Landslide kinematics and their potential controls from hourly to decadal timescales: Insights from integrating ground-based InSAR measurements with structural maps and long-term monitoring data

Knowledge of kinematics is rudimentary for understanding landslide controls and is increasingly valuable with greater spatiotemporal coverage. However, characterizing landslide-wide kinematics is rare, especially at broadly ranging timescales. We used highly detailed kinematic data obtained using photogrammetry and field mapping during the 1980s and 1990s and our 4.3-day ground-based InSAR survey during 2010 to study kinematics of the large, persistently moving Slumgullion landslide. The landslide was segregated into 11 kinematic elements using the 1980s-1990s data and the InSAR survey revealed most of these elements within a few hours. Averages of InSAR-derived displacement point measures within each element agreed well with higher quality in situ observations; averaging was deemed necessary because adverse look angles for the radar coupled with tree cover on the landslide introduced error in the InSAR results. We found that the landslide moved during 2010 at about half its 1985-1990 speed, but slowing was most pronounced at the landslide head. Gradually decreased precipitation and increased temperature between the periods likely resulted in lower groundwater levels and consequent slowing of the landslide. We used GPS survey results and limit-equilibrium modeling to analyze changing stability of the landslide head from observed thinning and found that its stability increased between the two periods, which would result in its slowing, and the consequent slowing of the entire landslide. Additionally, InSAR results suggested movement of kinematic element boundaries in the head region and our field mapping verified that they moved and changed character, likely because of the long-term increasing head stability. On an hourly basis, InSAR results were near error bounds but suggested landslide acceleration in response to seemingly negligible rainfall. Pore-pressure diffusion modeling suggested that rainfall infiltration affected frictional strength only to shallow depths along the landslide's marginal faults, highlighting their importance in controlling landslide stability. Hourly results also suggested that motion propagated along the 3.9-km length of the active landslide, even following sub-millimeter displacements, while strengthening of landslide shear boundaries during faster movement was likely critical in regulating the landslide's motion. Hence, detailed kinematic characterizations obtained from traditional and emerging approaches helped to reveal that mechanisms controlling landslide movement and evolution over decades also are critical to sub-millimeter movement on a nearly continuous basis.