New mobility measure and scoring system for predicting long-runout landslides, A

Abstract

Landslides are hazardous natural processes that threaten communities and infrastructure every year throughout the United States. Long-runout landslides, which are highly mobile and travel long distances from their sources, can be especially hazardous and unpredictable. The prediction of long-runout landslides has proved to be challenging because (1) it is unclear exactly what factors tend to influence long runout, and (2) the mobility measure most commonly used to measure runout (H/L) has limited physical significance. In this study, a new mobility measure, L/A1/2, is proposed and its effectiveness is evaluated. In general, this parameter provides a more meaningful assessment of landslide mobility than H/L because it describes the elongation of the deposit, rather than the overall slope gradient of the runout path. Using the new mobility measure L/A1/2, three geomorphological factors (planimetric curvature, sand content, and upslope contributing area normalized to landslide area) are identified as variables that influence landslide runout. Using these variables as input parameters, a Landslide Runout Score (LRS) system is developed and optimized to provide a method of predicting landslide runout behavior. This work is conducted in geographic information systems (GIS) using regional-scale landslide inventories and spatial data that are publicly available from government sources; therefore, the findings of this study are intended to be used with relatively coarse, regional scale data in GIS. The LRS system predicts short, medium, and long runout with accuracies of 75, 58, and 72 percent, respectively, for a combined accuracy of approximately 65 percent. The results of this work are summarized in a worksheet that can be used by geologists and engineers to develop preliminary, first-order predictions of landslide runout behavior, which can be incorporated into or used alongside regional-scale landslide hazard assessments.