Goodbye to SoSAFE (Southern San Andreas Fault Evaluation): review and a recent workshop

The SoSAFE activity was a very successful data gathering and interdisciplinary science rallying activity in the Southern California Earthquake Center (SCEC) for 10 years. Its early leader was Dr. Ken Hudnut from the USGS. The original aspiration was to develop understanding of the last 2000 years of activity along the Southern San Andreas Fault. I was fortunate to be invited to help co lead SoSAFE in 2011. Dr. Kate Scharer (USGS) was the other co-leader. Her recent research has been largely focused on understanding the southern San Andreas Fault paleoseismic history. I very much appreciated the chance to work with Kate to help coordinate the research activity of SoSAFE withing SCEC as part of its Planning Committee. Our main efforts included evaluating proposals and helping coordinate and promote the research of our colleagues. We did a few additional activities including the "FieldShop" in which we made a group field trip to discuss and assess small offset landforms along the San Andreas Fault Mojave segment near Pearblossom. The outcome of the FieldShop was a paper lead by Kate: Scharer, K. M., Salisbury, J. B., Arrowsmith, J R., Rockwell, T. K., Southern San Andreas Fault Evaluation field activity: Approaches to measuring small geomorphic offsets and challenges and recommendations for active fault studies, Seismological Research Letters, v. 85, no. 1, ppl 68 - 76, 2014. In addition, we also organized (along with Prof. Mike Oskin from UC Davis) the SoSAFE and Earthquake Geology Geochronology workshop in 2014. The discussions of novel applications of geochronology for earthquake geology were really interesting and helpful for the community. SoSAFE has been sunsetted as an activity within SCEC as it moves into its 5th iteration (SCEC5). Much of the important work of SoSAFE will be subsumed (and hopefully continued) in the San Andreas Fault System Working Group to be led by Kate and Prof. Michele Cooke (UMass). It makes some sense I suppose as a refresh on SCEC structure, but I and others are concerned about the loss of emphasis on paleoseismic data gathering.

Nice cover image produced by Barrett Salisbury (ASU).

In our last SoSAFE activity, Kate and I organized a workshop for Sept. 10, 2016: SCEC SoSAFE Workshop: Recent Successes and Future Challenges. The workshop had 3 main themes:

1. Earthquake recurrence and slip over short and long term: how does it all add up?
2. Integrating earthquake and paleoclimate/paleoenvironmental chronologies on the SoSAFE System
3. Outside looking in: Broad applications of Behavior of high slip rate faults

The workshop was well attended (we capped it at 40 people, but many more wanted to join).

Several synthesis points were evident from the presentations and discussions:

• The opportunity to and importance of documenting the full spectrum of slip behavior at a range of slip speeds (creep to seismic) and both on and off fault. Where in space and time along and adjacent to the fault surface is there deformation, and how might the behavior vary over time? How to capture the integrated effect of fine scale fracturing? (Toké, Milliner, Lindsey)
• Variable slip at a point in successive earthquakes is suggested by detailed evidence from many sites. In some paleoseismic records, we may be seeing both small and large events mixed in the same record. This is a good picture of the earthquake phenomena but a greater challenge for interpretation. (Dawson, Salisbury, Rockwell, Biasi)
• Paleoseismic event recognition and resolveability. Are we under or over counting paleo earthquakes? The consensus among the group (and as analyzed by Biasi) was that both happen and so suggestions of systematic overcounting (e.g., D. Jackson) were not supported by experience. But, there certainly is value in exploring ways to systematize paleoseismic data (evidence, age control, correlation, etc.). (Dawson, Rockwell, Biasi, Milner). Tim Dawson reminded us of the important work of Bonilla and Lienkamper, 1991. The time scale during which many of the comparisons are being made (~1000 yrs) may be too short to completely assess the question of moment rate fluctuation across the SoSAFE system.
• Slip rates vary in space and time probably as a function of evolving fault geometry (at multiple length scales) and mechanical interaction (Cooke, Onderdonk).
• Interpreting offset per event from the reconstruction of fine scale landforms remains challenging. 3D excavation and reconstruction is desired but time consuming (and itself can have subtleties and ambiguities). There is a "... tension between collection of observations at as many locations as possible (assuming there will be signal in the noise [Large N]) versus inclusion of only data that are clearly offsets (rather than deflections) and have good quality ranking." (Scharer, et al. 2014). Interpretation of small geomorphic offsets will not go away given its convenience, the availability of high resolution imagery and topography, and the potential to assess remote structures. But, we must continue to validate and push to understand what we are measuring, entertain alternatives, identify 3D anchor sites, etc. (Salisbury)
• Earthquake simulators are moving forward as the preferred integrative tool for forecasting earthquake behavior across the SAF system ("Paleoearthquake data and slip rates battle it out in UCERF3--RSQSIM to the rescue"--Kevin Milner, USC). Jaqui Gilchrist (USC) gave a nice presentation reviewing the simulator approach. Her faults are quite smooth from the perspective of the many geologists in the room. Importantly, we learned about the tuning that is done by adjusting fault normal stresses to match paleoearthquake rates across the system. She showed that there is a gap between the curated paleoearthquake datasets she has used and the data producers. There is a need for deeper access to paleoearthquake data and metadata. SCEC-VDO was rebuilt in 2016 and is a valuable tool for visualization and exploration.
• Field earthquake geology is time and resource consuming. Assuming that the site conditions are good enough to preserve a high quality record, it is important to recognize that it takes at lot to produce high quality field studies, both in the field, as well as in the geochronology laboratory.

The early afternoon featured thought provoking talks about paleoclimate and paleoenvironmental chronologies. Here are a few of my takeaways:

• Kate showed in her 2014 paper (Scharer, K. et al., 2014b. Paleoearthquakes at Frazier Mountain, California delimit extent and frequency of past San Andreas Fault ruptures along 1857 trace. Geophysical Research Letters, 41(13), pp.4527– 4534.) how using paleosol and sediment accumulation curves would illuminate similarly timed landscape variation in California. She demonstrated that it was likely that earthquakes with overlapping ages at separate paleoseismic sites were different because their evidence was above and below a period of slow sediment accumulation (a paleosol). This is quite exciting and a frontier for earthquake geology. The challenge is what is the best proxy? Paleoprecipitation indicators? or Pollen or fire? How to balance convenience and the ability to measure with environmental sensitivy and ability to date?
• Prof. Matt Kirby (CSU Fullerton) presented a nice review of paleoclimate and paleoenvironmental proxies (temperature, precipitation, circulation, flooding, fire, other geomorphic disturbance). He focused on the rare lake records of southern California and differentiated millenial, centennial, sub-centenial to decadal, and annual time scales and their different drivers (mostly interactions with the Pacific Ocean). The millenial scale offers an opportunity to look at (synchronous formation of large markers along the SoSAFE System--e.g., Wallace Creek at 3,700 years BP and other similarly aged offset landforms). The sub-centennial to decadal (or finer scales) offer the opportunity for further differentiation of paleoearthquakes (e.g., Scharer, et al., 2014b) as well as the formation of small scale markers for single or few event offsets.
• Matt called out the need for common protocols, the value of looking at sections together, and that there were lots of interesting possible sites out there. We can use existing sites as benchmarks and prospect for new ones. This work costs $$!
• Prof. Nick McKay (NAU) offered valuable perspective as well as a potential path forward for organizing paleoseismic and paleoclimate data. He talked about the PAGES2K effort as a distributed global collaboration on past climate. It relies on a cyberinfrastructure of linked paleodata (McKay, N., and Emile-Geay, J., 2015, Technical Note: The linked paleo data framework – a common tongue for paleoclimatology, Climate of the Past, 11, 4309-4327.). He ended with some commentary on the challenges of supporting such a data effort (1) Identify metadata, 2) Structure metadata hierachically, 3) Be thoughtful about Selection Criteria, 4) Iterative data and metadata assimilation, and 5) Flexible scientific control).

Many thanks to our colleagues for their great ideas and community spirit in support of SoSAFE. Adieu!!