Documentos de Académico
Documentos de Profesional
Documentos de Cultura
" ~
Rodney White
I ThomasTelford
I
\
I
..,, :-
'\
Published by Thomas Telford Publishing, Thomas Telford Ltd, 1 Heron Quay, London E14 4JD.
URL: http://www.thomastelford.com
Distributors for Thomas Telford books are
USA: ASCE Press, 1801 Alexander Bell Drive, Reston, VA 20191-4400, USA
Japan: Maruzen Co. Ltd, Book Department, 3-10 Nihonbashi 2-chome, Chuo-ku, Tokyo 103
Australia: DA Books and Journals, 648 Whitehorse Road, Mitcham 3132, Victoria
First published 2001
- j
i
--I
,i
:-1
~I
-j
-I
A catalogue record for this book is available from the BIitish Library
ISBN: 07277 2953 5
Rodney White and Thomas Telford Limited, 2001
All rights, including translation, reserved. Except as pennitted by the Copyright, Designs and Patents Act
1988, no part of this publication may be reproduced, stored in a retrieval system or transmitted in any form
or by any means, electronic, mechanical, photocopying or otherwise, without the prior written permission of
the Publishing Director, Thomas Telford Publishing, Thomas Telford Ltd, 1 Heron Quay, London E14 4ID.
II
:--1
-I-~
This book is published on the understanding that the author is solely responsible for the statements made and
opinions expressed in it and that its publication does not necessarily imply that such statements and/or
opinions are or reflect the views or opinions of the publishers. While every effort has been made to ensure that
the statements made and the opinions expressed in this publication provide a safe and accurate guide, no
liability or responsibility can be accepted in this respect by the authors or publishers ..
Typeset by Apek Digital Imaging, Bristol, UK
Printed and bound in Great Britain by MPG Books, Bodmin, Cornwall
-I
Preface
In many areas of the world the life span of reservoirs is determined by the rate
of sedimentation which gradually reduces storage capacity. Eventually, this
process destroys the ability of the scheme to deliver the benefits for which it was
built. Many major reservoirs are approaching this stage in their life.
There are various options available for positively managing sedimentation in
reservoirs.
2. Minimising deposition
of sediments in reservoirs
There are two main ways of passing sediments through reservoirs without
deposition:
sluicing - the process of passing sediment laden flo09 waters through the
reservoir. This method involves the reduction of water levels in the reservoir
during the flood season and is applicable mainly to very fine sediments (clays
and silts).
" density current venting - this method has the attraction that it is not necessary
to lovver water levels but is only applicable in very exceptional circumstances
where sediment-induced density currents carry very fine sediments towards the
dam. The number of cases where density current venting has been successful
is minimal.
a
iii
EVACUATION OF SEDIMENTS
It
iv
PREFACE
worldwide and draws together this existing knowledge into a concise manual for
practising engineers.
The book begins by assessing the scale of the problem of reservoir
sedimentation. It assesses the volume of storage that is likely to be lost to
sedimentation and compares this volume with the net volulne of storage that is
likely to be required to meet continuing demand. The book provides a review of
the current state of knowledge of reservoir flushing, and then considers the
worldwide experience of flushing to date. Areas of the world are then identified
where flushing is likely to be most useful. The final section of the book describes
the more detailed investigations which must be carried out when considering
sediment flushing at a particular dam site.
Acknowledgements
This book describes work which was funded principally by the Department of the
Environment, Transport and the Regions (DETR) as part of the Partners in
Innovation programme. The work was undertaken jointly by HR V/allingford, as
the lead partner, TAMS UK, Binnie Black and Veatch, and LAWGIBB.
It is a pleasant duty to acknowledge the valuable contributions made by
Laurence Attewill and Atila Bilgi of TAlVIS UK, Ed Atkinson and Andrew Nex
of ILl{ Wallingford, John Ackers, Chris Scott and Robert Jones of Binnie Black
and Veatch, and Richard Wingfield and Mary-Ellen Cromack of LAWGIBB.
HR Wallingford is an independent specialist research, consultancy, software
and training organisation that has been serving the water and civil engineering
industries worldwide for over 50 years in more than 60 countries. \Ve aim to
provide appropriate solutions for engineers and managers working in:
water resources
II
h~gation
groundwater
urban drainage
rivers
tidal waters
ports and harbours
coastal waters
offshore.
Address:
Internet:
vii
Notation
DDR
DSOT
D50B
F~VR'
H fiush
Hmax
LTCR
lv!in
LV
P sand
Qf
Qm
Qs
S
SBR
SBR d
SSR
TE
Tf
TIVR
\tV
Ibed
the ratio of the height of water at the dalTI during flushing to the
maximum height of water at the dam (to reservoir retention
level), both measured above original river-bed level (nondimensional)
50 percentile size of sediment in transport (IILrn)
50 percentile size of river-bed material (mm)
the ratio of the natural width of the flushing channel and a
representative bed width for the reservoir (non-dimensional)
the height of water at the datu during flushing, measured above
original river-bed level (m)
the maximum height of water at the dam (to reservoir retention
level), measured above original river-bed level (m)
the sustainable storage capacity divided by the original storage
capacity of the reservoir (non-dimensional)
the average sediment inflow rate (t/yr)
the interval between flushing operations (yrs)
proportion of total sediments in motion which exceed 0-06 rom
in size (sand and coarser material)
the flushing discharge (m3/s)
mean annual flow (m 3/s)
the sediment transporting capacity of the flow in the incised
flushing channel (tis)
the longitudinal energy gradient trliough the reservoir (nondimensional)
the ratio of sediment flushed to sediment depositing (nondimensional)
the specific value of SBR related to flushing with maximum
reservoir drawdown (non-dimensional)
.~
sand-size ratio, DsoTIDsoB
the trapping efficiency of the reservoir, i.e.
ratio of sediments
retained within the reservoir to sediments~.;,entering (nondimensional)
.
the duration of flushing (days)
the ratio of the natural top width of the flushing channel and a
representative top width for the reservoir
the bed 'width of the incised flushing channel (m)
the representative bottom width of the reservoir, taken as the
bottom width of the reservoir just upstream of the dam (m)
ix
EVACUATION OF SEDIMENTS
Wtep
Wmin
a
p
"
Contents
Dlustrations
1.
Executive summary
1.1.
1.2.
2.
xv
Review of sedimentation
2.1.
2.2.
2.4.
2.5.
2.6.
Introduction, 3
Summary of conclusions, 4
1.2.1.
Review of sedimentation in reservoirs, 4
1.2.2. Research into factors Wllich influence
sediment flushing, 7
1.2.3. vYorldwide experience of sediment
flushing, 9
1.2.4.
Geographical areas suited to flushing, 11
1.2.5.
Site-specific investigations and design
considerations, 13
reservoirs
15
Summary, 17
World total reservoir storage, 17
2.2.1. ICOLD World register of dams, 17
2.2.2. Other sources, 18
2.2.3.
Conclusion, 18
vVorldwide distribution of existing storage, 18
2.3.1.
Global water resources, 18
2.3.2.
Geographical distribution, 19
World demand for more storage, 19
2.4.1. Population, 19
2.4.2.
Irrigation, 21
2.4.3.
Hydropower, 21
2.4.4.
Conclusion, 23
Distribution of demand for more storage, 23
2.5.1.
Europe, 23
2.5.2.
North America, 24
2.5.3.
South and Central America, 25
2.5.4.
Aflica, 26
2.5.5.
Asia and Oceania, 27
2.5.6.
Summary, 28
Rate and distribution construction of new
xi
EVACUATION OF SEDIMENTS
2.7.
2.8.
2.9.
2.10.
xii
reservoirs, 28
2.6.1. Worldwide, 28
2.6.2. Distribution of construction of storage, 30
2.6.3.
Comparison of storage construction with
demand, 30
Rate and distribution of loss of storage, 31
2.7.1.
Rate of loss of storage, 31
2.7.2. Distribution of loss of storage, 32
Trends in the rate of loss of storage, 34
Reservoir size and rate of loss of storage, 35
Requirements for new storage, 36
3.
37
4.
63
CONTENTS
5.
6.
125
7.
References
131
8.
Bibliography
141
Appendices
Appendix 1.
Appendix 2.
.8..Dlue:nWlX 3.
91
149
Reservoir data, 151
Numerical model case study, 163
Flushing case studies, 171
Erosion, 211
251
xiii
I!' ..
I
Illustrations
Tables
Table 1.L
Table 1.2.
Table 1.3.
Table 1.4.
Table 1.5.
Table 2.1.
Table 2.2.
Table 2.3.
Table 2.4.
Table 2.5.
Table 2.6.
Table 2.7.
Table 2.8.
Table 2.9.
Table 2.10.
Table 2.11.
Table 2.12.
Table 2.13.
Table 2.14.
Table 2.15.
Table 2.16.
Table 2.17.
Table 2.18.
Table 2.19.
Table 2.20.
Table 2.21.
Table 2.22.
Table 3.1.
5
5
6
8
19
20
21
21
22
22
23
24
25
26
26
27
28
29
29
29
30
31
32
33
34
35
46
xv
EVACUATION OF SEDIMENTS
Table 3.2.
Table 3.3.
Table 3.4.
Table 3.5.
Table 3.6.
Table 4.1.
Table 4.2.
Table 4.3.
Table 4.4.
Table 4.5.
Table 4.6.
Table 5.1.
Table 5.2.
Table 5.3.
Table 5.4.
Table 5.5.
Table 5.6.
Table 5.7.
Table 5.8.
Table 5.9.
Table 5.10.
Table 5.1l.
Table 5.12.
Table 5.13.
Table 5.14.
Table 5.15.
Figures
Figure 2.1.
Figure 2.2.
Figure 2.3.
xvi
52
54
55
56
58
68
69
70
70
72
74
94
94
96
97
99
112
112
113
113
114
114
115
115
117
119
20
23
30
ILLUSTRATIONS
Figure 2.4.
Figure 2.5.
Figure 3.1.
Figure 3.2.
Figure 3.3.
Figure 3.4.
Figure 5.1.
Figure 5.2.
Figure 5.3.
Figure 5.4.
Figure 5.5.
Figure 5.6.
Figure 5.7.
Figure 5.8.
Figure 5.9.
33
35
41
44
48
49
102
104
106
107
109
110
111
116
118
xvii
e u
rn a y
;'.: '
-- ..
....
I. Executive summary
1.1.
INTRODUCTION
There are around 40 000 large reservoirs worldwide used for water supply, power
generation, flood control, etc. Between a half and one per cent of the total storage
volume is lost annually as a result of sedimentation and 300 to 400 dams, at the
cost of around 5 million per dam, would need to be constnlcted annually to
maintain current total storage. The introduction of flushing systems in some old
dams, where appropriate, and in the design of new dams could save 10 per cent
of these costs, i.e. 200 million annually. This book provides guidelines on the
design aspects of flushing systems and indicates where such systems could be
used beneficially.
'
The benefits attributable to dams and reservoirs, most of which have been built
since 1950, are considerable and they have improved the quality of life
worldwide. These benefits can be classified under three main headings.
Irrigation
About 20 per cent of cultivated land worldwide is irrigated, some 300 million
hectares. This irrigated land produces about 33 per cent of the worldwide food
supply. Irrigation accounts for about 75 per cent of the world water consumption,
far outweighing the domestic and industrial consumption of water.
Hydropower
About 20 per cent of the worldwide generation of electricity is attributable to
hydroelectric schemes. This equates to about 7 per cent of worldwide energy
usage.
Flood control and storage
Many dams have been built with flood control and storage as the main motivator,
e.g. the Hoover dam, the Tennessee Valley dams and some of the more recent
dams in China.
In many areas of the world the life span of these reservoirs is determined by the
rate of sedimentation which gradually reduces storage capacity and eventually
destroys the ability to provide water a.Tld power when sedinlents clog low level
outlets. Many major reservoirs are approaching this stage in their life.
One way of preserving reservoir storage is to flush sediments through purposebuilt outlet works within the dam. This technique can be applied to existing dams
(with adaptation of the
works) and to ndw dams. However, the
technique is only effective under certain favourable conditions and is not
applicable universally. The alternative is to build more dams to replace the
EVACUATION OF SEDIMENTS
depleting storage of the existing stock. However, there are fewer and fewer good
dam sites available and new dams can have serious environmental and social
consequences.
Dams designed within the last ten years or so, have sometimes incorporated
design features which will allow flushing to be undertaken when appropriate.
However, these designs have been based on considerations which are sitespecific. The purpose of this project was to produce a generally applicable design
manual which provides guidance on the necessary hydrological, hydraulic,
sedimentological and topographical features for successful flushing. It is based
on a review of recent research and field experience worldwide and draws together
this existing knowledge into a concise manual for practising engineers.
The book starts by assessing the scale of the problem of reservoir
sedimentation. It compares the volume of storage that is likely to be lost to
sedimentation and compares this volume with the net volume of storage that is
likely to be required to meet continuing demand. The book provides a review of
the current state ofknowledge of reservoir flushing, and from this proceeds to
consider the worldwide experience of flushing. Areas of the world are then
identified where flushing is likely to be most useful. The final section of the book
describes the more detailed investigations which must be carried out when
considering sediment flushing at a particular dam site.
1.2.
SUMMARY OF CONCLUSIONS
-.i
-~
EXECUTIVE SUMMARY
Table 1.1.
Demand/or storage
Period
Distribution of demand
Europe. Although the dema.'1d for new storage is sensibly zero in much of
Western Europe, it does appear that for the region as a whole there is a small
demand, of the order of 1% per annum, for new storage for hydropower, mainly
concentrated in Eastern Europe.
North America. Although the data show that the energy generated by
hydropower, as well as the area of land under irrigation, continued to ,grow
through to the 1990s, the fact that no new storage was constructed in that period
..;"'.... ~;".::, ....'v .. ..., that the data are influenced by operational factors. Therefore, although
Table 1.2.
Region
49
51
54
467
495
424
Africa
167
203
248
EVACUATION OF SEDIMENTS
Table 1.3.
The data show that the energy generated by hydropower, as well as the
arya of land unde~ irrigation, grew weakly through the 1980s and into the 1990s
at a rate well below the rate of growth of population. This trend is likely to
continue, despite strong demand and great potential, so an annual growth rate of
2% is postulated.
Africa.
')0')0..')010
~
EXECUTIVE SUMMAR.Y
apparent fall in the rate of construction during the 19908 is dramatic and may, in
part, be due to inadequate data.
The distribution of the growth of new storage is set out in Table 2.14.
EVACUATION OF SED1MENTS
Table 1.5.
Continent
Gross requirement
New demand
2000-2010
Loss to sedimentation
up to 2000
49
54
103
North America
112
112
South America
467
17
484
Aftica
167
35
202
315
349
664
998
567
1565
Europe
I.
Total
I
i
EXECUTIVE SUMMARY
Site-specific factors
The most suitable conditions for flushing are to be found in reservoirs that are
approximate i.T1 shape to the incised channel which develops during flushing.
Long, relatively narrow, reservoirs are better suited to flushing than short, wide,
shallow reservoirs.
Summary
Reservoirs in tb.e 'upper and middle reaches' of rivers are likely to be best suited
to sedinlent flushing for the following reasons.
\I
fII
II
In the lower reaches, reservoirs are likely to have inundated areas that have
previously been flood plains and these areas would not be reached by the
incised flushing channel which is inevitably of limited width.
The longitudinal slope available for the flushing channel is relatively small,
thus limiting the amount of sediment transport.
Reservoir volumes in the lower reaches are likely to be larger compared with
.
run-off and hence water availability becomes a restraint on
the mean ap. .11ual
sediment flushing.
.-
of sediment flushing
EVACUATION OF SEDIMENTS
10
!"
Transport
of sediment
Climatic zones
An understanding of the precipitation regimes throughout the world may allow
the definition of climatic zones based on temperature and precipitation regimes.
This may permit
definition of areas of high and low erosion rates. It is
difficult to classify distinct climatic zones as they tend to merge into one another
rather than have sharp boundaries, but a number of general models have been
produced.
II
EVACUATION OF SEDIMENTS
There have been many climatic classifications produced but one of the most
common is based on the original Koppen classification, with eight climatic
regions based on four temperature zones and one moisture zone and the seasonal
domination of air masses. Details of this classification are given in Chapter 5 and
a discussion of alternative classifications is given in Appendix A4.3.
The eight Koppen climatic regions are as follows.
Tropical wet: classification Af.
Tropical wet and dry: classification Aw, Am and BS.
e Tropical desert: classification BW.
e Mid-latitude wet: classification Cf and Df.
Mid-latitude winter dry: classification Cw and Dw.
Mid-latitude summer dry (Mediterranean Climate): classification Cs and Ds.
Polar wet and dry: classification ET.
Polar desert: classification EF.
Hydrological characteristics
Experience has shown that low reservoir water levels provide the most effective
conditions for sediment flushing. To allow water levels to be lowered requires
confidence that rainfall can be relied upon to refill the reservoir. It follows that
well defined wet and dry seasons will be favourable for a sediment flushing
regime. Such a climate is defined by Koppen as tropical wet and dry: Aw, Am and
BS. Also, there are areas in the mid-latitudes where spring snowmelt provides a
regular and predictable annual pattern of high flows.
River discharges must also be sufficient to transport sediment loads through
the reservoir. Regions of low precipitation like the Sahara and other desert
environments therefore will not be suitable for flushing even if they exhibit a
defined seasonal effect. The availability of water will also affect the duration and
discharge rate of the flow required for flushing. Where there is a limited amount
of water it is better to use a high discharge for a short period of time than a low
discharge for a long period of time. This increases the amount of sediment that
is removed.
Areas of the world which are best suited to reservoir flushing
It is not possible to define precisely which specific areas of the world will provide
conditions for successful flushing. In reality there is a spectrum of conditions
ranging from those sites where conditions are ideal to those sites which are quite
unsuited to sediment flushing.
From the Koppen classification of climatic zones and the mid-latitude spring
snowmelt effect, the requirements for successful flushing are most likely to be
met in the following locations:
.
parts of Central America extending into South America
areas in North and South America where the rivers are fed by the Rockies and
the Andes
parts of Central Africa from the Ivory Coast in the west to Sudan in the east
12
EXECUTIVE SUMMARY
areas in Central Asia where the rivers are fed by the Himalayas, including
Pakistan, India and Nepal
& parts of Asia, including Calnbodia, Vietnam and Thailand.
Site investigations
Site investigations are required to identify t.~e most compact and efficient
geometry for the flushln.g outlets and the energy dissipation works. The reservoir
itself requires a detailed survey to establish its topography.
Hydrological investigations
Inflows to the reservoir need to be established with confidence. This involves the
acquisition of historical records of river flows going back at least 30 years and
preferably longer and/or the development of a longer sequence from rainfall
records using catchment modelling.
Sediment investigations
The amount and nature of the sediment entering, or likely to enter, the reservoir
needs to be established. This requires measurements of sediment transport rates
in the rivers feeding the reservoir over many years to establish the results with the
confidence that is required.
In the case of existing reservoirs, information about the amount of sediments
entering the reservoir can be augmented by surveys of the amount and nature of
the material settling within the reservoir.
is required, however, to allow. for
the amount of material, mainly fine, which passes through the reservoir without
deposition.
Bed material sampling should be undertaken in the reservoir and in the rivers
which feed the reservoir. A sound knowledge of the nature of these sediments,
including their size, specific gravity and degree of compaction, is an essential
requirement to provide inputs for numerical models which simulate sediment
moven1ent, see below.
Hydraulic modelling
Numerical (computer) modelling of the way sediment is likely to behave within
the reservoir and the amount and nature of the sediment which will be passed to
the downstream reach is the cornerstone of any detailed evaluation of flushing
facilities.
Computer simulations of reservoirs ideally use representative, long-term
sequences of water and sediment inflows to the reservoir. The models are capable
13
EVACUATION OF SEDIMENTS
14
atio
er\f Irs
SUMMARY
In this chapter a summary of the total volume of reservoir storage, and its
distribution is given. An attempt is made to quantify the future demand for new
storage, especially for hydropower and irrigation, and this estimate is compared
with the historic rate of reservoir constnlction throughout the twentieth century.
The rate of loss of storage due to sedimentation is made, so as to arrive at a
prediction of both the net and gross future storage requirements.
'
2.2.
.c_
""
_"
'' . '''., . .
Some countries failed to respond 8....1J.d for these countrie$ .data:vvasretained
from the earlier edition.
--
The register gives the total number of dams reported by the 80 member countries
and the 60 non-member countries as 25 410. No exact "sUlTilllary of storage
volume is provided but in the introduction it is stated that the total volume of
storage is 6000 km3 From the analysis of the data in the register, the total gross
storage volume of the reservoirs reported by ICOLD is 6465 km3 This includes
490 km 3 of storage registere9 as under construction.
In order to estimate the total world storage it is necessary therefore to assess
the extent to which the register under-reports the total number of dams and the
number of dams less than 5 m high (and their storage). It is evident from the
res!:ister that the ratio of dams less than 30 m high to the total number of dams
varies from about 90% in the case of India to 5%~in the case of China. From this
it can be infened that many countries, but China in particular, under-report dams
17
EVACUATION OF SEDIMENTS
2.2.3. Conclusion
The best estimate of world storage in reservoirs (excluding natural lakes used as
storage for power and irrigation) is 6815 km3 ,
2.3.
18
SEDIMENTATION IN RESERVOIRS
2.4.
2.4.1. Population
The world's population in 1990 is estimated to have been 5286 million, growing
at an annual rate of 15%. This rate of growth is forecast to decline in the coming
decades so that the predicted future world population is as shown in Table 2.2
and Figure 2.1.
Water demand is expected to continue to grow at a faster rate than that
predicted by population growth alone. This is because the present -per capita
Table 2.1.
Region
North America
Number
of dams
1498
South America
Northern Europe
7205
2277
I
I
!
1845
1039
938
Average size of
reservoir:
Fraction of
world total
Gross
storage:
km3
M.m3
I
I
29%
256
16%
694
412
15%
I
3220
145
2%
45
Sub-Saharan Africa
966
575
9%
595
North Africa
280
188
3%
652
China
1851
649
10%
351
Southern Asia
4131
319
5%
77
44
148
2%
3364
277
117
2%
424
Pacific Rim
2778
277
4%
100
Middle East
895
224-
3%
250
25422
6464
100%
254
Southern Europe
Central Asia
South-East Asia-
World total
19
EVACUATION OF SEDIMENTS
Table 2.2.
Year
Population:
millions
1990
5286
153
2000
6158
134
2010
7032
115
2020
7887
2030
8671
072
2040
9318
054
2050
9833
095
18
(J)
8000
'E
c0
~
'3
0..
a..
6000
10000
--- -- --
......
------ .....
-...----- -- ---
..... ------
--
16
14
..c:
08
4000
06
04
2000
02
__~--~--------~----~O
OL-------~------~--------~
1990
2000
2010
2020
Year
Figure 2.1.
20
*'
12 1Y
2030
2040
2050
0)
Cii
::I
c
c
<:
,I
SEDIMENTATION IN RESERVOIRS
Table 2.3.
Year
Total
.,
Reservoirs
1950-1960
388
13-47
1960-1970
267
1112
1970-1980
251
616
1980-1990
221
354
1990-2000
231
261
,1 ~
~
!
.,
the two. It is pertinent therefore to examine the historic growth in irrigation and
energy generation.
2.4.2. Irrigation
The gro\vth in demand of water for irrigation is illustrated by the growth in
irrigated area, given in Table 2.4.
These data give only an ~indication of the growth of demand for water since
they are a record of the irrigated area actually planted and are subject to annual
variations in water availability: the effects of the drought in the 1980s is apparent.
Moreover, no data are available on the proportion of the total demand met from
storage reservoirs compared with other sources.
2.4.3. Hydropower
Historic groV'lth
\-Vorldwide power consumption is growing at a faster rate than population growth
as nations industrialise. The share of energy generated by hydropower is difficult
Table 2.4.
Year
Irrigated area:
million ha
1975
190
222
1980
212
102
1985
223
123
1990
237
187
1995
260
21
EVACUATION OF SEDIMENTS
Table 2.5.
1990
2 112000
1995
2474000
321
to predict as the viability of new schemes is influenced heavily by the price of oil
and gas and the safety and environmental concerns associated with nuclear
power. The cost of new hydroelectric schemes increases in real terms as the best
sites are used up.
The historic energy generated by hydropower is shown in Table 2.5.
As with the data for irrigation, the link between these data and reservoir
storage is tenuous: the energy generated is affected by water availability and does
not directly reflect either the growth of installed capacity or storage. The high
growth rates in the 1970s is a reflection of the quadrupling of oil prices in 1972.
Potential
The potential for new hydropower development is indicated in Table 2.6, which
compares average energy generated by hydropower with the estimated
economically-feasible energy in each continent.
These data (Table 2.6) show that future hydropower development is unlikely
to be constrained by the lack of suitable sites within the foreseeable future.
Table 2.6.
Region
Percentage
developed
Europe
525000
800000
66
North America
627000
1000000
63
491000
2325000
21
67 000
1000 000
750 000
3700000
20
2440000
8825 000
28
Africa
Asia and Oceania
Total
22
..
SEDIMENTATION IN RESERVOIRS
5
-- - - -_ Water demand
- - - - -
Hydro demand
- - - Population
-
1
1970
1975
1980
Storage
---
.--.---.... ..
1985
1990
1995
2000
2005
2010
2015
2020
Year
Figure 2.2.
2.4.4. Conclusion
The rates of growth of population, global water demand, irrigation area and
hydropower generation are compared in Figure 2.2.
From the rates of growth for population, water consumption, irrigation area
and hydropower, the following growt...1-t rates for demand of storage are postulated
and are shown in Table 2.7.
2.5.
2.5.1. Europe
Population
The population of Europe (including Russia) is predicted to grow from 722
million in 1990 to a maximum of 730 million in the year 2000, and thereafter
decline to 723 million in 2020 and to 678 million in 2050. The overall effects of
Table 2.7.
Period
2000-2010
156
998
2010-2020
139 .
1032
?Cnn_"il~il
23
EVACUATION OF SEDIMENTS
Table 2.B.
Year
Irrigation
area:
million ha
1975
273
33
1980
321
1985
Energy:
GWh/yr
Annual growth in
following 5 years:
%
535 000
403
21
652 000
140
357
00
699 000
-032
1990
357
-68
688 000
139
1995
251
Annual growth in
following 5 years:
%
737 000
24
~.
.~
i
~<;
SEDIMENTATION IN RESERVOIRS
Table 2.9.
Year
lITigation
area:
million ha
Annual growth in
following 5 years:
180
28
520 000
10
207
-19
547 000
22
1985
188
02
611 000
00
1990
190
13
614000
24
1995
203
1975
1980
Energy:
GWhlyr
Annual growth in
following 5 years:
%
693 000
of population will result In regional shortages and surpluses which may generate
demand for reservoir storage.
Conclusion
, Although the data show that the energy generated by hydropower, as well as the
area of land under irrigation, continued to grow through to the 1990s, the fact that
no new storage was constructed in that period suggests that the data are
influenced by operational factors. Therefore, although there remains a large
undeveloped potential resource in Canada, environmental pressures will
probably preclude any further development.
25
EVACUATION OF SEDIMENTS
Table 2.10.
Year
In'igation
area:
million ha
Annual growth in
following 5 years:
%
Energy:
GWh/yr
1975
117
08
117 000
Annual growth in
following 5 years:
%
112
I
1980
122
1985
14.. 8
1990
1995
167
198
39
199 000
72
24
282 000
53
35
365 000
55
477 000
I
2.5.4. Africa
Population
The population of Mrica is predicted to grow from 663 million in 1990 to 832
million in the year 2000, rising to 1348 million in 2020 and 2141 million in 2050
- with the growth rate declining from 277% at the present time to 153% by
Table 2.11.
26
SEDIMENTATION IN RESERVOIRS
Table 2.12.
Year
Annual growth in
foUowing 5 years:
1975
94
12
1980
100
14
1985
107
13
1990
114
15
1995
123
Energy:
GWh/yr
Annual growth in
following 5 years:
%
37000
10-9
62 000
45000
I
I
-09
5-g
43 000
57 000
-62
2050. Because of the relatively low per capita consumption at the present time,
water demand could increase at a rate well in excess of these rates.
Irrigation and hydropower
The historic growth in irrigation and hydropower in Africa is shown in
Table 2.12.
The annual growth rate has been remarkably constant over this 20-year period
at an average rate of 135 % per annum - about half the population growth rate.;
The hydropower data reflect more upon the extremity of the 1980s drought than
upon the growth of new hydropower capacity.
Conclusion
The data show that the energy generated by hydropower, as well as the area of
land under irrigation, grew weakly through the 1980s and into the 1990s at a rate
well below the rate of growth of population. This trend is likely to continue,
despite strong demand and great potential, so an annual growth rate of 2% is
postulated.
27
EVACUATION OF SEDIMENTS
Table 2.13.
Year
Annual growth in
following 5 years:
1975
1232
19
1980
1350
1985
I
I
Energy:
GWh/yr
I
I
Annual growth in
following 5 years:
235000
49
12
291000
37
1432
15
342000
52
1990
1555
34
437000
32
1995
1820
511 000
The annual growth rate has been reasonably constant over this 20 year period
at an average rate of 197% per annum - well above the population growth rate.
The average growth rate of hydropower over this 20-year pedod is 425%, about
three times the population growth rate, reflecting the strong growth of the
economies of many of the Asian countries in this period. It is unlikely that this
differential is likely to persist in the future.
Conclusion
From the rates of growth for population, irrigation area and hydropower, the
growth rates for demand of storage are postulated and are shown in Table 2.14.
2.5.6. Summary
Based on the above, the forecast future demand for storage is shown in
Table 2.15.
2.6.
2.6.1. Worldwide
The historic rate of constluction of storage worldwide is shown in Table 2.16 and
Figure 2.3.
The overall annual growth rate for the century as a whole has been 65%. It is
interesting to note that neither the two World Wars nor the Depression have made
any serious impact on the rate of growth: in this context the apparent fall in the
rate of construction during the 1990s is dramatic and may in part be due to
inadequate data.
28
SEDIMENTATION IN RESERVOIRS
Table 2.14.
Peliod
2000-2010
20
2010-2020
15
2020-2030
10
Table 2.15.
Reglon
.
2000-2010
2010-2020
I
I
2020-2030
Europe
South and Central America
Africa
Asia and Oceania
Year
1900
51
467
495
167
203
248
281
213
1032
939
315
998
Total
Table 2.16.
49
54
424
I
I
11
67
1910
21
1920
63
1930
121
1940
252
1950
116
6-7
I
I
76
51
414
I
I
112
1960
1196
.\
98
1970
3035
45
1980
4708
1990
5581
2000
5976
I
I
17
0-7
I
I
29
EVACUATION OF SEDIMENTS
6000
12
A
I
5000
CO)
E
.::<!.
CD
4000
E
"0
CD
Ol
I '"
::::l
>
............ ,
"'............"
3000
I
,
.e
til
ta
;e
2000
I
'J
"
\
\
\
\
\
1.
10
1000
,,
,,
,,
.........
.......... .....
2
,
oL-~~~~==~~~~==~--~----~--~__-L__~O
1900
191.0
1920
1940
1930
1950
1960
1970
1980
1990
2000
Year
Figure 2.3.
2.6.2. Distribution
1
1
of construction of storage
The distribution of the growth of new storage is set out in Table 2.17.
2.6.3. Comparison
I
1
Table 2.17.
,1
Europe
North America
South America
Africa
Asia
1950
005
939
617
540
1084
1960
005
739
11,87
3307
1361
1970
0-05
369
824
585
1980
0-05
099
562
072
213
1990
0-05
0-03
159
049
116
2000
30
476
1
1
1
1
SEDIMENTATION IN RESERVOIRS
2.7.
2.7.1. Rate
of /OS5 of storage
R~gion
Estimated
annual loss of
storage due to
sedimentation:
Estimated
reservoir
balf-life:
yrs
North America
020
250
South America
010
500
Northern Europe
020
250
Southern Europe
017
294
Sub-Saharan Africa
023
217
Northern Africa
008
625
China
230
22
Southern Asia
052
96
Central Asia
100
50
South-East Asia
030
167
Pacific rim
027
185
Middle East
150
~"I
~J
31
EVACUATION OF SEDIMENTS
J
The estimates are based on varying quantities of data gathered, as shown in
Table 2.19.
The estimates of annual loss of storage due to sedimentation have been used
in conjunction with the gross storage volume data available in the ICOLD World
register of dams (1998) to estimate the magnitude of the sedimentation problem.
The results of the analysis are displayed in Figure 2.4.
In summary, the analysis shows that by the year 2000 approximately 567 km3
(10% of the current gross available storage in the world) has been lost to
sedimentation. From the data available from the 1325 registered dams under
construction, it can be seen that the average gross storage volume of new
reservoirs is approximately 370 M.m3 Therefore, in order to replace the volume
lost to sedimentation, over 2000 average-sized dams would have to be
constructed around the world.
Total storage
used in
estimating
regional loss:
M.m3
Gross storage
used in
esth'llating
regionalloss:
%
109980
60
1038913
3832
04
Northern Europe
938 168
3067
02
Southern Europe
145 162
24030
165
Sub-Saharan Africa
575352
252 168
438
Northern Africa
188473
181 760
964
China
649322
42804
66
Southern Asia
318602
92712
291
Central Asia
148032
Nil
N/A
South-East Asia
117371
Nil
N/A
Pacific rim
277 124
20192
73
Middle East
223683
9006
40
Global total
6464730
738552
114
-.
Current gross
storage in region:
M.m3
North America
1 844 530
,
i
South America
32
SEDIMENTATION IN RESERVOIRS
7000
6000
5000
C'l
.::
Q.j
01
4000
~
m
3000
;0
~ ::;:f'-_- -'-,~.= =~ ; .; ; J.'- ':-" "L. .-: " -;'. .l:. . -_-.l-: ~ _- L_- -JL. -. _J-. _. . l. -_- -L. ._- L_.- -J
Sedimentation
.-.=..-=
..- ""_1.
1900
Decade
Figure 2.4.
Table 2.20.
Region
North America
South America
I
)
Gross yolunie
in 2000:
Annual
sedimentation:
km3
1845
973
I
I
Ian3
Lost to
sediments:
%
Total storage
loss:
km3
369
79
112
104
25
17
Northern Europe
822
188
68
Southern Europe
135
025
56
574
132
78
32
Northern Aflica
188
015
2-4
China
526
1493
458
Southern Asia
233
166
131
Central Asia
132
1-48
269
035
80
Sub-Saharan Africa
South-East Asia
117
230
31
29
Pacific rim
232
0 75
76
IS
Middle East
199
3 36
277
38
Giobal total
5976
3085
11 8
567
33
EVACUATION OF SEDIMENTS
It is interesting to note that the annual loss of world storage estimated by this
study, 0-48% per annum, is almost half that estimated by Mahmood (1987), who
quotes a figure of 1%. Although it is not clear how Mahmood estimated his
figure, it can be argued that the figure determined by this study may
underestimate the problem. This is further supported by estimates made by other
authors, such as Goldsmith and Rildyard (1984), who estimate the annual loss of
storage in central Europe to be 05%, compared with the estimates for Europe by
this study of 02% and 017% per annum.
2.8.
Table 2.21.
Year
34
Annual sedimentation:
. M.m3/yr
1965
165
003
1980
275
0-05
SEDIMENTATION iN RESERVOIRS
40
35
::::l
c:
c:
...
CiS
30
<V
c..
~
25
1ri
CI
e:!
0
t;
20
'0
Co"
tfl
10S
.Q
(l)
CI
e
ID
10
05
"i"":
0.0 1
0001
ttlt'
0-01
, "
It I
01
10
100
1000
10000
Figure 2.5.
2.9.
Continent
demand
2000-2010
Europe
49
South America
467
Africa
167
35
202
315
349
664
Total
998
567
1565
35
EVACUATION OF SEDIMENTS
1% per annum. At the other extreme, 23 of the reservoirs studied had a storage
volume in excess of 1233 M.m3 and an average rate of loss of storage of 0.16%
per annum.
2.10.
New storage will be required in the future both to satisfy increasing demand
generated by the growing world population and to replace the storage lost due to
sedimentation. The estimate of the gross storage required in the next decade is
given in Table 2.22.
36
earc
In
ue
fa tor
u Ing
INTRODUCTION
This review of the factors which contribute to the efficiency of sediment flushing
operations' for reservoirs is based both on field experience, which has been built
up over the past 30 years or so at dams that have fluspjng systems LTl regular
operation, and on research findings using simulation models, either numerical or
physical, which in tllln rely on fundamental experimental data concelning the
detailed physics of the movement of sediments in water.
Field installations provide information which includes:
the hydrological conditions at the site
" details of the flushing system
0._ the way in which the flushing system is operated
the development and re-erosion of the sediment delta
details of the development of the incised channel during flushing
sediment inputs, throughputs and outputs.
(9
The data are very valuable in looking at the efficiency of flushing operations.
They represent 'real' situations and there are no scale effects or other simulation
deficiencies to mask the findings. However, tt:lere are shortcomings:
there are only a few reservoirs being flushed at present and these do nbfcover
the full range of conditions where flushing might be considered: i.e. the data
forms a sparse matrix
the data are rarely comprehensive enough to cover the complex situation found
in the field
the records are short in terms of sediment deposition and they do not,
therefore, necessarily represent the hue long-term situation
o the historic development of flushing systems started, for economic reasons,
with very modest installations of low capacity - which turned out to be
relatively inefficient. There is thus a dearth of information for installations
with high capacity, more efficient, flushing systems:
Fortunately, there have been major advances in our understanding of sediment
transport by water and this has facilitated the development of numerical
simulation models. It has also facilitated a better understanding of the strong and
weak points in physical model simulation techniques. These models have also
39
EVACUATION OF SEDIMENTS
had the benefit of the field data in order to check their validity. Simulation
techniques can thus be used to:
extend the range of information beyond that covered by field data
systematically look at the importance or sensitivity of individual variables.
This holistic approach, using field data, models and fundamental knowledge of
the sediment transport process, has enabled a clearer understanding of the
requirements for efficient flushing systems to be developed.
3.2.
reservoir is too low to move much sediment. Only close to the flushing outlet
are the flow velocities high enough to erode sediment, and a flushing cone is
formed close to the flushing outlet.
(b) At intermediate water levels, water velocities at the upper end of the reservoir
increase and sediment is transported towards the flushing outlet. There
remains a flushing cone close to the outlet.
(c) When the water level falls to the top of the flushing outlet, scouring velocities
can be generated throughout the length of the reservoir. Retrogressive erosion
of previously deposited sediment occurs.
Shen (1999) concludes that stage three removes far more of the deposited
sediments, thereby regaining storage capacity. He also comments that stage three
'uses more water'. Indeed, stage three is the only realistic scenario for removing
significant quantities of previously deposited sediments from reservoirs.
When flushing is attempted without drawing down water levels, the high flow
velocities at the outlets are very localised and the impact is insignificant. The
water level in a reservoir must be drawn down close to the bed elevation at the
dam before flushing can be effective (Figure 3.la). Many authors have confirmed
this with observation, theory or modelling, including Mahmood (1987), White
and Bettess (1984) and Atkinson (1996), However, moderate lowering of water
levels during flushing will still increase flow velocities significantly at the
upstream end of the reservoir, where bed levels will be above the water level at
the dam (Figure 3.lb). Large sediment volumes will be scoured from these
upstream reaches and will re-deposit nearer the dam. Eventually, bed levels
upstream from the dam will rise to the water level during flushing and then
significant sediment quantities will be transported through the low level outlets
(Figure 3.1c).
Thus, flushing represents an extreme change in reservoir operation. It requires
draw down of the reservoir so that the velocity and volume of the flow are
40
----------
'.
(a)
(c)
Figure 3.1. Longitudinal profiles during flushing: (a) flushing with full drawdol,w1.;
(b) flushing yvith insu:fficient drmvdown; (c) final conditions after a long period of
flushing with insufficient drawdown
41
EVACUATION OF SEDIMENTS
sufficient to scour and remove sediment. This raises many technical, economic
and environmental issues.
The shape of the reservoir may preclude the formation of a scouring channel
that is capable of removing significant quantities of sediment.
River discharges may be insufficient to transport large sediment loads through
the reservoir.
The drawdown of reservoir level reduces the capacity to generate power and
the release of high volumes of water for flushing may also reduce the annual
water yield from the reservoir.
. There is a need to be able to predict ahead so that the flushing operation is not
undertaken if it may jeopardise future power or irrigation supplies.
..... The environmental consequences of passing sediments, that may have been in
, the .reservoir for .s.ome considerable time, to the downstream reach.
-.,
- \
-Ij
-I
!
3.3.
J
J
where Qs is the sediment transporting capacity (tis) of the flow in the incised
flushing channel, n is 86400 (seconds per day), Tf is the duration of flushing
(days), N is the interval between flushing operations (yrs), Min is the sediment
inflow rate (tiyr) and TE is the trapping efficiency of the reservoir.
42
J.
Efficiency of flushing
More generally, the non-equilibrium state of sedimentation in a reservoir can be
expressed as a sediment balance ratio SBR, the ratio of sediment flushed to
sediment depositing, which can be defined as:
(2)
The transporting capacity Qs' (tis) will be a function of discharge, channel
roughness, width and slope, and the properties of the deposited material. A low
sediment balance ratio indicates low flushing efficiency and the continued build
up of deposited s~~iments in the reservoir. A value of SBR> 1-0 indicates high
efficiency and the lpng -term stability of sediments within the reservoir.
(3)
43
EVACUATION OF SEDIMENTS
The coefficient, '1', varies with particle size, the larger the sediment size the
smaller the value of W. It follows, therefore, that fine sediments are more easily
flushed from reservoirs than coarser sediments.
2. The longitudinal energy gradient, S.
This is a parameter which depends on the degree of draw down at the dam
during flushing. The maximum energy gradient is obtained when the drawdown
is maximum and under these conditions it approximates to the slope of the
original river bed prior to impoundment. It is a parameter which can be
controlled by the choice of the amount of drawdown and which influences the
efficiency of flushing operations because of its effect on sediment transport rates,
see Equation (3)~
3. The bed width of the incised channel, W.
Equation (3) requires the bed width of the incised channel, W, to be input.
Channels formed by flushing in reservoir sediment deposits correlate well with
flushing discharge. Figure 3.2 shows the relationship and the data from which it
was derived. The fitted line is described by the equation (in SI units):
_.0
(4)
In some cases, channel bed widths may be constrained by the reservoir width:
In general, though, the width of the incised channel is determined by the flow and
is independent of sediment size.
Vl't = 128 0.0.5
1000
Sanmenxia
Guanting
@ Guernsey
Baira
EB
.s=
:0
.~
1i5
c:
c:
ctS
..c:
()
100
"'0
<l.'l
(/)
.0
10~--~~~~~~--~--~~~~~--~~~~~~----------
10
Flushing discharge,
Figure 3.2.
44
1000
100
0.: m3/s
Ii
the need for the operation in the ~ght of the success of previous years'
operations
hydrological forecasts for the follovving months
likely demands for power and irrigation water supplies for the following
months, taking into account alternative sources of supply.
45
EVACUATION OF SEDIMENTS
of the inputs to repeat predictions. For example, a low SBR may imply flushing
should be performed at a time of higher discharge and a high SBR may imply that
flatter slopes can be expected upstream from the dam (as shown on Figure 3.lc).
If a value of SBR well above 1 cannot be achieved, then flushing is not feasible.
In the comparison with data presented in Table 3.1, SBR was computed for a
slope defined as the drawdown water surface elevation below maximum water
surface elevation divided by reservoir length.
Table 3.1. Application of sediment balance and long-term capacity ratios to existing
reservoirs
Reservoir
. Initial
. capacity:
M_m3
Country
Sediment
balance ratio
(SBR)
Estimated
from
reservoir
surveys
Calculated
Calculated
Baira
India
9-6
085
0-85
Gebidem
Switzerland
9-0
Approx_
1-00
0-99
Gmund
Austria
0-93
0-86
098
21
Hengshan
China
13-3
075
0-77
Palagnedra
Switzerland
5-5
100
1-00
33
Santo Domingo
Venezuela
30
0-97
1-00
11
I
i
China
2270
Low
0-20
0-2
Guernsey
USA
91
Low
0-26
10
Heisonglin
China
86
023-0-35
030
Approx_
0-70
Ichari
India
116
Approx.
035
0-36
Ouchi-Kurgan
Former
USSR
56
Low
Approx.
China
9640
Sanmexia
010
031
0-39
3-4
46
Sefid-Rud
Iran
Shu1caozi
China
1760
<026
013
96
Low
039
46
In the list above only the side slope steepness is not known. In wellconsolidated sediments, near vertical channel sides can occur, while slopes as
low as 25% have been observed for poorly consolidated material. Therefore, a
technique to predict this slope is vital to a reliable prediction of sustainable
capacity.
Atkinson (1998) recommends, with some reservations, the use of the
following expression:
,..,.,Ian a. = 06'"1
. .Jp4,7
(5)
'where CL is the angle of the side slope (zero is horizontal) and p is the density of
3
the deposits expressed as weight of dry material per unit volume (tlm ). p ca.TJ. be
47
EVACUATION OF SEDIMENTS
. ---I
-I
I
1
-II
1750
E
C
~I
>
Q)
Q5
~l
"C
Q)
a:l
1745
I
1
-I
1740~----------------~-------------------L--------------__~
100
200
300
(a)
-\
(Before flushing)
Q5
Q)
-i
320
~
~
"C
325
E
C
-I
-I
315
310
r)
~I
Aug. 1960
(at end of
construction period)
305
0
1000
2000
3000
4000
Lateral distance: m
(b)
i
1
~.
predicted from the composition and age of the deposits using Lane and Koelzer's
(1953) method. Atkinson (1998) found that slopes computed using Equation (5)
could be in enor by as much as a factor of ten and clearly this method needs to
be treated with caution.
A simple criterion for assessing sustainable reservoir capacity can be
developed by fitting a simplified reservoir shape as shown in Figure 3.4. A cross
section just upstream from the dam can be taken as representative of the entire
reservoir, and then the area of the trapezoidal flushed section can be compared to
the original cross-section area. The ratio of these areas then gives a long-term
capacity ratio (LTCR), which is an estimate of the reservoir capacity that can be
sustained in the long term by flushing.
----~.
: ----------------------------------~----------------------------
48
(a)
(b)
~ River channel
.-/-
Dam
------
Section
Section
"'C7
""=7'
(e)
""':;:-------or---------,-----""""7'
~\ B_~
__
I I
lA'
),t
,-- - - - - - - - -
--It- .
"I"
Bed width,
WbO!
Area B
Area A plus Area 8
49
EVACUATION OF SEDIMENTS
If the long-term capacity ratio, LTCR, is greater than 05 then flushing is likely
to be successful in terms of maintaining live storage in the reservoir and is likely
to be economic if the shortfall in generating capacity during the flushing period
is not too severe.
3.3.3. Evaluation
of flushing criteria
The sediment balance ratio, SBR, is a measure of the propol1ion of the incoming
sediments which may be flushed from the reservoir. The long-term capacity ratio,
LTCR, is a measure of the proportion of the initial storage capacity which may
be retained by flushing from the reservoir.
Some of the factors which determine the values of the sediment balance ratio,
SBR, and the long-term capacity ratio, LTCR, are inherent characteristics of the
site. These include:
the shape and size of the reservoir
the imposed hydrological conditions
the imposed sediment inputs.
Some of the factors are controllable. These include:
the operation of the reservoir between flushing operations
the design of the flushing system, including elevation and capacity
/I the operation of the flushing system, including discharge and duration.
A review of the literature on reservoir flushing, produced information from 14
reservoirs where flushing had been attempted and where sufficient data were
available to test the criteria (Atkinson, 1996). The 14 reservoirs can be divided
into two categories: six where observations indicated that flushing would sustain
a long-term capacity in excess of half the original capacity, and eight where it
would fail to do so. Table 3.1 presents the results of the application of the two
assessment criteria. The criteria performed very well in distinguishing between
the six reservoirs where flushing was successful and the eight where it was not.
The predicted LTCR also proved to be a good indicator of the long-term
capacities that were estimated from the observations. The sediment balance ratio,
SBR, was, in most cases, not a constraint to successful flushing.
50
r'"
It
\Aihere Hfjush is the height of water at the dam during flushing and Hma,'( is the
maximum height of water at the dam (to reservoir retention level), both measured
above original bed-level. DDR less than about 07 indicates some degree of
constraint owing to insufficient drawdown.
51
EVACUATION OF SEDIMENTS
of the sediment balance ratio is designated SBR d and is calculated using the
original river slope, that is for conditions of full drawdown. SBR d < 10 indicates
a constraint due to the inadequate capacity to flush sediment.
The amount of sediment removed during the flushing period depends, for a
particular reservoir, on the flushing flow and the duration of flushing. The relative
importance of discharge and duration are illustrated in the following example.
By using Equation (4) to determine flushing channel widths and then
substituting in Equation (3), this yields values of sediment discharge rates, Qs'
The total quantities of sediment removed during the flushing period are then
obtained by considering the duration of flushing. Table 3.2 shows an example of
the results from this type of analysis. It is based on a notional reservoir with a bed
slope during flushing of 00006 and a flushing water volume of 864 x 109 m3 It
is assumed that the reservoir contains 01 mm sand and hence the coefficient in
Equation (3) is 100 (Atkinson, 1998).
If there is a restriction on the quantity of water available for flushing, it is
clearly better to use a high discharge for a short period than a low discharge for
a long period. This increases the amount of sediment that will be removed. The
penalties of extended flushing periods are considerable, quite apart from the
Table 3.2. The relative importance of the discharge used for flushing and the duration
of flushing for a particular volume offlushing water
Flushing
discharge:
m 3/s
Flushing
duration:
days
Mean sediment
concentration:
ppm
Sediment
removed per
day:
Mt
Sediment removed
during flushing
period:
Mt
500
200
9600/5760*
0411025* .
82/50*
1000
100
14550
126
126
1500
67
18560
241
160
2000
50
22060
381
191
2500
40
25220
545
218
3000
33
28130
729
243
3500
285
30860
933
267
4000
25
33430
1155
289
4500
222
35880
1395
5000
20
38220
1651
330
5500
182
40470
1923
350
6000
167
42640
2210
369
I
I
I
52
310
FvVR
=VV/W
(7)
bed
Reservoirs that. are too wide for the flushing channel to reach tie perimeter
The natural top width of the flushing channel may be less tha..TJ. the representative
top width of the reservoir and under these circumstances accumulated sediments
vviU remain along the perimeter of the incised channel forming a high level
terrace.
If the top width of the section scoured by L.~e flushing channel is not restricted
by the reservoir sides, then the top width of the flushing channel is a constraint.
top width ratio is defined as:
TVVR
..
(8)
where vVmin is the lesser of Wand }Vbed and V/top is a representative top width in'
the reservoir. If the reservoir can be approximated to the shape shown in Figure
3.4, the representative top width should be taken as that which occurs just
upstream of the dam. TVVR> 10 is required.
53
EVACUATION OF SEDIMENTS
Table 3.3. The influence of sediment size on the amount of sediment removed from,
reservoirs
Sediment removed during the flushing period: Mt
Flushing conditions
Sediments with a
mean size less
than 01 mm
Sediments with a
mean size greater
than 0-1 mm
Conditions
with low
flushing flow
N/A
N/A
N/A
50
1750
740
330
N/A
m3/s
Duration:
days
Loess
sediments
500
200
20
Flow:
5000
By taking the example given in Table 3.2 and by considering the specific
combinations of flushing discharges of 500 m3/s and 5000 m3/s for 200 days and
20 days respectively, the effects of sediment size are as given in Table 3.3.
The sizes of the sediments deposited in reservoirs are an important factor in
deciding whether flushing will be effective.
The effect of widely graded sediments
In many rivers there is a mixture of sand and gravel in the bed material. In these
circumstances, there may be a fairly small proportion of the bed material
consisting of fine sand, but owing to its high transportability, a large proportion
of the material being transported by the river is fine sand. When an impoundment
is introduced to such a river, the material depositing is dominated by fine sand
together with finer cohesive material that is transported in the river as wash load.
These deposits are relatively mobile compared with the general river-bed
sediments and hence are amenable to flushing. A parameter that can identify
these favourable circumstances for achieving a sediment balance by flushing is
the sediment size ratio, SSR:
(9)
Table 3.4 presents the relationship between this ratio and the number of days
of flushing required annUally. It is assumed that flushing is performed at a
discharge of twice the mean annual flow, as recommended above. Full drawdown
of water levels is also assumed and the calculations were performed for a series
of assumed sediment size ratios, SSR. In preparing the table, the updated Ackers
and White sediment transport predictor, Ackers (1993), was used and the riverbed material was divided into ten fractions. The equation derived by Tsinghua
University and reported by IRTCES (1985) could not be used as it cannot be
applied to sediment transport rates in rivers.
There is no universal relationship between the sediment size ratio, SSR, and
the number of days of flushing required annually because the reservoirs
compared are of different sizes. If a general rule is to be made, then it may be
suggested that rivers where the SSR is less than about 0-03 are generally suited
for flushing.
54
It should be noted that flushing has been proposed at the Tarbela reservoir, but
not, to our knowledge, at the other two reservoirs. The Tarbela and Tungabhadra
reservoirs are large (>3000 M.m3), while the planned Rooiport reservoir is about
800 M.m3
The parameters required to determine the SSR can be derived as follows.
DSOT: the 50 percentile. size of the sediment in transport in the river can be
obtained, if possible, from sediment sampling during periods of high
river discharge or (if the reservoir has been constructed) from samples
taken from deposits. Otherwise values can be obtained either from
estimates derived from other rivers in the region or by prediction using
the bed material grading .
Dsos: the 50 percentile size of the river-bed material can be obtained from
Tarbela,
Pakistan
Tungabhadra,
India
Rooiport.
South Africa
(> 100)
(> 100)
(> 100)
04
62
94
(> 100)
0-2
38
58
84
0-1
23
30
58
0-06
15
Ii
21
46
0-05
13
19
43
0-04
11
17
40
15
36
0-03
9 .
55
EVACUATION OF SEDIMENTS
The sediment sizes will also affect where in the reservoir the material will
settle. Some of the silt, the sand and the coarser material tend to deposit in a delta
at the upstream end of a reservoir, while the finer silts and the clay can deposit
throughout a reservoir. After a period of flushing an incised channel will be .
formed in the deposits in the delta, which will quickly refill with incoming sand
and coarser sediments. Thus, most of this coarser material will be flushed from
the reservoir when the incised channel reforn1s during the subsequent flushing
operation. In contrast, the finer deposits formed nearer the dam will be
distributed across the reservoir, and so will be only partially removed by the
flushing of an incised channel. The impact of this process will be to extend
reservoir life at sites with less fine sediment, even when the long-term capacity
achieved by flushing is quite smalL
Table 3.5 quantifies this. effect. It gives predictions of increases in reservoir life
due to flushing for various values of the proportion of sand and coarser materials,
P sand ' and for flushing discharge at the three reservoirs listed in the Table 3.4. Psand
is defined as the proportion of the liver sediment load that consists of sand and
coarser material. In each case, the following assumptions were made:
there is a single flushing period of 30 days annually
sediment inflow to the reservoir for all material (wash load and bed material
load) can be described by the simple relationship:
Concentration = Constant x Discharge1.2
Table 3.5. Effect of the proportion of sand and coarser material on extending the life
of reservoirs
PropOition
of coarse
sediment
Qllushin/ Qmean
I
02
04
Tarbela
Tungabhadra
19
14
25
18
Rooiport
1-4
18
!
I
I
06
38
27
27
08
76
5-4
55
09
152
108
110
02
27
16
16
04
36
21
21
06
54
32
31
109
63
62
217
127
125
08
09
56
Sensitivity to the exponent in this equation was slight (about 5% when the
exponent \-vas doubled to 2.4)
the silt deposits downstream from the main sedimentation delta (this is a
conservative assumption, if a proportion of the silt is known to deposit in the
delta then that proportion can be included in P sand)
a sediment balance is achievable.
The analysis technique outlined in Atkinson (1998) was used and the assumed
proportion of sand and coarser material, P sand ' was varied from 02 to 09 in each
case.
These results indicate that where a large proportion of the material deposits in
the delta, say P sand >O8, then flushing for 30 days annually can greatly extend
reservoir life. This would apply even at sites where flushing does not produce an
acceptable reservoir volume in the very long term. Sensitivity to other flushing
periods "vas found to be slight, for example reducing the period to 10 days only
reduced the factor by which reservoir life is extended by between 2% and 120/0.
(It has been assumed that a sediment balance can still be achieved with L.ie
reduced period of flushing.)
Summarising both the analyses presented in this section provides the
following conclusions .
., The sediment sizes in transport in the river can be of paramount importance to
the success of flushing in a reservoir.
From the point of view of achieving a sediment balance, a large factor is
required benveen the sediment sizes being transported in the river and the sizes
found in the river-bed material. Such conditions are typical for gravel rivers
with a widely-varying bed material composition.
If a sediment balance can be assured, then a predominance of fine sand, and
other material that deposits in the delta at the head of a reservoir, ensures that
flushing greatly extends reservoir life.
$
Therefore, from the point of view of sediment size alone, delta deposits of fine
sand and coarse silt are the most likely to produce success in flushing a
reservoir. Coarser material may inhibit a sediment balance arid finer material
will deposit in the body of the reservoir outside any incised channel and so will
not be available for reworking during flushing.
Evaluation
57
EVACUATION OF SEDIMENTS
Table 3.6.
SBRd value
FWR value
TWR value
India
068
24
3-4
16
Gebidem
Switzerland
093
20
67
15
Gmtind
Austria
089
58
5-2
13
China
077
Approx.4
0,1
71
Palagnedra
Switzerland
100
33
14
10
Santo Domingo
Venezuela
1-00
11
14
18
004
05
Reservoir
Country
Hengshan
Guanting
China
081
03
Guernsey
USA
044
3-2
14
026
Helsonglin
China
0'77
Approx.1
006
08
Ichari
India
031
33
99
14
Ouchi-Kurgan
Former USSR
014
110
Approx.2
Approx.03
Sanmenxia
China
075
48
026
09
Sefid-Rud
Iran
0'96
43
03
0-1
Shulcaozi
China
037
15
10
2-1
The results in the second part of Table 3.6 indicate that at two reservoirs, Ichari
and Shuicaozi, changes to the outlet structures at the dam could potentially
remove all constraints to successful flushing, while at the other reservoirs, site
conditions constrain the success of flushing.
3.4.
58
level must be held constant at as Iowa level as possible throughout the flushing
period. To achieve this:
ED
Flushing discharges of twice the mean annual flow are recommended and the
quantity of water required for flushing is unlikely to be less than 10% of the mean
annual run-off. This is based on worldwide experience from reservoirs which are
being flushed on a regular basis, see Atkinson (1996, 1998), Basson and
Rooseboom (1997a and 1997b) and Mahmood (1987), together with detailed
numerical modelling of proposed flushing systems, see Attewill et al. (1998) for
example. Note:
flushing discharges of at least twice the mean annual flow are required
e flushing volumes of at least 10%
the mean annual run-off should be
anticipated.
Graded bed sediments produce conditions which are the most conducive to the
efficient fiusr.ting of sediments. Such conditions are typical of gravel rivers
with a varying bed material composition. In large rivers this situation is found
where the longitudinal bed gradient is between, say, 0001 and 0002. In
smaller rivers the eauivalent
range
may
between 0002 to 0005.
1.
...
59
EVACUATION OF SEDIMENTS
.. From the point of view of sediment size alone, delta deposits of fine sand and
coarse silt are the most easily flushed. Coarser material is difficult to move and
tends to deposit at the upstream end of the reservoir. Finer material which
deposits in the body of the reservoir outside any incised channel will not be
available for reworking during flushing.
J
J
L()ng, relatively nan"ow, reservoirs are better suited to flushing than short,
wide, shallow reservoirs.
The greater the sustainable live storage the more attractive flushing systems
become. Subject to the economic circumstances, flushing systems will
normally be worthwhile if a sustainable live storage of more than, say, 35% of
the original live storage can be achieved by flushing.
A full economic analysis covering the whole life costs and benefits of the
flushing system should be undertaken.
.
3.4.7. Summary
Reservoirs in the 'upper and middle reaches' of rivers are likely to be best suited
to sediment flushing for the following reasons.
60
3.5.
In the lower reaches, reservoirs are likely to have inundated areas that have
previously been flood plains and these areas would not be reached by the
incised flushing channel which is inevitably of limited width.
The longitudinal slope available for the flushing channel is relatively small,
thus limiting the amount of sediment transport.
Reservoir volumes in the lower reaches are likely to be larger compared with
the mean annual run-off and hence water availability becomes a restraint on
sediment flushing.
NUMERICAL MODELS
The previous section described the factors which influence the efficiency of
. sediment fiuslling and gave guidance on some of the hydrological and design
parameters which need to be satisfied.
Detailed analysis of specific sites requires the use of numerical models that can
provide much firmer estimates of flushing performance. Numerical models can
take into account many details that are precluded from the simpler desk
calculation techniques. These include:
e
4D
61
EVACUATION OF SEDIMENTS
J
1
1
1
1
1
1
1
1
1
J~
I
62
1
1
I
I WI
e Imen
e
fit
len e
G
..
flu hing
4. Worldwide experience
4. J.
of sediment flushing
INTRODUCTION
The main purpose of this chapter is to answer the following questions:
how many reservoirs are being flushed?
where are they?
are they used for water supply (potable and irrigation) hydropower or flood
detention?
., what methods are employed for flushing (the facilities and the operational
regime)?
Q
what is the nature of the sediment?
how successful is the flushing?
iii what constraints (operational, economic and environmental) affect the flushing
sy'stems?
/I what downstream effects occur?
e
The answers to these questions will allow general findings to be made with
regard to the success (or failure) of current flushing operations, the factors that
influence the outcomes, and will provide an indication of the trends in the
designs for flushing systems.
The principal method of appraising worldwide experienGe_has been a
comprehensive literature review, drawing on the reference lists of previous
reviewers and including fresh searches of library references available orieD and
the Internet. Appendix 3 contains descriptions of a number of case studies, where
sufficient information has emerged from the literature searches. For each case
study, the history and physical features of the reservoir are described, the
sedimentation evidence reviewed and the flushing measures which have been
implemented are described.
In the overall stock of dams worldwide (over 40 000 with dams higher than
15 m according to Morris and Fan, 1997), flushing, in one form or another, must
have been attempted in many hundreds, probably th"Ousands of dams.
Unfortunately, the amount of accessible documentary evidence amounts to only
about 50 cases, with substantial quantitative and qualitative data readily available
for only about half of these.
Inevitably, the degree of science applied to the design and execution of the
flushing process must have varied considerably, while the degree of success
would depend on factors such as:
65
EVACUATION OF SEDrMENTS
-
.j
whether the reservoir and dam were designed taking account .of local
sedimentation data, with facilities to enable flushing to take place
the fundamental suitability of the reservoir and dam for undertaking successful
flushing
the degree of operational flexibility to allow an effective regime of sediment
fl ushing to be undertaken as needed
the application of sufficient know ledge and experience to allow the optimum
flushing regime to be developed.
4.2.
FLUSHING
J
r
J
J
66
-{
J.
4.3.
4.3. I. Overview
Table 4.1 lists the reservoirs for which the literature search has revealed evidence
of flushing, although in a few cases it appears that SOlne form of sediment routing
(sluicing or density current venting) may be the major method of sediment
discharge. Excluding those cases where there is no hard evidence of flushing,
leaves 50 cases, for which the locations and purposes are summarised in Table
4.2. In many cases, the purpose is not provided and in some cases multiple
purposes apply, so that the total numbers are not equal to the sums of the
purposes. (In a few instances where the reference cites two or three reservoirs
or parallel for flushing, only a single case is included in this
operated in
table.)
By far the greatest number of examples is in China, but this is not surprising,
because of the size of the country, the numbers of reservoirs (18 800 dams higher
than 15 ffi, according to the 1998 ICOLD vVorld register of large dams), and the
high sediment yield, particularly in the basin of the Yellow River. It is notable,
however, that 42% of the reservoirs listed in Table 4.1 in China, which contains
52% of the dams higher than 15 m and 30% of those higher than 30 m (ICOLD,
1988).
"". . . ,............. , it may also be noted that the majority of the examples are from
........ ,..".............. . ., with high sediment yields.
Attempts to relate L;e number of examples of flushing in different countries to
their stock of large dams and their typical sediment yields would not be fruitful,
because of the relatively small sample sizes in most cases, together with a
number of other factors which come into play, such as:
available to those countries for
t.1.e resources - fin~ncial and technical
researching and dealing with sedimentation problems
policies for open dissemination of the lessons learned from sedimentation and
flushing experience
11
the financial resources to allow attendance at international conferences
Of those flushed reservoirs for which the purpose is k..l1own, Table 4.3 lists the
numbers falling into each purpose or combination of purposes.
67
EVACUATION OF SEDIMENTS
Table 4.1.
Reservoir/dam
Baira
India
Barenburg
Switzerland
Bajiazui
China
Cach!
Chiyu
Dalingkou
I China
""0
::lE
3 .~
'" ,t:tl
J.6
F
F
M&F(l997)
China
Fergoug
Algelia
Ferrera
Switzerland
-.
F
F
1968
2)
1945
07
F :
Guanting
Guernsey
Heisonglin
Hengshan
~
China
China
Honglingjin
China
China
FWH
1953
182
HI
1927
IF
IF
I
1959
43
61
FRD
1966
84
1960
38
F
R
1975
02
76
1938
30?
=1
Sudan
HIW
1964
Kunda Pal am
India
Liujixia
China
Jensanpei
Taiwan
!China
Jiaojiazhuang
Khashm El Girba
F
D
FR
La
M&F (1997)
I
141-435
China
IF
1974
84
Naodehai
China
IF
1942
63
Nebeur
Thnisia
Ouchi-Kurgan
1961
04
PaJagnedra
Former USSR
Switzerland
PD
FR
1952
28
Prieto
Puerto Rico
Rioni
Former USSR
Sakura
Japan
60
1956
1960
USA
Sanmenxia
China
Sanshenggong
Santa Maria
Santo Domingo
China
Guatemala
Venezuela
1974
0-7
Sefid-Rud
Iran
HI
1962
35
Shiaodaokuo
China
Shimalin
China
Shuicaozi
China
Warsak
Yanouxia
Pakistan
China
F
F
F
F
9
M
Lo
M&F(1997)
M&F(l997)
ht
Yoon, 1992
i M&F(l997)
Yes M&F(1997)
M&F(l997)
I
F
F
F
L
N
FM
LoP I
8-1]
15-45
La
HI
1960
08
F
F
68
Zemo-Afchar
Former USSR
! 1927
Zhenziliang
China
1958
Drawdown
Full
Partial
M&F(1997)
Yes K&C (1979)
Yes M&F(l997)
M&F(l997)
M&F(1997)
F
1958 I 19
Qian (1982)
Yes UNESCO (1985)
F
H
Intake forebay
M&F(l997)
F
FRD II F
22
2 in series
M&F(1997)
San Gabriel
DF
Yes M&F(1997)
F
F
HI
H
LI
F
P
Nanqin
Mode
Flushing
Density current
Routing/sluicing
Uncontrolled
5880
FIFI
USA
1924
(1982)
Yes B&P(1986)
I
Morris
1954
i SNCOLD
Yes M&F(l997)
Puerto Rico
M&F(l997)
M&F(1997)
La
F
F
FR
M&F(l997)
Yes M&P(l997)
Yes i R&S (1982)
11
New Zealand
Qian (1982)
38
17-21
M&F(l997)
La
Mangahao
Hydropower
Irrigation
Flood control
Water supply
Multipurpose
F iF
F
F
Loiza (Carraizo)
Purpose
FD
1953
i Algeria
12-18
I
Parallel resrs
M&F(1997)
India
Iril Emda
EI
F
H
M&F(1997)
M&F(1997)
China
, Austria
Hongqi
Icbmi
F
1985
M&F(l997)
Yes M&F(I997)
, M&F(l997)
25
M&F(1997)
China
China
Comment
Principal
reference
Yes J&K(1984)
Donfanghong
Grimsel
:l~
~.g
E
Dashikau
Guanshan
,I
~~
17
Dashidaira
Groiind
jL
E-
'00:
c c
!:
;::
0
01
1966
China
' China
Genshanpei
68
Costa Rica
Chirurt
1981
""0
.~
~0
..,
~~
.;:.'"
c~
<.)0-
Sediment removal
~
~
I:l..
Country
P
P
23-83
592
10-67
I
Season
Flood
Early flood
Late flood
Non-flood
Yes IWHR(1983)
Yes Mahmood (1987)
M&F(1997)
I
Yes UNESCO (1985)
Yes Zhang et (d. (1976)
Enhancements
Mechanical
Lateral channels
Longitudinal channels
Piping to induce lateral erosion
Fluctuating pool level
3 in series
~.
Table 4.2.
Country
Hydropower
Irrigation or
water supply
Flood
control
Algeria
Austria
China
Costa Rica
Former USSR
Guatemala
India
I
II
Iran
Total
21
15
2
1
l
Japan
Unknown
New Zealand
Pakistan
Puerto Rico
Sudan
Switzerland
I
1
1
1
1
1
2
1
Taiwan
Tunisia
US A
Venezuela
Totals
19
I
,
11
25
50
cases, five were considered to involve a degree of sluicing and in five densitycurrent venting was considered to be an important contributor to sediment
removal.
Table 4.4 Slll1lll1arises worldwide evidence regarding the flushing techniques,
covering whether the drawdown is partial or complete, the season when it is
undertaken and whether it is enhanced by techniques such as lateral channels or
a fluctuating pool level. The total sample number for the amount of drawdown
and the flushing season is taken as the number flushed in Table 4.1. In all cases
where a flushing season is given, it is also stated whether the drawdown is full
69
EVACUATION OF SEDIMENTS
Table 4.3.
Flood control
Multipurpose
25
Total
Table 4.4.
Flushing season
Enhancements
Full
21
Early flood
Mechanical
Partial
Lateral channels
Late flood
3 . Longitudinal channels
Non-flood
Not stated
36
Not stated
22
Lateral piping
Fluctuating pool
Total sample
50
Total sample
I 50 I Total number*
12
70
by flushing are silts and fine sands. A proportion of finer materials nlay either be
discharged with the water passing through the impoundment under normal
operations~ or may be discharged as a density current.
Coarser sands, gravels and cobbles are likely to be deposited in the upstream
part of the reservoir basin. Given suitable flushing (or sluicing) conditions, these
can be drawn down into the lower part of the impoundment and ultimately
discharged downstream, but this tends to be a longer-term process, generally
associated with a permanent rise in the form of a delta and braided channel at the
upstream end of the basin and with associated permanent loss of reservoir storage
capacity.
4.4.
4.4.1. Summary
Over 20 case studies are included in Appendix 3. These appear in chronological
order of constnlction completion or first impounding, in order, to some extent, to
illustrate the development of knowledge in sediment problems and remedial
measures, including flusriing. The salient features and key findings from the case
sIDdies are given below, in the same sequence. Table 4.5 summarises the mrun
descriptive and quantitative information on these and a few other reservoirs.
Table 4.6 presents the main quantitative infonnation for the case studies only,
including the key ratios concerning reservoir volume, annual inflow and sediment
load and flushing discharges and volumes, where available. Also included in
Table 4.6 is a subjective assessment of whether flusPing has been successful,
'which is discussed furt.~er below.
JiJangahoo reservoir (New Zealand, 1924)
This example suggests that sedimentation was a consideration in the design, but
that no specific planning for sediment flushing was included, because it was
expected that more dams were to be built upstream. Fortunately, a generous lowlevel outlet could be made available by recommissioning gates on the diversion
tunnel, which had remained out of use after problems 25 years earlier. This was
successful in removing a large proportion of the accumulated sediment, after
which annual flushing has been undertaken.
71
Table 4.5.
Reservation/dam
Country
Purpose
Year built or
sll\rlcd
impounding
Year
modified
I Calchmenl
Max depth: m
Surface
area: km'
i area: km'
Original
capacity:
M.m'
Basin
1enlllb:km
1990
AM",'
sediment
inHow:Mt
India
Baita
Cachi
Costa Rica
Gebidem
Switzcrland
Hydropower
1981
Hydropower
1966
Hydropower
1968
5] (dam height)
35 (diversion
lunnel)
1996
785
69
200
lB
4-1
I
324
14
24
.0.3
(Atkinson,
1996)
(Atkinson,
1996)3500
(J984ref)
54
1500 (appro~.)
9'()
420
()'81
05
I
Gmund
Austria
Hydropower
1945
Gaunting
China
Flood control
1953
1967
1563
311
0J24
094
093
43400
43
229
30
2270
135
02107
1250
73 (19505)
7 (19805)
Yes
13 (B&P
I
USA
Guernsey
Heisonglin
Chin.
lITigation &
hydropower
1927
1959
42000
29
96 (1963)
24
91
value)
J.7
2100
No
(1927-57)
370
30
29
86
0-70
142
Hengshan
China
1966
163
65
I'()
133
158
118
Honglingjin
China
Irrigalion (probnbly)
Icbari
India
Hydropower
Jensanpei
Taiwan
1938
Khashm EI Gima
Sudan
lITigation, hydropower
1964
I
!
1960
1364
<42
16-6
432
11-6
5300
077
I
1975
37
1955
II
70
106
950
84
Loiza (Carraizo)
Pucno Rico
Water supply
1954
Mangahao
New Zealand
Hydropower
1924
534
23
27
27
449
038
I
!
Chillll
Nan'lin
Naodehai
\ China
1974
Originally flood
control. lanerly also
1942
453
1970
4-5
29
4501
121
069 or
0531
168
265
16
102
Yes
I irrigation
35
17
564
IS 000
13
138
55
26
55
3041 (199 in
SINCOLD
1982)
008
1966-71 &
1990
688400
55
120
9640
43000
1600
None
427
65
()'2
Yes, apparently
56200
50
Ouchi-Kurgan
FonncrUSSR
Hydropower &
irrigation
1961
Plllagnedru
Swil.7.crland
Hydropower
1952
1974
Sanmenxia
China
Muhipurpose .
1960
(m340)
Venezuela
Sanlo Domingo
Hydropower
1974
1962
Hydropower
1958
! Irrig.tion&
1960
l'()
3'()
450
82
25
1760
5000
28
"
96
514
()'63
42
170
21 100
(1961-70)
153
Apparently nol
6600
Approx.5
53-8
342
Approx.02 -[
Sefid-Rud
lImn
Shuicaozi
China
Pakistan
Warsak
67340
hydropower
lemo-Akhar
Fonner USSR
Hydropower
1927
Zilcnziliang
China
lrri~li"n
1958
(prohably)
18
<14
1740
I<
36-6
S~"\Iillh"11t:nit1R
Mudd studics'!
e.\pcricm:e
Rcfcrcm:cs
C'lS&!
LTCR:
Slud i l!~"
Mt::m mit:
~ri~inall\'
interrupt ion lO
thcrt!aftc:"
mCf':lhs
nUW! thtl~uyh
wiltKJut deposition.
Yes. :t((c:rrm.'hlern~
futlnd
54~
A6.1
DETR
~5
Cil 19
DETR
da.rifyin~
h..:C:lU~C
of gOl".;e:lyre
Phy!\il:al
Yc.,
Ao.3
DETR
A7.1
DETR
20
J.r~"Cki
A7.2
DETR
26
A7.J
CIt25
DETR
AM
I
I
77
DETR
SOLH't.~)
A7A
DETR
DETR
( '.~,!):
99
rut
A6.!
Ch~1
DETR
36
SLOr:J~t:
loss
~26
year
I
Cap~ity
seriously de~ie1l!d
(I'1~7)
lCJ5J-~~~
:.hrcc {lOU mm
lowlcv~J
~...'!!im"nt
DETR
Ch2Q
OETR
.~cdimc:H.:ltk'll':;
CilPllCity reduced
LO iiOOUL
50% hy ! 950.
bUl
:ccuvcr:t.! to
p~ing :lbout
DETR
I
I
19i7;
.
Lie::hi & Hacoerli (I 97O'J: SNCOLD
(1996)
119~2):
AtKinscn
A7.5
DETR
10
A6.3
OETR
100
11.7.6
Ch14
DETR
39
A6.6
100
I
osa M.m depo;'.e!! in two years 1975-78: on "'I.m' in
Yes
rmp!:::,'!.:n:et.!
(1~79);
Atkinson (1996)
DETR
A7.7
DETR
(WeHR i
!9~~):
IRTCES
A'kin,on ( 1996)
I
I
A7.Pt
DETR
DETR
right b.u:k. lc:lding to puw'.:r
int~~.;s
43 M.m.' dcp0:iitl"C.I
r.1vUl!. rcp(!!..:cntin~
p~r
I
:
13
ChD
Zh:;.ng
t:(
tTl.
!!~76j
DETR
{Ii.j~j,;
II
39
~~
()
~
Table 4.6.
Reservoir
Year
Capacity:
M.m3
;;:.,'
-~
Mean
annual flow:
M.m 3
Mean or
median
sediment
inflow: Mt
Mangahao
Guernsey
1927
Zemo-Afchar
1927
Jensanpei
1938
81
Naodehai
1942
168
265
16
Gmund
1945
093
135
007
Palagnedra
1952
55
199
Guanting
1953
2270
Shuicaozi
1958
Heisonglin
91
Flushing
discharge:
m 3/s
Flushing
volume:
M.m3
Ratios: %
CIl
SC:
tlm3
VII
Q/I
2100
17
125
54
6600
450
30
43
19
188
26
215
05
Atkinson
(1996)
95
07
75
008
28
15
1250
13
182
06
96
514
063
50
19
66
307
1959
86
142
071
10
61
83
2221
Sanmenxia
1960
9640
43000
1600
22
Warsak
1960
170
21 100
153
Ouchi-Kurgan
1961
564
15000
13
-~:
-1. - --'
25
045
432
2800
-l - j -
-- - L.~
583
Current
or
recent
estimate
26
03
Comment
~rn
3:
a.>
()
()
::l
tI)
Insufficient data
Note I
70
Y?
98
85
100
100
Assisted by bulldozers
Note 2
Insufficient data
Y?
Y
Note 3
17
50
Note 4
08
90
10
0-4
23
L-J
'-
-' -~
Vl
30
---L-J'
-t
45
30
589
""T1
Vl
39
08
C"-.
20
"
63
J~"-=",,,
LTCR%
1924
-1~",,";,,,j
o
o
10
40
L_-J
L-......:
Y? Note 4
--L-J
,_.
I --.
L.- . ..J_
.I
Sdid-Rud
1962
1760
Khashm EI Girba
1964
950
5008
50
100
R4
7100
1067
35
28
63
2]
1966
1].3
15g
Cach(
1966
54
1500
Y? Note 6
84
17
08
36
71
3393
429
20
2}
56
Note 7
Y
-
05
75
15
90
Y Note 5
--
I-Jt>ngshan
1968
75
88
Gebidclll
13
147
--.
..
07
99
99
100
96
75
36
34
N?
Note 9
85
85
Note 10
~-
SanlO Domingo
1974
30
450
02
10
07
67
70
Nunq'in
1974
102
121
05
14
8-4
52
365
1975
116
5300
22
02
19
1981
24
2700
03
0]
13
Note 8
---.. lchari
- -.
naira
100
117
03
NOles:
I. Flushing diseharge given as 120-1L10 ml/s at 12-13 III drawdown; Q and V from Atkinson (1996)
2. Value or sediment inflow ('or planning, as recommended in 13 innie & Partners (1986) report; flushing not acceptable due to downstream constraints
3. Sediment management includes sluicing during 1100d scason and lise of lateral channels
I
4. Main sediment management technique is sluicing during flood season
5. By using longitudinal flushing channels, anticipated Ihal long-term capacity could be up to 90%
6. More inrormalioll may be available in 1980 references by EI Hag (1980) and El Faith Saad (1980)
7 . Much greater Ilushing disc\larges possible via a higher outlet (abollt 25% of water deplh above base of dalll)
8. Mllch t~realer Illlshing discharges possible if needed
9. No hollom ourlet; llushing via gated spill.way only
10. Mean anllual !low taken as mean of two values reported in li rerallll'e
~
0
;0
0
n1
(/I
rn
0
3:
fl1
-I
11
r
C
VI
Z
(.)
"
. :..
.:0,
EVACUATION OF SEDIMENTS
76
77
J
EVACUATION OF SEDIMENTS
J
about 30% of the original capacity. The sediment released from the reservoir is
used beneficially for agricultural warping.
1
1
r
1
]
I
I
1
I
I
I
I
78
79
EVACUATION OF SEDIMENTS
reservoir was operated without flushing and the total sediment accretion was
about 25%. The first flushing operation, in 1978, with full draw down , was
estimated to have removed 50-60% of this accretion in three or four days. Over
a further three weeks, with assistance from bulldozers, the original storage
capacity was virtually restored, with an estimated 3-5% loss remaining.
It was concluded that, in the future, flushing should be undertaken annually,
preferably towards the end of the high-flow season. It was also considered that
empty flushing should occasionally be inten"upted and be followed by a short
period of pressure flushing to concentrate sediment removal on the immediate
areas of the bottom outlets.
Nanqin reservoir (China, 1974)
80
The Baira reservoir is hydrologically very small with a capacity 1.1 the order of
01 % of the mean annual run-off from the catchment. In the first 18 months of
operation, almost 20% of the original capacity had been consumed, representing
at least double the average annual sediment load assunled during the design. The
construction period diversion tunnel had been fitted with gates to facilitate
flushing and model studies had suggested that this wouldbe capable of removing
virtually all the deposited sediment. The first flushing operation was successful,
removing over 80% of the deposition in 40 hours, and it appears that annual
flushing should be effective in maintaining a large proportion of the original
storage capacity.
4.4.2. Findings
The findings from this review of t.~e case histories, together with the limited
information on a number of other reservoirs where flushing has been undertaken,
can be considered under the following subhea?ings:
, the hydrology of the catchment
the sedimentology of the catchment
" the storage capacity of the reservoir
o the sediment deposition'-potential
8
the shape of the reservoir basin
e the low-level outlet facilities provided
operational limitations
e whether full or partial drawdown is to be deployed
e the scope for enhancements to flushing
o downstream impacts
o the critelia for judging the success of flushing.
Hydrology.
"';'
The hydrology of the catchment needs to be properly researched and understood,
as it is central to the consideration of the other issues which affect the
practicability and likely success of flushing. It is necessary to k.TJ.OW the typical
patterns of run-off within the year, together with the ranges of variations
encountered within the year and from year to year. This information is important
both for 'broad-brush' assessments and for mathematical model simulations of
reservoir sedimentation.
If local fio\v gauging records are inadequate, additional expert hydrological
appraisals will be needed. These might make use of national or regional
hydrological paran1eters, supplemented by techniques such as flow gauging data
transposition (from within or outside the drainage basin) and flow gauging data
record extension, using correlations with longer periods of rainfall records.
The data intervals used need to take account of the size of the catchment,
varying from an hour or less when considering local floods in very small
81
1
EVACUATION OF SEDIMENTS
1
.catchments, up to. perhaps a week fer very large catchments, particularly these
where a large part o.f the flo.o.d flew is go.verned by glaciers and sno.wmelt.
Upstream reservo.irs, that are already in existence o.r planned in the future, can
have the fo.llo.wing effects en water inflews to. the reserveir:
en the to.tal annual inflo.w, if flews are diverted fer irrigatio.n er transferred into.
ether catchments, o.r subject to. increased evapo.rative lesses in sto.rage
en the seasenal distributien ef inflo.ws (unless the upstream reservo.irs are few
in number and all hydro.lo.gically small)
en the availability and co.ntro.l o.f flushing flews when they are needed .
. In the case o.f Guanting reservo.ir, fer example, the censtructio.n o.f ever 300
reserve irs upstream and increased water use fer irrigatien between the 1950s and
1970s, reduced the tetal river flews entering the reservo.ir. At the Gmund
reservo.ir, the censtructio.n o.f an upstream reserveir allo.wed a high degree o.f
co.ntrel to. be exercised ever inflews during flushing.
Acco.unt also. needs to. be taken o.f ether land-use trends affecting the
catchment hydro.lo.gy, such as urbanisatio.n and defo.restatien.
Sedimentology
The cellectio.n o.f useful sediment data is a vitally impo.rtant issue, because large
errers in predictien can be made if proper acceunt is net taken o.f the large
variatio.ns in sediment co.ncentratio.n which naturally o.ccur. An insufficiently
frequent and rigo.reus sediment data co.llectio.n pro.gramme is" liable to.
underestimate severely the large co.ntributien to. the annual sediment lead - beth
suspended and bed lead - which derives fro.m the highest discharges. Wherever
po.ssible, the available sediment data sheuld be tested against experience o.f
reserveir sedimentatio.n in the area, to.gether with natio.nal and regienal data en
sediment yields fer the seils and geelo.gical co.nditio.ns feund in the catchment.
The ultimate ebjective ef the sediment studies is to. o.btain reliable values fer
the mean annual sediment lead and the degree o.f variability fro.m year to. year,
tegether with particle size distributio.ns and to. derive hydro.graphs that give the
seasenal prefiles o.f sediment lead which co.mplement the seaso.nal flew
hydro.graphs.
Catchment changes also. need to. be taken into. acco.unt, as in the hydro.lo.gical
studies. Reductio.ns in sediment leads entering reservo.irs due to. the develepment
o.f upstream reservo.irs have been repo.rted fer the Guernsey, Gmund and
Guanting reservo.irs. At the Guanting reservo.ir, fer example, the reductio.n in
sediment leads, due to. reserveir and irrigatio.n develepment between the 1950s
and the 1970s, was much greater than the reductio.n in annual inflo.ws.
Co.nversely, fer the Mangahae reserveir, the expected develo.pment o.f additienal
upstream reservo.irs, to. reduce sediment leads, did net o.ccur. Where the
censtructio.n o.f upstream reserveirs prevides a respite in sediment lead, it sheuld
be remembered that this may be ef limited duratio.n, as the upstream reserveirs
fill with sediment, er as sediment flushing and sluicing measures are
implemented.
82
!
,.--L
1
_1
1
1
1
1
1
Storage capacity
The key storage capacity parameter is the ;hydrological size' which is the ratio
of the storage capacity to the mean annual inflow. Table 4.6 lists reservoirs with
ratios of between 01 % and almost 200%.
A ratio of 30% may be considered as an approximate boundary between
hydrologically large and small reservoirs. If the ratio is less than about 30%,
there is a reasonable prospect of having sufficient flow available to allow the
reservoir to be emptied for flushing annually, generally in the early part of the
flood season, so that it can be reliably filled in the later part of the flood season.
The smaller the ratio, then the more practicable flushing becomes, from a water
resource standpoint. Smaller ratios, of perhaps 5% or less, allow more rapid
emptying and refilling of the reservoir, and so suit relatively short periods of
flushing.
Su.bject to tlle other constraints (in particular u~e adequacy of the bottom outlet
and the suitability of the basin shape), most hydrologically small reservoirs
appear to have been flushed successfully, examples being Gmiind, Palagnedra,
Cacm, Gebidem, Santo Domingo and Baira. However, this is not a guarantee of
successful flushing if other requirements are not met, such as at Guernsey and
Warsa..1(.
If the ratio is much larger than 30%, then it becomes increasingly difficult to
schedule a flood-season flushing regime that will still meet the water storage
objectives, which generally require the reservoir to be full by the end of the flood
season. Successful flushing has nevertheless been undertaken at reservoirs witJ.1.
a higher ratio, an example being the Hengshan reservoir, aided in that case by its
small size and narrow steep Valley.
Once the ratio approaches or exceeds 100%, it is clear that an impounding
reservoir (other than one used solely for flood control) is designed for the carryover of water from one year to the next, to cover shortages in drought years.
Annual empty flushing of such reservoirs is not possible, but there may be the
possibility of empty flushing once every decade or so, if this is beneficial. There
may also be some benefit in flushing with partial draw down , at the lowest annual
water levels.
Reservoirs which are initially hydrologically large and impnlctlcable to flush,
may become practicable to flush as sedimentation reduces the storage capacity,
allowing an acceptable residual capacity to be sustained in the long term. The
Heisonglin reservoir probably falls into that category.
o
83
EVACUATION OF SEDIMENTS
Basin shape
Narrow steep-sided 'gorge-like' reservoirs are clearly more amenable to effective
flushing, pruticularly where the longitudinal gradient is steep. To some degree,
this is often offset by poorer mobility of the coarser bed load present at many of
these reservoirs. Several of the hydrologically small reservoirs listed in Table 4.6,
such as Gmund, fall into this category and have been successfully flushed. The
Hengshan reservoir, Which is hydrologically large, but steep-sided and with a
steep valley gradient, has also been successfully flushed.
Broader reservoir basins are vulnerable to sediment deposition over the
flooded floodplains, leading to two problems with flushing:
when the deposits are exposed during draw down of the reservoir, they tend to
dry out and consolidate
they are isolated from the flushing flows, which tend to be in line with the
original watercourse, so are not subject to significant erosion.
The operational regime (for example, whether the reservoir is empty or full at
the time when the flood anives) can have an impact on the vulnerability of the
former floodplains to progressive deposition.
As a result, the residual storage capacity resulting from flushing broader
reservoirs is generally limited by the geometry of the channel that can be eroded
by the flushing flows. The eroded channel gradient, width and side-slope angle
are generally functions of the sediment characteristics, valley gradient and the
flushing discharge, but may also be limited by other features of the geometry of
the basin. Guernsey, Naodehai, Heisonglin and Sefid-Rud reservoirs are
examples of broad reservoir basins where the basin geometry acts as a constraint
on the efficacy of flushing.
The flushing of reservoir basins, which include the valleys of tributaries, is
also likely to be influenced by the relative magnitudes of the tributary flows. An
84
Low-level outlets
The primary requirement, even in reservoirs where empty flushing may not be
desirable for operational reasons or considered likely to be necessary for many
years, is that there should be effective low-level outlets near the bottom of the
basin. \Vithout these, there is no possibility of undertaking empty flushing if and
when required in the future.
If the lowest outlet is at mid-height, for example, this provides a pelmanent
constraint to flushing, limiting it to the less effective partial drawdown. This is
the case at the Guernsey, Shuicaozi, Warsak and Ichari reservoirs.
J.4. common rule of thumb for successful flushing is that the discharge capacity
of L.~e low-level outlets should
be sufficient to Dass at least twice the meaTl annual
.
inflow at a drawdown of the pond level by at least 50%. This drawdown elevation
may allow sediments to be effectively scoured from Ll.e upstream half of the
reservoir length, although some of the coarser material would be expected to be
redeposited in the downstream half. However, this criterion should not be ta.l(en
to imply that only partial drawdown is needed. Full drawdown clearly has the
potential to be more effective.
If the reservoir is hydrologically smail, so that a relatively short flushing
.. period is possible, this discharge capacity criterion is probably a reasonable one
for flushing outside the flood season. In a larger reservoir, where the time taken
to draw the reservoir down would be longer, this discharge capacity may be
inadequate from an operational standpoint, even outside the high-flow season.
There would also be the risk of the drawdown being interrupted and extended by
periods of higher flow, which would also tend to result in further deposition.
For full draw down flushing in the flood season, the discharge capacity would
probably have to be significantly greater than the above rule-of-thumb value. For
t~e design or checking of a flushing system, consideration therefore should be
given to actual hydro graphs at the proposed time of flushing, simulating the time
ta..ken for the reservoir to empty for flushing, then to refill under a range of flow
conditions, together with th~f'robabilities of being able to keep the reservoir pool
at the required level during the flushing period ...
The Sanmenxia reservoir is an example of a reservoir where obtaining an
adequate discharge capacity through the low-level outlets for flushing (or
sluicing) through the flood season was vital for the success of the sediment
managemen t measures.
~
Operational considerations
The relevant operational considerations regarding flushing are mainly those
associated with the lack of water supplies and/or the reduction in operating head,
as they affect hydropower generation or irrigation supplies. These considerations
vary considerably from site to site. For example, if the irrigation supplies are
abstracted downstream and the irrigation system is designed to accept turbid
water for warping the agricultural land, such as at the Heisonglin reservoir, there
85
EVACUATION OF SEDIMENTS
may be no disruption and even a benefit from flushing. Seasonal demands will
also influence the time when flushing is most convenient and the amount of time
and water which can conveniently be used. At the Jensanpei and Sefid-Rud
reservoirs, industrial water supply and irrigation demands respectively favoured
flushing during the low-flow season.
In the final analysis, of course, operational considerations must not be allowed
to prevent effective flushing to be undertaken, if that is needed for the
preservation of the resource for future generations.
Drawdown
The case histories . show that full draw down to achieve empty flushing is
preferred, but there are cases where this is not practicable, owing to limitations
imposed by the available flushing facilities or from operational considerations.
The available outlets may prevent full draw down due to either their elevation (as
at the Guernsey, Shuicaozi, Warsak and Ichari reservoirs) or because of
insufficient discharge capacity, or a combination of reasons. In cases where a
limited discharge capacity prevents full draw down during the flood season, it
may be possible and more effective to undertake flushing at lesser discharges
during the non-flood season.
Enhancements
Of the enhancements listed in Table 4.1, fluctuating water levels during flushing
have been reported to be beneficial at the Gebidem and Santo Domingo
reservoirs. In practice, the technique has probably been more widely employed
than reported and would be expected to be always beneficial in terms of
enhancing sediment outflows. Fluctuations in flushing discharge have also been
found to be beneficial to the encouragement of slumping failures of the channel
banks.
The other enhancements all involve human intervention, preferably widl
earthmoving plant, on the reservoir deposits. In some of the cases where
bulldozers have been used to shift sediment towards the main 'channel and
increase its rate of disposal, the intervention was experimental in the first flushing
and may not be found worthwhile after a regular flushing regime has been
instigated.
The two cases (Heisonglin and Sefid-Rud) where lateral channels, lateral
piping and longitudinal channels have been successfully deployed (and where
sufficient information is available to judge) are hydrologically large reservoirs
with significant areas of deposits over the fonner floodplain. Without these
measures, there would be a significantly worse prognosis for the preservation of
storage in the long term.
Downstream impacts
There have been severe impacts on the downstream aquatic environment in a
number of cases, principally where heavy deposition or high suspended sediment
concentrations affect the habitat and the survival of fish and other wildlife. In this
respect, Sh011 peliods of flushing are particularly problematic. Sediment routing,
86
.!
by sluicing through most of the flood season, on the other hand, is "much more
benign environmentally.
The potential adverse environmental impacts downstream should always
considered in the light of the alternatives, such as an acceptance of long-term
sediment accretion, requiring the further use of natural resources for such things
as the construction of additional reservoirs or the development and use of
alternative sources of power.
In some cases, reservoir operators are subject to regulatory limits on
downstream sediment loads or concentrations, wrJeh have to be taken into
account in the detailed planning of every flushing operation. Because of the
variable nature of the phenomenon, unless a very large factor of safety is
employed vvith implications for the duration of flushing needed, occasional noncomplial"lce is almost inevitable.
The downstream impacts can be rnitigated substantially if there is dilution
available a short distance downstream, for example, from hydropower releases or
at the confluence wit.~ a larger river, such as in the case of the Rhone downstream
of the Gebidem reservoir. The control of flushing - for example, in response to
downstream concentration monitoring
can be aided if inflows to the reservoir
can be controlled, for example, by releases from an upstream reservoir, as at the
Gmund reservoir.
Other downstream interests that may be affected by sediment releases
include:
recreation, such as boating or swimming
water supply intakes
III hydropower intakes
reservoirs.
@
Ill)
L~at
87
,
I
EVACUATION OF SEDIMENTS
J
The key test is that a practical compromise should be achieved between the
processes of sedimentation and the requirements for beneficial use of the
reservoir. The long-term beneficial use may be much diminished from what
might have been expected at the time when the reservoir was designed and built,
but in comparison with the poorer or even catastrophic outcome that may result
without intervention, any tangible improvement from flushing must be judged a
success to some degree.
In these terms, therefore, the flushing undertaken at most of the reservoirs
presented in the case studies can be considered successful. Several of these are
cases that were considered unsuccessful by Atkinson (1996), based on the more
restrictive criteria listed earlier. The reservoirs at which flushing is not possible
or has been unsuccessful are probably limited to the following:
Guernsey
Guanting
Warsak
Ichari
J.
J
J
are
88
...l
J
1
"
"
89
ro hi
it d to
. (a) Climate
(i) precipitation and run-off
(ii) temperature
.
(iii) wind speed and direction.
(b) Geotechnics
(i) geology
(ii) volcanic and tectonic activity
(iii) soils.
(c) Topography
(i)
slope
5. 1.2. Estimates
The estiInates derived from more than a dozen studies of global average rates of
denudation have ranged from 006 to 016 rnmlyr (Morris and Fan, 1997).
Estimates for the aggregate worldwide sedilnent yield of between 15 and
20 x 109 t/yr have been given (\Valling and \Vebb, 1996). Areas with sediment
yield over 1000 tlkIn2/yr are 88% of the total land area and account for 690/0 of
93
EVACUATION OF SEDIMENTS
the total sediment load. Regions with less than 50 tJkm2/yr account for about half
of the land area and 21 % of the sediment yield. Case studies of erosion rates are
presented in Appendix A4.2.
Table 5.1.
~I
Area
I
mm
km3
North America
756
158
154
Asia
740
257
250
Africa
740
197
1600
790
South America
Europe
Sediment:
Mtlyr
Run-off:
km3
Precipitation
Yield:
tlkm2Jyr
66
171
1460
109
84
108
280
6350
474
380
192
42
109
530
39
35
270
262
118
305
1790
133
97.
75
73
27
230
17
50
60
0-4
28
3000
224
1000
13420
1000
165
70
I
Australia
791
71
25
69
65
Oceania
I
i
Total
1028
1000
386
1000
Table 5.2. Continental variations in sediment yield (Jolly, 1982, taken from Gregory
and Walling, 1973)
Suspended sediment:
tllcrn 2Jyr
Continent
Suspended sediment:
10 Mtlyr
27
550
600
16160
Australia
45
230
Europe
35
330
96
1990
63
1220
Mrica
Asia
North America
~j
J
J
South America
94
approximately 80% of the world sediment total (Jolly, 1982). The largest
sediment yields occur in Oceania at 1000 t/km2/yr including ntllnerous
catchments in. New Zealand, New Guinea and Taiwan with sediment yields two
to three times the world average.
There are considerable differences between the continental figures produced
by the two studies. The table produced by Mahmood (Table 5.1) distinguishes
between yield rates in Australia and Oceania and this leads to the highest
sediment yield rates in Oceanian rivers at 1000 t/km2/yr. This is hidden in the
study by Gregory and Walling (Table 5.2), where Australia is taken to include
Oceania. This produces a higher rate for Australia than found in the study by
Mahmood but a much lower rate than for Oceania. The next highest sediment
yield is produced by Asia with a yield of 380-600 tlkm?/yr. The lowest rates of
sediment yield occur in Australia (owing to aridity) in the Mahmood study, at 28
tfJs:m?/yr, and in Africa, at 27 tlkm?/yr, in the study by Gregory and Walling.
These rivers do not produce the world's highest sediment yields as the list
excludes drainage basins smaller than 10000 km2 which are likely to produce the
highest yields per unit area.
95
J
EVACUATION OF SEDIMENTS
1
Table 5.3. World maximum recorded suspended-sediment yields greater than
2000 tlkm 2/yr (Jolly, 1982, from Gregory and Walling, 1973)
River
Location
r--'-
Ching
Changchiashan, China
8040
Lo
Chuantou, China
7922
Waipaoa
6982
Tjatabon
Java, Indonesia
6250
Lo-Lo
Luyang, China
6068
Pietracuta, Italy
4570
Semani
4150
Soldier
4072
Shkum Bini
Paper, Albania
3590
Kosi
Chatra, India
3130
Yellow
Shenhsien, China
2957
Indus
Kalabagh, Pakistan
2498
Santa Anita
2374
Eel
2292
Marecchia
...,;. .
1
1
I
valley, and areas of volcanic soils, such as Java, South Island of New Zealand,
Papua New Guinea and parts of Central America (Morris and Fan, 1997).
The Pacific Asiatic-Australian sector demonstrates the most intensive rates of
erosion. Figures in the range of 10 000 to 50 000 tJkm2jyr have been reported
at stations in China, Taiwan, the Philippines, Indonesia, Java, Kenya, New Guinea
and New Zealand (Walling, 1994) due to active tectonics and volcanism, steep
slopes, high precipitation amounts and intensities, high and irregular lun-off, .
dissected mountain relief composed mainly of sedimentary rocks, and human
influence by agriculture and logging (Dedkov and Moszherin, 1992).
Taiwan discharges more sediment to the ocean per unit area than any other
country in the world. Streams draining the central range produce suspended
sediment yields of 13 760 tlkm2jyr. One small basin exports 31 700 tJkm2jyr
(Li, 1976). The sediment discharge of Taiwan is nearly five times larger than
that from the continent of Australia, even though it is 210 times smaller. Lower
values of 11-12000 tJkm2 have been reported in Java (Walling and Webb,
1983).
In New Zealand values reach between 20 and 28 000 tlkm2jyr with a mean
value of around 2000 tJkm2jyr. The highest mean annual sp~cific suspended-
96
.---L
1
1
1
f
1
j
Table 5.4. Rates of sedilnent yield for the world's major rivers at ocean level,
excluding basins with an area less than 10 000 km 2 (IYlalirJ'lood, 1987)
Continent
River
Country
Drainage area:
million km 2
Run-off:
cm/yr
Oceania
New Guinea
Purari
0031
248
South AmeLica
Peru
Chira
002
25
Sediment:
tlkm 2/yr
Yield:
ppm
2581
1039
2000
8000
Asia
China
Daling
Asia
China
Haiho
Asia
China
Yellow
India
Damodar
Asia
I
I
II
I Copper
1800
36 000
005
1620
40500
077
1403
22041
01.02
50
1400
2800
65
1167
1795
1128
1720
1083
1057
917
928
01.06
66
148
Bangladesh
002
Asia
Vietnam
Hungho
012
103
South America
Colombia
Magdelena
9.24
99
043
100
616
619
80
500
625
0061
126
492
390
097
25
454
1849
31
27
310
1143
Asia
I Burma
Irrawaddy
USA
Susitna
Oceania
New Guinea
Fly
Asia
Pakistan
Indus
Asia
I India
I China
Orinoco
99
111
212
191
Colorado
9.64
211
6750
Europe
Mekong
Brazil
017
I
I
I
I Vietnam
I China
340
59
Pearl
9.44
69
157
228
Amazon
~'15
102
146
143
al1
145
2286
II
C1
30
130
433
(09
S4
111
204
Negro
Rhone
i
203
79
Brazos
531
326
Mexico
South America
214
North America
I USA
II Ar2:en tina
I France
66
Venezuela
North America
246
9.07
South America
South America
46
6833
Po
I
I
241
Italy
Asia
r
194
Europe
Asia
Yangtze
I
I Liaohc .
USSR
~; ~
005
Godavari
4l,,:)Ld.
1\
..
North America
Asia
.'
97
1
EVACUATION OF SEDIMENTS
Table 5.4.
continued
Continent
Country
River
Drainage area:
million krn 2
Run-off:
crnlyr
Sediment:
t/km 2/yr
Yield:
ppm
North AmeIica
USA
Mississippi
327
18
107
602
Africa
Tanzania
Rufiji
018
94
1889
North America
Canada
Fraser
022
51
91
179
Europe
Romania
Danube
081
25
83
325
Africa
Mozambique
Limpopo
041
80
6600
North America
USA
Yukon
084
23
71
308
North America
Canada,",.
181
17
55
327
North America
USA
002
60
50
83
"
"
"
1
1
i
1
I
Asia
Iraq
Tigris-Eupha
105
50
1152
Europoean AItic
USSR
Indigirka
036
15
39
255
Africa
Egypt
Nile
296
38
3700
South Amelica
Argentina
La Plata
283
17
33
196
Africa
Nigeria
Niger
121
16
33
208
Asia
USSR
Amur
185
18
28
160
Oceania
Australia
Murray
106
28
1364
Africa
South Africa
Orange
102
17
1545
Africa
Mozambique
Zambesi
12
19
17
90
Asia
' India
Mehandi
013
52
15
30
European Artic
USSR
Yana
022
13
14
103
European Artic
USSR
Sev. Dvina
035
30
13
42
North America
USA
Columbia
067
37
12
32
Africa
Zaire
Zaire
382
33
11
34
South Amelica
Brazil
Sao Francisco
064
15
62
European Artic
USSR
Kolyma
064
11
85
European Artic
USSR
Ob
25
15
42
1
1
j
-1
" I'
98
European AItic
USSR
Yenisei
258
22
23
European Artic
USSR
Lena
25
21
23
North America
Canada
St Lawrence
103
43
1
I
..J..
1
T
AREAS SUITEDTO
sediment yield is 53 SOD t/la.n2/yr for the Huangfuachan river (3199 k1112), a
tributary of the Yellow River in China (Walling and Webb, 1983).
Table 5.5 lists a number of basins with very high yields in various countries,
which were reported in 1983 and which exceed the values of record yields
published in 1973 and shown in Table 5.3. For tributades of the Yellow River,
highly erodible loess, lack of vegetation and the semi -arid climate are the major
controlling factors. The semi-arid climate is a factor in the Kenyan example, but
severe disturbance due to agriculture is also a factor. For Java and New Guinea
steep relief, high rainfall and agriculture are important and in New Zealand the
steep relief, high rainfall up to 9000 mm1yr, and tectonic activity play a role
(vValling and Webb, 1983).
Country
River
China
Dali
China
China
Kenya
II
Mean annual
sediment yield:
Source
tfkrn'1jyr
961
25600
Dali
187
21 700
Dali
3893
16300
Perkerra
1310
(Unknown)
Taiwan
Drainage
area: km 2
I!
(Unknown)
19520
31 700
Dunne (1979)
Li (1976)
Java
Cilutung
600
12000
I
I Hardjowitjitro (1981)
Java
Cikeruh
250
11 200
Hardjowitjitro (1981)
New Guinea
Aure
4360
11126
Waiapu
1378
19970
Griffit.~s
Waingaromia
175
17340
Griffiths (1982)
Hikuwai
307
13 890
Griffiths (1982)
South Island
Hokitika
352
17070
Griffiths (1982)
New Zealand
Cleddau
155
13 300
Griffiths (1981)
North Island
New Zealand
New Zealand
J
I
(1982)
99
EVACUATION OF SEDIMENTS
New South Wales, Australia. Values of less than 1 t/km 2/yr have been reported in
Poland (Branski, 1975) (from Walling and Webb, 1983). Areas of low sediment
yield are usually flat, arid with inadequate streamflow to transport large sediment
volumes, or arctic regions with low relief, little precipitation and human impact
(Morris and Fan, 1997) or low mountains of the temperate zones that are
underlain by crystalline rocks and covered by dense deserts e.g. Scandinavia, the
Urals, the mountains of South Siberia and the Trans-Baikal region (Dedkov and
Moszherin, 1992). The results quoted for Poland appear to be anomalous.
100
",
- ,.
5.2.
In northern USA and in Canada there are low sedin1ent yields, ' except in
Alaska and the mountains, in southern USA and along the coast' of California
(Jansson, 1988).
5.2./. Introduction
An understanding of the precipitation regimes throughout the world may provide
a key to the definition of areas of high and low erosion rates. We briefly describe
these in the section which follows. It is difficult to classify distinct climatic zones
as they tend to merge into one another rather than have sharp boundaries but a
number of general models have been produced. These are discussed in Appendix
A4.3.
Seasonal variation
Precipitation distribution during winter (December to February). Figure 5.4 shows
the global distribution of precipitation during the northern winter of 1998.
During this period there were precipitation values of over 100 mm per month
along the western fringe of North America into Canada, a region to the north of
the Gulf of !vIexico, a region of Central America, mainly in Brazil, southern
Africa in the region of fvlozambique, Madaascar, Tanzania, Zambia and Angola.
In Asia, high p;ecipitation values- occur in the Pacific Islands and in the northern .
part of Australia.
101
()
.......
. ..
.. .. . . . .. .
.. ... .. . .. .. ...... ...... .. ...... .
...-...............
...
. .. .. ., .. ..............
. . . . .. . . . .. . .
.............................
. .. . ....................
. . . . . . . .. . .. . . . . . . .. . . . . .
1 --'
o
z
o-n
..
..
"!j
..
..
..
.. . .
It
.,
..
(/)
3:
m
(/)
t km-2 yr- 1
240
100
50
10
,0
Arid regions
(a)
Figure 5.1.
't"
'
'
Global patterns of sediment yield: (a) after Strakhov (1967); (b) after Fournier (1960)
"
,-
,"
,., '
L--- '
' . L ,,--
'" L.--,- ,
'
103
<
()
~
0
z
0
"m
Vl
-0
3:
-I
Vl
20
..
(a)
Figure 5.2. Global patterns of suspended sediment yield: (a) from Lvovich (1991) in Walling and Webb (1996); (b) frOln Walling and
Webb (1983)
1--
l- -
l--.
~ .
1--
"I
"
~,
~ .~
xv
1000
;0
m
750
500
Vl
Vl
250
100
50
VI
Deserts and
permanent ice
(b)
),
I@
-;
0
I~
C
Vl
Vl
Figure 5.2_
continued
Z
G)
()
o
z
o
"m
(/)
3:
-I
30N - t - - - - -- - t - " I
(/)
EQ~--------~----~
3~~---------+----~--~
6~~--------~----------~----------r----------+--------~~-------~~
9~~--------~----------~----------~---------+----------~--------~
1BO
12ml
60W
(I
GOE
1eo
1
Figure 5.3.
10
25
50
75
tOO
150
1 :
~
OD
300
0400
800
800 1000
. .
.I
~_. _
...._... . ..
~_"'
:..a ._~
__
~ ; ~ !a.'
. . ..;:J~ 5~ ~~
Ea~-----------~-----~~~
:;;0
VI
VI
C
-l
9~~----------~----------~----------~~------------~----------~~--------~
1fiD
12ml
!)OW
0
tiOE
120
1SO
1Q
25
50
15
100
1M
200
300
400
-l
o
r"C
VI
I:
Figure 5.4.
Z
C)
1
EVACUATION OF SEDIMENTS
Precipitation distribution during spring (March to May). Figure 5.5 shows the
global precipitation distribution during the nOl1hern spring of 1998. In spring,
high precipitation values extend to roughly the same area with a gradual
northwards movement of peak values. In Africa the high values are now more
concentrated on the central-western coast around Cameroon and Gabon and
values in Asia are higher in eastern China and Japan with the islands of the
Pacific still experiencing high values.
Precipitation distribution during summer Oune to August). Figure 5.6 shows the
global precipitation distribution during the northern summer of 1998. This shows
high precipitation values in America extend from-the northern regions of South
America into Central America. In Africa, high values are founq in central regions
and in the westenl zone from Guinea, Ivory Coast, Cameroon, Nigeria and into
Zaire. In Asia, the monsoon period brings high amounts of rainfall to India, Tibet,
China, the other eastern Asian countries and to the Pacific Islands.
Precipitation distribution during autumn (September to November). Figure 5.7
shows the global precipitation distribution during the nOl1hem autumn of 1998,
with regions of high precipitation now more isolated in America and confined to
a narrow strip along the western coast, Central America and the northern region
of South America. In Africa, high values occur in the western region around
Cameroon and Gabon. In Asia, the highest values are again in the islands of the
Pacific stretching into Japan, India, Vietnam and Cambodia.
The information shown in the four seasonal maps is available on the Internet.
It would be of interest in the present context to derive a world map showing the
degree of seasonal polarity of precipitation, as this may correlate with erosion
rates and sediment yields. However, this exercise was not possible within this
study.
Tropical wet
e
e
The climate has consistent high daily temperatures ranging from 20-30C.
Monthly temp~rature averages range from 24-30C . . The annual range of
monthly temperatures is about 3C. Precipitation is uniform with a total over
108
1
1
1
1
1
1
I
J..
1
1
1
1
1
I
J-
, 1
-I
[O-~-----------+---~'-
3(}S - i l - - - - - - - - - I - - - --
-+.i
rn
;0
Ul
Ul
C
-l
D~~----------~-----------F~---------+~--------~----------~~----------,
1
12[}11(
150'1/
0
1 ZOE
1BO
-I
o
."
10
25
50
15
, 00
t flO
200
300
.4QO
600
800
100'D
Ul
:c
Figllre 5_5_
zC)
~
()
9{JN------...----~--.....--...------.......------..,._----..-;...,.....----_.,
z
o
-n
Vl
OON
3:
m
-I
Vl
EC~----------~--~~
3~~----------r-------~~
6(}S-l-------+------+-------+------+--------t----~-____1
90~J.BO------12.j..rn-'I-----6+0.,,-t-----..f..D------a+oE------1+20-E------IllBO
1
Figure 5.6.
L-
' L-
' 1_
L- '
.1_
M~.' on month,~
__-----""""lI
m D - - - - - - - - I-----.;..,:-;+-I\I
3~) -~------------1----------~~),~~---------1--~~~~~~~-~-----------~
;::0
6liS u--------l-----------::--+-~----_il_~,..,.,...;..~~,-_I---------II_--------_I
Vl
Vl
..
90S -\-------+------""i-~----_Ii__------_I--1BO
12mI'
M't.'
tiOE
---__I------~
120E
1BO
C
-l
m
Q
-l
."
10
50
75
100
ffl<J
200
300
400
600
800
1000
r
C
Vl
Figure 5.7.
Z
C)
EVACUATION OF SEDIMENTS
Table 5.6.
Temp: C
Feb.
27
27
Mar. , Apr.
28
28
May
June
July
Aug.
Sept.
Oct.
Nov.
Dec.
Yr
27
27
27
27
27
27
27
27
27
[
-
Pptn: mm
554 / 519
557
620
655
655
572
574
561
563
563
512
,
,
6905
J
2000 mm. The region lies within the effects of the intertropical convergence zone
all year. Convergence and high maritime humidities create cumulus clouds and
thunderstorms regularly. A typical monthly distribution of temperature and
rainfall is given in Table 5.6.
....
.-J
- I
-!
.._ \
The climate has distinct wet and dry periods. The seasonal pattern is due to the
movement of the intertropical convergence zone. The wet season coincides with
the high sun and the presence of the convergence zone. The dry season is due to
more stable air associated with the presence of the subtropical high zone during the
low sun season.
During the rainy season the climate is similar to the tropical wet climate. During
the dry season, semi-desert conditions prevail. Some regions experience intensification of rainfall due to monsoons and orographic uplift. A typical monthly
distribution of temperature and rainfall is given in Table 5.7.
J
i
Tropical desert
Koppen classification BW.
Continental tropical air mass all year.
This region is found near the tropics usually, but not always, on the western side
of continents and covers 25% of all land area. It is characterised by:
low relative humidity (10-30%) and cloud cover
low frequency and amount of precipitation
high mean annual temperature
J
J
It
Table 5.7.
Feb.
Mar.
Apr.
May
June
July
Aug.
Sept.
Oct.
Nov.
Dec.
Yr
Temp: C
27
27
28
29
29
27
26
26
26
27.
27
27
27
Pptn: mm
40
233
982
1059
577
267
206
71
18
3467
J
I
112
J:
Table 5.8.
I Feb.
Mar.
Apr.
May
June
July
32
33
Temp: C
15
17
21
26
31
32
Pptn: mm
, 1
Aug.
I Sept. I Oct.
Nov.
30
28
22
I Dec.
I
I
Yr
17
35
, 3
~,r': ~
.'
_" ......-
The region is influenced {~y "'upp~r ail<:~:$ta.bility and subsidence owing to the
presence of the subtropicalhfgh pressure:~'zone . Temperatures are highly variable
daily and annually. vVith theayerage monthly temperatures ranging fronl 29-35C
and the average diurnal rangejs :}Jetween 14-25C. A typical monthly distribution
of temperature and rainfall is tsiven in Table 5.8.
Mid-latitude wet
"Koppen classification Cf a#d~f.~: .
.;, :."}~.7:' :'
NIaritime tropical in sUII1IIl~r:~'d:maritime p~Iru<ln"mnte~~:~.
Table 5.9.
I Jan .
Temp : cC
Pptn: rnm
I
[
Feb.
54
40
I Mar. I APr.'
,
,
' 37
38
I
I
May
I June
12
16
46
46
I July
i
18
56
I
I
I
I
Aug.
17
59
15
50
I
I
11
10
48
, 5"95
57
64
113
EVACUATION OF SEDIMENTS
Table 5.10.
1
Mar.
Feb.
Apr.
Temp: C
-10
-9
-4
17 I 20
, 26
35
i
I
Pptn: mm
i
i
:
10
52
July i Aug.
June
May
32J m
Sept.
Oct.
Nov.
Dec.
Yr
-7
15
444
13
17
15
11
-2
88
58
! 59
35
23
16
Cold winters, cool summers with a summer rainfall regime. The areas experiencing
this climate are the North American Arctic coast, Iceland, coastal Greenland, the
Arctic coast of Europe and Asia and the Sou~hern Hemisphere islands. Annual
precipitation is less than 250 mm with precipit4tion during the summer. A typical
monthly distribution of temperature andrainfal~ is given in Table 5.12.
Polar desert
Koppen classification EE
I
Continental Arctic and continental polar air masses.
These regions occur in continental areas of thd high-latitudes, such as Greenland
and Antarctica. No solar radiation is received for about half the year while during
I
Table 5.11.
I Jan.
I
I
Temp: C
I Julyl
Feb.
Mar.
Apr.
May
10
13
17
22 I 24
88 I 77
72
63
48
June
Aug.
I Sept. Ii Oct.
i
Nov.
Dec.
Yr
24 , 21
16
12
15
22
128
116
106
881
Pptn: mm
76
114
14
70
Table 5.12.
Temp: C
Pptn: mm
I Oct.
Nov., Dec.
Yr
-8
1-
19
-24
I -26
-16
21
16
35
204
I Feb. I Mar.
Apr.
May
June
July
Aug.
Sept.
-30 /-30
-33
-23
-11
12
19
Jan.
23
20
37
the summer insolation is high with long days, however the albedo of the snow
surfaces reflects up to 90% back. Average monthly temperatures are generally
below OC. A typical monthly distribution of temperature and rainfall is given in
Table 5.13.
In the cold steppe climate (BSk) the range in value of annual sediment yield
is betvveen 1 and 16 300 t/km2 Only 13 out of the 75 rivers have yields of
more than 600 tJkm2; these are mainly in China and Argentina with one in
South Africa and one in ~he USA.
The two warm temperate humid climates with no dry period (Cfa and Cfb)
have high standard deviations. For the Cfa climate, Taiwan and Italy are
responsible for the high standard deviations. The ten rivers with the highest
sediment yield are in Taiwan with a range of 2605-18 339 tJkm2 while 106 of
the rivers have yields of less than 300 tlkm2 In the Cfb climate there are 144
rivers with less than 100 tf1lffi12 and 38 rivers (23 in New Zealand) which yield
more than 500 t/km2,
The cool Mediterranean climate (Csb) has limited data with a wide variation
2
in yields. Four rivers in Australia have yields of less than 2 t/km and four
rivers in the USA yield more than 1000 t/km2
., The boreal climate without a dry period (Dfb and Dfc) has a low sediment
2
yield. The median value for Dfb is 33 tJkm2 and the mean is 104 tlkrn . There
I.
Table 5.13.
I Jan.
Temp:
Pptn: mm
[ Feb., Mar.
I APr.
, May
Sept.
Yr
-17
-17
-14
-7
-3
1-12
I 95
52
43
46
26
I 625
115
EVACUATION OF SEDIMENTS
~----------~~~--------------------------------~~~~
Figure 5.B.
B Dry climates
s steppe
w desert
h hot, mean temp. >18
k cold, mean temp. <18
D Boreal climates
"
E Snow climates
T tundra
F always frost
are eleven mountainous rivers in Romania and two Soviet rivers in the
Caucasus area with more than 500 t/km2 The median value of Dfc climate is
9 t/km2 with a mean of 76 t1km2 The data are highly variable, with a standard
deviation of 35 times the mean.
The boreal climate with dry winter (Dwb) has too few data points to give
reliable values. The snow climate ET consists of two populations - rivers in
the fonner Soviet Union with up to 50 t/km2 and rivers in Alaska and Iceland
with yields of between 375 and 4000 t/km2 (Jansson, 1988).
The data from the climatic regions were amalgamated by Jansson (1988) into
simplified homogeneous climatic groups, depending on their characteristics. The
groups are set out in Table 5.14.
The classification of sediment yield for each climatic group shows a number
of distinct features which can be seen in Figure 5.9. Over half the rivers in the
Af group have sediment yields greater than 1000 t/km2, with only one river
exhibiting a value less than 100 t/km2 The groups Cwa and Cs have the next
largest percentages in the highest sediment yield class, with more than 50% of
the basins having sediment yields above 100 tIkln2 Dfa-d and Dwc have 67%
and 100% of the rivers respectively with sediment yields less than 100 t/km2 Cfb
116
Table 5.14. Reasons for com,bining climates into homogeneous climatic groups .
(Jansson, 1988)
Climates
I
Af,Cf
Climatic
group
Af
AW,Am, Cw,
outside Argentina
Aw
BSh,BSk
BS
BW
I
I
I
BWh, BWk
Cfa
-.
Cfa
Ctb, Cfc
Ctb
Df
Df
Csa, Csb, Cs
Cs
Dsb, Ds
Ds
.-
Cwa, Cw in
Argentina
Cwa
Dfa-d
Dwe
D we. D\vd
ET, ET (Mt)
__::"._0
I
I
ET
Similar climate
has about 450/0 within the two lowest classes. Low values are found in Central
and Western Europe, with high values tn the Southern Hemisphere and in
southern-most Europe.
Generally, a tropical Af climate has high sediment yields while Aw has more
variable values. In arid climates BS has high values, while the desert group BW
has low and intermediate values. There are few rivers with yields less than
10 tf1.i.<ll12 in the warm temperate clirrlates except in the Cfb zone. In the boreal
region sediment yields are low, except in the mountains of the warm temperate
zone, Ds and Df, where elevated yields occur (Jansson, 1988).
Jansson (1988) provided a list of countries in each climatic zone in order to
show the geographical distribution of each. This allows an estimate of erosion
117
EVACUATION OF SEDIMENTS
c:::J
0-10
CJ
11-50
51-100
h',~; :J,hl
101-500
Snow
501-1000
_
>1000
Boreal
1
r--'-
Warm
temperate
,--
At
Tropical
and arid
---'-
Figure 5.9.
rates to be made for each country based on climatic classification. Table 5.15 lists
the climate zones in each country and indicates the number of rivers that occur
in each zone.
A similar exercise has been carried out in this study for a more comprehensive
list of countries in the world, the results of which are shown in Appendix A4.4.
The updated map of climatic zones produced by Alexandersson (1982) based on
the Koppen classification, was used to determine the climate classes that occur in
each country and the rates of sediment yield were based on the map produced by
Walling and Webb (1983) (see Appendix A4.4).
There is a reasonable similarity between the data listed in Table 5.15 and the
results we have obtained using the world climate map. Differences arise mainly
because the data in Table 5.15 are based on a number of river basins in each
118
Table 5.15. . Countries classified into climatic zones showing nwnber of river basins in
each zone (modijiedfrom Jansson, 1988)
Climate zone
Albania (312)
Cfb/Csa
Algeria (27/1)
CsafBSk
Argentina (22/13/5/5/2)
BSkiCwafCw/ET (Mt)IBSh
Australia (4/3/1/1/1/1)
CsbfBSh/Csa/CfbIBWhiAm
Austria (4/3)
Df/Cfb
Bolivia (3)
Cwa
Brazil (12)
Aw
Bulgaria (24/6/6)
Cfb/CfafCsa
Cameroon (311)
AwlBSh
Canada (49/18/2)
DfblDfclBSk
Chad (4/1)
-- ~
AwlBSh
Chile (2112)
ET (Mt)/Cs
China (1015/4/3/1/1)
BSkJCfaIDwa/Dwb/Cwa/BWk
Colombia (1512)
Cf/Aw
Costa Rica (10/6/1)
Awl Af/Cf
Cuba (1)
Aw
Czechoslovakia (7/5)
Cfb/Dfb
Denmark (2)
Cfb
Ecuador (1012)
Cw/Aw
El Salvador (2)
Aw
Finland (16/3)
DfclDfo
France (4)
Cfb
Germany (55)
Cfo
Great Britain (9)
Ctb
Greece (414/1)
Csa/Cfb/Cfa
Haiti (1)
Aw
Honduras (1)
Cw
Hungary (6/2)
Cfb/Cfa
Iceland (8/1)
ET/Cfc
India (5/5/2/1/111)
CwalAwfBSfET (Mt)/CwfBSh
Iran (6/2/1)
Csa/BSk./BWh
Iraq (1)
BWh
Israel (2/1)
CsaIBSh
Italy (26114/811)
CfalCs;?}CfbIDf
Java (6)
Af
Kenya (611)
CwaIBSh
Lesotho (9/5)
Df/Cfn
Madagascar (1)
BSh
Malaysia (3)
Af
Morocco (14/2)
CsaIBSk
Nepal (1)
Cwa
New Zealand (42)
Cfb
BWh
Niger (1)
Nigeria (11n)
BShlAw
Pakistan (1)
Ds
Panama (2)
Af
Papua New Guinea (3/2)
Af/Cf
Peru (2)
Cw
Philippines (5)
Am
Poland (53/28)
Dfb/Cfb
Romania (50/412)
Dfb/Ctb/Cfa
South Africa (17/8/7/4/2/1/111)
I' BSklCwb/CtbfBWklBWhlCsb/CfaJBSh
Soviet .(55/351221 IS!) 41 121 10/9/5/4/3/2/111) , DfblDfclD wcfDflDs/BSklETICfaJDwblDwdIDfaJCslDfdlCtb
119
EVACUATION OF SEDIMENTS
Table 5.15.
continued
Climate zone
Spain (18/3)
Sweden (14/3/2)
Switzerland (9/5)
Taiwan (16111)
Tanzania (1)
Thailand (25/13111)
Trinidad (1)
Tunisia (6)
Turkey (1)
USA (83116/14/12/916/4/212)
USA-Alaska (7/5)
Venezuela (4/312)
Yugoslavia (28/311)
Zimbabwe (111)
CsaJCfb
- 1
Dfc/CfblDfb
Df/Cfb
CwaJCfa
Bsh
Aw/CwaJAm
BS
Csa
Ds
CfalBSk/CsblDfblDfaJDflDsb/CfblB sh
DfclET
Cf/AwIBsh
Cfb/CfaJCs
BShlCwa
1
I
.....I-
country rather than on a global assessment of which climatic zone the country fits
In.
This study emphasises the diversity of climate classes and rates of sediment
yield within many individual countries and shows that this base unit will often
not be small enough to produce homogeneous conditions. On a global scale, a
compromise must be reached between generalisation and accuracy and the
country provides a manageable unit for some purposes.
1
1
,
1
i
5.3.
5.3.1. Introduction
There are a number of factors common to areas suitable for the application of
reservoir flushing techniques. A first prerequisite is that there must be ,a high to
medium erosion rate within the catchment. Secondly, the sediment must then be
transported down the river system to the reservoir resulting in the requirement for
its removal for flushing. These two prerequisites are discussed in the first two
sections of this chapter. The hydrological characteristics required for successful
flushing are then considered.
120
Precipitation
High rates of erosion occur in regions where there is high intensity of rainfall. It
is not just high precipitation totals that result in high erosion rates but it is the
relationship between precipitation and vegetation. Global relationships between
erosion rates and precipitation show variable results.
Geology
The geology is an important factor deteITPjning the susceptibility of the rock to
the effects of erosive forces. Erosion rates are generally highest in areas of soft
sedimentary rocks.
Soils
The key characteristics of a soil that influence erosion rates are texture, structure,
organic matter content, shear strength and infiltration capacity. High erosion rates
occur where the texture of the soil is high in silt and fine sand and low in clay,
and where the structure is compacted and the organic matter, shear strength and
infiltration rates are low. All these factors cause high run-off rates, leading to
erosion of the soil.
Slope
The gradient and length of the soil sUlface influence the velocity and direction of
run-off and therefore its erosivity. High erosion rates occur where there are long,
steep slopes resulting in movement of water downslope at a high velocity.
Vegetationlland use
Vegetation depends on the interaction of a number of factors including rainfall,
temperature, soils and topography. The presence of a vegetation cover reduces
the erosive power of rainfall by dissipating its energy, increasing infiltration,
reducing the velocity of nln-off and by holding soil particles together. High
erosion rates therefore occur where there is sparse vegetation cover either due to
natural climatic conditions or due to land-use practices.
Human impact
Activities such as deforestation, urbanisation and agriculture all affect the
erodibility of the soil. Current erosion rates are more than two and a half times
the historic, mainly as a result of human influences.
121
EVACUATION OF SEDIMENTS
Erosion processes
The sediment delivery ratio is generally higher for sediment derived from
channel-type erosion which delivers sediment to the main channels of the
transport system more quickly and more directly than from sheet erosion.
Distance from basin outlet
Channel networks with a high drainage density are more efficient for transporting
sediment than basins that have a low channel density, meandering low gradient
channels, or those clogged with debris.
Soil and vegetation
Finer particles are transported more easily than coarse particles, therefore higher
delivery ratios occur for soils with fine grained erosion particles. However, silts
tend to be more erosive and produce higher delivery ratios than clays.
DepOSitional features
The presence of a depositional area decreases the sediment delivery ratio. Most
of the sediment eroded from the steep uplands of basins may be redeposited at
the base of slopes.
122
5.3.5. Areas of the world which are best suited to reservoir flushing
It is not possible to define precisely which specific areas of the world will provide
conditions for 'successful' flushing. In reality there is a spectrum of conditions
ranging from tllOse sites where conditions are ideal to those sites which are quite
unsuited to sediment flushing.
From the hydrological a11d hydraulic conditions neCeSSfuy' for successful
reservoir flushing the luost likely locations in which to use this technique are
those which are within the Koppen climate classification tropical wet and dry:
classifications Aw, Am and BS. Also, there are areas in the mid-latitudes where
spring snowmelt provides a regular and predictable annual pattern of high
flows.
From the Koppen classification of climatic zones and the mid-latitude spring
snowmelt effect, the requirements for successful flushing are most likely to be
met in the following locations:
parts of Central America extending into South America
o areas in North and South America where the rivers are fed by the Rockies and
the Andes
Q
parts of Central Africa from the Ivory Coast in the west to Sudan in the east
19 areas in Central Asia where the rivers are fed by the Himalayas, including
Pakistan, India and Nepal
" parts of Asia including Cambodia, V-ietnam and Thailand.
o
123
Si
e
In
.m:i
pecific
sti
d sIgn
an
era ons
: .' ,
Site investigations
Flushing outlets have to be able to withstand high velocityfiows with high
concentrations of sediment. Such flows are pJghly abrasive and expensive steel
lining will normally be required to avoid undue damage to the structures. Hence
it is important that the site allows for the construction DT relatively compact
flushing facilities, either orifices within the dam itself or relatively short tunnels
or channels. Energy dissipation works will normally be required at the
downstream side and it is an advantage if these facilities can be shared with other
outlets such as high head spillways or irrigation outlets. It is advantageous if the
flushing facilities discharge to the downstream channel well away from any
power station outlets as any local deposition of sediments will increase tailwater
levels and reduce power output.
The reservoir itself requires a detailed survey to establish its topography. This
is required to check whether the reservoir basin is a suitable shape for sediment
flushing and also to provide input data for detailed modelling of the
sedimentation process within the reservoir.
Hydrological investigations
It has been stated that there are certain requirements for successful sediment
fiushin2: which are related to the amount of water available and its reliability year
on yea"r and season by season. Hence inflows to the reservoir need to be
127
-f
EVACUATION OF SEDIMENTS
-!
128
~-
-/
:4
J
SITE-SPECIFIC INVESTIGATIONS
new reservoirs are being investigated. When used on existing reservoirs they have
the added benefit of measured sedimentation data for verification purposes.
Computer simulations of reservoirs ideally use representative, long-term
sequences of water and sediment inflows to the reservoir. The models are capable
of looking at the effectiveness of various aspects which affect reservoir
sustainability over periods of up to 50 or 60 years, including:
measures to reduce the amount of sediments entering reservoirs such as
catchment conservation or upstream storage
measures to manage the sedimentation process within reservoirs such as
variations in the operating rule curves for the reservoir
measures to evacuate sediment from the reservoir including dredging and
sediment flushing.
Q)
129
7. References
Ackers, P. (1993). Sediment transport in open channels: Ackers and \Vhite update. Proceedings
of the Institution of Civil Engineers Water, Maritime and Energy, 101, 247-249.
Albertson, M. L., Malinas, A. and Hotchkiss, R. (eds) (1996). Proceedings of the International
Conference on Reservoir Sedimenation, Fort Collins, Colorado, USA.
Anderson, H. W. (1975). Sedimentation and turbidity hazards in wildlands. In: Watershed
Nfanagement. Proceedings of a Symposium of the Irrigation and Drainage Division. ASCE,
New York, pp. 347-376.
Atkinson. (1984). Consolidation of reservoir.peposits. In: Sedimentation in reservoirs in the Tana
River basin, Kenya, R. Wooldridge (ed.). Report OD 61, HR Wallingford, UK.
Atkinson, E. (1996). Feasibility of flushing sediment from reservoirs. Report OD 137, HR
Wallingford, UK.
Atkinson, E. (1998). Reservoir operation to control sedimentation: techniques for assessment.
Proceedings of the Conference of the British Dams Society, Bangor.
Attewill, L. J. S., \vrute, W. R., Tariq, S. M. and Bilgi, A. (1998). Sediment man(!gement shldies
of Tarbela Dam, Pa..1dstan. Proceedings of the Conference of the British Dam Society,
Bangor.
Basson, G. R. and Olesen, W. (1997). Modelling flood flushing. International Water Power and
Dam Construction, 49, No.6.
Basson, G. R. and Rooseboom, A. (1997a). Dealing with reservoirsedinuntation: guidelines and
case studies. Bulletin 115, ICOLD, Paris.
Basson, G. R. and Rooseboom, A. (1997b). Dealing with rese:rvoir sedimentation. "Vater
Research Commission Report No. TT91197, Pretoria.
Bhargava, D. N., Narain, L., Tiagi, S. S. and Gupta, P. P. (l987).Sediirientation problems at low
dams in the flimalayas. Water Power and Dam. Construction, Jan., 30-33.
Binnie & Partners. (1986). Report on Guanting reservoir - sedimentation. Report for
Government of PRC (Beijing Municipal Engineering Administration Division), under
assignment by ODA.
Bordas, M. P. and Canali, G. E. (1980). The influence of land use and topography on the
hydrological and sedimentological behaviour of basins in the basaltic r~gion of South Brazil.
In: The influence of man on the hydrological regime with special reference to representative
and experimental basins. International Association of Hydrological Sciences, Publication No.
130, pp. 55-60.
Branski, J. (1975). Oeena denudacji dorzecza Wisley na podstawie wynikow pomiarow
rumowiska unoszonego. Prace Instytuto Nfeteorologii i Gospodarki Iodnej, 6, 1-58.
efficiency of reservoirs. Transactions of the A,m.erican Geophysical
Bnme, G. 11. (1953).
Union,
No.3, Washington, DC, 407-418.
Budyko, !vi. 1. (1956). Heat balance of the Earth's surface (in Russian). Giclrvn1eteoizdat,
Leningrad, 255 pp.
133
EVACUATION OF SEDIMENTS
134
, REFERENCES
Gregory, K. J. and vValling, D. E. (1973). Drainage basin form and plVcess. Edward Arnold,
London
Griffiths, G. A. (1981). Some suspended sediment yields from South Island catchments, New
Zealand. Water Resources Bulletin, 17, 662-67l.
Griffiths, G. A. (1982). Spacial and temporal variability in suspended sediment yields of North
Island basins, New Zealand. Water Resources Bulletin, 18, 575-584.
Hardjowitjitro, H. (1981). Soil erosion as a result of upland traditional cultivation in Java Island.
In: Proceedings of South-East Asia Regional Symposium on Problems of Soil Erosion and
Sedimentation, T. Tingsanchali and H. Eggers (eds). Asian Institute of Technology, Bangkok,
pp.173-179.
Hwang, J. S. (1985). The study and planning of reservoir desilting in Taiwan.
International, 10, 7-13.
~Vater
International Commission on Large Dams (ICOLD). (1988). World register of dams: updating.
ICOLD, Pa..ris.
International Commission on Large Dams (ICOLD). (1997). General report on question 74.
ICOLD 19th Congress, Florence, Vol. 3.
International Commission on Large Dams (ICOLD). (1998). World register of large dams. CD
ROM.
In[ernational Research and Training Centre on Erosion and ~edimentation (IRTCES) (1985).
Lecture notes of the training course on reservoir sedimentation. IRTCES, Tsinghua"
University, Beijing, Chh"la.
Institute of vVater Conservacy and Hydroelectric Power Research (IWHR). (1983). The rational
operation of drawdownfiushing in Shuicaozi reservoir on the flU River. Summary in English
of Paper SR8301, IWHR.
Jaeggi, A. L. and Kashyap, B. R. (1984). Desi1ting of Baira reservoir of Baira Siul project.
Irrigation and Power, Oct., 375-380.
Jansson, M. B. (1988). A global survey of sediment yield. Geographical Annals, No. 70, Series
A (1-2), 81-98.
Jansson, M. B. (1992). Suspended sediment outflow from Cach! reservoir during the flushing in
1990. In: Sedimentological Studies in the Cach{ reservoir, Costa Rica, IVI. B. Jansson and A.
Rodriguez (eds). UNGI Report No.8l. Department of Physical Geography, Uppsala
University, Sweden.
Jantawat, S. (1985). An overview of soil erosion and sedim,entation in Thailand. In: Soil erosion
and conservation, S. A. El-Swaify, W. C. Moldenhauer and A. Lo(eds). Soil Conservation
Society of America, Ankeny, Iowa. pp. 10-14
Jarecki, E. A. and Murphy; T. D. (1963). Sediment withdrawal investigation - Guen'lsey
reservoil: Paper No. 91, Miscellaneous Publication 970, US Department of Agriculture.
Jowett, I. (1984). Sedimentation in New Zealand hydroelectric schemes. Water International, 9,
172-176.
Jolly, J. P. (1982). A proposed method for accurately calculating sediment yields from reservoir
deposition volume. In: Recent developments in the explanation and prediction of erosion and
sediment yield (Proceedings from the Exeter symposium, July 1982), IAHS Publication No.
137.
Jowett, I. (1984). Sedimentation in New Zealand hydroelectric schemes.
172-176,
~Vater
International, 9,
135
EVACUATION OF SEDIMENTS
Judson, S. and Ritter, D. F. (1964). Rates of regional denudation in the United States. Journal of
Geophysical Research, 69, 3395-3401.
Krumdieck, A. and Chamot, P. (1979). Sediment flushing at Santa Domingo reservoir. Water
Power and Dam Construction, Dec., 172-176.
Lagwankar, V. G., Gorde, A. K., Borikar, D. A. and Patil, K. D. (1995). Trends in reservoir
sedimentation in India. In: 6th International Symposium, on River Sedimentation, New Dehli,
pp.91-111.
J,
Lai, 1. S. and Shen, H. W. (1996). Flushing sediment through reservoirs. Jountal of Hydraulic
Research, IAHR, 34, No.2.
Lane, E. W. and Koelzer, V. A. (1953). Density of sediments deposited in reservoirs. Report No.
9: A study of methods used in measurement and analysis of sediment loads in streams, St Paul
l!te~. States Engineering District, St Paul, Minnesota, US.
I'
~.
,
!
Li, Y. H. (1976). Denudation of Taiwan Island since the Pliocene Epoch. Geology, 4, 105-107.
.~
J
J
I
!
,-1..
Morgan, R. P. C. and Davidson, D. A. (1986). Soil erosion and conservation, Longman Group,
UK.
Morris, G. L. and Hu, G. (1992). HEC-6 Modeling of Sediment Management in Loiza Reservoir,
Puelto Rico. In: Hydraulic Engineering Water Forum '92, M. Jennings and N. Bhowmik
(eds), ASCE, New York, pp. 630-635.
Monis, G. L. and Fan, J. (1997). Reservoir sedimentation handbook: design and management of
dams, reservoirs and watershedfor sustainable use. McGraw Hill, New York ..
i
r
Mou, J. and Meng, Q. (1980). Sedilnent delivery ratio as used in the ' computation of the
watershed sediment yield. Beijing, China.
Neil, C. R. and Mollard, J. D. (1982). Erosional processes and sediment yield in the upper
Oldm,an river basin Alberta, Canada. Recent developments in the explanation and prediction
of erosion and sediment yield. Proceedings of a symposium held during the first lARS
scientific assembly at Exeter, 1982, Publication No. 137.
Noble, E. L. (1965). Sediment reduction through watershed rehabilitation. In: Proceedings of the
Federal Inter-Agency Sedimentation Conference, 1963. US Department of Agriculture,
Miscellaneous Publication 970, pp. 114-123.
136
REFERENCES
O'Loughlin, C. L., Rowe, L. K. and Pearce, A. 1. (1980). Sediment yield and water quality
responses to clearfe1ling of evergreen mixed forests in western New Zealand In: The influence
of man on the hydrological regime }vith special reference to representative and experimental
basins. International Association of Hydrological Sciences, Publication No. 130, pp.
285-292.
Ongweny, G. S. (1979). Patterns of sediment production within the upper Tana basin in eastern
Kenya. In: The hydrology of areas of low precipitation (Proceedings of the Canberra .
Symposium, December 1979), IAHS Publication No. 128.
Palmieri, A. (1998). The role of technical solutions in the context of international developnzent
and dam investments. International Atomic Energy Agency, Consultant Meeting, Thematic
Planning for Technical Cooperation Isotope Hydrology in Danl Management, Vienna.
Paul, T. C. and Dhillon, G. S. (1988). Sluice dimensioning for desilting reservoirs. Water POrver
& Dam Con.struction, May, 40-44.
Pickup, G., Higgins, R. J. and vVarner, R. F. (1981). Erosier. and sediment yield in Fly River
drainage basins. In: Erosion and Sediment Transport in Pacific Rim Steeplands, T. R. H.
Davies and A. 1. Pearce (eds). International Association of Hydrological Sciences, Publication
No. 132, pp. 438-456.
Pidwimy, M. (1999). Climatic regions of the world: clim.a.tology - the geographic study of the
atmosphere, introduction to physical geography II. Department of Geography, Okanagan
University College, http://www.artslouc.bc.ca/geog/G121Iln.html. pp. 1-9.
Qian. N. (1982). Reservoir sedimentation and slope stability technical and environmental effects.
14tll Congress on Large Dams, leaL!!, Rio de Janeiro. __
RaInfrez, C. and Rodriguez, A. (1992). History of the Cachf Reservoh; Costa Rica. UNOI Report
No. 81, Department of Physical Geography, Uppsala University, Sweden.
Rhoades, E. D., Welch, N. H. and Coleman, G. A. (1975). Sediment-yield characteristics from
unit source watersheds. In: Present and prospective technology for predicting sediment yields
and sources. US Department of Agriculture. Publication No. ARS-S-40, pp. 125-129.
Rien6ssl, K. and Schnelle, P. (1982). Sedimentation of small reservoirs in the high Alps. ICOLD
14th Congress, Rio de Janeiro, Question 54, Reply 54, pp. 65-82.
Rooseboom, A. and Van Harmse. H. R. (1979). Changes in sediment load of the Orange River
1929-1969. Hydrology of areas of low precipitation. IAHS publication 128, proceedings of
conference, Canberra, Dec.
Salas et al. (1997a). Capacity situation in Spanish reservoirs. ICOLD 19th Congress, Florence,
Vol. 3, Question 74, Reply 53.
Salas et at. (1997b). Sediment yield at Spanish reservoirs and its relatior.ship with the drainage
basin area. ICOLD 19th Congress, Florence, Vol. 3, Question 74, Reply 54.
Shahin, M. IV!. A. (1993). An overview of reservoir sedimentation in some African river basins.
In: Sediment problems: strategies for monitoring, prediction and control, lARS Publication
No. 217, Wallingford UK, pp. 93-100.
Sharma, K. D. and Chatterji, P. C. (1982). Sedimentation in Nadis in the Indian arid zone.
Journal of Hydrological Sciences, 27, 345-352.
Shen, H. "'vY and Lai, 1. S. (1996). Sustain reservoir useful life by flushing sediment. International
Joumal of Sediment Research, IRTCES, 11, No.3.
Shen, H . \v. (1999). Flushing sediment through resen'oirs. Journal of Hydraulic Research, 37,
No.6.
137
EVACUATION OF SEDIMENTS
Strakhov, N. M. (1967). Principles of Lithogenesis, Vol. 1, 245 pp. Oliver and Boyd,
Edinburgh.
Subramanian, V. (1982). Erosion and sediment yield in Krishna River Basin, India Recent
developments in the explanation and prediction of erosion and sediment yield. Proceedings of
a Symposium held during the first lARS scient~fic assembly at Exeter, 1982; Publication No.
137.
Swiss National Committee on Large Dams (SNCOLD) (1982). General paper ICOLD 141h
Congress, Rio de Janeiro, pp. 889-958.
Ta Wei Soong, M. and Yean Zhao. (1994). Flood and sediment characteristics of the Lower
Yellow River in China. Water International, 19, 129-137.
Tabuteau, M .. M. (1960). Etude graphique pour les consequences hydro-erosives du climat
mediterrane. Association Geographique Fran{:aise Bulletin, 295, 130-142.
Tolouie, E. (1993). Reservoir sedilnentation and de-siltation. PhD thesis, University of
Birmingham, UK.
United Nations Educational, Scientific and Cultural Organisation (UNESCO). (1985). Methods
of computing sedimentation in lakes and reservoirs. A contribution to the International
Hydrological Programme, IHP-II projectA.2.6.1 Panel (Stevan Bruk, Rapporteur) (essentially
the same as IRTCES, 1985).
United Nations Education and Scientific Organisation (UNESCO) (1990). UNESCO sourcebook
in climatology for hydrologists and water resource engineers. UNESCO Publication,
Catalogue No. 89598, Paris.
Ursic, S. 1. and Dendy, E E. (1965). Sediment yields from small watersheds under various land
uses and forest covers. In: Proceedings of the Federal Inter-Agency Sedimentation Conference
1963. US Department of Agriculture. Miscellaneous Publication. No. 970, pp. 47-52.
US Global Change Research Information Office. (1999). Soil erosion and sediment erosion,
pp. 1-3. Web site: http://www.gcrio.org/geo/
Walling, D. E. (1994). Measuring sediment yields from river basins. In: Soil erosion research
methods, R. Lab (ed.). Soil and Water Conservation Society, Ankeny, Iowa.
Walling, D. E. and Kleo, A. R. (1979). Sediment yields of rivers in areas of low precipitation,
a global view, pp. 479-493. The Hydrology of Areas of Low Precipitation, lARS Publication
No. 128, Wallingford, UK.
Walling, D. E. and Webb, B. W. (1983). Patterns of sediment yield. In: Background to palaeohydrology, K. J. Gregory (ed.). John Wiley, New York, pp. 69-100.
Walling, D. E. (1984). Muddy waters move mountains. The Geographical Magazine, 262-267.
Walling, D. E. and Webb, B. W. (1996). Erosion and sediment yield: global and regional
perspectives. IARS Publication No. 236, Wallingford, UK.
Webb, R. M. T. and Soler-Lopez, L. R. (1997). Sedimentation history of Lago Loiza 1953-1994.
USGS, San Juan.
White, P., Labadz, 1. C. and Butcher, D. P. (1996). The management of sediment in reservoired
catchments. Chartered Institution of Water and Environmental Management Journal, 10,
June, 183-189.
White, W. R. (1982): Sedimentation problems in river basins. Project 5.3 of the International
Hydrological Programme. UNESCO.
138
REFERENCES
vVhite, vY. R. and Bettess, R. (1984). The feasibility of flushing sediments through reservoirs.
Proceedings of the Symposium 'Challenges in African Hydrology and Vlt~1ter ResoLlrces\
Harare, July, 1984. IAHS Publication No. 144.
Wilson, L. (1969) . Les relations entre les processus geomorphologiques et Ie climat modern
comme methode de paleoclimatologie. Revue de Geographie PhysiqLie et de Geologie
Dynamique, Ser. 2, 11, 303-314.
Xia, M. (1986). New desilting methods used in Heisonglin reservoir: lateral erosion on
floodplain . In: Collected Research Papers, Northwest Institution of Hydraulic Research, 2,
Part 2, Yangling.
Yoon, Y. N. (1992). The state and the perspective of the direct sediment removal method from
reservoirs. International Journal of Sediment Research, 7, No.2, 99-116.
Zhang Hao, Xia Mai-ding, Chen Shi-ji, Li Zhen-Wu Xia Heng-Bin ,Jiang Nai-Sen and Lin Benwen (1976). Regulation of sediments in some medium and small sized reservoirs on heavily
silt laden streams in China. 12th Congress on Large Dams, ICOLD, ~lexico.
Zhang Qishun and Long Yuqian. (1980). Sediment problems of Sanmenxia reservoir. 1st
Symposium on Reservoir Sedin'Lentation, Beijing, pp. 707-716.
139
Be lio rop
r
I
I
I
II
I
I
8. Bibliography
1
!
!
i
Abdelhadi, :M. L. (1995). Environment and socio-economic impacts of erosion and sedimentation in Northern African countries. 6th Symposium on River Sedimentation, New Delhi, Vols.
1-4, pp. 1141-1152.
Ackers, P. and Thompson, G. (1987). Reservoir sedimentation and influence of flushing. In:
Sediment transport in gravel bed rivers, C. R. Thome, 1. C. Bathurst and R. D. Hey (eds).
John Wiley and Sons, pp. 845-868.
Ackermann, Wbite and Worthington (eds). (1973). Geophysical 1'IIonograph No. 17, American
Geophysical Union, Washington.
Adams, 1. (1980). High sediment yields from major rivers of western southern Alps, New
Zealand. Nature, 287, 88-89.
Ahmed, T. E. S. (1992). Sedimentation in Sudan mUlti-purpose reservoirs. 5th International
Symposium on River Sedimentation, Karlsruhe.
Aitken, A. P. (1981). Aspects of erosion and sediment transport in Java. Problems of Soil Erosion
and Sedimentation, South-East Asia Regional Symposium, Thailand.
Ando, N., et al. (1994). SedLllent removal project at :NIiwa Dam. ICOLD 18th Congress, Vol. Ill.
Question 69, Reply 27.
Annandale, G. W. (1987). Reservoir sedimentation. Elsevier Science Publishers, Amsterdam.
Anon. (1982). Sedimentation of small reservoirs in the high Alps. ICOLD 14th Congress, Rio de
Janeiro, Question 54, Reply 5.
Anon. (1982). Estimation of silting load wid capacity loss rate in irrigation reservoirs in Korea.
ICOW 14th Congress, Rio de Janeiro, Question 54, Reply 12.
Anon. (1982). Some problems concerning reservoir sedimentation on Romanian rivers. ICOLD
14th Congress, Rio de Janeiro, Question 54, Reply 14.
Anon. (1982). Sedimentation aspects of the :tYnssouri river dams. ICOLD 14th Congress, Rio de
Janeiro, Question 54, Reply 15.
Anon. (1982). Environmental impact a.l1d control of reservoir sedimentation. ICOLD 14th
Congress, Rio de Janeiro, Question 54, Reply 23.
Anon. (1982). Studies on sediment transport in mountain streams of Eastern Alps. ICOLD 14th
Congress, Rio de Janeiro, Question 54, Reply 28:
Anon. (1982). Reservoir sedimentation - technical and environmental effects in Indonesia.
ICOLD 14th Congress, Rio de Janeiro, Question
Reply 31.
Anon. (1982). Investigation of average long-term suspended sediment discharge in Brazil.
ICOLD 14th Congress, Ria de Janeiro, Question 54, Reply 32.
Anon. (1982). Silting of reservoirs in southern Brazil - first results and methodological
approach. ICOLD 14th Congress, Rio de Janeiro, Question 54, Reply 33.
Anon. (1982). Sediment deposition in reservoirs on the River Tana, Kenya. ICOLD 14th
Congress, Rio de Janeiro, Question 54, Reply 37.
143
EVACUATION OF SEDIMENTS
Avendano Salas, C., Sanz 1v10ntero, E. and Gomez Montana, J. L. (1997). Sediment yield at
Spanish reservoirs and its relationship with drainage basin area. ICOLD 19th Conference,
Florence 1997, Vol. 3.
Bister. (1991). Ageing of dams. ICOLD, Vienna, Question 65.
Biswas, A. K. (1991). Water for sustainable development in the 21st century perspective. Water Intemational, 6, No.4, 219-224
a global
Bollinne, A. (1985). Adjusting the universal soil loss equation for use in Western Europe. In: Soil
erosion and conservation, S. A. El-Swaify, W. C. Moldenhauer and A. Lo (eds). Soil
Conservation Society of America, Ankeny, Iowa, pp. 206-213.
Borland, W. M. and Miller, C. R. (1958). Distribution of sediment in large reservoirs. Journal of
the Hydraulics Division, ASCE, 84, No. HY2.
Borland, W. M. and Miller, C. R. (1960). Distribution of sediment in large reservoirs.
Transactions ASCE, 125.
Brune, G. M. (1953). Trap Efficiency of Reservoirs. Trans A117,erican Geophysical Union, 34, No.
3,407-418.
Bruk, S. and Zebidi, H. (1996). Reservoir sedimentation. Proceedings of the St Petersburg
Workshop (May 1994), International Hydrological Programme, UNESCO, Paris.
Buff, B., et al. (1997). Survey of sedimentation in dams in the free sate of Saxony, Federal
Republic of Germany. ICOLD Conference.
Butcher, et al. (1992). Assessment of catchment erosion in the southern Penines using reservoir
sedimentation monitOling. International Association of Hydrological Sciences, No. 210.
Buttling and Shaw. (1973). Predicting the rate and pattern of storage loss in reservoirs. 11th
ICOLD Congress, Madrid, Q40-R37.
Central Board of Inigation and Power. (1981). Sedimentation studies in reservoirs. Technical
Report No. 20, Vol. I and II. Central Board Irrigation and Power, New Delhi, India.
Chain. (1986). Reservoir sedimentation and desilting in Taiwan. Proceedings Conference Water
Forum, Vol. 1, World Water Issues in Evolution, USA.
Chen, 1. L. (1994). Sedimentation studies at Three Gorges. International Water Power and Dam
Construction, Aug., 54-58.
Coleman, D. J. and Scatena, F. N. (1986). Identification and evaluation of sediment sources.
Drainage Basin Delivery IAHS. Pub. 159, Wallingford, UK, pp. 3-18.
Crowder, B. M. (1987). Economic costs of reservoir sedimentation: a regional approach to
estimating cropland erosion damages. Journal of Soil and Water Conservation, 42, No.3,
194-197.
Dai, D. (1991). River sediment problems in China. In: Water conservancy in China. Water
Resources and Electric Power Press, Beijing.
Delft Hydraulics, et al. (1992). The control of reservoir sedimentation Ministry of Development Cooperation, The Netherlands.
a literature review.
Dendy, F. E. (1967). Sediment deposition in reservoirs in the United States. ARS #41-137, US Department of Agriculture.
Paper
Dendy, F. E., Champion, W. A. and Wilson, R. B. (1973). Reservoir sedimentation surveys in the
USA. In: Man-nwde lakes: their problems and environmental effects, W. C. Ackerman, E. F.
White and E. B. W0l1hington (eds). Geophysical Monograph, No. 17, AmeIican Geophysical
Union, Washington, DC.
~I
144
BIBLIOGRAPHY
Dendy, F. E. and Champion, W. A. (1978). Summary of reservoir deposition surveys made in the
USA through 1975. US Department of Agriculture, rvIiscellaneous Publication No. 1362.
De Vera and Rewtarkulpaiboon. (1981). Analysis of sediment yield and estimated soil loss of
selected Philippine watersheds. Proceedings of SouthEast Asian Regional Symposium orz
Problems of Soil Erosion and Sedimentation, Bangkok, Thailand.
Dunne, T. (1979). Sediment yield and landuse in tropical catchments. Joun-zal of Hydrology, No.
42, 281-300.
Engelman, R. and Leroy, P. (1994). Sustaining water: population and the future of renewable
water supplies. Population Action International, Washington DC.
Fan, J. and Morris, G. L. (1992). Reservoir sedimentation II: reservoir desi1tation and long-term
storage capacity. Journal of Hydraulic Engineering, 118, No.3, ASCE, 370-384.
Fan, 1. and Morris, G. L (eds). (1993). Notes on sedinzent .'1-zanagement in reservoirs: natiortal
and international perspectives. Federal Energy Regulatory Commission, Washington DC.
Fan, J. (1995). An overview of preserving reservoir storage capacity. International ~Vorkshop on
Reservoir Sedil7'zentation, San Francisco. Federal Regulatory Commission, Washington, DC.
Fan, J. and Pan, S. (1996). Control of sediment deposition by flood flushing. Proceedings of the
International Conference on Reservoir Sedimentation, Fort Collins, Colorado.
Fan, S. (1988). Twelve selected computer stream sedimentation models developed in. the United
States. Interagency Sedimentation Work Group and Federal Energy Regulatory Commission.
Fan, S. (1989). Reservoirs. In: Civil Engin.eering guidelines for planning and designing
hydroelectric developments, American Society of Civil Engineers.
Fan, S. (1990). lvIajor sedimentation issues and activities at the selected Federal Agencies.
Federal Energy Regulatory Commission.
Fan, S. (1999). Sediment management at hydropower reservoirs. International Journal of
Sediment Research, 14, No.2, 391-397.
Fleming, G. (1969), Design curves for sllspended load estimation. Proceedings of the Institution
of Civil Engineering, 43, No. 1.
Frenette, M. and Julien, P. Y. (1986). Advances in predicting reservoir sedimentation.
Proceedings of the 3rd International Symposium River Sedimentation, U. lVIississippi,
Oxford.
Galay, V. (1987). Erosion and sedimentation in the Nepal Himalaya. "Vater and Energy
Commission Secretariat. Kathmandu.
Garde, R. J. and Kothyari, V. C. (1987). Sediment yield estimation. Journal of Irrigation and
Power, 44, No.3.
Gogus and Yalcinkaya. (1992). Reservoir sedimentation in Turkey. 5th International Symposium
on River Sedimentation, Karlsruhe.
Goldsmith, E. (1985). Sedimentation - the way of all dams - the social and environmental
effects of large dams. European Ecological Action Group, Vol. 1, pp. 226-239.
Gong, S. and Jiong, D. (1979). Soil erosion and its comrol in small watersheds of the loess
plateau. Saentia Sinica, 22, No.1, 1302-1313.
145
EVACUATION OF SEDlfVlENTS
Griffiths, G. A. (1979). High sediment yield from major rivers of western southern Alps, New
Zealand. Nature, No. 282, 61-63.
Hadley, R. F. and Shawn, L. M. (1976). Relation of erosion to sediment yield. Proceedings of the
3rd Federal Interagency Sediment Conference Water Resources Council, Washington DC,
pp.1.l32-1.139.
Harrison, L. L. and .Fan, S. (1996). Sedil1'lent management solutions for FERC licensed
reservoirs. Draft Staff Paper, Federal Energy Regulatory Commission.
Harrison, L. L. and Fan, S. (1997). Needs for sediment management at non-federal reservoirs.
International Water Power Conference.
Holeman, J. N. (1968). Sediment yield of major rivers of the world. Water Resources Research,
4, No.4.
Hotchkiss, R. H. (1999). Reservoir sedimentation: research needs and refocusing perspectives.
International Journal of Sediment Research, 14, No.2.
International Association for Hydrological Sciences (IAHS) (1979). The hydrology of areas with
low precipitation. IAHS Publication 128, Wallingford, UK.
International Association for Hydrological Sciences (IAHS) (1986). Challenges in African
hydrology and water resources planning. IAHS Publication 144, Wallingford, UK.
International Association for Hydrological Sciences (IAHS) (1986). Drainage basin sediment
delivery. IAHS Publication 159, Wallingford, UK.
International Commission on Large Dams (ICOLD) (1984). Deterioration of dams and
reservoirs.
International Commission on Large Dams (ICOLD) (1989). Sedimentation control of reservoirs.
Guidelines.
International Commission on Large Dams (ICOLD) (1997) . ICOW 19th Congress, Florence,
Vol. 3, Question 74, Reply 34.
Jiang, J. and Fu. (1997). Problems on reservoir sedimentation and measures for reducing
sediment deposition in China. ICOW 19th Congress, Florence, Vol. 3, Question 74, Reply
5.
Julien, P. Y. (1995). Erosion and sedilnentation. Cambridge University Press, Cambridge, UK.
Lauffer, H. and Sommer, N. (1982). Studies on sediment transport in mountain streams of the
eastern Alps. ICOLD conference, Vol. ill, Question 54, pp. 431-451.
Liu, X., Yueping, H. E., Zhang, X., Qin, Y. and Shuyou, C. A. O. (1999). A study of sediment
problem in the Gongzui reservoir. International Journal of Sediment Research, 14, No.2.
Meade, R. H. and Parker, R. S. (1985). Sediment in rivers in the United States. National Water
Summary 1984, USGS Water Supply Paper 2775, Washington.
Milliman, J. D. and Meade, R. H. (1983). Worldwide delivery of sediment to the oceans. Journal
of Geology, 91, No.1, 1-21.
Mizuyama, T. (1989). Sediment yield and river bed change in mountain rivers. Proceedings of
the IAHR, International Workshop on Fluvial Hydraulics of Mountain Regions, Trento, Italy,
pp.1-5.
Monis, E. L. (1995). Reservoir sedimentation and sustainable development in India: problem
scope and remedial measures. 6th Symposium on River Sedimentation, New Delhi.
Newson, M. D. (1998). The role of geomorphology in monitoring and managing river sediment
systems. Journal of the Chartered Institute for Water and Environmental Management, Feb.
146
-f
".t.,","
;,
.1
BIBLIOGRAPHY
,,~
I1
f
Palmieri, A., Shah, F. and Dinar, A. (1998). Reservoir sedimentation and the sustainable
management of dams. World Congress of Enviromnerztal and Resource Economists, Venice',
June, pp. 23-27.
i
i
1
j
I
I
I
~
~
~
Roehl, J. "V. (1962). Sediment source areas, delivery ratios and influencing morphological
factors. IAHS, Publication No. 59.
Rogers, R. S. and Schumm, S. A. (1991). The effect of sparse vegetation cover on erosion and
sediment yield. Journal of Hydrology, 123, Elsevier Science Publication, Amsterdam.
!
~
basins.~
Thesis. University-"of
147
EVACUATION OF SEDIMENTS
United States Department of Agriculture (USDA) (1943). The control of reservoir silting.
Miscellaneous Publication No. 521.
Varma, C. V. J., Rao, A. R. G. and Natarajan, S. (1992). Sedimentation of Indian Reservoirs. An
assessment. 5th Symposium on River Sedimentation, Karlsruhe, pp. 919-923.
Veltrop, J. A. (1992). The role of dams in the 21st century. US COLD, Denver.
Walling, D. E. (1989). The erosion problem. International Journal of Sediment Research, 14, No.
1, 1-11.
Walling, D. E. and Probst, J. L. (1997). Human impact on erosion and sedimentation. IAHS
Publication No. 245, Wallingford, UK.
Water Power. (1979). World's largest capacity reservoirs. Intenwtional Water Power and Dam
Construction, 31, Nov., 98.
White, W. R. and Rofe, B. H. (1996). Responsibilities and lists associated with large reservoirs.
Intenwtional Conference on Aspects of Conflicts in Reservoir Development and Management,
London.
Wolman, M., et al. (1989). Erosion control and reservoir deposition. Proceedings of Bilateral
Seminar on Problems in the Lower Reaches of the Yellow River, China. pp. 139-163. Kluwer
Academic Publishers, The Netherlands.
148
Appendix I.
Reservoir data
!
!
This appendix provides data which ICOLD has obtained from member
countries. Table A1.1 provides, on a country by country basis, data on
reservoir capacity development in the twentieth [century. Table A1.2
provides sedimentation data for individual dams.
lSi
Table AI.I.
Region
Country
Notes
Africa
Africa
Africa
Africa
Africa
Aldca
Al'rica
Africa
Africa
Africa
Africa
Africa
Africa
Africa
Africa
Africa
Afiica
Africa
Africa
ArriclI
AfriclI
Afril:a
Africa
Africa
Africa
Africa
Africa
Africa
Africa
Africa
Africa
Africa
Angola
Benin
Botswana
Cameroon
Congo
Congo (ORC)
Ethiopia
Gabon
Ghana
Guinca
Ivory Coast
Kenya
Lesotho
Liberia
Madagascar
Mal awi
Mali
Mlluritius
Mozambiljue
Namibia
Nigeria
Scnegal
Seychelles
Sierra Leone
South AI'rica
Sudan
Swaziland
Tanzania
Togo
Uganda
Zambia
Zimbabwe
Africa
Total
C. Asia
Kazakhstan
Kirghizstan
Tadjikis(an
Uzhekistan
Volume
3
213
967
575352
595
12
11
7
14
89786
23303
28182
6761
7482
2118
4026
483
2
I
c
c
1910--19
1900--9
1920--29
1930--39
1940-49
i950-59
Av. vol No. Volume No. Volume No. Volume No. Volume No. Volume No. Volume
630
9446
1734
867
77
231
14325
1592
44
22
5319
443
2624
328
220
220
150279 37570
237
119
37917
1724
162
2428
1970
281
0
0
431
43
I
4
13 440
6720
61
7
57 102 9517
51
662
40414
898
11520 5760
I
1
22
22
57
30583
5587
1117
42
250
1135
1135
I
I
200
200
47
16
187117
878
IS
2
9
2
12
8
I
4
2
22
15
7
I
10
4
2
9
6
13
45
2
2
I
539
5
6
<1900
Totals
No.
C. Asia
C. Asia
C. Asia
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
' 0
0
0
1267
0
0
0
I
0
125
1960-69
No.
Volunic
No.
142
44
I
5
2
0
I
I
0
8
2417
7
146
0
36
35
1925
0
0
I
0
0
Volume
7
2
293
<>
ISO 000
237
9
16
3
0
0
2
I
I
0
8
1536
4
3
I
I
6
36
33
295
15079
I
0
1100
16
0
I
2
2
I
2
I
1980-89
1970-79
No.
Volume
Volume
5
I
5124
1710
10915
()
I
0
2
0
14
()
5
0
12
0
11
0
164
0
26
I
679
931
22
I
3534
0
15
0
173
0
413(>
24
85
3410
3
4
0
692
1600
192
220
0
87
14
2
0
28444
170
0
3
9
I
8339
2155
4
0
I
I
0
70
1950
13
125
0
3
16053
0
23
124
0
2
2194
0
177
()
n
0
(,
1134
()
I 1270
I
0
1
I 1270
44
0
5099
0
0
0
41
()
0
183266
2
36
6
1431
51
0
1016
0
82
0
1201.1
0
5
66
6440
164
358850
235
124018
219
46580
132
30900
17
1326
630
56888
723
180
2928
29878
20060
4160
4
4
3
6
6
3
3
2
4
2380
10542
1564
2390
140
13 300
15830
0
0
I
2
41
39
0
2
0
39
0
5
0
56
16
11
164
31
1696
27
4967
26
251
96
144
l'
3
0
I
I
0
0
8
I
2170
4
1273
98
9331
250
I
I
25
0
21
0
4
54260
222
14856
200
0
0
104
2
I
I
0
0
3
6
23
2269
()
C. Asia
Total
44
148032
3364
4790
17
60719
13
52207
14486
China
China
1851
649322
351
15
24057
177
26113
479
256549
583
103536
220
58090
120
57278
260 123667
1851
649322
351
15
24057
177
26113
479
256549
260 123667
2
1
1579
6500
70
48
340
126
3158
37
0
0
4
2101
0
0
267
39
0
0,
2
826
0
0
157
82
0
77
41
2368
41
983
China
Total
S. Asia
S.Asia
S.Asia
S. Asia
S. Asia
S. Asia
Afgh anistan
Danghtdesh
India
Nepal
Pakistan
Sri Lanka
4010
3
69
46
3158
6500
279548
144
23436
5816
S. Asia
Total
4131
318602
- )
45
0
0
I
3287
0
0
101
40
0
0
0
6658
0
0
0
35
0
2446
0
27
87
85
2560
46
3388
40
6658
213
0
0
9
25601
0
550
0
7
0
1329
414
0
19
12
36
557
224
30088
446
I)
o
z
3
I
I
4
707
0
61
I
0
8
0
0
Volumt!
144
0
No.
0
22
1824
4656
50
1135
I
1
2
No.
Volume
No.
I
I
()
ute
1990-1998
583
103 536
220
58090
120
57278
()
0
6500
55938 1128
0
0
7961
16
621
8
57076 1178
3
0
15209
II
527
8
38901
144
184
2727
116
0
21
0
1045
0
48
0
683
85119
0
0
0
71020 1152
72 812 1200
41956
137
1093
683
85119
()
o
"
m
V1
3:
m
Z
-f
V1
.Mid. E"sl
Mid. Easl
Mid. Easl
Mid . 1;",1
Mid.E'lsl
(vlid.E:ISI
Mid. Easl
Mill. Hasl
Ivlid. ['ast
Mid. Fasl
Mid. E'lSI
Aruh.'nill
Iii
Azerbaidj'lIl
CYI'I1IS
17
l.d>:IIIOII
Saudi Ambia
Syri'l
Turl,ey
5
311
41
1i25
I7H9
183K5
2B6
2S!!1
32935
':>7183
203
161
2114
342
109233
'lblal
895
22:161B
5'2
14
66
(iL."Ofgia
11';111
Inil(
JII.-.I:1I1
- - --
-----.-
N. "f,-il' a
N. Arri.:a
N. Africa
N. A!'rica
N. Afrim
1\lgcri;1
EUypl
Lihya
t\'tllruccCl
"ruilisia
- --.- - - - N. Afri':3
'lllial
_- _--0_._.N. AIII('ri~a Call:llla
N.A llwric'l USA
--
---
N. AlIler;":1 'lhlal
N. ElIr<lp"
N. Ellrllp"
N. Ellmp~
N. EllfllpC
N.Ell rll!,c
N. Ellrupe
N. ElIwpe
N. EUI"llI'"
N.l'lIl"1lp"
N. Europe
N. Eul'Opc
N. Eurup~
N. Eurllpc
N. Eurnpl!
N. Europe
N. EUIll(1c
N. EllrClfX'
N. Europe
N. Europe
Pacilic l~illl
j'al'ifiL: \(jill
P:tdlkRil\l
Pacific Ilim
Parili.: Rim
1"lcilkRill1
P;ll'ilic Rim
Pacilie Rim
Pa.:ilic Rilll
Pa~ilic Rim
P;wilic Rim
Pacilic Ililll
Ln
W
- --
---
()
II
()
(I
I)
I)
II
1/
()
175
()
()
2511
3
0
5
0
0
-- --
I)
0
n
II
I0
0
()
II
(I
t)
II
I)
'l
tl
IJ
I)
(I
I)
()
10
132
48
1638
652
5
6115
6T!4
517 670
9
665
47.'i~
- -I
3
()
1929
19 Inn
160SY
()
0
141
15
0
17
14
0
0
0
0
Il
0
0
0
9
14
0
403
0
(I
t)
II
II
II
6
I
1414
331
0
5
6
'2
6104
\0
1911
381
36~63
414
II
0
2
158
92
0
6
I
0
2
35
27
0
3
281
22
9')
19
2142
71
1S1.S
90
5202 153
484
H7
2
9ti
1077
7(-,5
70
86
89855
10 775
153
74348
21 OKS
161161
34312
16515
333
3301
6714
2808
186
124
17
774
33
2
240
10
14
I
627
95
21
6
144
30
21
5102
24
21
20
2050
2
I
1
7
0
17
10
0
3
43
15
50
._-
:ms
- - -- .
277 124
I
26
14
()
I)
1221
15
\I
543
83
14
8
I
2
3
/I
612
0
0
1532
54
30110
5
177
9
15
I
9
II
168
43
68429
\06
6
I
II
259
1620nO
290
2238
31
6423
0
41i
563
1137
0
29
61
473
0
0
5143
199
tl
40
0
4
37
12
93
6')696 349
84923
592755 1393
I
0
6
38
9
I
'2
I
25
0
5726
-II
0
21
826
60S
II
654
~ - --
()
906
2\J
31
j
I
'2
14
75
34
Il
4
51:2()
162
201
1859
0
14
2
.5
3
18
10929
3320
74J
II
II
954
276
10281l
27
1006
17
166
24457 117
19631
--
3
5
7639
2755
850
0
307
2
4
16
90
49
-~--.--
t-506874
r--
0
485
35
2994
1811
225700
47
5329
18216
0
3
3
5
7
2953
3998
SKI
\3
0
5\0
462
79
3
101
9
19
4
53
7
58
204
11 765
883
173475
41
7927
19 7~('
38!li!
Y
0
II
93
76
4
2
2
12
0
I
2
75
10
IJ
4
30
6
32
425
226517
382
195444
R6
II
45850
64
16
0
')
0
:2
37
3
20
"
0\7
3
68
t----.
16
5
1187
71
51
7
S'S\()
1608
93
12
121'13
5901
,II
22
9
9
106
255
86
2332
322
4
37
62
7
3(1'l3
832
1361
2112
94
2
137
102
12
\0
128
3504
191
16350
nll5
4
3
8
322
424
198
I
208
140
18
12
I
8
I
6
2556
6444
1138
4316
2907
0
875
1!50
367
336
341)20
5(11)
37,197
52
II
.6
51
II
12
31
57
160
332
I
4
31
12
II
100
53
493
7J
5655 135
~449
142
6(i74
44H
4371
0
2
14
885
12 632
3
I)
II
60
0
39
0
\0 995
248SS
t--
()
[I
II
"I -0
0
122
5815
15
195
1i32
)-125
2111
4057
0
I
3
54
6
4
I
0
1900
72
3133
181
420
5
0
0
0
0
0
0
672
100
0
0
3
2
96
48
7
0
277
3984
132
340
65
0
735
1I2U
9154
724
162265
20
11028
2510
541
116 155495
641 16 MIl
2539
2054
639
1575
(--
264
335
168 972
40
43811
455767
51 \07
2777
221
23
35
151
55
114
161
23
41 178
432092 142
160663 1251
I
2
I
--
122
/0
(- -
3526
459
64 050
9
69
0
J2
8
I
0
15
7
151
52
133
0
6
18
13
3
0
76
14
3490
8551
26000
32428
52495
20
3
I
21
4
593
13
231
6
1206
1756
28
21330
80
U
0
165
U
6
18437 295
0
I
--
787
2182
17
8176 55
61520 294
:!I)
22
490
192
III
77
15H
I)
75.)
0
0
595
197
II
II
II
III
165
II
22061
0
482
86
106
0
27
0
7
8
4
3
8
I
I
0
12
6
72
0
0
6
0
0
I
0
(I
131
190
29
350
5568
3053
II
160
I
327
12241
3
I
13
29
'/31
5000
0
373
0
0
'2
- - - - - --- I - : - - - - -
0
2
2
2
0
6
0
0
0
2
56
2298'2
98
(I
21
52Y
(I
()
6800
314
7(1
5
0
0
.5
I
0
0
' :\
5'/
188
3
0
120
18
248
J
I
CiK
7
16
'2
5
3
18
412
()
14
56
117
0
tl
'2
I
I
I
160011
I
-- ---
9JS 168
I)
(I
'1277
35
15
0
3
0
II
0
312
0
1
9
321l
69
91
28
19[1
20
517
.--- - -
2679
994
H66
6"1
79
1220
9259
53496
3598
731 637
206
3:l216
46 B8'1
6864
<In!)
(J
II
()
19:1
3'16
28
14
149
62
173
31
26
6\0
1029
163
52
R040
7
201
234 '1
13
12 12247 85 17679 82
246 III 747 296 19284 332
256 674
5364 142 36R6 28K 30994
._r---'t - - 1-- - --. - . 252
Iii
0
0
0
I
0
II
0
--
-.
t)
1597
81
--._0
()
0
0
0
I 3~6 860
16
5
20
3
UK
'Ihiw;ln
1111
IS
Ukr:linc
I<llr~:llN)
47:1
55
569
311
Sw'-'lIcn
N. Z"'II:md
I'. N.Guillca
l'elll
Phjlipllincs
18~
111
Rllssia
J{orl~il
289
16
Slnwnia
Auslralia
Chik
Fiji
Indonesia
Japan
' -'
4732
48
Ifl71100 27833
3]6
28
I~ 54\)
15~
l11fi5
II
I
I
7205 IM'14530
lkigilll\l
[klllll;1rk
I'illiand
Fr:mc.:
Gl'I"IIHIIIY
kdand
Ireland
Lllvi"
Lillluania
Luscmhllrg
Mllld'lI'ia
Nl'lherlalld,;
Nlww:IY
Polalld
99
6
12
92
8tl
112
1081
6
20ti
49<)
4399
25
32
7
68
17
77
1322
463
640
()
4
0
4K
1
51
8
5
6
12
2
II
2780
2U944
2RO
10493
52
631
92
4011
0
4044
653
327liU
6
,9
1.
I
3
<)
95
II
I
5
0
0
---II
I
0
46
0
175
236
111805
"
\I
()
II
0
JOUO
0
69
38307
93
43341
92
14
2
27
165
'187
6
19
9808
1222
153
9277
1579
63911
3139
614
36
7
271H
291
0
2
0
10011
27
198
68
3
0
14515
3004
217K
6726
0
\I
56
0
41997
16119
II
690
2425
505
157
52
299
3581U
3115
29888
75
4'1738
._257
-
10.
202
170
12
14
2
5
4
574
3586
5453
1053
8978
33)
1069
3003
I)
1)95
4
6
9
525
70958
531
II
28 116034
II
-0
-U
~
n
Table Ai.i.
Region
c
~
continued
Country
Volume
No.
1920-29
1910-19
1900-9
<1900
TOlals
NOles
1930-39
1940-49
1950-59
Av. vol No. Volume No. Volume No. Volume No. Volume No. Volume No. Volume
No.
1960--69
Volume No.
Volume
------
S. America
S. America
S. America
S. America
S. America
S.America
S. America
S.Americ:n
S.America
S. America
S. America
S.America
S. America
S.America
S. America
S. America
S. America
S.America
S.America
S.Amcrica
S.America
S.America
S.America
Antigua
Argentina
Bolivia
Brazil
Colombia
Costa Rica
Cuba
Dominican R.
Ecuador
EI Salvador
Guatemala
Guyana
Haiti
Honduras
Jamaica
Mexico
Nicaragua
Panama
Paraguay
Surinam
Trinidad
Uruguay
Vene..:uela
5
130380
372
540386
10913
2288
3755
2373
70()3
2430
460
44
I
9035
220
121253
1252
5184
33690
20
48
12335
155467
5
1304
53
922
223
254
77
216
226
486
115
22
1498 I 038913
694
a, d
a, b
d
S. America Total
100
7
586
49
9
49
II
31
5
4
2
I
8
2
536
4
5
4
I
4
5
74
0
0
77
0
0
0
0
1874
0
0
0
0
3
0
0
0
0
179
0
0
0
0
16
0
0
0
0
540
0
0
3
0
35
0
0
32
0
2029
0
0
Albania
Austria
Bosnia
Bulgaria
Croatia
Czech Rep.
Macedonia
Greece
Hungary
Italy
Ponugal
Romania
Slovakia
Spain
Switzerland
Yugoslavia
S. Europe
Total
306
148
25
180
29
118
18
46
15
524
103
246
50
1\87
156
69
117
10
4144
544
0
I
15
92
3220
14
0
100
9
1057
161
1665
0
0
4
2
0
14
3
16
0
0
I
I
0
220
39
20
2658
1
0
624
0
I
0
0
5
0
0
0
I
44
0
0
9678
447
0
11825
250
0
5
17
2
103
10
2
15
0
6
620
0
I
I
0
0
0
0
0
0
0
0
0
0
0
0
0
I
0
3
92
1966
II
296
27
5228
47
2323
0
0
I
0
0
0
0
17706
1841
56477
4094
4360
0
0
53
0
0
I
10
0
0
3
113
1
0
145162
45
85
182
----
62
----
0
7
0
0
0
4
0
2
0
12
5
0
0
0
0
0
0
7
0
0
0
17
0
0
0
57
3
0
0
3
0
0
0
7
0
0
0
16
0
I
0
44
5
0
30
84
76
0
0
0
0
0
35
0
0
0
28
0
0
0
38
0
I
0
656
94
0
4
0
0
0
2
0
0
0
79
I
0
0
41
1I
0
852 138
0
5
0
0
0
3
0
0
0
1375
22
0
0
636
303
0
0
4
0
0
0
10
I
1
0
57
3
2
1
45
10
2
2344 136
I
5
3
4
I
1057
6330
380
460
Volume
No.
~ ~
----
21
2
3
34593
0
85063
85
13
2
3
175
465
()
Volume
----
I
29
3
0
197
54
38298
1999
0
0
0
44
(I
20357
I
2
0
&04
21000
(I
532
0
4
0
3981
58 141283
44
1940
10
7085
133
I
16573
23
4000
12690
65
15293
21
I
42
2
2946
16 142726
0
3
357
8800
345
I
I
3
I
29
0
I
I
16
202
31466
322
147118
0
26
0
3126
o
3:
m
Z
--i
66377
-------
1
30
3
19
5
3922
424
21629
JO
577
712
2289
2145
151
1567
1211
5875
19
2159
1627
2062
611
19353
1703
703
125
26
4
54
4
23
3
3
6
43
15
74
9
195
18
II
3857
700
814
2865
24
729
76
2721
30
634
899
6097
428
7444
343
1502
629
42764
613
29163
73
25
9
78
4
23
8
4
5
77
13
29
13
211
47
98
29
7
23
14
11
2
7
3
33
21
81
7
186
9
28
596
389
343
1034
231
187
19
1246
12
1070
1103
2674
131
7723
17
1620
9
9
I
4
2
5
0
26
0
31
21
20
7
139
2
2
29
94
5
43
0
253
0
2924
0
1273
644
728
258
5037
559
18395
278
0
2
0
0
0
0
2
..
4
0
0
0
0
0
42
0
28
0
34
0
38
0
3
1100
0
1355
0
6120
0
1865
0
143
11330
109
10587
()
o
..."
(.I')
(.I')
2
3
3910 143
II
31
6
22
I
0
9530
18
I
6
60 161247
14
3162
2
4
145
I
2
0
27
0
91
10
13
2
3
0
116
22
2
7
159
32
10
0
65
No.
18282
698
123
2221
608
404
359
447
0
3174
3011
10
408
9093
893
178
0
2
0
4
0
0
0
34
7
Volume
68
0
183
0
16
0
31
0
0
0
882
291
0
0
2073
278
168
22
0
0
0
192
4
41
0
715
31
14
2
2427
459
4
No.
5
79705
0
135220
3821
2220
2957
1141
190
1430
0
34020
425
16
0
20
0
560
4754
~217tO
o
4204
151
98590
1593
51
782
0
12
0
0
Volume
utC
1990-1998
I
0
13
0
0
1
6
0
0
6
52
6
0
~~~~
615
0
3856
3
0
7
0
17
19
143
46
35
29
93
312
4
24
74
72
37
48
26
63
5061
2841
3576
8324
1014
3454
1669
14354
62
3
0
35
-~-
S. Europe
S. Europe
S. Europe
S. Europe
S. Europe
S. Europe
S. Europe
S. Europe
S. Europe
S. Europe
S. Europe
S. Europe
S. Europe
S. Europe
S. Europe
S. Europe
No.
1980--89
-~~~~~~
1129
110
226
313
1037
8423
20
12
2467
2101
1970-79
_"
,.
S.E. A~ia
S.B. A$iu
S.E.Asia
Brunei
C:unhmlia
::i.E. Asia
Millays; ..
S.E.Asi:1
S.E.l\siil
M,YIIIIHlur
Sil1gllllure
Thailand
S.n.Asia
VkWiltll
S.r:. Asia
'2
'2
Lnos
45
2325
22
119
7(nn
491
465
204
75
78534
15
385
16.5
165
277
117 311
424
I
59
5
:I
23'1
7030
28 ')(,0
1blal
Worltllllllli
25432 6464730
I
14
18
2442
1
II
I
28
18536
165
33
21 18(i
I
4
-1-
S.I1.Asia
I
I
'11
0
II
()
(J
Il
.,'--
9986
0
0
96
96
()
75
0
I
75
0
'2
44
44
4
I
I
20
27
7030
6657
B27
)0
30976
44
19
3
19720
1423
17
27 ROI
I
135
~~
49005
223
3ri
1150
37
1J73
0
0
---0
600 411117 808 57579 979 131 139 911 162241 2734 781847 4793 I 838603 5425 I 673793 '1426 872998 1867 J9478K I32S 488566
NOles:
:\ AI'IlClllill:t ~.,c1udesYmyr~la 21 OIlO M.llll underconslruclion - indudcd in Paraguuy
b Paraguuy ex<:1udt'sllllil)U 24 (Jon M.II1) (1983) - included in Bl1Izil
c Zmllbiu .:xclmJes Kuritm 1806()O M.m' (1959)- included In Zimbubwil
d Ul1IglIUY .:xllld.:s SUlll~ Grande 5500 M.ll1) (1979) - included ill Argenlinn
e Yugu~lavia excludes Djcrdall 1 255 M.m' (1972) and Djerdap 11868 M.m' (1987) - included ill Homania
r 'lllgU exdudes Nangheln 1710 M.nl) (1988) - il1c1uded io lIel1in
[tala cXlracl.::t1 from ICOLD World ,.esister of ifillll.r (1998)
'1J
-0
o
X
EVACUATION OF SEDIMENTS
Table Al.2.
CountTY
Catchment:
km 2
Capacity:
M.m'
Survey dales
Stan
India
India
India
India
India
India
India
India
India
India
India
India
India
India
India
India
India
India
India
India
India
India
India
India
India
India
India
India
India
India
India
India
India
India
India
India
India
India
India
India
India
India
India
India
India
India
India
India
Uri
Maharashta - Koyna
Maharashta - Khadakwasala
Maharashta - Yeldari
Maharashta - Ramtek
Maharashta - Dama
Maharashta - Gima
Maharashla - Ekruk
Maharashta - Mhaswad
Maharashta - Visapllr
Maharashl3 - Mangi
Maharashta - Asolmendha
Maharashta - Nalganga
A.P. - Sriramsagar
A.P. - Nizamsagar
A.P. - Himatyasagar
A.P. - Cembum Tank
A.P. - Kaddam
A.P. - Ramappa Ll!ke
A.P. - Lakhamvaram
A.P. - Dindi
A.P. - Palair
A.P. - Manjira
Kanataka - Tungbhadra
Kanataka - Bhadar
Tamilnadu - Lower Bhawani
Tamilnadu - Vaigai
Tamilnadu - Mettur
Tamilnadu - Upper Bhawani
Tamilnadu - Setnur
Tamilnadu - Aliyar
H.P. -Pong
Punjab - Bhakra
Uttar Pradesh - Mamtila
U.P. - Ramganga
U.P. - Dhukwan
Uttar Pradesh - Tehri
Bilhar - Panchet Hill
Bilhar - Maithon
M.P. - Gandhisagar
M .P.-Tawa
GUjrat - Ukai
Gujrat - l(adana
W.B. - Mayurakshi
W.B. - Kangsabati
Orissa - Hirakud
Kerala - Malampuzha
Kerala - Peec.hi
India
China
China
China
China
China
China
China
China
China
China
China
China
China
China
China
China
China
China
China
China
China
China
China
China
China
China
China
China
156
12750
892
507
7362
212
404
4729
412
1243
412
304
246
316
91751
21694
1308
993
2656
184
268
3919
1687
16770
28 180
2435
4200
2253
42200
34
10826
195
12562
56980
20720
3134
21340
5188
1087.8
6294
23025
5980
62224
25520
1860
3790
83395
148
107
2797
110
934
117
237
609
94
87
43
34
93
76
3172
841
108
106
124
82
60
74
57
51
3751
239
933
195
2709
101
235
109
8579
9869
1133
2450
106
3540
1581
1349
7740
3645
8510
1543
608
1135
8105
228
113
Total
591737
78412
688000
16200
9640
2560
2270
721
5720
216
606
49
16000
200
36
168
357
521
105
34
203
389
183
43402
181 800
182700
285000
215900
77
175
348
105
496
120
185
1320
End
TOlal
sedimentation:
M.m'
Vol. lost:
2
1961
1870
1963
1914
1910
1965
1871
1888
1902
1957
1918
1963
1970
1930
1927
1956
1958
1919
1909
1943
1928
1966
1953
1963
1953
1958
1934
1965
1957
1962
1974
1958
1956
1974
1907
1986
1940
1983
1987
1941
1979
1991
1990
1988
1989
1987
1985
1984
1975
1976
1978
1977
1975
1975
1976
1977
1977
1985
1974
1983
1983
1984
1985
1982
1981
1986
1987
1990
1986
1980
175
23-9
841
145
21
495
264
45 3
296
33
:27-0
42
7949
5336
286
21
459
2g
20
22
11
187
5882
310
378
22-4
5283
36
27 6
31
422J
9158
2735
970
472
06%
21 7%
90%
12-4%
09%
81%
281%
520%
688%
96%
290%
55%
251%
635%
265%
20%
371%
34%
33%
30%
20%
367%
157%
13 0%
41%
115%
195%
36%
118%
29%
49%
93%
241%
40%
445%
1956
1955
1960
1974
1972
1977
1955
1965
1957
1955
1957
1985
1979
1976
1980
1984
1984
1975
1972
1984
1977
1982
1852
1554
3591
96
5412
702
620
57
13349
80
254
117%
115%
46%
03%
6-4%
46%
102%
05%
165%
35%
225%
7514
9-6%
1960
1960
1960
1953
1960
1968
1958
1967
1975
1968
1960
1959
1963
1967
1976
1973
1960
1972
1971
1960
1959
1959
1960
1959
1958
1970
1970
1960
1978
1989
1987
1994
1989
1989
1968
1980
5450
5690
670
630
330
1410
J61
566
15
1130
205
29
2
206
218
28
7
77
63
97
41
35
47
20
249
17
336%
590%
262%
278%
458%
247%
745%
93-4%
306%
71%
1025%
806%.
12%
577%
418%
267%
206%
379%
162%
530%
532%
200%
135%
190%
502%
142%
43%
47%
N/A
1986
1983
1973
1986
1987
1986
1988
1986
1990
1990
1989
1983
1988
1986
1988
1990
1990
1990
1983
8
62
Annual sedimentation
Notes
%
M.m'/ann
550
784
673
571
938
166
747
534
357
835
336
1588
601
619
547
446
96
910
267
112
14
102
652
1158
300
398
250
5347
102
838
2800
554
388
2580
30
1400
587
1029
975
268
725
393
1667
216
593
2465
9505
701
070
034
4 20
020
007
353
022
044
034
010
039
019
5678
11 86
058
010
242
005
003
007
002
170
1838
282
126
090
1057
018
110
016
3517
31 58
804
809
065
726
639
648
2245
160
45 10
1003
310
082
4944
036
\02
003%
031%
045%
017%
003%
058%
023%
051%
080%
030%
042%
025%
179%
141%
054%
009%
195%
006%
005%
009%
004%
334%
049%
118%
014%
046%
039%
018%
047%
015%
041%
032%
071%
033%
061%
021%
040%
048%
029%
004%
053%
065%
051%
007%
061%
016%
090%
604
35729
0-46%
4401
30278
19621
2481
1537
1138
6714
1610
4354
187%
204%
097%
068%
158%
117%
7-45%
718%
6278
891
207
009
1030
2180
)87
027
428
332
334
171
121
IS1
069
778
085
040
270
039%
446%
575%
005%
289%
418%
178%
079%
211%
085%
183%
222%
069%
052%
066%
157%
071%
022%
020'11.
17
3540
3693
881
1528
Mt/ann
% vol.
m'/km /yr
N/A
j
i
J
J
APPENDIX I
Table Al.2.
continued
Country
Re~ervoir
c~,_,
Ic",,""
I
km z
t.Lm}'
Survey dates
Start
I 42804
China
Total
I
I
End
ChilUl
China
Total-on3!) reservoirs
1981
80986
83 357 reservoirs with 460000 M.m' storage willI an aver.lge annual loss af 23%
Netherlands
HlIringvUet
Romania
Romania
Romania
Romania
Romania
Romania
Romania
Romania
Romania
Romania
Romania
Romani:!
Romania
Romania
Romania
Romania
Romania
Romania
Pangamli
V:lduri
Balea
I Tumu
21
11
21
62
Total
K...pelilny
Hrieov
Nosice
Slovakia
Total
Japan
29 dams
13.5 d:lms
169 dams
15S dams
152 dams
J.:lI:l!ln
207
S
8
36
Spain
Spain
Saain
Spain
Spain
Spain
Spain
Spain
SP:1!tl
202
26
149
123
570
308
138
86
23
, T:lb'lls - Buendia
; ucar - Alcon
Ebro - La Tranqu:::ra
Sur - R~(Ie"'o.do
CaL:1.lonia -'"Riudecanll5
S<!gura - Camarillas
Norte - Rioseco
I
049
094%
075
360
450
1104
1431
132
096
108
017%
016%
015%
0-22%
025%
022%
038%
051%
3755
22-4%
6-3%
Tot;u-Gains
Spain
Spain
Spain
Spain
Sp;z.in
284%
\.,5
247%
69%
2737
Spain
14969
1870
579
1520
2
84
2
3
35
4
7
4
Tagus
Ta!<u$ - Palmaces
Tagus - El Vado
Jlicar - Amadorio
Ebro - Arguis
275
379
Jucar - .~eIlQS
Iuco.r - Bus(!o
Ebro G:tl!iu~uen
luco.r - A:ouillo de Dan Bias
NOlie -Alfilorios
Gu:!d:llqtlivi~ - G'.lado.lmena
220
32
57
17
3
1311
8
4-
1955
1942
1965
1973
1969
1984
1956
1962
1954
1909
557-9
647
569
-92742
-10835
--4962
-403
-3724
-3478
-2014
-2
-In':::
-0504
-t:!6
1957
1958
1960
1974
19i5
1960
1973
1959
1948
1954
1972
1960
19:.8
1979
1912
1927
330
690
H,)
1973
2:2
1350
3J..7
1960
1990
1969
1984
1995
1993
199[
1977
1990
1985
1%9
1990
1983
324
855
3201
198J
1976
1994
1992
1981
1993
1994
1979
1968
1984
1979
1991
1980
1994
19S0
1979
[981
1988
1994
1 1989
0
0()75
0121
0154
oun
0189
0285
0326
0561
0629
0691
0723
07:58
014
008
028
166% 1
091%
077%
4500
034%
-22%
-047
-O5S
-007
-019
-005
-001
-469'0
-526 j
-0\9%
-320
-020
-018
~7*
-Q~:!2
-83%
-5-4%
-30%
-06%
-07%
-14%
-2~1%
000
000
000
001
000
0-01
(}O2
002
003
002
Q.1O
002
()O2
005
001
002
0-1 I
00%
38%
OI%8-6%
56%
0-5%
6-7o/c
47%
12-8%
20%
12%
44%
25~3~c
0. 793
06%
0-807
0-S4t
e9!
0965
0994
\0-1%
210%
9-1%
44%
106%
H),
03%
0-220/0
-029%
-0[0%
-067%
-015%
-038%
-010%
-002%
-021%
-006%
-003%
-IS8%
I 321%
6,65
32-4%
581%
825
\994
1994
1994
19941994
4825
2089
8055
t3 200
1112
Spain
1994
1994
1994
6707
Spain
1991193
i98!/90
1971180
1961170
1951160
1941/50
1931/40
tol930
1-3.5
052
096
043
063
[.63
Spain
Souin
Spain
Spain
Soain
Spain
SOllin
Spain
1992
1989
1992
39
691
1305
256%
41-2%
440%
535%
358%
323%
210%
683
Jupan
Spain
Spain
1957
1962
1963
12
0132
H
54
:5-15
1147
011
OCO
021
0C3
005
007
007
007
005
016
[ISH
Japan
Spain
214%
069%
/-17%
031%
241%
)50%
064%
120%'
3-93%
4<12%
4<88%
197%
1030%
440%
446%
398%
294%
262%
i7 322
23i7
2942
4911
5672
599
251
[88
27 dams
Spain
024%
03%
14%
2-9%
65%
98%
108%
227%
365%
35 dams
24 dams
Spain
53
443
JaPan
Jacun
Japan
Japan
lapan
Japan
Sp:!in
241
014
003
10%
Ramnieu
Raureni
Govan
' Babeni
Romania
099%
493%
110%
280%
75%
505%
255%
lQ.8%
241%
746%
74-l%
781%
Slovnkiil
Slovakia
Slovakia
800CO
70
Dnesti
1/9%
142%
33
055
28
003
437
13
08
13
132
% vol.
! 1500
1987
1981
19&7
1987
1986
1982
1983
1986
1986
1986
1986
j
1986
19S~ 1 1986
1976 1986
1974 1986
1977 1986
1975 1986
\978 1986
Notes
Mtlann
51071
1999
I~I
~edimcntJtion
M.mJ/a,.'l11
I 281%
\970
1966
1966
\967
1968
1970
1973
Annuul
rn"lkm!/yr
12013
1964
1965
1963
1963
1965
10
0
9
.5
7
5
2
2
I
\3 j
IJ tI
Garleni
Lilteci
Bucau
Oiesti
Cerbureni
I",
.5
Vanarori
RaCOV!
1 Vol.
1000
I
T~,'
sedimcntation:
M.m"
000%
O':H%
0-00%
0480/"
009%
002%
042%
023%
064%
007%
017%
014%
0-60%
0-04%
015%
0-40%
0-03
1-14%
0:6%
025
005
002%
2~%
157
J
EVACUATION OF SEDIMENTS
]
Table Al.2.
Country
Spain
Spain
Spain
Spain
Spain
Spain
Spain
Spain
Spain
Spain
Spain
Spain
Spain
Spain
Spain
Spain
Spain
Spain
Spain
Spain
Spain
Spain
Spain
Spain
Spain
Spain
Spain
Spain
Spain
Spain
Spain
Spain
Spain
Spain
Spain
Spain
Spain
Spain
Spain
Spain
Spain
Spain
Spain
Spain
Spain
Spain
Spain
Spain
Spain
Spain
Spain
Spain
Spain
Spain
Spain
Spain
Spain
Spain
Spain
Spain
Spain
Spain
Spain
Spain
Spain
Spain
Spain
Spain
Spain
Spain
Spain
continued
Reservoir
Catalonia - Foix
Guadalquivir - Gergal
Guadalquivir - Bennejales
Jucar - Forata
Segura - Argos
Ebro - Santa Maria de Belsue
Segura - Anchuricas
Sur - La Vinuela
bro - Santolea
Ebro - Moneva
Guadiana - Valuengo
Jucar - La Toba
Guadalquivir - CubiJIas
Segura - La Cierva
Catalonia - San Pons
Douro - Burgomillodo
Jucar - Sichar
Guadalquivir - La Bolera
Douro - Linares del Arroyo
Jucar - Guadalesl
Guadiana - Torre de Abraham
Ebro - Las Torcas
Guadalquivir - Cala
Ebro- Pena
Guadalquivir - La Minilla
Guadalquivir - El Tranco de Beas
Jucar - Maria Cristina
Norte - Penarubia
Jucar - Beniarres
Sur - Guadalhorce
Tagus - CazaJegas
Guadalquivir - Bernbezar
Guadalquivir - Puente Nuevo
Tagus - Guajarez
Tagus - Entrepenas
Guadalquivir - Torre del Aguilla
Segura - Cenajo
DOUTO - Agueda
Ebro - Cueva Foradada
Jucar - Benageber
Guadiana - Zujar
Ebro - La Estanca de AJcaniz
Ebro - Sotonera
Jucar - Ernbarcaderos
Catalonia - Sau
Guadalquivir - Guadelen
Tagus - EI Burguillo
Sur - Conde del Guadalhorce
Guadalquivir - Guadalcacin
Ebro - Ribaroja
Segura - Talave
Segura - Valdeinlierno
Tagus - Gru;el y Galan
Douro - Santa Teresa
Tagus - Riosequillo
Guadalquivir- Bornos
Ebro-Oliena
Guadalquivir - La Brena
Guadalquivir - Guadalmellato
Guadalquivir - Pedro Marin
. Segura - Alfonso XII
Segura - Puentes
Jucar - Contreras
Ebro- Yesa
Guadalquivir - Dona Aldonza
Tagus - San Juan
Ebro - Barasona
Segura - La Fuensanta
Ebro - Talarn - Tremp
Ebro - Mequinenza
Guadiana - Cijara
Spain
Total-Loss
Turkey
Turkey
Turkey
Turkey
Turkey
Turkey
Caygoren
Demirkopru
Buldan
Kerner
Yalvac
Kararnanli
158
Catchment:
km 2
300
1058
190
117
1221
737
167
800
187
760
60
761
64
965
1334
469
965
1955
1665
559
375
439
1060
644
3495
323
16952
1281
268
763
31 I
1846
1858
1361
2694
465
1195
420
852
1042
2344
2181
3766
863
1250
1201
7456
Capacity:
M.m;
6
36
104
39
12
13
8
170
49
10
20
II
21
8
25
15
52
56
58
16
60
9
59
22
60
500
23
12
31
134
11
347
289
25
8911
70
472
22
29
228
309
14
189
9
177
173
209
78
77
219
47
25
924
496
49
215
101
116
163
19
42
32
872
471
23162
71
235
258
1530
1670
Start
End
Total
sedimentation:
M.mJ
1928
1979
1958
1969
1970
1931
1957
1986
1932
1939
1959
1935
1956
1929
1957
1953
1960
1967
1951
1965
1974
1946
1927
1930
1956
1945
1920
1961
1971
1972
1949
1963
1972
1971
1956
1947
1960
1931
1926
1955
1989
1944
1963
1952
1963
1954
1931
1921
1917
1969
1918
1897
1961
1960
1956
1961
1959
1935
1965
1954
1916
1884
1975
1960
1955
1955
1932
1933
1916
1966
1956
1983
1985
1978
1983
1991
1980
1979
1994
1993
1984
1985
1980
1990
1987
1968
1989
1976
1979
1980
1989
1988
1979
1984
1989
1984
1990
1991
1994
1991
1991
1990
1994
1994
1982
1979
1992
1992
1980
1992
1992
1994
1971
1986
1983
1979
1977
1991
1991
1969
1982
1993
1984
1990
1989
1970
1990
1985
1991
1992
1977
1985
1985
1994
1986
1977
1992
1993
1991
1990
1982
1983
1158
1305
1395
1542
1666
1742
1759
1799
1851
1991
2131
2174
2299
2429
2591
2603
2729
2'828
2954
3008
315
3195
3603
3619
364
3675
375
3788
3-831
397
4052
4899
4 97
5391
5611
5643
6403
6582
6617
6663
7122
7133
7288
7539
8-495
9718
10936
11051
11972
12224
12344
12473
1284
13387
14024
14815
1518
15869
16323
17-893
18184
18726
19595
2078
220439
24258
24764
25 273
69592
92-822
138 111
193%
36%
13%
40%
142%
134%
220%
11%
38%
199%
107%
198%
109%
32-4%
104%
174%
52%
51%
51%
188%
53%
355%
61%
168%
61%
07%
161%
316%
124%
30%
368%
14%
].7%
21 6%
01%
81%
1-4%
299%
231%
29%
2-3%
510%
39%
838%
48%
56%
52%
142%
155%
56%
262%
499%
14%
27%
289%
69%
150%
137%
100%
942%
43-3%
593%
22%
44%
97' 6%
150%
349%
108%
270%
6 1%
83%
002
022
007
011
008
004
008
022
003
004
008
005
007
004
024
007
017
024
010
0]3
023
010
006
006
013
008
005
011
019
021
010
0 16
023
049
024
013
020
013
010
018
142
026
032
024
053
042
018
016
023
094
016
014
044
046
100
051
058
028
060
078
026
019
103
080
102
066
041
044
094
580
5 12
035%
060%
007%
028%
068%
027%
100%
0 13%
006%
044%
041%
044%
032%
056%
094%
048%
033%
042%
018%
078%
038%
108%
011%
029%
022%
002%
023%
096%
062%
016%
090%
005%
008%
196%
000%
018%
004%
061%
035%
008%
046%
189%
017%
270%
030%
024%
009%
020%
030%
043%
035%
057%
005%
009%
207%
024%
058%
024%
037%
4 09%
063%
059%
0 12%
017%
4 43%
040%
057%
0 19%
036%
038%
031%
859
37%
3240
0 14%
246
5636
315
2098
59
99
190%
692%
685%
563%
684%
397%
088
1445
098
466
023
038
068%
177%
214%
125%
263%
153%
Survey dates
23323
1510
6590
180
2500
133
164
130
814
46
373
9
25
\1971
I ::~
1954
1973
)973
1999
1999
1999
1999
1999
1999
Vol. lost:
%
Annual sedimentation
m"/km 2/yr
583
2193
5467
1865
1697
2311
M.mJ/ann
Mtlann
Notes
% vol.
J
- -J
'-1
-1
1
1
j
1
J
1
J
J
4
J
1
APPENDlX I
II
Table Al.2.
continued
C(luncry
Turkq
Turkey
Turkey
Turkey
Turkey
Turkey
Sdc:vir
Cubuk- I
Bayi ndir
Hilfanli
Kcssikkopru
Alcinapa
Sevhan
Kmalkaya
Cip
Surgu
Turkey
TOlal
Pakistan
Tarbela
Germany
Germeny
Saxonian Reservoirs
Bal c!eney
Ba\'aria - Forggensee
Bavaria - Saalachsee (flushed)
B :lv:ui~ - Sylven5rei nsee
Turk~y
Turkey
Turkey
Turk~y
G~r:nany
Germolny
Germany
Germany
Malaysia
Malaysia
Malaysia
R<!sc:rvair
~~
721
660
70
261 70
360
589
19 :!54
11 30
236
2i5
7'
14;;
940
I US
R i ng! ~l
Tunisia
Tunisia
Tunisia
Tunisia
Tunisia
Kasseb
EI Kebir
Mellegut:
Nebhana
Bezlk
Chiba
L:tkmess
.sou Her1.'na
Iuumine
Lebna
Sisi Saad
Sidi Salem
Silianu
Marg'Jellil
Tunisia
Totai
42122
~rasri
UK
95 reservoi rs surveyed
l[~ iy
Csdore Valley
S<lbcr..::
r~l y
To~i
USA
USA
US A
USA
US A
USA
USA
Total
Morroco
Morroco
Morreco
Marreco
MOCTecO
Morroco
Morroco
NakhLa
Mohamed V
Lalla Takerkousc
~~O!"TCCO
EI Kansera
Moue KiTalcabi
Ibn Batouca
My. YaU.>sef
Mar.sour Ed D!lhoi
Bin EI Ou idane
Hassan Addakhi1
Y B Tachfine
Morroco
Morroco
Morr.x o
SM3 Abdeilah
Morroco
Mor.oc o o EI Makhazine
Mor.oco I Hnssanler
;':[c;m:;co 1 AI M:mira
508
8
150
4
32
I""
1965
:960
1966
1
I
1913
i959
236
33-9
12-6
35
3936-0
28-1
14-2
4297
42-1
55-8%
2150%
50-0%
296%
43-9%
358%
284%
371 %
394%
5377
597%
2900
203%
183
36'70
2408
1-6%
100-L%
i3%
1379
303
1464
3760
2364
752
519
1379
3i O
3388
65-8%
8
8
7
11 8
130
29
209
~55
iO
110
1776
I
I
12-38
2374
232-;6
54 12
6 23
743
1002
606
11 00
i 50
600
9 180
561 0
38-50
4400
60764
15 1%
91 3'70
701 %
626%
965%
945%
125-2%
83-9%
94%
5-8%
207%
439%
101'70
550%
40')%
1203 875
109 980
107
49920
1710
45oiO
13
084%
0-06
022
009
;49
G' ll
I0
1 ' 15%
2-5t %
Q4[0/0
I
020%
I
165
275
550
1360
0030
0050
0 101
0249
4228
1200
514
1918
1952
3516
2465
3811
1282
1196
5051
603
18L
3365
4911
0-427
03252
529
164
0 164
0225
0313
0202
0-5
05
05
5-4
J.3
35
55
042%
01 2%
()'05%
019%
020%
0-35%
025%
038%
0-13%
0 L2%
05 1%
006%
002%
034%
0-49%
27-79
Oo();%
0 10%
360
160
723
96
330
43
436
198
592
1484
369
320
51]9
807
I
I! 8
I
196 1
1967
1935
1927
198 1
1979
1970
1972
1974
197:2
;07 1
i979
271
1987
2724 I 1976
O - I~
143
235
209
23-6
l804
19 1
1066
2563
3-4%
002
0 15
097
449
2069
57-92
134-44-
186
424 1
3-')%
218-69
603
2569 1
468'70
354%
265
64-66
6-96
56
22
62-88
9932
2096
16-49
li 6O/o
023
- L1 17
050
122
0S7
058
1-10
22 97
4 59'9
4SB
i 1987
1990
1988
1980
1989
1989
1'790
1988
,990
1989
: 1985
1 1990
11 990
I :937
0
2
23
94
257%
193%
165%
1304%
91%
36%
196%
162%
13%
111 %
106<;;,
6-7%
5-7o/c
51%
33-99
4 '2L~
to
82-94
3-7%
3-0%
014
00 1 I
I
11
138
700
5331
29249
74499
6400
1208)
770
154
93-6
342%
178
1 1.50 %
I
1969
1925
1954
1965
1960
1'765
1966
1968
1976
1983
1986
1981
1981
1987
1990
82
26
332
86
6
IMO
!5 C:("J
1-31%
135
043
I 1996
no
1-64%
3-57')I>
147%
165%
0-90%
137%
083%
105 %
109 %
100
010
30%
Notes
Muann % vol _
0-10
9840
0-85
044
9-99
156
01 1
093
I
I
I M_m-'/~nl1
m-'/km%/yr
280
1 1
Annual
1 1990
1980
19~
425
1132
94 12
39 345
2265 i6
454897
477 14B
4570
3780
9SCO
1821)
1670
28500
Val. la st:
3-644
TOl;ll
sedilO~maci o n:
M.m!
J.6
380
93
USA
1998
Tunisi a
Tunisia
Tlmisia
Tunlsia
Tunisia
Tu nisia
Tu nisia
Tanisin
Tunisia
Italy
101
271
10 300
855
84
64
127
53
390
418
99
8950
18 250
1040
1110
T~~ ~si:\
1999
1999
1999
_1999
1999
1999
1999
1999
1999
1999
183
183
183
183
Ringlet
Ringlet
To~1
1965
1936
1965
1959
1966
1967
1956
1972
1965
1969
1974
14300
Torn I
M~ lolysia
61
6
7
5980
95
3:!
120U
148
10
71
End
9006
I Ring let
Malaysia
Survey dates
Smn
3093
3-22
I
I
131
0-97
1-77
3-09
333
i -54
356%
2-00%
102%
081%
043%
023%
01 6%
0-20%
1770/0
1-54%
052%
0-37%
2-02%
0 13%
056%
0-66% I
012% j
03 6%
030%
Q35 ',t..
0-38%
122%
1 0.28 %
159
EVACUATION OF SEDIMENTS
Table Al.2.
continued
Country
Reservoir
Mormeo
Morroeo
ldriss ler
I Abde1moumen
MOITOCO
Total
Algeria
Algeria
Algeria
Algeria
Algeria
Algeria
Algeria
Algeria
Algeria
Algeria
Algeria
Algeria
Algeria
Algeria
Algeria
Algeria
Algeria
Algeria
Algeria
Algeria
Ghrib
DjorfTorba
Bough7..0ul
Bouhanifa
5MBA
Chcurfas
K'Sob
Bakkhadda
Foum EI Gharza
Beni Bahdels
Oued Fodda
Ighil Emda
La Chaffia
Zardesas
Sarno
Foum E1 Gheiss
Hamiz 1& 2
Mef\'rouch
Reservoirs on the Atlas Mountains
Reservoirs all flatter land
Algeria
Tolal
TaiwlIn
Taiwan
Taiwan
Taiwan
Taiwan
Taiwan
Taiwan
Taiwan
Taiwan
Taiwan
Shihmen
TenKee
WuSheh
Tsengwen
Pai ho
Gen Shai Pei
A Kung Ticn
Lu Liao
Tapu
Ku Kuan
Catchment:
km 2
Stlllt
End
1972
1981
1986
1987
169
1120
55
73
235
1974
31
1939
1974
1952
1938
1974
1954
1965
1986
1986
1986
1986
1986
1986
1986
1986
1986
1986
1986
1986
1986
1986
1986
1986
1986
1986
3680
1300
1217
216
135195
10351
9099
34118
28571
1414
246
1132
443
1277
873
288
1453
309
255
150
713
25
7
36
3094
138
2944
1525
Indonesia
Ind{lIlesia
Indonesia
Karangkates
Selorejo
Wonogiti
2050
238
1262
343
62
730
Indonesia
Total
3550
1135
Sudan
Sudan
Sudan
Sudan
Total
77108
74%
1126
336
22-3
214
94
12
1316
81
205
65
923
349
73
165
07
14
665
J4
667%
30%
409%
293%
40%
425%
40%
477%
10-3%
636%
349%
43%
148%
32%
436%
348%
20%
Egypt
Ethiopia
CYPl1ls
Cyprus
Cyprus
Cyprus
Cyprus
Cyprus
Cyprus
Galini
Petra
Kalokhorio
Lynlhroclhonda
Lymbia
Kophinou
Akrounda
Cyprus
Venezuela
221
023
)]1
85
21
59
213
41
198
122
433
i36
2098
2498
573
938
234
187
640
648
1100
2800
1982
1982
1985
930
1300
3354
167000
I
I 0-42%
JOI
290
060
056
118
060%
026%
110%
077%
050%
028
003
049
014
093
319
050
027
034
003
0-48
0-03
090%
001%
114%
021%
064%
319%
029%
024%
J55%
093%
249%
004%
1014
008%
060%
016%
031%
17%
4-41
I 039%
560
740
1127
602%
569%
336%
918
2846
5932
099%
219%
177%
2427
1 435%
9696
I 174% i
2100
13%
8077
0-05%
1986
1990
1985
1964
1990
1961
1981
]700
340
10
0216
0026
JI9%
000%
Sanlo Domingo
427
307
1202
148
2350
1400
1082
4S85
"
1976
1978
0575
15
50
8
143
109
32
229
II
1905
1913
1918
1932
1947
1937
1941
1960
1930
1959
1933
1961
J961
1965
1%5
1%7
0188
092
0138
85
03
38
4183
103
13%
18%
17%
59%
03%
118%
183%
56706
95%
192%
94<;i-
24
17
63
125
15
125
364
118
029
958%
001
002
OOJ
029
002
01.:f.
)78
0\5
005%
004%
012%
021%
002%
042%
078%
: ]35'k
242
i
I
i 040%
166
1
1
1
19
TOlal
597
204
010
227
410%
390%
303%
114%
7,13%
1433%
555%
30%
19%
12%
82%
78%
91%
57%
285%
430%
167%
mo
067%
167
000192
00024J
000628
000232
000739
0-00159
000370
067%
044%
072%
059%
151%
144%
143%
001%
254%
2-48%
0023
0031
0069
0041
0026
0004
0022
0-49%
26
37
24
9
33
12
26
l..esna
Pikhowkc
Lubachow
OtmuchQw
Turawa
Porabka
Roznow
I Myczkowce
0. 18 %
011%
1022
IJ6
907
1925
1964
1966
5584
% vol.
206
112
j07
423
038
010
052
000
023
0-42
I
1977
1970
1981
IMtlann
4327
13
2700
1900
4900
8800
14300
9400
16300
30
2200
600
1938
NOles
155%
409
1963
1998
M.mo/ann
m>lkm /yr
25%
06%
Total
Annual sedimentation
2
9
17
Total
160
1953
1939
1972
1937
Vol. lost:
%
M.m~
2633
763
592
219
481
27
11
32
8
104
708
Taiwan
1948
1978
203
43
63
145
100
171
111
22
3
19
70
Taiwan
Poland
Poland
Poland
Poland
Poland
Poland
Poland
Poland
Tota!
sedimentation:
Capacity:
M.ml
1
1
II
r"
.1
APPENDIX I
Table Al.2.
continued
Catchment:
kml
I
[
ICaDadty: I Survey
I
~l.m'
! I
Start
Avisio
P"rnegg
Pt.mtebbn
Steyerdurchbrudl
Tar:ento
Austria
Austria
Austria
Austria
Austnn
Austrit,
Austria
Austria
Wet;:man
Margarilze - study rt!f. further
ITotal
956
6150
10
575
62
324
085
0[5
0-60
8177
410
Switzerland
Switzerland
Kallna..:h
Petolle.
Switzerland
Total
2621
New Zealand
New Z:!\]and
Roxburgh
MutOlhina
8826
28+1
5740
34
i7:
::[~C~land ~lomi
Total
S'lutn Africa
South Africa
Suuth Africa
SO(1tb Africa
South Africa
South Africa
South Africa
South Africa
Afhasini
End
M.il,.'
1925
, 189()
1927
1892
ISS]
1908
1884
1919
: 1872
1886
1961
1979
[966
193.;1.
1980
1959
1979
Vol.losl;
%
1000%
700%
035
025
071
015
06
840%
1000%
1000%
381
930%
556%
100,0%
Annual sedimentation
m)/km~/yr
261
3704
1545
003
193
1852
12$
56
M.m}l:mn
Notes
I Mtlanll
% vol.
025
023
002
12:50%
4670%
380%
00\
060
800%
10000%
114
2HI%
017
007
945%
713%
714%
IOH
497
475
01
i 1939
~~:
26
liS
A1!cmanskraal
Bt:ervld
Boegoeberg
Bon Accord
Bronkhorstpruit
Bulshoel(
9t
'9
Darlil1l!:ton
Drid -
years
2628
2-982
95
00296
ill
2~~~~~:
1923
1995
1995
1986
1995
1995
1986
1995
1917
1995
1935
1922
1995
98
1995
1399
1995
1995
1995
1993
1975
[929
1925
S9
6
1950
to
E!!monc
Elnndsdrift.
Erfenis
Aornkraal
Gamlmpoort
1952
1960
36
124
169
to
Somh Afri:a
7
208
50
42
i973
1937
1977
1960
1957
1969
H:utbefspoort
Hazdmere
K:Ilkfolltein
Kamn1<lnassie
Keerom
91
19~4
195
18
355
36
]9!5
1977
[(l:l%rric
Kommandodri ft
KGPpi~~
Kiommenellenboog
74
41
9
73
10.5
14
Leeu Gllmka
]g4
Marico
NooU!!edacht
Phalaborwa ba.crll"e
Pietersfont.t!in
'17
79
9
PUnl!alaDCOn
S~'u(h.Ar.';ca
Poo;;jie'
E'rinsrivier
Ril!tvld
Rust de Wimer
S(ompcirlf.
11
27
55
33
2536
V:lU Rr:-:e'.'e!c?u$
Welbedacht
Wentzel
Windsor
SQuth Africa
To!:!!
1923
[954
1960
1956
1911
1955
1970
1924
1959
1939
1933
196::!
1966
1963
1917
1933
1934
1965
1990
1938
1936
Vaaiha.ctZ
S"uth Airica
SO'Jtn Africa
SQuth Arrie:!
1938
196&
[973
"
1995
1995
1995
.1995
5673
22
South Afric:l
SoulhAfric:a
South Africa
South Africa
SDuth Africn
South Africa
South Africa
South Africa
Toml
~:~:
200%
210%
192%
165
75
66
22
[4600
139%
g:~~~~ I
043%
080%
053%
00007
20537
IO!% I
II'
13%
263
025
0l4%
0..()9%
396
314
08%
1
7920
36j0
9 rellrs I
2822
-----------4------+---~--~~--+_------+_--~----~----~--_4----~--009%
382
08%
3-39
30471125
8870 ! 3329
Bmzil
SuutnAfrica
S~'I.!th Africa
SomhAfricn
South Africa
Suuth Africa
South Africa
South Africa
South Arne:!
SoudlAfrica
SilUlh Afril.:a
South Afric:t
South Africa
South Africa
South Africa
South Africa
S')llth At'ricn
South Afric:!
SuuchAfric:!
Soui.hArrica
SourhAfrica
South Aftica
Seurh Afric:!
SomhAfrica
South Africa
South Africa
South Arrka
SOllth Africa
Scutt: Africa
! 18~2
1360
J261
B.'aZi!
2()O
050
dUlt:$
112
'6
19:!.5
1973
1934
1950
1995
1995
199.:5
1995
1995
1995
19~6
1995
J995
1995
1995
1995
1995
1986
1995
41
4C.~
6643
14J
\3
68913
I
I
35
25
::!31
16g
101
5.:l..59
29
412
H
09
0435
1.:1.7
12
23
114
111%
066
044
039
2::!75
24}%
9-6%
132%
453<;7.
113%
2-19%
062%
198%
0,32%
038%
093%
040%
049%
064%
015%
199%
06"
005
002
002
038
018%
014'1&
030%
029%
051%
295%
014
035%
2459
006
050
061%
122%
ij,}'e
169'7~
551%
59%
119%
07%
64
011
058
037%
066%
066%
061%
042%
033%
024%
027%
013%
029
034
73
22-656
Q033
210
016
192
3;14%
71-2%
5)
064.;1.3
1-1968
022
002
019
001
001
023
006
014
7j.S
3-2
060
79%
828%
.;!.g.!%
360%
357%
103%
3
06
55
1996
174'*
207%
34.:2%
103%
[00%
05516
1995
1995
1995
1995
1995
1995
117%
20
1995
1995
1995
1995
1986
1986
1995
399%
297%
61
366
1985
1995
160%
230%
73%
3J,,1~;
22,)%
22"%
90[%
663';7~
52t;~
449'c
IHj%
01%
107
0-22
0-48
005
002
010
002
250
006
007
001
002
021
001
190%
068%
102%
153%
013%
019%
003%
[20%
0&3%
010%
282%
08:5%
010%
008%
039%
002%
015%
065%
83~:tt
36~
41-(;
3g4%
071
3!4
397%
O57'V~
967
13
045
440
:mJ%
{JO::!,
O3Y'lc
~31';i
OOS
135~1:
107
066';:;'
1577
)93'ij,
034%
161
EVACUATION OF SEDIMENTS
Table Al.2.
Country
continued
Reservoir
Catchment:
km 2
ICapacity:
M.m)
Survey dales
Kenya
Kenya
Kenya
Kenya
Kindaruma on Tana
Kamburu on Tana
Gitaru on Tana
Masinga on Tana
9810
9520
9540
7335
Start
1967
1974
1978
198]
7
150
20
11560
End
Total
sedimentation:
M.m3
IVol.%lost:
117
1980
Annual sedimentation
78%
m;/km'/yr
M.m 3/ann
246
2-34
Kenya
Portugal
Portugal
Portugal
Portugal
Portugal
Portugal
Portugal
Portugal
Korea
Santa Luzia
Montargil
Campilhas
Burgaes
Idanha
Arade
ROKo
Lake Kariba
50
164
22
0
79
29
96
I Total
1960
1961
1966
1962
1963
1973
1980
156%
440
85
26 years
002
002
003
000
015
011
007
003%
002%
015%
026%
019%
039%
008%
05%
00%
16%
58%
32%
71%
10%
&239
1-4%
041
009%
077012
111%
003
043%
116000 M.m' - no re-survey data available. but estimates = 1600 to 16000 yrs to !ill dead storage
650000
1000000
(inc. Kariba)
310
21
289
33
404
497
215
025
005
0347
0019
254
205
0983
Sedimentation may
I
in
NOles:
1. Reservoir is dis of Lake Wular
2. Catchment area not permanantly covered by snow only
3. 12 M.mJ of sediments dredged
4. Values calculated from hydrographic survey results
5. Approlt 50% of sediment is polluted
6. Example of the effect of deforestation
7. Typical losses from 95 reservoirs surveyed
8. dis of cascade of 3 dams & 100's of check dams - dredged at a rate of 600 000 m'Janl1 since J985
9. Flushed since 1955 at an average of 328 000 ml/ann
10. Reservoir will not be able to function after 2010 unless rules changed
11. Reservoir Hushed in 1978 for 3 weeks clearing 620000 m3 of sediments
162
% vol.
11737
Beaggog
Notes
Mtlann
Appendix'2o
Numerical model case study
A2.1. TARBELA DAM, PAKISTAN
The Tarbela dam, a key component of the Indus basin scheme in Pakistan, was
completed in 1974 for the purpose of irrigation and hydropower. With an annual
sediment inflow into the reservoir of over 200 million tonnes, the live storage is
being rapidly depleted and liPless action is taken hydropower generation could
cease within a decade, with irrigation releases declining over the next 30 years.
The overall purpose of the feasibility study ca..rried out in 1998 by HR
vVallingford and TAMS UK, was to determine a strategy for the economical
preservation of h1e assets at Tarbela on a more sustainable basis.
A2.1.I. History
Tarbela dam was constructed in the 1970s to help regulate the seasonal flows of
the upper Indus both for irrigation of the Indus plains downstream and for the
generation of hydropower. It is still, 30 years on, the only major storage reservoir
on the Indus and, as such, plays a key role in the provision of dry season releases
of water for irrigation. Tarbela irrigation releases amount to 11 600 M.m3 , or 50%
of the WAPDA (Water and Power Development Authority of Pakistan) total, with
a corresponding agricultural revenue of Rs 28 billion. In addition, with an
installed capacity of 3478 MW and a finn electrical energy of 148 GWh/yr the
Tarbela dam provides 32% of both Pa.1dstan's total power and energy needs with
a corresponding annual revenue of Rs 6 billion. It is, therefore, a strategic
national resource whose continuing future efficient operation is of paramount
national interest.
A2.1.2. Sediment
The Tarbela dam impounds the waters of the Indus, which carry a heavy
sediment load. This is the case particularly in the spring and summer when the
rnelting snows cause heavy erosion of the uplal'1d catchment. I\t1ost of the
sediments brought down by the Indus are trapped in the Tarbela reservoir. Thus,
with an average annual sediment inflow into the reservoir of approximately
240 Mt per year, the live and dead storages of the reservoir have diminished by
16% and 21 % to 9000 1V1.m3 and 1360 M.m3 respectively in 1997.
163
EVACUATION OF SEDIMENTS
A2.2. ENVIRONMENT
A2.2.1. Hydrology/climate
The source of the river Indus is situated in the Tibetan Plateau, at an elevation of
5500 metres above sea level. From there it flows across some of the highest
mountain ranges in the world before emerging onto rain-fed lower-lying country.
Downstream of Tarbela, the Indus flows along a broad valley until it reaches
Attock Gorge, some 51 km downstream. On leaving the gorge the Indus flows
onwards for a further 1600 km to its mouth on the Arabian Sea.
The Indus basin upstream of Tarbela Dam, an area of 169 650 km2, consists of
two distinct hydrological regions. Over 90% of this basin lies between the
Karakoram and Himalayan mountain ranges; the meltwaters from the snow and
ice that cover approximately one quarter of this mountainous portion of the basin
cont1ibute a major part of the annual flow reaching the Tarbela Dam. Seven of the
ten highest lnountains in the world reside within the catchment. The remainder
of the basin, about 11 700 km2, lying immediately upstream of the dam, is subject
to monsoon rainfall, primarily dllling the months of July, August and September,
the run-off from which causes sharp floods of short duration that are
superimposed on the slower responding snowmelt run-off.
Climate in the Indus basin is subtropical and semi-arid in the headwaters. It is
divided to form two distinct seasons: kharif (summer), extending from April to
September; and rabi (winter), coveling the remaining months. The annual rainfall
averages around 900 mm of which two-thirds fall between June and October.
164
:f
r
]
APPENDIX 2
-l
:]
1
t
l
14 000
12 000
Rainfall
10 000
IJl
---I
8000
iti
3:
0
6000
4000
2000 I-
J
J
Month
Figure A2.1.
165
EVACUATION OF SEDIMENTS
maintained. During the initial years of operation, until the year 1988, the
reservoir was drawn down close to the minimum operating level of 1300 ft every
year and, consequently, the delta advanced towards the dam. As the rate of
advance of the delta downstream is related to the extent to which the reservoir is
drawn down, the policy since this time has been to operate the reservoir with a
higher minimum water level. This, however, has encouraged the deposition of
sediment further upstream, in the middle reaches and within the live storage.
A2.2.6. Verification
The model was verified by simulating the observed sediment deposition from the
time the reservoir was impounded to 1996 and comparing the profiles predicted
by the model in 1996 with those observed. The model gave excellent predictions
in the 20 km immediately upstream of the dam, see Figure A2.2.
166
APPENDIX 2
455
435
c:
.2
415
ca>
Q)
395
"0
--1974observed
_. 1996 observed
Q)
co
375
- - - - 1984 mode!
1996 model
355
335
10 000
20 000
30 000
40 000
50 000
60 000
70 000
80 000
90 000
100 000
Figure A2.2.
Pakistan
It
It
Case study -
Sand 2 _
DiSChaigel120 000
450
100000
i'! ::~
,..,
E
80 009
c::
'E
(Ii
60 000
250[
(3
CIl
200r
150
'6
40000
-<
ca::l
C
c:
-<
100
50
OL-~
1975
1985
1995
2005
2015
2025
2035
2045
2055
Year
Figure A2.3.
Pakistan
Case study -
167
EVACUATION OF SEDIMENTS
500
120000
c:::::J Silt 1 c:::J Silt 2 E::El Silt 3
Sand 1 _
Sand 2 -
Discharge
450
400
'E
350
-0
Qi
':;"
100000
80 000
'"E
~
'E
300
a.;
C 250
OJ
E
60000 . ~
200
'5
'5
OJ
C/)
C/)
(i'l
::!
c
c
40000
150
(i'l
::!
C
C
100
20000
50
o~~--~--~--~~--~--~~~~--~--~~--~--~--~--Uo
1975
1985
1995
2005
2015
2025
2035
2045
2055
Year
Figure A2.4.
Pakistan
Case study -
-L:r12000
D5 -
~ D4 -1500 ft min
- 0 - F4 - High flush from 2021
.......
1350-1500 ft min
-+-
10000
85 -
11
10
8asha in 2016
.......................... . . :. . . . :. . : i
4:ir.~~..-;-;-:-F:.~-~-.~.-.o-.
u.
0- - 0 -
,I
6000
.....~
~ -
4000
.() -
-0
0-
0 -
-0- -0- -
-----~-----x------x-~~~..~--~--~---..~
~
-&'--A-=--
2000
::J
-1
3
2
1
O~
__
1996
__
2001
____
2006
2011
__
____
2016
2021
__
__
2026
____
2031
2036
__
__
2041
____
2046
2051
__
~O
2056
Year
Figure A2.S.
Pakistan
Case study -
II
..-1
I
j
168
~
j
APPENDIX 2
Figure A2.5 shows the live storage volumes attainable with the chosen flushing
regime, compared with some of the other operational policies that could be
pursued.
These results were subsequently analysed along with the costs and the benefits
associated with each scenario in order to determine the most appropriate course
of action at Tarbela.
169
Appendix 30
A3.I.I. Sedimentation
By 1958, the original live storage capacity of the reservoir had-reduced by 59%
and the intake structure and screens were aLTJ10st buried. In addition, the highlevel outlets of the upstream sediment trapping reservoir had failed, with the
result that water and sediments were being discharged through the low-level
diversion tunnel. The problem became increasingly serious by the mid 1960s,
tr-l1'eatening the continued operation of the power station.
A3.I.2. Hushing
In 1969, it was decided to attempt sediment- flushing at Mangahao reservoir
through the low-level diversion tunnel. The tunnel had not been used for 25
years, owing to problems with the gate, but had been used routinely between
1925 and 1944. For a period of 24 hours 'nothing happened, then on the second
day silt began to extrude from the tunnel and the reservoir emptied, leaving a
crater-like depression in the 13 metres of sediment which had overlain the tunnel
entrance' (Jowett, 1984).
A month of flushing, using water released from the upstream reservoir,
resulted in 088 M.m3 of sediment being scoured from the reservoir basin,
equating to 75% of the sediment that had accumulated since 1924. It is reported
that large logs and tree roots had to be cleared from the entrance, but that much
debris passed through the tunne1.
Since the initial flushing operation the reservoir has been emptied and flushed
annually (Jowett, 1984). The power station is closed for three weeks for the
flushing operation, reducing the annual energy produced by 4%. The annual
EVACUATION OF SEDIMENTS
flushing of the reservoir has resulted in the removal of most of the sediment from
-Mangahao Reservoir and a considerable amount from the upper reservoir.
No information is available regarding the flushing discharges used, although
the fact that -the original diversion tunnel is used suggests that the available
discharge capacity is generous.
A3.2.1. Sedimentation
From the time of construction, about 66% of the total catchment fell within the
catchment of the Pathfinder dam, reducing sediment inflows to Guernsey
reservoir substantially. After the constluction of Glendo dam in 1957, only 40/0 of
the catchment was expected to be contributing significant sedilnent inflows to
Guernsey reservoir. Until 1957, the reservoir was subject to a high sedimentation
rate, losing 39% of its oliginal capacity over a period of 30 years, with deposits
cOlnprising 170/0 sand, 61 % silt and 220/0 clay. The maximuln depth of deposit
repol1ed was about 12 m.
A3.2.2. Flushing
Partial draw down (by 12-13 m) was carried out at Guernsey reservoir annually
between 1959 and 1962 and data were collected to determine inflow and outflow
rates and sediment movement within the reservoir. No definitive details are
available in the references concelning the bottom outlet or other flushing
facilities, although Morris and Fan (1997) suggest that the 'overflow spillway'
was used, which is possible if there are large spillway gates. The elevations of the
power intake and the sediment deposits in the vicinity of the dam are about 10 In
above the original bed and the aInount of draw down would be consistent with an
outlet at about this elevation. Flushing discharges were typically in the range of
120-140 In3/s, cOlTesponding to about double the nlean annual inflow.
172
APPENDIX 3
Although sediment was scoured from the upper portions of the reservoir
during the four years of drawdowil, n10st of this was apparently redeposited in
the lower part of the basin nearer to the dam, and the suspended solids
concentration in the water discharge from the reservoir never exceeded 08 gil.
From the inflow and outflow data during the period 1957-62, it was estimated
that only 144 000 m 3 of accumulated sediment was relTIoved from the reservoir
basin, the equivalent of less than 02% of the original capacity. The long time
before first flushing was probably a factor in reducing the erosion of the deposits
(Atkinson, 1996).
It was concluded (Jarecki and Murphy, 1963) that, with future annual
drawdowns following a similar pattern, only about 2% of the original capacity
could eventually be recovered. As sediment inflows had been severely reduced by
t.1}e construction of the upstream dams, this appeared to represent a satisfactory
state of affairs.
.
A3.3. I. Sedimentation
During the first two years of operation, the storage capacity reduced by 22% a
year and during the following eight years a further 32% of the capacity was lost.
Only 4% more was losfduring the next 18 years (1937-54), suggesting that an
equilibrium had been reached.
A3.3.2. Flushing
No details of the flushing facilities are given, but they are sufficient to pass over
double the Inean annual flow when the reservoir is emptied.
Plior to 1939, the reservoir apparently operated with a lilnited annual
draw down of 23 ill, but this was not effective. Between 1939 and 1966, it is
reported that 38 flushing operations (between one and four per year) were
undertaken with full drawdown. A wide range of results were obtained, due to
influences such as the duration of flushing and the amount of accumulation
173
EVACUATION OF SEDIMENTS
between flushing operations (Morris and Fan, 1997). Peak concentrations were
reported in only two events, but were very high at 270 and 370 gil.
Each flushing event comprised two stages: partial drawdown, as the reservoir
was being emptied, followed by total drawdown, with flows along the bottom of
the scoured channel through the basin. The duration of flushing varied from 85
to 655 hours, with a mean of 185 hours, and was canied out mainly in the
month of April, Mayor November.
The volumes scoured each year ranged between about 05 and 2 M.m3, with an
average of approximately I M.m3 This is substantially less than the reported
mean annual sediment inflow, suggesting that most of the annual sediment load
passes through during routine operation.
The data presented by UNESCO (1985) suggests a reduction in the total
volume of sediment contained in the reservoir since full draw down flushing
began, although there is some ambiguity in the plotted data.
General conclusions drawn (UNESCO, 1985) from the flushing operation at
the ZelTIo-Afchar reservoir were:
l
JI
j
!
~
i
an optimum flushing discharge between 400 and 500 m3/s produces the most
effective evacuation of sediment (higher discharges cause a greater water depth
at the dam, reducing the effectiveness of flushing)
during the process of sediment flushing, the most active erosion occurs in a
period of 8 to 10 hours after effective erosion starts
when the effectiveness of flushing starts to fall, it can be restored by
temporarily raising the water level for a short period.
J
A3.3.3. Downstream impacts
No information available.
A3A.I. Sedimentation
The erosion rate in south-west Taiwan is particularly high, due to climatic
conditions and geological conditions of soft and erosive rock. In an 18-year
period from 1938 to 1955, the storage depletion at the Jensanpei reservoir due to
silting was 426 M.m\ an average annual loss of 34% of the storage capacity
.
(Hwang, 1985).
174
- I
J
J
J
J
J
J
J
APPENDIX 3
A3.4.2. Flushing
L., 1955 a 15 m diameter flushing tunnel was built through the base of the darn,
as a result of the sedimentation problems, and annual flushing commenced.
Flushing is arranged by emptying the reservoir between May and July and
allowing free flow through the reservoir. This suits the water'demands, as sugar
mills do not use water between May and October.
Between 1955 and 1980, the sediment volume contained in the Jensanpei
reservoir remained almost constant, showing that the adopted flushing regime
was highly effective, albeit retaining only about 45% of the enlarged capacity.
The N aodehai reservoir is a flood detention reservoir situated on the Liuhe River
in an arid region in the north-eastern part of China. As the main purpose of the
reservoir was for flood control, it was built initially with ungated outlets near the
river's original river-bed. Control gates were subsequently installed (apparently
1970) to preserve clearer water for irrigation in the non-flood season. The
reservoir was designed to attenuate a peak: inflow flood of 3500 m3/s to 1640 m3/s
(UNESCO, 1985). The original design flood storage volume was a little less than
the mean annual inflow.
A3.5. f. Sedimentation
The heavily silt-laden Liuhe River has an annual average sediment concentration
of 77 gil, so that detention
floods resulted in the deposition of sedilnent
deposits on the floodplain wit.~in the basin. High volumes of deposition were
reported during floods in 1949 and 1963. Table A3.1below gives some data for
flood peaks which occurred over a few successive days in 1963.
A3.5.2. Fiushinoo
The flushing wpich occurs in a reservoir of this sort is essentially uncontrolled.
Although there is some scope for control since the installation of gates on the
outlets, no information is available to judge if their use has had any
on
sedimentation in this case.
UNESCO (1985) shows the reported variation in available storage capacity in
N aodehai reservoir, reducinQ: from 168 M.m3
1942 to a minimum of 97 M.m3
3
in 1950 and varying up to about 134l\1.m in 1972. Overall, the storage loss has
ranged between 20% and 42% and it appears to be dominated by massive
deposition in the largest floods, followed by a period of progressive erosion.
175
J
EVACUATION OF SEDIMENTS
J
Table A3.1.
Date
July 20-22
July 23-27
July 28-31
8457
8862
8308
]928
7980
]160
760
2470
440
617
3280
388
67
168
]47
23 0
673
129
99
113
104
]31
563
25
J
I
Silt quantity
i~
outflow (Mt)
-,
Between 1950 and 1958, for example, there was progressive erosion along the
valley bottom, vi11ually reaching the original 1940 thalweg. However, this did not
influence the levels of deposition over the flood plain, which continued to lise.
It is doubtful whether the uncontrolled erosion of sediment deposited in the
basin that occurs duling operation could be significantly enhanced without
mechanical intervention.
176
J.
1
~
APPENDIX 3
A3.6.I. Sedimentation
vVith the bypassing of bedload, settlement of suspended-sedilnent still posed a
problem for the GmUnd reservoir. After the Durlassboden reservoir was
commissioned, the annual sediment load entering Gmund reservoir reduced from
O 2 Mt to 007 Mt, comprising mainly sands and gravels. Since the 1960s, the
annual sediment load has been estimated to be equivalent to about 16% of the
reservoir's volume, illustrating the need for effective flushing measures.
Aggradation of sediment in the Gmund reservoir increased between 1948 and
1960, reaching a maximum of about 02 M.n13 (over 20%) in the early 1960s.
From then until 1981, as a result of annual flushing and the construction of
Durlassboden reservoir, the total sediment volume was generally less, with a
typical value of about 015 M.m3
A3.6.2. Flushing
The bottom outlet passes around the right abutment of the dam in a curved
tunnel, with an inlet elevation 28 m below the crest. (It presumably occupied the
diversion tunnel used during construction.) A second outlet was added in the
middle of the dam during the reinforcement of the dam in 1963, with an inlet
elevation 27 m below the crest.
During the first period of flushing, from 1948 to 1960, flushing was not
executed every year, but was carried out depending on the amount of sediments
accuTIlulated. However, from 1960 flushing was carried out every year. Initially,
flushing proved to be difficult due to the low flow from the GerIos stream which
was the only flow available. There was also a problem with the positioning of the
entrance to the bottom outlet, apparently some distance upstream of the dam. In
most instances flushing was carried out for a week. The efficacy of flush1."1g
improved from 1964, after t."1e addition of the second outlet.
From 1967 and the beginning of the operation of Durlassboden reservoir, tIle
period required for flushing reduced to a one day. Flushing efficiency was also
increased by the increased flow, by using the turbine water released from
Durlassboden reservoir as well as the natural flow from the Gerios stream.
The flushing operation for the reservoir since 1967 is as follows:
the reservoir is drawn dovvn by 9 m one week before flushing is to commence
the newer bottom outlet is op~ned the evening before flushing day and water
is discharged until the drawdown is 14 m
\I the original bottom outlet is opened and the reservoir is emptied
~ flushing is then performed through the night with the natural flow from the
GerIos stream
III the following day flushing continues with the turbine water released from
Durlassboden reservoir (25 m?/s) over a period of 3 to 5 hours.
s
177
EVACUATION OF SEDIMENTS
-r
J
The construction of Durlassboden reservoir upstream of Gmund reservoir has
-_been a key factor, reducing the incoming sediment load and aiding the flushing
process. The idea of separating the bedload from the suspended load has not
proved successful in the long run (Rienossl and Schnelle, 1982), due to the
maintenance costs resulting from wear on the long bypass tunnel.
-- f
J
-,
J.
i
i
~,
I
I
I
I
J.
178
J,
1.
1.
APPENDIX 3
The materials deposited during the flood ranged from silt, sand and gravel to
14% wood.
In the period 1953-78 a total of 379 M.m3 had been deposited, representing
69% of the original volume and an average annual accretion rate of 015 M.m 3,
or almost 3 % of the capacity per annum.
A3.7.2. Flushing
The flushing of the alluvium following the 1978 flood was accomplished in two
different phases and required a period of four and a half mont.1.s, commencing in
mid-November 1978 and finishing by the end of March 1979. Phase '1 was
accomplished by flushing th.rough the upper of the outlets, about 44 m below the
spillway crest, with low flow of approximately 300 lis, coming from a lateral
valley.
phase ran until the end of December and evacuated about 03 M.m3
of materiaL
Phase 2 started with the opening of the lower bottom outlet, at the base of the
impoundment, at the beginning of January. The water used for flushing was
successively increased from 1 to 15 m3/s by using part of the river run-off.
During this three-month phase it was estimated that approximately 21 M.m3 of
material was flushed from the reservoir. Flushing was assisted by the use of
bulldozers and shovels to remove the wood buried in the alluvium and to push the
fflaterial into the eroded channels.
During the fiusping operation it is understood that L1}e balance of the inflows
was passed through the sediment diversion tunnel, which bypasses the reservoir
basin.
No information is available on subsequent flushing.
179
EVACUATION OF SEDIMENTS
arm about 30 km to its confluence, with the Yongding about 5 km upstream of the
dam. The original surface area at the flood storage level was 229 kln2
The catchment area is 43 400 km2, which has a semi-arid continental climate,
with dry and very cold winters and 750/0 of the annual rainfall occurring between
June and September. The average annual rainfall (1951-84) is 420 mm, with a
range of 278-545 mm. The mean annual run-off in the period 1925-85 was
1250 M.m3 (Binnie and Partners, 1986), although it was also noted that there had
been a progressive decrease in flow between the 1950s and the 1970s, probably
due to the construction of about 300 reservoirs in the catchment and the increased
use of water for irrigation. Only about 2% of the total catchment drains to the
Guishui arm of the reservoir.
The original storage capacity was 2270 M.m3, which is about 80% greater than
mean annual run-off. This original storage capacity comprised 600 M.m3 of dead
storage, 660 M.m3 of 'benefit' storage (for river regulation and hydropower) and
1010 M.m3 of flood storage. Of the total original storage, about 60% lay in the
Guishui arm.
The dam has a gated spillway, which was under reconstruction in 1986 (Binnie
and Partners, 1986), to give a discharge capacity of 2950 n13/s.
There is an 8 m diameter bottom outlet tunnel, through which flows are
controlled by two sets of four sluice gates, with invert levels 27 In and 39 In
below the flood storage level. The maximum discharge capacity is about
560 m3/s, which is about 14 times greater than the mean run-off from the
catchment.
A3.8.1. Sedimentation
About 40% of the total catchment is classified as loess areas, with friable soils
and sparse vegetation cover.
There has been a wide range in annual sediment loads in the rivers entering the
reservoir, with a nlaximum of 132 Mt and minimum of 13 Mt in the period
1951-84 (Binnie and Partners, 1986). As shown in Table A3.2, there was a
Table A3.2.
Mean annual
sediment
inflow: Mt
Mean trap
efficiency:
%
In period
Cumulative
73
73
352
352
1961-70
19
96
151
503
1971-80
11
96
85
588
1981-84
100
23
611
1953-84
28
611
611
Period
1953-60
I
!
8]
180
APPENDIX 3
181
EVACUATION OF SEDIMENTS
accretion raising bed levels. Proposals were made for a two-phase approach
comprising:
(a) divelting sediment-laden flows from the Yongding arm of the basin into the
Guishui arm, to store sediment in its dead storage zone for a period of about
eight years
(b) subsequently raising the flood storage elevation and developing a system of
polders to contain the bulk of the sediment inflows at the upstream end of the
Yongding arm.
No information is available on whether these or other proposals were adopted,
or on sedimentation and flushing experience since 1986.
The Shuicaozi reservoir is located on the Yili River in south-west China and
forms part of a hydropower scheme involving four hydropower stations in
cascades on two adjacent river basins (IWHR, 1983). The reservoir is 6 Ian long
with a 37 m high dam, giving the reservoir an original storage capacity of
958 M.m3, which is less than 2% of the average water inflow to the reservoir of
514 M.m3
A3.9.1. Sedimentation
The mean annual sediment inflow was estimated at 063 Mt (UNESCO, 1985).
With no bottom outlet for sediment flushing available, sediment deposition was
severe, amounting to,85% of the original storage capacity of the reservoir by
1981. The remaining 1-4 M.m3 of volume was insufficient for flow regulation,
which required 36 M.m3 ,
182
"
1,
APPENDIX 3
The general procedure is that, after drawing down the reservoir, about 50 m 3/s
is released from the upstrealll reservoir fOl~ flushing. The duration of effective
flushing is generally about one day, during which time typically about 02 M.IU3
of sediment can be removed by between 1 and 3% of the annual flovv. This is only
about one third of the average annual sediment inflow. It was noted that, in the
vicinity of the dam, the amount of erosion was about 2 m, but 4 krn upstream
there was none.
The quantity of sediment flushed is reported to be limited by a combination of
factors, including consolidation of the fine silts and the deposition of bedload in
the upper part of the reservoir basin, but an inlportant factor must be the
relatively high elevation of the spillway through which the flushed discharge
must pass.
IWHR's (1983) characterisation of the flushing in Shuicaozi reservoir can be
interpreted as follows.
Retrogressive erosion of the bottom of the basin commences at the upstream
end of the pool, whose position is controlled by the spillway level. This limits
the erosion capacity of the flow.
In the early stages of flushing, the sediment derives mainly from the deepening
of the main channel.
o Subsequently, the sediment derives mainly from collapsing of the banks.
1"1 the top portion of the delta, on deposition, the main channel is scoured
res lIlting in a progressive increase in accretion on the flood plains.
In the lower part of the delta and in the reach in front of the dam where the
river valley is narrow, unconsolidated silts migrate towards the main channel
and the level of the entire cross-section can be lowered.
$
In the light of these observations and theoretical studies into the behaviour of
the sediments, nVHR (1983) proposed the following refinements to the flushing
procedure.
Flushing should con1IDence after achieving the maximum p~acticable. drawdown.
e In the early stages of flushing, ~he maximum practicable discharge should be
deployed to deepen the main channel.
The discharge should then be reduced, to lower water levels and to induce
more bank instability.
e Finally, the discharge should be raised again to flush out the sediment which
has accumttlated from bank sliding, etc., and enlarge the cross section of the
main channel.
C)
183
J
EVACUATION OF SEDIf'lIENTS
}
A3.IO. HEISONGLIN RESERVOIR (China, 1959)
J
J
I
!I
,1
J
,
II
184
J
J
J
J
APPENDIX 3
Drawdown flushing
The lowering of the reservoir pool in July initiates the erosional processes
associated with flushing. A large and highly concentrated discharge of sediment
occurs at the transition from drawdown to flushing. Smaller amounts of sediment
are removed thereafter, by base flow and by the smaller inflow events that can be
released through the bottom outlet. Due to the operational objectives and
constraints at Heisonglin, in particular the supply of irrigation water at
mal1ageable rates and the limited discharge capacity of the bottom outlet,
flushing ,"vith large flows to widen the. main channel cannot be undertaken.
Detention flushing
vVhen floods entering the reservoir storage during the drawdown period exceed
the release rate to irrigators, the pool retains the silt-laden water for petiods of
hours or days. As the silts settle slowly, if the flow can be discharged within a
couple of days, approximately 70% of the sediment can normally be evacuated
with the water releases. In addition, as the reservoir is emptied at the end of each
flood and riverine flow is again established, deposits of silt are scoured in t..l-J.e
same manner as for the initial draw down ..
Lateral erosion
Attempts were made to create a longitudinal channel running parallel to the main
flushing channel for hydraulic scouring of the floodplain deposits. Accidental
overflow from this channel initiated lateral erosion, which formed a gully across
the deposits. From this, it was recognised that erosion of the deposits would
proceed much faster by directing the flow along the high lateral slopes from the
sides of the reservoir towards the main channel. Because of the high gradient that
can be achieved by lateral drainage, even small discharges were highly effective
in eroding deposits of non-cohesive silt. This technique was applied for a total of
68 months between 1980 and 1985. Using a flow of only 02 m 3/s, 816000 m 3
of deposits were eroded into the main channel. This equated to a remarkably
higher sediment/water ratio of 023.
185
j,
EVACUATION OF SEDIMENTS
1,
the irrigation canals, are designed to accept very high concentrations of sediment,
in order to enhance soil fertility. ,
Environmental penalties are clearly either non-existent or very minor.
1
A3.11. SANMENXIA RESERVOIR (China, 1960)
Sanmenxia dam was completed in September 1960 and was the first to be built
on the middle reaches of the silt-laden Yellow River. The 96 m high concrete
gravity danl, which controls a drainage area of almost 700 000 km2, was planned
as a multiple-use project for flood control, hydropower, inigation, navigation and
ice jam control. The maximum historic flood at the site was 36000 m 3/s in 1843
(Morris and Fan, 1997).
It was designed originally with a full reservoir level of 360 m, giving
65 000 M.m3 of storage capacity, inundating 3500 kn12 of floodplain and
requiring the relocation of 870 000 people. To reduce these impacts during the
first stage of construction, the dam was built to an elevation of 350 m with a
maximum operating level of 340 m and an original storage capacity of
9640 M.m3 Because of the high sediment loads in the Yellow River, the original
plans included two sediment control measures:
the reservoir was to impound water continuously, but release 350/0 of the
sediment inflow as turbidity currents, through 12 outlets at an elevation of
300 m (40 m below top operating level)
"
total sediment inflow was to be reduced by 3% annually by soil conservation
works in the catchment, resulting in a 60% reduction over 20 years.
, These figures proved to be extremely optimistic (Monis and Fan, 1997).
1
1
1
I
,..l
t
1
J
, -'
A3.II.!. Sedimentation
The estimated sediment load prior to construction of the dam was 1600 Mt per
annum, with a median diameter of 003 mm, representing a mean concentration
of 38 gIl. Extreme concentrations of up to 940 gil had been measured in the
Yellow River in the vicinity of the dam site. About 60% of the annual sediment
load occuned with about 300/0 of the annual run-off in July and August.
Iffilnediately after impounding began, severe sediment problems became
evident. During the first 18 months of operation, 1800 Mt of sediment had
accumulated in the reservoir, representing 93 % trapping and the loss of about
20% of the storage capacity. In the next four years, 3400 Mt was deposited and
the total loss of storage reached 3700 M.m3 , or 400/0 of the original capacity
(Monis and Fan, 1997). This rate of accretion threatened to elilninate all the
project benefits, in addition to sedinlent deposits which were raising the bed
elevation and flood levels in the Yellow River as fat as 260 kin upstream of the
daln.
186
APPENDIX 3
Stage 1 (1960-62)
Impounding of the reservoir began in September 1960 and seli.ous deposition
was occurring by the time the water level reached 3355 m (45 m below the
planned impounding level). Sediment accumulation raised bed levels by 45 m
. near the upstream end of the reservoir and caused backwater effects 250 kID
upstream of dam.
This was endangering agricultural land in the floodplain of the Wei River,
which joins the Yellow River near the upstream end of the impoundment, and
industrial developments in Xian City.
Total deposition up to the start of the 1962 flood season was estimated as
1800 Mt.
Stage 2 (1962-66)
The reservoir operation was changed from April 1962 to maintain a lower water
level throughout the year, by using the 12 outlets at an elevation of 300 m.
However, the outlet capacity proved insufficient and water levels during periods
of high discharge were too high for efficient sediment release during the flood
season, resulting in the trapping of a further 3400 l'vlt over four flood seasons.
It was also noted that sediment trapping in the reservoir was having an effect
downstream of the dam, where degradation was occurring.
It became clear that additional sluicing capacity would be required.
187
EVACUATION OF SEDIMENTS
[
Stage 3 (1966-70)
Prior to the start of the 1966 flood season, additional sluicing capacity was
provided by the excavation of two 11 m diameter bypass tunnels, with an invert
elevation of 290 m and controlled by 8 x 8 m radial gates, around the left
abutment of the dam. Four of the eight power intakes were convelted to sediment
sluices and the pool level was lowered duling the flood season. This had the
effect of reducing the trap efficiency to 175%, but did not lower the bed
elevation at the upstream end of the reservoir basin, and deposition in the Wei
River was still a problem. Consequently, it was decided that still more low-level
sluicing capacity would be needed.
Stage 4 (1970-73)
In time for the start of the 1970 flood season, 8 of the original 12 river diversion
outlets, filled with concrete, were reopened at an elevation of 280 m. Reservoir
operation was changed to flood detention and sediment sluicing, with all outlets
constantly open. Sediment release efficiency reached 1050/0, representing a yearon-year reduction in the amount of accumulated sedin1ent in the reservoir basin.
The bed elevation at the upstream end of the basin fell by nearly 2 m. In 1973
five generating sets of 50 MW each were installed, replacing the original eight
125 MW sets.
Stage 5 (1973-78)
Once an overall sediment balance had been achieved and bed levels at the
upstream end of the reservoir were under control, it was decided that reservoir
operation could be modified after the 1973 flood season, to increase the project
benefits.
This regin1e, which is understood to have continued to the present day,
provides water for irrigation, hydropower, and ice jam control during the non-
flood months. At the start of the flood season, in July, all the outlets are opened
and the high-capacity bottom outlets allow a low pool level to be maintained. The
high discharges can-ying the .sediment load also prevent excessive deposition in
the Yellow River downstream of the dam.
Stage 6 (1978 onwards)
Repair of serious abrasion in the bottoln outlet was carried out, decreasing the
cross-sectional area of the outlets and reducing the low-level flushing capacity.
To cOinpensate for this, two additional bottom outlets were opened in 1990.
Since 1980, to avoid turbine abrasion by high sediment concentrations,
hydropower generation has been halted during the flood season.
Long-term storage capacity
Since the sediment balance was brought under control from about 1975, the net
storage capacity below an elevation of 330 In (10 In below top water level) has
fluctuated between about 3000 and 3300 M.m3 , representing 50-55% of the
original capacity at that elevation.
188
[
[
[
[
r
I
I
I
L
[
1
r
APPENDIX 3
A3./2.1. Sedimentation
During the period 1961-70, the average measured suspended-sediment concentration was 727 mg/l, giving an average sediment inflow of 153 Mt, suggesting
a maximum accretion potential of the order of 80/0 per annum. The particle size
distribution of measured suspended load comprised 12% sand, 60% silt and 28%
clay (Mahmood, 1987). In addition, the Kabul River carries a bedload of gravel
and cobbles, which were not included in the rneasuredconcentrations.
After the first year's operation, 30 M.m3 of sediment had deposited in the
reservoir, increasing to 70 M.m3
five years. By 1980, after 20 years of
operation,
reservoir had completely silted to the conservation pool elevation,
except for a 60 m wide by 6 rn deep channel on
right bank, where the po\ver
and ilTigation intakes are located. The reservoir deposits had an accumulation of
cobbles and boulders on the surface and, in 1983, NIahmood observed
189
EVACUATION OFSEDIMENTS
gravels and cobbles up to 75 nun being passed from the reservoir with the
irrigation supplies.
A3.12.2. Flushing
Five flushing operations were performed during the period 1976-79. The
flushing was carried out by lowering the water level to the spillway crest level.
The total duration of flushing was about 20 days and these operations removed
an estimated 42 M.m3 of sedimel1t deposits from the reservoir, amounting to
about 6% of the probable sediment inflow over the same period.
No information is available on any later attempts at flushing, but it appears that
the reservoir has essentially reached an equilibrium condition with virtually no
residual live storage capacity, and that it will not be practicable to increase the
live storage unless deeper high-capacity outlets are provided at the dam.
A3.I3. J. Sedimentation
The annual sedinlent discharge into the reservoir is reported as between 12 and
14 Mt. The volume of deposited sediment reached about 30 M.m3 by 1968 and
was reasonably stable at 50-55% of the original storage capacity up to 1970,
after which no further data are available.
r
190
APPENDIX 3
A3.13.2. Flushing
The dam has eight bottom outlets, 35 m below impounding level and 21 m below
the elevation of the power intake. These are reported to have a discharge capacity
of about 350 m 3/s at maximum impounding lev~l. From other data available, it
appears that this is the discharge capacity for each, giving a maximum discharge
capacity of 2800 m 3/s when the reservoir is full. This is much larger than the
mean inflow of about 500 m3/s.
Since 1963, drawdown flushing of the reservoir has been operated, which was
achieved by lowering the water level by 4-5 m during the May to August flood
season. The fact that the lowering is so modest suggests that it may have
depended on the use of a gated spillway, as well as the bottom outlets. The
available data are not entirely consistent, as a plot of sediment concentrations
entering and leaving the reservoir during the 1964 flood season suggests that
there would be net accretion, rather than an approximate equilibrium.
A3.14.1. Sedimentation
The vast majority of the annual sediment inflows occur with high discharges in
the months of March to June.
Sedimentation was a serious problem in the first 17 years of operation and
caused an average storage loss of 365 M.m3 per annum, equivalent to an annual
rate of 21 %. The trap efficiency during this period was estimated as 730/0, with
19/
EVACUATION OF SEDIMENTS
A3./4.2. Flushing
The reservoir was built with three bottom outlets on the right-hand side (total
discharge capacity 430 m3/s, elevation 1913 m) and two bOttOlTI outlets on the
left-hand side (550 m 3/s, 1938 m). These are near the bottom of the reservoir,
close to the original river-bed level and their total discharge capacity of 980 m3/s
compares well with the mean annual flow of 160 m3/s. (With the reservoir level
drawn down to 25%, the bottom outlet discharge capacity would be reduced by
about a factor of two, so would still be three times the nlean annual discharge.)
Because of the rapid and continuing reduction in storage capacity, a decision
was made in 1980 that the operating regime should be changed to incorporate
more pro-active sediment removal. A number of alternatives were considered,
before selecting the option of annually elnptying and flushing the reservoir.
The peak annual inflows to Sefid-Rud reservoir occur in the months of March
to June and the inigation period is from May to September. Accordingly, the
flushing programrne was designed to occur from October to February, virtually
emptying the reservoir down to an elevation of 197 m, then allowing the reservoir
to fill in time for the start of the inigation season. (During the first two years of
flushing, the reservoir was not completely drawn down because of fears that
unstable sediment would block the bottom outlets.) It would have been desirable
to flush the reservoir for a further period, to take advantage of the higher flows
that would further SCOUT the deposits and widen the main channel, but the
uncertainty of filling the reservoir for the irrigation period dictates the closure of
the outlets in February. Furthennore, the bottom outlets may not have sufficient
discharge capacity to allow this to be done.
The consequence of this operating regime is that the majority of the annual
sediment load enters the reservoir when it is either filling or full, so is likely to
be deposited until it can be subjected to erosion during the subsequent flushing
period.
The initial years of flushing provided very high sediment outflows and a rapid
recovery of storage capacity up to about 75% of the original in 1992. This was
expected to be approximately sustained with a continuing flushing regime, with
the possibility of an increase to about 900/0 with supplementary measures (as
described later).
The gross benefits of the flushing operations, which comprise the volume of
sediment removed plus the vollune of deposition averted, amounted to 320 M.ln3
over the first 10 years of flushing, equivalent to 19% of the storage capacity per
annum. The average suspended-sedinlent content in the flushing flow was 48 gil,
with a peak of up to 670 gIl. Selected statistics on the annual flushing period up
to 1990 are summarised in Table A3.3.
92
...
; '.r. . r
.:~
APPENDIX 3
Table A3.3.
F1ushing
year
Dra\vdown flushing
duration: days
Empty flushing
duration: days
'Water volume
used: M.m'
Sediment
removed: Mt
1980-81
61
536
10
24
1981-82
6S
390
11
12
1982-83
117
10
1513
26
52
1983-84
16
80
795
23
68
1984-85
19
138
1810
29
142
1985-86
18
129
1131
29
46
17
85
26
27
1987-88
24
86
1988-89
113
1989-90
103
681
351
744
10667
1986-87
Total
942
]812
22
1057
31
54
22
32
21
514
57
It was observed that sediInents were eroded during draw down and flushing by
three processes:
sheet erosion
channel erosion
bank failures.
Sheet erosion was the most important type of erosion during the first
draw down operation, comprising sheet flow and scour of recentiy deposited fine
sediments in the lower reaches of the reservoir. After the first couple of years of
drawdown flushing, the amount of sediment removed by this process was less
important, removing only a small part of the deposition which had occurred onto
the submerged floodplain in the previous period when the reservoir was filling or
full.
Channel erosion was the most important process from the third drawdown
period onwards, when the first full draw down was undertaken. The channel
banks would be near-vertical initially, but would then fail. The rate of sediment
removal was found to be sensitive to changes in discharge, which would trigger
off accelerated erosion.
It \vas concluded that the long-term storage recovery would be limited by the
narrow width of the Inain channel in relation to the overall width of the reservoir
basin. Furthermore, sediment would continue to be deposited on the submerged
floodplains during impounding periods, suggesting that, after the initial period of
recovery, the storage capacity would begin to decrease again in the longer term.
193
EVACUATION OF SEDIMENTS
Two novel methods were investigated to promote the removal of deposits from
the floodplains and to prevent a progressive long-term loss of storage, as
described below.
Lateral erosion by piping
-. ~
I
I
.--l
._- /
:--I
. -
APPENDIX 3
re-exposed for the next flushing season. By December 1989, after a total" of 95
days of operation, the diversion channel had reached an essentially stable
condition. The eventual channel top width ranged between 50 m and 200 m, but
no information is given on the depth or total volUlne eroded.
_I
Long-term predictions
Studies by Tolouie (1993) estimated that, by creating a new diversion channel
each year and by deploying 75% of the total flushing season inflow, it would be
possible to recover lost storage and maintain a long-telID storage capacity of
about 90%, compared with about 75% by flushing alone.
A3.15.1. Sedimentation
The capacity of the reservoir was seriously depleted by an average annual
sediment inflow of about 84 Mt (UNESCO, 1985). Morris and Fan (1997) show
a photograph of a water supply intake in the delta upstream of the dam,
apparently completely surrounded by sediment, but no further details are
available.
A3.15.2. Flushing
Little infom1ation is available on the flushing operations. Table A3.4 lists some
data for flushing operations carried out in July 1971 and July 1973.
195
EVACUATION OF SEDIMENTS
Table A3.4.
Flushing period
Water used:
M.rn 3
Silt inflow:
Mt
Net sediment
release: M.rn 3
612
35
175
545
33
125
A3.16.1. Sedimentation
From 1966 to 1973, the first eight years of the reservoir's operation, 319 M.m3
of sediment had deposited in the reservoir, representing 24% of the original
storage, with the height of the deposits behind the dam reaching 27 m. Deposits
near the dam were described as fine, with a Dso of 002 mm, becoming coarser
at a distance of 350 m to 800 m from the dam.
A. 3. 16.2. Flushing
The dam has a small outlet, 26 m above the base of the danl, with a discharge
capacity (at full impounding level) of 17 m 3/s, but there is also an outlet for flood
196
APPENDIX 3
Ij
i
i
I
I
Ij
I
discharge 145 m above the river bed, capable of passing a maximum discharge
of 1260 m 3/s.
Flushing was first can-ied out in July 1974, when the reservoir was emptied
and flushed for 37 days. During this flushing period, 08 M.m 3 of sediInent was
removed from the reservoir. The reservoir was then impounded for five years to
June 1979, before flushing for the second time for a period of 52 days during the
flood season. The second flushing period removed 103 M.m3 of sedilnent,
reducing the volume of sediment in the reservoir to 262 M.m3 (20% of the
original storage capacity). Emptying and flushing were subsequently undertaken
in 1982 and 1986.
During emptying and flushing, it was reported that strong retrogressive erosion
occurred as a result of lowering the water level. A channel was rapidly fonned in
the floodplain deposits regressing upstream and deepening continuously. In the
first 350 m from the datu, deposits on the floodplain collapsed and slid into the
main channel. In the upstream reaches, where the sediments were coarser, the
cross-section eroded was initially rectangular in form and was followed by the
collapse of the floodplain deposits into the main channel. Outflow concentrations
were reported to reach about 1000 gIl, irrespective of the flushing discharge
(UNESCO, 1985).
Experience at Hengshan reservoir suggests that flushing every few years is
sufficient in this case, which is probably aided significantly by the high gradient
of u~e original stream bed and the steepness of the valley sides. The efficiency of
the flushing was high when the main channel, which had been eroded in the
previous flushing, had been silted up by deposited sediments during a period of
several years. It was thought that greater recovery of storage capacity could be
achieved if the reservoir was to be emptied prior to the start of the flood.
197
J
EVACUATION OF SEDiMENTS
,J
A3./7./. Sedimentation
The average annual sediment inflow is 081 Mt, which would have a deposited
volume of the order of rather over 1% of the original storage volume. The annual
load is estimated to be distributed as follows:
18 % throughflow from normal hydropower and gate operations;
21 % deposited on telTaces;
7%
bedload trapped in reservoir; and
54% thalweg deposits, removed by flushing.
A nalTOW section of the reservoir 4 km upstream of the dam divides the basin
into upper and. lower parts. The upper basin is being progressively filled with
sand and coarse matelial, which is generally not removed by flushing, whereas
'the lower basin consists of a deep river channellnaintained by flushing, between
a series of relatively flat river terraces, onto which fine sediment is deposited.
For the first seven years it was apparently operated without flushing, with the
reservoir trapping 82% of the incoming sediment. Part of the suspended load was
transported by turbidity CUlTents to the area of the dam and, after several years,
were starting to interfere with hydropower production.
A3./7.2. Flushing
The dam has a single bottom outlet located near the thalweg of the original river
channel and immediately adjacent to the intake screen, a location that facilitates
flushing of sediment from in front of the intake.
Flushing operations at eachi reservoir have been well documented and have
been considered successful in preserving the storage capacity of the reseIVoir
(MolTis and Fan, 1997).
The first flushing operation was canied out in October 1973, to flush
sediments that had accumulated near the power intake. Owing to the success of
this operation, it was decided to carry out flushing every year during the wet
. season. During the 18 years from 1973 to 1990 the reseIVoir was flushed 14
times. Flushing was cmTied out in three stages:
Slow drawdown: the reseIVoir level was lowered from 990 m (full impounding
level) to 965 m at a rate of 1 mlday, with the turbines operating at full capacity
and supplemented by opening the spillway gates and the bottom outlet as
necessary
Rapid drawdown: the turbines were stopped and the bottom outlet opened to
evacuate the remaining water from the reservoir, which typically took between
5 and 10 hours
Free flow: this typically lasted 2 to 3 days and occulTed once the reseIVoir was
empty and the river was flowing freely along the original river channel.
At the end of the flushing operation, the outlet was closed up and the reservoir allowed to refill, typically taking between 16 and 21 days. The amount of
198
J
J
APPENDIX 3
sediment released during each stage varied considerably from one event to
another, reflecting variations in the rates of sediment inflow and different
intervals between flushing operations, as illustrated in Table A3.5.
Little erosion of the sandy or gravely material in the upper part of the reservoir
basin was observed during flushing operations. In the lower basin, minor gullies
developed across the terraces during the slow draw down period and on the
terrace slopes there was a tendency for deposited sediments to be eroded by wave
action. However, there was no general erosion of sediment from the surface of
the terraces.
The zone of ma"'{imum erosion was along the main channel, which is also
where most of the incoming sediment was deposited. The slow draw down
exposed channel sediments to scouring action, and finer sediments were
transported nearer the dam.
The erosion and release of sediment during the rapid draw down phase was
reported to be spectacular (Monis and Fan, 1997). During the last few metres of
rapid draw down hyper-concentrated flows were observed. A major part of the
Table A3.5.
Date
Slow
drawdown
Rapid
drawdown
Freefiow
Total
Oct. 1973
Aug. 1974
186200
225200
411400
Oct. 1975
Oct. 1977
40700
44000
19500
5000
Oct. 1981
14600
348900
113400
476900
Oct. 1982
5800
111 600
250900
386300
402400
114300
May 1980
Sept. 1983
I!
28700
84700
24500
545400
I
665500
Oct. 1984
23300
604600
June 1985
July 1987
Sept. 1988
61600
627 000
577100
Sept. 1989
42400
144300
482200
r---moo
Gsoo
278 700
347 100
653000
Oct. 1990
32600
1265700
199
EVACUATION OF SEDIMENTS
total sediment flushed on each occasion occun-ed during the final few hours of
the rapid drawdown phase and the first few hours of free flow conditions.
A3.IB.I. Sedimentation
As a result of glacial activity the sediment inflow to Gebidem reservoir is very
high, with an annual average of about 0-4 M.m3 , equivalent to over 4% of the
storage capacity. This is mainly granular material, ranging from very fine sand to
gravel, of which about 20% is between 1 mnl and 100 nlm in diameter. The
sediment load is strongly correlated with the flow hydro graph during the summer
months.
A.3.IB.2. Flushing
Because of the high sediment load in relation to the reservoir capacity, sediment
managelnent was planned for in the initial design. Consideration was given to the
alternatives of sedilnent bypassing and dredging before selecting flushing as the
. most practicable and economic option. Venting of turbidity cun-ents was also
considered, but was rejected because the sediments would be too coarse for it to
be effective:
The danl was designed with two flushing tunnels located directly beneath the
. power intakes and close to the original streanl-bed level. Originally, the low-level
200
APPENDIX 3
outlets each contained two gates; a radial service gate at the downstream end, and
flap gate at the upstream end that could be closed in emergencies or for
maintenance of the service gate and outlet tunneL To resist erosion, the entire
sutface of the outlet tunnel was lined with steel plate.
After 25 years of operation, erosion of the service gate seal on the bottom
outlet had become a problem, preventing an effective watertight seal to be
maintained. In .1995-96 a third gate was added to each outlet, for use as the
discharge control during flushing operations, allowing the original service gates
to be used only fully open or fully closed, without significant wear on the
replaced seals.
The reservoir is flushed between May and July every year, for 2 or 3 days.
Owing to the gorge-type geolnetry of the impoundment, flushing has resulted in
the entire reservoir basin being kept virtually sediment free. Flushing is carried
out prior to late SUnL.T..er floods, \vhen conditions favourable to flushing occur:
the flow of the Massa River is low enough (less than 20 rn3/s) to allow full
drawdovvn
the flow of the Rhone is large enough (greater than 40 rn3/s) to dilute and
transport the sediment-laden flows, but not sufficient to pose excessive flood
risks downstream
the 0 isotherm is located around 3000 m, which corresponds to stable
meteorological conditions (presumably indicating that a summer storm would
not interfere with the flushing operation).
I)
201
EVACUATION OF SEDIMENTS
Table A3.6.
Year
Duration:
h
Water
volume used:
M.m 3
Mean
flushing
flow : m 3/s
Gate
operations *
1982
56
238
11 8
1983
48
338
196
Sediment
removed:
M.m 3
Solids
concentration: t
0143
60
0175
52
0178
60
1984
68
297
121
1985
49
250
142
0150
60
1986
45
353
218
0212
60
1987
45
320
198
13
0192
60
1988
79
293
103
13
0176
60
1989
49
249
141
0150
60
1990
40
318
. 221
12
0191
60
1991
96
235
68
0270
115
1992
151
328
60
61
0197
60
1993
101
248
68
0260
105
Table A3. 7.
Parameter
Volume: m3
270000
70000
. ~
Deposited in gorge
81000
32000
-it
Delivered to Rhone
Delivered to Rhone via turbines
Total
202
157 000
70000
340 000
f[
tf
APPENDIX 3
The release of sediment into the fast flowing Rhone has appeared not to have
had any significant adverse effect on river morphology, but has helped maintain
sediment loads in the face of a long history of gravel extraction. However,
temporary high suspended-sediment loads and deposition on the river bed have
caused some problems at water supply intakes, and have also been linked to fish
kills in the Rhone.
Al/9.1. Sedimentation
There was scant information available to predict likely sediment loads at the time
of design, but the adopted design suspended load of 160 000 m 3 per year (based .
on lirnited actual data) was considered to be conservative. The bedload was
estimated at half the suspended load for design purposes. Thus, there was
expected to be the potential (with 100% trap efficiency) for the loss of up to 8%
of the storage capacity per annum.
.,
During the first four years of operation, from 1974 to 1978, the scheme
operator was required to continue generation without any interruptions for
sedii'1lent flushing. Generally, the reservoir was held at the highest level possible
at the time, as this minimised the passage of the highly abrasive sediment through
the turbines, maximised the generating head and allowed for the easy release of
floodwater over the spillway.
Surveys of the reservoir bed carried out in February 1976 and April 1978
indicated that 058 M.m3 of deposition had occurred over two flood seasons,
suggesting a rate of accretion of 029 M.m3 per annum, which was rather greater
tt~an the design estimate of 024 M.m3 This was attributed, at least partly, to road
building and deforestation in the catchment. On the other hand, the estimated
accretion over the first four years amounted to about 25 % of original storage
capacity, so was a little less than might have been expected.
203
..
EVACUATION OF SEDIMENTS
A3.19.2. Flushing
The dam is provided with three bOttOITI outlets for flushing sediment froin the
reservoir, two in the deeper San Domingo river valley and one in the Aracay river
valley. The outlets are each equipped with a 3 x 5 rn radial gate and a 32 x 25 m
sliding gate. At normal top reservoir level (1585 m), the total discharge capacity
of the three bottom outlets is 170 m 3/s. They were also considered large enough
to deal satisfactorily with major obstructions, such as tree trunks.
Hydraulic model studies during the design stage had indicated that sediment
could be effectively flushed from the reservoir both under pressurised and freeflowing conditions, which_ are described below.
Pressurised flushing. In this condition, the bottom outlets are submerged by
sediment and flushing is started by inducing a 'piping' failure of the overlying
sediments, which are discharged through the outlet at a high concentration;
This causes sufficient sediInent to be eroded in order to provide a clear
pathway to the bottom outlet. Special attention should be paid during the
operation to check on the size of sediment load and to ensure that an adequate
flow is discharged, capable of transporting the material to the river-bed
downstream.
Free-flow flushing. This is used when the outlets are clear of deposited
sediment and usually begins when the level of the reservoir is already low and
the sediment load is moving towards the central channel of the reservoir. Freeflow conditions are capable of flushing greater sediment loads, but at the cost
of consuming greater volumes of water. Experiments showed that 12 000 n13
to 15 000 m 3 of sediment could be passed through the bottom outlets per day.
., First flushing operation. The first flushing of Santo Domingo reservoir took
place in May 1978, after four years of operation, when the powerhouse was
closed to enable a complete inspection of the reservoir and the plant to be
undertaken. It was estimated that the bottoin outlets flushed between 50% and
60% of the deposited sediment in a period of only three or four days of freeflow flushing, at a time when the inflow was 8-10 m 3/s. The bottom outlets
were covered by deposits by the fourth day, so the reservoir was allowed to fill
overnight, following which flushing under pressure cleared the botton1
outlets.
Some three weeks later, with inflows remaining low in both rivers, it was
decided to attempt to accelerate the flushing operation by using two bulldozers
to move deposits out of reach of active erosion towards the main strealns of
each river.
The entire flushing operation was sufficient to relTIOVe the majority of the
sediment which had been deposited in the reservoir basin over the four years
of operation. A subsequent topographic survey established that 062 M.m 3 of
deposits had been removed, restoring the storage volume to 2 85 M.m3
From the first flushing operation, the following conclusions were drawn
(Krumdieck and Chan10t, 1979) specifically for the Santo Domingo reservoir, but
also having application for other small reservoirs with heavy sedin1ent loads:
204
APPENDIX 3
flushing should take place annually, during, and preferably towards the end of
the high-flow period
'I even under low-flow conditions, hydraulic flushing can be effective
e flushing operations should begin when the sediment deposits are not less than
100-200 m from the face of the dam
free-flow flushing is generally more effective than pressure flushing, but freeflow flushing should be intenupted occasionally, to catTY out pressure flushing
of deposits around the entrances and exits of bottom outlets (for up to 10
minutes at a time, eroding up to 5000 m 3).
e
A3.20.1. Sedimentation
The mean annual suspended-sediment load inflow is given as 053 Mt, the vast
majority of which enters during the flood season, July to September. The annual
average covers a wide annual variation, of between 012 Mt and 134 Mt in the
period 1974-83 (Chen and Zhao, 1992).
In 1975, the second year of operation, a major flood occurred which deposited
gravel 3-4 ill thick, 1 krn upstream of the oliginal impoundment.
By the end of 1983, 53%
storage capacity in the Nanqin reservoir was
reported to
occupied by deposited sediment and it was estimated that the
span of the reservoir would end by the
2000 (Chen and Zhao, 1992). (From
the quoted data, this percentage loss apparently
to an intermediate
205
EVACUATION OF SEDIMENTS
impoundment level of 118 m.) The Inaximum depth of deposition near the dam
was of the order of 12 m.
A3.20.Z. Flushing
A 3 m diameter tunnel, 3 m above the original river-bed level, was built into
the dam for the purpose of sediment flushing. This has a discharge capacity of
14 m 3/s when the pool level reaches the soffit, rising to 110 m 3/s at maximum
impounding level.
206
APPENDIX 3
A3.21.1. Sedimentation
Suspended sediment inflows to the reservoir have been estimated by one or two
daily samples, supported by more detailed sediment concentration profiles. There
are no direct measurements of bedload entering the reservoir, but values have
apparently been inferred from outflow sediment measurements and surveys of the
reservoir basin. Between 1976 and 1984 the estimated total annual amounts of
sediment inflow have ranged between 049 an.d 29 M.m3, with a median value of
22l\1.m3, which is about 20% of the original storage (Bhargava et al., 1987).
The reservoir began impounding in March 1975 and the sediment deposited
was surveyed after one year of operation, by which time it had reached the crest
of the spillway, which is 16 m below the full reservoir level, reducing the storage
capacity by 23%. Between then and 1981, the sediment level throughout the
reservoir basin rose progressively, reaching a total storage loss of 60%. Table
A3.8 summarises some key data regarding the sedimentation of Ichari reservoir
(Bhargava et al., 1987). It is notable that over 90% of the very high sediment load
in 1978-79 was passed downstream.
In an inspection of the roller bucket of the gated spillway in 1984, severe
damage of the teeth was found, including exposed concrete surfaces (which'Ii'ad
been eroded sufficiently to expose the reinforcement) and steel plate armouring
(some of which had been totally removed and washed away). The damage was
att.ibuted to the impact of cobbles and pebbles passed through the spillway after
the reservoir had silted up (Bhargava et aI., 1987).
PJ.21.2. Flushing
The po'wer intake incorporates facilities for sediment exclusion (with the
excluded sediment discharged downstream of
dam), but no details are given
of any facilities for flushing sediment from the dead storage of the reservoir
basin. It is understood that the only facility for sediment flushing from the
reservoir basin is by opening the
spillway, which is done during the rainy
207
EVACUATION OF SEDIMENTS
Table A3.8.
Year
(June to
May)
Total
water
inflow:
M.m3
1975-76
Total
sediment
inflow:
M.m3
Sediment
trapped:
M.m3
262
1976-77
5049
163
061
1977-78
6455
371
012
1978-79
7825
2902
186
207
114
capacity: M.m3
Dead
Live
Total
393
500
893
340
492
832
480
820
425
634
160
360
520
209
.l
1979-80
3420
f
::
4585
480
060
114
346
460
5445
109
-{)41
108
393
501
1982-83
4716
049
035
108
358
466
1983-84
5148
243
023
120
323
443
1980-81
1981-82
season, whenever the powerhouse is closed. The spillway gates are fully raised,
to allow free flow through the reservoir along the top of the deposits.
Measurements are made during these periods, froin which the quantities of
sediment flushed can be calculated. It appears that flushing by this method has
been undertaken annually since 1976-77 and accounts for between about 30% of
the annual sediment discharge in years with low sediment loads, increasing to
70% or more in years with high sediment loads.
It appears from the information given in Table A3.8 that the regin1e of annual
flushing is likely to result in a fairly stable residual storage capacity of the order
of 4 M.m3 (Atkinson, 1996), but no more recent data are available in the
Iiterature.
downstream reach of the liver. The sediment flushed from the reservoir is carried
by the water discharged. No details are given of possible environmental
impacts.
Baira reservoir forms part of a hydropower project which utilises the combined
flow of three tributaries of the River Ravi in the nOl1h west of India. The 51 m
208
APPENDIX 3
high elnbankment dam (earth core, rockfill shoulders) diverts the flow of the
Baira River to a network of tunnels leading to the powerhouse. The oliginal
storage capacity of the reservoir was 2-4 M.m3 (Paul and Dhillon, 1988)
representing only about 01 % of the annual inflow (which is variously reported
as 1900 M.m3 or 3500 M.m 3).
A3.22.1. Sedimentation
The reservoir is subject to both monsoon and winter floods carrying high silt
loads of up to 100 gil (Jaeggi and Kashyap, 1984). A mean annual rate of siltation
had been estimated at 0092 M.m3 , but in the first 18 months of operation, a silt
volume of 045 M.m3 had accumulated, representing allnost 20% of the original
capacity and suggesting an annual sediment load of at least 03 Mt (Atkinson,
1996).
A122.2. Flushing
The 5 x 7 m diversion tunnel for the construction of the dam, with an upstream
invert level of 1088 Ill, which is believed to be at least 35 m below the maximum
impoundment level, was equipped with a service gate and an emergency gate to
facilitate flushing. Model studies carried out during the design stage (albeit with
a different design of flushing tunnel) had indicated that almost the entire silt
content upstream of the tunnel could be flushed out.
The first flushing operation. was undertaken in August 1983, with the objective
of achieving the maximum possible volume of silt removal, adopting the
following sequence:
e
e
1/
CD
the reservoir was drawn down to the minimum normal operational water level
(1113 m), following which flows to the powerhouse were stopped
the diversion tunnel was opened to allow a discharge of 150 m 3/s until
reservoir was almost empty
the diversion service gate was opened fuHy
water was fed to t.~e reservoir from,the Suil and Bhaledh to clear silt from
around the associated structures
flushing ceased when the concentration in the discharge had decreased to
about 10 gil
the diversion tunnel gate was closed and the reservoir refilled.
209
EVACUATION OF SEDIMENTS
Kashyap, 1984). Bearing in mind the steepness of the valley sides and the small
size of the reservoir basin in relation to the annual inflow, it appears that the
recommended flushing regime should be capable of maintaining a high
proportion of the original storage capacity in the Baira reservoir.
210
Appendix 4.
Erosion
A4.1. FACTORS THAT AFFECT EROSION
A4.1.1. Definition
A fundamental definition of erosion is the detachment and removal of rock
particles by water and other geological agents such as wind, waves and ice
(Mahmood, 1987). A broader definition would include the subsequent removal of
material deposited temporarily at another point in the catchment. The rate of
erosion is generally expressed as the mass of sediment removed from a given
area per year (tlkm2/yr). It vruies with climatic, geological, topographic and
human factors. Sediment yield expresses the quantity of material that reaches a
defined point on a river draining the catchment and therefore the quantity
entering a reservoir created by the construction of a dam at this point. The
quantity will depend on the effectiveness of sediment transport in the basin. The
majority of sediment yield studies consider only the suspended part of the total
load. Bedload is generally assumed to be a minor part, representing about 10%
of the total, even though in extreme cases it can vary between 4% and 60%
(Jansson, 1988).
A4.1.2. Climate
Precipitation
The rate of erosion depends on the erosive power of the rainfall which is related
to the intensity, droplet size and total quantity. High intensity, short duration
events produce more erosion than long duration staTInS of low intensity. Storms
with large rain drops are more erosive than drizzle with small droplets (Goldman
et at., 1986). Tne effect of rainfall intensity is illustrated in Table A4.1 by data
for 183 events which caused erosion at Zanesville, Ohio, between 1934 and
1942. They show that the average soil loss per rain event increases with the
intensity of the storm (Fournier, 1972, reported by Morgan and Davidson,
1986).
Seasonal variations in erosion rates are influenced by previous meteorological
conditions. The moisture content of the soil and hence the infiltration capacity
will depend on previous rainfall and this will affect the amount of run-off which
in turn has a direct
on erosion rates (rv10rgan and Davidson, 1986). Highest
erosion rates are likely after a long dry period when there will be a supply of
readily erodible material (Morris and Fan, 1997).
211
J
EVACUATION OF SEDIMENTS
]
Table A4.1.
40
037
255-508
6]
06
509-762
40
118
763-1016
19
114
1017-1270
l3
342
1271-152-4
363
1525-1778
387
177 9-2540
479
0-254
],
j
i
212
~.
APPENDIX 4
800
>-
C)'
><'t'
600
.:s::
~
Qi
Qi
';:;"
"E
CD
E
E
'5 200
CD
's:,
400
"E
tlJ
'6
U)
en
250
500
750
1000 1250
Effective precipitation: mm
(i)
1250
>-
c::::
-s:,
a:;
1::
tlJ
E
1:: 500
tlJ
E
a:; 100
10
'5
1'
tlJ
(j)
150
100
10
100
1000 10000
400
800
1200
1600
(ii)
L 1200
>-
~ 800
-c
Q.i
':;'
'5-
C
<ll
E
50
en
25:
(i)
><'t'
'6
750
';:;"
-L
-c
a:;
l-
tlJ
1200
1000
;;'1000
oX
en
800
-L
1000
'6
600
(a)
10 000
C\I
400
200
'E 400
<ll
E
'0
Q)
(j)
200
400
600
800
1000
400
(iii)
Figure A4.1.
800
1200
1600
2000
2400
(iv)
213
EVACUATION OF SEDIMENTS
and shows the first peak at a preCIpItation of 450 mm, which reflects the
relationship proposed by Langbein and Schumm (1958). The two initial peaks
are similar to the curve produced by Wilson (1969) but the magnitudes of the
peaks are 450 mm compared with 750 mnl, and 1350 mm compared with 1750
mm, which conesponds to the troughs in these curves. The relationship between
sediment yield and Inean annual run-off is similar to that produced by Douglas
(1967) although the peak in semi-arid areas is less pronounced, with maximum
yields occurring in areas of high annual run-off.
Other factors such as relief, geology and human impact may be more
important controls at a global scale than precipitation. The seasonality, intensity
and type of rainfall and its effect on vegetation cover are also important measures
of the effect of precipitation on erosion rates .
Run-off. Run-off results from an excess of precipitation over the sum of
infiltration and evapotranspiration and is the quantity of water available to
convey the products of erosion. Factors that encourage infiltration and thereby
reduce run-off will decrease the quantity of erosion. Run-off may provide a
better correlation with rates of erosion than precipitation does. Low run-off
rates indicate aridity and hence poor vegetation cover compared with high
rates that indicate an excess of water and therefore dense vegetation cover. A
10000
'0
....
:.
>-
.
. ~:
: ..
..
00
1000
i:i
Q)
':;"
c
Q)
E
:0
(!)
C/)
"C
Q)
100
"C
c:
Q)
c..
C/)
:J
C/)
(ij
:J
c:
c:
ns
ns
Q)
:2
c:
10
1L-__
L-~
200
__
400
__
600
__- L_ _- L_ _- L_ _
800
1000
1200
__
1400 1600
__
1800
~~~
____
____
6000
__
__
8000
Figure A4.2.
tion
214
APPENDIX4
similar general relationship will exist between run-off and erosion as for
precipitation and erosion; with maximum erosion "levels at intelmediate
values.
o Temperature. The temperature will affect vegetation growth and evapotranspiration rates. Where there are high temperatures, higher rates of
evapotranspiration occur and therefore larger amounts of rainfall are required
to cause erosion. High temperatures will also cause more rapid rates of
vegetation growth which will reduce run-off rates and erosion. The ilnportance
of temperature will depend on the quantity of precipitation. In regions with
high precipitation quantities the relative importance of temperature is likely to
be reduced.
Wind speed and direction. The wind speed and direction will affect the
movement of soil particles. In areas where the wind speed is high and there is
a lack of vegetation to hold the soil pa.'1:icles togell-J.er high rates of wind
erosion are likely to occur. Wind erosion is inlportant in arid or semi -arid
regions as an agent that can transport sediment from ridges into depressions
which can then be transported by run-off.
1200
r-
j\ .. .. \. //
4DO
\.of
;/0 0 \./
Qi
's;.
400
800
Q)
ca::l
1600
1200
2000
2400
2800
3200
3600
4000
'6
<ll
en
o~
0/\ ./
800
1200
c::
c::
ctI
Q)
800
400
------------
~.~./.
0\./ . 0
1L-____- L______L -____
400
800
______
1200
1600
____
______
2000
2400
____
____
2800
~~
3200
Figure A4.3.
215
EVACUATION OF SEDIMENTS
J
Table A4.2.
Lithology
143
95-143
571
523
1237
761
A4./J. Geology
Rock type
The geology of the area has a major impact on the rate of erosion as it determines
the susceptibility of the rock to the effect of other factors. Table A4.2 shows that
under the same climatic conditions geology can produce a variation of ten times
in the sediment loss.
Generally, suspended-sediment loads are greater in areas of sedin1entary rocks
compared with crystalline rocks by a factor of 2-4 and .compared with areas of
mixed rocks by a factor of 14 (Dedkov and Mozherin, 1992).
J
.J
J
A4.1.4. Soils
The soils in arid and semi-arid environments with sparse vegetation cover are
very different from soils in more humid regions. The key soil characteristics
influencing erosion rates are the texture, structure, organic matter content, shear
strength and infiltration capacity.
Texture
The texture describes the sizes and proportions of the particles making up the
soil. Soils with high sand contents are coarse textured with high infiltration rates,
low run-off and relatively low erosion potential. Soils with a high content of silts
and clays are fine textured, the clay binds the soil and makes it resistant to
erosion. Soils high in silt and fine sand and low in clay and organic matter are the
nlost erodible. Well drained sandy and rocky soils are the least erodible as they
have large particles which require large forces to transport theln (Goldman et ai.,
1986).
216
.,.....L
APPENDIX 4
Soil structure
Soil stnlcture is the anangement of particles into
The soil structure
affects the soil's ability to absorb water. When the soil
is compacted or
crusted, water tends to run off rather than infiltrate. Granular structure is the most
desirable to minimise erosion as it absorbs and retains water, reduces run-off and
encourages plant growth (Goldman et al., 1986).
Organic content
Organic matter improves the soil structure and increases
permeability, water
holding capacity and soil fertility (Goldman et al., 1986). Clay content can be
used as an indicator of erodibility as it combines with organic matter to form soil
aggregates and it is the stability of these particles which determines the resistance
of the soil. Soils with an organic content of less than 3 5% are highly erodible.
Shear stren~J1
This is a measure of the cohesiveness of a soil and its resistance to shearing
forces exerted by gravity, moving fluids and mechanical loads. Its strength is
derived from frictional resistance met by its constituent particles when they are
forced to slide past one another or to move out of interlocking positions. The
higher the shear strength of a soil the more resistant it is to erosion (Morgan and,
Davidson, 1986).
Infiltration ratelpermeabiliv/
infiltration capacity is the maximum sustained rate at which soil can absorb
water and is influenced by pore size, pore stability and the form of the soil
profile. Soils with stable aggregates maintain their pore spaces better while soils
with swelling clays or minerals that are unstable in water tend to have low
infiltration capacities. Where infiltration varies with depth, the horizon with the
lowest infiltration capacity is critical. Texture, structure and organic matter all
contribute to the pelweability of a soil. High erosion rates occur where
rates are low and large volumes of run-off are ....,"". . . ""............ ' -' .
,...
1"\ .....
't"I
217
EVACUATION OF SEDIMENTS
they experience greater fluctuations in air and soil temperature. North facing
slopes are cooler and more moist with less sun.
x
x
104
i::i
05
";;:..
C
~
'5
Q)
103
o
o
o
102
1::..
o
+OX
10 1
o
+
q.
0
C/)
1::..0
-k
1::..
10oL-~--~---L--~--~~~~--~--~--~--~~
10-3
10-2
10-1
Drainage area: 106 km2
Figure A4.4.
218
100
101
APPENDfX4
relationspip between suspended sediment yield and drainage basin area (Walling
and Webb, 1996),
Drainage density
Drainage density is all expression of the distribution of streams in the drainage
system. It is a crude indicator of run-off and is often used as an index of the
severity of erosion - areas of high drainage density being associated with
elevated erosion rates. Broadvariations in drainage density on a macro-scale are
associated with differences in climate. At the meso-scale, regional variations can
be related to differences in rainfall volume but are complicated by lithology and
relief. Micro-scale differences in soil type and frequency and intensity of
individual climatic events are important (Morgan and Davidson, 1986).
Vegetation
Vegetation is the most important erosion control factor. It dissipates the energy
of rainfall, prevents rain impact on the ground, reduces splash erosion, increases
infiltration, decreases surface nln-off volumes and velocity, holds soil particles in
place and maintains the soil's capacity to absorb water.
The type of vegetation cover is dependent on the raiilfall, temperature, soils
and topography of a region. These factors interact to produce distinctive zones
called biomes. Climates with relatively mild year-round temperatures and
frequent~ regular rainfallare favourable to plant growth. Cold and dry climates
are less favourable to growth and therefore more susceptible to erosion (Goldman
et ai., 1986).
Land use
The land use of an area is influenced by the topography, geology, soils and
climate of a region. These factors determine the use to 'which land is put by
humans. Cultivation may decrease the erodibility of clay soils but increase that
of sandy soils (Morgan and Davidson, 1986).
Soil loss from hillslopes in West Africa between a gradient of OJ:) and 4
experienced mean annual erosion rates of 0015, 002 and 0003 kglm2 under
natural conditions of open savanna grassland, dense savanna grassland and
tropical rain forest respectively. Clearance of the land for agriculture increased
rates to 08, 26 and 90 kg/m2 while leaving the land as bare soil produced rates
of 2, 3 and 17
. The removal of rain forest produces greater rises in erosion
rates than the removal of savanna grassland (Morgan and Davidson, 1986).
Areas of low precipitation are more vulnerable to land use changes. Changes
in one part of the ecosystem may produce changes in the basin condition and
In semi -arid regions the
response and recovery may take a long period of
recovery time is four times that of humid areas (Walling and Kleo, 1979).
on erosion rates can
The effect of cultivation or lack of soil cover by
be seen in Table A4.3.
to bare ground can also
The increase erosion rates from natural
be seen graphically in Figure A4.S which shows the results of soil erosion tests
under different vegetation cover at Mpwapwa, Tanzania.
219
EVACUATION OF SEDIMENTS
Table A4.3. Rates of erosion in selected countries in kg/m2/yr (after Morgan and
Davidson, 1986)
China'
Natural
Cultivated
<02
15-20
Bare soil
28-36
I
USA
05-17
0003-03
04-90
0003-002
001-90
10-750
Nigeria
005-01
001-3 5
03-15
India
005-01
003-20
10-20
Belgium
001-005
03-30
07-82
UK
001-005
003--03
10-45
Ivory Coast
Human impact
It is estimated that human activities have degraded 15% (200 million ha) of the
land between 72N and 57S. Around half of this is due to hUlnan-induced water
erosion, a third due to wind erosion with most of the balance due to chemical and
physical deterioration (US Global Change Research Information Office, 1999).
Activities such as deforestation, urbanisation and agriculture all increase the
erodibility of soil. Present rates of erosion are approximately two and a half times
historic rates mainly due to human influences. With the conversion of forest to
agricultural land there Inay be an increase in sediment yield at the basin mouth
by three and a half times (Mahmood, 1987) - see Table A4.4.
Explanation:
Ungrazed thicket
Ot
19%
rI
260%
50-4%
Figure A4.5.
220
J.
. i
APPENDIX 4
Table A4.4.
Erosion rates for different land use categories (Morris and Fan, 1997)
Land Llse
Forest
Under natural conditions erosion rates in mountain zones are 27 times greater
than in lowland areas. The influence of man has increased sediment yield from
mountainous regions by 14 times; however, larger increases in lowland areas
have reduced the difference between mountainous and low land regions to 32
. times. For eX3J.I1ple, sediment yields in the sout.h and middle Urals are up to
~ 30 tlk..rn2/yr, less than the neighbouring eastern part of the Russian plain where
rates of up to 200 tlkn12/yr occur (Dedkov and Moszherin, 1992).
Increases in sediment yield caused by human activity are demonstrated in
Table A4.S.
10
221
I'
EVACUATION OF SEDIMENTS
Table A4.S.
1983)
Increases in sediment yield due to land use changes (Walling and Webb,
Region
Factors for
increase in
sediment yields
Source
Rajasthan, India
Overgrazing
4-18
Utah , USA
Overgrazing
10-100
Noble (1965)
Oklahoma, USA
50-100
Oklahoma, USA
Cultivation
5-32
Texas, USA
340
Northem California,
USA
5-25
Anderson (1975)
Mississippi, USA
10-100
Southern Brazil
4500
Westland,
New. Zealand
Clearfelling
Oregon, USA
Clearcutting forest
39
Fredriksen (1970)
The altitude has a distinct effect on the climate causing a knock-on effect on
the vegetation of each zone, as shown in Table A4.6. There are two rainy seasons:
from March to May and from October to December.
Sediment yields from the forested areas of Mount Kenya are around
20 t/km2/yr rising to 1000 t/km2/yr on grazed areas and lnore than 3000 t/km2/yr
on steep cultivated areas of the basin. This shows the strong influence of land use
on sediment yield. Under natural conditions cultivated areas have ground cover
of crops for around eight months of the year. Soil losses from grazing lands are
therefore generally higher and increase markedly as basal cover declines. In
cultivated areas, rural roads yield 5-20% of the sediment yield (Ongweny,1979).
Table A4.6.
Altitude: m
Vegetation
Rainfall: mm
Soil
>1800
1800
Clay loam
>1400
1400-1800
Clay loam
1100-1400
<1000
222
900-1400
Farming
Marginal farming
APPENDIX4
Comparisons between the data for the Upper Tana basin and that produced by
Dunne, relating mean annual suspended-sediment yield to annual run-off
depending on land-use, showed good agreement. This can be seen in Figure
A4.6, which shows the relationship between mean annual suspended-sediment
yield and mean annual run-off.
-- --
__~-L__~~~~~~--~~--~--~~--~
100 200 300 400 500 600 700 800 900 1000 1100 1200 1300 1400 1500
Mean annual run-off: mm
1~~~
III 0 Forest
r:.
Forest> Agriculture
Agriculture> Forest
Grazing
Scrub lorest
Figure A4.6.
223
EVACUATION OF SEDIMENTS
In all three regions the slopes are longer than 500 m with only a small portion
of gradients over 01. The n1aximum rates of erosion occur at the central, steepest
part of the hillslopes. The sediment yield on the slopes is 2500 t/km 2/yr to
3150 tlkm2/yr, on the basement rocks 11 300 tlkm2/yr and on Kilimanjaro lavas
17 600 t/lGl1 2/yr (Dunne et a!., 1979).
Table A4.7.
Region
224
NOlthern
12-2045
Central plains
20-570
East
27-356
South
30-1787
. - :!
. i
-~.
APPENDIX 4
Table A4;8.
Sediment contribution for sections of the Yellow River (Tal Wei Soong and
Yean Zhao, 1994)
Upper
Area: km'
385000
Length of
channel: km
Middle
Contribution:
% water
Contribution:
% sediment
111000
48-7
9-0
36-6
893
345000
1206
111400
22000
786
1/8000
Lower
3472
Average
slope: m/m
11
17
Sanmenxia reservoir
The drainage basin to the reservoir is 688 400 km2, constituting 92% of the
Yellow River basin. The Yellow River drains China's semi-arid loess plateau
composed of thick aeolian deposits of silty soils. Due to the high erodibility of
this soil, intensive land use, inadequate soil conservation practices and virtually
limitless supply of sediment, the load through the valley is high. Rates of
sediment trarlsport are especially high in July and August which account for 60%
of the total annual sediment yield and 30% of the annual run-off. Sediment
discharge averages 16 billion tlyr, equivalent to an annual sediment yield of
2300 t!k..rn2 and a mean suspended-sediment concentration of 38 gil ~Aorris and
Fan, 1997).
S = 0-065 A
S = 0278 AO- 8i5
S = 0200 A 0-887
225
EVACUATION OF SEDIMENTS
The Krishna River rises in the Western Ghat mountains at an elevation of 1400 m
and flows 1400 k:m to the Bay of Bengal. 40% of the area is mountainous. 80%
of the basin is formed on archaen and younger crystalline rocks and 20% on
Deccan Traps and recent sedin1ents. The discharge is 3 m 3/s to 3400 m 3/s with the
majority of the flow occurring during the monsoon in July and August. The
maximum sediment yield occurs between July and October with 95% of the
annual sediment load derived froln the monsoon period. The erosion rate is
highest in small sub-basins of the Krishna dver where rates are up to
4000 t1km2/yr (Subramanian, 1982).
226
APPENDIX 4
tough and easily erodible siltstone or sandstone with nlore granitic rock above
Caonillas reservoir.
Loiza reservoir
The region has an average rainfall of 1900 mmJyr (841 mrn run-off) and
teluperature of 25C. The dam impounds 534 kn12 of the Loiza catchment of
which over half the area has slopes greater than 35. The sediment yield of the
region is high, between 1000 tfKm2/yr and 2000 ttlCIl12/yr (Monis and Fan,
1997).
Qs
= 13959*A
1,213
(r
=0946)
Qs
E: == erosion index
S == slope of telTain
P r = annual average precipitation (mm)
227
EVACUATION OF SEDIMENTS
Sf = 617 AO- 67
(r
=077, n =20)
Group 2
150-1000 t/km 2/yr. This group contains over 500/0 of the reservoirs.
SY =202 A I -07
(r
=092, n =33)
Group 3
Over 1000 t/km2/yr. This group includes 7 basins which are less than 420 km2 so
that eroded sediment is likely to be transported into the reservoir.
Sf =3137 Ao- s7
(r=O91,n=7)
These relationships describe the sediment yield at the reservoirs, which is not all
the material eroded. The sediment delivery ratio varied from 08% to 67-470/0
with most ratios less than 25%. Those catchments with the lowest specific
sediment yields are not always those with the lowest erosion rates. The surface
area and location of the sediment source, relief, slope, transpo11 system and
vegetation cover all affect the delivery ratio. The predominant factor is the
drainage basin area (Salas et aI., 1997a and 1997b)
228
APPENDIX 4
A4.3.2. Thornthwaite
.il..n alternative climatic classification was put forward by Thornthwaite (see Table
A4.9), based on the relationship between precipitation and potential evapotra...l1spiration.
The classification calculates an index I which expresses the relationship
between surplus moisture, 5, moisture deficiency, d,and potential evapotranspiration, pe.
1= (5 - d)lpe x 100
The zero value separates moist (positive) from dry (negative) climates. To
incorporate a thermal parameter in the classification Thomthwaite used potential
evapotranspiration as this parameter expresses water need as a function of
temperature and length of day (UNESCO, 1990).
A4.3.3. Alisov
A classification by Soviet climatologist Alisov is based on the conditions of
circulation of the atmosphere. He identified seven main climatic zones which can
be seen in Figure A4.8. Each zone is characterised by the predominance of the
air mass corresponding to the name of the zone (UNESCO, 1990).
A4.3.4. Map
of aridity
(PIETP
(PIETP
(PIETP
(PIETP
< 003)
< 02)
< 05)
< 075)
In addition to the four aridity classes, temperature was used to further define
the arid regions. The subdivisions were warm winter, mild winter, cool winter and
229
IV
<
()
(:)
--I
0
Z
-n
l/)
0
3:
--I
l/)
A. T,op"=-,
,ainy clim.',"
8. Dry
cli~l"
C. Humid
nwtOlMtm.1
cllmal"
O. Humid
mlcrotMfma'
ctifNl"
D
'0
f11M
ILl""""
I.,-t-J~~
~ ImonJOOf\ IYpe)
. f'ollf
clime..,
Figure A4.7.
_1
.J
' ]
APPENDIX4
Table A4.9.
Symbol
Thermal province
Annual pe: em
Index: I
Perhumid
Megathermal
>114
>100
B4
Humid
I
I
80-100
998-114
60-80
Bz
856-998
40-60
B,
713-855
20-40
B3
C2
Meso thermal
Moist sub-humid
Microthennal
0-20
42-8-570
I
Dry sub-humid
286-427
(-33)-0
Semi-arid
Tundra
143-285
(-66)-(-33)
Add
Forest
0-142
(-100)-(-66)
60
60
120
120-"
180
5
60~----~----~~~-------4------~~~------r---'------+--------~
60
1. Equatorial
2. Subequatorial
3. Tropical
4. Subtropical
Figure A4.B.
60
120
180
120
5. Temporate
6. Subpolai
7. Polar
231
EVACUATION OF SEDIMENTS
cold winter. Consideration was also given to the time of the rainy periods and to
the length of the dry peliod (UNESCO, 1990).
A4.3.5. Budyko
Budyko (1956) devised a climatic classification based on a 'radiational index of
dryness' as a means of rating the efficiency of the moisture supply. This is
illustrated in Figure A4.9.
The index is given by:
1= RnlLn
232
APPENDIX4
:!
~
0
~
0
:!
0
..
<0
0
'Of
~
0
0
..-."
Cl
CI
~.
:..
0
~
:....
~
--
Co
c..::
c..::
-e
C)"
)...,>
.::;
'';::
.~
'\::
,..~
0\
~
~
::::
CJ:
t;:
233
EVACUATION OF SEDIMENTS
Table A4.10.
Climatic index
Vegetation type
Run-off coefficient
< 1/3
Tundra
07
1/3-1
Forest
03-07
1-2
Steppe
01-03
2-3
Semi-deselt
<01
>3
Deselt
<01
30~----~--------~~~~-----------
25
20
15
Warm temperature
Cold temperature
-5
-10
-15L-----~~-----L------~------~----~------~------~------~----~
50
100
150
200
250
300
350
Figure A4.10.
234
400
450
APPENDIX 4
Rain forests
Abundant ai110unts of rainfalL Olympic Rainforest: located on the west coast of
USA, on the OlYlnpic peninsula in \Vashington State. Warm climate.
Tropical rainforest
Abundant amounts of rainfall. Located near the Equator. Wann climate.
Grasslands
The typical rainfall is 25 cmJyr to 30 crrJyr. The lowest rainfall occurs in a desert
biome. The grass has roots that can penetrate into the soil to obtain some water,
but the amount of water available is not enough to sustain much tree life.
Biomes are explained more fully at:
o
http://sheepshead.usl. edufT......aCEPTIrainforest.ht!pl
The following tables (Tables A4.11-A4: 16) provide, country by country, data on
climatic conditions and rates of sediment yield" Examples from case studies are
included where available.'-The tables.also.contain data on the number and storage
capacity of dams in each country together. with .estimates of the loss of storage
due to sedimentation.
235
Table A4.11.
Country
Afghanistan
BwklBsk
n
Rates of sediment
yield: tlkm 2/yr
(Walling and Webb,
1983)
Example rates:
tlkm 2/yr
(from literature)
No. of
dams
data for
Total
capacity
of dams:
% Li:lP,lL Il Y
lost to
sediments
--I
o
Annual
loss of
storage
M.m}
100-250/250-500
28
8000%
."
00%
Vl
Armenia
DfiDs
CJ
100-250
3:
----
Azerbaijan
DflDs
100-250
Bahrain
Bwh
<50
Z
--I
Vl
-----
Bangladesh
Am/Cwa
1128
500-750/> 1000
Bhutan
Bs
>1000
Brunei
Af
<50
Burma
Am
500-750/750-10001
>1000
Cambodia
Aw
250-500
China
ET/BwkIBs/Cwbl
Cf/CfalCw/Dwl
Dwa/Dwb
50-100/250-5001
500-750/7 50-1 0001
>1000
Df
100-250
Hong Kong
Cwa
250-500
India
Aw I AflAmIAs/Cwa
] 00-2501250-5001
500-750/>] 000
3130/310115
47
784]2
96%
05%
250-500/500-750
1135
17%
04%
616
804012957/1 8001
1620/1400124611571
25 600/2] 7001
16300/2300
28
42804
28}%
]2%
--
TncioI1P"i::J
---,----BwkIBsk
50-100
Iraq
Bwh
50-100125()-500
l~rael
Bwh
50-100
DfC/D IlJ/Cfa
<50/50-tOOI
250-500
Tnm
-50
. - -------- -----
--
--
Japan
729
17322
69%
02%
111%
0-4%
.Jordan
Bwh
50-100
Kazakstnn
DI11/Bsk/Bwk
<50/50-100
ClhlOwa
500-750
Kuwait
Rwh
250-500
Kyrgy:-:tan
Df
100-250
Laos
Aw/Cf
250-500
Lebanon
Bwh
50-100
Malaysia
Af
250-500
Maldives
Csb
Island
Mongolia
Bwk/Bsk
50-100
Cwa
500-750
Cwa
>1000
Om:m
[3wh
50-100
Pakistan
Bw
50-1 OOt:!50-500
.-
--
~~---~
--------
Myanmar
-------~---
Nepal
----~---
~-~
..
2800
~
~---
-~
-u
iJ
2498/454/1179
14300
Pnpu:1 New
Guinea
.
AI'
250-500/500-750
2581/492/11 126
..
-------~
203%
08%
m
Z
o
X
..t:>.
Table A4.11.
continued
~
()
Climatic
classification
(Koppen)
Rates of sediment
yield: tlkm2/yr
(Walling and Webb,
1983)
Philippines
Am
500-750
Qatar
Bwh
50-100
DfclDtbfDfalDfdF-
DwclDwdlET
<50/50=100
Saudi Arabia
Bwh
50-100
Singapore
Af
250-500/500-750
Sri Lanka
Af
100-250
Syria
Bwh
50-100
Taiwan
Cwa
>1000
Country
Example rates:
tlkm2/yr
(from literature)
No. of
dams
data for
% capacity
lost to
sediments
Total
capacity
of dams:
M.m3
Annual
loss of
storage
~
o
z
o
-n
Vl
Russia-
..
-~
~----
----
~241128/39/141B/9/
CJ
3:
. ...
6/5/5
Z
-I
Vl
.-~
31 700
1525
10
......
Tajikistan
Ds
250-500
Thailand
Aw
<501250-500
Turkey
DslDf
100-250/500-750/
>1000
Turkmenistan
Bwk
50-lO0/lO0-250
United Arab
Emirates
Bwh
50-100
Uzbekistan
BwkfBsk
<50/50-100
Vietnam
Aw
250-500
Yemen
Bwh
50-100
........
07%
.......
76-3874
16
9006
..
-------
--------
--------
1083/203
597%
15%
Table A4.12.
Country
Climatic
classification
(Koppen)
Rates of sediment
yield: tllcm 2/yr
(Walling and Webb,
1983)
Example rates:
tlkm1/yr
(from literature)
No. of
darns
data for
Total
capacity
of dams:
M.m)
% capacity
lost to
sediments
Annual
loss of
storage
L7
26328
155%
05%
124%
OO~%
-Algeria
BWh/Csa
<50
Angola
Aw/Cwa
]00-250
Benin
Aw
50-100
Botswana
BSh
100-250
Burkina Faso
Bsh/Aw
50-100
Burundi
AflAw
500-750
Aw/Am
100-250
lslmul
Is1ancl
Aw/Am
<50
BwhlBsh/Aw
<50
Comoros
ls1and
Is1and
Congo
Aw/Af
<50
Congo,
Democratic
RepUblic
Aw
<50
Djibouti
13sh
50-100
Egypt
BWh
<50
--
--
--------Carneroon
Verde
Central Africa
Republic
Chad
--
--
38
'1
1689000
Table A4.12.
Country
continued
Climatic
classification
(Koppen)
Rates of sediment
yield: tlkro2/yr
(Walling and Webb,
1983)
Aw/Am
50-100
ExampJe rates:
tlkm2/yr
(from literature)
No. of
dams
data for
()
TQtal
capacity
of dams:
M.m3
% capacity
lost to
sediments
Annual
loss of
storage
,!
Equatorial Guinea
---.
Eritrea
CwlBsh
50-100
Ethiopia
Bshl
Bwh
5<J-I 00/500-750
Am/Aw
100-250
Gambia
Bsh
50-100
Ghana
Aw
50-100
Guinea
Aw
50-100
Guinea-Bissau
BWh
50-100
Ivory Coast
Aw
50-100
Kenya
Bsh/C
50-100/250-500
Lesotho
Ctb
250-500
Liberia
Am
50-100
Libya
BWh
<50
Madagascar
Af/Aw
250-500/500-750
Malawi
Aw/Cwa
50-100
Mali
Bwh/Bsh/Aw
<50
"Tl
Vl
3:
Z
-I
-----
Vl
---
'-------------
19 520/20117 600/
20-3000/2500-17 600
--------
-----
o
z
o
o
-----
Gabon
?:j
150-0
780%
1-56%
Mauritania
DwhlBsh
Mauritius
Isl.and
<50
Island
~
Morocco
Bsh/Bw
Mozambique
Aw
<501750-1000
100-2501250-5001
10 351-0
17
7-45%
0-42%
4346%
174%
80/17
500-750
Namibia
BwlBsh
50-100
Niger
BWh
<50
Aw
50-100
At'
<50
SnoTome and
Principe
Island
Island
Senegal
Aw/l3sh
50-100
Seychelles
Islands
Nigeria
33
~-~.
Rwanda
-----
Island
--
--.--Sierrn Leone
Am
50-100
Somalia
Bsk
100~-250
South Africa
Csb/Cll)/I3W
<50/50-1001
17/10-1000
100-250/250-5001
5,00-750
----
-----
--
Slldan
BWh/Bsh/Aw
<50
Swaziland
eft)
250-500
55840
TLltlZania
Bs/Aw
250-500/500-750
Aw
50-100
-Togo
94
iJ
-u
rn
Z
o
x
,.tI..
()
o
z
o
."
c.n
m
3:
m
Z
Table A4.12.
Country
Tunisia
-I
continued
c.n
Climatic
classification
(Koppen)
Rates of sediment
yield: tlkm2/yr
(Walling and Webb,
1983)
BWh/Csa
<501750-} 000
Example rates:
tlkm2/yr
(from Ii terature)
No. of
dams
data for
Total
capacity
of dams:
M.m3
% capacity
lost to
sediments
Annual
loss of
storage
15
17760
3421%
0-07%
---
Uganda
Aw
50-100/500-750
Zambia
Cwa
50-100
Zimbabwe
Aw
100-250
-~--
---
.1
'j
Table A4.13.
Country
Country data/or-Australasia
Climatic
classification
(Koppen)
Rates of sediment
yield: t/km2/yr
(Walling and Webb,
1983)
Example rates:
tlkm2/yr
(from literature)
28
Australia
Aw/Bsb/Cfh/CsICsa/
Bs/Bwh
<50/50-100
Fiji
Af
Island
Kiribati
Af
Island
Marshall Islands
Af
Island
Micronesia
Af
Ishmd'
Nauru
AI'
Island
New Zealand
Ctb
250-500/500-7501
>1000
Palau
Af
ISland
Solomon Islands
Af
Island
Tonga
AI'
Island
Tuvnla
Af
Island
Vanuatu
At"
Island
6982119 9701
1'7 340/13 8901
17 070113 3001
12736
No. of
dams
data for
Total
capacity
of dams:
M.m3
% capacity
Annual
lost to
sediments
loss of
storage
202
192%
10%
---Western Samoa
~
w
Af
I----- .
Island
iJ
iJ
ITI
o
x
Table A 4.14.
Climatic
classification
(Koppen)
Example rates:
t/km2/yr
(from literature)
Albania
Ctb
500-750
4150/3590
AndolTa
Ctb
250-500
o
z
o
"m
250-500/500-750
3:
Country
No. of
darns
data for
Total
capacity
of dams:
M.m3
()
Rates of sediment
yield: tlkI1l2/yr
(Walling and Webb,
1983)
% capacity
lost to
sediments
Annual
loss of
storage
Vl
Ctb
Austria
930%
-
Dtb
<50
Belgium
Cfb
<50
Bosnia and
Herzegovina
Cfb
500-750
Bulgaria
Dfa/Ctb
100-250/250-500
Croatia
Ctb
100-250
Cyprus
Csa
50-100
Czech Republic
Ctb
<50/100-250
Denmark
Cfb
<50
Estonia
Dtb
<50
Finland
DfclDtb
<50
France
Ctb
<50/50-100/
100-250/250-500
GelTOany
Ctb
<50/100-250
Greece
CsalCfb
500-750
Belarus
279%
-l
Vl
119%
39%
236
30%
02%
-,-~-~
"J
"\
!
'I
~--
--
111
Hungary
Ctb
100-250
Icchmd
efelET
<50
Ireland
cn)
50-100
Italy
Cfa/Csa
50-1001100-250/
250-500
Latvia
Dtb
<50
Liechenstcin
Cfb
100-250
Lithuania
Dfb
<SO
Luxembourg
Cfb
<50
Macedonia
cnl
500-750
Malta
Csa
250-500
Moldova
Of
<50
Monaco
Cfb
250-500
Netherlands
Ctb
<50
Norway
orc
50-100
Poland
Otb
<50
597
95%
04%
Portugal
Ctb
100-250
440
14%
01%
Romania
CfblDtb
18
207
324%
32%
San Marino
Ctb
250-500
--_._--
45701214
83
"'0
.--
Serhia and
Montenegro
------
Ctb
500-750
"'0
rn
><
..r:...
~
()
C
o
z
o
II
(f)
Table A4.14.
continued
Climatic
classification
(Koppen)
Rates of sediment
yield: tlkm2/yr
(Walling and Webb,
1983)
Slovakia
Ctb
100-250
Slovenia
Cfb
500-750
Spain
Cfb
100-250
Sweden
DfclDfb
<50
Switzerland
Cfb
50-100
Ukraine
Dfb
<50
United Kingdom
Cfb
<50/50-100
Vatican City
Cfb
250-500
Country
Example rates:
tlkm2/yr
(from literature)
84-2703
No. of
dams
data for
Total
capacity
of dams:
M.m3
% capacity
lost to
sediments
Annual
loss of
storage
53
28-4%
0-9%
91
23323
3-7%
0-1%
--
71-4%
8-6%
-----
95
-)
0-1%
3:
m
Z
-I
(f)
Country
Climatic
c1nssi1ication
(Koppen)
Rates of sediment
Antiglla amI
Barhuda
At"
1.00-250
Bahamas
Af
100-250
Barbados
At'
100-250
Belize
Af
lOO-250
Cl.1nadn
ETICfc/C1b/Dfcl
Dlb
yield: tlkm2/yr
Example rates:
t/km2/yr
(from literature)
No. of
dams
data for
Total
capacity
of dams:
M.m3
_..__.
91/55/4/0-350
>1000
Costa Rica
Aw
100-250
Cuba
Af
100-250
Af
100-250
Af
100-250
EI Salvador
Aw
100-250
Greenland
Hf
50-100
Af
100-250
Guatemala
Aw
100-2501250-500
Hniti
Af
100-250
Dominica
Dominican
Republic
-------.---
Grenada
--
----
% capacity
lost to
sediments
Annual
loss of
storage
()
Table A4.15.
continued
Climatic
classification
(Koppen)
Rates of sediment
yield: tlkm 2/yr
(Walling and Webb,
1983)
Honduras
Aw
]00--250
Jamaica
Af
100-250
Mexico
Bw/Cw/Af/Aw
50-100/100-2501
250-5001
500-750
Nicaragua
Aw
100-250
Panama
Af
100-250
Af
100-250
Saint Lucia
Af
100-250
Af
]00-250
Trinidad and
Tobago
Af
100-250
USA
DfblDfaiB w/Cfal
Country
Example rates:
tlkm2/yr
(from literature)
No. of
dams
data for
Total
capacity
of dams:
M.m 3
% capacity
lost to
sediments
Annual
loss of
storage
3:
-I
(/)
211
.-.. ---..
(/)
--------_.
CsblBs/Df
o
z
o
"
m
407212374122921
1167/500114511 071
71150112
1105
--~-----
109980
39%
02%
Table A4.16.
Count}")1
Climatic
classification
Country
(Koppen)
Rates of sediment
yield: tlkm 2/yr
(Walljng and Webb,
1983)
Example rates:
tlkm 2/yr
(from literature)
130/33
Argentina
Bw/Cfa/Bs
< 501250-500/
750-1000/> 1000
l10livia
AwlBsc
100-250/250-500/
> 1000
<50/100-250
No. of
dams
data for
Total
capacity
of dams:
% capacity
lost to
sediments
Annual
lo~s of
storage
M.m3
]46/9
3829
0-8%
--0-1%
192%
--96%
100-2501>1000
----~--~----.
Colombia
------~---
50-100/> 1000
Af/Aw
..--.-
917
-,
Ecuador
Aw/Af
100-250/> 1000
French Guiana
A1'
<50
Guyana
AI'
<50
Paraguay
---
Cw..uAw
<50
Peru
Af/AwlBsc
100-250/250-5001
>1000
Af
<50
2000
----~---
Suriname
---UllIgtluy
Venezuela
~
~
Cra
Aw/Af
<50
<50/100-250
212
-0
-0
o
x
Index
Africa
country data 239-242
erosion rates 221-224
hydropower 27
irrigation 27
population 26-27
Alisov classification 229, 231
Americas
country data 247-249
hydropower 24-26
irrigation 24-26
population 24-26
annual precipitations 106, 213-214,213
Aracay River, Venezuela 202
areas
erosion rate 95-100
suited to flushing 93-124
aridity maps 229,232
Asia
country data 236-238
hydropower 27-28
irrigation 27-28
population 27
Atbara River, Sudan 195
Athi-Kapiti plains 223
Australasia 243
Austria 68,72-74, 76
case study 176-178
downstream impacts 178
flushing 177-178
sedimentation 177
autumn 1998 precipitation 108,111
251
EVACUATION OF SEDIMENTS
205-207
density current venting 206
downstream impacts 176,182-186, 189,
197,207
erosion rates 224-225
flushing 175-176, 181-189, 196-197,
206
lateral erosion 185
Liuhe River 175
sedimentation 175, 180-186, 196,
205-206
storage capacity 188
Yellow River 186, 225
Yeyu River 184
Yili River 182
Yongding River 179-180
climate 211-215
classification 229-235
combining homogenous groups 117
erosion rate 93, 115-120
Tarbela Dam, Pakistan 164-165
climatic zones 12
country classification 118-120
earth 231
erosion rate 115-120
river basins 118-120, 118
sediment yield 118, 118
world 101-120
cold steppe climates 115
concentration inflow 56
252
INDEX
India 225-226
Kenyan grazing lands 223-224
Krishna River, India 226
land use effect 93, 121
Pakistan 226
precipitation effect 121
Puerto Rico 226-227
slope effect 121-122
soil effect 121-122
Spain 228
Switzerland 227
Taiwan 96
Thailand 224
topography 93
Turkey 227-228
vegetation effect 93, 121-122
worldwide 93
Europe
_
country data 244-246
hydropower 24
irrigation 24
population 23-24
evaluation
flushing criteria 50
45-46
253
EVACUATION OF SEDIMENTS
254
204
INDEX
sedimentation 196
high annual precipitation 10 1
high erosion rate areas 95-99
Himalayas, erosion rates 226
historic growth .
hydropower 21-22,23
reservoir 30
homogenous climatic groups I 17
human impacts
erosion control 220-221
erosion rate 93, 121
humid climates 115
hydraulic conditions, flushing 7-8, 58-59
hydraulic modelling 13-14, 128-129
hydrology
catchment 81-82, 88, 123
characteristics 12, 123
investigations 13, 127-128
sedimentology 9
Tarbela Dam, Pakistan 164-165
hydropower 3
Africa 27
Americas 24-26
Asia and Oceania 27:-:-28
Europe 24
historic growth 21-22, 23
potential 22-23
Iehari, India
concentration 56
hydro graph 165
Tarbela Dam, Pakistan 165
insufficient drawdown 41
insufficient flushing flows 51-52
investigations
sediment 128
site-specific 13
Iran 68, 72-73, 75, 78-79
bank failure 193
case study 191-195
channel erosion 193
diversion channels 194
downstream impacts 195
flushing 192-195
lateral erosion 194
long-term predictions 195
sedimentation 191-192
sheet erosion 193
irrigation 3, 21
Africa 27
Americas 24-26
Asia and Oceania 27
Europe 24
189
land use
erosion control 219-220
erosion rate effect 93, 121
lateral erosion 185, 194
Liuhe River, China 175
255
EVACUATION OF SEDIMENTS
256
Oceania
hydropower 27-28
irrigation 27-28
population 27
operation duration, flushing 45
operation policies, Pakistan 166
operational considerations 85-86, 88
operational limitations 10
optimum locations, flushing 12-13
Orange River, South Africa 224
organic content, soil 217
orientation, catchments 217-218
Ouchi-Kurgan, fOlmer USSR 68,72-74, 78
case study 190-191
downstream impacts 191
flushing 191
sedimentation 190
50 percentile size
river-bed matelial 55
transported sediment 55
INDEX
257
EVACUATION OF SEDII'1ENTS
258
South America
country data 249
hydropower 25-26
irrigation 25-26
population 25
Spain, erosion rates 228
spring 1998, precipitation 108, 109
SSR see sediment size ratio
steppe climates 115
storage 3-4
capacity 9, 83, 188
construction vs demand 30
demand diSllibution 4-7, 23-28
distribution 31-34
gross requirements 35-36
increase distribution 30
loss rate 7,31-36
lost to sedimentation 33
requirements 7, 35-36
trends 34
volume disllibution 19
volume predictions 168-169,168
world demand 19-23
world total 17-18
worldwide distribution 18-19
structures, soil 217
Sudan 68, 72-75, 79
Atbara River 195
case study 195-196
downstream impacts 196
flushing 195-196
sedimentation 195
summer 1998, precipitation 108,110
suspended-sediment yields 214-215,223
world maximum 96
sustainable reservoir capacity 47-50
Switzerland 68, 72-77, 79
case study 178-179,200-202
downstream impacts 179, 201-202
erosion rates 227
flushing 179, 200-201
Masse River 200-201
Melezza River 178
sedimentation 178-179,200
system simulation modelling 14, 129
INDEX
flushing 172-173
sedimentation 172
USSR see former USSR
value, flushing 10
vegetation
climate classification 232, 235
erosion control 219
erosion rate effect 93, 121-122
Venezuela 68,72-75, 79-80
Aracay River 202
case study 202-205
downstream impacts 205
flushing 202-205
sedimentation 203
volcanic activity 216
221-223
Yellow River, China 186, 225
Yeyu River, China 184
yields, sediment 93-105, 213
YiIi River, China 182
259
EVACUATION OF SEDIMENTS
179-180
260
68, 72-74, 76
72-74, 76
68,