WEBVTT Kind: captions Language: en-US 00:00:02.160 --> 00:00:05.040 All right. Welcome, everybody, to the ESC seminar this week. 00:00:05.040 --> 00:00:08.320 Our speaker will be Kate Scharer, who will be talking about 00:00:08.320 --> 00:00:11.976 rupture models for the San Andreas and San Jacinto faults. 00:00:12.000 --> 00:00:15.176 We don’t have any announcements this week. 00:00:15.200 --> 00:00:20.480 But we do want to just let you know as a reminder to mute yourselves – 00:00:20.480 --> 00:00:22.960 mute your audio and turn off your video to save bandwidth. 00:00:22.960 --> 00:00:25.785 So we hopefully will have a stronger connection. 00:00:26.560 --> 00:00:29.200 Also, we do have live captions available 00:00:29.200 --> 00:00:32.536 under the More Actions icon in Teams. 00:00:32.560 --> 00:00:38.960 And then we look forward to hearing from Kate, and now I’ll turn over 00:00:38.960 --> 00:00:42.614 the time to Elizabeth Cochran to introduce her for us. 00:00:42.639 --> 00:00:46.960 - All right. Thanks, Tamara. So it is my great pleasure to introduce 00:00:46.960 --> 00:00:52.456 Kate Scharer from the Pasadena office, where we both are. 00:00:52.480 --> 00:00:58.240 So I actually met Kate at a UJNR meeting in Nagaoka, Japan. 00:00:58.240 --> 00:01:05.176 It was held October 20th to 22nd, 2010, so exactly a decade ago. 00:01:05.200 --> 00:01:09.120 We had some fun times wandering around Japan trying to find our way 00:01:09.120 --> 00:01:14.560 to the meeting and back. So, at that time, neither of us were 00:01:14.560 --> 00:01:19.417 working at the USGS, but both of us joined the USGS the following year. 00:01:20.080 --> 00:01:25.096 So whoever made the invitation list did a good job there. [laughs] 00:01:25.120 --> 00:01:28.880 Prior to coming to the USGS, Kate was a professor at 00:01:28.880 --> 00:01:32.536 Appalachian State University in North Carolina. 00:01:32.560 --> 00:01:36.856 She got her Ph.D. from the University of Oregon and 00:01:36.880 --> 00:01:40.776 a B.S. in geological sciences from the University of Washington. 00:01:40.800 --> 00:01:44.640 Of course, we all know Kate for her impressive and impactful 00:01:44.640 --> 00:01:49.040 contributions to our understanding of the timing of earthquakes 00:01:49.040 --> 00:01:53.896 along the San Andreas and other faults worldwide. 00:01:53.920 --> 00:01:58.800 She’s also made a lot of contributions to looking at deformation produced by 00:01:58.800 --> 00:02:05.736 tectonic motion using field mapping, Lidar analysis, and other methods. 00:02:05.760 --> 00:02:11.360 Most recently, she served as the NEHRP investigations coordinator 00:02:11.360 --> 00:02:16.536 for the 2019 Ridgecrest earthquake sequence. 00:02:16.560 --> 00:02:21.176 And she was also, this year, named a GSA fellow. 00:02:21.200 --> 00:02:28.240 And I’m going to read the [laughs] – so, for her comprehensive research on 00:02:28.240 --> 00:02:33.680 San Andreas Fault paleoseismology, application of novel Quaternary 00:02:33.680 --> 00:02:37.040 geochronology methods to estimate earthquake timing, 00:02:37.040 --> 00:02:40.536 and commitment to earthquake hazards outreach and education 00:02:40.560 --> 00:02:44.456 that will have lasting impacts on the field of paleoseismology. 00:02:44.480 --> 00:02:49.336 And her leadership distinguishes her as one of the best in our profession. 00:02:49.360 --> 00:02:52.560 So I think that’s an accurate characterization of Kate, 00:02:52.560 --> 00:02:57.816 and so I will now turn it over to her for her talk with the title, 00:02:57.840 --> 00:03:01.040 a maximum rupture model for the southern San Andreas and 00:03:01.040 --> 00:03:04.400 San Jacinto faults, California, derived from paleoseismic 00:03:04.400 --> 00:03:10.256 earthquake ages – what limits does it provide for MFDs? 00:03:11.207 --> 00:03:14.880 - Okay, well, I’m sufficiently embarrassed now [laughs] – 00:03:14.880 --> 00:03:16.878 or humbled. 00:03:18.800 --> 00:03:22.501 And we’ve already run into our – there we go. 00:03:22.501 --> 00:03:24.800 The first exciting moment here. So, thanks. 00:03:24.800 --> 00:03:28.296 That was a really nice introduction. [laughs] 00:03:28.320 --> 00:03:35.760 I am going to talk about many of the field studies that Elizabeth just 00:03:35.760 --> 00:03:39.976 mentioned along the San Andreas, but it’s going to be a much broader view. 00:03:40.000 --> 00:03:44.320 What I’m going to go over is a study that Doug Yule and I published 00:03:44.320 --> 00:03:48.240 recently examining the record of paleoseismic events along 00:03:48.240 --> 00:03:52.225 the southern San Andreas. Can you see my cursor? 00:03:53.480 --> 00:03:56.055 No? - No. I cannot see it. 00:03:56.696 --> 00:03:58.400 - That’s the worst. 00:03:58.400 --> 00:04:02.264 - Might try sometimes switching over to laser pointer mode works better. 00:04:02.264 --> 00:04:08.131 - Okay. How about that? Red dot? 00:04:08.897 --> 00:04:11.121 - Yep. We can see that. - Great. 00:04:12.160 --> 00:04:15.200 So – that Doug Yule and I recently published 00:04:15.200 --> 00:04:17.760 in Geophysical Research Letters. 00:04:17.760 --> 00:04:21.760 So many of you – I spend a lot of my time working on problems 00:04:21.760 --> 00:04:26.640 related to earthquake rupture forecasts. And many of us at the Earthquake 00:04:26.640 --> 00:04:30.640 Science Center investigate different aspects of this problem. 00:04:30.640 --> 00:04:33.920 The magnitude-frequency distribution of a set of faults. 00:04:33.920 --> 00:04:36.880 We are asking, are there any spatial patterns going on 00:04:36.880 --> 00:04:39.760 in the earthquakes that occur? Segmentation, sequencing, 00:04:39.760 --> 00:04:43.520 or regional activity might be underlying some of the patterns. 00:04:43.520 --> 00:04:47.280 And then, are there any temporal patterns in the timing 00:04:47.280 --> 00:04:50.456 of these big events, in particular? 00:04:50.480 --> 00:04:54.640 And the same types of questions are what are the sort of fundamental 00:04:54.640 --> 00:04:57.200 reasons we produce paleoearthquake rupture histories. 00:04:57.200 --> 00:05:00.880 In those, we’re using paleoearthquake data to determine the timing and 00:05:00.880 --> 00:05:03.680 extent of past ruptures. So you can imagine, if we know 00:05:03.680 --> 00:05:06.560 how old they are and where they occurred, we can address 00:05:06.560 --> 00:05:11.256 some of these questions with empirical data directly. 00:05:11.280 --> 00:05:14.400 So, as I said, this was recently published in Geophysical Research 00:05:14.400 --> 00:05:16.960 Letters, and I just want to make sure that everyone’s aware of 00:05:16.960 --> 00:05:20.376 Doug Yule’s contributions here, which are substantial. 00:05:20.400 --> 00:05:24.160 The talk will be in three parts. I’ll talk about how I constructed 00:05:24.160 --> 00:05:29.280 the maximum rupture model itself and sort of the data that go into it. 00:05:29.280 --> 00:05:32.320 We’ll focus on new and exciting data that was really – you know, 00:05:32.320 --> 00:05:36.800 we were able to crack the nut of how this part of the world works. 00:05:36.800 --> 00:05:39.680 And so we’re looking forward to sharing that with you. 00:05:39.680 --> 00:05:42.480 And then we’ll evaluate the maximum rupture model 00:05:42.480 --> 00:05:47.416 in comparison to other data sets and constraints. 00:05:47.440 --> 00:05:52.000 And, as I go from each part, I’ll pause for questions, but people are welcome 00:05:52.000 --> 00:05:55.680 to interject at any time if they want clarification on anything. 00:05:55.680 --> 00:05:58.216 I’m happy to have that. 00:05:58.240 --> 00:06:00.960 So, throughout the talk, what we’ll be looking at is the 00:06:00.960 --> 00:06:04.800 southern San Andreas Fault here and the San Jacinto Fault here on its side. 00:06:04.800 --> 00:06:09.280 So north is pointed up and to the left. You’re going to see this plot quite a bit, 00:06:09.280 --> 00:06:12.640 or parts of this plot. In all of the imagery, the San Andreas 00:06:12.640 --> 00:06:17.416 Fault are in these blue tones, and the San Jacinto Fault will be in green. 00:06:17.440 --> 00:06:20.320 You can see each one of the major sections listed here – 00:06:20.320 --> 00:06:23.040 Carrizo, Big Bend, Mojave, San Bernardino, etc., 00:06:23.040 --> 00:06:26.856 and the strand names for the San Jacinto Fault below. 00:06:26.880 --> 00:06:29.280 And then we’ll be looking at these paleoseismic sites, 00:06:29.280 --> 00:06:32.800 each of which has a two-letter character that identifies its name. 00:06:32.800 --> 00:06:39.496 For example, this is the Frazier Mountain site that I published in 2017. 00:06:39.520 --> 00:06:42.800 So the data set we’ll be examining to produce this maximum rupture 00:06:42.800 --> 00:06:47.040 model includes data from 31 sites. And the data are only 00:06:47.040 --> 00:06:50.080 site paleoearthquake data. So, for example, here’s a little 00:06:50.080 --> 00:06:53.600 picture from Pallett Creek. You can envision, we have trench 00:06:53.600 --> 00:06:57.520 information that shows rupture extending up to some paleo 00:06:57.520 --> 00:07:00.640 ground surface. We date a series of layers 00:07:00.640 --> 00:07:04.136 through that. Here might be the radiocarbon ages for those layers. 00:07:04.160 --> 00:07:07.680 And we determine the age of the paleoearthquake through Bayesian 00:07:07.680 --> 00:07:12.800 methods, but it’s simply defined, or expressed, as related to the 00:07:12.800 --> 00:07:17.600 bounding ages of the sediments above and below that earthquake horizon. 00:07:17.600 --> 00:07:20.960 And usually, we’re using radiocarbon for all of these. 00:07:20.960 --> 00:07:23.600 For the rest of the talk, rather than worrying about the probability 00:07:23.600 --> 00:07:26.480 distribution functions we get, I’m only going to pay attention 00:07:26.480 --> 00:07:30.320 to the 95% ranges for those earthquake ages. 00:07:30.320 --> 00:07:35.141 Because there’s sufficient information in that to do what we need to do. 00:07:36.000 --> 00:07:40.160 So, again, here’s the plot. And I’m showing you now all of the ages. 00:07:40.160 --> 00:07:45.840 In any column, you’re looking at fault distance from the end of the creeping 00:07:45.840 --> 00:07:48.480 section – basically Highway 64 00:07:48.480 --> 00:07:53.096 down to the southeast. And you’ll see these vertical bars 00:07:53.120 --> 00:07:58.000 sort of arranged below each site. Each vertical bar is that 95% range 00:07:58.000 --> 00:08:00.960 of the paleoearthquake date. And then sometimes you’ll see 00:08:00.960 --> 00:08:04.240 a thinner bar where you’re with a little number down below. 00:08:04.240 --> 00:08:07.280 And that’s the number of paleoearthquakes in some interval 00:08:07.280 --> 00:08:09.840 where, at the time of this publication, they weren’t well-dated. 00:08:09.840 --> 00:08:13.336 We just knew there were maybe four to five earthquakes at the site. 00:08:13.360 --> 00:08:18.240 The sites with bold print here are the ones that have the 00:08:18.240 --> 00:08:22.136 best recorded and best dated information. 00:08:22.160 --> 00:08:27.680 So the data set is extensive. Here’s a list of all of the papers 00:08:27.680 --> 00:08:31.896 and masters theses that I culled to get this information. 00:08:31.920 --> 00:08:34.880 And what we’ll do next is show you how this maximum 00:08:34.880 --> 00:08:38.640 rupture model would be constructed. It’s really quite simple. 00:08:38.640 --> 00:08:43.736 Again, you have a series of sites and time going down on the Y axis. 00:08:43.760 --> 00:08:48.856 Where we see ages that overlap in time, we say that that’s going to be a rupture. 00:08:48.880 --> 00:08:54.856 The rupture ends between the last observed earthquake and the next site, 00:08:54.880 --> 00:08:58.560 where there is no evidence of rupture. So, for example, this bar is halfway 00:08:58.560 --> 00:09:02.480 between C and D. Down here, you can see a set of overlapping 00:09:02.480 --> 00:09:07.520 ages for an older rupture. And then A has no equivalent 00:09:07.520 --> 00:09:10.240 earthquake in that time period, so this rupture terminates halfway 00:09:10.240 --> 00:09:13.600 between sites A and B. It’s just a – most simple rule 00:09:13.600 --> 00:09:15.920 to follow in constructing this. And you can envision 00:09:15.920 --> 00:09:19.760 more complicated ones, but this allows you to stay 00:09:19.760 --> 00:09:22.936 in this upper end member of a maximum rupture model. 00:09:22.960 --> 00:09:27.280 It’s important to note that the height of the horizontal bar 00:09:27.280 --> 00:09:30.560 is not a signal of the strength of the quality of it. 00:09:30.560 --> 00:09:34.056 It’s just the age uncertainties of the contributing earthquakes. 00:09:34.080 --> 00:09:37.360 So, for example, this is the maximum permissible age 00:09:37.360 --> 00:09:40.400 that’s shared by all these sites. And this is the minimum 00:09:40.400 --> 00:09:44.000 permissible age. So the width of this band is fairly thick. This one just 00:09:44.000 --> 00:09:48.376 happens to be better constrained, in particular by the dating at Site A here. 00:09:48.400 --> 00:09:51.600 So that’s how it’s made. Here’s the data again. 00:09:51.600 --> 00:09:54.640 If you follow those rules, you end up with a kind of 00:09:54.640 --> 00:09:57.520 complicated-looking diagram like this. Again, the San Andreas Fault on the 00:09:57.520 --> 00:10:00.376 left and the San Jacinto Fault over here on the right. 00:10:00.400 --> 00:10:04.000 In the next image, what I’m going to do is just take away the vertical bar 00:10:04.000 --> 00:10:06.696 so you’re looking just at the ruptures themselves. 00:10:06.720 --> 00:10:10.160 You’ll see they’re numbered in red – with red values. 00:10:10.160 --> 00:10:15.736 There’s a total of 50 ruptures that I modeled in this end member model. 00:10:15.760 --> 00:10:18.480 Of course, the other end member is that you take each one of those 00:10:18.480 --> 00:10:24.240 157 earthquakes as its own rupture. If you do that, you get something 00:10:24.240 --> 00:10:29.760 like a recurrence interval for magnitude 6s of one every – 00:10:29.760 --> 00:10:34.240 mid-6s of one every decade. So that seems like an unlikely scenario 00:10:34.240 --> 00:10:37.760 given the historic information we have about the San Andreas Fault. 00:10:37.760 --> 00:10:41.255 And I want to stay up here on the San Jacinto Fault, up here in the 00:10:41.280 --> 00:10:44.080 sort of upper-left corner of the maximum rupture model because it’s 00:10:44.080 --> 00:10:49.920 defensible ground to examine just this really simple level of data 00:10:49.920 --> 00:10:53.840 that we’re putting into it. I also want to emphasize that all 00:10:53.840 --> 00:10:59.920 of the data, including the earthquake, each site age is available in 00:10:59.920 --> 00:11:02.000 supplementary information – you can just email me if you 00:11:02.000 --> 00:11:05.920 don’t want to find it at GRL – and also information about each rupture. 00:11:05.920 --> 00:11:12.000 So here is 1 through 13 of 50 with information about the extent, 00:11:12.000 --> 00:11:18.160 the contributing age limits, and then the magnitude given 00:11:18.160 --> 00:11:22.640 the length of the rupture as well as comments either about historic 00:11:22.640 --> 00:11:25.200 information related to it or other constraints that 00:11:25.200 --> 00:11:30.446 we have about that earthquake. So any questions so far? 00:11:32.240 --> 00:11:34.720 Nope. Okay, good. Let’s check it out. 00:11:34.720 --> 00:11:37.920 So what I want to show first is data that I’m really excited 00:11:37.920 --> 00:11:42.776 to present to everybody from the San Gorgonio Pass area. 00:11:42.800 --> 00:11:47.600 And then we’ll move on to Cajon Pass. So, for folks that don’t remember 00:11:47.600 --> 00:11:51.623 southern California, Los Angeles is over to the left. 00:11:52.320 --> 00:11:55.360 And, at this intersection between the southern-most San Andreas – 00:11:55.360 --> 00:11:58.456 this is the Salton Sea in this image – and the San Jacinto Fault, 00:11:58.480 --> 00:12:03.096 You get to interact with both Cajon Pass here and San Gorgonio Pass here. 00:12:03.120 --> 00:12:07.680 The San Gorgonio Pass was a focus of quite a bit of study 00:12:07.680 --> 00:12:13.440 in past SCEC iterations. And, in part, that was because of the 00:12:13.440 --> 00:12:16.560 attention that was given to the area as part of the ShakeOut scenario. 00:12:16.560 --> 00:12:20.560 So these plots, for example, show you the number of crossings 00:12:20.560 --> 00:12:23.680 of major infrastructure. Those are the little yellow dots 00:12:23.680 --> 00:12:27.200 of the San Andreas Fault that would be impacted by this ShakeOut scenario, 00:12:27.200 --> 00:12:31.280 which went all the way from the Salton Sea up to Palmdale, basically, 00:12:31.280 --> 00:12:35.120 across this image. So you can see, for infrastructure purposes, 00:12:35.120 --> 00:12:39.496 we obviously want to know more about what ruptures are like through this area. 00:12:39.520 --> 00:12:43.120 And it continued to be of interest because it’s obviously quite 00:12:43.120 --> 00:12:46.856 geometrically complex. So everything we know about 00:12:46.880 --> 00:12:52.080 ruptures going through geometric complexities are tested in this area. 00:12:52.080 --> 00:12:56.960 This is a slightly different view of it. If you are headed to SCEC, 00:12:56.960 --> 00:13:02.696 you’re driving through Cajon Pass itself over to Palm Springs here. 00:13:03.440 --> 00:13:06.800 And you can see that you go from basically sea level down the 00:13:06.800 --> 00:13:10.480 Salton Sea, you’re rising up to – this is San Gorgonio Peak. 00:13:10.480 --> 00:13:15.656 It’s the highest peak in southern California at 11,500 feet. 00:13:15.680 --> 00:13:19.040 And these are the sites that I’ll be talking about through 00:13:19.040 --> 00:13:22.960 the rest of this talk. We’ll be seeing this little cartoon image 00:13:22.960 --> 00:13:26.136 here so I can show you where they’re located. 00:13:26.160 --> 00:13:30.240 But before I want to – before I go into the details, I wanted to make sure 00:13:30.240 --> 00:13:33.600 to give special acknowledgement to several master’s students that have 00:13:33.600 --> 00:13:39.096 been really the backbone of the data sets that I’m going to show here. 00:13:39.120 --> 00:13:43.760 Paul McBurnett, Shahid Ramzan – both worked in the Millard 00:13:43.760 --> 00:13:47.040 and Cabazon areas. And then the last three I had the 00:13:47.040 --> 00:13:51.040 pleasure of serving on their master’s thesis committee and doing 00:13:51.040 --> 00:13:55.313 quite a bit of field work with them. Lisa Wolff – she is here. 00:13:56.320 --> 00:14:00.880 Jose Cardona – that’s him. And Bryan Castillo here smiling in 00:14:00.880 --> 00:14:04.000 front of the 18th Avenue Trench. And, of course, I also want to 00:14:04.000 --> 00:14:08.400 acknowledge their primary advisers – Doug Yule at Cal State-Northridge 00:14:08.400 --> 00:14:12.536 and Sally McGill at Cal State-San Bernardino. 00:14:12.560 --> 00:14:15.120 Trenching takes a considerable amount of work, and, as you’ll see 00:14:15.120 --> 00:14:18.960 in the next slides, the trenches that we excavated for these studies 00:14:18.960 --> 00:14:22.480 were really large, and so they take even more work. [chuckles] 00:14:22.480 --> 00:14:26.560 So it couldn’t have been done without a cadre of folks from – students from 00:14:26.560 --> 00:14:30.936 Cal State-Northridge. In particular, Brittany Huerta was really helpful 00:14:30.960 --> 00:14:33.120 in doing quite a bit of logging and field work. 00:14:33.120 --> 00:14:36.160 And she was great at finding radiocarbon samples. So just want to 00:14:36.160 --> 00:14:39.896 make sure that all of these people are properly acknowledged. 00:14:39.920 --> 00:14:41.440 Oh, I should say – this is Doug here. 00:14:41.440 --> 00:14:45.576 He’s running in to make it into the picture. 00:14:45.600 --> 00:14:49.600 So the first sites that I want to show are the Cabazon and Millard sites 00:14:49.600 --> 00:14:52.320 here on the San Gorgonio Pass thrust fault system. 00:14:52.320 --> 00:14:58.776 It’s the most oblique-to-plate motion of the San Andreas as a whole. 00:14:58.800 --> 00:15:02.400 There were a series of trenches excavated both at Millard Canyon 00:15:02.400 --> 00:15:06.560 and then here at the Cabazon site. And these are available in these 00:15:06.560 --> 00:15:13.120 master’s theses online at CSUN. I got involved in about 2013 when 00:15:13.120 --> 00:15:19.176 we excavated these larger trenches over here at the Cabazon site. 00:15:19.200 --> 00:15:22.240 So this trench was ginormous. We called it the megatrench. 00:15:22.240 --> 00:15:27.016 It’s 9 meters deep, 45 meters long, and about 30 meters wide. 00:15:27.040 --> 00:15:31.360 This picture is taken from about here looking towards the south. 00:15:31.360 --> 00:15:34.800 These are the San Jacinto Mountains. And you can see people down here 00:15:34.800 --> 00:15:37.976 at the bottom and some vehicles for a sense of scale. 00:15:38.000 --> 00:15:42.616 This picture is taken at the south end about where this truck is 00:15:42.640 --> 00:15:47.176 and looking at the fault zone. So this is my dog [Tocket] for scale. 00:15:47.200 --> 00:15:52.960 And you can see two primary fault zones coming up-trench right at you. 00:15:52.960 --> 00:15:56.240 And the nice interplay between different sediment sources 00:15:56.240 --> 00:15:59.920 in these trench walls. So, to show you what some of 00:15:59.920 --> 00:16:05.416 the evidence is like for this system, it’s an oblique thrust fault system. 00:16:05.440 --> 00:16:09.816 We have about five earthquakes that we could observe in the last 6,000 00:16:09.840 --> 00:16:13.980 years, and they all had fairly modest displacement, which is interesting. 00:16:15.040 --> 00:16:18.880 The record, we think, is best and most complete in the last 2,000 years. 00:16:18.880 --> 00:16:24.000 So this includes the most recent event, which you can see in the cartoon 00:16:24.000 --> 00:16:28.960 below, the faulting extending up, and here’s the photomosaic showing 00:16:28.960 --> 00:16:32.056 nice offset of this gravel unit here. 00:16:32.080 --> 00:16:35.120 And then the sort of decay of that offset. 00:16:35.120 --> 00:16:39.360 This was probably the ground surface at the time and diffused out and 00:16:39.360 --> 00:16:42.960 smoothed out the original rupture before you had subsequent deposits. 00:16:42.960 --> 00:16:47.416 So you can see a nice onlapping of material here. 00:16:47.440 --> 00:16:51.600 Cabazon 2 – you can see faulting going through bedrock here, 00:16:51.600 --> 00:16:55.520 going up into the photomosaic here. This has also fairly modest 00:16:55.520 --> 00:17:00.160 displacement. You can see this tan unit has an apparent offset of sort of 00:17:00.160 --> 00:17:06.080 30 to 40 centimeters in this section. And maybe it’s best imaged here with 00:17:06.080 --> 00:17:12.240 a white polygon showing this thickened deposit on top of that thrust motion. 00:17:12.240 --> 00:17:16.320 So these both occurred between about 300 and 1320 A.D. 00:17:16.320 --> 00:17:19.336 I’ll show you the dating towards the end. 00:17:19.360 --> 00:17:24.480 The 18th Avenue site was also recent work. 00:17:24.480 --> 00:17:27.920 It’s – Bryan Castillo from Cal State-San Bernardino, 00:17:27.920 --> 00:17:31.680 I’m really proud to say, has just submitted the final version 00:17:31.680 --> 00:17:35.336 to Geosphere, so it’s – will be published there. 00:17:35.360 --> 00:17:38.240 It’s an important site because it represents the only data trench 00:17:38.240 --> 00:17:41.680 on this 40-kilometer-long strand of the San Andreas. 00:17:41.680 --> 00:17:45.336 So it’s really our only window into rupture history. 00:17:45.360 --> 00:17:49.840 It was also really huge. It was a part of a AP fault zone study, 00:17:49.840 --> 00:17:53.760 so it’s a really, really long trench. We focused here on the north where 00:17:53.760 --> 00:17:58.400 we saw the only expression of faulting at this little sort of minor 00:17:58.400 --> 00:18:01.936 step-over – right step, or releasing step, in the fault system. 00:18:01.936 --> 00:18:05.280 To give you a sense of how very large this was, this was actually dug by 00:18:05.280 --> 00:18:08.560 a grader – one of those giant blades that they usually do, like, 00:18:08.560 --> 00:18:13.760 highway underpasses with. And then Sally McGill here knew 00:18:13.760 --> 00:18:16.720 the consultant, and he gave her a call because he realized what a gold mine 00:18:16.720 --> 00:18:20.160 this was for understanding earthquake behavior, and the landowner was 00:18:20.160 --> 00:18:24.000 amendable to us continuing work there. Unfortunately, one of the pieces of 00:18:24.000 --> 00:18:27.360 infrastructure I mentioned earlier – the fiber optic lines – runs through the area. 00:18:27.360 --> 00:18:32.480 This is – apparently, right along this tract is one of the major internet tubes 00:18:32.480 --> 00:18:36.240 [laughs] between southern California and the rest of the U.S. 00:18:36.240 --> 00:18:39.920 And so we weren’t able to extend the trench particularly farther north. 00:18:39.920 --> 00:18:44.960 We did do some hand-dug slot trenches to sort of improve our understanding of 00:18:44.960 --> 00:18:49.656 the expression to the north, but that was a minor default from this study. 00:18:49.680 --> 00:18:54.296 I’ll also mention other people that were important for this was Devin McPhillips 00:18:54.320 --> 00:18:58.560 and Seulgi Moon and Sourav Saha from UCLA 00:18:58.560 --> 00:19:01.816 that did the luminescence dating. 00:19:01.840 --> 00:19:06.080 So, at this site, again, it’s located about here in our diagram. 00:19:06.080 --> 00:19:10.480 We have five earthquakes in the last 3,000 years and two between about 00:19:10.480 --> 00:19:15.096 300 and, again, in 1260 A.D. Actually, that’s a typo from the other one. Sorry. 00:19:15.120 --> 00:19:17.920 I’ll show you the dates in a minute. So the most recent event – here’s 00:19:17.920 --> 00:19:22.320 Bryan sitting in front of it logging it. You can see faulting that deforms 00:19:22.320 --> 00:19:27.600 up through at least this orange layer. This blue unit is showing the 00:19:27.600 --> 00:19:31.840 warped ground surface. And then deformation is not expressed 00:19:31.840 --> 00:19:37.416 in this upper unit, so we would say that occurred about at this paleosurface. 00:19:37.440 --> 00:19:43.360 The second earthquake back is similar in that you can see a little bit of vertical 00:19:43.360 --> 00:19:48.240 separation across this yellow horizon. What’s important about this site is 00:19:48.240 --> 00:19:52.080 it’s clear there was always quite a strong component of lateral slip. 00:19:52.080 --> 00:19:56.560 We often have very strong mismatch of units across these faults that 00:19:56.560 --> 00:20:00.080 helped us understand the strength of these earthquakes. Unfortunately, 00:20:00.080 --> 00:20:05.196 the trench was so wide, we weren’t able to get 3D separations across them. 00:20:05.196 --> 00:20:10.376 And the last slide I want to show is the work of Jose Cardona in 2016. 00:20:10.400 --> 00:20:13.840 This was on the Garnet Hill strand. It’s actually a tiny, little connector 00:20:13.840 --> 00:20:17.120 between the Banning and the San Gorgonio Pass thrust fault system. 00:20:17.120 --> 00:20:21.360 It’s tiny but very important. As you can see here, it’s got a complex 00:20:21.360 --> 00:20:27.120 surface trace and warped some of the older alluvium units in the area. 00:20:27.120 --> 00:20:30.856 And then runs sort of sub-parallel to Interstate 10. 00:20:30.880 --> 00:20:34.080 If you’re driving from L.A. to Palm Springs, you know 00:20:34.080 --> 00:20:38.240 this pass particularly well. And this is about where the fault 00:20:38.240 --> 00:20:43.416 would [audio cuts out] the interstate at that location. 00:20:43.440 --> 00:20:46.880 I also like to point out for fun is that this sign right here is 00:20:46.880 --> 00:20:52.536 a welcome-to-California sign. And so, if this fault ever goes, 00:20:52.560 --> 00:20:55.600 I expect that this damaged or crumpled sign would be 00:20:55.600 --> 00:20:59.336 one of the most favorite memes that we’ll see on the internet. 00:20:59.360 --> 00:21:02.400 So, jokes aside, this site was pretty brutal. 00:21:02.400 --> 00:21:07.256 It was located right along I-10. It was loud and dusty and painful. 00:21:07.280 --> 00:21:10.560 And also, we have no evidence of faulting. 00:21:10.560 --> 00:21:14.480 Our oldest layer is 1400 A.D. This ends up providing 00:21:14.480 --> 00:21:16.880 a pretty interesting constraint compared to the other sites 00:21:16.880 --> 00:21:19.016 in the area, as we’ll see in a minute. 00:21:19.040 --> 00:21:22.865 But no evidence of earthquakes in the last 600 years. 00:21:23.680 --> 00:21:27.920 So, as I said, I wanted to just provide all the dates all together. 00:21:27.920 --> 00:21:32.480 It’s pretty hard to read these without having spent some time at it. 00:21:32.480 --> 00:21:35.360 So much easier is to just count how many earthquakes 00:21:35.360 --> 00:21:39.360 are expressed at each site. If we look just at San Gorgonio Pass 00:21:39.360 --> 00:21:44.000 and then examine the sites to either side, the Burro Flats site is very 00:21:44.000 --> 00:21:47.896 well-dated. It has eight earthquakes in the last 1,500 years. 00:21:47.920 --> 00:21:52.136 Millard and Cabazon have two. I just talked about those. 00:21:52.160 --> 00:21:57.440 The East Whitewater Wash site has zero in the last 600 years, 00:21:57.440 --> 00:22:00.480 so that’s why it’s in parentheses. 18th Avenue has two. 00:22:00.480 --> 00:22:03.600 And the Coachella site – Belle Philibosian published 00:22:03.600 --> 00:22:08.080 that work – has eight over here. So, right off the bat, if you’re asking 00:22:08.080 --> 00:22:11.760 about ruptures that are making it through San Gorgonio Pass, you know 00:22:11.760 --> 00:22:16.560 that only two of eight can cross through that system in the last 1,500 years. 00:22:16.560 --> 00:22:22.480 And this is a really new and exciting result that’s consistent across many 00:22:22.480 --> 00:22:26.160 trench sites, multiple depositional environments, 00:22:26.160 --> 00:22:28.536 and also different dating constraints. 00:22:28.560 --> 00:22:31.840 So these two earthquakes – one is relatively short. 00:22:31.840 --> 00:22:34.936 It would have occurred about 1210 A.D. 00:22:34.960 --> 00:22:39.016 At least 63 kilometers long. It would be about a magnitude 7. 00:22:39.040 --> 00:22:43.280 It may have extended farther to the northwest on some faults that aren’t 00:22:43.280 --> 00:22:46.800 part of our study, given the offsets that are observed in 00:22:46.800 --> 00:22:52.616 some of the trenches at Millard. The second one back is about 620 A.D. 00:22:52.640 --> 00:22:57.200 It’s almost 200 kilometers long, equivalent to about a magnitude 7.5. 00:22:57.200 --> 00:23:01.200 And it’s intriguing. It goes all the way from Coachella past Wrightwood. 00:23:01.200 --> 00:23:05.280 It’s not quite as large as a ShakeOut event, and of course, because this is 00:23:05.280 --> 00:23:07.760 a maximum rupture model, this could also be two separate 00:23:07.760 --> 00:23:11.760 earthquakes closely spaced in time. So what I would say is, from the 00:23:11.760 --> 00:23:19.255 trenches that we see, less than or equal to one permissible ShakeOut size event. 00:23:19.280 --> 00:23:21.920 Spending a little bit more time here, there’s also some interesting 00:23:21.920 --> 00:23:26.400 information about how the Coachella section and the Mission Creek and 00:23:26.400 --> 00:23:31.680 the Banning strands interact. If we look at the available earthquakes 00:23:31.680 --> 00:23:35.520 at the Coachella site, we see that three of those would terminate 00:23:35.520 --> 00:23:38.000 on the Mission Creek. These are these ones in gold here 00:23:38.000 --> 00:23:41.176 that are going into the gray, which is the Mission Creek strand. 00:23:41.200 --> 00:23:43.520 Two would terminate along the southern Banning. 00:23:43.520 --> 00:23:47.200 These I have in orange. Two would terminate closer to the 00:23:47.200 --> 00:23:50.640 splay between the two – basically at the southern end of the Indio Hills. 00:23:50.640 --> 00:23:52.640 And then the one I mentioned that would make it all the way 00:23:52.640 --> 00:23:55.176 through San Gorgonio Pass. 00:23:55.200 --> 00:24:00.240 What’s pretty intriguing here is – to me [laughs] – is that Douilly et al. – 00:24:00.240 --> 00:24:04.480 you guys heard him speak probably a month ago now about dynamic rupture 00:24:04.480 --> 00:24:09.440 modeling that his group had been working on in this area, and you can 00:24:09.440 --> 00:24:12.320 see, if your read this paper, that it turns out to be pretty hard for 00:24:12.320 --> 00:24:15.360 them to push ruptures onto the Banning. They had to really change 00:24:15.360 --> 00:24:19.120 the regional stress field. It’s permissible, but it was – 00:24:19.120 --> 00:24:21.440 required a strong rotation in the stress field in order to 00:24:21.440 --> 00:24:24.400 get ruptures onto the Banning. Given what we think we know 00:24:24.400 --> 00:24:28.320 about the stress field, it was more likely to go onto the Mission Creek. 00:24:28.320 --> 00:24:32.061 And that’s actually consistent with the earthquake ages we have here. 00:24:32.640 --> 00:24:36.800 So let’s move a little bit to the northwest and look around Cajon Pass, 00:24:36.800 --> 00:24:40.080 which is another geometric complexity that’s been of interest 00:24:40.080 --> 00:24:43.520 to the community lately. There is not too much known 00:24:43.520 --> 00:24:48.856 historically about this area, but one event is the 1899 Cajon Pass event. 00:24:48.880 --> 00:24:52.480 If you take a read through each one of these quotes, you’ll find that 00:24:52.480 --> 00:24:58.000 a common feature is landsliding. That’s consistent with the mapping 00:24:58.000 --> 00:25:01.256 that’s been done in the area. There’s no discrete evidence for 00:25:01.280 --> 00:25:06.080 a post-European arrival rupture anywhere in the area. 00:25:06.080 --> 00:25:11.280 It hasn’t been established exactly if this rupture did have – 00:25:11.280 --> 00:25:14.400 or, if this earthquake did have rupture – surface rupture or not. 00:25:14.400 --> 00:25:17.040 So therefore, it’s not included in my maximum rupture model. 00:25:17.040 --> 00:25:19.656 It’s just shown here by these dashed lines. 00:25:19.680 --> 00:25:24.080 I also like to note that, even in 1899, scientists would have been dealing 00:25:24.080 --> 00:25:29.440 with journalists who are asking if features halfway around the globe are 00:25:29.440 --> 00:25:32.400 relevant to activity we have locally. 00:25:32.400 --> 00:25:37.519 Here is says strangely coincident with a volcanic eruption in Hawaii, so … 00:25:39.610 --> 00:25:45.350 That’s all we know about Cajon Pass earthquakes from the historic record. 00:25:46.800 --> 00:25:51.200 Except for the 1812 earthquake, and even this one has different 00:25:51.200 --> 00:25:55.440 interpretations out there. So the December 8th, 1812, earthquake 00:25:55.440 --> 00:25:59.840 has long been argued to occur at Wrightwood based on a detailed study 00:25:59.840 --> 00:26:04.560 of tree ring growth in the Wrightwood area and also historic records that 00:26:04.560 --> 00:26:07.680 suggest it was on the San Andreas Fault on the San Bernardino 00:26:07.680 --> 00:26:11.120 and Mojave sections. And that’s also based – 00:26:11.120 --> 00:26:15.496 the dates based on damage to Spanish missions in the area. 00:26:15.520 --> 00:26:20.320 However, folks also have noticed that there is this earthquake here, shown by 00:26:20.320 --> 00:26:24.800 the big green swath on the San Jacinto Fault that has a compatible age 00:26:24.800 --> 00:26:30.880 with the 1812 rupture. And also, in consideration of the 00:26:30.880 --> 00:26:37.440 precariously balanced rocks and the distribution of damage to the missions, 00:26:37.440 --> 00:26:42.080 which are shown by these colored circles here, Julian Lozos, following 00:26:42.080 --> 00:26:46.640 on other investigations, did dynamic rupture modeling in the area and argued 00:26:46.640 --> 00:26:50.640 that, in fact, it was most likely that the 1812 event starts probably on the 00:26:50.640 --> 00:26:54.320 San Jacinto Fault and ruptures onto the San Andreas Fault. 00:26:54.320 --> 00:26:59.816 So this is pretty intriguing, giving you a rupture that would take up a much 00:26:59.840 --> 00:27:05.680 wider, longer bit of real estate to – and also involve both the San Andreas 00:27:05.680 --> 00:27:09.840 and the San Jacinto Fault. So I’m going to be a little bit agnostic about that. 00:27:09.840 --> 00:27:13.200 I think it’s certainly permissible given all the information we have. 00:27:13.200 --> 00:27:17.576 But I wanted to ask if that was observed at other times in the past. 00:27:17.600 --> 00:27:20.560 And I’ll call these things Cajon Pass sequences. 00:27:20.560 --> 00:27:24.160 So all they are is simply the temporal overlap of earthquakes on the 00:27:24.160 --> 00:27:29.040 San Bernardino and the Claremont sections. So here and here. 00:27:29.040 --> 00:27:33.736 And this will be dominated by the records at Burro Flats and Mystic Lake, 00:27:33.760 --> 00:27:38.000 since they provide the best constraints for that stretch of fault. 00:27:38.000 --> 00:27:41.520 So, if we go back in time, you can see that there was another event that’s on 00:27:41.520 --> 00:27:44.240 the San Bernardino strand, and then, in green, you should be able to make 00:27:44.240 --> 00:27:49.416 out a San Jacinto Fault. So this would be second Cajon Pass sequence. 00:27:49.440 --> 00:27:53.040 And now we’re extended back 1,500 years, and if you look at the 00:27:53.040 --> 00:27:58.960 record there, highlighted in those red ellipses are seven permissible Cajon 00:27:58.960 --> 00:28:04.080 Pass sequences in the last 1,500 years. Each one is a overlap that occurs 00:28:04.080 --> 00:28:10.056 in sort of a 75-or-so-year window. And two times, the Clark Fault is also – 00:28:10.080 --> 00:28:13.440 has a coincident earthquake. And so these could be very large 00:28:13.440 --> 00:28:17.896 ruptures that make it all the way down onto the San Andreas Fault. 00:28:17.920 --> 00:28:22.000 So the question I have about it and why I’m sort of agnostic on it and 00:28:22.000 --> 00:28:26.080 call these sequences rather than for-sure ruptures that do this is 00:28:26.080 --> 00:28:31.680 the implication due to the locations of the Burro Flat and Mystic Lake site is 00:28:31.680 --> 00:28:36.320 that you would have to have co-rupture on both the San Bernardino section 00:28:36.320 --> 00:28:40.640 and the Claremont for 50 kilometers. So this distance is about 50 kilometers 00:28:40.640 --> 00:28:46.216 past the splay point here where the two faults come closest to each other. 00:28:46.240 --> 00:28:53.040 This would be hard to buy, given that, if you look at historic compilations, 00:28:53.040 --> 00:28:58.400 for example, the one by Biasi and Wesnousky in 2016, after a splay point, 00:28:58.400 --> 00:29:02.760 they tend to see that a maximum of about 15 kilometers of one splay 00:29:02.760 --> 00:29:06.400 or the other is actually ruptured. So to envision that seven times 00:29:06.400 --> 00:29:10.800 in the last 1,500 years, we’ve had 50 kilometers on both 00:29:10.800 --> 00:29:12.480 is a little harder to accept. 00:29:12.480 --> 00:29:16.376 But maybe some of these are, in fact, such a beast. 00:29:16.400 --> 00:29:19.600 Another interesting feature that comes out of this – and, again, we’re dealing 00:29:19.600 --> 00:29:22.960 with a maximum rupture model, so we’re up in one quadrant of this – 00:29:22.960 --> 00:29:27.040 is perhaps episodes of these Cajon Pass sequences alternate 00:29:27.040 --> 00:29:30.640 with the San Gorgonio Pass events, which now I’m showing in pink here. 00:29:30.640 --> 00:29:33.440 And you can see that we seem to have sort of clustered activity, 00:29:33.440 --> 00:29:37.440 especially on the Claremont. And that is during times of 00:29:37.440 --> 00:29:41.920 quiescence on the San Gorgonio Pass thrust fault system. 00:29:41.920 --> 00:29:44.240 So perhaps something to investigate. 00:29:44.240 --> 00:29:48.209 So I’m going to pause there and ask if there’s any questions. 00:29:52.714 --> 00:29:58.102 Nope. Okay. So the last thing we’re going to do is evaluate the model. 00:29:59.120 --> 00:30:03.760 So here it is again. We can take a look at the historic record. 00:30:03.760 --> 00:30:08.720 If you accept that this 1812 event ruptured both the San Andreas and 00:30:08.720 --> 00:30:12.560 the San Jacinto Fault, there’s an 1800 event that’s fairly well-located 00:30:12.560 --> 00:30:15.600 and also observed in paleoseismic studies. 00:30:15.600 --> 00:30:18.776 And then the famous 1857 earthquake here. 00:30:18.800 --> 00:30:24.960 You have about 85% of the total domain of this model rupturing 00:30:24.960 --> 00:30:29.200 in 60 years, which is pretty stunning, given what we’ve had 00:30:29.200 --> 00:30:32.696 in my career, for example [chuckles], thankfully. 00:30:32.720 --> 00:30:37.120 So you could ask, well, do I see such a thing just visually or qualitatively 00:30:37.120 --> 00:30:40.000 in the record as you go back? And you can certainly pick out 00:30:40.000 --> 00:30:43.280 some time periods. This one here in the 1500s actually 00:30:43.280 --> 00:30:48.720 sees basically the entire system go off in about a 60-year span. 00:30:48.720 --> 00:30:53.336 And there’s other periods as well, where you have heightened activity. 00:30:53.360 --> 00:30:59.520 So comparing it to the historic record discretely, if you look at just from 00:30:59.520 --> 00:31:04.696 1800 to present, I’m showing those earthquakes in yellow bars. 00:31:04.720 --> 00:31:09.200 And then compare that with the magnitude predicted from the rupture 00:31:09.200 --> 00:31:15.360 model – that’s the gray bars – using the scaling out of Wesnousky 2008 because 00:31:15.360 --> 00:31:18.930 that’s only dependent on length, and I don’t have to worry about rupture area. 00:31:19.840 --> 00:31:24.480 What you can see is things are not totally dissimilar, which is intriguing, 00:31:24.480 --> 00:31:27.440 given this is a maximum rupture model. You can envision that the 00:31:27.440 --> 00:31:31.096 Venn diagram of these two wouldn’t overlap at all, but in fact, 00:31:31.120 --> 00:31:37.016 above about a magnitude 7.3, they’re fairly similarly stacked. 00:31:37.040 --> 00:31:39.600 What we lack, of course, in the maximum rupture model are 00:31:39.600 --> 00:31:45.840 these smaller moderate magnitude 6.5 to 7 events. 00:31:45.840 --> 00:31:48.560 I’d say that, you know, it would be really easy to produce some of those. 00:31:48.560 --> 00:31:53.680 All you have to do is cut one of these longer ruptures by about 15 kilometers, 00:31:53.680 --> 00:31:59.200 and you’ve produced a magnitude 6.5 earthquake there. 00:31:59.200 --> 00:32:01.520 You just would need some rules to do it. 00:32:01.520 --> 00:32:06.480 I didn’t have any handy, so I just have left it as is, noting that it 00:32:06.480 --> 00:32:12.000 would be easy to accommodate that and make a pretty similar distribution 00:32:12.000 --> 00:32:14.640 if you wanted to match the historic record, of course. 00:32:14.640 --> 00:32:17.680 It’s also a short record, so I’m not sure you want to do that, but it’s 00:32:17.680 --> 00:32:21.483 nonetheless the one we know about, so it’s important to test against. 00:32:22.560 --> 00:32:26.000 The challenge there, though, is that, as you go back in time, several of 00:32:26.000 --> 00:32:28.400 these sites have shorter records. 00:32:28.400 --> 00:32:33.040 Frazier, for example, kind of gets weak after about 1000 A.D. 00:32:33.040 --> 00:32:36.536 Bidart’s ends at about 1350. 00:32:36.560 --> 00:32:40.960 So, in order to get around that, I am going to plot cumulative 00:32:40.960 --> 00:32:44.320 moment release as a fraction of the available fault length. 00:32:44.320 --> 00:32:49.680 So we have about 500 kilometers of the San Andreas Fault plus sort of 00:32:49.680 --> 00:32:56.880 230 kilometers of the San Jacinto Fault available to us back to about 1350. 00:32:56.880 --> 00:33:01.416 And then that fraction goes down as you go back in time. 00:33:01.440 --> 00:33:04.720 And, if you do that, you get a plot that looks something like this. 00:33:04.720 --> 00:33:10.376 So cumulative moment release, scaled by modeled length, over time. 00:33:10.400 --> 00:33:14.000 The horizontal bar represents the uncertainty on the age of the rupture, 00:33:14.000 --> 00:33:17.520 and they’re color-coded – again, blue for San Andreas and green for San 00:33:17.520 --> 00:33:23.496 Jacinto, and these purple-y ones for the San Gorgonio Pass thrust fault system. 00:33:23.520 --> 00:33:28.240 And interestingly, it’s fairly linear over time. 00:33:28.240 --> 00:33:33.176 That wasn’t required as a – as a product of the inputs. 00:33:33.200 --> 00:33:36.880 What we see is that the biggest spikes visually, there is some – 00:33:36.880 --> 00:33:39.040 it wobbles a little bit around the average. 00:33:39.040 --> 00:33:43.760 And these biggest spikes are largely events that are on the – 00:33:43.760 --> 00:33:47.938 that cross the Mojave section, like the 1857 rupture does here. 00:33:48.720 --> 00:33:54.720 Another way to look at this is to examine the average of 00:33:54.720 --> 00:33:59.120 a couple of ruptures at a time. Because we have uncertainties in 00:33:59.120 --> 00:34:02.160 the ages, we don’t actually know, for example, if this San Jacinto 00:34:02.160 --> 00:34:05.840 rupture happened before or after the San Andreas rupture. 00:34:05.840 --> 00:34:11.040 So if I average over 2, shown in the dark line, or 3, shown in the red line, 00:34:11.040 --> 00:34:16.000 you can still, but you do – you see that you do see highs and lows, or lulls, 00:34:16.000 --> 00:34:20.400 where you have a drop in the available moment that’s happening. 00:34:20.400 --> 00:34:23.920 And then time periods like here in the medieval warming anomaly 00:34:23.920 --> 00:34:26.949 that sort of chatters at this moderate level. 00:34:27.600 --> 00:34:32.136 I think it’s worth comparing to what we have for the historic record here. 00:34:32.160 --> 00:34:38.536 There’s 61 years between the 1857 and the 1918 event. 00:34:38.560 --> 00:34:42.480 There’s 50 years up until the 1968 event. 00:34:42.480 --> 00:34:44.560 And so far, we’ve had 52 years since then. 00:34:44.560 --> 00:34:48.160 I’ll point out that those historic ruptures on the San Jacinto Fault are in here. 00:34:48.160 --> 00:34:53.040 They’re just these tiny, little green dashes. Because they were sub-7s, 00:34:53.040 --> 00:34:59.416 and they just barely expressed in this plot of cumulative moment. 00:34:59.440 --> 00:35:03.840 So the last thing we’re going to do is look at sort of a different measure 00:35:03.840 --> 00:35:07.600 for the quality of the maximum rupture model, or we’re really looking for 00:35:07.600 --> 00:35:12.160 ways to test how far off might it be, as the producing ruptures 00:35:12.160 --> 00:35:14.616 that are egregiously large. 00:35:14.640 --> 00:35:19.360 And what we’re going to do there is examine the – something called 00:35:19.360 --> 00:35:22.056 the average slip, or the predicted average slip. 00:35:22.080 --> 00:35:25.096 And we’ll get there in kind of a funny way. 00:35:25.120 --> 00:35:28.560 What I’ve done first is I’m going to take the rupture rate from the model. 00:35:28.560 --> 00:35:32.240 So, for each kilometer bin, I just calculated how many earthquakes 00:35:32.240 --> 00:35:35.600 per modeled century, and that’s what the bars are, and kind of 00:35:35.600 --> 00:35:39.280 visually smoothed it with this line. You can see it wobbles around a bit, 00:35:39.280 --> 00:35:43.920 depending on where ruptures end and then also, due to the model 00:35:43.920 --> 00:35:47.760 length changing a little bit. It’s a little lower in the Mojave, 00:35:47.760 --> 00:35:50.880 picks up again, and then drops down San Gorgonio Pass. 00:35:50.880 --> 00:35:52.800 This was – remember, there was about two earthquakes 00:35:52.800 --> 00:35:55.016 in 1,500 years through here. 00:35:55.040 --> 00:35:57.840 And rises up again towards Coachella. 00:35:57.840 --> 00:36:01.760 The San Jacinto Fault in green is staying fairly level, picks up 00:36:01.760 --> 00:36:05.736 a little bit here, although that’s a very short record. 00:36:05.760 --> 00:36:12.480 We’ll then compare this, or calculate this, with the slip rates from UCERF3. 00:36:12.480 --> 00:36:17.520 And then I added some updates here. So here the shaded areas represent 00:36:17.520 --> 00:36:19.920 the uncertainties on the slip rates in UCERF3. 00:36:19.920 --> 00:36:24.320 Some of the added ones, for example, since UCERF3 was published, we now 00:36:24.320 --> 00:36:28.000 have, from Heermance and Yule a couple years ago, a slip rate for the 00:36:28.000 --> 00:36:32.616 San Gorgonio Pass thrust fault – thrust fault zone, which is shown in purple. 00:36:32.640 --> 00:36:36.080 And what we can do is simply take the slip rate, divide it by 00:36:36.080 --> 00:36:41.096 the rupture rate, and that [audio cuts out] of slip. 00:36:41.120 --> 00:36:44.720 That’s shown in this shaded bar here. 00:36:44.720 --> 00:36:48.000 So what this would be is what you would predict is the average slip 00:36:48.000 --> 00:36:52.856 given the rupture rate and the geologic slip rate. 00:36:52.880 --> 00:36:55.600 And then I can compare those. Well, first, let me say, 00:36:55.600 --> 00:36:59.920 it gives a fairly expected value. You can see that, for the Carrizo 00:36:59.920 --> 00:37:04.400 and the Mojave, it hovers sort of around 3 to 4 meters – this would 00:37:04.400 --> 00:37:10.647 be a average slip per event – and begins to drop as you head into Cajon Pass. 00:37:11.440 --> 00:37:15.736 Things bobble around a little bit on the Banning and Coachella. 00:37:15.760 --> 00:37:21.360 And they’re fairly flat – sort of in the 2 to 4 meters of predicted 00:37:21.360 --> 00:37:26.776 average slip for the San Jacinto Fault. What we can do is compare that 00:37:26.800 --> 00:37:31.520 predicted value against independent measures of slip per event. 00:37:31.520 --> 00:37:36.800 And there’s two types that I’m showing by either the solid or the open ellipses. 00:37:36.800 --> 00:37:40.960 The solids are individually dated offsets, or average slip. 00:37:40.960 --> 00:37:43.120 And, by average, I mean short-term averages. 00:37:43.120 --> 00:37:47.016 So an example is this nice study by Onderdonk et al. in 2015, 00:37:47.040 --> 00:37:51.280 where he’s got a paleo channel – a buried channel through here. 00:37:51.280 --> 00:37:57.040 He excavated it and dated it, and the dates that that offset has accumulated 00:37:57.040 --> 00:38:01.096 are equivalent to three earthquakes at the nearby Mystic Lake site. 00:38:01.120 --> 00:38:04.720 And so, from that, I can say – I can get a estimate of what 00:38:04.720 --> 00:38:08.800 the average slip per event is. And that represents this green 00:38:08.800 --> 00:38:13.920 ellipse here at about 3 to 4 meters – 3-1/2 meters. 00:38:13.920 --> 00:38:17.576 And then, other places, which are geomorphic slip, this would be the 00:38:17.600 --> 00:38:22.000 work, for example, by Zielke et al. along the Carrizo section where 00:38:22.000 --> 00:38:26.000 they looked at what typical geomorphic offsets are. 00:38:26.000 --> 00:38:30.320 So these are generally undated but represented in the landscape 00:38:30.320 --> 00:38:32.400 by cumulative offset distributions. 00:38:32.400 --> 00:38:35.959 And we have a lot more of those because they’re much easier to do. 00:38:36.640 --> 00:38:40.800 What you see is that things actually are positioned pretty well 00:38:40.800 --> 00:38:43.360 on top of each other. The slip is – the predicted average 00:38:43.360 --> 00:38:50.856 slip is a little lower than these undated offsets along the northern part of the 00:38:50.880 --> 00:38:54.880 San Andreas Fault that we show here. And I would argue that’s because 00:38:54.880 --> 00:38:57.680 these things tend to miss the number of earthquakes 00:38:57.680 --> 00:38:59.736 that have actually occurred on them. 00:38:59.760 --> 00:39:04.696 In other places, we do really well. You can see nice overlap between, 00:39:04.720 --> 00:39:08.960 for example, a couple of offsets here at the Thousand Palms site and the slip 00:39:08.960 --> 00:39:14.240 rate, which is has large uncertainties, at the Mission Creek strand. 00:39:14.240 --> 00:39:18.720 What’s interesting and important here, if Ned Field is on the line, is that, 00:39:18.720 --> 00:39:22.480 if we had missed paleoearthquakes in the trenching records, we’d actually 00:39:22.480 --> 00:39:25.840 reduce the predicted average slip. So this shaded bar would drop further 00:39:25.840 --> 00:39:29.440 down, and it would be less likely to have done a good job of capturing 00:39:29.440 --> 00:39:33.176 what we have from independent data about slip per event. 00:39:33.200 --> 00:39:37.840 So this presents a way to look at the maximum rupture model in a different 00:39:37.840 --> 00:39:42.160 way and ask if it’s egregious, and the answer would be no. 00:39:42.160 --> 00:39:46.720 So, to summarize, this is definitely an end member model, and it produces 00:39:46.720 --> 00:39:50.376 the longest possible and thus the largest magnitude earthquakes. 00:39:50.400 --> 00:39:54.480 If you examine it qualitatively, it’s comparable to the historic record. 00:39:54.480 --> 00:39:59.360 We know we’re missing moderate events – sort of this mid-6s range, 00:39:59.360 --> 00:40:03.736 but they would be easy to produce if you had a rule for doing it. 00:40:03.760 --> 00:40:06.880 And, like the historic record for southern California, we see that 00:40:06.880 --> 00:40:10.616 we have epochs of activity and these permissible lulls. 00:40:10.640 --> 00:40:13.680 As well as time periods where the whole fault system seems to 00:40:13.680 --> 00:40:17.840 go off in a fairly short order. The average magnitude from 00:40:17.840 --> 00:40:22.216 the model is about a 7.2. The maximum is 7.8. 00:40:22.240 --> 00:40:25.360 And, as I just went over, it has a predicted slip that 00:40:25.360 --> 00:40:29.016 is reasonable with no to few missing events. 00:40:29.040 --> 00:40:31.840 The main thing I want to actually highlight is these new observations 00:40:31.840 --> 00:40:34.720 from San Gorgonio Pass fault and the Banning strand. 00:40:34.720 --> 00:40:38.720 They’re really important data that should really open up our 00:40:38.720 --> 00:40:45.440 understanding of this part of the system. And importantly, if you look at those 00:40:45.440 --> 00:40:49.200 data sets, only one in four ruptures from either side, either on the 00:40:49.200 --> 00:40:53.440 Coachella strand or the San Bernardino strand can make it through that system. 00:40:53.440 --> 00:40:57.120 That means they have one or less than one ShakeOut-like event that really 00:40:57.120 --> 00:41:01.656 makes it all the way from the Salton Sea to Palmdale, for example. 00:41:01.680 --> 00:41:04.480 There’s a lot of testable ideas put forward in this model. 00:41:04.480 --> 00:41:08.616 The idea of the Cajon Pass sequences certainly needs more investigation, 00:41:08.640 --> 00:41:13.840 whether or not they are close enough space and time to matter, that geometry 00:41:13.840 --> 00:41:18.160 is quite permitted by both static and dynamic rupture modeling, but how 00:41:18.160 --> 00:41:22.720 you get both the San Bernardino and the Claremont to rupture is – 00:41:22.720 --> 00:41:25.920 remains a open question. And then the idea that we threw out 00:41:25.920 --> 00:41:30.480 there, that periods of activity in Cajon Pass are more frequent and then they – 00:41:30.480 --> 00:41:34.616 but they alternate with the San Gorgonio Pass events. 00:41:34.640 --> 00:41:41.200 So I want to conclude by emphasizing this one San Gorgonio ShakeOut-ish- 00:41:41.200 --> 00:41:46.000 like event around 620 A.D. and then acknowledge Doug Yule here. 00:41:46.000 --> 00:41:49.680 This is Doug with his family standing at the Cabazon trench site. 00:41:49.680 --> 00:41:54.160 You can see really nicely expressed here a fault ripping through the 00:41:54.160 --> 00:41:58.960 stratigraphy there offsets this buff-to-red contact here. 00:41:58.960 --> 00:42:02.320 Amusingly – I know it looks like a COVID picture, but this work 00:42:02.320 --> 00:42:06.800 was done in about 2013, and we had some problems with people 00:42:06.800 --> 00:42:09.680 getting valley fever at the site, so we were always wearing masks then. 00:42:09.680 --> 00:42:15.200 So luckily, I had, still, a lot of masks in my field truck when COVID happened. 00:42:15.200 --> 00:42:19.696 So, with that, I’m done. And I would love any questions. 00:42:20.800 --> 00:42:24.000 - All right. Thanks, Kate. That was a wonderful talk, and we 00:42:24.000 --> 00:42:27.680 really appreciate you sharing it with us. We do have a couple questions in the 00:42:27.680 --> 00:42:31.120 meeting chat, but I also want to invite people to unmute and 00:42:31.120 --> 00:42:32.160 ask your questions yourself. 00:42:32.160 --> 00:42:35.722 You can also raise your hand to let us know if you have a question. 00:42:36.960 --> 00:42:39.896 So, to get started, we’ll ask some of the ones from the chat. 00:42:39.920 --> 00:42:43.360 Janet Watt asked, do you know where the San Jacinto and 00:42:43.360 --> 00:42:48.000 San Andreas Fault meet at depth? - [laughs] 00:42:48.000 --> 00:42:51.360 No. And if you know someone who does [laughs], they want to – 00:42:51.360 --> 00:42:54.616 they have – a lot of people are interested in knowing as well. 00:42:54.640 --> 00:42:57.600 We don’t even know – they don’t appear to meet at the surface. 00:42:57.600 --> 00:43:02.320 They get sub-parallel, and then there’s no obvious geomorphic 00:43:02.320 --> 00:43:07.016 connector between the two. The existing models out there 00:43:07.040 --> 00:43:11.360 tend to have a couple of options. One is the San Andreas being vertical, 00:43:11.360 --> 00:43:14.519 which I think is kind of a generic option through there. 00:43:15.760 --> 00:43:20.136 And then, of course, Gary Fuis has put forward the idea that the 00:43:20.160 --> 00:43:22.560 San Andreas through there is actually dipping to the northeast. 00:43:22.560 --> 00:43:26.480 So they wouldn’t necessarily connect unless the San Jacinto Fault also did that. 00:43:26.480 --> 00:43:32.000 So, so far, the jury is out on the exact geometry through there. 00:43:32.000 --> 00:43:36.240 I will state, for the dynamic rupture models would maybe say, 00:43:36.240 --> 00:43:38.800 ah, it’s not so important because they do come within, I think it’s 00:43:38.800 --> 00:43:45.096 2 to 3 kilometers of each other, so it’s a easy enough gap to traverse. 00:43:45.120 --> 00:43:48.720 Is that okay an answer? - I think so. 00:43:48.720 --> 00:43:51.656 The person who asked it can pop up if they need more. 00:43:51.680 --> 00:43:55.440 Okay. From – we had a couple hands raised. David Schwartz, 00:43:55.440 --> 00:44:00.196 do you want to turn on your audio? - Kate. Great talk. 00:44:01.120 --> 00:44:05.360 I was curious. How do your results fit in with the 00:44:05.360 --> 00:44:11.920 paradigm in UCERF3 of both the San Jacinto and the San Andreas 00:44:11.920 --> 00:44:17.040 being involved with ruptures that extend into northern California 00:44:17.040 --> 00:44:20.400 and perhaps up to the Mendocino fracture zone? 00:44:20.400 --> 00:44:29.280 So you’re talking about maximum ruptures in southern California. 00:44:29.280 --> 00:44:32.480 What about the connections and possibilities of ruptures 00:44:32.480 --> 00:44:37.029 going further north or starting north and coming south? 00:44:37.029 --> 00:44:42.640 - That’s a good question. The argument we put out there is that 00:44:42.640 --> 00:44:47.680 the creeping section, most of the time, is a pretty strict boundary. 00:44:47.680 --> 00:44:51.280 As well as the Imperial. So you’ll see that, in the model domain, 00:44:51.280 --> 00:44:55.760 I ended it, basically, at the creeping section and to the south at the Imperial. 00:44:55.760 --> 00:44:58.400 So I’m not assessing whether or not they go out there. 00:44:58.400 --> 00:45:01.360 If they do, then there are larger ruptures out there. 00:45:01.360 --> 00:45:06.880 I don’t worry about it that much. I think it is rare, geologically, 00:45:06.880 --> 00:45:09.840 from the information that Nathan Toké and others have published. 00:45:09.840 --> 00:45:13.280 And it’s also rare in UCERF, in that they are not frequent events, 00:45:13.280 --> 00:45:17.376 but they are large when they happen. You’re right about that concern. 00:45:18.368 --> 00:45:21.404 - Great. Thank you. Really good presentation. 00:45:21.404 --> 00:45:22.761 - Thanks. 00:45:23.816 --> 00:45:27.196 - All right. You had another question from Scott Bennett. 00:45:29.282 --> 00:45:31.016 - Hey, Kate. Great talk. 00:45:31.040 --> 00:45:33.983 Congrats on the new paper with you and Doug. 00:45:34.960 --> 00:45:38.616 I was wondering about your comparison of the historic 00:45:38.640 --> 00:45:42.240 earthquakes and the prehistoric – that plot – the histogram where 00:45:42.240 --> 00:45:48.936 the yellow and the gray lines divert, kind of in the mid-6 range. 00:45:48.960 --> 00:45:51.976 I was wondering – kind of a two-part question. 00:45:52.000 --> 00:45:55.440 Are the historic earthquakes, are they are all surface rupturing? 00:45:55.440 --> 00:45:58.776 Or is that just kind of anything that’s been reported? 00:45:58.800 --> 00:46:02.704 I only included the surface-rupturing ones there. 00:46:02.704 --> 00:46:04.560 - Okay. - Because that would be all I could see 00:46:04.560 --> 00:46:09.360 in my tally of the paleoearthquakes. There are a couple of other – 00:46:09.360 --> 00:46:13.360 like, there’s an 1899 Casa Loma, which I didn’t include. 00:46:13.360 --> 00:46:19.040 It does have – some folks went out and toured after that event 00:46:19.040 --> 00:46:21.760 and then wrote it up later. And so that would be on the 00:46:21.760 --> 00:46:26.960 San Jacinto Fault, for folks – and it does seem – in the – in the records, 00:46:26.960 --> 00:46:30.080 they talk about – basically, it looks like they saw liquefaction and that 00:46:30.080 --> 00:46:32.640 there was – they talked about it, kind of a fissury thing. 00:46:32.640 --> 00:46:36.960 But it’s pretty short. At any rate, it’s not included in that. 00:46:36.960 --> 00:46:40.080 But those types of things – definitely, like, any paleoseismology, 00:46:40.080 --> 00:46:42.535 start to get in the 6.5s, and our … - Right. 00:46:42.535 --> 00:46:45.027 - .. lower threshold gets pretty murky. 00:46:45.027 --> 00:46:48.240 - Sure. Yes, okay. I was thinking – so it sounds like 00:46:48.240 --> 00:46:52.000 you’re comparing apples to apples. You know, surface-rupturing historic 00:46:52.000 --> 00:46:56.880 versus surface rupturing prehistoric. Do you have any insights on the 00:46:56.880 --> 00:47:01.280 likelihood of, you know, 6.5s that are sneaking in that are blind, and you’re 00:47:01.280 --> 00:47:08.501 just – you just can’t see or – in the trenches? Or how likely that might be? 00:47:08.501 --> 00:47:12.080 - I think we’d probably pick a few of them up, and the maximum rupture 00:47:12.080 --> 00:47:14.400 model makes them longer than they should be. 00:47:14.400 --> 00:47:20.080 And example would be one at Frazier that has fairly modest surface 00:47:20.080 --> 00:47:26.056 expression. There isn’t a correlable earthquake at either site to either side. 00:47:26.080 --> 00:47:28.640 But the maximum rupture model, because it’s 50 kilometers, 00:47:28.640 --> 00:47:32.485 or something like that, to the next site, makes it a pretty long event. 00:47:33.680 --> 00:47:39.360 So I’m sure – I’m sure we both – my view is, we probably capture 00:47:39.360 --> 00:47:44.240 a couple mid-6s, and we also miss a lot of mid-6s. 00:47:44.240 --> 00:47:46.400 - Sure. That makes sense. Well, thanks again. 00:47:46.400 --> 00:47:48.080 That was a wonderful talk. - Thanks. 00:47:48.080 --> 00:47:50.000 Thanks for your field work help. [laughs] 00:47:50.000 --> 00:47:52.899 - Yeah, you bet. - Elizabeth Lake is coming out soon. 00:47:55.082 --> 00:47:57.200 - Great. Thank you. And then, in the chat, we have, 00:47:57.200 --> 00:48:00.400 Victoria Langham asked, what Coulomb stress calculations 00:48:00.400 --> 00:48:04.240 indicate whether Cajon Pass – Cajon Pass sequenced earthquakes 00:48:04.240 --> 00:48:08.176 trigger San Gorgonio Pass earthquakes or vice versa? 00:48:09.191 --> 00:48:12.175 - I look forward to someone doing that. [laughs] 00:48:12.175 --> 00:48:16.720 I guess I could do that. The studies that are out there – 00:48:16.720 --> 00:48:20.240 there’s a paper for static. There’s a paper by Anderson et al. 00:48:20.240 --> 00:48:28.000 2003, and he – they look – instead of San Gorgonio Pass, they basically 00:48:28.000 --> 00:48:31.120 don’t take it that far to the east, but they do examine whether or not 00:48:31.120 --> 00:48:35.200 the Cucamonga would be – where the clock would be advanced 00:48:35.200 --> 00:48:39.336 or delayed from San Jacinto or San Andreas ruptures. 00:48:39.360 --> 00:48:41.920 Overall, it fits the geometry as you would expect. 00:48:41.920 --> 00:48:44.480 You know, the San Jacinto promotes San Andreas ruptures. 00:48:44.480 --> 00:48:48.160 San Bernardino promotes San Andreas. 00:48:48.160 --> 00:48:52.376 And you don’t tend to promote that around the hairpin curve. 00:48:52.400 --> 00:48:55.920 But, to my knowledge, no one’s done static modeling of 00:48:55.920 --> 00:49:00.845 San Gorgonio Pass thrust. - Okay. Something still to do then. 00:49:00.845 --> 00:49:03.576 - Many testable ideas. - Yeah. 00:49:03.600 --> 00:49:06.464 And then, Austin Elliott, you had your hand raised? 00:49:07.542 --> 00:49:10.160 - Yeah, sure. Well, I think you sort of answered 00:49:10.160 --> 00:49:12.936 the question I had in response to Scott’s question. 00:49:12.960 --> 00:49:16.560 But – and you may have said this. I apologize if I sort of missed it in 00:49:16.560 --> 00:49:22.000 the talk, but in defining the maximum rupture model – so you define – 00:49:22.000 --> 00:49:25.600 you draw these rupture lengths, and how do you really sort of 00:49:25.600 --> 00:49:29.360 pin down what the endpoint is? When you have 40 or 50 kilometers 00:49:29.360 --> 00:49:34.400 between sites, shifting that makes a big difference in the magnitude 00:49:34.400 --> 00:49:36.560 that you get on either side and then the implications for the 00:49:36.560 --> 00:49:39.816 scaling relationships. So what sort of uncertainties 00:49:39.840 --> 00:49:44.800 does that add in, overall, the magnitudes of earthquakes that you 00:49:44.800 --> 00:49:52.080 interpret, and I guess, is it relevant? - The rule is they end halfway between 00:49:52.080 --> 00:49:55.520 the last site that had a correlable earthquake and the next site over. 00:49:55.520 --> 00:49:58.960 So it depends where you are. There’s generally about 25 kilometers 00:49:58.960 --> 00:50:03.256 between sites, on average. It gets denser in some spots. 00:50:03.280 --> 00:50:10.616 So a 25 – that’s, like, a magnitude 6.7, maybe 6.8, off the top of my head. 00:50:10.640 --> 00:50:14.000 If you – if you think that the rupture happened in the 00:50:14.000 --> 00:50:17.760 kilometer after the site. [laughs] You could envision a more complicated 00:50:17.760 --> 00:50:20.240 scenario, where you say, okay, I’m going to apply kind of a 00:50:20.240 --> 00:50:25.360 random length or tail to that based on the distance or some fraction 00:50:25.360 --> 00:50:29.816 of the distance to the next site. And fiddle with it that way. 00:50:29.840 --> 00:50:38.640 What the maximum – the rule ends up basically filling out the rupture rate so 00:50:38.640 --> 00:50:43.760 that the rupture rate is sustained, more or less, between sites with sort of 00:50:43.760 --> 00:50:50.536 modest changes to adjust between sites is the best way to think about it. 00:50:50.560 --> 00:50:55.680 But it – again, it would be really easy to produce some magnitude mid-6s 00:50:55.680 --> 00:51:03.440 just by – and still retain the total event evidence by just sort of cutting some 00:51:03.440 --> 00:51:06.880 of those ruptures into smaller ones. But I don’t know – you know, 00:51:06.880 --> 00:51:09.440 I think there’s lots of seismologists that would say that I shouldn’t 00:51:09.440 --> 00:51:14.376 try to match the historic record. That’s a little short. [laughs] 00:51:14.400 --> 00:51:18.016 I think it’s worth just observing it and knowing that they could be done. 00:51:19.027 --> 00:51:20.336 - Yeah. Thanks. 00:51:20.871 --> 00:51:23.613 - You had a question from Glenn. 00:51:25.480 --> 00:51:32.320 - Hi, Kate. Great talk. Could you review why we – 00:51:32.320 --> 00:51:36.880 so you presented evidence for two potential earthquakes through – 00:51:36.880 --> 00:51:41.016 based on trenching at Millard and the super trench in the pass. 00:51:41.040 --> 00:51:47.280 Could you review for us the evidence against big ruptures 00:51:47.280 --> 00:51:50.720 taking some other path? I mean, you found two. 00:51:50.720 --> 00:51:53.040 Could there be others hiding in the mountains, 00:51:53.040 --> 00:51:57.776 say, one flake back in the San Bernardinos? 00:51:58.657 --> 00:52:03.416 - Yes. [laughs] There certainly could be. 00:52:03.440 --> 00:52:07.976 The Millard site – so there’s basically, in terms of the dominant 00:52:08.000 --> 00:52:10.080 mapped faults – I don’t know if Doug Yule made it to this, 00:52:10.080 --> 00:52:13.680 but Doug Yule is the person who’s done the – most of the mapping 00:52:13.680 --> 00:52:17.600 back there with John Matti Kerry Sieh and others, but there’s sort of a – 00:52:17.600 --> 00:52:21.576 within the San Gorgonio Pass thrust fault zone, there’s two dominant 00:52:21.600 --> 00:52:24.960 geomorphic expressions of it. The Millard site is on the 00:52:24.960 --> 00:52:27.760 more northerly one. And the Cabazon site is on the 00:52:27.760 --> 00:52:30.640 more southerly one, and they both show the same record. 00:52:30.640 --> 00:52:35.040 So that gives us some confidence that – and they all connect in kind of 00:52:35.040 --> 00:52:39.280 complicated ways. But that gives us some confidence that we are seeing 00:52:39.280 --> 00:52:43.055 the same event at both sites and that it really ruptured through there. 00:52:44.400 --> 00:52:47.520 There are – I guess the other reason to argue for it would be 00:52:47.520 --> 00:52:51.280 that the Banning site, at least – so incoming to the southeast, 00:52:51.280 --> 00:52:55.840 also only has two earthquakes. And so we see – you know, the 00:52:55.840 --> 00:53:00.683 ruptures coming in from the east – southeast also, are only two. 00:53:01.360 --> 00:53:03.840 That said, you know, the Palm Springs event happened 00:53:03.840 --> 00:53:07.920 somewhere within that massif. So they could certainly be these 00:53:07.920 --> 00:53:11.760 magnitude 6s, or even 7s, that are pretty massive but don’t produce 00:53:11.760 --> 00:53:17.576 a very long surface rupture to connect out to other parts on the San Andreas. 00:53:17.600 --> 00:53:21.851 But we could always do some more trenching to examine that. [laughs] 00:53:21.876 --> 00:53:27.040 - Can we – can we discount a Mission Creek-type – or, Mill Creek-type thing 00:53:27.040 --> 00:53:29.472 that’s yet farther in the mountains? 00:53:29.513 --> 00:53:32.320 - That’s a very important, outstanding question. 00:53:32.320 --> 00:53:35.280 The Thousand Palms site, which is at the very southern 00:53:35.280 --> 00:53:39.520 end of the Mission Creek strand, actually matches – has many more 00:53:39.520 --> 00:53:42.256 earthquakes – has five, I think. - Five, yeah. 00:53:42.256 --> 00:53:48.160 - Yeah. And so what ends up in sort of the discussions that happen around that 00:53:48.160 --> 00:53:53.120 part of the world is that some folks think that the Mission Creek expression 00:53:53.120 --> 00:53:57.200 as you move towards the northwest, that they argue for termination or 00:53:57.200 --> 00:54:01.600 aggressive slowing of that system down to more like a 1 to 3 millimeters a year, 00:54:01.600 --> 00:54:05.040 which is what is in UCERF. Other folks – Kim Blisniuk 00:54:05.040 --> 00:54:10.536 and I have work in Thousand Palms that suggests it’s higher. 00:54:10.560 --> 00:54:15.176 We all diverge a little bit on where that slowdown might happen. 00:54:15.200 --> 00:54:20.160 But, so far, there’s not a great paleoseismic or short-term slip rate 00:54:20.160 --> 00:54:27.483 farther into San Gorgonio Pass on that northern branch to test these arguments. 00:54:27.483 --> 00:54:30.936 - Thanks. - Thank you. [laughs] 00:54:30.960 --> 00:54:33.200 Scott, you still have your hand up. I don’t know if you had another 00:54:33.200 --> 00:54:37.840 question or if that’s just a leftover hand from earlier. 00:54:39.440 --> 00:54:41.976 Well, seeing lots of thank-yous in the chat. 00:54:42.000 --> 00:54:44.560 People really appreciating the talk that you gave and say 00:54:44.560 --> 00:54:47.485 it’s great – you did a great job. - Thanks. 00:54:50.087 --> 00:54:52.080 - Yeah. So I don’t know if there are any other questions that 00:54:52.080 --> 00:54:55.766 anyone wants to speak up? Another thank-you. 00:54:57.120 --> 00:55:01.440 So, yeah, I think we’ll kind of wrap up the formal Q-and-A session right now, 00:55:01.440 --> 00:55:04.320 but, people, we ask you – ask people to please stick around if 00:55:04.320 --> 00:55:09.200 you would like to talk with Kate more. Sort of unmute and turn on your video. 00:55:09.200 --> 00:55:12.640 We’ll turn off the recording so we can kind of continue a more informal 00:55:12.640 --> 00:55:16.560 discussion and just chat that way. All right. 00:55:16.560 --> 00:55:18.320 So thank you again for a wonderful talk. 00:55:18.320 --> 00:55:22.000 It was great to have you [inaudible]. - [inaudible]