Gravel and cobble transport by rivers governs channel change, and is therefore of great importance to geomorphologists and river engineers. The most commonly used approaches to estimating bedload transport rates are empirically calibrated relationships based on flume experiments. However, it is difficult to assess how effectively such relationships predict transport rates during extreme events in large rivers because quantitative measures of bedload transport are labor intensive and, at high flows, often dangerous to obtain. For logistical reasons, most bedload studies have been carried out in small mountain streams. Particle impacts on the river bed transfer momentum, which in turn generates elastic (seismic) waves. Therefore, seismology (the study of elastic waves in the Earth) can potentially constrain bedload transport rates in rivers. Taking advantage of a well-instrumented catchment with independent constraints on when and at what rate sediment moves in the channel, we will deploy an array of seismometers to test theoretical predictions for the generation of seismic waves from bedload sediment impacts with the bed during transport.

Understanding more clearly when and at what rate coarse sediment moves in river channels is essential for planning infrastructure in and adjacent to rivers (e.g., bridges, dams, roads). Approaches to predicting coarse sediment transport are typically developed in controlled settings and can be inaccurate in large floods, when it is most critical to be able to predict sediment movement. Using a novel application of an existing science (seismology), we are developing an approach to measuring bedload transport in large rivers without requiring the deployment of any instrumentation in the river. This work has the potential to greatly improve our capacity to cheaply, safely, and efficiently monitor sediment movement in rivers while simultaneously advancing understanding of the processes influencing bedload transport rates in rivers. Additionally, this project will have an outreach component in which an undergraduate geology class will conduct a sediment transport monitoring exercise using the approach developed here.

Project Report

The purpose of our research has been to better understand how and measure how much sediment moves in upland rivers. Intellectual merit: The academic impact is to deepen our understanding of the dynamics of mountain landscapes. Broader impacts: The societal impact of this work is to build towards tools for monitoring how much coarse sediment and debris is transported during dangerous floods in mountain environments. Our work began with the deployment of an array of seismometers in one of the most geomorphologically dynamic environments in the world, in a national park in Taiwan. We placed sensitive instruments close to a river channel about to be disturbed by removal of a check dam and the release of all the sediment trapped behind it. Using monitoring data gathered over several years, we were able to study how individual floods moved the coarse sediment (sand, gravel and cobbles) down the river, generating weak but measurable seismic signals. Analysis of these signals confirmed our suspicion that seismic amplitude is proportional to the amount of bed sediment transport, and that this is the result of the changing frequency of impacts of sediment grains on bedrock. We were able to show that much more sediment is moved during the rising stage of the flood compared to the later waning stage, in a phenomenon called hysteresis. By quantifying this hysteresis we were able to show that a wave of sediment, triggered by dam removal, can be tracked as it travels downstream. The ways in which sediment grains are transported along a mountain river bed, strike bedrock, and generate measureable seismic waves, is complex. Part of the complexity lies in the shape of the bedrock bed of such channels: erosion of the bed carves out bedforms that make the bed rough, help to store sediment in patches, and modify the flow. We carried out field and theoretical studies to help understand these phenomena, with the ultimate goals of understanding how the seismicity comes about and what it can tell us about erosion. The field aspect of this work was a detailed study of the morphology of a mixed bedrock alluvial channel in Japan - the Shimanto River - which exhibits a range of forms from pure bedrock to entirely alluviated. We developed a new technique for the mapping of the three-dimensional topography of the channel bedrock using structure-from-motion and multi-view stereo algorithms - this method allows us to build 3d maps at close range (meters) and at centimeter resolution and low cost. Our main result, aside from establishing a how-to for this exciting new field technique, is a large dataset of bedrock channel morpholgy that can be used in future studies. The theoretical aspect of this work was the development of micro-scale model of how sediment grains move along a river bed and either strike bedrock (generating seismicity and causing erosion) or bed sediment (causing neither). We had the novel idea of using queueing theory to treat this problem mathematically. Queueing theory (QT) was developed by researchers in the field of operations research, and was once used by dam engineers and hydrologists to help design dams, but it has never before been used in this field of geoscience. We have been able to show how QT simplifies the mathematics of the sediment transport problem, and how it can be used to study the details of fluctuating sediment motions and storage in a relatively simple fashion. As such, our QT work provides a useful theoretical tool for future studies of sediment transport in mountain rivers.

Agency
National Science Foundation (NSF)
Institute
Division of Earth Sciences (EAR)
Type
Standard Grant (Standard)
Application #
1148176
Program Officer
Paul Cutler
Project Start
Project End
Budget Start
2012-07-01
Budget End
2014-06-30
Support Year
Fiscal Year
2011
Total Cost
$63,859
Indirect Cost
Name
Columbia University
Department
Type
DUNS #
City
New York
State
NY
Country
United States
Zip Code
10027