HIGH-RESOLUTION TOPOGRAPHIC MAPPING OF ACTIVE FAULTS IN SOUTHERN CALIFORNIA WITH SATELLITE OPTICAL IMAGERY

Digital elevation models (DEMs) play a range of important roles in the study of active and Quaternary faults. These topographic models provide a core input for mapping active structures, queuing and generating digitized maps, characterizing local and regional tectonic geomorphology, and quantifying co-seismic displacements. Traditionally, local and regional DEMs are derived from three sources: digitization of topographic maps, space-borne radar systems (i.e. the Shuttle Radar Topography Mission, SRTM), and airborne LiDAR surveys (i.e. the B4 LiDARproject). Topography models derived from these sources suffer from several shortcomings that limit their utility in the study of active landscapes: spatial resolution (~30 m ground resolution for radar-derived DEMs), spatial coverage (highly limited with LiDAR), and repeatability following an earthquake (space-borne surveys are often not repeated while LiDAR acquisitions may be cost prohibitive).

Here, we present preliminary high-resolutions DEMs (0.3-3 m per pixel) generated from satellite stereo electro-optical imagery from the WorldView constellation of satellites. We generate DEMs over the active faults of southern California in an effort to supplement current high-resolution topography data sets from LiDAR, expand the spatial coverage of high-resolution DEMs where LiDAR observations are currently unavailable, and develop workflows for the rapid generation of post-earthquake DEMs in anticipation of geodetic analysis of forthcoming surface rupturing earthquakes. The DEMs are generated through NASA’s Ames Stereo Pipeline algorithm and SETSM. Our preliminary work focuses on the Ventura fault and the San Gorgonio Pass of the San Andreas fault – regions identified as special fault study zones by the Southern California Earthquake Center. We also present preliminary quality assessments of the generated DEMs through comparisons to the 30 m SRTM DEM and previously acquired LiDAR swaths that intersect our imaging regions. Initial results indicate that we are able to generate DEMs of exceptional quality and spatial resolution from the WorldView imagery, particularly in urban and vegetation-free regions.