Multi-Year, Three-Dimensional Landslide Surface Deformation From Repeat Lidar and Response to Precipitation: Mill Gulch Earthflow, California

Published In


Document Type


Publication Date



Slow-moving landslides are often the dominant process that shapes hillslopes and delivers sediment to channels in weak lithologies. Understanding what controls their velocities is therefore essential for deciphering their role in landscape evolution and estimating hazards. In this study, we used four sequential airborne lidar data sets to derive the three-dimensional (3D), 1-m spatial resolution surface velocity field of the Mill Gulch earthflow, northern California, for three periods of time spanning a decade. Phase correlation, an image processing technique, applied to the precisely aligned lidar digital elevation models confidently resolved horizontal velocities from 0.2 to 5 m year−1. The velocity field defined three distinct kinematic zones of the landslide with different sensitivities to precipitation, such that the head moved slowly at 0.3–0.5 m year−1, the transport zone moved fastest at 2–5 m year−1, and a forked toe moved intermittently at 0–4 m year−1. Inverting the 3D surface velocity field to infer landslide thickness suggested that the head was underlain by a 6-m-deep, concave-up slip surface, while the transport zone likely had a 1–2.5-m-deep translational slip surface. We hypothesized that velocity in the head was least sensitive to changing precipitation because of its greater inferred depth, which likely dampened the amplitude of pore pressure fluctuations. A lack of major head scarp retrogressions also may have contributed to the relatively steady velocity in the head, while buttressing and debuttressing interactions between the landslide’s toes and the creek running in Mill Gulch may have contributed to the more dramatic changes in velocity in the transport zone and toes. Although predicting landslide velocity from precipitation data alone may therefore be challenging, producing detailed 3D deformation fields from repeat topographic data can be an important tool for deciphering interactions between internal controls, such as changes to landslide geometry, and external controls, such as climate, on landslide motion.


© Springer-Verlag GmbH Germany part of Springer Nature 2020



Persistent Identifier