T83 Motion of Landslides, Debris Flows, and Avalanches

Monday, 1 November 2010
8:00 AM - 12:00 PM
Mile High Ballroom 4AB

MOTION CHARACTERISTICS OF THE SLUMGULLION LANDSLIDE DETERMINED FROM GROUND-BASED InSAR SURVEYS

Paper number: T83-7
Program: Technical Sessions
Time: 9:20 AM – 9:35 AM

Authors

Abstract

Most landslides consist of interacting kinematic units with differing stress and strain characteristics. Landslides generally consist of an extending unit upslope, a compressing unit downslope, and a relatively neutral central unit. These units may be further subdivided due to topographic and geological complexity and other factors. Proper identification and characterization of kinematic units is critical for understanding landsliding mechanisms and for designing mitigation strategies, largely because of disparities between internal earth pressures. For example, lateral earth pressure coefficients can range from nearly zero in extending units to more than 3 in compressing units. Knowledge of the temporal relations between kinematic units may provide valuable insight into landslide mechanisms, but has only been attainable over longer durations (days to years) so provides little more than a qualitative view; data spanning minutes to hours may prove much more useful. Ground-based, interferometric, synthetic aperture radar (InSAR) can detect sub-mm line-of-sight displacement at high (seconds to minutes) frequency over large areas so can potentially reveal temporal interactions between kinematic units at a scale fine enough to highlight strain and implied stress transfers that affect landslide motion.

We performed ground-based InSAR surveys of the Slumgullion landslide in Colorado to study interactions between kinematic units and subsequent effects on landslide motion. This landslide moves reliably at rates of cm/day so provides a suitable displacement signature. Additionally, previous studies have characterized the many kinematic units comprising the landslide and documented its primary movement characteristics. Our surveys were performed at 10-minute intervals for five days during June-July 2010. About 53,000 locations were repeatedly surveyed. Concurrently, we continuously measured movement at five locations using GPS and extensometers. Preliminary results suggest episodic acceleration and deceleration at different times and locations within the landslide, suggesting temporal variability in the interactions between different kinematic units. We are hopeful that the data will provide insight into the importance of this temporal variability as a control on landslide motion.