Professional Documents
Culture Documents
Ecohydrol. (2012)
Published online in Wiley Online Library
(wileyonlinelibrary.com) DOI: 10.1002/eco.1301
State Key Laboratory of Urban and Regional Ecology, Research Center for Eco-Environmental Sciences (RCEES), Chinese Academy of Sciences
(CAS), Beijing, China
Department of Limnology of Shallow Lakes and Lowland Rivers, Leibniz-Institute of Freshwater Ecology and Inland Fisheries (IGB), Berlin, Germany
3
Laboratory of Hydraulic Constructions (LCH), Ecole Polytechnique Fdrale de Lausanne (EPFL), Lausanne, Switzerland
4
Instituto Superior Tcnico, Lisbon, Portugal
ABSTRACT
The role of turbulence in the dislodgment of benthic stream invertebrates from the riverbed was investigated experimentally in a
laboratory ume. For the rst time, technological advances allowed measuring the spatio-temporal patterns of turbulent
ow around two free-moving invertebrates (Aeshna cyanea and Somatochlora avomaculata). A specic methodology was
developed for the analysis of turbulence around benthic invertebrates. The results conrmed two hypotheses: (i) on the contrary
to sediment particles, invertebrates are not only sensitive to the peak values of the turbulent ow forcing but also to the temporal
uctuations in this ow forcing; and (ii) the dominant temporal uctuations are not due to local turbulent structures of the size of
the invertebrate, but to turbulent structures that scale with the ow depth and are inherited from upstream. In 15 of the 17
conducted tests, important turbulent events that scale with the ow depth accompanied by rapid temporal ow uctuations
occurred at the moment of dislodgement. The dominant forcing was consistently a threefold increase in shear stress, and was
related to a sweep event in 12 of the 17 tests. Thereby, the increase in longitudinal velocity was typically about 40%, which led to
a 100% increase in drag force in comparison with the time-averaged drag force. These results enable a new understanding of
the detailed hydraulic conditions leading to passive drift of stream invertebrates. In addition, they open new perspectives to
improve models predicting the distribution of benthic invertebrates based on hydrodynamics by accounting for turbulence.
Copyright 2012 John Wiley & Sons, Ltd.
KEY WORDS
turbulent coherent structure; invertebrate drift; invertebrate dislodgment; ume experiment; Odonata
INTRODUCTION
Benthic invertebrates are key components of aquatic
ecosystems, as they often occur in high densities and take a
central position in aquatic food webs, linking microbiota and
mesobiota with sh (Allan, 1995). Benthic invertebrates
living on the streambeds of running waters have to cope with
a relatively harsh hydraulic environment. They are exposed to
ow, which may dislodge them from the streambed and force
them to drift (Ambhl, 1959; Statzner and Holm, 1989;
Statzner, 2008; Gibbins et al., 2010). Also, higher ow
velocities may cause sediment movement and thus initiate
invertebrate drift (Gibbins et al., 2007, 2010). The ow eld
near invertebrates has been identied as a dominant factor
governing the spatial distribution of benthic invertebrates on
streambeds (Statzner and Higler, 1986; Hart et al., 1996; Hart
and Finelli, 1999; Reid and Thoms, 2008; Statzner, 2008;
Oldmeadow et al., 2010; Long et al., 2011). However, even
after more than a century of research, we do not fully
understand what the relevant hydrodynamic forces acting
on invertebrates are, which hampers the development of
process-based models describing and quantifying the
K. BLANCKAERT et al.
Flow
Sediment particle
FL
Pivotal
point
Streambed invertebrate
Flow
FL
FD
Ws
FD
Finv
Ws 0
(Eckman et al., 1990; Hart et al., 1996; Hart and Finelli, 1999;
Statzner, 2008; Reid and Thoms, 2008; Oldmeadow et al.,
2010), but remains poorly understood. This can largely be
attributed to the difculty in accurately measuring the pattern
of turbulence near the invertebrate body. The investigations
by Ambhl (1959) and Statzner and Holm (1982, 1989) were
largely motivated by the evolutionary question of how stream
invertebrates have adapted their shape in order to resist
hydrodynamic forces. These investigations mainly focused
on the local interactions between the invertebrate and the
boundary layer. Statzner (2008) noted, however, that the local
ow eld cannot fully explain the ow adaptation of
streambed invertebrates, because of the following: (i) the
local ow eld exerts different constraints during different
life stages of the stream invertebrate; and (ii) the ow eld
around the invertebrates is intrinsically so complicated and so
diverse (dependent on the invertebrate body shape and size,
Reynolds number of the ow, boundary layer conditions),
that it cannot be generalized. Similarly, Hart and Finelli
(1999) already attributed the diversity and complexity of local
ow elds to the unique arrangement of the sediment particles
on the streambed. With the observed levels of turbulence that
were greater than predicted from traditional boundary layer
theory, Hart et al. (1996) also suggested that much of the
turbulence is not produced by local shear, but inherited from
upstream. These observations suggest that the dominant
hydrodynamic force exerted on benthic invertebrates is not
necessarily determined by local ow conditions.
The previously mentioned debate on the importance of
temporal uctuations and non-locally produced turbulent
coherent structures concurs with Hart and Finelli (1999), who
noted that ow characteristics in a stream vary over a broad
range of space and time scales that can span more than
six orders of magnitude, and who postulated that the
identication of the most relevant scales is one of the central
challenges in the study of organism-ow interactions.
Owing to limitations in measuring technology, previous
investigations could only perform point-wise measurements,
which only allowed visualizing spatial patterns of timeaveraged characteristics of the ow and the turbulence, but
not of coherent turbulent structures. Moreover, the velocimeters did not allow tracking moving invertebrates. Hence,
Statzner and Holm (1982, 1989) measured the ow around
dead invertebrates glued on the streambed, whereas Hart et al.
(1996), Lancaster et al. (2006), Oldmeadow et al. (2010),
Schnauder et al. (2010) and Long et al. (2011) decoupled
invertebrate observations and ow measurements, i.e. ow
was measured at locations where invertebrates were previously observed. As a consequence, Hart and Finelli (1999),
Statzner (2008) and Lancaster and Downes (2010) argued for
the increased integration of concepts from uid mechanics
and the use of modern measurement technologies with high
spatial, and particularly high temporal velocity resolution.
The experiments presented in this paper use an acoustic
Doppler velocity proler (ADVP) that has the ability to
measure turbulent coherent structures with a high spatial and
temporal resolution as well as to track moving invertebrates.
The experiments aim to enhance insight in dominant
hydrodynamic forces acting on streambed invertebrates, and
Ecohydrol. (2012)
(1)
1
FL CL S rv2bed
2
(2)
K. BLANCKAERT et al.
(3)
(4)
(5)
(6)
(7)
10 mm
10 mm
1
0.9
0.8
0.7
0.6
0.5
0.01
0.4
0.3 1
0.2
0.1
0
3
3.5
X=3.60
Location of invertebrate introduction:
1: upstream of tributary on horizontal bed, in straight uniform flow
2: downstream of tributary on heterogeneous bed , without tributary inflow
3: downstream of tributary on heterogeneous bed , with tributary inflow
0.06
0.05
0.03
0.02
0.01
0
0.04
0
4
4.5
(b)
5
X [m]
(c)
Y [m]
(a)
5.5
100
6.5
80
60
40
S2
20
S1
S3
0
0.0
0.1
d [mm]
1.0
10.0
Figure 3. (a) Schematic representation of the laboratory ume including the main channel, the tributary channel, and the three different locations where
tests on invertebrate dislodgment were performed. The contours and colour code represent the bathymetry. (b) Spatial heterogeneity of the bed surface
material in the reach upstream of the conuence. (c) Sieving curves of the bed surface material at different locations in the ume.
4
er
eiv
Emitted pulse
Re
c
r2
ive
ce
Re
Emitter
water-filled
housing
Water
surface
i=1
Backscattered pulse
v4,i
Flow
i
Streambed
Acoustically
transparent
mylar film
z = 4[mm]
-v2,i
vi
vx,i
vz,i
Figure 4. Schematic representation of the acoustic Doppler velocity proler (ADVP). Only two of the four receivers are drawn. Modied from Blanckaert
and Lemmin (2006).
Copyright 2012 John Wiley & Sons, Ltd.
Ecohydrol. (2012)
K. BLANCKAERT et al.
!
"
velocity vector, !
v t vx ; vy ; vz t , simultaneously in an increased until it attained hydraulic conditions that trigger
entire vertical water column that ranges from the water invertebrate dislodgment (Figure 5). These critical conditions
surface to the streambed, from which all relevant hydro- for invertebrate dislodgment had been estimated in prelimdynamic characteristics can be computed [Equations (3)(7)]. inary experiments to avoid tiredness of invertebrates caused
The vertical water column situated above the invertebrate by very long exposure to ow. The ow (discharge and depth)
(Figure 4) was measured with a sampling frequency of was then maintained constant at a period that encompasses the
31!25 Hz. The measured water column was divided into bins dislodgment moment (Figure 5).
with a height of 4 mm (Figure 4). This proling capacity is a
Experiments were replicated three to six times for each
crucial methodological progress as compared with veloci- hydraulic condition with larvae of A. cyanea and once for
meters used in previous studies that only measured the each hydraulic condition with larvae of S. avomaculata,
velocity in one single point (LDA in Statzner and Holm, resulting in a total of 17 conducted tests. Such number of
1982, 1989; hot-lm in Hart et al., 1996; ADV in McReid and tests corresponds to the number of repetitions used by
Thoms, 2008; Oldmeadow et al., 2010). The proling Lancaster et al. (2006), Oldmeadow et al. (2010) and Long
capacity allows visualizing the ow patterns with high spatial et al. (2011) in laboratory experiments on invertebrate
resolution, and thus may especially show turbulent coherent hydraulics. The larvae were given at least 2 h of recovery
structures occurring in the vicinity of the invertebrate. The between two experiments.
ADVP was mounted on a carriage that allows its quasiinstantaneous displacement along the horizontal, transverse Data treatment and analysis
and vertical directions (Figure 4). This allowed the ADVP to The methodology used for data treatment and the experifollow the movement of the invertebrate on the streambed, mental analysis is illustrated by means of test 1 (cf. Table 1).
which is another important progress as compared with
velocimeters used in previous studies.
Time-averaged ow characteristics. The characteristics of
In order to accurately detect the moment of inverte- ow are commonly parameterized by the time-averaged
brate dislodgment, ADVP velocity measurements were longitudinal velocity v! , longitudinal-vertical shear stress
x
synchronized with a video acquisition of the invertebrate !
v 0x v 0z and turbulent kinetic energy, !
tke. Movement of the
movement at a frequency of 5 Hz. Hence, the moment of invertebrate required the displacement of the ADVP, which
invertebrate dislodgment could be determined with a induced perturbations in the velocity measurements
precision of (0!1 s.
(Figure 5). To obtain reliable estimates of the timeaveraged ow characteristics at invertebrate dislodgement,
Experimental design
the longest possible time window that was free of such
Three different hydraulic conditions were investigated to perturbations was determined. This determination was
cover the high variability inherent in turbulent ow based on the measured longitudinal velocity, which has
(Figure 3a). In the rst, the invertebrate was introduced been averaged over the central part of the water column
upstream of the conuence where the bed is at and the (0!25 < z/h < 0!75, where h is ow depth) to minimize the
ow quasi uniform. In the second, the invertebrate was uctuations caused by turbulence (Figure 5). This time
introduced in the conuence zone, where important gradients window free of perturbations was on the average 33!9 s long.
existed in bathymetry, streambed surface material and ow It varied considerably between experiments, with a minimum
characteristics. In the third, the invertebrate was also duration of 7!6 s, a maximum duration of 94!6 s and a
introduced in the conuence zone, but an additional discharge standard deviation of 24!5 s. We postulate that invertebrate
was provided by the tributary in order to modify local ow dislodgment is mainly determined by local ow characteristics around the invertebrate specimen. Therefore, Figure 5
characteristics.
Invertebrate specimens were introduced on the streambed also illustrates the temporal dynamics of the longitudinal
under quasi-stagnant ow conditions (Figure 5) and were velocity averaged in the zone of 01!6 cm height above the
given 5 min to become accustomed to the slowly owing bed. This vertical position has been chosen because it is near
water and ume environment. The ow was then gradually the top elevation of the invertebrate body of the two species
Longitudinal
velocity [m s-1]
0.5
0.4
0.3
0.2
0.1
0
0
10
20
1
30 Time [s] 40
2
50
3
60
4
70
5
Figure 5. Experimental design illustrated with the temporal dynamics of measured longitudinal velocity (example of test 1, see Table 1). (1) Introduction of the
invertebrate under quasi-stagnant ow conditions; (2) gradual increase of ow until conditions that trigger invertebrate dislodgment; (3) displacement of the
acoustic Doppler velocity proler to follow invertebrate movement, inducing a perturbation in the velocity signal; (4) dislodgment moment identied from
synchronized video acquisition; and (5) longest time window of constant ow conditions used to estimate the time-averaged ow characteristics.
Copyright 2012 John Wiley & Sons, Ltd.
Ecohydrol. (2012)
A. cyanea
A. cyanea
A. cyanea
A. cyanea
A. cyanea
A. cyanea
A. cyanea
A. cyanea
A. cyanea
A. cyanea
A. cyanea
A. cyanea
Somatochlora
avomaculata
S. avomaculata
5
6
10
11
12
13
14
15
1
1
28
29
27!5 + 3!3
28!6 + 3!6
28!0 + 3!4
28!5 + 3!0
30!5
30!5
31!5
31!5
30
27
27
30
30
30
30
[1 s#1]
Test
condition Discharge
0!16
0!18
0!17
0!16
0!14
0!20
0!18
0!18
0!20
0!20
0!20
0!20
0!20
0!20
0!20
0!20
0!33
0!33
0!34
0!38
0!38
0!36
0!34
0!30
0!32
0!32
0!30
0!29
0!27
0!28
0!29
0!29
0!30
U
[m s#1]
h
[m]
0!20
Depth-averaged
velocity
Flow
depth
1!2
1!2
1!2
1!3
1!3
1!5
1!5
1!5
1!5
1!5
1!5
1!0*
1!0*
1!5
1!5
1!5
0!25
0!18
0!21
0!24
0!28
0!15
0!18
0!17
0!17
0!17
0!14
0!18
0!14
0!12
0!14
0!14
0!14
vx(1!6)
[m s#1]
u*
[10-2 m s#1]
1!6
Velocity
Shear
velocity
*Unreliable value.
TCS, turbulent coherent structure; SS, shear stress; LV, longitudinal velocity; AF, accelerating ow.
17
S. avomaculata
A. cyanea
16
Aeshna cyanea
Taxa
Test
1!71
2!41
2!57
2!68
2!63
1!85
1!86
2!02
1!96
1!96
#0!75
#1!13
#0!73
#1!46
#1!21
#0!71
#1!07
#0!94
#0!95
#0!95
#1!87
#1!34
#0!49
2!00
1!41
1!90
#0!99
#1!26
#1!26
#1!29
[10#4 m2 s#-2]
Shear stress
!
v 0x v 0z 1:6
1!79
1!97
1!97
1!82
[10-3 m2 s#2]
tke(1!6)
Table I. Time-averaged and instantaneous hydraulic conditions at invertebrate location at the moment of dislodgment for the 17 tests conducted. Test condition: (1) Uniform ow upstream of
conuence, (2) downstream of conuence without tributary inow, and (3) downstream of conuence with tributary inow.
Ecohydrol. (2012)
(a)
z/h [-]
0.8
0.6
U = 0.30 m s-1
K. BLANCKAERT et al.
(c)
(b)
tke(1.6) =
1.82 x 10-3 m2 s-2
0.4
0.2
vx(1.6) =
0.14 m s-1
1.6 cm
0
0
0.1 0.2 0.3 0.4
Longitudinal velocity, vx [m s-1]
vxvz (1.6) =
-1.29 x 10-4 m2 s-2
1.6 cm
-2
-1
0
Shear stress, vxvz [m2 s-2 x 10-4]
0
0
1.6 cm
1
2
tke [m2 s-2 x 10-3]
Figure 6. Vertical proles of time-averaged ow quantities estimated from measurements in the longest time window with constant ow conditions
corresponding to dislodgment (Figure 5). (a) Longitudinal velocity; (b) longitudinal-vertical turbulent shear stress; and (c) turbulent kinetic energy.
Averaged values of the ow characteristics in the 01.6 cm height zone above the bed are indicated in boxes. Measurements close to the water surface
were not reliable because of the local ow perturbation induced by the acoustic Doppler velocity proler instrument that touches the water surface.
z [m]
z [m]
0.18 (a)
0.15
0.12
0.09
0.06
0.03
0
Q4
Q3
Q2
Q1
1.6cm
0.18 (b)
0.15
0.12
0.09
0.06
0.03
0
x 10 -4 (c)
5
0
-5
-10
-15
-20
vv
vv
[m s ]
5
0
5
10
vv
vv
0.18 (d)
0.15
0.12
0.09
0.06
0.03
0
0.12 (e)
0.08
0.04
0
-0.04
-0.08
(1.6)
[m2 s-2]
vx [m s-1]
z [m]
z [m]
0.18 (f)
0.15
0.12
0.09
0.06
0.03
0
-1
-0.8
-0.3
x 10-4
10
0.08
0.06
0.04
0.02
0
0.02
0.04
0.06
0.08
vx(1.6) [m s-1]
-2
tke' [m s ]
-0.6
-0.4
-0.2
0
0.2 0.4 0.6 0.8
1
Time in seconds before drift, t [s]
1.2
1.4
1.6
1.8
x 10-3
2
1.5
1
0.5
0
0.5
1
1.5
2
-0.2
-0.1
0
0.1
0.2
0.3
0.4
0.5
Distance in meters based on Taylors hypothesis of frozen turbulence, x = Ut [m]
Figure 7. Patterns of some turbulence quantities measured (example of test 1) during a time window that extends from 2 s before invertebrate
0
0
dislodgment to 1 s after dislodgment. The vectors in the plots (a), (b), (d) and (f) represent the longitudinal-vertical velocity uctuations v x v z, represented
with a vertical spatial resolution of 4 mm and a temporal resolution of 31!25 Hz. The colour plots represent patterns of the following: (a) A quadrant
analysis of the longitudinal-vertical velocity uctuations.
% 0 0 % The coloured zones are characterized by quasi-instantaneous shear stresses that are larger in
magnitude than the time-averaged bed shear stress, %v x v z % > u2) (Figure 8). Each colour represents one quadrant as indicated in Figure 8. (b) Deviations
0
0
v 0x v 0z [Equation (4)]. (d) Longitudinal velocity
of the quasi-instantaneous longitudinal-vertical turbulent shear stress from its time-averaged value, v x v z #!
0
uctuations, v x . (f) Deviations of the turbulent kinetic energy from its time-averaged value, tke [Equation(5)]. The line plots show the temporal
evolution at 1!6 cm above the bed of the following: (c) Deviations of the quasi-instantaneous longitudinal-vertical turbulent shear stress from its time0
0
0
v 0x v 0z . (e) Longitudinal velocity uctuations, v x . The scaling of the axis is chosen such that coherent turbulent structures are
averaged value, v x v z #!
undistorted in space. The vertical dashed line indicates the dislodgment moment identied from the video imaging. The horizontal dashed white line
delimits the zone of 1!6 cm near the bed, which is highlighted in the analysis.
Ecohydrol. (2012)
K. BLANCKAERT et al.
0.04
0.03
Q2: Ejection
Q1: Quadrant 1
vxvz = u*2
0.02
0.01
0
-0.01
-0.02
-0.03
Q3: Quadrant 3
(vxvz)Q3 : vx<0, vz<0
Q4: Sweep
(vxvz)Q4 : vx>0, vz<0
-0.04
0
0.05
0.1
0.15
-0.15 -0.1 -0.05
Longitudinal velocity fluctuations, vx [m s-1]
Figure 8. Quadrant representation of the longitudinal-vertical velocity
uctuations measured in a zone of 01!6 cm height above the ume bed
(example of test 1). Points situated outside the red hyperbolic lines
represent quasi-instantaneous shear stresses that are larger than the timeaveraged bed shear stress. Turbulence events corresponding to these points
are highlighted in Figure 7(a).
RESULTS
At the moment of dislodgement of the invertebrate specimen,
ow variables varied within only a relatively restricted range:
ow depth h varied between 0!14 and 0!20 m, discharge Q
between 27 and 32 l s#1, depth-averaged velocity U between
0!27 and 0!38 m s#1, and shear velocity u* between 0!012 and
0!016 m s#1 (Table 1). The particle Reynolds number in all
Copyright 2012 John Wiley & Sons, Ltd.
DISCUSSION
The results conrmed two hypotheses. (i) Invertebrates are
not only sensitive to the peak values of the turbulent forcing
by the ow but also to the temporal uctuations in this
turbulent ow forcing, i.e. the time required for the ow
forcing to uctuate between its average and peak values; This
constitutes a fundamental difference with respect to the
passive transport of sediment particles that is only sensitive to
the peak values. (ii) The dominant temporal uctuations are
not due to small-scale turbulent structures of the size of the
invertebrate body that are inuenced by the interaction
between the boundary layer and the invertebrate body, but to
large-scale turbulent structures that scale with the ow depth
and are inherited from upstream.
From an applied perspective, these results imply that
predictive habitat models for invertebrate distribution,
intending to reect physicalbiological coupling, should
primarily quantify and parameterize two aspects: (i) local
substrate heterogeneity and its hiding potential for streambed
invertebrates; and (ii) peaks in ow forcing and the temporal
uctuations of ow related to the dominant turbulent coherent
structures. Boyero (2003) has proposed a method to approach
the former aspect on the basis of multiple metrics able to
quantify two components of the substrate heterogeneity: the
composition and the spatial conguration of substrate
patches. The following discussion therefore focuses on the
second aspect.
The order of magnitude of the temporal uctuations
associated with the turbulent peak values can be estimated
Copyright 2012 John Wiley & Sons, Ltd.
v x v z 3u2)
U
U3
* 3u2) 3Cf
t
T
H
H
(8)
ACKNOWLEDGEMENTS
This research was sponsored by the Deutsche Forschungsgemeinschaft (DFG) and the Netherlands Organization for
Scientic Research (NWO) under grants SU 405/3-1 and
DN66-149, respectively, in the framework of their bilateral
cooperation programme. The rst author was partially
funded by the Chinese Academy of Sciences Visiting
Professorship for Senior International Scientists, grant
number 2011T2Z24, and by the Sino-Swiss Science and
Technology Cooperation for the Institutional Partnership,
grant number IP13_092911.
Ecohydrol. (2012)
K. BLANCKAERT et al.
REFERENCES
Allan JD. 1995. Stream Ecology: Structure and Function of Running
Waters. Chapman and Hall: London; 388.
Ambhl H. 1959. Die Bedeutung der Strmung als kologischer Faktor.
Schweizerische Zeitschrift fr Hydrologie 21: 133264.
Anasco NC, Koyama J, Imai S, Nakamura K. 2008. Toxicity of residual
chlorines from hypochlorite-treated seawater to marine amphipod Hyale
barbicornis and estuarine sh Oryzias javanicus. Water, Air, and Soil
Pollution 195(14): 129136.
Antonia RA, Atkinson JD. 1973. High-order moments of Reynolds shear
stress uctuations in a turbulent boundary layer. Journal of Fluid
Mechanics 58(3): 581593.
Ashworth P, Bennett S, Best JL, Lelland MC. 1996. Coherent Flow
Structures in Open Channels. Wiley: Hoboken, N. J. ISBN 978-0-47195723-2.
Batchelor GK. 1970. An Introduction to Fluid Dynamics. Cambridge
University Press: Cambridge.
Blanckaert K. 2010. Topographic steering, ow recirculation, velocity
redistribution, and bed topography in sharp meander bends. Water
Resources Research 46: W09506.
Blanckaert K, de Vriend HJ. 2004. Secondary ow in sharp open-channel
bends. Journal of Fluid Mechanics 498: 353380.
Blanckaert K, Graf WH. 2001. Experiments on ow in an open-channel
bend. Mean ow and turbulence. Journal of Hydraulic Engineering,
ASCE, 12710: 835847.
Blanckaert K, Lemmin U. 2006. Means of noise reduction in acoustic
turbulence measurements. Journal of Hydraulic Research, IAHR, 441:
317.
Boyero L. 2003. The quantication of local substrate heterogeneity in
streams and its signicance for macroinvertebrate assemblages.
Hydrobiologia 499: 161168.
Bratrich C, Jorde K. 1997. Hydraulische und morphologische Modellierung von Fliessgewssern mit dem Simulationsmodell CASIMIR:
Gewsserbiologie und Habitatmodellierung. Wasserwirtschaft 78:
370371.
Cellio M, Lemmin U. 2004. Inuence of coherent ow structures on the
dynamics of suspended sediment transport in open-channel ow. Journal
of Hydraulic Engineering 13011: 10771088. DOI: 10.1061/ASCE073394292004130:111077.
Chang WY, Constantinescu G, Tsai WF, Lien HC. 2012. Coherent
structures dynamics and sediment erosion mechanisms around an instream rectangular cylinder at low and moderate angles of attack. Water
Resources Research DOI: 10.1029/2011WR010586, in press.
Constantinescu G, Koken M, Zeng J. 2011. The structure of turbulent ow
in an open channel bend of strong curvature with deformed bed: insight
provided by detached eddy simulation. Water Resources Research 47:
W05515. DOI: 10.1029/2010WR010114.
Dwivedi A, Melvill BW, Shamseldin AY, Guga TK. 2011. Flow structures
and hydrodynamic force during sediment entrainment. Water Resources
Research 47: W01509. DOI: 10.1029/2010WR009089
Eckman JE, Savidge WB, Gross TF. 1990. Relationship between duration
of cyprid attachment and drag forces associated with detachment of
Balanus amphitrite cyprids. Marine Biology 107: 111118.
Einstein HA, Li H. 1958. The viscous sublayer along a smooth boundary.
Transactions, ASCE, 123, Paper No. 2992, pp. 293313.
Escauriaza C, Sotiropoulos F. 2011. Lagrangian model of bed-load
transport in turbulent 799 junction ows. Journal of Fluid Mechanics
666: 3676.
Gibbins C, Vericat D, Batalla RJ. 2007. When is stream invertebrate
drift catastrophic? The role of hydraulics and sediment transport
in initiating drift during ood events. Freshwater Biology 52:
23692384.
Gibbins C, Batalla RJ, Vericat D. 2010. Invertebrate drift and benthic
exhaustion during disturbance: response of mayies (Ephemeroptera) to
increasing shear stress and river-bed instability. River Research and
Applications 26: 499511.
Grass AJ. 1971. Structural features of turbulent ow over smooth and
rough boundaries. Journal of Fluid Mechanics 50: 233.
Hart DD, Finelli CM. 1999. Physical-biological coupling in streams: the
pervasive effects of ow on benthic organisms. Annual Review of
Ecological Systems 30: 363395.
Hart DD, Clark BD, Jasentuliyana A. 1996. Fine-scale eld measurement
of benthic ow environments inhabited by stream invertebrates.
Limnology and Oceanography 412: 297308.
Hinze JO. 1959. Turbulence. McGraw-Hill: New York.
Hurther D, Lemmin U. 1998. A constant beam-width transducer for threedimensional Doppler prole measurements in open channel ow.
Measurement Science & Technology, IOP, 910: 17061714.
Copyright 2012 John Wiley & Sons, Ltd.
Ecohydrol. (2012)