The Use of Imagery in Laboratory Experiments - Fluvial Remote Sensing for Science and Management - page 290

13

The Use of Imagery in
Laboratory Experiments

Michal Tal 1 , Philippe Frey 2 , Wonsuck Kim 3 , Eric Lajeunesse 4 ,
Angela Limare 4 and Fran ¸oisMetivier 1
1 Aix-Marseille Universite, France
2 Irstea, Unite de Recherche ETNA, St-Martin-d'Heres, France
3 University of Texas, Austin, Texas, USA
4 Institut de Physique du Globe de Paris, Sorbonne Paris Cite, Paris, France

controlled conditions. Full control means that variables
can be held fixed and the role of individual variables can
be isolated. Experiments enable observation of long-term
behaviour that can often only be predicted or inferred in
the field. This is important because processes being stud-
ied in the field typically evolve well beyond the typical time
frame of a grant or monitoring program and insight from
long-term observation is often missed. Finally, the ability
to collect data from an experiment using several differ-
ent techniques permits researchers to draw conclusions
based on an aggregate of measurements, while multi-
ple runs allow for more rigorous statistics and greater
confidence in results.
This chapter deals with the use of imagery to acquire
data from laboratory experiments. Imagery in the labo-
ratory provides the same advantages as in the field: data
acquisition at very high temporal and spatial resolutions
in an entirely non-invasive manner. Experiments have the
intrinsic advantage offered by satellites and airplanes to
study natural systems: a view of the system through a lens
zoomed way out. Imaging techniques in the laboratory
are enhanced by the high degree of control over lighting
and materials which can both simplify image processing
and greatly enhance the results. Despite these inherent

13.1 Introduction

Experimental-based research in fluvial geomorphology
constitutes an important tool for studying processes
occurring in natural rivers. Laboratory experiments are
useful for inspiring questions, testing hypotheses, and
identifying key processes and parameters. Insights from
studies using laboratory stream-tables (relatively wide
shallow containers in which the experimental river sets
its own width and pattern) and flumes (relatively long
narrow channels with fixed parallel side walls) underlie
our understanding of processes such as bedload transport
(e.g., Gomez and Church, 1989), controls on channel
width (e.g., Ikeda et al., 1988), downstream fining (e.g.,
Seal et al., 1997), and the mechanics of flow through
bends (e.g., Hooke, 1975), and have led to important
insights about the underlying physics of large and chaotic
systems such as braided rivers and deltas (e.g., Ashmore,
1991b; Sapozhnikov and Foufoula-Georgiou, 1997, Kim
and Paola, 2007; Hoyal and Sheets, 2009).
Laboratory experiments lend themselves well to studies
that are relevant to stream restoration and management
because they provide a relatively inexpensive, fast, and
simple way to collect large quantities of data under