SCEC Award Number 16147 View PDF
Proposal Category Collaborative Proposal (Data Gathering and Products)
Proposal Title High-Resolution Topographic Mapping of Southern California with Optical Imagery
Name Organization
William Barnhart University of Iowa Michael Willis Cornell University Rich Briggs United States Geological Survey
Other Participants 1 SCEC Intern (possible Terryl Bandy, University of Iowa undergrad)
SCEC Priorities 4a, 4b, 4c SCEC Groups USR, Geology, Geodesy
Report Due Date 03/15/2017 Date Report Submitted 03/31/2017
Project Abstract
Digital elevation models (DEMs) are a fundamental input for a range of applications and studies in active tectonics, including mapping active faults, modeling landscape evolution and fluvial response to lateral and vertical fault displacements, and measuring co-seismic displacements following large earthquakes. There is, however, a large discrepancy between the spatial resolution of freely available global DEMs (SRTM, USGS, 10-30 m/pixel) and ultra high-resolution (<5 m/pixel) LiDAR and UAV-generated DEMs that are spatially limited around faults in southern California. Additionally, it is impossible or cost intensive to reacquire many of these datasets to form time series of spatially and temporally evolving landscapes. This project sets out to take advantage of the increasing volume of stereo-optical imagery acquired by commercial satellites as means to generate regional, high-resolution DEMs over southern California. Using imagery acquired from the World View and GEOEYE constellation of satellites, we implement a processing routine that uses the open source SETSM program, developed by the Ohio State University, to generate 2m DEMs spanning the San Gorgonio and Ventura Special Fault Study Zones, downtown Los Angeles, and portions of the southern San Andreas and San Jacinto Fault systems. We compare the generated DEMs to B4 LiDAR acquisitions where the two data sets overlap and find that the datasets agree within an uncertainty of ~1m. Our current efforts focus on expanding the spatial coverage of the DEM dataset, filling voids in the generated DEMs with supplemental imagery, and providing the DEMs to the broader science user community via OpenTopography.
Intellectual Merit This research contributes to a more detailed understanding of fault zone geomorphology and supports efforts to better ascertain the surface geometry of spatially complex faults in southern California. The generated datasets likewise provide a “pre-seismic” dataset that can be used with “post-seismic” datasets to rapidly evaluate earthquake impacts, both geologic and societal, following future earthquakes in southern California.
Broader Impacts We are currently making this dataset available to the scientific community as a community data resource via OpenTopography. We have already, and continue to, made the DEMs available on an as-requested basis. The datasets also provide an important resource for the rapid assessment of the human impacts of potential large southern California earthquakes in the future. The project additionally supported an early career, female undergraduate student and continues to support a second undergraduate student at the University of Iowa.
Exemplary Figure Figure 1 (Project #16147): Hillshades of three 2-meter DEMs generated from WorldView-2 stereo imagery. The first highlights the trace of the San Andreas Fault and near-fault tectonic geomorphology near the town of Valyermo. The second highlights the locations of landslides on the flank of Anderson Peak near the San Gorgonio Pass (red arrows). These landslides are not readily apparent in optical imagery. The third highlights the high-quality of these DEMs over urbanized regions, as highlighted by Los Angeles International Airport (LAX). The quality of the results in urban areas was unexpected, and the production of DEMs in these regions is motivated by the capability to image building damage following earthquakes in conjunction with methods such as Damage Proxy Maps [Yun et al., 2012]