 |
| Titel |
Continuous monitoring of river bed load transport from the analysis of the induced seismic noise |
| VerfasserIn |
Arnaud Burtin, Laurent Bollinger, Jérôme Vergne, Rodolphe Cattin |
| Konferenz |
EGU General Assembly 2010
|
| Medientyp |
Artikel
|
| Sprache |
Englisch
|
| Digitales Dokument |
PDF |
| Erschienen |
In: GRA - Volume 12 (2010) |
| Datensatznummer |
250034194
|
|
|
|
|
|
| Zusammenfassung |
| River bed load transport, through the induced erosion, is a key-actor of the landscape
evolution. Despite this relevance, continuous measurements of bed load remain highly
difficult or even impossible to assess during extreme hydrologic conditions: the methods of
sediment sampling are generally in situ techniques, which are not adapted for large flood
events. Here we present our ongoing developments of the use of seismic sensors to monitor
continuously bed load transport within a large spatial coverage. The approach relies on the
monitoring of ground vibrations produced by the impacts of sediments on a river bed and that
propagate to seismic stations installed nearby the stream. Since these stations are
sheltered from the largest floods, the seismic monitoring offers a great potential for the
survey of bed load transport during extreme events. We explore the interest of the
spectral analysis of continuous seismic noise produced by rivers. We apply this
method along two rivers with different hydrological characteristics including water
discharge and grain size distribution: the trans-Himalayan Trisuli River and the
“torrent de St Pierre” in the French Alps. In both cases, we observe strong spatial and
temporal variations of the high-frequency (1-80 Hz) seismic energy produced along the
river. Comparisons with the regional meteorological and hydrological data along
both rivers reveal some clear correlations. Along the Trisuli River, a hysteresis
pattern between the high-frequency seismic noise level and the river water level
attests that a significant part of the observed seismic energy is caused by bed load
transport. In addition, analyses of seismic signals and their comparisons with sediment
sampling data suggest a frequency signature related to the sediment particle size. Other
methods of signal processing, as noise correlation, show our ability to locate river
segments that produce an intense sediment transport and where erosion processes are
potentially increased. With a demonstration that such a seismic monitoring is still
feasible along rivers with low water discharges and a reduced grain size distribution
(French Alps), we confirm the strong potential of seismic bed load measurements.
Further developments are still required to improve the method. Thus, calibration
experiments in natural sites and in laboratory are expected to link spectral content
and the seismic energy to a direct quantification and characterization of sediment
transport. |
|
|
|
|
|
|