Document Actions

IJEGE-11_BS-Liu-et-alii

background image
Italian Journal of Engineering Geology and Environment - Book www.ijege.uniroma1.it © 2011 Casa Editrice Università La Sapienza
231
DOI: 10.4408/IJEGE.2011-03.B-027
INFLUENCE OF DEBRIS FLOW ON CHANNEL EVOLUTION
J
ian
k
anG
LIU
(*, **, ***)
, Z
un
L
an
CHENG
(*, **)
, X
iao
G
anG
ZHANG
(*, **)
& S
i
CHenG
(****)
(*)
Key Lab of Mountain Hazards and Surface Processes, Chinese Academy of Sciences, Chengdu 610041, China
(**)
Institute of Mountain Hazards and Environment, Chinese Academy of Sciences, Chengdu 610041, China
(***)
Graduated University of Chinese Academy of Science, Beijing 100049, China
(****)
China University of Geosciences(Beijing), Beijing 100083, China
charge and low sediment concentration, incision plays
a more important role in mountainous rivers, while it
makes few changes for channel evolution in a short
period. However, landslides, debris flows and col-
lapses may produce significant impact on river chan-
nel. Channel segments in the dammed-lake reach will
experience sediment deposition in the upper and ero-
sion in the lower portion of the reach, and bed gradient
is also undergoing continuous adjustment (C
osta
&
s
CHusteR
, 1991; s
CHusteR
, 2000; k
oRuP
, 2002; 2005;
H
su
& H
su
, 2009; d
anG
et alii, 2009).
Debris flows have great impact on channel change
and always induce rapids and pools. The field investiga-
tion in Colorado Canyon by d
olan
et alii (1978) points
out that, steep tributaries flowing within the zones of
bedrock weakness move large debris to the Colorado,
forming the major rapids along the 450-kilometer river
in the Grand Canyon, and accelerated flow through
the rapids scours the deep pools that are located below
them. According to Zhong (1999), there are more than
400 rapids along Jinsha River from Jinshajie to Xinshi
town, of which 85 were formed by debris flows. l
ianG
et alii (2001a; 2001b) conducted a lot of research on
the impact of debris flow on river channel change. H
e
(2003) combined the data of Xiaojiang River in Yunnan
with model experiment results, concluding impact of
debris flow on channel change from 4 aspects. Studies
mentioned above have concluded what changes occur
when the mainstream is injected by debris, but without
explain of how the impacts work.
ABSTRACT
Affected by the active plate motion, the upper Min-
jiang River develops more than 21 debris flow valleys.
After the Wenchuan earthquake on May 12, 2008, the
geohazards occur more frequently and seriously be-
cause of the increased debris source, and 9 debris flows
blocked the mainstream. Based on the characteristics of
debris-flow dams, model tests have been accomplished.
Given the junction angle of 90°, influence of debris flow
on channel evolution is discussed for different flow den-
sity, and discharge ratio and momentum ratio between
debris flow and the mainstream water flow. The results
show that there is a linear relationship between momen-
tum ratio R
M
(ratio between momentum for flow in trib-
utary and mainstream) and shrink ratio S (index defined
to describe the magnitude of channel shrink after inflow
of debris) of mainstream, and also an exponential rela-
tionship between the ultimate average width B
m
(index
defined to describe the average width of river channel
after dam failure) of mainstream and the coefficient of
velocity variation F
v
(ratio between velocity for flow in
mainstream after dam failure and before its formation).
K
ey
words
: debris flow, channel evolution, model test,
earthquake
INTRODUCTION
Channel evolution depends on interaction be-
tween water flow and river channel, including incision
and deposition. Because of steep gradient, high dis-
background image
J.k. LIU, Z.L. CHENG, X.G. ZHANG & S. CHENG
232
5th International Conference on Debris-Flow Hazards Mitigation: Mechanics, Prediction and Assessment Padua, Italy - 14-17 June 2011
MODELING EXPERIMENTS
Debris flow which blocked Minjiang River several
times after Wenchuan earthquake in Mozigou is the pro-
totype for this simple model (Fig.1, Table 1). Confined to
the site condition, geometrical scale is defined as 1:250
in the experiment. According to the similarity criterion,
scales for other parameters can be obtained as follows:
λ
v
= λ
L
1/2
, λ
Q
= λ
L
5/2
, λ
pi
= 1
Where λ
L
, λ
v
, λ
Q
and λ
pi
are scales for length, ve-
locity, discharge and grain size, respectively. Accord-
ing to the geometrical scale, λ
L
=250.
The upper Minjiang River lies in
Longmenshan earthquake belt, where
landslides and debris flows are common
because of the influence of plate move-
ment. At the same time, Wenchuan Earth-
quake produced significant influence for
the initiation of geological hazards (x
u
et alii, 2009; C
ui
et alii, 2009; l
iu
et
alii, 2009). Before the earthquake, small
debris flows were common and large
debris flows used to break out in Baihua
Gully, Fotang Gully and Luojuan Gully
in Yingxiu town, but with low frequency.
After the earthquake, both the magnitude and frequen-
cy of debris flow increased obviously. Nine debris
flows blocked Minjiang River several times, of which
a most serious one blocked mainstream every year
after the earthquake in Mizigou (Fig.1). During Aug.
2010, several large debris flows occurred and the one
in Yingxiu Town blocked Minjiang River, constrain-
ing the flow to rush into the newly built town.
There are plenty of debris flows in upper Min-
jiang River, making it an ideal area to study the ef-
fect of debris flow on channel evolution. According
to our investigation, there were 9 debris-flow dams
within 30km along the river after the earthquake,
20km away from the earthquake fault and 30km
away from the epicentre (Fig.2). This paper tries to
explain the influence of debris flow on channel evo-
lution through a series of model tests based on our
field survey in Mozigou.
Fig. 1 - Debris flow dam in Mozigou(June
17
th
,2008)
Fig. 2 - Distribution of debris flow dams in
the upper Minjiang River
Tab. 1 - Parameters for debris flow dam in Mozigou
background image
INFLUENCE OF DEBRIS FLOW ON CHANNEL EVOLUTION
Italian Journal of Engineering Geology and Environment - Book www.ijege.uniroma1.it © 2011 Casa Editrice Università La Sapienza
233
DEPOSIT OF DEBRIS FLOW IN THE RIVER
After entering the river, debris flow deposits in
three distinctive zones: the erosion zone near the junc-
tion, where the grain size increases from the edge to
the centre; the deposition zone, where debris deposits
en masse and keeps the original configuration; and the
washed zone, where the grains washed from the dam
are well sorted in size (Fig.4).
The setting of the modelling experiment (Fig.3)
consists of a concrete flume which is 0.4m wide and
5m long, with one-side toughened glass wall, a ma-
terial box of 0.5m x 0.4m. x 0.75m, and a tributary
flume orthogonal to the mainstream flume, which ac-
cords with characteristics of the debris dams in field.
13 sets of tests have been conducted, with param-
eters listed in Table 2.
Fig. 3 - Layout of the experiment setting
Tab. 2 - Parameters for the experiment
Fig. 4 - Deposit formation of debris flow after entering the river
background image
J.k. LIU, Z.L. CHENG, X.G. ZHANG & S. CHENG
234
5th International Conference on Debris-Flow Hazards Mitigation: Mechanics, Prediction and Assessment Padua, Italy - 14-17 June 2011
stream more strongly, which increases the shrink of
mainstream eventually.
VARIATION OF MAINSTREAM VELOCITY
Deposition of debris flow into the mainstream
shrinks the cross-section, and accelerated flow
through the dam is formed. In order to explain the ve-
locity variation, another index is defined:
F
v
= v
t
/ v
m
where v
m
is the velocity of water flow before the forma-
tion of debris dam and v
t
is the velocity after the dam
failure. In the experiment, F
v
, the shrink ratio S and the
average channel width B
m
(B
m
= S
m
/L, where L is the
length of confluence section), are recorded, while no re-
markable relationship is established among them. These
parameters are listed in Table 4, and the Pearson matrix
of the correlation coefficients are listed in Table 5.
But there is roughly a power-law relationship be-
tween F
v
and B
m
(Fig. 6):
B
m
= 16.917 F
v
- 0.6876
(R
2
= 0.743)
The outcomes above can be well exemplified in
field investigations. When the average width of river
channel gets bigger, velocity diversification for water
flow overtopping the debris dam turns to be smaller,
and vice versa. However, there is no obvious relation-
ship between the shrink ratio S of mainstream and the
The area of each zone changes with the density
and discharge of debris flow. However, limited by
conditions, a quantitative relationship between the ra-
tios of these zones is beyond the experiments.
SHRINK OF THE RIVER
Even if there is no blockage, debris flow into the
river may cause the shrink of channel. In order to ex-
plain the channel shrink easily, an index(S) is defined:
S = 1 - S
m
/S
sum
where S
m
is the channel area after dam failure, i.e. the
area shrunk by the debris flow; and S
sum
is the sum
of the erosive area S
t
, the deposition area S
b
and the
shrunk channel area S
m
(Fig.4), following a equation
as S
sum
= S
b
+ S
t
+ S
m
. Calculation for the experiments
are listed in Table 3.
Furthermore, an empirical relationship between
the shrink ratio S and the momentum ratio R
M
is estab-
lished as follows (Fig.5):
S = 0.0012R
M
+ 0.6538
(R
2
= 0.716)
This relationship is proved in our field investiga-
tion. First, an increased density of debris flow raises
the strength and height of the deposit (or the dam),
and thus the channel shrinks further; second, if the
debris flow has larger discharge or velocity, it will
carry more sediment into the river and constrain the
Tab. 3 - Shrink ratio of the river channel
Fig. 5 -
S
-
R
M
relationship
background image
INFLUENCE OF DEBRIS FLOW ON CHANNEL EVOLUTION
Italian Journal of Engineering Geology and Environment - Book www.ijege.uniroma1.it © 2011 Casa Editrice Università La Sapienza
235
tio RM and shrink ratio S of mainstream: S = 0.001
2RM + 0.6538, which shows that the magnitude of
debris flow has great impact on channel evolution.
• An exponential relationship is established betwe-
en the ultimate average width Bm of mainstre-
am and the coefficient of velocity variation Fv:
Bm=16.917Fv- 0.6876, which indicates that if
the ultimate average width of mainstream gets
smaller, rapids overtopping the debris dam will
be formed more easily.
ACKNOWLEDGEMENTS
This research was supported by the Special Item
of the Non-profit Industry Project of Ministry of Wa-
ter Conservancy (200801032) and Special Item of the
National Key Technology R&D Program of China
(2008BAKSOB04-5).
variation of mainstream velocity F
v
, because some
factors, referring to the size and configuration of dam
breach, may play an indispensable function.
CONCLUSIONS
Influence of debris on channel evolution involves
the following aspects:
• There is a linear relationship between momentum ra-
Table 4 Velocity variation index
F
v
Table 5 Pearson matrix of the correlation coefficients
Fig. 6
B
m
-
F
v
relationship
REFERENCES
C
osta
J e & s
CHusteR
R l. (1991) - Documented historical landslide dams from around the world. Open-File Report (United
States Geological Survey), 91-239.
C
ui
P, z
Hu
y y, H
an
y s, C
Hen
x Q & z
HuanG
J Q. (2009) -The 12 May wenchuan earthquake-induced landslide lakes:
distribution and preliminary risk evaluation. Landslides, 6: 209-223.
d
anG
C, C
ui
P & C
HenG
z l. (2009) - The formation and failure of debris flow-dams, background, key factors and model tests:
case studies from China. Environ Geol, 57: 1901-1910.
d
olan
R, H
owaRd
a & t
Rimble
d. (1978) - Structural control of the rapids and pools of the Colorado River in the Grand
Canyon. Science, 202(10): 629-631.
H
e
y P. (2003) - Influence of debris flow on river channel change of mountains. PhD, Dissertation, Graduate Institute of Chinese
Academy of Science. Chengdu: Institute of Mountain Hazards and Environment, CAS: 108-181.
background image
J.k. LIU, Z.L. CHENG, X.G. ZHANG & S. CHENG
236
5th International Conference on Debris-Flow Hazards Mitigation: Mechanics, Prediction and Assessment Padua, Italy - 14-17 June 2011
H
su
y s & H
su
y H. (2009) - Impact of earthquake-induced damned lakes on channel evolution and bed mobility: Case study of
the Tsaoling landslide dammed lake. Journal of Hydrology, 374: 43-55
l
ianG
z y, l
i
m, k
uanG
s f & x
u
y n. (2001a) - River channel changes due to converge of debris flow. Journal of Soil and
Water conservation, 15(5): 112-132.
l
ianG
z y, l
iu
x, x
u
y n & s
ui
z C. (2001b) - River channel changes due to convergence of debris flow. Journal of Natural
Disaster, 10(1): 45-50.
l
iu
n, z
HanG
J x, l
in
w, C
Hen
w y & C
Hen
z y. (2009) - Draining Tangjiashan Barrier Lake after wenchuan earthquake and
the flood propagation after the dam break, Sci China Ser E-Tech Sci, 52(4): 801-809.
k
oRuP
O. (2002) - Recent research on landslide dams - a literature review with special attention to New Zealand. Prog Phys
Geogr, 26: 206-235.
k
oRuP
O. (2005) - Geomorphic hazard assessment of landslide dams in South westland, New Zealand: Fundamental problems
and approaches. Geomorphology, 66: 167-188.
s
CHusteR
R L. (2000) - Outburst debris-flow from failure of natural dams. Proceedings 2nd International Conference on Debris
Flow Hazard Mitigation, 16-20 August, Taipei, 29-42.
x
u
Q, f
an
x m & H
uanG
R Q. (2009) - Landslide dams triggered by wenchuan Earthquake, Sichuan Province, south west
China. Bull Eng Geol Environ, 68: 373-386.
z
HonG
D L. (1991) - Documented debris flows and landslides from the downstream of Jinsha River. Reports of Geomorphic
hazards investigation in Sichuan Province, 34-119.
Statistics