An empirical equation to determine the threshold for rainfall-induced landslides developing to debris flows

Abstract

This study derives an empirical equation to determine the rainfall threshold that triggers landslides and subsequent debris flows. In this research, the contribution of different factors, especially various sediment layer thicknesses, was assessed on the landslide leading to debris flows. Bed slope, sediment thickness, sediment mean diameter, sediment layer length, rainfall intensity, and time of landslide occurrence were selected as the effective factors. Rainfall simulator device was used to create rainfall on the sediment sample. The effect of these factors on the landslide which leads to debris flow occurrence was investigated using dimensional analysis of experimental data. Then, linear and power regression models were tested by 30% of the total experimental data to estimate the rainfall-induced landslide. Determination coefficient (R2), mean absolute relative error (MARE), and Akaike information criterion (AIC) were applied to determine the best equation. Results revealed that landslide often occurs at a rapid rate in the thicker deposits than the thinner one. The evaluation of linear and power equations demonstrated that two-variable power model was the best empirical equation for estimation of rainfall threshold in initiation of landslide and debris flows. The evaluation criteria including MARE, R2, and AIC were estimated 0.044, 0.984, and − 25.93, respectively. In this equation, bed slope and the ratio of the time of landslide occurrence to the sediment layer length were used as the effective factors. This study can be used for assessment of the threshold of rainfall-induced landslide which finally leads to debris flow hazard.