SCEC Award Number 19225 View PDF
Proposal Category Individual Proposal (Data Gathering and Products)
Proposal Title Collection of Very High-Resolution Aerial Imagery For Assessment of Creep on the Southernmost San Andreas Fault
Investigator(s)
Name Organization
Thomas Rockwell San Diego State University Allen Gontz San Diego State University
Other Participants Yuri Fialko
Roger Bilham
SCEC Priorities 1a, 5c, 5d SCEC Groups Geology, SDOT, SAFS
Report Due Date 04/30/2020 Date Report Submitted 11/08/2020
Project Abstract
 We collected aerial imagery along the creeping southernmost San Andreas fault to resolve the
displacement field from offset rills that post-date the most recent Lake Cahuilla highstand (~ 1726 CE),
rills and stream channels that survived the last inundation of Lake Cahuilla and record displacement in
the most recent large earthquake (MRE), and to provide an archive for future studies of surface creep and
earthquake displacement through repeat and post event surveys. Specifically, we produce very high-
resolution imagery and associated dense point clouds when we fly at 25 m agl, and subsequently, these data and imagery can be used to compare to future imagery with optical correlation algorithms (COSICOR
and similar programs).
Our results are mixed but provide useful information on the limits of UAV surveys under less than
ideal conditions. Our initial surveys were conducted during periods of higher wind and a non-favorable
GPS satellite constellation, which resulted in poor imagery collection, even at 25 m flight elevation.
Resurveys during ideal conditions resulted in excellent imagery with sub-cm pixel resolution. Analysis
of offset and beheaded channels suggests about 3-3.5m of lateral slip occurred in the MRE along the
Durmid section of fault, very similar to what we documented in the Mecca Hills (Blanton et al., 2020),
indicating of 3-3.5 m of slip for at least 40 km of fault length in the ca 1726 earthquake.
Intellectual Merit Resolving slip distribution in past large San Andreas earthquakes and comparing to the strain that has accumulated in the past 300 years provides information on earthquake repeatability, strain cycles, and the broader questions of impending hazard from a large southern San Andreas earth-quake. Analysis of the spatial distribution of individual creep events, and comparison to longer term InSAR measurements and creep-meter signals may provide better information on the thick-ness of the creeping zone, and therefore the seismic thickness, as well as the variability in creep rate as a precursory signal. If we are able to resolve deformation signals as small as 1-2 cm, this will be an advance in field applications of UAV technology.
Broader Impacts The field data acquisitions were conducted with students at SDSU. In the winter/spring of 2020, field surveys were part of Rockwell’s neotectonics class, and the gathered data were intended to be used by the students to learn to process and interpret the collected data. In addition to the UAV data collection, students collected data on geomorphic offset rills and channels in the field. Then COVID hit and we were unable to complete the project with the students, with the ultimate goal being a student-led publication of the results, similar to the Blanton et al. (2020) work.
Exemplary Figure Figure 5: Example of a small truncated channel offset about 3.2 m, with sediment ponded behind a small scarp. Flow was to the southwest (lower left), with the southwest side of the fault uplifted, resulting in deeper channel incision to the southwest. A younger, post-lake channel is deflected about a half meter, presumably due to creep. The contour interval in the topographic map, based on the DEM created in Agisoft, is 10 cm.