Quantifying Seismic Properties of a River Channel at Mount Rainier for Use in Debris Flow Monitoring and Analysis

Abstract Theoretical models that relate debris flow properties to their seismic signature suggest seismic methods may be used to remotely characterize properties of these events. However, the complexity of debris flow sources and poorly constrained material properties in near-surface environments limit our ability to determine important attributes of debris flows, such as volume and particle size distribution, from seismic records alone. In this study, we explore the sensitivity of debris flow seismic signals to subsurface seismic characteristics using an established debris flow seismicity model and subsurface properties derived from active-source measurements in the Tahoma Creek stream channel at Mount Rainier, United States. Using refraction and multichannel analysis of surface waves, we estimate 1D primary and secondary wave velocity profiles to a depth of 11 m and calculate the frequency-varying phase velocity (vc) and Rayleigh-wave quality factor (QR) for frequencies between 9 and 50 Hz. We find that the Tahoma Creek stream channel has low vc, varying with frequency between 226 and 434 m/s, and is highly attenuating, with QR estimated below 13.3 at all analyzed frequencies. We model the seismic signal of a hypothetical debris flow in Tahoma Creek using our measured values and compare the results against the same model but varying vc and QR over a range of frequency-independent, single values commonly used in the literature when measurements are not available (vc=250–750 m/s, QR=3–33). We find that the power spectral densities of the modeled debris flows vary by orders of magnitude within the subset of our test values, highlighting the benefits of measuring material properties when using modeled debris flow seismic signals for quantitative monitoring.