AUTOMATIC MEASUREMENT OF SCARP HEIGHT ALONG NORMAL FAULTS FROM HIGH RESOLUTION TOPOGRAPHY IN THE VOLCANIC TABLELANDS, CALIFORNIA

Observations of slip distribution along faults offer insight into the seismic history, the mechanics of faulting, and seismic hazard. Slip magnitude, fault segmentation, and fault length hold critical information about long-term fault growth and the elastic properties of faults and the surrounding rock. Observations of fault zones from high resolution topography (less than 1 m per pixel) acquired by satellite and airborne-based platforms facilitate constraining geometric fault parameters over a range of spatial scales. Systematically extracting the relevant features from topography in a fashion that is both reproducible and uses all of the available data is aided by tools that automate mapping.

We apply a new algorithm for mapping faults and measuring scarp height to a suite of normal faults located within the Volcanic Tablelands of Owens Valley, California. These faults break the relative smooth surface on the top of the Bishop Tuff (758.9 +/- 1.8 ka), where they form well-preserved topographic scarps tens of meters to several kilometers in length. We measure topography along the scarps from three data types: (1) a point cloud derived from structure-from-motion techniques and unmanned aerial vehicle (UAV) imagery (2 cm resolution), (2) lidar imagery hosted on the OpenTopography archive (30 cm resolution; Delano (2014), doi.org/10.5069/G9RJ4GCH), and (3) Pleiades stereo satellite imagery (70 cm resolution). Our new algorithm measures the cumulative throw by distinguishing between steep faults and the adjacent flatter hanging and footwall by topographic steepness averaged over several meters. The estimated scarp heights have an uncertainty of 0.5 m. Initial results for the algorithm and hand-mapped fault architecture agree well for the master faults, although the automatic approach is less sensitive to the secondary faults. We compare the vertical slip estimates derived from the three datasets discussed above to field-based laser range finder measurements of scarp height and a second algorithm that measures scarp height based on the local curvature of topography. We anticipate that our results will support characterizing key fault properties in a systematic manner, and ultimately refining fault length, width, and slip scaling laws.