Paper No. 345-9
Presentation Time: 3:35 PM
LANDSLIDE, EARTHQUAKE AND CRYOSPHERE STUDIES USING HIGH RESOLUTION DIGITAL SURFACE MODELS FROM COMMERCIAL IMAGERY AND DRONES (Invited Presentation)
WILLIS, Michael J.1, HIGMAN, Bretwood2, SHUGAR, Dan H.3, LYNETT, Patrick4, BARBA, Magali1, TIAMPO, Kristy5, CLARK, Marin K.6, ZHENG, W.7, PRITCHARD, M.E.7, RAMAGE, Joan M.8, STEARNS, Leigh A.9 and MORIN, Paul10, (1)University of Colorado Boulder, Cooperative Institute for Research in Environmental Sciences (CIRES), 216 UCB, Boulder, CO 80309, (2)Ground Truth Trekking, Seldovia, AK 99663, (3)University of Washington Tacoma, Tacoma, WA 98402, (4)University of Southern California, Los Angeles, CA 90089-2531, (5)Department of Geological Sciences and the Cooperative Institute for Research in Environmental Sciences (CIRES), University of Colorado, Boulder, CO 80309, (6)Earth and Environmental Sciences, University of Michigan, 2534 C C Little Bldg, 1100 N University Ave, Ann Arbor, MI 48109-1005, (7)Cornell University, Ithaca, NY 14850, (8)Earth and Environmental Sciences, Lehigh Univ, 31 Williams Hall, Bethlehem, PA 18015, (9)Department of Geology, University of Kansas, Lawrence, KS 66045, (10)Polar Geospatial Center, University of Minnesota, Pillsbury Hall, 310 Pillsbury Drive SE, Minneapolis, MN 55455, Mike.Willis@Colorado.Edu
We use very high resolution digital surface models derived from along track stereo satellite commercial imagery and from airborne structure from motion to examine landslides, earthquake fault motion and glacier disintegration. We combine open source software and NSF/NGA/Digital Globe imagery to rapidly respond to landslides such as the one that generated the 2015 Taan Fjord, Alaska Tsunami and the 2017 Karrat Fjord, Greenland Tsunami. We show the value of rapidly combining imagery and numerical models with DSMs to provide estimates of tsunami run up, landslide volume and landslide dynamics. The rapid response to the Greenland event has been combined with Synthetic Aperture Radar studies to provide local authorities with estimates of future inundation risk and regional slope stability. In both studies, volumetric calculations from repeat DSMs match, or refine those made from seismic estimates.
The November 2016 Kaikoura, NZ Earthquake provides examples of three dimensional displacements from DSM differencing and cross-correlation, that are compared to ultra-high resolution DSMs produced using Structure from Motion from low flying drone imagery. We push the limits of the SETSM software to produce repeat 0.5 m posting DSM models around fault ruptures in the rugged, forested areas of the eastern South Island, NZ. Horizontal fault motions from pixel tracking on half meter imagery allows near fault motion and shallow rupture dynamics to be more accurately modelled.
Finally, we use our workflow, which is similar to that of the ArcticDEM project, to examine the rapid collapse of a cold-based ice cap in the Russian High Arctic. The bed of the ice cap switches from a high-resistance frozen bed to a very low friction sliding bed within the interval of one to two years. Speeds along the glacier accelerate from 20 m per year to 25 m per day, with thinning rates of over 0.30 m/day observed. Our precise altimetry and speed measurements combined with passive microwave observations suggest that this surge like event is initiated at the ice front and is driven by continuity, drawing down the interior of the ice cap, where the ice-bedrock interface is above sea level, into the Kara Sea. This is an unexpected situation, as cold-based ice bodies above sea level are generally thought to be stable.
![[Visit Client Website]](/img/gsa/banner.jpg)