Assessing the seismic signature of turbulent flow and intense bedload transport from designed laboratory experiments

Abstract

The field of fluvial seismology has undergone significant advances over the past decade. The development of dedicated physical theories and their applications in various contexts have allowed separating the respective contributions of turbulent flow and bedload transport, such that physical parameters like flow depth and sediment flux may be inferred from seismic observations. However, the quantitative link between signal characteristics (amplitude, frequency) and the underlying physics yet involves simplified considerations that do not necessarily apply to more complex situations, such as for example under rough flow conditions or during extreme floods. In this talk I will present results from laboratory experiments that we designed specifically in order to quantify the seismic signature of flow turbulence and intense bedload transport under a range of conditions using force sensors coupled to the river bed. On one hand, I will show that existing theory regarding turbulent flow properly captures the main characteristics of the seismic source, but that additional dependencies on flow conditions and particle-wake development need to be included for more accurate predictions. On the other hand, I will show that existing theory regarding bedload transport fails at capturing the main characteristics of the seismic source under intense bedload transport conditions associated with complex changes in internal flow dynamics. In this case the seismic source appears to be a decreased function of solid concentration, as opposed to an increased function such as considered in current theories, which we suggest is due to grain impacts being agitation-controlled rather than bed-roughness controlled. Finally, I will discuss possible ways towards building more generic theories of ground motion induced by sediment transport.