andreas Fault System in the Bay Area with the San Andreas fault on the west, the Hayward-Rogers Creek Fault System in the middle and the eastern margin of this fault system is the Concord Green Valley 00:05:20.000 --> 00:05:26.000 fault. Starting in the bottom right and in the south 00:05:26.000 --> 00:05:42.000 this fault system includes the Concord fault which extends from the north edge of Mount Diablo to Suisun Bay, and it passes through the city of Concord, which has a population of greater than 100,000 people. 00:05:42.000 --> 00:05:53.000 The Green Valley fault, which is 43 kilometers, at least 43 kilometers long, and it extends from Suisun Bay to around Lake Berryessa. 00:05:53.000 --> 00:06:13.000 Passing through the town of Cordelia, and near the town of Fairfield and then to the north, the Hunting Creek-Berryessa fault, which is 44 kilometers long, and Lienkaemper marks this as the northern, 00:06:13.000 --> 00:06:17.000 Crosses the Northern Green Valley fault. For this talk, 00:06:17.000 --> 00:06:27.000 I'll be talking about the Green Valley fault proper here, and the Concord fault because these are the best studied sections of the fault system. 00:06:27.000 --> 00:06:29.000 The black circles mark structural complexity which could be boundaries to earthquake rupture. 00:06:29.000 --> 00:06:33.000 These are typically step-overs that are at least one kilometer wide. 00:06:33.000 --> 00:06:47.000 The largest and most complicated step-over is at the southern end of the fault system, between the Greenville fault 00:06:47.000 --> 00:06:51.000 and the Calaveras fault. 00:06:51.000 --> 00:06:56.000 Slip rates for the Green Valley and Concord faults were developed at two sites. 00:06:56.000 --> 00:07:10.000 The Lopez Ranch site on the Green Valley fault and the Galindo Creek site on the Concord fault. At Lopez Ranch two channels were studied, one that's a few 100 years old; 00:07:10.000 --> 00:07:27.000 another one that's 15,000 years old, and these resulted in rate of estimates of 3.9 to 4.8mm a year for the younger channel, and 2.2 to 4mm a year for the older channel. On the Concord fault a 6,000 year 00:07:27.000 --> 00:07:34.000 old channel, yields a rate of 3.4 millimeters a year. 00:07:34.000 --> 00:07:51.000 The UCERF3 model assigns preferred rates of 4 to 4.3mm a year for the Concord and and Green Valley faults, with a range of 2 to 9mm and 3 to 9mm a year for the different sections. 00:07:51.000 --> 00:08:07.000 This accounts for the fact that the Calaveras fault to the south, which has a rate of 6mm a year, could be transferring some slip to the Concord fault. in addition to the Contra Costa shear zone just marked here. And then the Greenville 00:08:07.000 --> 00:08:08.000 fault could be transferring slip via the blind Mount Diablo 00:08:08.000 --> 00:08:19.000 thrust. The Greenville fault rate is 2mm a year. 00:08:19.000 --> 00:08:40.000 As I mentioned above, the Concord-Green Valley Fault System is creeping theodolite network with multiple stations along those faults yields a rate of 2.9 to 3.5mm a year for the Concord fault and variable creep 00:08:40.000 --> 00:08:45.000 rates of less than 1mm a year to 3.7mm a year 00:08:45.000 --> 00:09:00.000 along the Green Valley fault. Just north of I-80 is a historic wall that was built around 1862, and Anderson and Mahrer, 00:09:00.000 --> 00:09:06.000 calculated a rate of 1.4 to 3.4mm a year, which is within the uncertainties of the other calculated creek rates. The photo on the right 00:09:06.000 --> 00:09:23.000 was taken in the city of Concord, which has ample evidence of creep throughout the town. 00:09:23.000 --> 00:09:40.000 The earthquake potential just looking at deterministic scenarios, data that you could get online from Caltrans that used UCERF sections calculated in magnitude 6.8 for the Green Valley fault 00:09:40.000 --> 00:09:44.000 and M6.6 for the the Concord fault. 00:09:44.000 --> 00:10:08.000 PG&E deterministic scenarios arranged from M6.4 to M7.2 on the Green Valley, and M6.0 to M7.1 on the Concord, and these account for ruptures that could include parts of each fault section, or adjacent fault sections and that's how you 00:10:08.000 --> 00:10:11.000 would get the higher magnitudes. Looking at the UCERF3 00:10:11.000 --> 00:10:14.000 incremental magnitude frequency rates for a magnitude 6, 00:10:14.000 --> 00:10:32.000 you get a return period of about 200 to 300 years for the Green Valley and Concord faults, and about 2,500 years for magnitude 7. 00:10:32.000 --> 00:10:44.000 Now instrumental seismicity has had no events that approach magnitude 6 on the Green Valley fault. 00:10:44.000 --> 00:11:00.000 There have been a couple of magnitude 4.1 along the fault, and between, 1984 and 2009 there were 33 earthquakes ranging from magnitude 2.5 to M4.1. 00:11:00.000 --> 00:11:21.000 On the Concord fault, there was a magnitude 5.4 in 1955 and a bunch of seismicity smaller than magnitude 4 directly along the fault, and then multiple magnitude 4's in the vicinity, but not directly on the fault. These include 00:11:21.000 --> 00:11:28.000 magnitude 4.5 earthquake that occurred in 2019 in Pleasant Hill, and then the Danville and Alamo earthquake swarms which had multiple magnitude 4's, 00:11:28.000 --> 00:11:41.000 but again, these aren't directly on the fault. Looking at the paleoseismic record, the Mason Road site north of I-80, 00:11:41.000 --> 00:11:55.000 and the Lopez Ranch site near Cordelia combined to produce a record of 3 to 4 earthquakes with the most recent event occurring around 1610 CE 00:11:55.000 --> 00:12:09.000 Because the fault is creeping, you have to be very careful in identifying larger events. 00:12:09.000 --> 00:12:13.000 This is done by identifying areas where fischer fills were developed. 00:12:13.000 --> 00:12:22.000 These are assumed to happen pretty suddenly during a large earthquake magnitude 6 or greater, and then filled in with colluvium shortly 00:12:22.000 --> 00:12:31.000 after the event. The average recurrence interval is 199 to 200 years, based on these two sites. 00:12:31.000 --> 00:12:35.000 But it's been 400 years since the last event. 00:12:35.000 --> 00:12:42.000 So the star here on the figure on the lower right shows that just given the average recurrence interval, the next earthquake should have happened in the 1760s. 00:12:42.000 --> 00:12:50.000 But we're far beyond that, with no event. 00:12:50.000 --> 00:13:00.000 The UCERF3 time dependent model calculates a 23 percent chance of a greater than or equal to magnitude 6.7 00:13:00.000 --> 00:13:05.000 in the next 30 years. There's no geologic slip per event 00:13:05.000 --> 00:13:26.000 data largely due to the fact that it's hard to get slip per event when you have variable creep rates along the fault because you're looking at coseismic displacement plus some unknown amount of creep unless you've got a station right at your 00:13:26.000 --> 00:13:37.000 site. Modeled estimates of deterministic displacements range from .6 to 1.7 meters and 1 to 1.4 meters. 00:13:37.000 --> 00:13:54.000 That's PG&E Model Working Group on California earthquake probabilities of 3, and then taking some of the UCERF data and calculating the displacements, using magnitude log area relations. 00:13:54.000 --> 00:14:06.000 So moving on to industry funded studies, and fault displacement studies that I've been involved in with consultants on the the Concord-Green Valley Fault System in the late 1990s, early 2000s, PG&E 00:14:06.000 --> 00:14:13.000 teamed with USGS to do reconnaissance level 00:14:13.000 --> 00:14:17.000 paleoseismology on most of the faults in the Bay Area 00:14:17.000 --> 00:14:35.000 looking at over 25 sites and several of these sites yielded good paleoseismic records that are the basis for slip rates and return periods and timing of the last event for faults in the Bay Area, so on the Concord-Green Valley System 00:14:35.000 --> 00:14:42.000 there were four sites. Two of which yielded good data, as I showed above. 00:14:42.000 --> 00:14:43.000 Close to 10 years that I've been at PG&E 00:14:43.000 --> 00:15:00.000 we've done multiple fault displacement, evaluations for gas transmission lines along the Concord and Green Valley fault in the area shown in the stars. 00:15:00.000 --> 00:15:07.000 Shown by the stars, and I'll focus on a couple of these in my next slides. 00:15:07.000 --> 00:15:08.000 So the Concord fault, because we've done so many studies on it. 00:15:08.000 --> 00:15:15.000 It's a great laboratory to look at uncertainties that we're interested in. 00:15:15.000 --> 00:15:24.000 For studying, a fault displacement hazard to our infrastructure. 00:15:24.000 --> 00:15:27.000 So one critical uncertainty is fault location, uncertainty. 00:15:27.000 --> 00:15:49.000 So, we and the consultants that we work with use geomorphology, creep evidence, subsurface data from trenching or drilling. The figure on the right shows two boreholes that were done by LCI to pinpoint the location of the Concord 00:15:49.000 --> 00:15:51.000 fault. And then we also use empirical data 00:15:51.000 --> 00:16:14.000 from statistics from Peterson 2011, that look at where geologists map faults and where ruptures actually occur, and judgment. We've been doing a big research push the last few years with Arizona State University developing data to 00:16:14.000 --> 00:16:23.000 improve, statistical models of fault location and uncertainty. Working primarily with Chelsea Scott, and Ramon Aerosmith. North of Interstate 80 00:16:23.000 --> 00:16:27.000 here's another site where we've got three fault strands 00:16:27.000 --> 00:16:35.000 their locations is well constrained by trenching, 00:16:35.000 --> 00:16:36.000 and geophysics, geomorphic mapping, 00:16:36.000 --> 00:16:45.000 and the question is, how to partition, slip from a large earthquake across these strands. So, do we put a 100 percent on each strand 00:16:45.000 --> 00:17:08.000 that would be very conservative. Do we do symmetrical partitioning 33 percent on each strand, or asymmetrical, and working with LCI came up with an asymmetrical partitioning system that is based on geomorphology. This middle strand 00:17:08.000 --> 00:17:15.000 is the best expressed. It offsets this historical rock wall that I mentioned earlier 00:17:15.000 --> 00:17:20.000 the most, and then also based on literature review 00:17:20.000 --> 00:17:31.000 looking at sandbox models. We've been funding work on new fault displacement models that can hopefully answer these types of questions. 00:17:31.000 --> 00:17:32.000 If you have multiple fault strands, how do you partition the slip? 00:17:32.000 --> 00:17:47.000 Greg Lavrentiadis at Berkeley and now UCLA with Norm Abrahamson developed a model that will allow us to do this under the UCLA fault 00:17:47.000 --> 00:18:01.000 displacement hazard initiative. So ending with research needs for the Concord-Green Valley Fault System specifically, it would be good to get better constraints on the slip rate. 00:18:01.000 --> 00:18:26.000 I showed that the uncertainty range for UCERF is 2 to 9mm a year, which is pretty large. Also better understanding the connectivity of the Concord fault to the Calaveras fault and Greenville fault to the South and how much slip is transferred from those 00:18:26.000 --> 00:18:29.000 faults to the Concord-Green Valley Fault System is important. 00:18:29.000 --> 00:18:33.000 There's no paleoseismic record, 00:18:33.000 --> 00:18:48.000 event record for the Concord fault, so that would be important to develop, but difficult because of all the development that's on the fault. And then slip per event would also be important. 00:18:48.000 --> 00:18:49.000 And then, in my previous examples, I discussed ongoing research. 00:18:49.000 --> 00:19:13.000 I should also mentioned that for fault complexity, we're working on the CRADA with Steve Delong and Austin Elliot and some post docs to develop a better understanding of what the predictors would be for fault complexity, and and how you would scale slip based on that so 00:19:13.000 --> 00:19:24.000 my closing question is the BAPEX initiative 20 to 25 years ago developed a lot of the paleoseismic data for the Bay Area. 00:19:24.000 --> 00:19:31.000 And do we need to revisit this with new methods and new sites, to develop more data for Concord-Green Valley and other faults? 00:19:31.000 --> 00:19:35.000 Okay. Thank you very much. 00:19:35.000 --> 00:19:44.000 Great. Thank you. Thank you, Chris. Sorry everybody for the slight snafu there. Gareth, we're gonna get back to you. 00:19:44.000 --> 00:19:55.000 John has asked that we go on to Lian Xue with PKU. I believe, that's what John wanted to do there, so Lian. 00:19:55.000 --> 00:19:59.000 Hello, everyone! What I'm talking today is about kinematics 00:19:59.000 --> 00:20:07.000 of the 2015 San Ramon, California earthquake swarm: implications for fossil structure and a driving mechanisms. I'm Lian from Peking University. 00:20:07.000 --> 00:20:10.000 Volume. [interruption] 00:20:10.000 --> 00:20:21.000 Here are my collaborators. In October 2015, there was an earthquake swarm that occurred near San Ramon, California. 00:20:21.000 --> 00:20:25.000 This swarm is indicated by the red color dots and beach balls. 00:22:29.000 --> 00:22:39.000 They are in an extensional rest step-over regime between the northern Calaveras fault and the Contra Costa Mount Diablo Thrust fault. 00:22:39.000 --> 00:22:48.000 This region has hosted 10 major swarms since 1970s they are indicated by the color dots here. Most of the selected swarms appear to occur 00:22:48.000 --> 00:23:01.000 along faults that are conjugate to the northern Calaveras fault at a high angle with a southwest strike. 00:23:01.000 --> 00:23:08.000 Earthquakes swarms represents transient increase in seismicity, and do not follow a typical main 00:23:08.000 --> 00:23:15.000 shock aftershock title, the elevated seismicity rate might increase the probability of a nearby larger earthquake. 00:23:15.000 --> 00:23:25.000 So it is important to investigate the evolution of the swarm and the driving mechanism. 00:23:25.000 --> 00:23:36.000 So, there are two questions that we want to address. The first one is, what is the spatial temporal evolution of the 2015 San Ramon swarm. 00:23:36.000 --> 00:23:40.000 The second one is, what are the possible driving magnetism? 00:23:40.000 --> 00:23:46.000 First, we use a template-matching detection to extend the events of the swarm. 00:23:46.000 --> 00:23:55.000 We use routinely cataloged events recorded by the Northern California Seismic System as waveform templates. 00:23:55.000 --> 00:23:56.000 Then use those templates to detect uncatalog small events to be similar 00:23:56.000 --> 00:24:20.000 waveforms through cross correlation. Paste, all the detected events we can, first to construct a matrix, including all the relative polarity measurements between other templates and the other detected events for all the data channel interface. 00:24:20.000 --> 00:24:34.000 Then we can decompose this matrix by using singular bio-digitization to extract the dominant signal from those numerous information. Based on this we can group the events into clusters 00:24:34.000 --> 00:24:44.000 with similar polarity pattern course out of the network. Events in each cluster should have similar focal mechanisms. 00:24:44.000 --> 00:24:50.000 Finally, we can use [indiscernilble] to estimate the catalog composite focal mechanism. 00:24:50.000 --> 00:24:56.000 All polarity observations in a cluster are used to constrain the focal mechanism. 00:24:56.000 --> 00:25:05.000 The details of this method that can be seen from Shelly et al. 2016 paper. 00:25:05.000 --> 00:25:14.000 Here are all the results. This is a short magnitude versus time. Colors indicate time sequence by using template matching detection. 00:25:14.000 --> 00:25:26.000 This swarm has over 4000 events; started from October 11 through November 18th. 00:25:26.000 --> 00:25:37.000 The smallest magnitude of this catalogue is about negative point 2, and the magnitude of the completeness is 0.2. 00:25:37.000 --> 00:25:43.000 The activity of this catalogue doesn't follow typical mainshock aftershock patterns. 00:25:43.000 --> 00:25:55.000 There are nine earthquakes larger than magnitude 3, and the largest one is M3.6, which occurred about 10 days after the initial of the swarm. 00:25:55.000 --> 00:26:03.000 Now we will use this catalog to illuminate those swarm structures and migration patterns. 00:26:03.000 --> 00:26:29.000 This is the radio showing an animation of the swarm activities, event waving or giving 6 hour long-time period are shown in color. Cool color, indicate earlier event. Warm color indicate later events in the 6 hour time windows. Gray dots are previous events. Right circles show the rupture 00:26:29.000 --> 00:26:34.000 dimension of the event which are larger than magnitude 2. 00:26:34.000 --> 00:26:38.000 By assuming three [indiscernible] stress drop. 00:26:38.000 --> 00:26:43.000 You can see the seismic activity started on the middle segment. 00:26:43.000 --> 00:26:46.000 First by migrating down deep followed by active extension 00:26:46.000 --> 00:26:54.000 days later. At the same time, you can see there was an initiation of the northern segment. 00:26:54.000 --> 00:27:02.000 It's seismicity first expand active, then propagate down deep. 00:27:02.000 --> 00:27:07.000 Ten days later the southern segment initiate a very small 00:27:07.000 --> 00:27:17.000 event. After a big earthquake it grows into a larger area. Later on those three segments slowly grew on the edges. 00:27:17.000 --> 00:27:26.000 Twenty-six days later the southernmost cluster of event occurred at the 4 kilometer far away 00:27:26.000 --> 00:27:30.000 and stayed in the small zone. 00:27:30.000 --> 00:27:47.000 Those events further away from the main event, and the number of those events is small we only focus on the three earlier segments which are relativity close in space. 00:27:47.000 --> 00:27:51.000 Here are the map view and 3D view of the swarm. 00:27:51.000 --> 00:27:55.000 The corners indicate the time sequence based on the map view. 00:27:55.000 --> 00:28:05.000 You can see all of the three segments have a clear southwest strike, and the base down the 3D view you could see 00:28:05.000 --> 00:28:14.000 all of them deeply deep into the northwest direction, and the middle segment has a "y" shape. 00:28:14.000 --> 00:28:36.000 The three segments, have a clear separation. The northern segment is located about 1.6 kilometers away from the middle segment. The southern segment is located about 1.3 kilometers away from the middle segment, and the depths of the major 00:28:36.000 --> 00:28:47.000 event is from 6 kilometers to 9 kilometers deep, and the depths range of the three segments slightly increased towards the south. 00:28:47.000 --> 00:28:54.000 Using the relative first motion polarity, here is our clustering result. 00:28:54.000 --> 00:29:02.000 Different colors indicate different clusters. Events in the same cluster have the same focal magnetism. 00:29:02.000 --> 00:29:09.000 You could see the hypocentral distribution of the clusters is very complex. 00:29:09.000 --> 00:29:24.000 Clusters in the middle, and the northern segment commingle together. To indicate this complex structure, we use ellipsoid to describe nodal plane feature. Ellipsoid is the 00:29:24.000 --> 00:29:46.000 same color indicates event within the same clusters. So based on the shape of the ellipsoids, you can see the hypocentral distribution of most clusters indicates a southwest striking and steeply northwest deep in structure, but there are some exceptions. The events of 00:29:46.000 --> 00:29:56.000 Cluster 2 in the middle segment, indicated by the blue color, have a nearly north striking at the west deep in structure. 00:29:56.000 --> 00:30:09.000 The events of cluster 4 in the northern segment indicated by the green color have a southwest striking, and the southeast deep in structure. 00:30:09.000 --> 00:30:15.000 The color beach balls show in the composite focal mechanism for each clusters. 00:30:15.000 --> 00:30:23.000 The right lines are the fitting planes of hypocentral distributions based on the principal component analysis. 00:30:23.000 --> 00:30:30.000 Multiple planes are fit to distinct subset of clustering events that occurred in 00:30:30.000 --> 00:30:41.000 different locations. Our preferred thought plane in the nodal plane in closed agreement with the hypocenter distribution. 00:30:41.000 --> 00:30:46.000 So you can see our open events have a similar focal mechanism. 00:30:46.000 --> 00:31:08.000 They are left lateral strike-slip events striking southwest at the deep end nearly vertical. By comparing the hypocentral distribution with the focal mechanism, you can see the hypocentral distribution has anti-clock wise rotation from the fault plane but consider the uncertainties 00:31:08.000 --> 00:31:21.000 of the focal mechanisms. The rotation of cluster 2, 3, and 4 are significant, and the rotation can reach as much as 20 degrees. 00:31:21.000 --> 00:31:26.000 So the multiple four sides with varying strikes in the swarm of clusters may reflect very complex default 00:31:26.000 --> 00:31:54.000 internal structure, which may involve an immature stage. After having the spatial temporal evolution of the swarm we want to explore the driving mechanism. The redistribution of fluid in the hydrogenous fault zone can 00:31:54.000 --> 00:32:01.000 cause earthquake swarms. To explore why there is evidence showing fluid involvement we compare the spatial migration of the swarm. 00:32:01.000 --> 00:32:12.000 We have a fluid diffusion front, the first outer formulation. There are three major bursts, the first one and the second one on the medium segment. 00:32:12.000 --> 00:32:35.000 The third one is on the northern segment. Those are the 2D plots for the bursts, the horizontal axis is the distance along the less direction, lens is down deep and getting deeper to the right. The vertical axis is the distance, along with, and the colors indicate the 00:32:35.000 --> 00:32:53.000 number of days since the start date of the burst. White dots shows the location of the event for the future burst; the black dots show the location of the events for the pre-dispersed. To explore the migration pattern 00:32:53.000 --> 00:32:59.000 we plot the distance of events from the initial point versus time. 00:32:59.000 --> 00:33:06.000 The black dash curves show the hypothetical slow, and fast diffusion front. 00:33:06.000 --> 00:33:13.000 The right dashed lines show the propagation distance assuming linear migration 00:33:13.000 --> 00:33:19.000 rate. So by this comparison you could say for the first burst 00:33:19.000 --> 00:33:25.000 that down deep propagation rates the fluid diffusion from water 00:33:25.000 --> 00:33:26.000 well, so this indicates it might be involved of fluid interaction 00:33:26.000 --> 00:33:37.000 process. For the second burst the initial propagation is 00:33:37.000 --> 00:33:47.000 very fast. It doesn't fit the fluid diffusion front, but it affects the linear migration. 00:33:47.000 --> 00:33:52.000 Migration rates as seen migration rate of slow slip 00:33:52.000 --> 00:34:00.000 events. At the later stage the burst has a slowly expansion at the edges of the rupture zone. 00:34:00.000 --> 00:34:10.000 So overall, it indicates it might include an unknown contribution of seismic slip. 00:34:10.000 --> 00:34:15.000 For the third burst the initial of the propagation fits the fluid diffusion 00:34:15.000 --> 00:34:20.000 very well. At the later stage, there is no expansion. 00:34:20.000 --> 00:34:24.000 This sort of event occurred at the rupture 00:34:24.000 --> 00:34:40.000 zone. So by all these comparisons here it shows the complicated spatial temporal evolution of the swarm, and this indicates a very complex supporting, and the possible fluid interaction process. 00:34:40.000 --> 00:34:48.000 It's also might include our own contribution of assessment for slip. 00:34:48.000 --> 00:34:53.000 Here are our conclusions. The 2015 San Ramon swarm activated 00:34:53.000 --> 00:34:59.000 three primary southwest striking, and Northwest-dipping 00:34:59.000 --> 00:35:00.000 fault zones. The focal mechanism resulted by composite catalog 00:35:00.000 --> 00:35:13.000 P-wave polarity and the hypothetical distribution of events indicates complex fault internal structures. 00:35:13.000 --> 00:35:31.000 The spatial-temporal evolution, the swarm indicates a complex faulting and a possible fluid interaction process, and might include an unknown contribution of aseismic fault slip. 00:35:31.000 --> 00:35:35.000 Thank you for your attention. Welcome any questions. 00:35:50.000 --> 00:35:51.000 Thank you guys for continuing to bear with our technical challenges. 00:35:51.000 --> 00:36:01.000 Next up. We're gonna hear from Artie Rogers talking about probabilistic, 00:36:01.000 --> 00:36:10.000 O,h that's the wrong one. A scenario earthquake simulation on the Greenville fault. 00:36:10.000 --> 00:36:17.000 This talk will describe some preliminary ground motion simulations for earthquakes on the Greenville fault near Livermore. 00:36:17.000 --> 00:36:21.000 This topic is close to work for me, and more than 8000 co- 00:36:21.000 --> 00:36:24.000 workers who work at Lawrence Livermore National Lab, 00:36:24.000 --> 00:36:31.000 and this photograph was taken from the main administration building, showing the Greenville fault lurking in the distance, less than a mile away. 00:36:31.000 --> 00:36:40.000 Parts of the lab, which is one mile square are just a few 100 meters from the fault. 00:36:40.000 --> 00:36:49.000 The Greenville fault is on the eastern edge of the San Francisco Bay Area, and is the last major fault going east toward the Central Valley. 00:36:49.000 --> 00:37:00.000 It is over 50 kilometers long, with a slip rate of one to 5mm per year, and as such is capable of earthquakes of magnitude 6.5 or larger. There have been earthquakes on the Greenville fault. 00:37:00.000 --> 00:37:17.000 The most recent large earthquake was a magnitude 5.8 in January of 1980, followed by a magnitude 5.4 aftershock. The most recent moderate earthquake was a magnitude 4.3 in July, 2019. The Livermore Valley is bounded by three faults and these basin 00:37:17.000 --> 00:37:21.000 edges play an important role in shaping ground motions that resulted in the [indiscernible]. 00:37:21.000 --> 00:37:27.000 In this talk I'll describe an attempt to model the largest of 1980 events. 00:37:27.000 --> 00:37:40.000 I'll also show results for a larger magnitude 6.2 scenario earthquake near Livermore, and this simulation shows complex basin and corner effects due to the geometry of the basin. 00:37:40.000 --> 00:37:51.000 The 1980 M5.8 main shock caused significant damage at Lawrence Livermore Lab, $10 million dollars in 1980, 00:37:51.000 --> 00:38:00.000 and it cost significant impact developments. And note that also that the surface rupture closed Highway 580. 00:38:00.000 --> 00:38:07.000 And today this highway carries 175,000 vehicles per day in each direction. 00:38:07.000 --> 00:38:16.000 The magnitude 5.8 mainshock in 1980 had clear north-to-south directivity, which contributed to the strong motion in Livermore. 00:38:16.000 --> 00:38:23.000 This is the described in a paper by Jack Boatwright and Dave Boore. Two strong motion to the south showed an acceleration as high as 0.26g 00:38:23.000 --> 00:38:38.000 at Del Valle Reservoir. But this reporting is probably contaminated by site and dam resonance effects, less bias estimate of the differential motion 00:38:38.000 --> 00:38:51.000 due to directivity is probably about a factor of four. And the magnitude 4.5 aftershock here on the right showed [indiscernible] of south-to-north directivity. 00:38:51.000 --> 00:39:06.000 Here is a map of the 1980 sequence which had over 300 events in 4 months, and for our scenario earthquake more than we considered a segment to the south of this activity. 00:39:06.000 --> 00:39:10.000 The goal of the study is to simulate ground motions, using SW4. 00:39:10.000 --> 00:39:11.000 SW4s is an anelastic 3D finite difference 00:39:11.000 --> 00:39:19.000 code for simulating seismic waves, and it has significant improvements 00:39:19.000 --> 00:39:23.000 in recent years. Thanks to the Department of Energy [indiscernible] 00:39:23.000 --> 00:39:28.000 scale computing projects. We use the USGS 3D model and run simulations on the [indiscernible] Livermore, 00:39:28.000 --> 00:39:48.000 and this follows other simulation studies that we published recently focusing on the Bay Area. A few years ago, I ran a simulation of a magnitude 6 rupture on the Greenville fault, though far to the south of Livermore and this resulted in the ShakeMap shown here on the right in terms of 00:39:48.000 --> 00:39:59.000 PGB. And we can see the strong ground shaking, extending into the sedimentary structure of the Livermore Valley, and then the following will revisit these simulations, using the improvements in SW4 00:39:59.000 --> 00:40:06.000 And so running high performance computers. 00:40:06.000 --> 00:40:23.000 This is the domain for the new simulations, with 15km x 40km x 35km in depth, and the simulations resolve frequencies up to 4 Hz using surface topography into natural [indiscernible]. And the surface trace of the 00:40:23.000 --> 00:40:36.000 magnitude 6.2 event is indicated here in red. These simulations require a 1 billion grid points and run on 12 nodes on Lassen in less than a minute. 00:40:36.000 --> 00:40:42.000 Ruptures were generated with the Graves & Pitarka methodology, and really just focused on 2 ruptures. 00:40:42.000 --> 00:40:48.000 here, the 5.8, 1980 mainshock, and this 6.2 00:40:48.000 --> 00:41:00.000 scenario earthquake to the south. And here we show the slip distributions which use a slip cache that concentrate the still toward the center of the file service. 00:41:00.000 --> 00:41:16.000 Here we zoom-in on the near fault region, showing the Greenville fault from the Livermore Valley locations of the three ruptures I've shown. In red, we have the 1980 mainshock of the north-to-south directivity. Cyan is the 5.4 aftershock and blue is 00:41:16.000 --> 00:41:34.000 the magnitude 6.2 scenario earthquake, which cuts closest to the population center and the Lawrence Livermore Lab. Also note that the Las Positas fault is indicated here by the green dash is from the southern edge of the Livermore Basin and conjugate fault to the Greenville fault. 00:41:34.000 --> 00:41:46.000 And there are many important critical infrastructure features that are located on or crossing the Greenville fault, including Highway 580, 00:41:46.000 --> 00:41:54.000 which carries a great deal of traffic; water for Hetch-Hetchy and EBMUD crosses the Greenville fault; 00:41:54.000 --> 00:42:04.000 Or close to the Sacramento-San Joaquin Delta, and of course, gas and electric lifelines through the regions as well. 00:42:04.000 --> 00:42:10.000 Here are some images of the USGS 3D models rendered on to our SW4 00:42:10.000 --> 00:42:18.000 domain, and these define Livermore Valley and Livermore Basin. On the left we show the shear wave speed at the surface and the minimum speed in this area is 500 m/s, 00:42:18.000 --> 00:42:25.000 although geotechnical management set within their lab indicate VS30 is 350 m/s. 00:42:25.000 --> 00:42:39.000 In the center we have the depth to the 1.0 km/s shear wave speed, and to the right the 2.5 km/s shear wave speed. 00:42:39.000 --> 00:42:47.000 And these very clearly show the basin edges defined by the Greenville fault, 00:42:47.000 --> 00:42:52.000 Las Positas fault, and the Calaveras fault. 00:42:52.000 --> 00:43:07.000 Here are PGB ShakeMaps from our simulations for the magnitude 5.8 1980 event on the left and the rupture indicated by the black, and on the right we show the M6.2 scenario earthquake. 00:43:07.000 --> 00:43:13.000 We see very strong shaking, very strong asymmetric shaking across the Greenville and Las 00:43:13.000 --> 00:43:18.000 Positas faults due to the Livermore Basin structure shown in the previous slide. 00:43:18.000 --> 00:43:21.000 It's notable that shaking for the M6.2 00:43:21.000 --> 00:43:22.000 scenario shows strong shaking along the Greenville and Las 00:43:22.000 --> 00:43:37.000 Positas faults in the associated basin edges of the Livermore Basin, and there's a strong band of shaking along the diagonal from the basin [indiscernible]. 00:43:37.000 --> 00:43:47.000 Our first attempt to model this 1980 M5.8 events does reproduce the directivity of some of the directivity scene. 00:43:47.000 --> 00:44:05.000 We see a factor of two differential between a north and a south site compared for sites that were considered by Boatwright and Moore; however, this doesn't reproduce the larger variations that they saw in the data and this is gonna have to wait for further study, 00:44:05.000 --> 00:44:16.000 and more analysis of the limited available data, and probably the consideration of the rupture analysis. 00:44:16.000 --> 00:44:24.000 Here's an animation of the magnitude 6.2 main shock, and this shows the PGB 00:44:24.000 --> 00:44:28.000 in terms of the magnitude of the horizontal velocity. 00:44:28.000 --> 00:44:39.000 We see the usual basin and edge effects here of the motions racing out ahead, and the hard rock, and delayed and amplified within the basin 00:44:39.000 --> 00:44:44.000 both along the Greenville fault, but also the Las Positas fault as well. 00:44:44.000 --> 00:44:58.000 Both of these basin edges generate basin edge generated surface waves, and they constructively interfere to amplify the motion within the Livermore Basin, and they find a strong pulse. [Indisernible] 00:44:58.000 --> 00:45:00.000 They form a strong pulse that travels into the basin 00:45:00.000 --> 00:45:15.000 seen here and this is followed by a very slow surface wave that travels in the diagonal from the Basin. 00:45:15.000 --> 00:45:33.000 So this rupture plumps a lot of energy into the very corner of the basin of two orthogonal basement edges and basin edges generated waves are formed on either of these boundaries of the basin, and they result in strong motion within the 00:45:33.000 --> 00:45:45.000 basin, and at late times we also see that the waves are reflected back by the more distant edges of the basin, such as the Calaveras fault. 00:45:45.000 --> 00:45:51.000 Alright, here are ground motions for the magnitude 6.2 scenario. 00:45:51.000 --> 00:45:58.000 Here for a site in Livermore on the top, and a site near Livermore lab at the bottom 00:45:58.000 --> 00:46:09.000 here. We can see the direct arrivals have a complex nature that results from the interfering surface waves generated from either edge of the basin. 00:46:09.000 --> 00:46:16.000 The Greenville being the primary, a source of the basin edge generated wave, but also Las Positas 00:46:16.000 --> 00:46:23.000 fault generates basin edge, generated basin, and then these are seen in the lead. 00:46:23.000 --> 00:46:33.000 Here we have velocity, on the left, and acceleration on the right, and there is the later slower wave that propagates along the diagonal of the scene 00:46:33.000 --> 00:46:42.000 it's a longer duration, longer period pulse. 00:46:42.000 --> 00:47:02.000 In terms of response spectra, here are the response spectra for two sites, one within the basin at Livermore, Livermore lab, and another just outside of the basin, and these RotD50 response spectra are compared to [indiscernible] two ground motion models with the 00:47:02.000 --> 00:47:07.000 response spectra that we only resolve the periods indicated by the gray shading with the number 00:47:07.000 --> 00:47:13.000 R, our maximum frequency is 4Hz or 0.25s period. Now within the basin the motions are uniformly higher than the [indiscernible] 00:47:13.000 --> 00:47:24.000 West ground motion models. However, outside the basin the motions are consistent with the [indiscernible] West models with a slight amplification 00:47:24.000 --> 00:47:33.000 between, say, 1 and 4Hz. So outside the basin, you know the fact that our motions agree or are recently well modeled or consistent with the NGA 00:47:33.000 --> 00:47:47.000 West too, gives us confidence that the absolute value of the mission is reasonable, and the differential between these response spectra really indicate 00:47:47.000 --> 00:47:55.000 the strength of amplification, and in this case you know this double edge or basin corner effect. 00:47:55.000 --> 00:48:06.000 We got basin edge generated waves from two orthogonal edges of the basin 00:48:06.000 --> 00:48:14.000 So in conclusion, we perform preliminary simulations of moderately large magnitude earthquakes on the Greenville fault using 00:48:14.000 --> 00:48:21.000 SW4 (Graves & Pitarka) rupture models and most 00:48:21.000 --> 00:48:32.000 recent USGS 3D models for the region. Ground motion show a complex pattern of response that are strong asymmetries due to geologic structure could be across the 00:48:32.000 --> 00:48:40.000 Greenville and Las Positas faults. You see, amplification in the Livermore Basin due to the sedimentary geology there, 00:48:40.000 --> 00:48:48.000 and importantly, we see this basin corner effect caused by the interfering basin edge generated waves from the orthogonal 00:48:48.000 --> 00:48:57.000 geometry of the basin edges. The coast ranges south of Livermore and Las Positas fault 00:48:57.000 --> 00:49:04.000 they clearly form a barrier to strong shaking comparison of ground motions with ground motion models. 00:49:04.000 --> 00:49:18.000 Periods greater than 0.25s, or frequencies, less than 4Hz is quite good outside of the basin we see our motions are consistent with the ground motion models, but within the Livermore Basin 00:49:18.000 --> 00:49:32.000 we see amplification due to this double edge basin formal effect. Our simple attempt to model the 1980, M5.8 rupture really fails to reproduce the strong directivity 00:49:32.000 --> 00:49:40.000 reported by Boatwright and Moore and this will require some further study to try to understand the available data. 00:49:40.000 --> 00:49:45.000 An investigation of moderate events in this region would be informative for evaluating 00:49:45.000 --> 00:49:51.000 USGS 3D model simulated waveforms and intensities we get from simulations and I'll stop there. 00:49:51.000 --> 00:50:00.000 Thank you very much. 00:50:00.000 --> 00:50:02.000 Thank you Arnie. That was great. Some great discussions in the chat area 00:50:02.000 --> 00:50:05.000 there too. We're on to Gareth Funning, University of California at Riverside. Here we go. Thanks! 00:51:50.000 --> 00:51:58.000 Hello! I'm Gareth Funning, and my talk today is going to be about our attempts to measure deformation between the San Francisco Bay Area and the Sacramento-San Joaquin Delta area. 00:51:58.000 --> 00:52:21.000 I just call it the Delta. This is work that I have been working on with Mike Floyd from MIT, and thank you to the conveners of the session for inviting me to talk. The picture you can see here is of Montezuma Slough from last summer. 00:52:21.000 --> 00:52:25.000 Last, September, when we were on our last field campaign in this area. 00:52:25.000 --> 00:52:35.000 This is a GPS receiver a GNSS receiver that we set up overlooking the slough on the bridge; on the bridge abutment, which is a nice stable benchmark. 00:52:35.000 --> 00:52:44.000 I remember this day very well, because it was 114 degrees, and we concentrated very strongly on our hydration this day. 00:52:44.000 --> 00:52:47.000 We made it through alive. Why are we interested in this? 00:52:47.000 --> 00:52:53.000 It turns out that an earthquake in this area could be very, very dangerous. 00:52:53.000 --> 00:52:58.000 It could produce a large risk to the whole State of California. 00:52:58.000 --> 00:53:04.000 This is a picture for a screen grab of the California Department of Water Resources website. 00:53:04.000 --> 00:53:11.000 And this quote is from the first paragraph that lays out how the Delta is the hub of California's water supply. 00:53:11.000 --> 00:53:21.000 It supplies fresh water to two-thirds of the State's population, mostly in the large population areas in Southern California and the Bay Area, and also to millions of acres of farmland. 00:53:21.000 --> 00:53:29.000 If an earthquake happened that disrupted this water supply, the results could be catastrophic to the state. 00:53:29.000 --> 00:53:39.000 This is a picture of a figure that I've made, showing the active faults in California, these red lines and these pink ones, and also the areas of liquefaction. 00:53:39.000 --> 00:53:50.000 These are highlighted in pink. The Delta is here, and you can see that within the Delta is a significant amount of land which is at high risk of liquefaction. 00:53:50.000 --> 00:54:05.000 Most of the the channels and levees in here that you can see here would be at risk, and the only reason that more of this area isn't covered in pink is that the data set I plotted ends right at the edge of the the 9 counties of the Bay Area which in here 00:54:05.000 --> 00:54:11.000 I would anticipate that nearly all of the Delta would be at risk. 00:54:11.000 --> 00:54:17.000 Zooming in a little. The major fault that is inferred to exist here is the Midland fault. 00:54:17.000 --> 00:54:27.000 This is found in seismic reflection profiles that were collected for gas. Gas exploration in the 1980s, and it is mapped through most of the Delta. 00:54:27.000 --> 00:54:43.000 Obviously, if if an earthquake happened on this fault, it could be a magnitude 6 or larger, and that would generate large shaking in the area, which would also induce liquefaction. So one thing that we really want to understand is how much strain is accumulating in this area and how 00:54:43.000 --> 00:54:54.000 much is accumulating on the Midland fault that could drive a future earthquake, and that is why we have been measuring with GNSS the strain across this area. 00:54:54.000 --> 00:54:57.000 We're not the first people to work in this area 00:54:57.000 --> 00:55:16.000 broadly speaking this is a figure from a paper right by Will Prescott, from the early 2000, and it shows the plate boundary normal velocity, so that if you drew a profile from Point Reyes to the Great Valley through Davis more or less that's where it goes 00:55:16.000 --> 00:55:21.000 and measured the displacement along, or the velocity along that profile 00:55:21.000 --> 00:55:40.000 this is what you get. So there's about 4mm a year of total difference between the plate boundary zone which starts at the Farallon Islands and heads basically to the edge of the Coast Ranges between that zone and the Great Valley blocks which includes Davis and 00:55:40.000 --> 00:56:00.000 most of the small towns in the Great Valley. There's about 4mm a year of relative displacement between those things, those two places, and that is manifested across a zone just as wide. So this implies that there's a shortening across the zone might drive 00:56:00.000 --> 00:56:07.000 thrust faulting and the Midland fault is inferred to be a reverse fault. So can we 00:56:07.000 --> 00:56:08.000 improve on the resolution of the data that you could collect in this area. 00:56:08.000 --> 00:56:19.000 Could we see how extensive and how concentrated this conversion zone is? 00:56:19.000 --> 00:56:22.000 Can we identify it? Can we localize it to false? 00:56:22.000 --> 00:56:27.000 That is why we are interested in doing GNSS field work in this part of the world. 00:56:27.000 --> 00:56:36.000 We've been measuring velocities in this region since 2008, a group from UCR/MIT with collaborations with Mike Floyd at MIT 00:56:36.000 --> 00:56:40.000 and with the help of lots of field assistants. This is our data set. 00:56:40.000 --> 00:56:46.000 This also includes data collected by the USGS, and also data from continuous stations. 00:56:46.000 --> 00:56:47.000 Our intent from measuring this was to understand the plate 00:56:47.000 --> 00:56:52.000 boundary, parallel velocities and 00:56:52.000 --> 00:57:06.000 also data from continuous stations. Our intent from measuring this was to understand the plate boundary parallel velocities, and the slip rates of the major active faults in this area, and we have modeled them and basically the deformation 00:57:06.000 --> 00:57:08.000 you can see here can be fit by 3 major fault systems through here, 00:57:08.000 --> 00:57:21.000 San Andreas, Rogers Creek, Maacama, and the Green Valley fault system, and that accounts from nearly all the deformation you see in this map. 00:57:21.000 --> 00:57:38.000 And while that's great for studying, the most active faults in the region. Obviously, if we're interested in studying deformation in this area, the Delta, we have a big hole to fill and our next thing to do is to try and figure out how to fill that gap. 00:57:38.000 --> 00:57:45.000 Plus to add new velocities to our map, we need to measure sites that already have 00:57:45.000 --> 00:57:56.000 GNSS measurements made at them. Luckily for us, the California Department of Water Resources had made two campaigns in 2011 and 2017 at all 00:57:56.000 --> 00:58:08.000 these sites, marked by triangles. So, in order for us to get new camp velocity estimates, we picked a subset of these sites and went and remeasured them again 00:58:08.000 --> 00:58:24.000 in the summer of 2022. However, one concern was that many of the benchmarks were in sites that may not give you a measurement that is actually diagnostic or representative of the motion of the crust rather a motion that's non-tectonic. 00:58:24.000 --> 00:58:34.000 We know this is an issue, because the levees in the Sacramento Delta area have been known to be subsiding for quite some time. 00:58:34.000 --> 00:58:47.000 This area is formerly a tidal marsh, and they accumulated a great thickness of peat in these areas and that peat over time has been oxidizing through microbial action and releasing CO2. 00:58:47.000 --> 00:59:06.000 So the carbon in the peat is basically being expelled as gas, which means that over time the amount of mass in the levee system is reducing, and the fear is that this will cause maybe the levees to fail all on their own even without an earthquake, 00:59:06.000 --> 00:59:15.000 but of course the fluid, saturated, substrate, and the mud and everything else that's in here makes it also much more susceptible to liquefaction. 00:59:15.000 --> 00:59:16.000 But a fear is that if we do GPS GNSS 00:59:16.000 --> 00:59:28.000 measurements in this area, then we might just be measuring the movement of the levee and not the movement of the crust underneath it. 00:59:28.000 --> 00:59:41.000 This fear is not unfounded. This is a good Google street view picture of a road in the Sacramento Delta area, and you can see that the road is full of cracks and repairs. 00:59:41.000 --> 00:59:42.000 And it was actually this side of the road when I drove on 00:59:42.000 --> 00:59:46.000 it was several inches higher than the other side of the road. 00:59:46.000 --> 00:59:48.000 It's quite gnarly. These are classic signs of lateral spreading. 00:59:48.000 --> 01:00:08.000 This is motion of outward motion that's stretching the road and breaking it, and if we were to make measurements here, maybe that's what we would be measuring and not the movement of the the crust underneath. This is just another picture gratuitously showing the location of the 01:00:08.000 --> 01:00:20.000 survey benchmark is actually over here. We found it, but we declined to measure it, because you could see in the road all these cracks again, and this road surface is broken up completely. 01:00:20.000 --> 01:00:23.000 This is another area where lateral spreading is occurring. 01:00:23.000 --> 01:00:30.000 So we had to figure out a way of selecting which sites we were going to measure without subjecting ourselves to 01:00:30.000 --> 01:00:36.000 maybe collecting things that would not be useful in the long run. 01:00:36.000 --> 01:00:41.000 And so what we used to solve this problem was InSAR. 01:00:41.000 --> 01:00:44.000 Of course, I like to talk about it. It's a technique that can measure deformation over large areas. 01:00:44.000 --> 01:00:51.000 It's a satellite based, remote, sensing technique. 01:00:51.000 --> 01:00:58.000 And in this particular case, and what I'm going to show you here is that you can actually use it to show areas that are affected by subsidence 01:00:58.000 --> 01:01:06.000 that is not tectonic. So I've been working with Sim Sangha he's pictured here from JPL 01:01:06.000 --> 01:01:15.000 for several years now to produce these maps in a SCEC-funded project called the Community Geodetic Model. 01:01:15.000 --> 01:01:28.000 The novelty here is that we were using standard product interferograms which were produced by JPL basically processed inside data so that you can use the data without having to process it. 01:01:28.000 --> 01:01:34.000 And then we just solved for the best fitting velocities from a stack of interferograms. 01:01:34.000 --> 01:01:48.000 And this is what they look like. At the top are the velocities in the bottom of the uncertainties, and you can see we have the whole state of California is covered here, but is shown in this kind of exploded view because these tracks 01:01:48.000 --> 01:01:58.000 overlap. So yeah, California isn't really this wide, but California is covered by nine tracks of the sentinel, 01:01:58.000 --> 01:02:08.000 one satellite, and in these images blue colors mean the ground is moving relatively away from the satellite the satellite line of site shown by these pink arrows. 01:02:08.000 --> 01:02:09.000 If you see the same feature, the same color in both the ascending and descending data sets 01:02:09.000 --> 01:02:19.000 that generally means the ground is moving vertically, and if it's blue in both the set data sets, it's moving vertically away. 01:02:19.000 --> 01:02:30.000 So this blue thing here is the the agricultural part of the Central Valley which is subsiding due to groundwater extraction, 01:02:30.000 --> 01:02:38.000 we think. You can see that the San Andreas fault is marked as a very abrupt change in velocity, which changes cyan between 2 data sets. 01:02:38.000 --> 01:02:42.000 That means it's moving horizontally that's the creep on the San Andreas. 01:02:42.000 --> 01:03:00.000 But for the purposes of this talk, the pertinent feature is this area out here the Delta, where the majority of the velocities are bluer than their surroundings, which means, again, that the ground is moving downwards, and this is signature of the subsidence in that 01:03:00.000 --> 01:03:06.000 area. Zooming in a little. I'm showing this 01:03:06.000 --> 01:03:14.000 inside later again at the ascending track, which is not very sensitive to fault parallel movement, so it's showing nearly everything you can see. 01:03:14.000 --> 01:03:34.000 Here is vertical movement, and many of the sites, as you can see, these triangles are in locations which might be subsiding, so we were careful to avoid those, and we selected a set of of stations that were mostly out of the blue, and to the west of the Delta and that's 01:03:34.000 --> 01:03:39.000 what we measured last summer. Just to say I didn't do this on my own. 01:03:39.000 --> 01:03:49.000 Mike came along. There he is, and 2 graduate students from UCR were also there, and I'm happy to say we all made it out of there alive. 01:03:49.000 --> 01:04:09.000 Having made it back to the lab with our data, we found that when we tried to process the previous campaign data, actually, the 2011 data that they were being thrown out by the processing software due to having faulty information in them, it turns out the L2 pseudorange was the same as the 01:04:09.000 --> 01:04:15.000 L1 pseudorange, which is weird and unexpected and wrong. 01:04:15.000 --> 01:04:24.000 But because Mike is a GPS processing genius, he figured out a way of getting those data processed so that we could make a velocity with the measurements we'd made in 2022. 01:04:24.000 --> 01:04:37.000 The trick was to treat them as if they were data from the codeless receiver which, as I understand it, is a strategy that was employed in the early days of GPS. 01:04:37.000 --> 01:04:54.000 So this is old school. So, having figured that out, we went from a map that had this coverage, with this coverage, with about 15 extra stations out here in the East. 01:04:54.000 --> 01:05:04.000 What's interesting is that these sites in the east seem to have a motion which is much more northeastward, at a rate of around 2mm a year. 01:05:04.000 --> 01:05:12.000 In contrast with the sites out in the west, which seem much more plate boundary parallel. 01:05:12.000 --> 01:05:30.000 It seems you can separate the sites out in the east here from the sites out in the west by drawing on the boundary between these two domains, which corresponds very closely with the range front of the Coast Ranges, which runs up to the east of Lake Berryessa, here. 01:05:30.000 --> 01:05:32.000 It divides this into these two zones. In the west, 01:05:32.000 --> 01:05:54.000 stations are moving much more parallel to the faults and to the plate boundary in general, and this plate boundary parallel motion seems to end the last major strike-slip faults, the Green Valley fault out here. Out in the east this is a 01:05:54.000 --> 01:06:10.000 zone of more plate boundary perpendicular motion, and if you look at the trend that these these stations are moving, it's actually quite close to perpendicular to the strike of the Midland fault suggesting indeed, that the interpretations of this being a 01:06:10.000 --> 01:06:18.000 reverse fault are supported by the contemporary deformation in this area. 01:06:18.000 --> 01:06:19.000 So that is the the headline result of our preliminary work in this area. 01:06:19.000 --> 01:06:37.000 It seems that the defamation and the delta appears to transition from plate boundary parallel in the West to plate boundary perpendicular in the East, as you cross the Coast Ranges the motion in the east and just to the west of the Delta is consistent 01:06:37.000 --> 01:06:46.000 with the interpretation of the middle and fault being a reverse fault, the deformation is complicated by the nontectonic subsidence of the Delta levees 01:06:46.000 --> 01:06:52.000 this makes it difficult to make GPS measurements or GNSS measurements that actually capture the tectonic motions 01:06:52.000 --> 01:06:59.000 and we have hit on the strategy of using our InSAR velocity maps to help identify sites that may be more stable. 01:06:59.000 --> 01:07:15.000 Obviously, we will try to find geological associations with these stable areas and try and use the map, the map mythologies as well as as a guide to this, and we still want to do more of that, because we do not yet have good deformation constraints in the east of the area. 01:07:15.000 --> 01:07:22.000 So east of the the Delta we don't have any data at all at the moment, and our future goal is obviously to go and expand out there. 01:07:22.000 --> 01:07:28.000 And that's what we hope to do in our next field trips. 01:07:28.000 --> 01:07:36.000 So thank you for your attention. I want to thank a whole slew of field assistants who collected data over the years and funding from the USGS 01:07:36.000 --> 01:07:44.000 and SCEC. Thanks. 01:07:44.000 --> 01:07:45.000 Thanks, Gareth! That was, excellent. Thank you for that. Thank you. All. :24:28.000 --> 03:24:45.000 Speakers, Chris, Dan, and Artie, very good topic for the Delta area 01:08:02.000 --> 01:08:06.000 Alright! 01:08:06.000 --> 01:08:09.000 Chad you out there? 01:08:09.000 --> 01:08:10.000 Yeah. 01:08:10.000 --> 01:08:17.000 Great, alright. 01:08:17.000 --> 01:08:26.000 Alright, so I'm going to start off. I have a question for for Artie on his interesting enough. 01:08:26.000 --> 01:08:41.000 You know this, all of these topics are very interesting to me specifically, because the California Department of Water resources has the California State Water project specifically, the South Bay Aqueduct in the East Bay. 01:08:41.000 --> 01:08:46.000 But going through the Delta, you know the headwaters to the California awkward are right there. 01:08:46.000 --> 01:08:53.000 So the the aqueduct and water transportation very vulnerable to to all the seismic issues. 01:08:53.000 --> 01:08:58.000 The Livermore Valley, especially so. Artie you brought up this point, which I thought was very interesting. 01:08:58.000 --> 01:09:14.000 The amplification in the SW4 simulation, where we've got the magnitude 6.2 01:09:14.000 --> 01:09:16.000 [indiscernible technical problems] any interference or or effects from the Verona fault. 01:09:16.000 --> 01:09:24.000 The thrust fault that dips under the Livermore Valley. 01:09:24.000 --> 01:09:25.000 As the animation was going on, I was looking pretty closely. 01:09:25.000 --> 01:09:36.000 Maybe not, but it's on our radar as a big seismic potential seismic source that the Devil Dam which is ours hangs there on the hanging wall. 01:09:36.000 --> 01:09:41.000 Did you see any effects in wave propagation from the Verona fault? 01:09:41.000 --> 01:09:47.000 Well, let's see. So the I don't know that those faults are explicitly in the model. 01:09:47.000 --> 01:09:55.000 I had a little back and forth with Evan Hirakawa last week, and I think I'm inferring 01:09:55.000 --> 01:10:11.000 Las Positas fault; I've never heard of the Verona fault so, but I believe it's coincident, you know, with the Las Positas fault or nearby forming sort of this southern edge of the Livermore Basin and if that's 01:10:11.000 --> 01:10:14.000 the case, then, yeah, that's the feature, you know, 01:10:14.000 --> 01:10:21.000 that's the structural impedance contrast between the hard rock kind of Diablo Range south of Livermore and the sedimentary structure in Livermore Basin. 01:10:21.000 --> 01:10:44.000 So what was so, you know, was interesting about this, and unexpected, you know I not until I looked at the seismograms in the animation, did I really see this sort of two orthogonal basin edges you know generating basin edge surface waves that travel into the valley. 01:10:44.000 --> 01:10:50.000 So that was pretty interesting. And just to back up, I mean on the infrastructure. 01:10:50.000 --> 01:11:07.000 You know, there's important infrastructure, very important infrastructure, water infrastructure crossing the Greenville fault to the south of Livermore, the Hetch Hetchy and EBMUD, and you know, but from the ground motion perspective you know I'd 01:11:07.000 --> 01:11:20.000 be more concerned about like fault displacement and near fault motions there, and in this particular case, those ruptures don't really intersect that infrastructure. 01:11:20.000 --> 01:11:31.000 I've run other simulations that generate a lot of motion in the Delta, there's a lot of uncertainty about the velocity and all the seismic properties of the Delta, so that needs to be taken into account. 01:11:31.000 --> 01:11:38.000 I kind of chose this M6.2, because you know, it's sort of not so large, 01:11:38.000 --> 01:11:44.000 has a very long return period, but is just a little bit bigger than the 1980 events. 01:11:45.000 --> 01:11:48.000 But it's also has this proximity to the infrastructure transportation; 01:11:48.000 --> 01:11:59.000 A lot of people live there. There's a lot of electric lines coming across there, and there's gas coming along, a big gas pipeline on Vasco road, and another one coming in on 580. 01:11:59.000 --> 01:12:14.000 So there is a confluence of a lot of infrastructure there, and of course, the 8,500 people that I work with at Livermore, you know, will be strongly affected, too. 01:12:15.000 --> 01:12:16.000 So we have that interest. So I mean, I'm kinda curious to learn more about this. 01:12:16.000 --> 01:12:20.000 Evan has a nice, cross-section, or a nice kind of 3D 01:12:20.000 --> 01:12:31.000 view of the basin, you know, and I it's the sort of the dip of these false and these structural features that shallow depths are really gonna control the ground shaking. 01:12:31.000 --> 01:12:38.000 And it's right there by Del Valle also. So you know, I think like a lot of good research 01:12:38.000 --> 01:12:44.000 we ask a lot more questions after we see some of these results 01:12:44.000 --> 01:12:50.000 Great. Thanks, Artie. Chad? 01:12:50.000 --> 01:12:58.000 I was looking back to see in the chat there any other questions for Artie. While I do that, if anybody has any. 01:12:58.000 --> 01:13:11.000 I'll take them now. Got a couple new ones for Gareth. 01:13:11.000 --> 01:13:19.000 Okay, I should probably be mining like my video going out something 01:13:19.000 --> 01:13:20.000 So real, quick before we move on to Garrett's talk. 01:13:20.000 --> 01:13:26.000 There's one additional question, for already from Ruth Harris in the chat. 01:13:26.000 --> 01:13:29.000 What do you think about the effect of fault creep? 01:13:29.000 --> 01:13:33.000 Hmm. 01:13:33.000 --> 01:13:39.000 All I know of this is from Harris and Abrahamson. [laughing] 01:13:39.000 --> 01:13:45.000 So I don't know. I don't know. I think you know there's a lot of uncertainties. 01:13:45.000 --> 01:13:54.000 There's, you know, we have to consider a lot of rupture models and maybe modulate the slip. 01:13:54.000 --> 01:13:55.000 I think there's big questions about paleoseismic in this area, too. 01:13:55.000 --> 01:14:04.000 So. 01:14:04.000 --> 01:14:11.000 Did you have one from the comments section Chad? 01:14:12.000 --> 01:14:21.000 I think that they mostly have been addressed at this point, although if anybody feels strongly that they have not had their question already answered, feel free to let us know. 01:14:21.000 --> 01:14:42.000 In the meantime, since we've got Garrett here. Garrett, I'm curious about your InSar data and your focus on the Midland fault specifically, because that's not the only fault out there. And in fact, on the maps that you were showing you had a couple of others 01:14:42.000 --> 01:14:48.000 that have a similar strike, that might be more or less along that 01:14:48.000 --> 01:14:55.000 break in vector direction. I'm thinking specifically about the the Pittsburgh Kirby Hills fault and the Great Valley Thrust System further north 01:14:55.000 --> 01:15:10.000 and I'm just curious whether you thought anything about additional structures and sort of how those fit into your model, or if that's sort in the future. 01:15:10.000 --> 01:15:22.000 That is definitely in the future. I'm happy to be instructed by people who know about the full thing out in that area better than I do, certainly, and we've only just started working there. It's true, 01:15:22.000 --> 01:15:37.000 yes, most of our stations are east of the Kirby Hills fault, which suggests that anything we see between that and the Great Valley block which is 0 on our maps would have to be a be expressed over structures to the east but yeah, you're right that at the 01:15:37.000 --> 01:15:45.000 point where the strike or the the orientation of our vectors changes is approximately the location of the Kirby Hills fault. 01:15:45.000 --> 01:15:50.000 I have a question for Chris Medugo, Chris. 01:15:50.000 --> 01:16:00.000 There at the location, interesting where your study was, you've got the junction of Highway 80, Highway 12, 680, an abundant of infrastructure, including PG&E gas transmission 01:16:00.000 --> 01:16:06.000 and electrical. Department of Water Resources 01:16:06.000 --> 01:16:26.000 we have our North Bay aqueduct going through there as well, and multiple telecommunications with the paleoseismic work that was done specifically on the Green Valley just to the east, of less than a mile, we have the Cordelia fault and I'm 01:16:26.000 --> 01:16:31.000 wondering has any evaluation from PG&E's perspective been done, or is that on your radar? 01:16:31.000 --> 01:16:42.000 You know, as far as it being so proximal to the Green Valley and potential future paleoseismic work 01:16:42.000 --> 01:16:46.000 Yes, [cough] excuse me, we did look at the Cordelia 01:16:46.000 --> 01:17:02.000 fault again with LCI. We did trenching, the work that we did was focused on location because for the gas lines that we have in the area location was really the most important factor. 01:17:02.000 --> 01:17:11.000 And so we pinned the location of the active strands of the fault, and I forget whether we need to do a mitigation there. 01:17:11.000 --> 01:17:30.000 The area that you mentioned that I talked about in my talk a lot of that work was initially done in the early 2000's, and there was mitigation done, and then we re-studied it again a few years ago and we'll probably be updating the mitigation and that 01:17:30.000 --> 01:17:40.000 highlights that we're just doing continuous improvement once we've mitigated a line we'll take new information down the road and use that for additional mitigations as necessary. 01:17:40.000 --> 01:17:51.000 But I believe we do have you know trench logs of of the location of the Cordelia fault and we came up with parameters for that based on the work that we did. 01:17:51.000 --> 01:18:11.000 I also wanted to mention for the the Midland fault if I'm remembering correctly, I think Chris Hitchcock and Jeff Unruly did a NEHRP back in the day, looking at gross strata and thickening of beets right against that structure showing 01:18:11.000 --> 01:18:25.000 that at that latitude that was the active structure, or the most active structure. 01:18:25.000 --> 01:18:29.000 I think you're muted. 01:18:29.000 --> 01:18:40.000 Thanks. The Concord fault, the urbanization there makes it very difficult for paleoseismic studies, are there prospective sites? 01:18:40.000 --> 01:18:45.000 You might be looking at that, have some potential in that area? 01:18:45.000 --> 01:18:46.000 Not at this time; it's really urbanized. 01:18:46.000 --> 01:18:51.000 It's just so hard. But if you look at maps of downtown Concord, there's trenches everywhere, just for AP 01:18:51.000 --> 01:18:54.000 Studies, so the fault is really well located. It's been identified in a lot of trenches. 01:18:54.000 --> 01:19:11.000 None that I know of are of, you know, research quality. Maybe there's stuff out there in the great literature that some consultants may be aware of. 01:19:11.000 --> 01:19:18.000 But right now we don't have our sites on any good paleoseismic sites. One thing that we are focusing on 01:19:18.000 --> 01:19:34.000 as I mentioned, working with Ben Brooks, is just getting a better handle of creep through the urban environment, using truck mounted terrestrial lidar, and for our gas lines, what we're really interested in is not creep right at the surface where you can see the 01:19:34.000 --> 01:19:54.000 offset curves, but we're interested 3 or 4 feet below the ground surface, and if you imagine that those creeping fault strands could be flowering up to the surface and more distributed at the surface and more focused deformation as you extend deeper beneath the 01:19:54.000 --> 01:19:56.000 surface even in the upper meter. That's important for our gas line. 01:19:56.000 --> 01:20:01.000 So that's a an area of research focus for us right now. 01:20:01.000 --> 01:20:03.000 Real good. Good. Gareth, with the interferometry work you did 01:20:03.000 --> 01:20:18.000 that was, from my perspective, pretty innovative. It sounded like you're doing something where you were stacking the interferometry, and it reduced the processing time. 01:20:18.000 --> 01:20:21.000 It actually sounded like you were able to get results very quickly. 01:20:21.000 --> 01:20:27.000 I'm real curious how quickly is quickly with the method you were using. 01:20:27.000 --> 01:20:32.000 Oh, yeah. So this is an effort. That's being put in by 01:20:32.000 --> 01:20:37.000 JPL to make what they call consensus interferon products. 01:20:37.000 --> 01:20:44.000 They're basically kind of processed in a standard way to a reasonable resolution. 01:20:44.000 --> 01:20:51.000 each interferon is a file is about 60mb, which is was way smaller than the actual data which is a gigabyte. 01:20:51.000 --> 01:20:54.000 So it's very efficient in terms of like data volume. 01:20:54.000 --> 01:21:00.000 And yeah, you can go to the Alaska Satellite Facility and download hundreds of these, 01:21:00.000 --> 01:21:11.000 now pretty easily. And then we feed it into a time series code which basically has been adapted to read it as input and you can get an answer within an afternoon. 01:21:11.000 --> 01:21:15.000 Oh, please. Okay. 01:21:15.000 --> 01:21:23.000 So yeah, it's great. We're also putting together part of this, the project that produced those that I'm working with Simpson at JPL 01:21:23.000 --> 01:21:29.000 and David backup is to is to make a kind of consensus product which we then share with everybody. 01:21:29.000 --> 01:21:33.000 So that you don't even have to do the time series 01:21:33.000 --> 01:21:38.000 analysis that's the goal. So we hope to have something at about a slightly reduced resolution, 01:21:38.000 --> 01:21:44.000 about 200 meters that we can share with everybody within months. 01:21:44.000 --> 01:21:48.000 I mean, this is something we've been working on at SCEC, Southern California, where 01:21:48.000 --> 01:22:00.000 the data basically exist for Northern California as well. We're writing a paper to document what we did and how we did it, that they should be shared as part of that. 01:22:00.000 --> 01:22:03.000 Thank you. 01:22:03.000 --> 01:22:13.000 Eric Fielding has his hand up. Eric, you wanna ask your question? 01:22:13.000 --> 01:22:16.000 Can't hear you 01:22:16.000 --> 01:22:32.000 Still, quiet. 01:22:32.000 --> 01:22:49.000 Eric, we can't hear you. 01:22:49.000 --> 01:22:51.000 Yeah. 01:22:51.000 --> 01:22:52.000 How's this? 01:22:52.000 --> 01:22:53.000 Yes. 01:22:53.000 --> 01:22:54.000 There we go! 01:22:54.000 --> 01:23:03.000 Okay. Sorry. I, I have a a new display here. The studio display had to switch to the different microphone. Okay. 01:23:03.000 --> 01:23:10.000 Sorry. That we have UAVSAR, our coverage of the Central San Francisco area 01:23:10.000 --> 01:23:17.000 with excellent coverage of the Hayward fault, Concord fault, 01:23:17.000 --> 01:23:36.000 Green Valley fault, Greenville fault, and Calaveras fault and of course, the San Andreas, so it shows the creep on the Concord fault very clearly. 01:23:36.000 --> 01:23:47.000 You know, we're we're working with Roland and Ben Brooks on the analysis, our main focus is the Hayward fault 01:23:47.000 --> 01:23:52.000 since that's what we got our NASA grant for. 01:23:52.000 --> 01:23:55.000 This is Chris. Is there a viewer where just a non subject matter 01:23:55.000 --> 01:24:07.000 expert consultant could go and look at a rate for a section of the fault for the UAVSAR data set? 01:24:07.000 --> 01:24:19.000 That's not available yet. And we're just still trying to... sort of raw data is available online that you can download it. 01:24:19.000 --> 01:24:22.000 But it requires some amount of processing to you. It make the velocity map. 01:24:22.000 --> 01:24:28.000 I can send you a KMZ file if you wanna take a quick look. 01:24:28.000 --> 01:24:31.000 Okay. Thank you. 01:24:31.000 --> 01:24:39.000 Chad, I'm seeing a question by Ruth. Did we get to Ruth Harris's question on the 3-D velocity structure 01:24:39.000 --> 01:24:41.000 I don't think so, Ruth. Do you want to ask that question? 01:24:41.000 --> 01:24:46.000 Yeah. 01:24:46.000 --> 01:24:47.000 Yeah, I was just wondering for the earthquake 01:24:47.000 --> 01:24:58.000 locations. That was for the swarm I was just wondering how or if the 3D-velocity structure would make a difference? 01:24:58.000 --> 01:25:17.000 So, if you have really complicated 3D-velocity structure with your earthquakes to be in, because they move around a little bit and maybe give different focal mechanisms or different earthquake locations, and then give a different story about the earthquakes swarm behavior 01:25:17.000 --> 01:25:21.000 So for the first question, for the location of the swarms. 01:25:21.000 --> 01:25:29.000 So the velocity model or structure model might change the absolute locations for the swarm. 01:25:29.000 --> 01:25:33.000 But it will now change the relative locations. So the relative locations, I mean the internal fault structures might be good 03:42:16.000 --> 03:42:44.00 even though there is uncertainties for the velocities models. For the focal mechanism, we used here have considerable uncertainties of the velocity. So I think the rotations of the focal mechanism writing to the hypocenter distribution so I 01:26:01.000 --> 01:26:06.000 think that some of them may be significant. 01:26:06.000 --> 01:26:07.000 Okay, thank you. Thanks. And I really like that work. 01:26:07.000 --> 01:26:22.000 Thank you. Thanks. 01:26:22.000 --> 01:26:31.000 Let's see another question by by Vicki. Vicki, would you like to ask your question? 01:26:31.000 --> 01:26:37.000 Yeah, I just wanted to have some clarification about the trend of those earthquake 01:26:37.000 --> 01:26:45.000 swarms were pretty much orthogonal to the dominant structural trend of the faults, 01:26:45.000 --> 01:26:50.000 and what might that actually mean? Is that the width of the fault? 01:26:50.000 --> 01:26:59.000 Basically, it's kind of unusual. 01:26:59.000 --> 01:27:06.000 Oh, yes. So the the swarm occurred either [indiscernible], transform place from 01:27:06.000 --> 01:27:17.000 I think that's the transfer fault from the Calaveras fault to 01:27:17.000 --> 01:27:24.000 the Mount Diablo Thrust fault so that's not the major fault of the region. 01:27:24.000 --> 01:27:35.000 So the striking direction of those swarm is different from the striking direction on [indiscernible] fault 01:27:35.000 --> 01:27:38.000 Okay, thank you. I'm kind of puzzled 01:27:38.000 --> 01:27:39.000 how does I'm assuming a near vertical strike-slip fault, step over to a thrust 01:27:39.000 --> 01:27:53.000 fault. Do these earthquakes tell us something about how that actually takes place? 01:27:53.000 --> 01:28:01.000 Yeah. What do you mean by... Sorry. Could could you say that questions again? 01:28:01.000 --> 01:28:04.000 Well, we usually when I think, of a step over in a strike-slip system, 01:28:04.000 --> 01:28:09.000 I think of a strike-slip fault transferring slip to another strike-slip fault. 01:28:09.000 --> 01:28:10.000 Yes. 01:28:10.000 --> 01:28:20.000 But the Diablo fault is a thrust, so I was just wondering how these faults tie into that transfer of slip. 01:28:20.000 --> 01:28:28.000 Oh, I think that the Mount Diablo thrust fault, 01:28:28.000 --> 01:28:49.000 I think that's one. I think the transform slip that's between the Calaveras fault to the [indiscernible] fault and the Mount Diablo fault is between those two. So that's now the major transform fault. 01:28:49.000 --> 01:28:53.000 So that's just, I mean... 01:28:53.000 --> 01:28:59.000 It's just an unusual geometry. Because it's going to a thrust fault in yet 01:28:59.000 --> 01:29:06.000 your faults were very steep, and not oriented the same way as the Diablo fault. 01:29:06.000 --> 01:29:09.000 No, I was just curious. It just seemed kind of interesting. Thank you. 01:29:09.000 --> 01:29:16.000 Okay, that's a good question. I will think of that later, thanks. 01:29:16.000 --> 01:29:30.000 So we have a question from Christine. Christine, you had a question for Artie. 01:29:30.000 --> 01:29:37.000 Sorry. I switched my... 01:29:37.000 --> 01:29:38.000 Yeah. 01:29:38.000 --> 01:29:39.000 Sorry about that, hey? Can you hear me? Okay, I've switched my, my listening and speaking devices. 01:29:39.000 --> 01:29:48.000 Yeah. Artie, I was just curious, you use the GP method, Grayson, Pitarka, and you just showed that it seemed to me that there was a different space between, 01:29:48.000 --> 01:30:02.000 well, the graduation, the high slip patch in the middle was not filling the fault plane, as I expected to be in the model. 01:30:02.000 --> 01:30:06.000 I was wondering if this something calibrated. No. Okay. 01:30:06.000 --> 01:30:17.000 So this is using the the hybrid method that you know incorporates some of these [indiscernible] concepts, you know, and written some papers on this. 01:30:17.000 --> 01:30:30.000 The idea was, really, you know, you throw the dice and you get this random slip distribution. 01:30:30.000 --> 01:30:31.000 Yeah. 01:30:31.000 --> 01:30:34.000 And I wanted to concentrate that, you know, on to this fault surface of these smaller faults, and that was kind of the purpose. 01:30:34.000 --> 01:30:35.000 Yeah. 01:30:35.000 --> 01:30:41.000 So this is not something calibrated. But you did, you know I'd have looked at, you know the spectral response inside and outside the basin. 01:30:41.000 --> 01:30:48.000 You know. They're not out of line, you know, but I wouldn't say calibrated. 01:30:48.000 --> 01:30:55.000 So you're mixing the [indiscernible] asperities with the slip generation. 01:30:55.000 --> 01:30:56.000 Okay. Yeah. 01:30:56.000 --> 01:31:03.000 Well, there's Arvin's [name?] got a couple Arban [name?] and Rob. I have a couple of papers on this, and you know the concept of the slip patches comes from the stochastic. 01:31:03.000 --> 01:31:19.000 You know, random, completely random slip, you know, generation, stochastic greens functions and you know, Erikora [name?] developed this method where you know if it's smaller than, say, 6 and a half. 01:31:19.000 --> 01:31:20.000 Yes. 01:31:20.000 --> 01:31:24.000 There's one patch, and if it's bigger to be two, and there's some scaling laws on those, but it's pretty, it's pretty 01:31:24.000 --> 01:31:39.000 ad hoc. Frankly, you know, and this was sort of just utilitarian to say, I want to have the slip on this section of the faults so that I can get some directivity, and I'm not gonna have some slip patch on the edge 01:31:39.000 --> 01:31:45.000 which is, you know, physically, undesirable, unrealistic. 01:31:45.000 --> 01:31:57.000 Thank you. This is Chris Madugo I wanted to jump back to the complicated Calaveras and other faults. 01:31:57.000 --> 01:32:15.000 We convened a workshop and Jeff Unru and David Schwartz, participate in and a bunch of consultants, because our question was, what happens to slip on the Calaveras fault, and one common held view is that it just jumps over to the 01:32:15.000 --> 01:32:23.000 Concord fault. But when you look at the topography there, that doesn't make sense or right lateral fault, jumps into a right lateral fault. 01:32:23.000 --> 01:32:33.000 You'd expect a big basin, but there's all this inverted topography in between, and so I think the answer is complicated. 01:32:33.000 --> 01:32:51.000 Some of it jumps onto the Contra Costa shear zone and makes its way up to the west Napa fault system and then Jeff Unru was suggesting you could have right-stepping if it was kind of an echelon stepping not your typical basin forming right- 01:32:51.000 --> 01:33:07.000 stepping between the Calaveras and Concord fault, but that maybe you aren't getting all the slip, transferring over there, and then I'm not quite sure how the other faults that the earthquake swarms have occurred on fit into there and then there's 01:33:07.000 --> 01:33:13.000 also the Pleasant fault coming up along the west side of Mount Diablo, which is also kind of a strike-slip fault. 01:33:13.000 --> 01:33:16.000 So it's a really complicated area. 01:33:16.000 --> 01:33:24.000 Also the Calaveras fault stops well short of the Concord fault, so it seems like more work could be done there. 01:33:24.000 --> 01:33:33.000 It's a very complicated step over. 01:33:33.000 --> 01:33:39.000 On that subject, Chris. I mean, how does the Greenville fault 01:33:39.000 --> 01:33:46.000 connect with Mount Diablo in the area you were just talking about in this Contra Costa shear zone? 01:33:46.000 --> 01:33:47.000 I don't know, I mean any insight on that. 01:33:47.000 --> 01:33:57.000 That one's easier, in my mind, and that's more like your typical left-step on a strike-slip system with Mount Diablo rising between. 01:33:57.000 --> 01:34:04.000 So if you have left-step on right-lateral, you'd expect inverted topography. 01:34:04.000 --> 01:34:08.000 And so that one makes a little bit more sense to me. 01:34:08.000 --> 01:34:17.000 Tom Sawyer and Jeff Unru have been working on that, looking at uplifted terraces in Mount Diablo, and I think that's their model here. 01:34:17.000 --> 01:34:18.000 Cool. 01:34:18.000 --> 01:34:25.000 You've got a left-step in a right-lateral or dextral strike-slip system, creating that uplift. 01:34:25.000 --> 01:34:31.000 Alex, I see you had a question, too. Did that get answered by Chris? 01:34:31.000 --> 01:34:38.000 A moment ago on the Concord in Green Valley fault. Well. 01:34:38.000 --> 01:34:45.000 Thanks. Yeah. Chris mentioned that there was like a bend and a maybe a small step in the Concord to Green Valley faults, 01:34:45.000 --> 01:34:52.000 but they seemed fairly continuous, and it's a bit of splitting hairs, but I'm just kind of curious 01:34:52.000 --> 01:34:59.000 if there was more information about their 3D geometries, Chris showed a little bit about the boreholes on the Concord. 01:34:59.000 --> 01:35:04.000 But I'm just curious if you could discuss the differentiation between those two faults 01:35:04.000 --> 01:35:07.000 I think it's just a step under the Bay, but I think they're basically the same fault. 01:35:07.000 --> 01:35:17.000 So if you look at the segmentation, the southern Green Valley and the Concord are basically the same fault. 01:35:17.000 --> 01:35:22.000 The big change in my mind that happens it's the northern Concord fault, 01:35:22.000 --> 01:35:36.000 it's really well defined like you can see it in the geomorphology, historical, aerial photos put your fingers on it, and then the southern half of the Concord fault becomes really diffuse multiple strands in the southern Ignacio Valley it's a lot harder to 01:35:36.000 --> 01:35:39.000 find in in the geomorphology. So there's no step there where that change occurs, but it just becomes a lot more diffuse 01:35:39.000 --> 01:35:56.000 about half way down the mapped length of the Concord fault. But, in short, I think the southern Green Valley and the Concord are basically the same fault with a minor structural complexity between them 01:35:56.000 --> 01:36:02.000 Thank you. 01:36:02.000 --> 01:36:04.000 John, are we okay on time or 01:36:04.000 --> 01:36:06.000 Yeah, I think we're well, we're 7 min past the official end of our block. 01:36:06.000 --> 01:36:17.000 So maybe we should wrap things up unless anybody has any last question. 01:36:17.000 --> 01:36:22.000 Everybody. Thank you to all of our speakers, and thank you, all of you who contributed questions for this li this lively discussion. 01:36:22.000 --> 01:36:33.000 It's fun to think about new and different, and maybe more complex than we actually imagine faults in the East Bay. 01:36:33.000 --> 01:36:41.000 So we will see all of you in 22 min for the next session.