Sediment budget of the Mekong basin: transfer processes and negative impacts of dams and extractions

rapidcrimsonMécanique

22 févr. 2014 (il y a 3 années et 6 mois)

62 vue(s)

“Nature and Culture Conservation Forum for

The sustainable development on the Mekong Delta”

5
th

Forum: Maintaining Ecosystem Services in the Mekong Delta of Vietnam



Sediment budget of the Mekong basin: transfer processes and negative
impacts of dams
and extractions


Prof. Jean Paul Bravard
, University of Lyon, France

Marc Goichot,
WWF
-
Greater Mekong, Vientiane



1.

What is the sediment discharge of the Mekong River?


The average yearly sediment discharge
of the Mekong River at its mouth
averages 150
+/
-

10
million tons according to scientific literature

and this value has not

been
discussed before 2005
.

This
rough
evaluation
, considered as a constant through years,

used results from suspended sediment monitoring undertaken in the early 1960’s at
diffe
rent stations of the Lower Mekong basin (Harden & Sunborg, 1992
; Ta
et al
., 2002).
D
ata of suspended sediment load
from monitoring stations
(Chiang Saen, Luang
Prabang, Ch
ia
ng Khan, Mukdahan, Khong Chiam, Pakse, Kratie) allowed scientists
to
model the impa
ct of the Lancang chain of dams and climate change (Fu
et al
., 2006
;
Kummu & Varis, 2007; Wang
et al
., 2011; Liu
et al
., 2012)
.

However Walling (2005)

warned the MRC

(Mekong River Commission)

to be very
cautious regarding the field sampling methods and

the

results obtained
to date
. Walling

pointed out that sand had been poorly taken into account. Geomorphic studies have
described the presence of sand in the channel without connecting landforms issuing
from flood deposition with the results of suspended
monitoring (Gupta et al., 2002;
Carling, 2009). The average value of sediment transport may the
n have been
underestimated in a
proportion

that remains unknown today
.


2.

New perspectives
for sand transport processes during floods


The field work for

this stu
dy

was performed
in 2011
-
2
0
12
by

WWF (
J
ean
-
P
aul Bravard
and Marc

Goichot
)

in the
main stem of the
Lower Mekong channel from Chiang Saen to
the upper Mekong delta.

This study first aimed at bridging the gap between landforms of
the banks, made of sand, and
transport processes. Indeed sand is continuously present
along the 2400 km stretch of the Lower Mekong, but at different elevations above the
low flow level. This study was based on field sampling, application of the so
-
called CM
image method (Passega, 196
4)
, and computation of unit stream power (a value of
stream energy during
small
floods).

The results are summarized
i
n fig. 1.
w
hich displays the long profile of the Mekong at
low flow, the values of energy


(function of slope,
of the
1 in 2 yrs flood and
channel
width)
, and the succession of processes involved in sediment transport:

-

Coarse bed load is a discontinuous process occurring in steep reaches controlled
by bedrock.
It is important to note that

fine gravel is present
a
s

far dow
nstream
as

the upper delta.

-

Sand is continuously transported as bed load, from the China border to the sea
and the delta shore,

-

Sand is
also
transported in suspension along most of the river reaches, except in
the Mukdahan

reach and downstream of the Cambodia
-
Vietnam border where


has very low values.
Sand displays a vertical gradient on the riverbanks

(fine
sand on upper banks, coarse sand closer to the bottom),

as it does in the water
body during floods (graded suspension): this is due to turbulence controlled by
velocity and roughness.

Sand is transported mostly during the sharp spates of the
annual flood,

-

When sand is not transported as graded suspension, it m
oves as bed loa
d, but the
two processes are usually simultaneous,

with relative proportions that change
depending on the level of the flood. Note that
sand transported in suspension
travels faster than sand transported as
bed load
.

-

Silt and clay are transp
orted as wash load (or uniform suspension) during floods
whatever energy may be.




Fig. 1
: Graph showing the long profile of the Mekong water level, the downstream variation of energy
(black squares) and the types of transport processes.


3.

What is the pra
ctical consequence

of the reevaluation of sand transport
for
the Mekong
sediment
budget
?


The reason why sand has been underestimated in the Mekong sediment budget
(and

in
some cases completely

forgotten) is due to the fact that
hydro
-
meteorological
stations
did not use sampling methods that allowed accurate monitoring

during flood spates.
R
eappraisal of suspended transport of sand
by
the MRC Information and Knowledge
Management Programme (
IKM
P
)
only very recently
confirmed that
sand

does constitute
a

significant part of
sediment transport (Koehnken, 2012). As a consequence sediment
budget of the Mekong should be revalued upward.
To date it

is not possible to provide a
revised average value
,
but we estimate
the
existing
figure
s
o
f 160 Mt per year could

be
increased by

a value

between 10 and
3
0 millions
.


4.

Why is the delta shoreline retreating?

D
am
s

and

upstream sediment
storage?


S
everal causes
have been put forward to

explain the
del
ta shoreline retreat
over the past

1
5

years.

Deltas are highly
dependant on sediment replenishment from the rivers that
have created the
m, thus reduction of sediment in
put from upstream are

prime cause
s to

explain coastal erosion on delta fronts.
We suggest two significant causes.


The first one relates to dam impacts.

4.1.
Several papers have held the Lancang dams liable for downstream sediment
shortage at the river mouth

(Lu & Siew, 2006; Liu
et al
., 2012)
.

If it may be true for the
si
lt
-
clay component
, as part of the total load

i
s

being

trapped in the
large hydropower
reservoirs

on the Chinese section of the main stem of the Mekong. M
ost of
the
sand is

probably
also
trapped

in those large
reservoirs

But t
he amount of
sand
stored

in the
channel

all along the river course make it du
bious that sand shortage at the mouth may
be due to the Lancang dams.

The impact of the Chinese hydropower dams will need
more time to affect the coast of Vietnam
, as it will occur only after the stock of sand in
northern Laos is depleted
.

This might take
20 to 50 years.

4.2. It is estimated that the

Three S


rivers provide
d about 10% of the overall
load before 1993

(Sarkkula
et al
., 2010)
, or

35
-
40% of the
present
sediment supply from
the Mekong tributaries
. The main

supplier
of sand
is the Se San
R
iver
which has been
dammed in Vietnam and which will be dammed downstream in Cambodia.
This
means
this
important source of sand will be blocked. Considering the downstream position of
this tributary in the Mekong basin,

the existing

hydropower
dam

projects

prob
ably play
an important role in the sediment shortage at the mouth.

4.3. The WWF study has considered the expect
ed

impacts of
the
Xayabury dam,
in
Lao PDR. The 80 km long reservoir
behind
the
3
5

m

high
dam
will trap a significant part
of the sand stored downstream of the lowest Lancang dams, but should let most of wash
load (silt and clay) pass through the dam (considering turbulence in the reservoir during
floods). However WWF stressed the importance of the

compaction of mixed sedimen
t
because it will decrease the
ability of floods to rework deposits

on the bedrock lateral
platforms of the channel bottom. Trapping of silty sand will affect the downstream
reaches with erosion of sand banks and bars, due to th
e reduction of
sediment
concentration during floods.

4.4. Also,
the
WWF study considered flushes

that have been proposed by the
Government of Lao PDR as a mitigation measure.

Flushes are

really implemented for
“cleaning” the reservoir,
but they
will have d
etrimental impacts on habitat (clogging of
substrate), and on fauna (lethal concentrations of suspended sediment). It is dubious
that the operator will be able to follow the instructions and provide the required low
concentrations. The dilemma
will

therefo
re
be to ch
o
ose between the
detrimental
sediment trapping or damaging flushes. In both cases the Mekong River will be
significantly
altered.

In conclusion of

point 4,
WWF stresses

that dams have already an unmeasured impact
on the

Mekong sediment load. Sediment shortage impacts, which are

non
-
reversible
, will
increase in the future
when those impacts will have prograded to downstream. It will be
all the more severe
i
f planned dams are built up on the main steam and on the
tributari
es
that provide an important
supply

o
f

sand.


5.

The severe detrimental impacts of sand and gravel harvesting


WWF launched a study in the four countries of the Lower Mekong
main stem
in order to
quantify

sand and gravel extraction
s
, the places of extraction

and the trends.

The volumes per
category of grain
-
size are presented in table 1:


Country

Extractions (1000 m
3
/yr)



Sand

Gravel

Pebbles

Total

%


Lao

904.1

10

454.5

1 369

4%


Thailand

3 677.2

857.75

0

4 535

13%


Cambodia

18 748.5

2 045

0

20
793.45

60%


Vietnam

7 750

0

0

7 750

22%


TOTAL

31 079.8

2 912.7

454.5

34 447

100%


%

90%

8%

1%

100




Table 1: Volumes and percentage of grain
-
size categories per country

(2010
-
2011)

In 2010
-
2011, most of the extractions were located in Cambodia but in part
because
sediment resource is exhaust
ed

in the branches of the delta

in
Vietnam. Out of the 118
sites operated, 88 are less than 3 years old. The trend is an increase of volumes

extracted

from Savannak
h
et to the delta.

The 34.4 millions m
3

extracted
is pro
bably an underestimate of the effective extraction

as, the surveys was not exhaustive (only targeting the largest operators), were only
targeting
the
main stem, thus not encompassing extraction activitie
s

on tributaries, and
finally
volumes

have probably b
een minimized by the interviewed dredgers. This figure
corresponds to 56.6 m
illions tons of sand and gravel and is probably far higher. We
suggest that
current rates of
sand extraction exceeds the va
lues of
annual
transit
, thus
extraction rates exceed
repl
enishment

rates
.


Conclusion


Sand and gravel extraction have probably been so far the most important factors of the
sediment depletion monitored along the Mekong delta shoreline. The second factor,
sediment trapping in
dam
reservoirs, will have increasing impacts in the future because
the impacts of the Lancang reach are partly delayed

and more dams are constructed and
planed
.
It is important to note that the impact of sand and gravel extraction can be
considered as much eas
ier to reverse than the impact of hydropower.

Extractions should be considered in the framework of a master plan taking the
total
sediment balance into account and dams should be built up
only
on
tributaries
which do not provide
significant quantities of
s
and and gravel to the
downstream course

and delta
, and not on the main stem of the Lower Mekong
.


References


-

Carling P., 2009b: Geomorphology and sedimentology of the Lower Mekong River. In I.C. Campbell (ed.):
The Mekong
,
Biophysical Environment of an
International River basin
, Amsterdam, Elsevier, p. 77
-
111.

-

Fu K.D., He D.M., Li S.J., 2006

: Response of downstream sediment to water resource development in
mainstream of the Lancang River.
Chin. Sci. Bull
., 51(Supp.): 119
-
126.

-

Gupta, A., Lim, H., Hua
ng, X., Chen, P., 2002. Evaluation of part of the Mekong River using satellite
imagery.
Geomorphology

44, p. 221

239.

-

Harden, P.O. and Sundborg, A. (1992)
The Lower Mekong basin suspended sediment transport and
sedimentation problems.
Hydroconsult,
Uppsala, Sweden.

-

Koehnken L., 2012: IKMP Discharge and Sediment Monitoring program Review, Recommendations and
data Analysis. Part 2: data analysis ad preliminary results. Phnom Penh, MRC.

-

Kummu M., Varis O., 2007

: sediment
-
related impacts due to upst
ream reservoir trapping, the Lower
Mekong River.
Geomorphology
, 85, p. 27
-
293.

-

Liu C., He Y., Wang J., 2012

: Changes in sediment load of the Lancang
-
Mekong River and its response to
the hydro
-
power development. 4th International Conference on Estuaries
and Coasts, 8
-
11 Oct. 2012,
Water Resources University, Hanoi, Vietnam, 10 p.

-

Lu X.X., Siew R.Y., 2006

:

Water discharge and sediment flux changes in the Lower Mekong River.
Hydrol.
Earth Syst. Sci. Discuss
., 2, 2287

2325.

-

Pas
sega

R., 1964

: Grain
-
size repre
sentation by CM pattern as a geological tool. Journal of sedimentary
Petrology, 34(4), 830
-
847.

-

Sarkkula J.,

Koponen J., Lauri H., Virtanen M., Kummu M., 2010

:
Origin, Fate, and Impacts of the Mekong
Sediment
. MRC, SYKE
Report, 53 p.

-

Ta T.K.O, Nguyen V.L., Tatheishi M., Kobayashi I. Tanabe S., Saito Y., 2002

: Holocene delta evolution and
sediment discharge of the Mekong River, southern Vietnam.
Quaternary Science Review
, 21, p. 1807
-
1819.

-

Walling D.E., 2005

:
Evaluat
ion and analysis of sediment data from the Lower Mekong River
. Unpublished
report for the Mekong River Commission, 61 p.

-

Wang J.
-
J., Lu X.X., Kummu M.,

2011

: Sediment load estimates and variations in the Lower Mekong River.
River Research and Applicatio
ns, 27, 33
-
46.



This reseach was made possible with the financial support of Fond Francais pour
l’Enviornnement Mondial (FFEM) and was conducted in the frame of a project co
-
implemented with the Information and Knowledge Management
Programme
of the
Mekong

River Commission Secretariat.