Looking Ahead by Looking Down – Potential Applications of Very High Resolution Drone-Based Imagery for Tectonic Geomorphology

Allen M. Gontz, Chelsea M. Blanton, Thomas K. Rockwell, & Joshua T. Kelly

Submitted August 15, 2018, SCEC Contribution #8467, 2018 SCEC Annual Meeting Poster #278

Geomorphic analyses have relied on various methods throughout the recent past. In the 1940’s, aerial imagery changed the way we look at the surface of the Earth. The 1980’s brought Landsat and new capacity to image the surface of the Earth. In the past few years, Planet Lab made daily 3 meter resolution satellite imagery available. Arguably, the most impactful tool in recent years is the explosion of small, easy to fly and cheap remotely controlled aerial systems… also known as DRONES. Today’s low-cost drones feature high-resolution cameras for still and video, GPS positioning, automated flight path planning and the ability to change cameras. We used a small commercially available drone, the DJI Phantom 4 Professional, to image sections of the southern San Andreas Fault. Flying at altitudes below 60 m and often as low as 25 m, we were able to acquire base imagery with a ground sample distance in the sub-centimeter range. The images were analyzed using structure from motion for creation of DEMs as well as orthophoto mosaics for sections along the southernmost San Andreas fault. Effective resolution of the data products is in the range of 1-10 cm, depending on the processing algorithm selected and altitude of imagery acquisition. Once a DEM is created, additional products, including hillshade models, surface curvature and slope-aspect maps can be created to facilitate identification of landscape features at various scales. One of the greatest potentials for very high-resolution drone-based imagery products is the speed at which data can be acquired. This approach facilitates rapid response to isolated events, such as earthquakes, and development of baselines for time series analyses. Time series data with centimeter-scale resolution has the potential to constrain slow-moving processes, such as creep, over reasonable time scales such as five years. We present examples of data products created from the imagery and use feature identification techniques that highlight offset geomorphic features that can be tied back to the earthquake record for this important fault.

Key Words
UAV, SfM, DEM, GIS, Remote Sensing, SAFS

Citation
Gontz, A. M., Blanton, C. M., Rockwell, T. K., & Kelly, J. T. (2018, 08). Looking Ahead by Looking Down – Potential Applications of Very High Resolution Drone-Based Imagery for Tectonic Geomorphology. Poster Presentation at 2018 SCEC Annual Meeting.


Related Projects & Working Groups
San Andreas Fault System (SAFS)