ALBERT

Description: On 25 May 2014, a rain-on-snow–induced rock avalanche occurred in the West Salt Creek valley on the northern flank of Grand Mesa in western Colorado (United States). The avalanche mobilized from a preexisting rock slide in the Green River Formation and traveled 4.6 km down the confined valley, killing three people. The avalanche was rare for the contiguous United States because of its large size (54.5 Mm 3 ) and high mobility (height/length = 0.14). To understand the avalanche failure sequence, mechanisms, and mobility, we conducted a forensic analysis using large-scale (1:1000) structural mapping and seismic data. We used high-resolution, unmanned aircraft system imagery as a base for field mapping, and analyzed seismic data from 22 broadband stations (distances 〈656 km from the rock-slide source area) and one short-period network. We inverted broadband data to derive a time series of forces that the avalanche exerted on the earth and tracked these forces using curves in the avalanche path. Our results revealed that the rock avalanche was a cascade of landslide events, rather than a single massive failure. The sequence began with an early morning landslide/debris flow that started ~10 h before the main avalanche. The main avalanche lasted ~3.5 min and traveled at average velocities ranging from 15 to 36 m/s. For at least two hours after the avalanche ceased movement, a central, hummock-rich core continued to move slowly. Since 25 May 2014, numerous shallow landslides, rock slides, and rock falls have created new structures and modified avalanche topography. Mobility of the main avalanche and central core was likely enhanced by valley floor material that liquefied from undrained loading by the overriding avalanche. Although the base was likely at least partially liquefied, our mapping indicates that the overriding avalanche internally deformed predominantly by sliding along discrete shear surfaces in material that was nearly dry and had substantial frictional strength. These results indicate that the West Salt Creek avalanche, and probably other long-traveled avalanches, could be modeled as two layers: a thin, liquefied basal layer, and a thicker and stronger overriding layer.

Description: We explored regional influences on debris-flow initiation throughout the Colorado Front Range (Colorado, USA) by exploiting a unique data set of more than 1100 debris flows that initiated during a 5 day rainstorm in 2013. Using geospatial data, we examined the influence of rain, hillslope angle, hillslope aspect, and vegetation density on debris-flow initiation. In particular we used a greenness index to differentiate areas of high tree density from grass and bare soil. The data demonstrated an overwhelming propensity for debris-flow initiation on south-facing hillslopes. However, when the debris-flow density was analyzed with respect to total rainfall and greenness we found that most debris flows occurred in areas of high rainfall and low tree density, regardless of hillslope aspect. These results indicate that present-day tree density exerts a stronger influence on debris-flow initiation locations than aspect-driven variations in soil and bedrock properties that developed over longer time scales.

Description: We explored regional influences on debris-flow initiation throughout the Colorado Front Range (Colorado, USA) by exploiting a unique data set of more than 1100 debris flows that initiated during a 5 day rainstorm in 2013. Using geospatial data, we examined the influence of rain, hillslope angle, hillslope aspect, and vegetation density on debris-flow initiation. In particular we used a greenness index to differentiate areas of high tree density from grass and bare soil. The data demonstrated an overwhelming propensity for debris-flow initiation on south-facing hillslopes. However, when the debris-flow density was analyzed with respect to total rainfall and greenness we found that most debris flows occurred in areas of high rainfall and low tree density, regardless of hillslope aspect. These results indicate that present-day tree density exerts a stronger influence on debris-flow initiation locations than aspect-driven variations in soil and bedrock properties that developed over longer time scales.

Description: Debris flows can travel great distances from their sources and pose hazards to communities downstream. Identification of soils susceptible to liquefaction during slope failure could aid in locating potential debris-flow sources. We used a modified version of the slump test, typically used to measure the consistency of wet concrete mix, to test the proclivity of soil to flow. The method can be used to identify a threshold density below which a soil will flow under field-saturated conditions. We performed 51 soil mobility tests on 13 sandy and gravelly soils with 0%–16% silt and clay (fines) and one sample with 72% fines collected from landslide and nonlandslide sites. Results are consistent with corresponding site porosity and failure mode for coarse-grained soils. Threshold porosity for flow ranged from 0.39 to 0.65. Highest failure velocities were associated with the loosest soils and soils with the lowest permeability. Failure velocities and modes observed in our tests were consistent with test results for identical soils in a large-scale flume on thicker samples (0.65 m) that approach average failure depths measured in the field (0.75 m). This novel test was successfully designed as a rapidly conducted (~24 h to prepare and test samples, including overnight saturation), low-cost alternative to large-diameter triaxial tests capable of testing gravelly soils that does not require a predefined shear rate, allowing pore pressure to evolve as soil fails under self-weight and drained conditions.

Description: Distributed models to forecast the spatial and temporal occurrence of rainfall-induced shallow landslides are based on deterministic laws. These models extend spatially the static stability models adopted in geotechnical engineering, and adopt an infinite-slope geometry to balance the resisting and the driving forces acting on the sliding mass. An infiltration model is used to determine how rainfall changes pore-water conditions, modulating the local stability/instability conditions. A problem with the operation of the existing models lays in the difficulty in obtaining accurate values for the several variables that describe the material properties of the slopes. The problem is particularly severe when the models are applied over large areas, for which sufficient information on the geotechnical and hydrological conditions of the slopes is not generally available. To help solve the problem, we propose a probabilistic Monte Carlo approach to the distributed modeling of rainfall-induced shallow landslides. For this purpose, we have modified the transient rainfall infiltration and grid-based regional slope-stability analysis (TRIGRS) code. The new code (TRIGRS-P) adopts a probabilistic approach to compute, on a cell-by-cell basis, transient pore-pressure changes and related changes in the factor of safety due to rainfall infiltration. Infiltration is modeled using analytical solutions of partial differential equations describing one-dimensional vertical flow in isotropic, homogeneous materials. Both saturated and unsaturated soil conditions can be considered. TRIGRS-P copes with the natural variability inherent to the mechanical and hydrological properties of the slope materials by allowing values of the TRIGRS model input parameters to be sampled randomly from a given probability distribution. The range of variation and the mean value of the parameters can be determined by the usual methods used for preparing the TRIGRS input parameters. The outputs of several model runs obtained varying the input parameters are analyzed statistically, and compared to the original (deterministic) model output. The comparison suggests an improvement of the predictive power of the model of about 10% and 16% in two small test areas, that is, the Frontignano (Italy) and the Mukilteo (USA) areas. We discuss the computational requirements of TRIGRS-P to determine the potential use of the numerical model to forecast the spatial and temporal occurrence of rainfall-induced shallow landslides in very large areas, extending for several hundreds or thousands of square kilometers. Parallel execution of the code using a simple process distribution and the message passing interface (MPI) on multi-processor machines was successful, opening the possibly of testing the use of TRIGRS-P for the operational forecasting of rainfall-induced shallow landslides over large regions.

Description: Distributed models to forecast the spatial and temporal occurrence of rainfall-induced shallow landslides are deterministic. These models extend spatially the static stability models adopted in geotechnical engineering and adopt an infinite-slope geometry to balance the resisting and the driving forces acting on the sliding mass. An infiltration model is used to determine how rainfall changes pore-water conditions, modulating the local stability/instability conditions. A problem with the existing models is the difficulty in obtaining accurate values for the several variables that describe the material properties of the slopes. The problem is particularly severe when the models are applied over large areas, for which sufficient information on the geotechnical and hydrological conditions of the slopes is not generally available. To help solve the problem, we propose a probabilistic Monte Carlo approach to the distributed modeling of shallow rainfall-induced landslides. For the purpose, we have modified the Transient Rainfall Infiltration and Grid-Based Regional Slope-Stability Analysis (TRIGRS) code. The new code (TRIGRS-P) adopts a stochastic approach to compute, on a cell-by-cell basis, transient pore-pressure changes and related changes in the factor of safety due to rainfall infiltration. Infiltration is modeled using analytical solutions of partial differential equations describing one-dimensional vertical flow in isotropic, homogeneous materials. Both saturated and unsaturated soil conditions can be considered. TRIGRS-P copes with the natural variability inherent to the mechanical and hydrological properties of the slope materials by allowing values of the TRIGRS model input parameters to be sampled randomly from a given probability distribution. The range of variation and the mean value of the parameters can be determined by the usual methods used for preparing the TRIGRS input parameters. The outputs of several model runs obtained varying the input parameters are analyzed statistically, and compared to the original (deterministic) model output. The comparison suggests an improvement of the predictive power of the model of about 10% and 16% in two small test areas, i.e. the Frontignano (Italy) and the Mukilteo (USA) areas, respectively. We discuss the computational requirements of TRIGRS-P to determine the potential use of the numerical model to forecast the spatial and temporal occurrence of rainfall-induced shallow landslides in very large areas, extending for several hundreds or thousands of square kilometers. Parallel execution of the code using a simple process distribution and the Message Passing Interface (MPI) on multi-processor machines was successful, opening the possibly of testing the use of TRIGRS-P for the operational forecasting of rainfall-induced shallow landslides over large regions.