Springer Tracts in Civil Engineering

Hui Liu Editor

Flood Prevention
and Drought
Relief in Mekong
River Basin
Springer Tracts in Civil Engineering

Series Editors
Giovanni Solari, Wind Engineering and Structural Dynamics Research Group,
University of Genoa, Genova, Italy
Sheng-Hong Chen, School of Water Resources and Hydropower Engineering,
Wuhan University, Wuhan, China
Marco di Prisco, Politecnico di Milano, Milano, Italy
Ioannis Vayas, Institute of Steel Structures, National Technical University of
Athens, Athens, Greece
Springer Tracts in Civil Engineering (STCE) publishes the latest developments
in Civil Engineering - quickly, informally and in top quality. The series scope
includes monographs, professional books, graduate textbooks and edited volumes,
as well as outstanding PhD theses. Its goal is to cover all the main branches of civil
engineering, both theoretical and applied, including:
• Construction and Structural Mechanics
• Building Materials
• Concrete, Steel and Timber Structures
• Geotechnical Engineering
• Earthquake Engineering
• Coastal Engineering; Ocean and Offshore Engineering
• Hydraulics, Hydrology and Water Resources Engineering
• Environmental Engineering and Sustainability
• Structural Health and Monitoring
• Surveying and Geographical Information Systems
• Heating, Ventilation and Air Conditioning (HVAC)
• Transportation and Traffic
• Risk Analysis
• Safety and Security
Indexed by Scopus
To submit a proposal or request further information, please contact: Pierpaolo Riva
at [email protected] (Europe and Americas) Mengchu Huang at
[email protected] (China)

More information about this series at http://www.springer.com/series/15088

Hui Liu
Editor

Flood Prevention
and Drought Relief
in Mekong River Basin

123
Editor
Hui Liu
China Institute of Water Resources
and Hydropower Research
Beijing, China

ISSN 2366-259X ISSN 2366-2603 (electronic)

Springer Tracts in Civil Engineering
ISBN 978-981-15-2005-1 ISBN 978-981-15-2006-8 (eBook)
https://doi.org/10.1007/978-981-15-2006-8
Jointly published with China Water & Power Press
The print edition is not for sale in China. Customers from China please order the print book from: China
Water & Power Press.

© China Water & Power Press and Springer Nature Singapore Pte Ltd. 2020
This work is subject to copyright. All rights are reserved by the Publishers, whether the whole or part
of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations,
recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission
or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar
methodology now known or hereafter developed.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this
publication does not imply, even in the absence of a specific statement, that such names are exempt from
the relevant protective laws and regulations and therefore free for general use.
The publishers, the authors, and the editors are safe to assume that the advice and information in this
book are believed to be true and accurate at the date of publication. Neither the publishers nor the
authors or the editors give a warranty, express or implied, with respect to the material contained herein or
for any errors or omissions that may have been made. The publishers remain neutral with regard to
jurisdictional claims in published maps and institutional affiliations.

This Springer imprint is published by the registered company Springer Nature Singapore Pte Ltd.
The registered company address is: 152 Beach Road, #21-01/04 Gateway East, Singapore 189721,
Singapore
Preface

This book was conceived as a product of the Lancang-Mekong Cooperation

mechanism. The LMC Fund supports work in flood and drought mitigation in the
Lancang-Mekong Basin, which is the common concern of the six member coun-
tries. This book draws together relative resources and materials on the flood and
drought topic of this region, as well as analysis and modelling, so that all member
countries and a wider group of people could benefit from them.
The Lancang-Mekong River nurtures the long history and the splendid culture
along with it, and is a common precious resource of countries in the basin.
Attaching great importance to the integrated management, rational development,
and effective protection of water resources, Lancang-Mekong countries are com-
mitted to propping up the sustainable development of society and economy
through the sustainable utilization of water resources. In March 2016, the First
Lancang-Mekong Cooperation (hereinafter referred to as “LMC”) Leaders’ Meeting
was held, marking the formal establishment of the Lancang-Mekong cooperation
mechanism, on which the Sanya Declaration that stresses water resources coop-
eration should be a prior field and flagship field was published. In the list of
early-harvest projects of LMC, “Strengthening coordinated flood and drought
management in the Lancang-Mekong basin” the Vietnamese side proposed was
among the two early-harvest projects in the water resources field, and it showcases
the six Lancang-Mekong countries are commonly concerned with and attach
importance to flood and drought prevention and mitigation. People of all countries
in the region will get more benefits if relevant countries step up communication and
coordination and carry out practical cooperation under the LMC mechanism.
Affected by tropical monsoons, typhoons, climate change, and El Nino, Mekong
River Basin countries are often subject to flood or drought disasters to a varying
degree. Effectively preventing and coping with flood and drought disasters is a
problem with socio-economic development and human well-being and also a
common challenge facing all countries. The Chinese side recognizes that carrying
out a joint assessment of flood and drought management together with all basin
countries is an important measure for commonly understanding the flood and
drought situation in the Mekong River Basin, enhance mutual trust and remove

v
vi Preface

misgivings, cope with flood and drought risks jointly, and build neighborly inter-
national relations. In view of this, the Chinese side takes the lead to apply for the
“Joint Assessment on the Current Status of Flood and Drought Management in the
Mekong River Basin (Phase I)” project as sponsored by the LMC fund. This project
is of great significance for bringing benefit to people of Lancang-Mekong countries
and promoting regional sustained stability and shared prosperity and development
because we will carry out communication and cooperation in water resources field,
especially in disaster prevention and mitigation, implement the spirit of the 2nd
LMC Leaders’ Meeting and the consensus of the Joint Working Group of
Lancang-Mekong Water Resources Cooperation, promote the implementation of
LMC early-harvest projects proactively, deepen the understanding of the charac-
teristics and current status of flood and drought disasters in the Mekong River
Basin, share disaster prevention and mitigation technology, and experience and join
hands to cope with flood and drought risks and challenges.
During the First Meeting of the Lancang-Mekong Water Resources Cooperation
Joint Working Group (hereinafter referred to as “LMC Water”) in February 2017,
the Chinese side released to representatives of all countries present at the meeting
information about project design, and the project proposal won support from all
representatives of countries. In January 2018, the 2nd LMC Leaders’ Meeting
published the Five-Year Plan of Action on Lancang-Mekong Cooperation (2018–
2022), stating the joint assessment of flood and drought management in the Mekong
River Basin would be implemented. On April 5, 2018, Mr. Yu Xingjun, Leader
of the LMC Water China, sent to the other five countries’ leaders of LMC Water a
letter giving an introduction to the joint research content and inviting experts to
found a joint assessment expert group; in early August, the Chinese expert group
carried out, at the invitation of Thailand’s leading unit (Ministry of Natural
Resources and Environment) and Viet Nam’s leading unit (Viet Nam National
Mekong Committee), a seven-day field investigation in Thailand and Viet Nam,
with in-depth understanding of structural measures and non-structural measures for
flood and drought management in Mekong River Basin countries; in mid-August,
experts of five Mekong River Basin countries came to China to have a week’s
technical exchange on current status and technical need of floods and droughts in
the Mekong River Basin; in mid-October, Chinese experts formulated the first draft
of the joint assessment report and submitted it to the experts of the member
countries for discussion and further improvement, and invited them to have tech-
nical exchange in China in 22–24 October.
With guidance of the LMC Water, and joint efforts of experts from all LMC
member countries, the joint assessment expert group finally formed a report with
general consent of the LMC member countries, and ready for sharing and publi-
cation in early 2019.

Beijing, China Hui Liu

Acknowledgements

This book is a team effort at its best. The work was supported by the Joint Working
Group on Lancang-Mekong Water Resources Cooperation under the Lancang-
Mekong Cooperation mechanism. We are more grateful than we can say for con-
tributions of colleagues from Lancang-Mekong Cooperation member countries and
its line agencies, who provided relentless help and valuable suggestions. We
especially owe thanks to Hlaing Tun, Thongthip Chandalasan, Winai Wangpimool,
Chea Monyvycet, Nguyen Dinh Dat, Toe Toe Aung, Aung Than Oo, Somphone
Khamphanh, Phaylin Bouakeo, Prasith Deemanivong, Wasna Roi-Amphaeng, and
Supapap Patsinghasanee.
The production of this book would not have been possible without the support of
China Water & Power Press. We would like to thank Ms. Lijuan Xu for copyediting
and coordinating the publication processes.

vii
Disclaimer

This book was prepared by the China Institute of Water Resources and Hydropower
Research (IWHR) and Tsinghua University. Although all effort is made to ensure
that the information, views, and analyses contained in this document are based on
sources believed to be reliable, no representation, expressed or implied, is made
guaranteeing accuracy, completeness or correctness. The opinions contained herein
reflect the judgment of the authors and are subject to change without notice. Neither
the IWHR nor the Tsinghua University, the Lancang–Mekong Cooperation
Member Countries and any agency thereof, their employees, contractors, subcon-
tractors or their employees, assumes any legal liability or responsibility for the
consequences of any third party’s use of the information, opinions, and analysis
contained in this document.

ix
Contents

1 Overview of the Mekong River Basin . . . . . . . . . . . . . . . . . . . . . . . . 1

Baiyinbaoligao, Hui Liu, Xingru Chen, and Xiangpeng Mu
2 Summary of Flood and Drought in Mekong River Basin . . . . . . . . . 27
Xingru Chen, Hui Liu, Baiyinbaoligao, and Xiangpeng Mu
3 Analysis of Flood Character in the Mekong River Basin . . . . . . . . . 55
Shiyu Hou, Hui Liu, and Fuqiang Tian
4 Analysis of Drought Character in the Mekong River Basin . . . . . . . 95
Xuejun Zhang and Hui Liu
5 Overview of Measures and Assessment of Capacity for Flood
Prevention and Drought Relief . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109
Zhixiong Ding, Xuejun Zhang, Song Han, and Hui Liu
6 Main Findings and Recommendations . . . . . . . . . . . . . . . . . . . . . . . 167
Hui Liu and Xiangpeng Mu

Appendix: Introduction of THREW Model. . . . . . . . . . . . . . . . . . . . . . . . 173

Fuqiang Tian and Shiyu Hou

xi
Chapter 1
Overview of the Mekong River Basin

Baiyinbaoligao, Hui Liu, Xingru Chen, and Xiangpeng Mu

Abstract The Lancang-Mekong River, the largest transboundary river in Southeast

Asia, flows through six riparian countries, nurtures splendid culture along it. The
river is called Lancang River in China, and Mekong River out of China. The Mekong
River region is featured by various landscape patterns and high variability of rainfall
caused by monsoon cycles. This region is also featured by high density of population,
high potential of development and high international attention, which make this study
more meaningful. The physical geographical feature of this region is illustrated from
the aspects of topography, meteorology and hydrology in this chapter. The Socio-
economy characteristics of this region is also reviewed in this chapter to provide
background knowledge for the further analysis of flood and drought management.
The irrigation projects and hydropower projects, main types of structural measures,
are also briefly introduced from the whole basin view.

1.1 The Lancang-Mekong River Basin

The water resources challenges of the Southeastern Asia regions is exceptionally

complex, with growing populations and economies needing reliable and predictable
water resources for livelihoods and food and energy production. The region is char-
acterized by monsoon cycles leading to exceptionally high rainfall and runoff vari-
ability, and extremes of severe flooding and drought, overlain with uncertainties of

Baiyinbaoligao (B) · H. Liu · X. Chen · X. Mu

China Institute of Water Resources and Hydropower Research, A-1 Fuxing Road, Haidian
District, Beijing 100038, China
e-mail: [email protected]
H. Liu
e-mail: [email protected]
X. Chen
e-mail: [email protected]
X. Mu
e-mail: [email protected]
© China Water & Power Press and Springer Nature Singapore Pte Ltd. 2020 1
H. Liu (ed.), Flood Prevention and Drought Relief in Mekong River Basin,
Springer Tracts in Civil Engineering,
https://doi.org/10.1007/978-981-15-2006-8_1
2 Baiyinbaoligao et al.

a changing climate. Many great rivers rise from the Himalayas and are shared by
more than one nation, with almost all of the region’s mainland nations dependent to
a greater or lesser extent on these transboundary waters. Most of the flow of these
rivers comes from midstream precipitation, with glacier melt providing less than
20% of these rivers’ overall flow, except in the Indus basin (c. 50%).1
Lancang River originates from Yushu Tibetan Autonomous Prefecture, Qinghai
Province, China, and is known as Mekong River2 after flowing out of China from
Xishuangbanna Dai Autonomous Prefecture, Yunnan Province. Mekong River flows
through 5 countries including Myanmar, Lao PDR, Thailand, Cambodia and Viet
Nam into South China Sea from the west of Ho Chi Minh City, Viet Nam, as shown
in Fig. 1.1.
The Lancang-Mekong River runs as long as 4880 km, with a drainage area of
795,000 km23 and an average annual runoff volume of 475 km3 . It is the tenth longest
river and ranks eleventh in terms of annual water yield in the world, and is amongst the
seven biggest rivers in Southeast Asia. In specific, Mekong River registers a drainage
area of 630,600 km2 , a length of 2750 km and an average annual runoff volume of
410.9 km3 , taking up 79%, 56% and 86.5% respectively of Lancang-Mekong River.
From the source in Qinghai-Tibet Plateau to the estuary, the river flows through a
slope of 5060 m, with an average gradient of 1.04‰.4
Distribution of area and river length of the Lancang-Mekong River Basin by
country is demonstrated in Table 1.1.

1.1.1 Lancang River Basin

About half (57%) of the total length of the Lancang-Mekong River of 4884 km is
located in the territory of China. Lancang River basin is mainly steep alpine valley,
located in the under-developed region with extremely inconvenient transportation
and deficient natural resources, except extraordinary rich hydropower resources.
Water utilization rate is about 3% in this region, and the water consumed is less
than 1% of the total runoff of Lancang-Mekong basin. This zone contributes about

1 Tian Fuqiang and Liu Hui. China-shared rivers, shared futures. Vientiane Times, pp 15–16. July
14, 2016.
2 In most MRC publications, it is usually called as Lower Mekong River. To keep insistent with the

former joint assessment carried out by MWR, China and MRCS in 2016, we use Mekong River
instead of Lower Mekong River. As to the Upper, Middle and Lower reaches of Mekong River,
Nong Khai and Kratie hydrological station is choosen as the division point for the three sections of
the mainstream Mekong River.
3 The total Lancang-Mekong Basin area of 795,000 km2 is used in MRC publications (e.g. Overview

of the Hydrology of the Mekong Basin), however, China suggests the total Lancang-Mekong Basin
area of 812,400 km2 .
4 Mekong River Commission and Ministry of Water Resources of the People’s Republic of China

(2016). Technical Report—Joint Observation and Evaluation of the Emergency Water Supplement
from China to the Mekong River. Mekong River Commission, Vientiane, Lao PDR.
1 Overview of the Mekong River Basin 3

Fig. 1.1 The sub-basins of the Mekong River Basin

4 Baiyinbaoligao et al.

Table 1.1 Distribution of the area and river length of Lancang-Mekong River Basin by Country
[Tang Haixing, Water resources in the Lancang-Mekong River Basin and analysis on the present
situation of its utilization. Yunnan Geographic Environment Research, 1999, 11(1), 16–2 (in
Chinese)]
China Myanmar Lao Thailand Cambodia Viet Total
PDR Nam
Drainage 16.5 2.4 20.2 18.4 15.5 6.5 79.5
area/104 km2
% of total 21 3 25 23 20 8 100
drainage area
% of national 1.7 3.6 85.2 35.9 85.6 19.7 –
territorial area
Length 2130 777 502 230 3639
(inland)/km
Length 31 234 976 – – 1241
(boundary)/km

13.5% runoff of Lancang-Mekong River. The runoff comes from rainfall, snowmelt
and groundwater seeping. This region has a distinguishing rainy season and dry
season, the runoff of rainy season accounts for 75% of the annual runoff of Lancang
River. The dry season lasts from November to April, during which the runoff mainly
depends on snowmelt and groundwater seeping.
The construction of cascade reservoirs in the mainstream in the middle and lower
reaches of the Lancang River has been completed. Xiaowan Reservoir and Nuozhadu
Reservoir have especially the multi-year regulating capacity, with regulating storage
of 21.2 billion m3 in total. By operating and regulating scientifically, Lancang River
cascade reservoirs are capable to balance the water discharge/volume between the
wet season and dry season, benefiting the Mekong River on the aspects of flood
control, irrigation, navigation and so on.5

1.1.2 Mekong River Basin

Locations and area of the primary sub-tributaries in the Mekong River Basin are
shown in Fig. 1.1 and Table 1.2.6 Main tributaries include Nam Tha, Nam Ou, Nam
Kam, Nam Ngum, Nam Cadinh, Se Bang Fai, Se Bang Hieng, Nam Mun, Nam Sang
and Tonle Sap River, and the Nam Mun is the biggest tributary.

5 Mekong River Commission and Ministry of Water Resources of the People’s Republic of China
(2016). Technical Report—Joint Observation and Evaluation of the Emergency Water Supplement
from China to the Mekong River. Mekong River Commission, Vientiane, Lao PDR, p 2.
6 The first-class tributaries are drawn with reference to the MRC publications. Table 1.2 is the

corresponding statistical area of the first class tributaries. The name of the tributaries is determined
by that of the rivers converge on the Mekong mainstream.
1 Overview of the Mekong River Basin 5

Table 1.2 Sub-basin area of Mekong River Basin [Mekong R C. Planning Atlas of the Lower
Mekong River Basin. Phnom Penh: Lao Uniprint Press Co. Ltd., 2011, p 9 (Though most of the
primary tributaries were referred to this publication, the author did not split larger tributaries like
Tonle Sap into sub-basins as the publication did)]
Name Area (km2 ) Name Area (km2 ) Name Area (km2 )
Ban Khai San 778 Nam Khan 7490 Nam Thong 455
Ban Nam Song 138 Nam Khop 1521 Nam Ton 587
Delta 48235 Nam Loei 4012 O Talas 1448
Doi Luang Pae 688 Nam Ma 1141 Phu Luong Yot 491
Muang Huai Dua
Huai Khok 538 Nam Mae Ing 7267 Phu Pa Huak 132
Huai Bang Bot 2402 Nam Mae 4079 Prek Chhlong 5957
Kham
Huai Bang Koi 3313 Nam Mae Kok 10701 Prek Kamp 1142
Huai Ma Hiao 990 Nam Mae 485 Pre Krieng 3332
Ngao
Huai Nam 1755 Nam Mang 1836 Prek Mun 476
Huai
Huai Sophay 186 Nam Mang 944 Prek Preah 2400
Ngai
Huaag Hua 626 Nam Mi 1032 Prek Te 4364
Huai Bang 938 Nam Mun 70574 Prek Thnot 6124
Haak
Huai Bang I 1496 Nam Nago 1008 Se Bang Fai 10407
Huai Bang 695 Nam Ngam 489 Se Bang Hieng 19958
Lieng
Huai Bang Sai 1367 Nam Ngaou 1495 Se Bang 3048
Nouan
Huai Ho 691 Nam Ngeun 1819 Se Done 7229
Huai 3762 Nam Ngum 16906 Se Kong 28815
Khamouan
Huai Luang 4090 Nam Nhah 316 Se San 18888
Huai Mong 2700 Nam Kadun 456 Nam Sing 2681
Huai Muk 792 Nam Kai 602 Nam 13123
Songkhram
Huai Nam Som 1072 Nam Kam 3495 Nam Suai 1247
Huai Som Pak 2516 Nam Keung 633 NamSuong 6578
Huai Thuai 739 Nam Nhiep 4577 Nam Tam 1548
Huai Tomo 2611 Nam Nuao 2287 Nam Tha 8918
Muang Liep 488 Nam Ou 26033 Nam Thon 838
Nam Beng 2131 Nam Pho 2855 Siem Bok 8851
(continued)
6 Baiyinbaoligao et al.

Table 1.2 (continued)

Name Area (km2 ) Name Area (km2 ) Name Area (km2 )
Nam Cadinh 14822 Nam Phone 664 Sre Pok 30942
Nam Chi 49133 Nam Phoul 2095 Tonle Repon 2379
Nam Heung 4901 Nam Phuong 4139 Tonle Sap Lake 85850
Nam Hinboun 2529 Nam Sane 2226
Nam Houng 2872 Nam Sang 1290

(1) Chinese border—Chiang Saen, Thailand

From the Chinese border to Chiang Saen, the Mekong river section is mainly the
boundary river between Lao PDR and Myanmar, and between Lao PDR and Thailand.
Mainly made up of canyon rivers, the river reach gets its runoff largely from the
Lancang River in China and several small tributaries in Thailand, Lao PDR and
Myanmar. Tributaries on the right bank of the Mekong River within the territory of
Myanmar are the Nam Loi, Nam Yang, Nam Leung and Nam Mae Kham, involving
a drainage area of 28,600 km2 and an average annual runoff volume of 17.63 km3 .
Tributaries on the left bank of the Mekong River within the territory of Lao PDR
include the Nam Ma, Nam Pho and Nam Nuao. Part of the Nam Mae Kham is located
on the right bank of the Mekong River within the territory of Thailand.
(2) Chiang Saen-Vientiane
From Chiang Saen to Vientiane, the mainstream Mekong River begins to be wider
with ladder-like floodplains and paddy and other crop fields. On Thailand’s side in
Chiang Saen are two big tributaries, the Nam Mae Kok and the Mae Nam Ing, and
such big tributaries as the Nam Tha, Nam Ou, Nam Soung and Nam Khan, which
originate from the Annam Mountainous Region, are located near Luang Prabang in
Lao PDR.
(3) Vientiane-south frontier of Lao PDR
On the right bank in Thailand are the Huai Lang River, Huai Luang River, Nam
Songkhram, Nam Mum and other tributaries; and tributaries originating from the
Annam Mountainous Region which are located on the left bank in Lao PDR include
the Nam Ngum, Nam Cadinh, Nam Hinboun, Se Bang Fai, Se Bang Hieng and Se
Done.
(4) Within the territory of Cambodia
Tributaries on the right bank within the territory of Cambodia mainly include the
Tonle Sap River and the Tonle Repon River. In specific, the Tonle Sap River has a
catchment area of about 85,850 km2 (including lake area), and at its upstream is the
Tonle Sap Lake that not only stores and retains floodwater from upper mountainous
regions but also stores and retains a large amount of floodwater of the Mekong River
in the wet season and replenishes it back to the Mekong River during the dry season.
1 Overview of the Mekong River Basin 7

On the left bank are three big tributaries, Se Kong, Se San and Sre Pok, which
originate from the Annam Mountainous Region in Lao PDR and Viet Nam. Their
catchment area amounts to 78,645 km2 , taking up roughly 13% of the drainage area
of the Mekong River, and their water volume makes up 23% of the average runoff
of the basin.
(5) Delta region

Also known as the Cuu Long River Delta, the Mekong Delta is situated at the south-
ernmost point of Viet Nam and the southeastern part of Cambodia. With an area of
48,235 km2 , it is the largest plain in Southeast Asia. After flowing through Phnom
Penh, the Mekong River gets divided into two rivers known as Tien Giang and Hau
Giang within the territory of Viet Nam and then, divided into six rivers. Because of
separation by sandbars, there are nine estuaries leading to the sea.

1.2 Physical Geography

1.2.1 Topography

The upper Mekong River covers the east of Shan State of Myanmar, the north of Lao
PDR and the north of Thailand. Mostly covered with hills and mountains, the east
of Shan State, Myanmar is a hilly and mountainous area. It is the central zone of
Golden Triangle. The north of Lao PDR is covered with plateaus and low mountains
including Xiangkhoang Plateau, Tran Ninh Plateau, Sipsongchutai Range, Luang
Prabang Range and Phetchabun Range. Mountains, hills and plateaus take up 85%
of its area and inter-mountainous plains, flat lands and basins only take up 15%.
Its altitude is between 1000 and 2000 m. The north of Thailand borders Luang
Prabang Mountainous Area to the east. Dean Lao Mountainous Region is located in
the northwest, and middle part is a broad plain. The average altitude is about 400 m.
This area is also a part of Golden Triangle.
The middle reaches of Mekong River include the central and southern parts of
Lao PDR and the northeastern parts of Thailand. The topography is complicated,
including Phou Luang Range, Khammouane Plateau, Bolovens Plateau, Vientiane
Plain, Thakhek Valley, Savannakhet Plain and Champasak Lowland in central and
southern Lao PDR. In northeastern Thailand are Khorat Plateau, Phu Phan Moun-
tains, Buri Ram Hills and Ubon Ratchathani Steppe. In addition to plateaus and low
mountains, the left bank of Mekong River is composed of plains and lowlands.
The lower Mekong River covers Cambodia and most parts of southern Viet Nam,
with high terrains in three directions and low terrains in the middle part. Eastern,
northern and western parts are covered with plateaus and mountains. Most Cambo-
dian areas in this region are plains, and central plains take up 46% of the total area of
Cambodia. Located in the northeast of the central plains, the Truong Son Range sep-
arates Lancang-Mekong River from rivers flowing to the South China Sea. Situated
8 Baiyinbaoligao et al.

in the west and the southwest, the Cardamom Mountainous Region is a watershed
between Mekong River and rivers flowing to the Gulf of Siam. The Dangrek Moun-
tains constitute a natural border between Cambodia and Thailand. Cambodia has
1,200,000 hm2 and Viet Nam takes up 3,900,000 hm2 in Mekong Delta. In Viet
Nam, Mekong Delta consists of Tien Giang Plain, Hau Giang Plain and Dong Thap
Muoi Plain.
Topography of the Mekong River Basin is indicated in Fig. 1.2.

1.2.2 Meteorology

Situated in the tropical monsoon region of Asia, the Mekong River Basin has distinct
wet and dry seasons under the influence of the monsoon climate, with extremely
uneven annual distribution of precipitation. The Southwest Monsoon starts from
May until the late September, and then the Northeast Monsoon lasts from November
to mid-March. As the Southwest Monsoon brings a lot of moisture from the sea to
coastal countries, abundant rainfall is enjoyed, especially in hilly and mountainous
areas. On the contrary, the Northeast Monsoon comes from mainland, so it is dry and
rainfall is rarely witnessed. Precipitation is uneven between seasons, which is very
evident between wet and dry seasons. 80% of precipitation gathers in the wet season
between May and October.
In the meantime, the spatial distribution of rainfall in the Mekong River Basin
is also very uneven, as shown in Fig. 1.3. Under the influence of terrain, movement
of the Southwest Monsoon, the left bank of Mekong River is multi rain belt, with
much higher runoff yield than the right bank, it takes up 70% precipitation of the
whole basin. Annual precipitation rises from 1000 mm in northeastern Thailand to
4000 mm in the mountainous borders of the basin within Lao PDR, Cambodia and
Viet Nam. In the Chi-Mun River Basin of Thailand and the Tonle Sap Lake Basin of
eastern Cambodia, average annual precipitation declines evidently from east to west.
Temperature change is even within the Mekong River Basin, with an average
annual temperature between 25 and 27 °C. Average relative atmospheric humidity
is the highest, slightly higher than 80%, in September, and the lowest, merely 60%,
in March.

1.2.3 Hydrology

Hydrology closely reflects the pattern in rainfall distribution described in Fig. 1.3,
but also affected by topography, vegetation and soil texture. Accordig to MRC pub-
lication, the flow contribution from tributaries to the mainstream Lancang-Mekong
River in various reaches is shown in Table 1.3. The flood season in the Mekong River
Basin lasts from June to November and accounts for 80–90% of the total annual flow.
1 Overview of the Mekong River Basin 9

Fig. 1.2 Topography of the Mekong River Basin

10 Baiyinbaoligao et al.

Fig. 1.3 Annual precipitation distribution of the Mekong River Basin [Based on monthly
precipitation data (1901–2016) from Climate Research Unit (CRU). The unit is mm]
1 Overview of the Mekong River Basin 11

Table 1.3 Proportional contributions to total Lancang-Mekong River mean annual flow by river
reach, distinguishing those made by the left and right bank tributary systems (Mekong R C. Planning
Atlas of the Lower Mekong River Basin. Phnom Penh: Lao Uniprint Press Co. Ltd., 2011, p 53)
River reach Left bank (%) Right bank (%) Total (%)
Lancang 16a 16a
Lancang–Chiang Saen 1 4 5
Chiang Saen–Luang Prabang 6 3 9
Luang Prabang–Chiang Khan 1 2 3
Chiang Khan–Vientiane 0 0 0
Vientiane–Nong Khai 0 1 1
Nong Khai–Nakhon Phanom 19 4 24
Nakhon Phanom–Mukdahan 3 1 4
Mukdahan–Pakse 5 6 11
Pakse–Stung Treng 23 3 26
Stung Treng–Kratie 1 0 1
Total 60 16 24 100
a Chinasuggests the proportional contribution of Lancang River is 13.5% [Mekong River
Commission and Ministry of Water Resources of the People’s Republic of China (2016). Technical
Report—Joint Observation and Evaluation of the Emergency Water Supplement from China to the
Mekong River. Mekong River Commission, Vientiane, Lao PDR, p 2]

The wet and steep left bank tributaries used to have high runoff depth.7 Nam Ngun
and Nam Kading-Nam Thuen produce the greastest runoff depth up to 2500 mm per
year, accounting for around 5% of the Lancang-Mekong total runoff for each catch-
ment. The left bank tributaries between Vientiane and Nakhon Phanom contribute
around 19% of the total runoff. The 3 s River (Se Kong-Se San-Sre Pok), entering
the mainstream between Pakse and Stung Treng, contritutes 17% of the total runoff.
The drier and flatter catchements on the right bank used to have low runoff depth.
Among them, Chi-Mun River Basin and Tonle Sap Basin have the largest catchment
areas, contributing around 5% to the total runoff.

1.3 Socio-economy

1.3.1 Administrative Division

Provincial-level administrative divisions of countries within the Mekong River Basin

are shown in Fig. 1.4 and Table 1.4.
All of the 17 provinces of Lao PDR fall within the Mekong River Basin, take up
85.2% of the total area of Lao PDR.

7 Runoff depth is calculated by dividing the runoff (discharge) by its catchment area.
12 Baiyinbaoligao et al.

Fig. 1.4 Administrative divisions of the Mekong River Basin

1 Overview of the Mekong River Basin 13

Table 1.4 Provinces of the 5 Countries within the Mekong River Basin [Mekong R C. Planning
Atlas of the Lower Mekong River Basin. Phnom Penh: Lao Uniprint Press Co. Ltd., 2011, p 5.
(Most of the information was referred to this publication except that of Myanmar)]
Cambodia Lao PDR Myanmar Thailand Viet Nam
Banteay Attapeu Most parts of Amnat Charoen An Giang
Meanchey Kengtung
Battambang Bokeo Almost all parts Burirum Bac Lieu
of Mongphat
Kampong Cham Borikhamxay Tarchilaik Chaiyaphum Ben Tre
Kampong Champasak A small part of Chantaburi Binh Phuoc
Chhnang Mongsat
Kampong Speu Huaphanh Chiang Mai Ca Mau
Kampong Thom Khammuane Chiang Rai Can Tho
Kampot Luangnaumtha Katasin Dak Lak
Kandal Luangprabang Khon Kaen Dak Nong
Koh Kong Oudomxay Loei Dien Bien
Kratie Phongsaly Maha Sarakham Dong Thap
Mondul Kiri Saravane Mukdahan Gia Lai
Otdar Meanchey Savannakhet Nakhon Hau Giang
Ratchasima
Pailin Sekong Nong Bua Kon Tum
Lamphu
Phnom Penh Vientiane Nong Khai Lam Dong
Preah Sihanouk Vientiane Phayao Long An
Capital
Preah Vihear Xayaboury Phet Chabun Quang Tri
Prey Veng Xiengkhuang Roi Et Soc Trang
Pursat Sa Keo Thua Thien Hue
Ratanak Kiri Sakon Nakhon Tien Giang
Siem Reap Si Saket Tra Vinh
Stung Treng Surin Vinh Long
Takeo Ubon
Rachathani
Tonle Sap Lake Udon Thani
Yasothon
Total
23 17 4 25 22
14 Baiyinbaoligao et al.

All provinces, except Koh Kong, of Cambodia, are wholly or partly contained in
the Mekong Basin, accounting for 85.6% of the total national area.
25 out of the 77 provinces of Thailand located in the Mekong Basin, accounting
for 35.9% of the total national area.
22 of the 63 provinces of Viet Nam fall in the basin, of which 13 gathered in the
Mekong Delta and another 5 located at the sources of Se San Basin and Se Prok
Basin, accounting for 19.7% of the total national area.
The area of Myanmar falling whthin the basin only accounts for 3% of the national
territory. 4 cities of Shan State locate in the basin, of which Taichilaik wholly and
the other 3 partly falling within the basin.

1.3.2 Population Distribution

Population of the Mekong River Basin gather largely in Cambodia, Lao PDR, Thai-
land and Viet Nam, and only a small population live in Myanmar, as shown in Fig. 1.5.
Table 1.5 demonstrates the share of each country’s population in the basin and the
share of the basin population in each country. Population distribution is uneven.
Cambodia and Lao PDR only make up 28% of the basin’s population though most of
their territory is located there. Thailand has 37% of its territorial area located within
the Mekong River Basin, and also takes up 37% of its population. 20% of Viet Nam’s
territorial area is located in Mekong Delta and highland in the middle part, and the
country takes up 34% of the basin’s population. 80% of the basin’s population live
in rural areas.
Population density varies greatly across the Mekong River Basin. Mekong Delta
within Viet Nam and the area around Phnom Penh, Cambodia have a dense popula-
tion. In Mekong Delta, average population density is 443 people/km2 . That of Phnom
Penh reaches 41,200 people/km2 . Area around Vientiane has a population density
of 739 people/km2 . Reasons behind the dense population in these regions include
the convenient transport of goods on land and in inland waters, the adjacency with
Hanoi, fertile land, and being close to irrigation facilities and waters. In contrast,
population density along the highlands bordering Viet Nam is 20 people/km2 , and
population density is inversely proportional to altitude.8

1.3.3 Land Utilization

Land utilization of the basin in 2015 is demonstrated in Fig. 1.6.9 The main land
utilization include forest and paddy field. Paddy rice makes up 22% of the total

8 Mekong R C. Planning Atlas of the Lower Mekong River Basin. Phnom Penh: Lao Uniprint Press

Co. Ltd., 2011, p 15.

9 Basedon landuse database of Tsinghua University. http://data.ess.tsinghua.edu.cn/.
1 Overview of the Mekong River Basin 15

Fig. 1.5 Population distribution of the Mekong River Basin, 2011 [Based on GWPv4 data, Center
for International Earth Science Information Network (CIESIN), Columbia University. 2018. Doc-
umentation for the Gridded Population of the World, Version 4 (GPWv4), Revision 11 Data Sets.
Palisades NY: NASA Socioeconomic Data and Applications Center (SEDAC). https://doi.org/10.
7927/H45Q4T5F Accessed 15 July 2019]
16 Baiyinbaoligao et al.

Table 1.5 Share of country population in the Basin and share of basin population in the Country
[Based on the following references: Mekong R C. Planning Atlas of the Lower Mekong River Basin.
Phnom Penh: Lao Uniprint Press Co. Ltd., 2011, p 13 and Myanmar Statistical Yearook 2016. (For
the data of Myanmar is supplemented by the author, the “Share of Country Population in the Basin”
is recalculated accordingly)]
Country Population in the Share of country Total population Share of basin
basin (million) population in the of country population in the
basin (%) (million) country (%)
Cambodia 11.6 19 14.4 81
Lao PDR 5.3 9 5.9 90
Myanmar 0.8 1 51.5 2
Thailand 23.0 37 63.9 36
Viet Nam 20.7 34 87.4 24
Total 61.4 100 223.1 28

area,10 which is the major land utilization mode in the flood plains of the Chi-Mun
Basin in northeastern Thailand, the Vientiane Plain of Lao PDR, the Tonle Sap Lake
Basin of Cambodia and the delta in southern Viet Nam.

1.3.4 Industry

Agriculture is the most important economic activity in the Mekong River Basin for
75% of its population live on agriculture. Other industries include fishery, animal
husbandry, forestry and etc. In the Mekong River Basin, agriculture is divided into
two categories—for own use and for commercial use. 70% of the population are
self-sufficient farmers who grow paddy for their own demand and sometimes, leave
a small part for sale. Commercial farming is largely seen in low-lying areas. In
the Khorat Plain in northeastern Thailand, crops, such as tobacco and sugarcane,
are planted once a year, mainly for commercial purpose. Thanks to the commercial
cultivation of paddy in Mekong Delta within Viet Nam, the country becomes the
second largest paddy exporter in the world. In addition, Viet Nam has formulated
policies to encourage people to move to the central highlands and grow commercial
crops such as coffee, tea and rubber.
In the Mekong River Basin, fishery is also very important for people. The 12
million rural households in the basin depend on both farming and fishing for their
livelihood. Fish is the main source of animal protein in people’s diet. In Cambodia,
more than 1.2 million people live in near the Tonle Sap Lake, where fishery is the
only source of income. The annual fishery output of Tonle Sap Lake in Cambodia is

10 http://portal.mrcmekong.org/tech_report.
1 Overview of the Mekong River Basin 17

Fig. 1.6 Land utilization of the Mekong River Basin, 2015

18 Baiyinbaoligao et al.

Table 1.6 Economic and industrial status of countries in the basin, 2016
Cambodia Lao PDR Myanmar Thailand Viet Nam
GDP (billion USD) 200.2 158.1 632.3 4117.6 2052.8
Population (10,000 People) 1576 676 5289 6886 6457
Per capita GDP (USD) 1270 2339 1196 5980 3179
Share of agricultural added 25 17 25 9 16
value in GDP (%)
Share of industrial added value 29 29 35 36 33
in GDP (%)

about 1.5 million tons (about 230 kg/hm2 ), which is much higher than that of other
regions in Asia (about 0.5 million tons, that is 100 kg/hm2 ).11
According to publications of World Bank, per capita GDP, Share of Agricultural
Added Value in GDP and Share of Industrial Added Value in GDP are as illustrated
in Table 1.6. It can be seen per capita GDP is 1270 USD in Cambodia, 2339 USD
in Lao PDR, 1196 USD in Myanmar, 5980 USD in Thailand and 3179 USD in Viet
Nam. In terms of the ratio of agricultural added value to GDP, Thailand ranks bottom
by 9%, followed by Lao PDR by 17% and Viet Nam by 16%, and Cambodia and
Myanmar rank top either by 25%. In terms of the ratio of industrial added value to
GDP, these countries have small differences, and the industrial added value-to-GDP
ratio is bigger than the agricultural added value-to-GDP ratio. It is a major source of
GDP growth for these countries.

1.4 Current Status of Development and Utilization

of Water Resources

1.4.1 Irrigation Projects

Before the 1980s, water resources of the Mekong River were used largely for agricul-
tural irrigation, followed by shipping and fishery. Though a mass of water resources
were used for irrigation, the utilization rate was not high and water consumption
was huge. Since 1992, Mekong River has been planned uniformly; in addition to
irrigation, water resources have been used for hydroenergy development, though at
a low rate.

11 MRC Annual Flood Report 2005. Mekong River Commission, Vientiane, Lao PDR. 82 pp. ISSN

1728 3248, p 10.

1 Overview of the Mekong River Basin 19

Cambodia has registered an irrigation area of 479,762 hm2 in wet season, and will
increase it to 774,000 hm2 as of 2030 and to 838,000 hm2 as of 2060. Lao PDR has
3094 irrigation projects that cover an area of 225,446 hm2 in wet season, less than
that of other countries in the basin. Thailand has built 134 irrigation projects in the
Mekong River Basin. Those projects cover an area of 903,946 hm2 in wet season, and
are mainly located on both banks of rivers and the flood plains in northern Thailand.
Viet Nam has registered an irrigation area of 2.38 million hm2 in the wet season, and
won’t increase it in next decades. According to MRC publications, existing projects
and irrigation area in the Mekong River Basin are shown in Figs. 1.7, 1.8 and 1.9.
Viet Nam has the biggest existing irrigation area, taking up 59.5% of the gross
irrigation area of the Mekong River Basin in wet season; following it is Thailand
that accounts for 22.5%; the rest 18% is situated in Cambodia and Lao PDR. In the
following decades, the irrigation area of Cambodia and Lao PDR in the Mekong
River Basin may increase.

1.4.2 Hydropower Projects

The United Nations and the World Bank give high priority to the development of
sustainable hydropower, which is deemed low carbon and renewable, Hydropower
currently comprises 80% of the World’s installed renewable energy, and is a key
energy option for both poverty reduction and the reduction of greenhouse gas emis-
sions.12 Most industrial nations, including China, have developed a relatively high
proportion of their hydropower potential. However, China has only developed 10%
of its hydropower potential on its transboundary rivers.
Table 1.7 shows hydropower projects in the Mekong River Basin. 11 cascade
hydropower station projects are planned on the mainstream Mekong River, of which
2 are located on the boundary between Lao PDR and Thailand, 7 within the territory
of Lao PDR and the rest 2 within the territory of Cambodia. In hydropower projects
on tributaries of the Mekong River, 91 are located within the territory of Lao PDR,
15 in Viet Nam, 12 in Cambodia and 7 in Thailand.
In Cambodia, 2 mainstream hydropower stations, and 11 tributary dams which are
all located in the Se Kong-Se San-Sre Pok Basin, are planned in the Mekong River
Basin. In Lao PDR, there are 100 hydropower stations (73.5% of total hydropower
stations in the basin) existing, under construction or planned in the Mekong River
Basin, of which 9 hydropower stations are located on mainstream and the rest 91 on
tributaries. In the Mekong River Basin in Thailand, 7 hydropower stations have been

12 http://www.un.org/wcm/webdav/site/sustainableenergyforall/shared/Documents/SEFA-

Action%20Agenda-Final.pdf; http://www.worldbank.org/en/topic/hydropower/overview#1.
20 Baiyinbaoligao et al.

Fig. 1.7 Irrigation projects and existing irrigation reservoirs in the Mekong River Basin (http://
www.mrcmekong.org/assets/Publications/MRC-Irrigation-Database-Improvement-final-for-web.
pdf, p 53)
1 Overview of the Mekong River Basin 21

Fig. 1.8 Existing irrigation headworks and reservoirs in the Mekong River Basin (http://www.
mrcmekong.org/assets/Publications/MRC-Irrigation-Database-Improvement-final-for-web.pdf, p
54)
22 Baiyinbaoligao et al.

Fig. 1.9 Existing irrigation areas in the Mekong River Basin (http://www.mrcmekong.org/assets/
Publications/MRC-Irrigation-Database-Improvement-final-for-web.pdf, p 55)
1 Overview of the Mekong River Basin 23

Table 1.7 Statistics of hydropower projects under construction or planned on the main stream and
tributaries of the Mekong River [According to Mekong R C. Planning Atlas of the Lower Mekong
River Basin. Phnom Penh: Lao Uniprint Press Co. Ltd., 2011, p 79 and updated information of
Lower Se San 2 project in Cambodia (http://www.hydrosesan2.com/project.php?id=119), Xayabury
and Don Sahong projects in Lao PDR by 2016]
Country Main stream Tributaries Total
Under Planned Existing Under Planned
construction construction
Cambodia 2 2 10 14
Lao PDR 2 7a 11 9 71 100
Myanmar (Data – – – – – –
of Myanmar is
not available)
Thailand 7 7
Viet Nam 7 5 3 15
Total 2 9 26 14 85 136
a Two
of them located on the transboundary between Lao PDR and Thailand on the mainstream
Mekong River

constructed on the tributaries in the north, no more hydropower station is planned

in following years. In Viet Nam, there are 7 hydropower stations existing, 5 under
construction and 3 planned, which are all located on the tributaries. In specific,
hydropower stations under construction or planned are situated in the Se San-Sre
Pok Basin. The distribution of hydropower stations in the basin is demonstrated in
Fig. 1.10.
The gross installed capacity of the hydropower projects existing, under construc-
tion and planned in the Mekong River Basin is 29,684 MW. According to statistic in
2011, existing hydropower stations register a gross installed capacity of 2688 MW,
hydropower stations under construction 5320 MW and planned hydropower stations
21,676 MW, as shown in Table 1.8.
Hydropower projects with large installed capacity are mainly located on the main-
stream, with an installed capacity of 14,697 MW or 49.5% of the gross installed
capacity. In specific, hydropower projects on mainstream Mekong River in Lao PDR
register an installed capacity of 10,417 MW or 70.9% of gross installed capacity
of mainstream. The gross installed capacity of hydropower projects existing, under
construction or planned on tributaries is 14,987 MW, mainly located in Lao PDR.
24 Baiyinbaoligao et al.

Fig. 1.10 Hydropower projects in the Mekong River Basin

1 Overview of the Mekong River Basin 25

Table 1.8 Statistics of gross installed capacity of hydropower in the Mekong River Basin [Mekong
R C. Planning Atlas of the Lower Mekong River Basin. Phnom Penh: Lao Uniprint Press Co. Ltd.,
2011, p 81 (The table was updated by the author according to hydropower constructions in Cambodia
and Lao PDR by 2016)]. Unit: MW
Country Main stream Tributaries Total
Under Planned Total Existing Under Planned Total
construction construction
Cambodia 4280 4280 1 1309 1310 5590
Lao PDR 1540 8877a 10417 738 2764 6847 10,350 20,767
Thailand 745 745 745
Viet Nam 1204 1016 363 2583 2583
Total 1540 13157 14697 2688 3780 8519 14,987 29,684
a 3879MW installed capacity is located on mainstream Mekong River on the border between Lao PDR and
Thailand

Dr. Baiyinbaoligao’s research primarily focuses on Eco-hydraulics, urban river, and river restora-
tion. He was the Program Manager of important national and provincial level projects. The num-
ber of investigated or participated projects and programs exceeds 70. He has written 4 technology
standards (or specifications) and 4 books, has published over 80 papers.

Dr. Liu’s research primarily focuses on hydrological processes and modeling, flood management,
and transboundary river water cooperation. She has over 5 years’ research experiences in trans-
boundary water cooperation of China. She was a participant of various Lancang-Mekong water
resources cooperation activities. She is an editor of a joint publication by MWR of China and
MRC in 2016. She is Program Managers of 3 international cooperation projects with Mekong
River countries. She is author of over 10 journal papers and 1 book.

Dr. Chen’s research primarily focuses on comprehensive management of urban rivers, river eco-
logical restoration, river sediment management, transboundary river water management and flood
management. She was the Program Manager of many important national and provincial level
projects. She has over 2 years’ research experiences in transboundary water cooperation of China.
The number of participated projects and programs exceeds 50. She has written 4 technology
standards (or specifications) and 4 books, has published over 50 papers.

Dr. Mu’s research primarily focuses on water resources management, river ecosystem restoration
and hydraulic computation. He has over 10 years’ research experiences in relative national level
projects. He is author of over 40 journal papers and 4 books, as well as dozens of patents for
inventions.
Chapter 2
Summary of Flood and Drought
in Mekong River Basin

Xingru Chen, Hui Liu, Baiyinbaoligao, and Xiangpeng Mu

Abstract Flood and drought disasters occur frequently in Mekong River basin owing
to ocean climate and climate change. Based on the collected floods and droughts in
Mekong River basin in recent 20 years, the losses, causes and effects of flood and
drought disasters are analyzed. The main conclusions are drawn as follows: (1) Flood-
caused fatalities in the Mekong River Basin were most serious, 825, in 2000, followed
by 2001 (489), 2011 (396), 2013 (247) and 1996 (173). Cambodia and the Cuu
Long Delta, Viet Nam take the largest shares while the share is small in northeastern
Thailand and Lao PDR. (2) Flood imposes significant impact on agriculture. 2000 saw
the biggest agricultural impact of floods, approximately 2.50 million hm2 , followed
by 2011, about 500,000 hm2 . Seen from the geographic distribution, agricultural
impact gathered in the Cuu Long River Delta, Viet Nam in 2000, and Cambodia. (3)
Flood incurs serious economic loss in countries in the basin. Cambodia ranks top by
1.4 billion USD, followed by Viet Nam by 980 million USD and Lao PDR by 590
million USD. Thailand ranks bottom by 310 million USD in 1996–2014. (4) Extreme

Myanmar’s territorial area in the Mekong River Basin takes up approximately 3% of the drainage
area of the Mekong River, and the mainstream Mekong River is the boundary river between Myanmar
and Lao PDR. According to the statistical analysis of flood disaster data of Mekong River Basin
countries, Myanmar’s flood disasters mainly took place outside the Mekong River Basin. Therefore,
emphasis is placed on the Mekong River Basin in Cambodia, Lao PDR, Thailand and Viet Nam
when it comes to the analysis of flood disasters in the Mekong River Basin in this paper.

X. Chen (B) · H. Liu · Baiyinbaoligao · X. Mu

China Institute of Water Resources and Hydropower Research, A-1 Fuxing Road,
Haidian District, Beijing 100038, China
e-mail: [email protected]
H. Liu
e-mail: [email protected]
Baiyinbaoligao
e-mail: [email protected]
X. Mu
e-mail: [email protected]

© China Water & Power Press and Springer Nature Singapore Pte Ltd. 2020 27
H. Liu (ed.), Flood Prevention and Drought Relief in Mekong River Basin,
Springer Tracts in Civil Engineering,
https://doi.org/10.1007/978-981-15-2006-8_2
28 X. Chen et al.

floods take place more frequently and cause a huge loss upon the beginning of the
twenty-first century. The flood loss amounted to 1.164 billion USD in 2011, taking
up 35% of the gross flood loss in 1996–2014; to 651 million USD in 2013, taking up
20%; and to 462 million USD in 2000, taking up 14%. (5) Drought features extensive
influence, long duration and huge economic loss. Seen from spatial distribution,
drought loss is heavy in northeastern Thailand, Cambodia and Viet Nam and relatively
mild in Lao PDR.

2.1 Objective

In order to effectively cope with flood and drought disasters in the basin in the future,
it is necessary to deepen the recognition of the current status of flood and drought
disasters in the Mekong River Basin, which will also lay foundation for the further
analysis and modelling.
Records of flood and drought was collected and summarized, with a statistic of
the loss in the past decades, which will provide a brief introduction of the spatial and
temporal distribution of disasters in the basin.
Typical flood and drought were illustrated with detailed description based on data
available, in order to provide the readers a better understanding of the whole picture
of the typical events.

2.2 Data and Methodology

2.2.1 Data

Data of natural disasters of the Mekong River Basin was collected and analyzed. The
data sources include:
• International Disaster Database (EM-DAT) established by the Center for Research
on the Epidemiology of Disasters (CRED) of Belgium’s University of Leuven;
• Asian Disaster Reduction Center (ADRC);
• ReliefWeb—a disaster statistics website sponsored by the UN Office for the
Coordination of Humanitarian Affairs (OCHA) (https://reliefweb.int);
• Disaster data from the Mekong River Commission, including Annual Mekong
Flood Report etc.
The EM-DAT database contains detailed recording of the occurrence time, place,
type, fatalities, affected population, economic loss etc. Those data are from a vari-
ety of sources, including relevant UN organizations, NGOs, insurance compa-
nies, research institutions and publishing agencies. The database is a free shared
2 Summary of Flood and Drought … 29

database that records the most and complete information of disasters, is very reliable
and authoritative, and has been widely quoted by many international institutions,
organizations and scholars.
A disaster won’t be included into EM-DAT until it meets one of the following
three conditions: 10 or more people reported killed; 100 or more people reported
affected; declaration of a state of emergency or request for international aid. There-
fore, seen from the completeness and comprehensiveness of disaster recordings for
years, international public data are inferior to national databases. But these data can
meet the requirements in the macroscopic analysis of the spatio-temporal distribution
of flood loss, especially as basis for relatively major disaster loss.
Every database has its characteristics. The EM-DAT database has a small number
of parameters about single disaster event, including fatalities, affected population,
occurrence place and economic loss parameters, but it is advantageous in the record-
ing length and can analyze the long-sequence variation trend of a given disaster, but
doesn’t present detailed description of disaster events. The advantage of the disaster
statistics web of the Asian Disaster Reduction Center (ADRC) lies in the relatively
detailed description of single disasters, but the disadvantage is the short length of
record and its data mainly start from the 1990s.
Disaster analysis of this assessment report is based on data from said various
international public authoritative webs. Those data are verified and supplemented
mutually, so the report is exempted from the problem of one-sided data that may be
caused by taking a single database as the data source, and the statistical analysis and
research of disaster loss in this project is based on solid fundamental data support.

2.2.2 Methodology

Based on the above-mentioned public data sources, we analyzed the spatio-temporal

distribution characteristics of the occurrence time, occurrence place, fatalities,
affected population, housing damage, affected farmland, traffic, school and eco-
nomic losses and other parameters of flood and drought disasters in the countries
located in the Mekong River Basin.
For the description of typical floods, the meteorological conditions, water level,
discharge and flood loss were analyzed. The findings from previous publications
were primarily used in this section.
For the description of typical droughts, some drought indicators (SPI and SRI)
were applied. The descripition of these indexes please refer to that in Chap. 4.
30 X. Chen et al.

2.3 Summary of Flood

2.3.1 Flood Loss

As a major disaster in the Mekong River Basin, flood threatens the life and property
safety of people there by leading to housing damage, road closure and school suspen-
sion, affecting the order of production and life and causing damage or reduction of
output in the most important industry—agriculture in the basin to a varying degree. In
the following, fatalities, affected population, affected agricultural area and economic
loss of flood disasters in the Mekong River Basin are analyzed respectively.

(1) Flood-caused fatalities

Table 2.1 showcases the fatalities caused by flood disasters in the Mekong River Basin
in the recent 20 years. Fatalities and a huge life and property loss were witnessed in
almost every flood disaster in the Mekong River Basin.
Figure 2.1 showcases the variation of yearly flood-caused fatalities in the basin.
2000 saw the most flood-caused fatalities, 825, followed by 2001 (489), 2011 (396),
2013 (247) and 1996 (173). The death toll is smaller than 100 in other years.
Figure 2.2 showcases the share of fatalities at each country in the basin. Cambodia
and the Cuu Long River Delta, Viet Nam rank top. In specific, the biggest death toll

Table 2.1 The spatio-temporal distribution of flood-caused fatalities in the Mekong River Basin
(Unit: Person)
Year Cambodia Lao PDR Northeast Cuu Long Eastern Total
Thailand River Delta Highlands of
Viet Nam
1996 169 – – – 4 173
2000 347 – 25 453 >20 825
2001 62 – 34 393 – 489
2002 – 3 – 71 2 76
2003 – – – 23 6 29
2004 – – – 38 – 38
2005 4 5 0 44 – 53
2006 11 5 – 55 0 71
2007 10 2 – 30 29 71
2008 – 7 – 7 7 21
2010 8 7 – – 4 19
2011 250 42 – 89 15 396
2012 26 5 – 0 0 31
2013 168 17 17 45 247
2014 49 5 4 17 75
2 Summary of Flood and Drought … 31

900

800

700

600
people killed

500

400

300

200

100

0
1996 2000 2001 2002 2003 2004 2005 2006 2007 2008 2010 2011 2012 2013 2014
year
Cambodia Lao PDR Cuu Long River Delta
Eastern Highlands of Viet Nam Northeast Thailand

Fig. 2.1 Variation of yearly flood-caused fatalities in the basin

100% 0 0 0 0 0 0
04 25
0 34 2 15 17 4
0 5
90% 6 4
89 45 17
7
80% 29 0
relative ratio of people killed

70% 453 45
42 17
60% 55 7 17
44
5
393 71
50% 169 38 7

40% 0 26
23 30
30% 250
168 49
20% 347 5 8
2 7
0 5
10%
62 11 10
4
0% 03 0 0 0
1996 2000 2001 2002 2003 2004 2005 2006 2007 2008 2010 2011 2012 2013 2014
year
Cambodia Lao PDR Cuu Long River Delta
Eastern Highlands of Viet Nam Northeast Thailand

Fig. 2.2 Relative proportion of yearly flood-caused fatalities by country

32 X. Chen et al.

of Cambodia was witnessed in 1996, 2000 and 2010–2014, and that of the Cuu
Long River Delta, Viet Nam was seen in 2000–2007. The share of fatalities is low in
northeastern Thailand. As for Lao PDR, the share reached 30% in 2008 and 2010,
but with a small death toll, so Lao PDR’s flood-caused fatalities are fewer than other
countries.
(2) Flood-affected population

Table 2.2 demonstrates the flood-affected population in the Mekong River Basin.
Figure 2.3 shows the yearly variation in the distribution of affected population. It can
be seen that 2000 ranks top by 13.40 million people affected by floods in the basin,
followed by 2001 by 2.26 million people and 2002 by 2.05 million people, and the
population fell between several tens of thousands and hundreds of thousands in other
years.

(3) Flood-affected agricultural area

Table 2.3 shows the flood-affected agricultural area in the Mekong River Basin.
Agriculture is an important industry at the Mekong River Basin and also an important
source of food for people there. Every flood disaster causes damage to agriculture to
a varying degree, in addition to human life loss in the basin.
Yearly flood-affected agricultural area at the Mekong River Basin is showcased
in Fig. 2.4. 2000 saw the biggest agricultural impact of floods, approximately 2.50

Table 2.2 Flood-affected population in the mekong river basin (Unit: Person)
Year Cambodia Lao PDR Northeast Cuu Long Eastern Total
Thailand River Delta Highlands of
Viet Nam
1996 – – – – – –
2000 3,400,000 – – 10,000,000 – 13,400,000
2001 600,000 – 660,000 1,000,000 – 2,260,000
2002 1,500,000 249,800 – 300,000 – 2,049,800
2003 – – – – – 0
2004 – – – – – 0
2005 – 480,900 305,000 – – 785,900
2006 – 89,800 – 77,700 – 167,500
2007 147,200 118,100 – 67,500 – 332,800
2008 – 95,200 – – – 95,200
2010 – 86,097 82,000 – 4000 172,097
2011 – – – – –
2012 – 3047 – – – 3047
2013 – – 429,000 – – 429,000
2014 – 92,165 400,000 800 – 492,965
2 Summary of Flood and Drought … 33

14000000

12000000

10000000
people affected

8000000

6000000

4000000

2000000

0
1996 2000 2001 2002 2003 2004 2005 2006 2007 2008 2010 2011 2012 2013 2014
year
Cambodia Lao PDR Cuu Long River Delta
Eastern Highlands of Viet Nam Northeast Thailand

Fig. 2.3 Variation of yearly flood-affected population in the Mekong River Basin

Table 2.3 Flood-affected agricultural area at the Mekong River Basin (Unit: hm2 )
Year Cambodia Lao PDR Northeast Cuu Long Eastern Total
Thailand River Delta Highlands of
Viet Nam
1996 25,020 67,500 – – – 92,520
2000 421,600 42,900 – 2,000,000 – 2,464,500
2001 164,200 42,200 – – – 206,400
2002 45,000 33,700 – – 9000 87,700
2003 – 800 – – 1000 1800
2004 247,400 14,400 – – – 261,800
2005 55,000 56,000 39,500 – – 150,500
2006 14,500 6900 – 14,700 130 36,230
2007 9500 7500 – 46,400 20,300 83,700
2008 18,900 28,500 – 28,500 80 75,980
2010 18,527 3861.1 14,000 – 5814 42,202
2011 284,300 77,000 – 250,000 – 611,300
2012 – 147,800 – 65,906 – 213,706
2013 – – 301,000 – – 301,000
2014 – – 34,000 3098 – 37,098
34 X. Chen et al.

Damaged Crops(hm2) 2500000

2000000

1500000

1000000

500000

0
1996 2000 2001 2002 2003 2004 2005 2006 2007 2008 2010 2011 2012 2013 2014
year
Cambodia Lao PDR
Cuu Long River Delta Eastern Highlands of Viet Nam
Northeast Thailand

Fig. 2.4 Yearly flood-affected agricultural area at the Mekong River Basin

million hm2 , followed by 2011, about 500,000 hm2 . Seen from the geographic dis-
tribution, agricultural impact gathered in the Cuu Long River Delta, Viet Nam in
2000, and gathered mostly in Cambodia and the Cuu Long River Delta, Viet Nam
and partially in Lao PDR in 2011. In 2013, floods mainly affected agriculture in the
irrigation regions of northeastern Thailand. In 2001 and 2004, affected agricultural
area was about 200,000–250,000 hm2 , largely located in Cambodia.

(4) Flood-caused economic loss

Floods cause reduction or damage of agricultural output, road closure, damage of

water conservancy facilities, school suspension and housing damage, and yearly
flood-caused economic loss at countries in the basin is summarized in Table 2.4. It
can be seen that Cambodia and Viet Nam rank top in terms of the share of flood-
caused economic loss by 43% and 29.6% respectively, followed by Lao PDR 17.9%
and northeastern Thailand 9.6%.
Though all located in the Mekong River Basin, Cambodia, Lao PDR, Thailand
and Viet Nam differ from one another in the temporal and spatial distribution of flood
disasters because they have different hydrological and geographical conditions. For
a detailed analysis on the inter-annual variation of flood loss in these countries, the
variation in the yearly flood-caused economic loss of each country is expounded
below.
Figure 2.5 shows the year-on-year flood-caused economic loss of Cambodia in
the Mekong River Basin between 1996 and 2014. It can be seen that the country’s
most serious economic loss was seen in 2011 (624 million USD), 2013 (356 million
USD) and 2000 (161 million USD).
Figure 2.6 shows the year-on-year flood-caused economic loss of Lao PDR in the
Mekong River Basin between 1996 and 2014. The country’s flood-caused economic
2 Summary of Flood and Drought … 35

Table 2.4 Flood-caused economic loss at the Mekong River Basin (Unit: Mn USD)
Year Cambodia Lao PDR Northeast Cuu Long Eastern Total
Thailand River Delta Highlands
of Viet Nam
1996 86.5 10.4 – 113 – 209.9
2000 161 30 21 250 – 462
2001 36 56 23.9 99 – 214.9
2002 12.5 61 – 0.3 3 76.8
2003 – 18.3 – 15 0.5 33.8
2004 55 4.1 – 3 – 62.1
2005 3.8 18.3 2.8 3.5 – 28.4
2006 11.8 3.1 6.8 15 – 36.7
2007 9 18 – 1.5 50.8 79.3
2008 5.8 56 – – 1 62.8
2010 62 21 47 55 – 185
2011 624 220 – 260 60 1164
2012 – 1.5 – 16 1 18.5
2013 356 62 210 23 0.2 651.2
2014 – 12 6 2.7 5.7 26.4
Total loss 1423.4 591.7 317.5 857 122.2 3311.8
Average loss 118.6 39.4 45.4 61.2 17.5 282.1
Percentage 43.0 17.9 9.6 25.9 3.7 100
(%)

700
economic loss (million dollars)

600

500

400

300

200

100

0
1996 2000 2001 2002 2003 2004 2005 2006 2007 2008 2010 2011 2012 2013 2014
year

Fig. 2.5 Flood-caused economic loss in Cambodia, 1966–2014

36 X. Chen et al.

250

economic loss (million dollars)

200

150

100

50

0
1996 2000 2001 2002 2003 2004 2005 2006 2007 2008 2010 2011 2012 2013 2014
year

Fig. 2.6 Flood-caused economic loss in Lao PDR, 1966–2014

Flood-caused Economic Loss
(Unit: Mn USD)

Fig. 2.7 Flood-caused economic loss of the Mekong River Basin in Thailand, 1989–2008 [MRC
(2009) Annual Mekong Flood Report 2008, Mekong River Commission, Vientiane. 84 pp, p 27]

loss amounted to 220 million USD in 2011, and was 50 million USD or below in
other years.
Figure 2.7 shows the year-on-year flood-caused economic loss of Thailand in the
Mekong River Basin between 1989 and 2008. The country’s flood-caused economic
loss in the Mekong River Basin stayed at a steady level, with the biggest economic
loss being about 370 million USD (in 2002), and the amount was also big, around
300 million USD, in 1989, 2000 and 2006. On the whole, Thailand’s flood loss in
the Mekong River Basin tends to fluctuate cyclically. Its economic loss caused by
extreme floods in the Mekong River Basin in 2011 and 2013 remains unknown,
which should be even hundreds of millions of USD according to Lao PDR’s flood
loss estimate.
2 Summary of Flood and Drought … 37

Flood-caused Economic Loss

(Unit: Mn USD)

Fig. 2.8 Flood-caused economic loss of Mekong Delta in Viet Nam, 1995–2008 (Unit: Mn USD)
[MRC (2009) Annual Mekong Flood Report 2008, Mekong River Commission, Vientiane. 84 pp,
p 28]

Figure 2.8 shows the year-on-year flood-caused economic loss of the Mekong
Delta in Viet Nam between 1995 and 2008. The most serious loss, 240 million USD,
was seen in 2000, and the amount was about 100 million USD in 1996 and 2001.
Figure 2.9 shows the flood loss of the Mekong River Basin between 1996 and
2014. The flood loss in 2000, 2011 and 2013 takes up 69% of the gross flood loss.
In specific, 2011 ranks top by taking up 35% of the 19-year flood loss, followed by
2013 by 20% and 2000 by 14%.

2000, 462, 14%

other years,
1034.6, 31%

2011, 1164,
35%
2013, 651.2,
20%

Fig. 2.9 Yearly flood loss in the Mekong River Basin, 1996–2014 (Unit: Mn USD)
38 X. Chen et al.

2.3.2 The Spatio-temporal Distribution of Flood Loss

For the results of the yearly statistical analysis of flood loss at the basin in the recent
20 years, please refer to Figs. 2.10, 2.11 and 2.12. The frequency of extreme flood
events increased and each flood event caused a great loss upon the beginning of
the twenty-first century. The flood loss amounted to 1.164 billion USD in 2011,
taking up 35% of the gross flood loss in 1996–2014; to 651 million USD in 2013,
economic loss (million dollars)

1400
1200
1000
800
600
400
200
0
1996 2000 2001 2002 2003 2004 2005 2006 2007 2008 2010 2011 2012 2013 2014
year
Cambodia Lao PDR
Cuu Long River Delta Eastern Highlands of Viet Nam
Northeast Thailand total

Fig. 2.10 Yearly flood loss of the Mekong River Basin by region

Thailand, 318,
10%
Cambodia,
Viet Nam, 979, Cambodia
1423, 43%
29% Lao PDR
Viet Nam
Lao PDR, 592,
18% Thailand

Fig. 2.11 Flood loss in the Mekong River Basin, 1996–2014 (Unit: Mn USD)
2 Summary of Flood and Drought … 39

Thailand, 80,
3%

Cambodia,
1104, 41%
Viet Nam,
1394, 52%

Lao PDR, 98,

4%

Fig. 2.12 Flood-caused fatalities in the Mekong River Basin, 1996−2014 (Unit: Person)

taking up 20%; and to 462 million USD in 2000, taking up 14%. The flood loss was
heavy in those three years. Compared to small floods, economic losses caused by
extreme floods in different regions at the basin in these countries are synergetic in
time. Cambodia and Viet Nam register the largest flood loss and the largest share of
fatalities.
Frequency statistics of flood disasters in the Mekong River basin was made from
1962 to 2017, the result is shown in Fig. 2.13. High flood risks are distributed in the
lower reaches, especially in Cambodia and some provinces in central Lao PDR that
close to Viet Nam. In the Mekong River Basin in Myanmar, Thailand and Viet Nam,
flood risk is at an intermediate level. In provinces located in the upstream basin, flash
floods are also caused by the oceanic climate, but at a low rate and with a limited
area and a small population affected.

2.3.3 Flood Benefit

Flood is two-sided. It imposes negative impact on human production and life on

the one hand and on the other hand, improves soil fertility by taking with it a great
many organic matters. As for the Mekong River Basin, Cambodia’s flood plains and
Mekong Delta are among the earliest origins of Southeast Asian civilization, where
people recognized early before 1800 that the millions of tons of sediment deposition
rich in organic matters taken by floods could improve the agricultural productivity
of the farmland in Cambodia’s flood plains and Mekong Delta. Table 2.5 shows the
40 X. Chen et al.

Fig. 2.13 Flood statistics of 1962–2017 in Mekong River Basin

2 Summary of Flood and Drought … 41

Table 2.5 Gross benefit of Mekong River Floods for regional agricultural production [MRC (2009)
Annual Mekong Flood Report 2008, Mekong River Commission, Vientiane. 84 pp, p 11]
Country Agricultural value of flood (108 USD) Remarks
Cambodia 10 2006 data. Basis of calculation: Paddy
area takes up 90% of gross agricultural
area, and 32% of paddy area is located
in flood plains
Lao PDR Non-significant Agriculture is largely dependent on
summer rain and autumn irrigation
Thailand Non-significant –
Viet Nam 35 2004 Mekong Delta data
Total 45 –

gross benefit the floods of the Mekong River bring to the agricultural production of
Cambodia and Viet Nam.
Floods also bring significant benefits to the aquaculture industry represented by
fishery, in addition to agriculture. When floods register long duration and wide
coverage, the biologic amount of fishery will be high.
Flood is also of important significance for maintaining the ecological service
value of wetland, which includes many functions such as food and freshwater supply,
climate regulation, hydrological rhythm, pollution control, leisure and entertainment,
aesthetics and education, biodiversity protection and nutrient cycling. Figure 2.14
shows wetland of Lao PDR along the Mekong River.

Fig. 2.14 Wetland in Lao PDR

42 X. Chen et al.

2.4 Summary of Drought

Drought means the precipitation in a period is apparently smaller than multi-year

average precipitation, leading to a relative shortage of water for a series of water cir-
culation elements. The water circulation process follows an axis from atmospheric
process (precipitation), surface process (surface water) to underground process (soil
water, groundwater). Water deficit in a link will affect the next link, and these links
will influence one another along with water circulation progress. Water deficit in
any stage of water circulation will lead to drought in the corresponding form. When
rainfall of a region is deficit compared to the historical level, the region will suffer
meteorological drought; along with the progression of the drought, soil moisture will
decline continuously because of the lack of rainfall replenishment, which will make
it difficult for the effective replenishment of river runoff and thereby, will cause the
hydrological drought that features slower and smaller flow. Based on the aforemen-
tioned formation mechanism, American Meteorological Society (AMS) classifies
drought as: meteorological drought (precipitation decreases evidently compared to
the same period of normal years), hydrological drought (long-term deficiency of
precipitation leads to shortage of surface water and groundwater compared with
multi-year average) and agricultural drought (water requirement for growth of crops
is unsatisfied because of soil moisture deficiency). In specific, meteorological drought
mainly affects rain-fed paddy, taking up around 90% of paddy in the Mekong River
Basin in Lao PDR, Thailand and Cambodia, so agriculture of the Mekong River
Basin is under great influence of drought.

2.4.1 Drought Loss

Drought loss in the Mekong River Basin is mainly made up of agricultural loss,
such as reduction or devastation of yield, especially for paddy. In addition, fishery
and livestock yields may be also reduced. According to MRC “Report of Flood and
Drought Management and Mitigation Programme 2011–2015”, EM-DAT database
and ADRC data, the recent drought losses in the Mekong Basin are as in Table 2.6.
It should be noted the drought loss in said table is merely the estimated loss in
a given aspect, so the drought loss is understated. Even so, the loss remains huge.
The drought loss of the Cuu Long River Delta in 2002 and May 2004 is much
higher than that of the loss of 7 out of the 9 floods in 2000–2008. In northeastern
Thailand, drought-caused paddy loss is about 10 million USD/year, and its impact
on aquaculture of the Tonle Sap Lake is also very apparent, with a yearly loss of 15
million USD/Year.1
According to the survey of drought loss of each country in the basin, droughts
cause bigger affected population and economic loss than floods, though drought

1 http://www.mrcmekong.org/assets/Publications/basin-reports/FMMP-working-paper-110820.

pdf, p iii.
2 Summary of Flood and Drought … 43

Table 2.6 Estimate of drought loss in the Mekong River Basin

Country Year Estimated loss Evaluation Data source
(Mn USD)
Cambodia 1987 – – EM-DAT
1994.6 100 – EM-DAT
1998.9 14.5 Fishing loss of WFP, 2009
the Tonle Sap
Lake
1999 6 Mekong Delta MRC working
paper
(2011–2015)
2001.9 – – EM-DAT
2002.3 22 Mainly paddy IRRI, 2007
output
2003.4 14.5 Fishing loss of MRC working
the Tonle Sap paper
Lake (2011–2015)
2004.5 21 Mainly paddy WFP, 2009
output
2007.8 14.5 Fishing loss of MRC working
the Tonle Sap paper
Lake (2011–2015)
2016.5 – Provinces near EM-DAT
Tonle Sap Lake
Lao PDR 1977 – Southern EM-DAT
provinces
1987 – Northern
provinces
1988.12 – Southern
provinces
1991.7 1 National
1999.1 – –
Myanmar No record in the EM-dAT or ADRC database
Northeastern 1991.3 – – EM-DAT
Thailand 1993 2 North and EM-DAT
Middle Thailand
1999.1 – – EM-DAT
2002.2 2.3 Northeastern EM-DAT
Thailand
(Mekong River
Basin part)
(continued)
44 X. Chen et al.

Table 2.6 (continued)

Country Year Estimated loss Evaluation Data source
(Mn USD)
2005.1–2005.3 – Thailand, EM-DAT
including some
provinces in the
Mekong River
Basin
2008.4 – Most provinces EM-DAT
of Thailand,
including some
provinces in the
Mekong River
Basin
2010.3 – Most provinces EM-DAT
of Thailand,
including some
provinces in the
Mekong River
Basin
2011.6 – Most provinces EM-DAT
of Thailand,
including some
provinces in the
Mekong River
Basin
2012.4–2012.8 1.2 Most provinces EM-DAT
of Thailand
2014.3 – 42 provinces in EM-DAT
northern
Thailand, 28
provinces in the
northeast
2016.5 – Only paddy EM-DAT
output, loss may
be
underestimated
Annual Loss 10 Thailand, MRC working
including some paper
provinces in the (2011–2015)
Mekong River
Basin
Viet Nam Cuu 1997.12–1998.4 – National EM-DAT
Long River delta 1998.12–1999.4 6 Cuu Long River EM-DAT
delta
2002 24 Mekong Delta Mekong Flood
Report
(continued)
2 Summary of Flood and Drought … 45

Table 2.6 (continued)

Country Year Estimated loss Evaluation Data source
(Mn USD)
2004.7–2004.11 42 Mekong Delta MARD, 2005
2004.11–2005.4 110 Middle and EM-DAT
South Viet Nam
2016.5 675 Mekong Delta EM-DAT

doesn’t take place frequently, and individual droughts feature extensive impact, long
duration and huge economic loss and the gross loss incurred thereof tends to increase
significantly.
Cambodia records 10 drought events, of which the 1987 drought isn’t recorded
with occurrence place, affected population and loss data. 6 of the other 9 droughts took
place after 2000, indicating drought frequency increased significantly in the recent
20 years in Cambodia. Seen from affected population, the 1994 drought incurred a
loss of 100 million USD, and the loss of other droughts fell between 6 million USD
and 22 million USD.
Lao PDR encountered 5 droughts, mainly between 1987 and 1999, and no drought
was reported since the beginning of the twenty-first century. In the country, a single
drought imposes extensive influence in many provinces and even across the country.
On the whole, droughts feature small frequency and extensive influence in Lao PDR.
In Myanmar, drought frequency is low, and the country has no drought report no
matter on Belgium’s EM-DAT web or Asian disaster reduction web.
In northeastern Thailand (the part in the Mekong River Basin), a total of 11
drought events took place, mainly after 2000, which indicates the country’s drought
frequency is also on a prominent rise upon the beginning of the twenty-first century.
Seen from the economic loss of droughts, drought loss of 72 provinces in Thailand
amounted to 3.3 billion USD in 2016 and 420 million USD in 2005. According to
conservative estimate based on nationwide drought loss data, the drought loss in
northeastern Thailand was 1.0–2.0 billion USD in 2016 and 100–200 million USD
in 2005.
Drought recording of the Mekong Delta in Viet Nam started from 1997 to 2016,
during which 6 drought disasters took place. Compared to other countries, its drought
loss is the most serious though at a low frequency, and single drought’s loss is about
24 million USD.

2.4.2 Typical Drought Disasters

Abnormally deficient rainfall is a critical factor driving the formation and occurrence
of drought. When the precipitation in a region is small for a long time compared to the
same periods in history, meteorological drought will be formed in the region; along
46 X. Chen et al.

with the further development of drought, soil moisture will start to decline because
of the persistent shortage of rainfall replenishment, effective replenishment of river
runoffs will be difficult and thereby, river discharge will decrease and hydrological
drought will occur. Over the past few decades, countries along the Mekong River have
experienced different degrees of drought events, causing tremendous impacts on agri-
culture, fisheries and production and life. This research selected two typical drought
events in 2004–2005 and 2016 to reveal the causes, development process and impact
of drought from the atmospheric circulation, meteorology, hydrology and other back-
grounds. Specifically, the standardized precipitation index (SPI) (SPI1, SPI3, SPI6
and SPI12)2 for diagnosing meteorological drought was established on different tem-
poral scales of one month, three months, six months and twelve months based on the
rainfall data of the past 116 years; and three typical hydrological stations, Chiang
Saen, Mukdahman and Stung Treng, were selected as representatives of hydrological
characteristics of upper, middle and lower reaches of the mainstream Mekong River,
as well as the one-month, three-month, six-month and twelve-month standardized
runoff indexes (SRI1, SRI3, SRI6 and SRI12) were calculated respectively based on
each station’s average monthly cross-section runoff sequence from January 1985 to
December 2016. Based on the two established representative indicators of SPI and
SRI, the causes and development of typical drought were analyzed.
1. Drought in 2004–2005
Since September 2004, most of the middle-east equator Pacific Ocean has been
controlled by positive SST anomaly above 0.5 °C and the SST anomaly index in the
NINO composite region (NINO1 + 2 + 3 + 4) has reached 0.6, indicating that the
atmosphere-ocean in the tropical Pacific Ocean has entered the El Nino state; during
the next 3–4 months (October 2004–January 2005), the sea surface temperature
in the middle-east equator Pacific Ocean continued to be warmer and the most of
the equatorial Pacific maintained a positive SST anomaly above 0.5 °C. Southern
Oscillation Index continued to be negative and the El Nino phenomenon continued;
until May 2005, the Pacific sea surface temperature returned to normal, marking the
end of the El Nino event.3
Under the background of tropical Pacific ocean-atmosphere circulation anomaly,
the early end of the wet season in 2004 in the Mekong River Basin led to widespread
rainfall deficit in Thailand, Cambodia, Viet Nam and other countries. Figure 2.15
shows the spatial distribution of SPI6 in the Mekong River Basin in the dry season
(November 2004–April 2005) and wet season (May 2005–October 2005) and SPI12
in the whole year (November 2004–October 2005). Figure 2.15 shows that in the dry
season of 2004–2005, except for a small part of the northern part of the basin, the SPI
of the other areas was generally smaller than −1, indicating that the precipitation
throughout the basin was apparently smaller than former years and indicating the
moderate or above meteorological drought. The SPI of the Thai-Lao border, central

2 Thedefinition of SPI and SRI please refer to Chap. 4.

3 Wang Qiyi, Ren Fumin. The latest ENSO monitoring. Date: Mar. 16, 2005; news source: Climate
System Diagnosis and Forecast Office. http://ncc.cma.gov.cn.
2 Summary of Flood and Drought … 47

Fig. 2.15 Spatial distribution of SPI6 in the dry and rainy seasons and SPI12 in the whole year in
2004–2005 in the Mekong River Basin. a November 2004–April 2005; b May 2005–October 2005;
c November 2004–October 2005

Lao PDR and the Mekong Delta Region was much smaller than −1.5, indicating
that the drought in the above areas has reached the degree of exceptional drought.
In the wet season of 2005, the drought coverage decreased significantly and the
severity declined significantly compared with the dry season. The areas with less
precipitation were mainly concentrated in northeastern Thailand, Cambodia and the
Mekong Delta in Viet Nam and other middle and lower reaches. The SPI value
of southwestern Cambodia was generally lower than −1, indicating the moderate
drought in this area. Relatively speaking, countries such as Myanmar and Lao PDR
in the upper reaches have abundant rainfall, high SPI and no drought. On the yearly
scale, the drought in November 2004–October 2005 was mainly concentrated in
the southern countries in the middle and lower reaches, including eastern Thailand,
southern Lao PDR, Cambodia and the Mekong Delta in Viet Nam and so on; in terms
of drought severity, the drought in the south was more severe than that in the north,
reaching severe drought.
Figure 2.16 further shows the spatial distribution of monthly
SPI3 in 2004–2005 dry season in the Mekong River Basin. In
November 2004, SPI was calculated based on the cumulative rainfall of Septem-
ber–November (three months) in 2004, and was analogized in other months. It is
known from Fig. 2.16 that in November 2004, the rainfall in the whole Mekong
River Basin was lower than the historical level of the same period and the SPI value
was negative, indicating that the whole basin was affected by drought. Among them,
the drought in the southern region of the basin (northeastern Thailand, southern
Lao PDR, Cambodia and Viet Nam and other countries) was particularly serious
and SPI values were lower than −1.5, indicating severe drought; in December, the
rainfall continued to be low and the drought in the Mekong River Basin continued to
aggravate, as well as the scope of severe drought continued to expand; in 2005, the
48 X. Chen et al.

Fig. 2.16 Spatial distribution of monthly SPI3 in 2004–2005 dry Season in the Mekong River
Basin (a November 2004; b December 2004; c January 2005; d February 2005; e March 2005;
f April 2005)

whole basin in the first two months (January and February) was still in a continuous
situation of less rain and drought was still severe; after entering the late dry season
(March 2005–April 2005), the basin began to rain gradually and the SPI values of the
northern parts of the basin (such as Myanmar, northern Lao PDR, etc.) returned to
positive value and the regional drought subsided; although the southern region was
still in a state of rainfall deficit, the overall drought has been effectively alleviated
and the degree of drought has gradually been weakened to a mild level.
Figure 2.17 shows the six-month SRI (SRI6) sequences of three typical hydro-
logical stations, Chiang Saen, Mukdahman and Stung Treng, in the upper, middle
and lower reaches of the mainstream Mekong River. The SRI of Chiang Saen station
in the upper reaches was lower than −0.5 in 2004–2005 and close to −2 in some
periods, indicating that the runoff of this station in 2004–2005 was much lower than
that of the same time in the previous years and the relatively severe hydrological
drought occurred; compared with Chiang Saen station, the runoff of Mukdahman
station was obviously lower in 2004–2005 dry season and the hydrological drought
reached moderate or above level; compared with two stations in the upper and middle
2 Summary of Flood and Drought … 49

Fig. 2.17 SRI (SRI6) sequences of Chiang Saen, Mukdahan and Stung Treng stations

reaches, the dry season runoff of Stung Treng station in 2004–2005 did not show
obvious anomalies, and the hydrological drought occurred mainly in the early months
of 2005.

Influence of drought in 2004–2005

In 2004–2005, drought events imposed very serious impact on Viet Nam, Cambodia
and Thailand in the basin.
In Cambodia, drought events that took place in 2004–2005 are the most serious
in recent years, affecting 14 out of the 24 provinces, reducing paddy cultivation in
all provinces and incurring a food shortage for 500,000 people.
Thailand suffered exceptionally serious droughts in 2004–2005, which affected
63 out of the 76 provinces. Across the country, 9 million people’s life was affected,
and irrigation water was restricted (and even prohibited) for assuring domestic water
supply. Its gross loss amounted to 193 million USD, and data about the Ko Kret
Plateau are absent.
The wet season of 2004 ended early, leading to the failure of autumn paddy
cultivation in Viet Nam, especially in the Cuu Long Delta where more than 104,000
hm2 of paddy was damaged. The impact was most serious in Ben Tre Province,
where about 7000 hm2 of paddy and 15,000 hm2 of fruit farms were damaged,
involving a value of 33 million USD. In addition, 82,000 households were forced
to buy drinking water (4.5 USD/m3 ). In the Cuu Long Delta, gross drought loss
amounted to 42 million USD.

2. Drought in 2016

In August 2015, the sea surface temperature anomaly over Nino 3.4 in the tropical
Pacific region (i.e. degree of deviation from the climate mean state) exceeded 2 °C
50 X. Chen et al.

(for the first time since the twenty-first Century), while in November 2015, the sea
surface temperature anomaly in this area reached 2.95 °C, exceeding the record of
1997–98 super El Nino event. During the winter (December 2015–February 2016) in
the Northern Hemisphere, the El Nino event remained at its peak until March, when
the unusually cold water on the thermocline began to surge toward the sea surface
along the coast of South America, and the sea surface temperature in the middle-east
equator Pacific Ocean began to drop sharply, and eventually returned to the neutral
state at the end of May, the El Nino event ended. According to the data of World
Meteorological Organization, the El Nino phenomenon in 2015–2016 was relatively
strong, comparable to the strongest El Nino in 1997–1998 since records began.4
Under the influence of strong El Nino, the rainfall in Viet Nam, Thailand, Myan-
mar and other countries in early 2016 was significantly less than that in previous
years. From Fig. 2.18a, it can be seen that on the one-month scale, the obvious
rainfall deficit generally occurred in the middle and lower reaches of the southern
Mekong River in April 2016 (SPI < −1) and the moderate or above meteorological
drought occurred. Especially in the southeastern pat of the basin, including southern
Lao PDR, eastern Cambodia, Viet Nam and a small part of the eastern Thailand,
their SPI values were lower than −1.5, indicating that the drought in these areas
reached the exceptional drought level in the same month. When the temporal scale
is extended to three months (Fig. 2.18b), it can be found that the drought coverage
based on three-month cumulative rainfall monitoring (February–April 2016) was
further extended to the entire Mekong Delta region, covering northeastern Thailand,
central and southern Lao PDR, Cambodia and the Mekong Delta; at the same time,

Fig. 2.18 SPI distribution in the Mekong River Basin on the one-month, three-month and six-month
scales in 2016. a SPI1: April 2016; b SPI3: February–April 2016; c SPI6: November 2015–April
2016

4 Mekong River Commission and Ministry of Water Resources of the People’s Republic of China
(2016). Technical Report – Joint Observation and Evaluation of the Emergency Water Supplement
from China to the Mekong River. Mekong River Commission, Vientiane, Lao PDR, p 12.
2 Summary of Flood and Drought … 51

the SPI value was generally lower than −1.5, indicating the severe and exceptional
droughts; the diagnostic results on the six-month scale (November 2015–April 2016)
show that the drought was mainly concentrated in northeastern Thailand and Cam-
bodia and reached severe drought level. Compared with the diagnostic results on the
three-month scale (Fig. 2.18b), there was no meteorological drought on the six-month
scale (SPI > 0) in the Mekong Delta.
Figure 2.19 further shows the spatial distribution of monthly SPI3 in the Mekong
River Basin from November 2015 to April 2016. It can be seen from Fig. that in
November 2015, the Mekong River Basin was slightly dry and only a small range of
mild drought took place in eastern Cambodia; in December 2015 and January 2016,
most of the basin was wet; in February, the drought occurred, but mainly in the area
with the border area of Thailand, Cambodia and Lao PDR as the center while most
of the northern region was still wet; in March, the drought was widened to most parts
of northern Thailand and Cambodia and generally reached the moderate level; as
rainfall continued to be low, the drought in April was further expanded and covered
all areas of the basin except northern Lao PDR and the severity of the drought reached
severe level.

Fig. 2.19 Spatial distribution of monthly SPI3 in 2015–2016 dry season in the Mekong River Basin
(a–f in Fig. is November 2015–April 2016 respectively)
52 X. Chen et al.

Based on Fig. 2.17, the hydrological drought situation of three typical hydrological
stations in 2016 was further analyzed. From the figure we can see that the SRI of
Chiang Saen station in the upper reaches of the river in 2016 was basically above
zero, indicating that the observed runoff of the station in 2016 was not lower than that
of the same period in the previous years and there was no hydrological drought; for
Mukdahan station in the middle reaches of the river, the SRI value in 2016 was lower
than −0.5, indicating that the station had a certain degree of hydrological drought,
but the drought was relatively mild. Compared with the two stations in the upper and
middle reaches of the basin, the runoff of Stung Treng station in the lower reaches in
2016 was obviously lower than that in previous years and SRI value was even close to
−2, showing a very serious hydrological drought. It can be seen that, compared with
the upstream stations, the downstream stations experienced more severe hydrological
droughts in 2016.
Combined with the above meteorological and hydrological analysis, February–
April 2016 is the most extensive and influential period of the drought event. Among
them, the sustained low rainfall in February–April triggered a certain degree of
meteorological drought in the Mekong Delta region, but far less than that in Cambodia
and Thailand in the upper reaches; however, the less rainfall and the reduction of
upstream inflow resulted in the severe hydrological drought in the Mekong Delta
region and further aggravated the seawater invasion. This series of chain reactions
has led to heavy losses caused by drought in the Mekong Delta in 2016.

Influence of drought in 2016

The beginning of 2016 saw serious drought in the Mekong River Basin, especially
in Viet Nam. According to the official media,5 Viet Nam was meeting with the most
serious drought in recent 90 years, which had affected 39 out of the 63 provinces;
saltwater intrusion started 2 months earlier in the dry delta, due to which 40–93 km of
the main channel had been intruded, 30% of the 1.15 million hm2 of winter and spring
crops were threatened, many coastal provinces would endure a long-time shortage
of water, and local agricultural production and life had been influenced severely.
Nearly one million Viet Namese were faced with a shortage of daily water, nearly
160,000 hm2 of farmland was affected, and the economic loss was estimated at 10.50
million USD.6 According to another media report,7 the Scientific Symposium on the
Environmental Impact from Hydropower Projects on the Mainstream Mekong River,
held in Can Tho Province on March 4, deems 2016 witnessed the most serious alkaline
erosion and drought in the Cuu Long River Plain in recent 100 years. The alkaline-
eroded distance was 10 and 9 km for provinces along Tien Giang and Hau Giang.
Because of climate change, aquatic product output of this region was estimated to
decline by 600,000 t/a and crop output would also decline by 224,000 t/a. Total loss
of agriculture and aquaculture was about 520 million VND (about 230 million USD),
taking up 2.3% of its GDP.

5 VN Express (March 14, 2016).

6 UN Office for the Coordination of Humanitarian Affairs (December 15, 2016).
7 Nhan Dan Newspaper (March 7, 2016).
2 Summary of Flood and Drought … 53

Other countries attacked by severe drought included Thailand and Myanmar.

In Thailand, the persistent drought for two consecutive years affected food output
seriously and also influenced the economy. Nearly 70 out of the 77 provinces went
through the most serious drought in decades, with more than 9.56 million people
affected. More than 2 million hm2 of farmland was damaged, and the economic loss
reached hundreds of millions of USD. Myanmar suffered persistent high temperature,
and agriculture and farmers there were suffering from drought.
As requested by the Vietnamese government, the Chinese government replen-
ished water emergently to the lower reaches on March 15–May 31, 2016, during
which period the Jinghong Hydropower Station increased the discharged flow from
1000 m3 /s to above 2000 m3 /s. The emergency replenishment is a spatio-temporal
regulation of water resources that was realized through water storage of upper reser-
voir in the special conditions of lower reaches, demonstrating humanitarian spirit
and positive and practical cooperation attitude of the Lancang-Mekong cooperation
mechanism

2.5 Summary

Through collecting and reorganizing data in international authoritative disaster

databases at and released officially by MRC and disaster bulletins these countries
released publicly, we preliminarily analyzed the flood loss of the Mekong River Basin
and got the following conclusions:
(1) Flood causes heavy fatalities in the basin. Flood-caused fatalities in the Mekong
River Basin were most serious, 825, in 2000, followed by 2001 (489), 2011
(396), 2013 (247) and 1996 (173). The death toll is smaller than 100 in other
years. Cambodia and the Cuu Long Delta, Viet Nam take the largest shares
while the share is small in northeastern Thailand and Lao PDR.
(2) Flood imposes significant impact on agriculture. 2000 saw the biggest agri-
cultural impact of floods, approximately 2.50 million hm2 , followed by 2011,
about 500,000 hm2 . Seen from the geographic distribution, agricultural impact
gathered in the Cuu Long River Delta, Viet Nam in 2000, and gathered mostly
in Cambodia and the Cuu Long River Delta, Viet Nam and partially in Lao PDR
in 2011. In 2013, floods mainly affected agriculture in the irrigation regions of
northeastern Thailand.
(3) Flood incurs serious economic loss in countries at the basin. Seen from the
statistical analysis of flood-caused economic loss at the basin in 1996–2014,
Cambodia ranks top by 1.4 billion USD, followed by Viet Nam by 980 million
USD and Lao PDR by 590 million USD. Thailand ranks bottom by 310 million
USD.
(4) Extreme floods take place more frequently and cause a huge loss upon the
beginning of the twenty-first century. The flood loss amounted to 1.164 billion
USD in 2011, taking up 35% of the gross flood loss in 1996–2014; to 651 million
54 X. Chen et al.

USD in 2013, taking up 20%; and to 462 million USD in 2000, taking up 14%.
The flood loss was heavy in those three years. In addition, flood losses of the
four countries were synchronous in time.
(5) Drought features extensive influence, long duration and huge economic loss.
Though drought doesn’t take place frequently, drought imposes heavy impact
on population and incurs huge economic loss in these countries, and individual
droughts feature extensive impact, long duration and huge economic loss and the
gross loss incurred thereof tends to increase significantly. The drought loss of the
Cuu Long River Delta in 2002 and May 2004 is much higher than that of the loss
of 7 out of the 9 floods in 2000–2008. In northeastern Thailand, drought-caused
paddy loss is about 10 million USD/year, very nearly the same with yearly
flood loss in the region. Besides, drought has very apparent influence on fishery
breeding at the Tonle Sap Lake, with a yearly loss of 15 million USD/year or so.
Seen from spatial distribution, drought loss is heavy in northeastern Thailand,
Cambodia and Viet Nam and relatively mild in Lao PDR.

Dr. Chen’s research primarily focuses on comprehensive management of urban rivers, river eco-
logical restoration, river sediment management, transboundary river water management and flood
management. She was the Program Manager of many important national and provincial level
projects. She has over 2 years’ research experiences in transboundary water cooperation of China.
The number of participated projects and programs exceeds 50. She has written 4 technology
standards (or specifications) and 4 books, has published over 50 papers.

Dr. Liu’s research primarily focuses on hydrological processes and modeling, flood management,
and transboundary river water cooperation. She has over 5 years’ research experiences in trans-
boundary water cooperation of China. She was a participant of various Lancang-Mekong water
resources cooperation activities. She is an editor of a joint publication by MWR of China and
MRC in 2016. She is Program Managers of 3 international cooperation projects with Mekong
River countries. She is author of over 10 journal papers and 1 book.

Dr. Baiyinbaoligao’s research primarily focuses on Eco-hydraulics, urban river, and river restora-
tion. He was the Program Manager of important national and provincial level projects. The num-
ber of investigated or participated projects and programs exceeds 70. He has written 4 technology
standards (or specifications) and 4 books, has published over 80 papers.

Dr. Mu’s research primarily focuses on water resources management, river ecosystem restoration
and hydraulic computation. He has over 10 years’ research experiences in relative national level
projects. He is author of over 40 journal papers and 4 books, as well as dozens of patents for
inventions.
Chapter 3
Analysis of Flood Character
in the Mekong River Basin

Shiyu Hou, Hui Liu, and Fuqiang Tian

Abstract Mekong River Basin (MRB) has suffered huge injuries and property losses
from frequent floods, which would very likely witness an increase of flood magni-
tude and frequency under climate change. In this chapter, we set up a distributed
hydrological model (THREW) to provide fundamental analysis of the flood char-
acteristics in the MRB, the simulation period is 1991–2016, the spatial coverage
is the whole basin except the delta region due to lack of reliable topographic data.
Two main types of flood in the MRB which are riverine flood and flash flood are
discussed. Flood peak frequency at mainstream stations along Mekong are achieved
by Pearson-III Frequency Curve Fitting. The annual flood volume and duration at
mainstream stations along Mekong River are calculated and analyzed. Taking the
flood volume of damaging floods at Stung Treng station (at lower reach of Mekong
mainstream) as subject, the THREW model is used to analyze flood’s travel time
and regional composition, which would benefit flood prevention and water resources
management from a whole-basin view.

3.1 Data and Methodology

3.1.1 Data

(1) The historical long-sequence section flow data of major hydrological stations of
Chiang Saen, Luang Prabang, Nong Khai, Nakhon Phanom, Mukdahan, Pakse,

S. Hou (B) · F. Tian

Tsinghua University, New Hydraulic Building, Beijing 100050, China
e-mail: [email protected]
F. Tian
e-mail: [email protected]
H. Liu
China Institute of Water Resources and Hydropower Research, A-1 Fuxing Road, Haidian
District, Beijing 100038, China
e-mail: [email protected]
© China Water & Power Press and Springer Nature Singapore Pte Ltd. 2020 55
H. Liu (ed.), Flood Prevention and Drought Relief in Mekong River Basin,
Springer Tracts in Civil Engineering,
https://doi.org/10.1007/978-981-15-2006-8_3
56 S. Hou et al.

Stung Treng, Kratie on mainstream Mekong River (from upper to lower reaches,
see Fig. 3.1 and Table 3.1) was collected from Mekong River Commission. The
sequence length is 1985–2016, and temporal resolution is daily.
(2) As input of hydrological modelling, meteorological data, soil data and vegeta-
tion information was collected and applied in this chapter. Including:
• Ground observation meteorological/rainfall data from Mekong River Com-
mission, including daily rainfall information from 502 rainfall stations of

Fig. 3.1 Hydrological stations along Mekong mainstream

3 Analysis of Flood Character in the Mekong … 57

Table 3.1 Main properties of hydrological stations along Mekong mainstream

No. Station Location Country Start/end year
Latitude Longitude
1 Chiang Sean 20.274 100.089 Thailand 1985–2016
2 Luang Prabang 19.893 102.134 Lao PDR 1985–2016
3 Nong Khai 17.881 102.732 Thailand 1985–2016
4 Nakhon Phanom 17.425 104.774 Thailand 1985–2016
5 Mukdahan 16.583 104.733 Thailand 1985–2016
6 Pakse 15.100 105.813 Lao PDR 1985–2016
7 Stung Treng 13.533 105.950 Cambodia 1985–2016
8 Kratie 12.481 106.018 Cambodia 2005–2016

which 32 are conventional meteorological stations offering daily meteoro-

logical elements such as barometric pressure, temperature, humidity, wind
speed and direction, sunshine duration and solar radiation etc. The period is
1991–2005.
• Satellite-based rainfall product TRMM3B42_7.01 was used for the period
1998–2016.
• Soil data was based on the global earth database provided by FAO (http://
www.fao.org/home/en/) of which the spatial resolution is 10 km.
• As to vegetation information, NDVI and LAI data products of MODIS were
used and NDVI data of Mekong Basin were downloaded from the technical
website of NASA (http://reverb.echo.nasa.gov/reverb).

3.1.2 Methodology

Recognizing there are two types of flood in the Mekong River Basin, and realizing
the fact of lacking first-hand local information on the flash flood, we made a brief
introduction of the two types of flood, and a deep analysis of the mainstream riverine
flood. For the mainstream riverine flood,

1 SatelliteTRMM, jointly developed and designed by NASA (National Aeronautics and Space
Administration) and JAXA (Japan Aerospace Exploration Agency), is mainly used to monitor and
study rainfall in tropical regions, actually covering a wide region ranging between 50° N and 50°
S on Earth. Satellite TRMM, a low earth orbit satellite at the dip angle about 35°, launched on
November 28 1997 in Japan, is the first meteorological satellite specifically used to observe rainfall
in tropics and subtropics, carrying sensors such as TMI, PR, VIRS, LIS and CERES. Of devices
carried by satellite TRMM, PR is ground breaking designed and made by Japan’s National Space
Development Agency (NASDA) and can provide 3D structure of rainstorm, capable of increasing
accurate estimation of rainfall. TRMM provides rainfall data ranging between 50° N and 50° S and
multiple time intervals in the globe, supplementing rainfall information of no-information regions
in the globe.
58 S. Hou et al.

(1) Flood peak, volume and duration

The characteristics of peak, volume and duration of the mainstream flood is analyzed
based on historical records and observed hydrological data. Pearson-III Frequency
Curve Fitting model was applied to reveal the frequency character of flood peak.
(2) Flood composition
The flood composition analysis is mainly based on THREW model (a hydrologi-
cal model developed by Tsinghua University, the introduction of the model is in
Appendix), and a statistic of flood travel time by corresponding discharge method.

3.2 Setting Up of the THREW Model

Please be noted that though aiming at setting up a hydrological model over the whole
Mekong River basin, the delta region is still missed due to flat terrain and complex
river network in the Mekong delta.

3.2.1 Forcing Data

(1) Precipitation
To extend the timespan coverage of the model as long as possible, two kinds of pre-
cipitation data were collected: information from meteorological/rainfall stations and
satellite-based rainfall products. Ground observation data come from Mekong River
Commission, including daily rainfall information from 502 rainfall stations of which
32 are conventional meteorological stations offering daily meteorological elements
such as barometric pressure, temperature, humidity, wind speed and direction, sun-
shine duration and solar radiation etc. The distribution of the rainfall stations and the
meteorological stations are shown in Fig. 3.2. The data covers the period of 1991–
2005. Based on meteorological stations, Thiessen Polygons2 was used to calculate
rainfall at each representative unit basin in the distributed model. Satellite-based
rainfall product TRMM3B42_7.0 was used for the period 1998–2016.
It is known from relevant researches that there are many south-north mountains
in Mekong Basin on both banks as a result of that the windward slope on the left

2 Thiessen Polygons is a method worked out by Holland climatologist Thiessen to calculate mean
rainfall on the basis of rainfall of discretely distributed meteorological stations, i.e. connect all
adjacent meteorological stations into triangles, draw perpendicular bisectors on each side of these
triangles, and connect the crossover points (i.e. center of circumcircle) of three sides of every triangle
to obtain a polygon. The rainfall intensity of the only meteorological station within the polygon is
used to express the rainfall intensity within the polygon region and such polygon is called Thiessen
Polygon.
3 Analysis of Flood Character in the Mekong … 59

Fig. 3.2 Distribution of meteorological stations and rainfall stations in Mekong Basin

bank is a high rainfall and runoff area. While the leeward slope on the right bank
(because of foehn effect) is a low-value area. In addition, within the basin, there are
some mountains which are not giant also become high rainfall and runoff areas, for
instance, Mount Douluo region in the lower Mekong Basin, and some giant upheaval-
like landforms, for instance, Highland Nakhon are low-value areas. In the north of
Chiang Saen, there are many high mountains and valleys, water vapor is hard to enter
for the deep valley and larger height difference, and the foehn effect, so these areas
become the low rainfall and runoff areas. The rainfall distribution obtained from the
research also reflects aforementioned spatial distribution characteristics.
60 S. Hou et al.

(2) Potential Evapotranspiration

On the basis of meteorological information about temperature, humidity and wind
speed from 32 meteorological stations, Penman-Monteith equation is used to cal-
culate the potential evapotranspiration and then are used as the forcing data of the
distributed hydrologic model THREW. The equation is:

0.408(Rn − G)
900
γ u 2 (es − ea )
ET0 = + T +273 (3.1)
 + γ (1 + 0.34u 2 )  + γ (1 + 0.34u 2 )

where, es is the saturated vapor pressure (kPa); ea is the actual vapor pressure (kPa);
 is the gradient of the curve saturated vapor pressure versus temperature (kPa/°C);
γ is the constant of hygrometer (kPa/°C); u2 is the wind speed (in m/s); Rn is the net
radiation (MJ/(m2 /d)); G is surface heat flux (MJ/(m2 /d)).
Similarly, Thiessen Polygons was used to calculate the potential evapotranspira-
tion of each representative unit basin, the results are shown in Fig. 3.3. It is observed
from the figure that the potential evapotranspiration has obvious spatial variability,
increasing gradually from north to south, and reaching more than 1200 mm/year in
south basin.

(3) Soil

Based on the global earth database provided by FAO (http://www.fao.org/home/en/)

of which the spatial resolution is 10 km, the soil thickness distribution in Mekong
Basin is shown in Fig. 3.4.
The data such as field water capacity, wilting coefficient, saturated moisture con-
tent, porosity, air entry value and pore-diameter distribution, were matched and aver-
aged for per grid unit and representative unit basin, to as soil parameter of the model
THREW.

(4) Vegetation Information

The model THREW needs two kinds of vegetation-related information: normalized

differential vegetation index (NDVI)3 and leaf area index (LAI). NDVI and LAI data
products of MODIS were downloaded from the technical website of NASA (http://
reverb.echo.nasa.gov/reverb), the NDVI data of Mekong Basin in July 2000 is shown
in Fig. 3.5, and it is seen from the figure that the whole Mekong Basin has higher
NDVI value while Chi-Mun basins in Thailand have lower NDVI value.

3 The value range of NDVI is −1 ≤ NDVI ≤ 1, where the negative means the surface is covered
with cloud, water and snow etc., high reflectance against visible light; 0 means there are rock or
bare soil etc.; positive means it is covered with vegetation, increasing with increasing coverage.
3 Analysis of Flood Character in the Mekong … 61

Fig. 3.3 Distribution of annual mean potential evapotranspiration in Mekong Basin (based on
observation data of rainfall stations in 1991–2005)

3.2.2 Calibration and Validation

Lacking of tributary observed hydrological data, the THREW model is calibrated

based on mainstream hydrological data with an upper to lower order to make sure
each section of the basin is fully calibrated. During model calibration, the Nash
efficiency coefficient (NSE) of different combinations of parameters were optimized
from 1991 to 2000, and the combination of parameters that made the best NSEs was
picked as calibration result. The validation period is 2001–2005.
62 S. Hou et al.

Fig. 3.4 Soil thickness distribution in Mekong Basin (plotted based on FAO’s global earth database)

Table 3.2 shows NSEs calculated at each stations in 1991–2005 and it is known
that the simulation results are good as a whole. 7 hydrological gauges were calibrated
and verified here, and each gauge was simulated for 15 years, and a total of 105 NSE
results were obtained. Among them, 62 NSEs exceeded 0.8, 82 NSE results were
more than 0.7, and most low NSEs are concentrated in years of small simulated
runoff. For the validation period 2001–2005, 23 out of 35 NSEs are larger than 0.7
and 30 NSEs are larger than 0.5, which means the model performance is good or
acceptable; 2 out of 35 is approximating 0.5; and the left 3 is 0.2 and less, which is
not acceptable, further discussion of these low NSEs is in Sect. 3.2.3.
3 Analysis of Flood Character in the Mekong … 63

Fig. 3.5 NDVI distribution in Mekong Basin (July 2000)

Figure 3.6 is the comparison diagram between the observed runoff and simulated
runoff at hydrologic stations in Mekong mainstream in 2000, where all NSEs are
over 0.75 of which three stations’ NSE is over 0.9, getting a good simulation result
in general.
Since it is in the period 1991–2005 that observed rainfall data at stations were
collected, to further extend the simulation period, TRMM rainfall data is used to
extend it to 2016. To verify the quality of TRMM rainfall data, the research calculates
NSE of TRMM simulated discharge using the simulated station results as reference
by comparing the discharge simulation results in the period 1998–2005 when the
64 S. Hou et al.

Table 3.2 Runoff simulation result with the model THREW on stations in Mekong main stream
(evaluated with NSE)
Chiang Luang Nongkhai Nakhon Mukdahan Pakse Stung
Saen Prabang Phanom Treng
1991 0.81 0.84 0.84 0.83 0.90 0.87 0.87
1992 0.68 −0.15 0.50 0.63 0.83 0.82 0.87
1993 0.56 0.88 0.87 0.78 0.79 0.75 0.76
1994 0.71 0.62 0.79 0.91 0.94 0.88 0.91
1995 0.79 0.84 0.90 0.92 0.91 0.80 0.86
1996 0.67 0.84 0.86 0.89 0.86 0.73 0.83
1997 0.92 0.89 0.90 0.94 0.96 0.90 0.92
1998 0.91 0.88 0.85 0.91 0.81 0.52 0.46
1999 0.93 0.74 0.76 0.80 0.76 0.48 0.72
2000 0.78 0.93 0.91 0.82 0.89 0.92 0.89
2001 0.57 0.86 0.89 0.71 0.84 0.92 0.88
2002 0.69 0.84 0.7 0.46 0.63 0.78 0.79
2003 0.00 0.89 0.81 0.20 0.65 0.79 0.73
2004 0.69 0.89 0.88 0.52 0.71 0.79 0.81
2005 0.82 0.92 0.90 −0.04 0.43 0.56 0.87

station rainfall data overlapped with TRMM rainfall data. As shown in Table 3.3
and Fig. 3.7, in 1998 and 1999 when TRMM just began operating, the simulated
results were not good, but with TRMM’s operating time increasing, the algorithm
being optimized constantly, NSEs are around 0.9 after 2000, indicating that the result
from model THREW simulation driven by TRMM rainfall data is good. Furthermore,
considering that the contribution rate of tributaries and sections to main stream floods,
instead of specific contribution discharge value, is used to analyze flood’s regional
composition, therefore, it is believed that the results from simulation that applied
TRMM rainfall data to extend the simulation discharge to 2016 are credible.

3.2.3 Discussion

(1) At the time of the calibrating, at Luang Prabang, if the NSE was high in 1991,
the NSE in 1992 was low, and vice versa. Considering that the NSE should be
raised as much as possible in most years, the balance was taken and parameters
of the 1991 NSE high was selected finally. The reason for the low NSE in 2003
at Chiang Saen and in 2005 at Nakhon Phanom may be due to the increasing
human consumption water in the relatively dry years, but there is no definite
evidence to prove it.
3 Analysis of Flood Character in the Mekong … 65

Fig. 3.6 Simulated and observed runoff NSEs in 2000

Table 3.3 NSE of Stung Treng discharge simulated with TRMM rainfall
Year 1998 1999 2000 2001 2002 2003 2004 2005
NSE 0.53 0.46 0.89 0.87 0.97 0.88 0.94 0.96

(2) Since observed gauge rainfall data was of 1991–2005, to further extend the
simulation period, TRMM rainfall data was used to extend to 2016. Though the
runoff simulation results indicate that simulation driven by TRMM rainfall data
is credible for the flood composition analysis, considering the spatial distribution
of TRMM rainfall data is different with gauge data, we found that the ratio of
rainfall on the right bank to the left bank of the Mekong is bigger than that of
66 S. Hou et al.

Fig. 3.7 Stung Treng discharge hydrographs obtained by using stations’ observation rainfall and
simulated with TRMM rainfall product

the gauged data. So the different rainfall resources might bring uncertainties,
which is not further studied here, to the results. If gauged rainfall data of recent
years could be collected and included in this study, the results would be more
credible.

3.3 Flood Types in the Mekong River Basin

Although the Mekong flood is considered beneficial and not a hazard there can
sometimes be too much water in the wrong place, causing loss of human lives and
damage. But it is important to distinguish between the types of floods causing the
hazards and understand the mechanism behind them, to be able to prepare for them
and mitigate the impacts. An obvious example is when extreme water levels in the
mainstream Mekong cause overflow of river banks and inundation of areas along
the river. Such floods have been preceded by a rising water level during a period of
days, and have thus given an alarm in advance. Given reliable weather and rainfall
forecasts this type of flood can be predicted with a high degree of confidence to make
appropriate preparations to combat it. This type of flood recedes in a few days, but can
become quite prolonged in the flood plains of Cambodia and Viet Nam. On the other
extreme, flash floods in the tributaries are caused by intense rainfall during a short
period of time, causing high rise of water levels within a few hours. They occur in
steep streams or small tributaries and the high speed of the water flow causes erosion,
land and mud slides and other damage, sometimes with human casualties. They are
not necessarily preceded by constant rainfall during a long time, but can occur at any
time during the rainy season, and they will recede rapidly. These floods are much
harder to predict, as the extreme, intense rainfall may be concentrated to a small
3 Analysis of Flood Character in the Mekong … 67

area in the watershed. It is often the case, though, that local damage of floods often
happens in connection with storms sweeping in over the Mekong Basin, bringing
rainfall that in some cases last several days. Soil conditions first become saturated
and then an intense but not necessarily an extreme rainfall event can rapidly trigger
a flood in small watersheds, as the soil cannot absorb any more of the rainfall. The
prediction of such floods calls for skills and systems especially designed for this.
Flash floods are potentially dangerous not only because water levels rise quickly,
but also because water flow velocities are high. The force of the water can cause
erosion, mudslides, uproot trees and tear away boulders, and the flow of debris
can sweep away houses and destroy bridges. The collection of debris in the river
bed should be cleared away as soon as possible to ensure that it does not block
the water flow should there arrive a new flood event. Therefore it is important that
disaster management agencies identify locations where flash floods and damages
have occurred, to provide immediate relief and aid. As flash floods are local in
character this information has in most cases to come from local authorities or village
people themselves. Recovery works, such as repairing damaged roads, bridges and
other infrastructure, usually needs broader support from national agencies.
As river valleys often are the most productive parts of the land it is natural that
there is pressure to establish settlements in these areas, despite the risk of floods.
Risk management, such as land use planning is the key to lessen the exposure to
risks, and at least public structures, such as schools, hospitals and the like should be
built in areas safe from floods.4 .
Since the flash floods only occur in the local watershed, the small area of the basin
increases the difficulty of the simulation. In addition, the data on the flash floods at
hand is limited, so it is difficult to carry out detailed analysis. Therefore, this chapter
mainly analyzes the characteristics and composition of the mainstream floods.

3.4 Peak, Volume and Duration

3.4.1 Flood Peak

Based on the available discharge data of hydrological stations along Mekong main-
stream, the annual maximum flood peak along the Mekong mainstream is illustrated
in Fig. 3.8 and Table 3.4. It could be seen, for the upper reach (Chiang Saen, Luang
Prabang and Nong Khai), the maximum flood peak occurred in 2002 and 2008; for
the middle reach, the maximum flood peak happened in 1991, 1996, 1997, 2000,
2001 and 2011 (Daily discharge data of Kratie is only available for 2005–2016,
which timespan is too short for frequency analysis, so flood character at Kratie is not
further analyzed in this section).

4 MRC (2015) Annual Mekong Flood Report 2014, Mekong River Commission, 70 pages, p 4.
68 S. Hou et al.

Fig. 3.8 Annual maximum flood peak along Mekong mainstream

Pearson-III Frequency Curve Fitting model is applied to reveal the frequency

character of flood peaks along Mekong mainstream. The results at main hydrological
stations are plotted in Figs. 3.9, 3.10, 3.11, 3.12, 3.13, 3.14, and 3.15.

3.4.2 Flood Volume and Duration

Caused by rainfall associated with the Southwest Monsoon, floods in Mekong shows
a remarkable regularity, the flood season normally starting in June and ending in early
November. But of course this seasonal pattern may be stronger or weaker, starting and
ending earlier or later and carry different water volumes, making the individual floods
having their own features. The onset of the flood season can be defined as the date
when the rising discharge of the river exceeds the long-term average annual discharge.
The end of the flood season is defined in a similar way as the date when the falling
discharge crosses the long-term average discharge. Based on daily discharge data at
mainstream stations from 1985 to 2016, the flood season duration and corresponding
volume is calculated (Tables 3.5 and 3.6). Box plots of the flood volume and flood
duration at mainstream stations are shown in Figs. 3.16 and 3.17. The flood volume
shows an increasing trend from upstream to downstream, with 54 km3 at Chiang
Saen and 306 km3 at Stung Treng. The flood duration varies among years and from
stations, with an average ranging from 128 days to 135 days.
3 Analysis of Flood Character in the Mekong … 69

Table 3.4 Annual maximum flood peak along Mekong Mainstream. Unit: m3 /s
Year Chiang Luang Nong Nakhon Mukdahan Pakse Stung
Saen Prabang Khai Phanom Treng
1985 11,700 18,200 17,200 22,700 25,900 34,000 42,878
1986 11,000 12,600 13,300 18,400 20,800 28,300 42,196
1987 7120 8760 10,900 24,000 28,200 37,900 49,020
1988 6620 11,900 12,100 20,900 24,500 28,800 30,532
1989 7590 9450 11,100 15,100 18,500 28,700 37,950
1990 11,000 15,600 16,100 23,900 28,300 34,100 49,752
1991 13,300 18,900 17,000 24,500 29,800 47,600 63,560
1992 5370 5680 7760 13,100 16,100 24,600 40,464
1993 10,100 11,000 15,100 22,900 26,000 27,900 38,356
1994 6419 13,125 16,275 29,645 29,188 34,699 55,534
1995 12,200 19,369 17,753 34,200 34,016 37,827 55,341
1996 12,500 18,732 20,500 29,400 32,400 40,342 69,803
1997 7720 18,754 18,300 31,200 31,400 41,847 66,119
1998 9810 14,119 15,900 23,500 21,700 26,797 31,763
1999 9560 18,135 17,400 24,000 25,700 30,467 52,222
2000 10,700 16,740 18,800 33,000 35,700 45,149 62,540
2001 10,700 16,644 17,500 34,200 36,800 42,319 67,552
2002 12,700 20,943 22,800 32,100 32,900 39,343 63,049
2003 6880 8960 13,200 30,300 25,800 34,159 50,908
2004 8910 12,896 15,100 29,700 31,900 38,510 50,826
2005 9580 14,145 14,900 34,500 35,700 39,560 46,751
2006 12,724 13,050 14,051 24,857 26,837 31,760 42,923
2007 9462 11,846 13,480 24,365 26,109 32,567 40,655
2008 14,395 21,841 19,595 32,683 36,005 34,099 41,171
2009 7658 10,735 12,245 22,880 25,615 29,042 45,421
2010 6590 9326 14,471 25,425 28,082 32,111 35,846
2011 6493 11,043 15,769 31,065 37,236 41,422 50,188
2012 8941 12,842 15,309 23,600 24,939 27,175 35,287
2013 7115 10,459 13,892 25,282 26,952 37,892 53,480
2014 5412 9905 12,846 23,877 26,914 36,119 51,964
2015 7936 12,346 14,775 26,506 28,674 28,747 31,743
2016 5427 13,627 15,643 23,359 24,086 29,375 39,119
70 S. Hou et al.

Fig. 3.9 Pearson-III Frequency Curve of flood peak at Chiang Saen

Fig. 3.10 Pearson-III Frequency Curve of flood peak at Luang Prabang

3 Analysis of Flood Character in the Mekong … 71

Fig. 3.11 Pearson-III Frequency Curve of flood peak at Nong Khai

Fig. 3.12 Pearson-III Frequency Curve of flood peak at Nakhon Phanom

72 S. Hou et al.

Fig. 3.13 Pearson-III Frequency Curve of flood peak at Mukdahan

Fig. 3.14 Pearson-III Frequency Curve of flood peak at Pakse

3 Analysis of Flood Character in the Mekong … 73

Fig. 3.15 Pearson-III Frequency Curve of flood peak at Stung Treng

3.5 Composition of Mainstream Flood

3.5.1 Flood Travel Time

By observing initially typical floods in recent years, i.e. the day-by-day runoff data
in 2000 and 2011, it is found that in 2000 a better corresponding relation exists for
day-by-day runoff of all stations while in 2011 all stations’ discharge hydrographs
and trends differ to some degree, suffering bigger impact of local tributary inflow,
therefore calculation of flood travel time based on 2000’s flood is mainly on the basis
of measured data of 8 hydrologic stations along the Mekong main stream, as shown
in Fig. 3.1 and Table 3.1. Using the typical floods in 2000 as subject, 8 hydrologic
stations in Mekong main stream (from upstream to downstream they are Chiang
Saen, Luang Prabang, Nong Khai, Nakhon Phanom, Mukdahan, Pakse, Stung Treng
and Kratie) are selected to calculate flood’s travel time (the water-level data are used
in calculation at the most downstream station Kratie). The full-year discharge (water
level) hydrographs of 8 hydrologic stations are shown in Fig. 3.18. It is understood
from the figure that the shapes and trends of the discharge (water level) hydrographs
of 8 stations are pretty similar, but that of 3 upstream stations is different from that of 5
downstream ones to some degree, mainly in a heavier flood process that happened in
the middle and last ten days of July during which there are two peaks in 3 upstream
stations and the 2nd peak is bigger while there is a single peak in 4 downstream
stations and water rises fast but subsides slowly.
Based on the full-year discharge hydrographs shown in Fig. 3.18, three levels
of flood processes are selected: large, medium and small, and the maximal daily
74 S. Hou et al.

Table 3.5 Annual flood volume along Mekong Mainstream. Unit: km3
Year Chiang Luang Nong Nakhon Mukdahan Pakse Stung
Saen Prabang Khai Phanom Treng
1985 77 108 123 151 181 248 327
1986 47 66 83 124 158 211 294
1987 51 59 64 91 120 169 234
1988 45 67 74 98 120 147 187
1989 50 69 81 122 152 197 254
1990 67 95 103 169 207 264 403
1991 74 107 110 142 180 257 351
1992 27 30 43 69 92 141 221
1993 56 70 92 131 145 178 239
1994 50 88 114 237 235 281 386
1995 76 97 113 213 205 244 319
1996 64 96 112 191 179 250 366
1997 49 85 95 195 195 250 340
1998 56 76 81 127 119 139 166
1999 61 99 114 206 207 266 371
2000 77 104 129 270 234 340 483
2001 84 124 145 292 277 329 455
2002 59 97 128 257 258 325 433
2003 41 52 67 153 141 174 271
2004 56 83 102 219 201 233 336
2005 47 70 82 251 230 269 321
2006 59 70 98 177 191 232 305
2007 61 73 88 174 189 228 282
2008 83 119 144 270 286 315 341
2009 44 59 75 160 168 216 284
2010 44 58 85 158 169 200 204
2011 46 81 118 252 275 337 406
2012 39 61 74 153 163 186 230
2013 44 73 92 178 190 251 304
2014 34 55 72 156 170 213 276
2015 27 46 60 136 144 160 179
2016 23 51 70 142 147 185 226
Average 54 78 95 177 185 232 306
3 Analysis of Flood Character in the Mekong … 75

Table 3.6 Annual flood duration along Mekong Mainstream. Unit: day
Year Chiang Luang Nong Nakhon Mukdahan Pakse Stung
Saen Prabang Khai Phanom Treng
1985 176 173 167 122 132 139 136
1986 122 119 127 117 135 140 136
1987 134 113 93 81 98 111 112
1988 125 122 116 89 98 102 106
1989 137 133 127 120 130 127 113
1990 170 165 157 146 152 155 165
1991 155 154 147 113 129 138 131
1992 98 81 93 81 92 97 110
1993 135 130 128 106 107 107 108
1994 138 139 140 143 142 140 138
1995 173 128 122 127 125 127 127
1996 141 136 121 125 104 129 132
1997 124 123 118 120 118 118 118
1998 118 105 99 97 94 95 97
1999 147 145 153 171 164 173 174
2000 162 150 159 173 148 168 170
2001 182 182 182 177 170 167 166
2002 132 127 136 148 141 152 149
2003 117 106 108 114 113 107 119
2004 132 127 139 149 130 117 130
2005 134 124 112 153 128 123 122
2006 154 118 133 124 125 127 127
2007 134 118 116 133 134 133 132
2008 174 168 168 171 170 174 165
2009 116 109 111 123 123 133 126
2010 126 110 113 110 111 114 104
2011 128 139 138 138 137 152 152
2012 106 106 103 129 128 126 119
2013 142 144 139 138 138 149 132
2014 114 108 105 118 122 122 119
2015 80 86 91 100 100 100 98
2016 78 134 136 142 143 159 157
Averagee 135 129 128 128 128 132 131
76 S. Hou et al.

Fig. 3.16 Box plot of annual flood volume at Mekong mainstream stations (1985–2016)

Fig. 3.17 Box plot of annual flood duration at Mekong mainstream stations (1985–2016)

discharge of each process is used as the peak discharge of the flood process to deter-
mine corresponding discharge and estimate the flood’s travel time. Taking Stung
Treng station as example, three flood processes occurred in the last ten days of May,
the middle/last ten days of June and the first ten days of September, and the flood
process that occurred in the last ten days of July is not selected as corresponding
3 Analysis of Flood Character in the Mekong … 77

Fig. 3.18 Discharge (water level) hydrographs of hydrologic stations in Mekong main stream in
2000

discharge because there are obvious difference in discharge between different sta-
tions and its discharge level is pretty close to that of flood process #3. Table 3.7
displays the peak discharges of three flood processes of all hydrologic stations and
the correlational relation in corresponding discharge between adjacent stations, and
their correlational relations are shown in Fig. 3.19.
It is understood from Table 3.7 that there are good corresponding relation in 3
flood-peak discharges between stations which can be used as corresponding discharge
to estimate the flood’s travel time. Furthermore, there is also good correlational
relation between the highest water level at Kratie and the flood-peak discharge at
Stung Treng (Zdown = 0.0002 ∗ Qup + 10.64, R 2 = 0.9697) and therefore the water
level is considered to be corresponding to the flood-peak discharges of upstream
stations and can be used to estimate the travel time of the section Stung Treng–Kratie.
On the basis of the flood processes and the flood-peak discharged determined
with aforementioned methods, the flood’s travel time between two stations can be
determined by comparing the time difference of corresponding discharges between
two stations. Table 3.8 and Fig. 3.20 show the time spent by flood peaks of each
discharge level traveling to hydrologic stations along the river. Because the obtained
discharge data sequences are on daily scale, the temporal resolution in analysis of
flood’s travel time can be day only.
It is known from Table 3.8 that floods at three levels of flood-peak discharges take
5, 10 and 11 days respectively to travel from Chiang Saen station to Kratie station.
 78

Table 3.7 Peak discharge of hydrologic stations in Mekong main stream and the correlational relation in corresponding discharge between adjacent stations in
2000
Station Flood peak 1 (m3 /s) Flood peak 2 (m3 /s) Flood peak 3 (m3 /s) Correlational relation R2
Chiang Saen 4350 5860 10,700
Luang Prabang 5443 8090 17,714 Qdown = 1.95 ∗ Qup − 3149 0.9995
Nongkhai 6620 9400 18,800 Qdown = 0.99 ∗ Qup + 1313 0.9998
Nakhon Phanom 11,500 19,400 33,000 Qdown = 1.68 ∗ Qup + 1753 0.9770
Mukdahan 10,100 17,800 35,700 Qdown = 1.20 ∗ Qup − 4454 0.9945
Pakse 13,415 23,550 45,149 Qdown = 1.23 ∗ Qup + 1215 0.9996
Stung Treng 18,201 32,356 62,540 Qdown = 1.40 ∗ Qup − 546 1.0000
S. Hou et al.
3 Analysis of Flood Character in the Mekong … 79

Fig. 3.19 Peak discharge of hydrologic stations in Mekong main stream and the correlational
relation in corresponding discharge between adjacent stations in 2000

Table 3.8 Flood’s travel time in Mekong main stream in 2000 (start point: Chiang Saen station)
Station River length to Flood peak 1 Flood peak 2 Flood peak 3
Chiang Saen Travel time Travel time Travel time
station (km) (day) (day) (day)
Luang Prabang 354 2 3 0
Nongkhai 815 4 5 3
Nakhon Phanom 1143 6 10 8
Mukdahan 1236 6 10 9
Pakse 1498 3a 12 8a
Stung Treng 1681 3a 9a 10
Kratie 1804 5 10 11
a The flood peak at the downstream hydrologic station occurs earlier than its upstream counterpart

But the calculated results indicate that peaks at some downstream station occurred
earlier than their upstream counterpart, possibly because inflow of local tributaries,
which will lead to error in calculation of flood’s travel time.
It is known from Fig. 3.20 that in the first 4 sections (from Chiang Saen to Mukda-
han) there are good correlational relations between the flood’s travel time and travel
distance, so it is believed that the changes of downstream discharge are mainly caused
80 S. Hou et al.

Fig. 3.20 Relation of flood’s travel time versus travel distance

by water from upstream within the sections, affected smaller by inflow of tributary.
However, the travel time of flood peak #1 is far smaller than that of flood peak #2 and
#3 in the section Nong Khai–Nakhon Phanom, possibly because of being affected by
local tributary, resulting in error in the relation of travel time versus travel distance.
Therefore flood peak #1 is used to verify corresponding discharges for flood peak #2
and #3 (Table 3.7 and Fig. 3.19) only and not to further estimate the travel time. On
the basis of the correlational relations between the travel time and the travel distance
of flood peak #2 and #3 at river sections above Mukdahan, the time spent by flood to
travel to the most downstream station Kratie and the travel speed of the flood wave
are estimated, with their results shown in Table 3.9.
It is known from Table 3.9 that by using the upstream sections with good corre-
lational relation of travel time versus distance in estimation, the time spent by flood
peak #2 and #3 to travel from Chiang Saen to Kratie is 14.9 and 14.5 days respectively
and the travel speed of flood wave is 1.40 m/s and 1.44 m/s respectively. According
to the report Joint Assessment of Results of China’s Emergency Water Make-up to
Mekong released by Mekong Commission, during the emergency water make-up of
China to Mekong, it takes 17 days for the making-up water (flow rate of net make-up
water is about 1000 m3 /s) to travel from Chiang Saen to Kratie, close to the result

Table 3.9 Correlational relation of flood’s travel time versus travel distance
Flood process Time–distance relation Number of days spent Ravel speed of flood
to arrive at Kratie wave (m/s)
Flood peak 2 T = 0.0086L − 0.6136 14.9 1.40
 2 
R = 0.9881
Flood peak 3 T = 0.0104L − 4.2551 14.5 1.44
 2 
R = 0.9609
3 Analysis of Flood Character in the Mekong … 81

in this report. According to the result, the bigger the flow rate is, the faster the flood
travels, which is consistent with the regularity of flood travel.

3.5.2 Flood Composition

Distributed hydrological model (THREW) was applied to simulate the discharge

processes of various cross-sections and tributaries, and based on continuous simula-
tion of long sequence, the annual maxima method and the super-quantitative method
are combined to select a series floods to analyze regularities on flood’s regional
composition.
While selecting floods, we mainly focus on floods happened at the flatter down-
stream section (modelled hydrograph at Stung Treng was used), and those that would
possibly cause huge damages. It is also needed to determine whether flood happens
and the start/end time of floods. Due to limited data, based on descriptions on flood
disaster, with water level information considered, corresponding flood time periods
and discharges are found out to select the heaviest floods in the period 1991–2016.
The warning water level at Stung Treng is 10.7m5 (red line in Fig. 3.21), and the
corresponding discharge is determined as 40,000 m3 /s (both flood peaks correspond
to different discharges, use the lower one 40,000 m3 /s), and based on this, floods
are selected: if the flood peak is more than 40,000 m3 /s, it is considered a flood

Fig. 3.21 Water level and discharge hydrograph at station Stung Treng in 2000

5 Obtained from MRC report Seasonal Flood Situation Report for the Lower Mekong River Basin—

Covering period from 1st June to 13rd November 2011.

82 S. Hou et al.

happened. Mekong Commission’s report 20 Years of Cooperation stated “in 2000,

the largest flood happened in Mekong downstream 20 km distance from Phnom Penh
in a few decades, which inundated villages around it and lasted 3 months.” The range
indicated with blue arrow in Fig. 3.21 is 98 days, consistent with 3 months described,
and therefore 30,000 m3 /s is selected as threshold of the flood.
Based on the discharge hydrographs at Stung Treng station in 1991–2016 and
aforementioned standards, a flood is recognized every year and its start/end date is
determined, of which the flood peaks in 1992, 1993, 2003, 2007, 2009, 2010, 2012
and 2015 were smaller than 40,000 m3 /s but their flood start/end dates are determined
by using the same standard as the heaviest flood of the year, thus there is a flood
every year from 1991 to 2016, 26 floods in total (Fig. 3.22; Table 3.10).

Fig. 3.22 Discharge hydrographs at Stung Treng station in 1991–2016 (results from simulation
with the model THREW)
Table 3.10 Typical flood information in 1991–2016
No. Year Start date End date Duration (day) Total flood volume (km3 ) Average discharge (m3 /s) Remarks
1 1991 July 25 October 16 84 266 36,680
2 1992 August 8 September 9 33 95 33,342 The flood peak smaller than
40,000 m3 /s
3 1993 July 5 September 27 85 234 31,914 The flood peak smaller than
40,000 m3 /s
4 1994 July 9 October 9 93 312 38,873
5 1995 July 19 October 2 76 278 42,328
6 1996 July 24 October 19 88 329 43,228
7 1997 July 16 October 11 88 306 40,236
8 1998 August 13 September 13 32 89 32,192
9 1999 July 26 October 11 78 274 40,653
3 Analysis of Flood Character in the Mekong …

10 2000 July 2 September 25 86 282 37,896

11 2001 August 4 October 1 59 209 40,984
12 2002 July 9 September 29 83 243 33,831
13 2003 August 26 September 22 28 71 29,287 The flood peak smaller than
40,000 m3 /s
14 2004 August 8 September 26 50 138 31,881
15 2005 July 28 September 24 59 168 32,888
16 2006 August 10 September 27 49 139 32,882
17 2007 July 19 October 19 93 269 33,445 The flood peak smaller than
40,000 m3 /s
18 2008 August 7 October 20 75 199 30,775
(continued)
83
Table 3.10 (continued)
84

No. Year Start date End date Duration (day) Total flood volume (km3 ) Average discharge (m3 /s) Remarks
19 2009 July 26 September 24 61 161 30,507 The flood peak smaller than
40,000 m3 /s
20 2010 August 31 September 7 8 21 30,461 The flood peak smaller than
40,000 m3 /s
21 2011 August 2 October 11 71 226 36,884
22 2012 August 26 September 18 24 65 31,424 The flood peak smaller than
40,000 m3 /s
23 2013 July 28 October 7 72 206 33,089
24 2014 July 24 August 17 25 87 40,307
25 2015 July 29 September 10 44 107 28,124 The flood peak smaller than
40,000 m3 /s
26 2016 August 21 October 3 44 116 30,459
S. Hou et al.
3 Analysis of Flood Character in the Mekong … 85

Table 3.11 Flood evolution time of main-stream stations and tributaries (the hydrologic stations
start at Jinghong, the tributaries start at Ziqu)
Station Corresponding Tributary Corresponding Tributary Corresponding
time (day) time (day) time (day)
Jinghong 0 Ziqu River 0 Se Bang 17
Hieng
River
Chiang 2 Angqu River 0 Mun River 17
Saen
Luang 4 Yangbi River 4 Se Done 17
Prabang River
Nongkhai 8 Nanlei River 7 Sekong 19
River
Nakhon 11 Nam Ou 11 Se San 19
Phanom River River
Pakse 14 Nam Ngum 13 Sre Pok 19
River River
Stung 15 Nam Theun 14
Treng River
Kratie 16 Songkhram 15
River

The flood travel speed fitting formula provided in the Sect. 3.5.1 is used to reckon
the flood evolution time (see Table 3.11), and the model simulation discharge process
is combined to calculate the contribution rate of tributary basin and local basins of
hydrologic stations to Stung Treng flood volume. Figures 3.23, 3.24, 3.25 and 3.26
respectively display the contribution rates of tributaries and station-related sections
to Stung Treng flood volume in 26 floods, and detailed information are present in
Tables 3.12 and 3.13. It is evident that, of tributaries, Mun River, Se Done River
and Tonle Sap River’s flood contribution rates are maximal, and of sectional basins,
Nong Khai–Nakhon Phanom, Pakse-Stung Treng, Nakhon Phanom-Pakse and Chi-
ang Saen-Luang Prabang’s contribution rates are maximal. Considering Fig. 3.27, it
is known that the high runoff-generation contribution rate of Mun River and Tonle
Sap River is due to large basin area and the high flood contribution rate of a few
tributaries like Segong River is due to large rainfall intensity within their basins.
For all rivers, the discharge reaches maximum in summer in general except Sre
Pok River and Tonle Sap River in the downstream of Mekong of which the dis-
charge in autumn is more than in summer because monsoon retreats from north to
south gradually. For tributaries near the upstream, water volume mostly reaches the
maximum in August, for those near the downstream, water volume mostly reaches
the maximum in September, while water volume of Tonle Sap River reaches max-
imum in October, so there are 3 months’ peak staggering from the upstream to the
downstream.
86 S. Hou et al.

Fig. 3.23 Contribution rate of main tributaries in Mekong to Stung Treng flood volume (the results
from simulation with the model THREW driven by the observed rainfall data at rainfall station from
1991 to 2005)

Fig. 3.24 Contribution rate of main tributaries in Mekong to Stung Treng flood volume (the results
from simulation with the model THREW driven by TRMM rainfall products)
3 Analysis of Flood Character in the Mekong … 87

Fig. 3.25 Contribution rate of sectional basins of Mekong to Stung Treng flood volume (the results
from simulation with the model THREW driven by observed rainfall data at rainfall station)

3.6 Conclusion

(1) There are two main types of flood in the Mekong River Basin. For the riverine
flood, it could be predicted and effectively prevented from thereby damages
with proper measures. For the flash flood, it is usually local and hard to predict.
It is important for the local agencies to develop monitoring and early warning
system, as well as carry out immediate relief and aid.
(2) Based on yearly flood peak analysis of 1985–2016 using observed discharge
data of hydrological stations, it shows that the maximum flood peak happened
in 2002 and 2008 at the upper reach of Mekong River Basin (Chiang Saen, Luang
Prabang and Nong Khai), with an extreme peak of around 20,000 cumec; while
that of the middle reach occurred in 1991, 1996, 1997, 2000, 2001 and 2011,
with an extreme peak of 70,000 cumec at Stung Treng. Flood peak frequency at
mainstream stations along Mekonog are also achieved by Pearson-III Frequency
Curve Fitting.
(3) By defining the onset and withdrawal date of flood season, the annual flood
volume and duration at mainstream stations along Mekong River was calculated
88 S. Hou et al.

Fig. 3.26 Contribution rate of setional basins of Mekong to Stung Treng flood volume (the results
from simulation with the model THREW driven by TRMM rainfall products)

and analyzed. The flood volume shows an increasing trend from upper to lower
reaches, with 54 km3 at Chiang Saen and 306 km3 at Stung Treng. The flood
duration varies among years and stations, with average ranging from 128 days
to 135 days.
(4) On the basis of daily runoff data of hydrologic stations in Mekong mainstream
in 2000, it takes 14.9 and 14.5 days for flood peaks at two discharge levels of
5860 m3 /s and 10,700 m3 /s to travel from Chiang Saen station to Kratie station,
close to the travel time for China’s emergency water supplement to Mekong in
2016.
(5) Taking the flood volume of damaging floods at Stung Treng station as sub-
ject, the model THREW is used to analyze flood’s regional composition in
main stream and the results indicate: of main tributaries, Mun River, Sekong
River and Nam Ngum’s contribution rates to Stung Treng total flood volume are
highest, 15.9%, 8.6% and 6.0% respectively; of sectional basins, Nong Khai-
Nakhon Phanom, Pakse-Stung Treng, Nakhon Phanom-Pakse’s contribution
rate are highest, 25.0%, 21.3% and 19.6% respectively.
Table 3.12 Contribution rate of main tributaries in Mekong to Stung Treng flood volume (the results from simulation with the model THREW from 1991 to
2016)
Flood North of Nanlei Nam Nam Nam Songkhram Se Bang Mun Se Sekong Se Srepok
Jinghong River Ou Ngum Theun River Hieng River Done San
1 5.4 2.5 3.9 6.2 5.3 2.2 4.8 15.1 2.2 9.2 4.0 4.6
2 6.2 1.0 3.5 6.9 4.0 3.3 6.1 16.6 2.6 17.2 7.8 4.7
3 5.2 1.4 4.9 8.3 8.2 2.7 3.2 10.6 1.3 10.7 5.9 3.6
4 7.2 1.7 4.7 6.0 6.7 3.4 3.8 9.7 2.3 11.0 5.2 4.1
5 6.1 1.9 4.9 7.3 5.0 3.0 3.7 13.1 1.5 9.9 5.0 2.9
6 3.4 1.6 4.3 7.5 6.2 2.8 4.9 13.7 1.7 10.3 5.9 4.5
7 6.9 1.7 3.8 5.4 5.3 3.1 4.3 14.2 1.7 9.3 5.4 5.9
8 8.7 3.4 4.9 8.3 3.4 1.5 2.5 14.8 0.7 7.2 3.0 3.8
9 5.9 2.1 4.3 4.5 2.9 3.0 4.8 12.3 1.8 11.1 5.5 3.9
3 Analysis of Flood Character in the Mekong …

10 5.1 2.2 5.0 8.8 5.2 2.1 5.3 13.1 2.1 11.3 6.0 4.0
11 4.1 1.9 4.9 6.9 5.3 2.5 6.8 11.8 1.9 7.8 3.9 5.1
12 6.9 1.7 6.1 5.6 5.8 2.6 4.7 10.5 1.8 10.4 6.0 4.1
13 6.8 1.8 7.0 6.0 4.1 2.8 4.2 14.3 2.5 8.5 4.5 4.4
14 4.4 2.0 4.2 8.3 6.4 2.4 4.6 10.2 2.0 11.0 6.1 4.1
15 5.4 3.5 4.4 4.7 6.0 2.1 4.6 7.9 1.3 14.6 7.7 6.3
16 5.7 3.1 3.9 4.3 6.1 4.5 6.1 21.8 2.2 6.0 3.4 4.9
17 5.1 3.0 4.2 4.5 5.8 3.5 4.6 23.6 2.3 7.6 4.7 6.7
18 7.0 3.2 3.6 4.5 5.5 3.6 5.4 25.2 1.8 5.9 3.9 5.6
19 5.6 3.4 5.4 5.8 6.4 4.1 5.2 14.3 1.2 5.0 3.1 3.6
20 3.6 2.4 2.6 4.5 5.9 3.8 6.3 24.4 1.4 3.4 2.0 3.7
(continued)
89
Table 3.12 (continued)
90

Flood North of Nanlei Nam Nam Nam Songkhram Se Bang Mun Se Sekong Se Srepok
Jinghong River Ou Ngum Theun River Hieng River Done San
21 4.5 2.3 4.1 5.5 6.5 4.0 5.7 22.9 1.3 5.6 3.0 4.5
22 4.3 3.3 5.3 5.3 5.2 3.4 4.8 19.4 0.9 6.0 3.3 4.8
23 6.2 3.5 5.5 5.4 5.2 3.0 3.8 19.5 1.1 5.9 3.2 4.3
24 5.1 2.4 2.6 5.0 6.5 3.5 3.6 27.0 2.2 8.5 4.3 5.6
25 6.6 4.1 6.0 6.4 8.8 3.9 3.9 11.5 0.6 3.0 2.2 2.6
26 6.6 3.6 5.9 5.2 5.8 3.9 4.8 16.8 0.7 6.1 3.9 4.2
Mean 5.7 2.5 4.6 6.0 5.7 3.1 4.7 15.9 1.7 8.6 4.6 4.5
Min 3.4 1.0 2.6 4.3 2.9 1.5 2.5 7.9 0.6 3.0 2.0 2.6
Max 8.7 4.1 7.0 8.8 8.8 4.5 6.8 27.0 2.6 17.2 7.8 6.7
S. Hou et al.
Table 3.13 Contribution rate of sectional basins of Mekong to Stung Treng flood volume (the results from simulation with the model THREW from 1991 to
2016)
Flood North of Jinghong–Chiang Chiang Luang Nongkhai–Nakhon Nakhon Pakse–Stung
Jinghong Saen Saen–Luang Prabang–Nongkhai Phanom Phanom–Pakse Treng
Prabang
1 4.8 8.7 14.4 3.3 21.5 27.7 19.7
2 5.4 8.2 12.5 4.1 24.2 21.7 24
3 4.8 7.7 17.9 5.0 41.4 −2.6a 25.9
4 6.7 17.7 21.5 7.3 29.4 −1.0a 18.3
5 5.5 8.8 12.6 2.2 18.8 20.7 31.5
6 3.1 7.5 14.6 3.6 29.4 10.2 31.6
7 6.2 8.3 14.2 3.0 21.3 23.2 23.9
8 7.6 11.9 16.2 4.8 29.4 6.9 23.3
3 Analysis of Flood Character in the Mekong …

9 5.5 11.0 17.4 4.0 30.6 7.2 24.2

10 4.7 9.1 20.2 4.8 32.8 3.4 25
11 3.9 9.5 17.4 5.7 28.8 9.3 25.3
12 6.6 11.7 18.2 5.4 24.4 7.5 26.2
13 6.5 6.0 14.6 3.3 20.6 27.2 21.8
14 4.1 10.3 22.0 7.4 32.8 1.2 22.2
15 5.1 7.6 17.3 5.4 31.6 9.4 23.7
16 5.0 7.7 10.7 2.8 20.3 35.2 18.4
17 4.2 7.5 12.6 3.4 18.0 31.5 22.8
18 6.0 7.8 11.0 3.2 18.1 34.6 19.3
19 5.1 9.3 17.0 5.9 22.3 25.7 14.7
(continued)
91
Table 3.13 (continued)
92

Flood North of Jinghong–Chiang Chiang Luang Nongkhai–Nakhon Nakhon Pakse–Stung

Jinghong Saen Saen–Luang Prabang–Nongkhai Phanom Phanom–Pakse Treng
Prabang
20 3.4 7.6 10.4 4.4 24.8 38.4 10.9
21 4.0 6.1 11.8 4.9 22.1 34.3 16.8
22 3.8 8.2 15.9 5.9 18.7 28.8 18.6
23 5.7 9.1 15.8 6.4 19.2 27.0 16.8
24 4.4 6.1 7.2 4.5 22.0 34.5 21.3
25 6.2 10.3 16.5 7.1 29.2 21.0 9.8
26 5.8 9.3 16.7 5.0 19.3 25.3 18.6
Mean 5.2 9.0 15.3 4.7 25.0 19.6 21.3
Min 3.1 6.0 7.2 2.2 18.0 −2.6a 9.8
Max 7.6 17.7 22.0 7.4 41.4 38.4 31.6
a Thecontribution rate of section is negative. The fitting formula in 3.3 is used to calculate the contribution rate, and the flood evolution time is considered,
which may result in the discharge at the downstream smaller than that at the upstream
S. Hou et al.
3 Analysis of Flood Character in the Mekong … 93

Fig. 3.27 Drainage area of main tributaries

Ms. Hou’s research primarily focuses on hydrological processes, hydrological predictions and
modelling, and flood management. She is in Department of Hydraulic Engineering of Tsinghua
University. She has published over 1 journal papers.

Dr. Liu’s research primarily focuses on hydrological processes and modeling, flood management,
and transboundary river water cooperation. She has over 5 years’ research experiences in trans-
boundary water cooperation of China. She was a participant of various Lancang-Mekong water
resources cooperation activities. She is an editor of a joint publication by MWR of China and
MRC in 2016. She is Program Managers of 3 international cooperation projects with Mekong
River countries. She is author of over 10 journal papers and 1 book.

Dr. Tian’s research primarily focuses on hydrological processes and modeling, flood manage-
ment, agricultural water management, and transboundary river water management. He is editor of
two journals of Hydrology and Earth System Sciences and Journal of Hydrology. He is the Chair
of Panta Rhei Initiative (the IAHS Scientific Decade 2013–2022) for the fourth biennium (2019–
2020), and also the chair of Biofuel Working Group of ICID from 2015 to the present. He has
published over 150 journal papers, 3 books and 1 textbook.
Chapter 4
Analysis of Drought Character
in the Mekong River Basin

Xuejun Zhang and Hui Liu

Abstract Drought is among the most costly natural disasters in the Mekong River
Basin (MRB). To understand the spatial and temporal characteristics of drought can
largely facilitate scientific drought management and risk mitigation. Using a set of
long-term (1901–2016) global monthly precipitation data and nearly 30-year (1985–
2016) in situ daily observed streamflow, this chapter investigates the long-term trend
and inter-annual variability of meteorological and hydrological droughts in MRB,
respectively, through estimations of multi-scale Standardized Precipitation Index
(SPI) and Standardized Runoff Index (SRI). Results indicate that while with slight
upward trend from basin-scale perspective, the SPI is found with obvious downward
trend over the northeastern Thailand, most of Cambodia and Myanmar, suggesting
these regions are overall subject to intensified drying during the past half-century.
The occurrence frequency of drought is over 25% across much of MRB, particularly
for southern Cambodia and Mekong delta where the occurrence of extreme drought
is around 10%. The hydrological drought analysis show that the trend of SRI greatly
varies by in situ station location over the past 30 years. The 12-month SRI (SRI-12)
for the Upper station (Chiang Saen) exhibits obvious decreasing trend, and mostly
falls into the negative range, suggesting the recent frequent below-normal streamflow
in Chiang Saen station. The occurrence of hydrological drought in the Middle portion
is found with downward trend, as indicated with the long-term increasing SRI-12 at
NongKhai and Mukdahan stations. When it comes to the downstream station (Stung
Treng), while the trend of SRI seems little change, the SRI-12 value is mostly below
zero since 2005, suggesting that this portion is susceptible to hydrological droughts
recently. These analysis provide us an overview insight into changes in meteorologi-
cal and hydrological droughts, and advance our understanding on drought variations
and its long-term trend in MRB.

X. Zhang (B) · H. Liu

China Institute of Water Resources and Hydropower Research, A-1 Fuxing Road, Haidian
District, Beijing 100038, China
e-mail: [email protected]
H. Liu
e-mail: [email protected]

© China Water & Power Press and Springer Nature Singapore Pte Ltd. 2020 95
H. Liu (ed.), Flood Prevention and Drought Relief in Mekong River Basin,
Springer Tracts in Civil Engineering,
https://doi.org/10.1007/978-981-15-2006-8_4
96 X. Zhang and H. Liu

4.1 Data and Methodology

4.1.1 Data

(1) Historical long-sequence rainfall data of the Mekong River Basin in the recent
100 years (1901–2016) was collected and compiled on the basis of the CRU
(Climatic Research Unit) global precipitation product. It is a full-coverage,
high-resolution and no-omission data set of average monthly surface climatic
elements rebuilt by the U.K. University of East Anglia’s Climatic Research Unit
(CRU) through integrating existing famous databases.1 The current sequence
length is 1901–2016, spatial resolution is 0.5° × 0.5°, and all land networks in
the world are covered.
(2) The historical long-sequence section flow data of major hydrological stations
of Chiang Saen, Luang Prabang, Nong Khai, Nakhon Phanom, Mukdahan,
Pakse, Stung Treng, Kratie on mainstream Mekong River (from upper to lower
reaches, see Fig. 3.1 and Table 3.1) was collected from Mekong River Commis-
sion. The sequence length is 1985–2016, and temporal resolution is daily-scale
observation.

4.1.2 Methodology

In this project, we established the standardized precipitation index (SPI) and stan-
dardized runoff index (SRI) as indicators for monitoring and diagnosis of differ-
ent types of drought and analyze the characteristics of drought, on different scales
and of different types, in the Mekong River, from meteorological and hydrological
perspectives.
(1) Definition and calculation of SPI
Generally speaking, precipitation abides by skewed distribution rather than normal
distribution. In drought monitoring and evaluation,  distribution probability is usu-
ally adopted to describe the variation of precipitation. The standardized precipitation
index (SPI), for measuring the excess and deficit of precipitation on various temporal
scales, is a widely adopted index for drought diagnosis.  distribution probability is
adopted to describe precipitation in the SPI calculation; then, normal standardiza-
tion of skewed probability distribution is conducted; finally, drought is graded using
the distribution of cumulative frequency of standardized precipitation. The SPI is an
indicator expressing the precipitation occurrence probability in a given period that
is applicable to meteorological drought monitoring and evaluation on or above the
monthly scale. With the advantages of easy access to data, easy calculation, flexi-

1 HarrisI, Jones PD, Osborn TJ, et al. Updated high-resolution grids of monthly climatic
observations-the CRU TS3.10 dataset. International Journal of Climatology, 2014, 34(3): 623–642.
4 Analysis of Drought Character in the Mekong … 97

Table 4.1 SPI-based drought

Grade Type SPI
classification
I No drought >− 0.5
II Mild drought (−1.0, −0.5)
III Moderate drought (−1.5, −1.0)
IV Severe drought (−2.0, −1.5)
V Exceptional drought ≤− 2.0

ble temporal scale and regional comparability, SPI has been widely applied to the
depiction of meteorological drought in recent years. SPI formula is2 :
 
(c2 t + c1 )t + c0
SPI = S t − (4.1)
[(d3 t + d2 )t + d1 ]t + 1.0

1
t = ln (4.2)
G(x)2

In specific, x is precipitation sample value; S is the positive and negative coefficients

of probability density; c0 , c1 , c2 and d1 , d2 , d3 are calculation parameters of the
simplified approximation analysis formula for converting  distribution probability
into cumulative frequency—c0 = 2.515517, c1 = 0.802853, c2 = 0.010328, d1 =
1.432788, d2 = 0.189269 and d3 = 0.001308. G(x) is rainfall distribution probability
related to  function. According to the probability density integral formula of 
function is:
x
2
G(x) = γ xγ −1 e−x/β dx, x > 0 (4.3)
β Γ (γ0 )
0

When G(x) > 0.5, S = 1; when G(x) ≤ 0.5, S = − 1.

Drought is graded with SPI ≤ − 0.5 being the standard for judging the occurrence
of drought. Table 4.1 shows the drought gradation based on SPI.
(2) Definition and calculation of SRI
By reference to the calculation principle of SPI, the Standardized Runoff Index (SRI)
was proposed. Based on long-sequence measurement or simulated monthly runoff
calculation, SRI measures effectively runoff deficit relative to multi-year average
runoff, expresses the probability of occurrence of the cross-section runoff of a given
period in the same period in history, and is used in hydrological drought diagnosis
and evaluation on and above monthly scale.

2 McKee T, Doesken N, Kleist J. The relationship of drought frequency and duration to time scales.

In Proceedings of the 8th Conference on Applied Climatology, Anaheim, CA, USA, 17–22 January
1993; American Meteorological Society: Boston, MA, USA, 1993; pp 179–184.
98 X. Zhang and H. Liu

Similar with the SPI-based gradation of drought, drought is also graded with SRI
≤ − 0.5 being the standard of judging hydrological drought, as shown in Table 4.1.

4.2 Character Analysis of Drought

4.2.1 Character of Meteorological Drought

Based on the long-sequence monthly precipitation data for 1901–2016 from CRU,
The one-month, three-month, six-month and twelve-month standardize precipita-
tion indexes (SPI1, SPI3, SPI6 and SPI12) was respectively calculated to depict the
short-term and long-term meteorological drought in the Mekong River Basin. Based
on the SPI calculation results on various temporal scales, we revealed the spatio-
temporal characteristics of meteorological drought in the Mekong River Basin from
the perspectives of drought severity and drought frequency.
(1) Drought severity

1) Inter-annual variation of SPI

Figure 4.1 demonstrates the inter-annual variation characteristics of SPI on different

temporal scales between 1950 and 2016. In specific, Fig. 4.1a–d show the one-
month, three-month, six-month and twelve-month SPI values. Different temporal
scales demonstrate great differences in the degree of drought. For instance, SPI1 of
the Mekong River Basin in March, 2013 is −1.80, indicating severe drought; SPI3
is −1.43, indicating moderate drought; SPI6 is −0.76, indicating mild drought, and
SPI12 is 0.48, indicating no drought. This means the mild or moderate drought on

Fig. 4.1 SPI Sequences on Different Temporal Scales of the Mekong River Basin. a SPI1; b 3
SPI3; c SPI6; d SPI12
4 Analysis of Drought Character in the Mekong … 99

Fig. 4.2 Annual precipitation anomaly of the Mekong River Basin (mm)

short-time scales may be changed into no drought on long-time scales if rainfall

deficit can be remitted in time. On the contrary, the severe or exceptional drought
on long-time scales may be also no drought or mild drought on short-time scales.
An example is July, 2005. SPI1 and SPI3 are 0.87 and 0.18, indicating no drought;
along with the growth of temporal scale, however, SPI12 is −1.94, indicating severe
drought relative to the same period in history. From Fig. 4.1d, we could see that
SPI varies from −2.38 to 2.70, with a slightly upward trend (0.037/10a), meaning
the drought severity is slightly decreasing. Based on SPI values on various temporal
scales and the gradation standard in Table 4.1, we can know in the past half century,
serious droughts gather in late 1950s, the end of 1970s/the beginning of 1980s, the
beginning of 1990s and around 2005. On yearly scale, serious droughts took place
in 1980, 1992, 1994, 1998, 2005, 2009–2010 and 2015–2016, basically consistent
with existing research findings.
Figure 4.2 shows the annual precipitation anomaly of the Mekong River Basin
in the past century. Since the 1950s, obvious rainfall deficit was seen in late 1950s,
the end of 1970s/the beginning of 1980s and the beginning of 1990s, which further
proves the result SPI reveals (Fig. 4.1).
2) Seasonal variation of SPI
The Mekong River Basin belongs to the tropical monsoon climate zone, with
distinct dry and wet seasonal variations throughout the year: wet season with abun-
dant precipitation from May to October and dry season with low precipitation from
November to April of the following year. Therefore, based on the six-month SPI
values (SPI6) from October to April, the severity of meteorological drought in rainy
and dry seasons of the basin was analyzed.
Figure 4.3a, b show the severity changes of drought in wet season (May–October)
and dry season (November-April) respectively in the Mekong River Basin. It can be
seen from Fig. that drought and flood events occur alternately in the wet season and the
SPI values are between −2.38 and 2.90. On the whole, the variation trend of severity
of drought in wet season is not obvious. In terms of inter-annual variations, the wet
100 X. Zhang and H. Liu

Fig. 4.3 Variation characters of severity of seasonal drought (SPI6) in the Mekong River Basin
with the Time. a Wet Season (May–October); b Dry Season (November–April of Next Year)

season precipitation in the 1960s–1990s and 2005–2010 was relatively high; and the
precipitation in the 1950s, 1990s and the last five years (2005–2010) was relatively
low. Among them, the wet seasons of 1993, 1998 and 2015 had serious droughts
(SPI6 < − 1.5); especially in 1993, SPI6 was only −2.38, indicating exceptional
drought.
Compared with the wet season, SPI6 value of dry season in the basin shows an
obvious upward trend, indicating that the drought severity in the dry season shows a
downward trend. In terms of inter-annual variation, the dry season had less precip-
itation before the mid-1990s and SPI6 values of 1995, 1959, 1980 and 2005 were
lower than −1.5, which was a extraordinarily serious drought year. Since the late
1990s, SPI6 value has been greater than 0 in most years, especially in the dry season
of 2013, the SPI6 value reached 2.70 and precipitation was abundant.

3) Spatial distribution of SPI trend

Figure 4.4 shows the variation trends of drought severity in the whole year, wet season
and dry season of the Mekong River Basin. On the yearly scale, the SPI of Myanmar,
4 Analysis of Drought Character in the Mekong … 101

Fig. 4.4 Spatial Distribution of Drought Severity trend in the Mekong River Basin. a Annual
drought severity trend of SPI (SPI12) on the twelve-month scale; b wet season drought severity
trend of SPI (SPI6) on the six-month scale; c Dry season drought severity trend of SPI (SPI6) on
the six-month scale

northeastern Thailand and most parts of Cambodia shows a downward trend, indi-
cating that the annual drought severity in the above areas is increasing, especially in
parts of northeastern Thailand, the downward trend of SPI reaches −0.15/10 years;
relatively speaking, SPI in Lao PDR has a significant increasing trend, indicating the
severity of drought in Lao PDR is decreasing. Figure 4.4b shows that the areas with
the increased drought severity in wet season are mainly distributed in northeastern
Thailand, central Lao PDR, Myanmar, Cambodia and other countries and regions.
The drought severity in northern Lao PDR is obviously weakened. Compared with
the wet season, the drought severity in dry season shows a weakening trend in most
parts of the basin (corresponding SPI value is increasing); relatively speaking, the
drought-enhanced areas are mainly concentrated in northeastern Thailand and a small
part of western Cambodia.
(2) Frequency of drought
The frequency of drought refers to the number of occurences of drought in the entire
period of time. The calculation formula is:

d = (n/N) × 100% (4.4)

In specific, n is the number of years in which drought takes place; N is precipitation

sequence length (116 years).
Taking SPI < − 0.5 as the criterion of the occurrence of drought, we made statis-
tics of the frequency of drought in each grid of the Mekong River region (Fig. 4.5a).
According to the result, the frequency of meteorological drought exceeds 25% in
almost all regions across the basin. In specific, drought frequency surpasses 30% in
Lao PDR, northeastern Thailand and northwestern Cambodia; relatively speaking,
102 X. Zhang and H. Liu

Fig. 4.5 Distribution of annual drought frequency (Left) and frequency of severe and exceptional
drought (Right) in the Mekong River Basin

the frequency is slightly lower in Mekong Delta in Viet Nam than in other regions.
On this basis, we further calculated the frequency of severe and exceptional droughts
(SPI < −1.5) (Fig. 4.5b). According to the result, the frequency of severe and excep-
tional droughts is high, close to 10%, in Cambodia, Viet Nam and other regions
in the lower reaches. Comparison of both figures indicates Mekong Delta registers
higher frequency of severe and exceptional droughts than upstream regions, though
its drought frequency is slightly lower. This means the Mekong Delta is more liable
to severe and exceptional meteorological droughts.

4.2.2 Character of Hydrological Drought

Three typical hydrological stations, Chiang Saen, Mukdahan and Stung Treng, were
selected as representatives of hydrological characteristics of upper, middle and lower
reaches of the mainstream Mekong River, to analyze the spatio-temporal characteris-
tics of hydrological drought in the mainstream Mekong River. Based on the day-by-
day runoffs between January 1, 1985-December 31, 2016, we figured out each sta-
tion’s average monthly runoff in January 1985-December 2016; based on the monthly
runoff sequence in the past 32 years (1985–2016), we figured out the one-month,
three-month, six-month and twelve-month standardized runoff indexes (SRI1, SRI3,
SRI6 and SRI12) respectively for diagnosing each station’s hydrological drought on
different temporal scales.
4 Analysis of Drought Character in the Mekong … 103

Figures 4.6 and 4.7 show SRI1, SRI3, SRI6 and SRI12 of Chiang Saen station,
Nongkhai station successively. From Fig. 4.6d we could see that there is a slight
downward SRI trend at Chiang Saen station at the rate of about 0.20/10a. Albeit
with the little change, the magnitude of SRI is almost negative during 2010–2016,
suggesting the frequent hydrological droughts since 2010 in this station. As for inter-
annual variation, in the end of the 1980s/the beginning of the 1990s, in 1993/1994,

Fig. 4.6 SRI sequence of Chiang Saen station on various temporal scales. a SRI1; b SRI3; c SRI6;
d SRI12

Fig. 4.7 SRI sequence of Mukdahan station on various temporal scales

104 X. Zhang and H. Liu

around 2004/2005 and the last few years since 2010, the streamflow was lower
than normal conditions. In particular during 1993/1994, the SRI value is as small
as −1.9.
As for the Mukdahan station in the middle reaches, there is an apparent upward
SRI trend, with a change rate of 0.05/a (see Fig. 4.7d), which means the hydrological
drought severity and frequency both declined. For interannual variation, low runoff
gathered in the end of the 1980s/the beginning of the 1990s, 1993–1994 and 1998–
1999, when severe hydrological drought, SRI less than −1.5, was witnessed. In
comparison, runoff was ample after 2000, the SRI is around 2.0 in 2008, 2009, 2011
and 2012.
For the downstream Stung Treng station, the SRI trend is not obvious (Fig. 4.8d),
the dry year alternates with wet year. Relatively severe hydrological drought was
seen in the end of the 1980s/the beginning of the 1990s, 1998–1999, 2010 and 2016.
The SRI of 1998–1999 was as small as −1.89. In the last few years, the downstream
Stung Treng station is frequently hit by the severe hydrological droughts, e.g., in the
year of 2010 and 2016. Though drought happened, most years between mid-1990
and 2005 were wet, with high SRI value approximate 2 between 2000 and 2005.
Furthermore, we calculated and presented the SRI series for the wet and dry
season, respectively, at the stations of Chiang Saen, Mukdahan, and Stung Treng.
The magnitude of SRI in wet season is consistent with the interannual variability
of 12-month SRI (shown in Figs. 4.6, 4.7 and 4.8), varying among different station
locations. The upstream Chiang Saen shows an evident downward trend at the rate
of −0.4/10a, especially for the last few years since 2010, while the wet season
SRI in middle Mukdahan station shows the 0.3/10a upward trend. Comparing with
the upper two stations, the downstream Stung Treng station seems little change of

Fig. 4.8 SRI sequence of Stung Treng station on various temporal scales
4 Analysis of Drought Character in the Mekong … 105

Fig. 4.9 SRI sequence in wet (left panel; May–October) and dry season (right panel; November–
April of Next Year) at the Chiang Saen (top panel), Mukdahan (middle panel), and Stung Treng
(bottom panel) station

streamflow in wet season. In contrast, all these three typical stations have seen the
obvious increasing SRI in dry season with the rate over 0.3/10a, suggesting the
overall decreased frequency and severity of hydrological droughts in dry season.
This seasonal analysis suggests that more attentions need to be paid on the hydro-
meteorological change (e.g., precipitation, streamflow) in wet season in the future
(Fig. 4.9).

4.3 Discussion

In this section, the current analysis and results to most extent are limited by the
data availability. Specifically, limited by the lack of in situ precipitation measure-
ments across the entire basin, the present analysis of meteorological drought is totally
based on the open-accessible global precipitation dataset with relatively low spatial
resolution (0.5° × 0.5°). In future, the results may need to be further verified if the
abundant ground observations are available (provided by the member counries in
106 X. Zhang and H. Liu

Mekong River Basin). Moreover, the unavailable water use or demand data from
various socio-economic sectors (e.g., agriculture, industry, and human livelihoods)
also hampers the drought analysis from the socio-economic perspective. To this end,
continued efforts should be further devoted to characterize drought with respect
to different aspects including meterology, hydrology, agriculture, and socioecon-
omy, and thus benefits us more deeply and comprehensively understand the drought
characteristics in the Mekong River Basin.

4.4 Summary

(1) Severity of meteorological drought: In terms of inter-annual drought severity,

the SPI values of the Mekong River Basin changed from −2.38 to 2.70 in the
past 60 years and severe drought took place in 1980, 1992, 1994, 1998, 2004–
2005, 2009–2010 and 2015–2016. As a whole, the drought severity shows a
slight weakening trend, but in terms of spatial distribution, the drought severity
in northeastern Thailand, most of Cambodia and Myanmar shows an increasing
trend, especially in parts of northeastern Thailand.
(2) Frequency of meteorological drought: The frequency of meteorological drought
(using SPI < −0.5 as the criterion) in the most areas of the Mekong River Basin
generally exceeds 25% and frequency of severe and exceptional droughts (using
SPI < −1.5 as the criterion) is mostly between 6 and 8%. As far as the whole
region is concerned, the areas more liable to severe and exceptional droughts
are mostly concentrated in southern Cambodia, Mekong Delta in Viet Nam and
other downstream regions.
(3) Severity of hydrological drought: The severe hydrological drought years diag-
nosed by runoff observations at Chiang Saen station and Mukdahan station in
the upper and middle reaches are the late 1980s–early 1990s, 1994, 1998/1999
and 2005; the severe hydrological drought took place in Stung Treng station
in the lower reaches in many years, besides the late 1980s–early 1990s and
1998/1999, this station also experienced significant hydrological droughts in
2010 and 2016.
(4) Taking two typical droughts in 2004–2005 and 2016 as examples and combined
with meteorological and hydrological factors such as atmospheric circulation
background, rainfall and runoff, the causes and development process of two typ-
ical droughts were studied. The results show that the El Nino phenomenon, char-
acterized by the atmospheric-ocean circulation anomaly in the tropical Pacific
Ocean, usually leads to the early end of the rainy season in the Mekong River
Basin and the abnormally less rainfall than that of the same period in history,
so that the Mekong River Basin is more liable to a large-scale meteorological
drought; and when the meteorological drought lasts for a long time, the corre-
sponding hydrological drought will take place from the upstream to downstream
in the basin. Among them, the hydrological drought in the downstream delta
4 Analysis of Drought Character in the Mekong … 107

usually leads to the decline of river water level, which makes the time of seawa-
ter invasion advanced and the degree enhanced; the superposition of both will
aggravate the losses caused by drought, and the drought in the Mekong Delta
in 2016 is a typical case.

Dr. Zhang’s research primarily focuses on drought monitoring and forecasting, hydrometeorol-
ogy, and climate change. He has developed an operational drought monitoring and forecasting sys-
tem over China and the Mekong River basin, respectively, in support of drought early-warning. To
date, he has been funded as the director of Chinese Postdoctoral Science Foundation, and partici-
pated in more than 10 national-level funding projects. He has published more than 15 papers in the
famous high-impact journals like Nature Climate Change, and also been involved in 3 published
books like the AGU Geophysical Monograph Series.

Dr. Liu’s research primarily focuses on hydrological processes and modeling, flood management,
and transboundary river water cooperation. She has over 5 years’ research experiences in trans-
boundary water cooperation of China. She was a participant of various Lancang-Mekong water
resources cooperation activities. She is an editor of a joint publication by MWR of China and
MRC in 2016. She is Program Managers of 3 international cooperation projects with Mekong
River countries. She is author of over 10 journal papers and 1 book.
Chapter 5
Overview of Measures and Assessment
of Capacity for Flood Prevention
and Drought Relief

Zhixiong Ding, Xuejun Zhang, Song Han, and Hui Liu

Abstract This Chapter introduces the structural and non-structural measures for
flood prevention and drought relief in every country of the Mekong River Basin
(MRB), and evaluates the capacities for flood prevention and drought relief to these
countries. In this chapter, Some specific situations (including location, quantity and
scale) of the flood prevention engineering, such as reservoirs, dikes, sluices, pumps,
flood storage and detention areas etc., are introduced. The non-structural measures for
flood control and disaster reduction of Mekong River Commission and other countries
in the basin are introduced from the aspects of hydrological monitoring, prediction,
early warning, organization and management system, emergency response, etc. The
flood prevention capacity of each country in the MRB and the overall flood control
capacity of the MRB are evaluated from the aspects of the per capita reservoir capac-
ity of each country in the basin, and the flood discharge capacity of the main stream
of the Mekong River, etc. In this chapter, the drought relief capacity is assessed from
an analysis of the available structural and non-structural measures in the MRB using
multiple disaster survey data. Specifically, three key indicators (including reservoir
irrigation pattern, the proportion of irrigation area, per capita GDP) are employed
to assess the water project capacity, economic strength, and drought prevention and
disaster reduction response capability. Results show that the flood and drought relief
capacity is mostly dependent on the support capacity from the national economy
and the water conservancy project. From the national view, the flood and drought

Z. Ding (B) · S. Han

China Institute of Water Resources and Hydropower Research, Research Center on Flood and
Drought Disaster Reduction, A-1 Fuxing Road, Haidian District, Beijing 100038, China
e-mail: [email protected]
S. Han
e-mail: [email protected]
X. Zhang · H. Liu
China Institute of Water Resources and Hydropower Research, A-1 Fuxing Road,
Haidian District, Beijing 100038, China
e-mail: [email protected]
H. Liu
e-mail: [email protected]

© China Water & Power Press and Springer Nature Singapore Pte Ltd. 2020 109
H. Liu (ed.), Flood Prevention and Drought Relief in Mekong River Basin,
Springer Tracts in Civil Engineering,
https://doi.org/10.1007/978-981-15-2006-8_5
110 Z. Ding et al.

relief capacity differs by countries. Comparing with other countries, the Thailand
and Viet Nam are of better capacity in response to flood and drought disasters,
attributing to the enough economic investment in emergency response system and
the relatively completed flood prevention and irrigation project system. In contrast,
the flood prevention and irrigation project and its spatial coverage in other countries
(Cambodia, Laos and Myanmar) are still in infancy, resulting in the relatively poor
flood prevention and drought-coping capacity. From the basin perspective, the inte-
grated flood prevention and drought relief capacity is weak, due to the lack of the
Upper-Lower coordination scheme. This suggests the countries in the Mekong River
Basin should further strengthen the coordination and communication to improve the
integrated water resource emergency management, and thus explore the potential
flood prevention and drought relief capacity at the basin-scale.

5.1 Data and Methodology

5.1.1 Data

In order to survey the flood prevention and drought relief measures of the Mekong
River Basin countries and evaluate their corresponding capability, the following
relevant information was obtained through on-site investigation and data collection.
(1) Maps, photos and videos from the site survey.
(2) Relevant public information downloaded from the Mekong River Commission
website, including: incident reports, hydrological sites and their sections, water
monitoring data, and mountain flood warning information. Including:

Planning Atlas of the Lower Mekong River Basin, 2011.

Overview of the Hydrology of the Mekong Basin, 2005.
Annual Mekong Flood Report, 2005–2014.

(3) The professional websites of relevant departments of the Mekong River coun-
tries and information from other third-party related websites. These web sites
are as follows.
Cambodia National Mekong Committee http://cnmc.gov.kh/cnmc/index.
php/en/.
Laos National Mekong Committee http://www.lnmc.gov.la.
Thailand National Mekong Committee http://www.tnmc-is.org.
Viet Nam National Mekong Committee http://www.vnmc.gov.vn/.
Thailand’s Department of Disaster Prevention and Mitigation.
http://www.disaster.go.th/en/index.php.
Central Steering Committee for Natural Disaster Prevention and Control of
Viet Nam.
http://phongchongthientai.vn/.
5 Overview of Measures and Assessment of Capacity … 111

Tonle Sap Lake Authority http://www.tonlesap.gov.kh/.

Asian Disaster Reduction Center http://www.adrc.asia/.
(4) The Google Earth satellite remote sensing image data in recent years in the
Mekong River Basin and topographic data of Mekong River Basin with 10 m
resolution provided by Google Map.

5.1.2 Methodology

Through on-site investigation, expert consultation, and analysis of relevant data,

study and analysis of flood prevention structural measures, drought-resistant struc-
tural measures, non-structural measures for flood prevention and drought relief by the
Mekong River Basin countries were done to assess the flood prevention and drought
relief of these countries. The specific content mainly includes:
• Using the Mekong River Basin Planning Atlas, in combination with on-site inves-
tigation, satellite remote sensing image interpretation and other related data anal-
ysis, and through the consultation and exchange with relevant experts from the
Mekong River countries, the dikes, reservoirs, gate pumps, section of Mekong
river mainstream and status, scale, location and related characteristic parameters
of flood area in these countries were studied and analyzed.
• Based on the Atlas of Mekong River Basin Planning, the distribution characteris-
tics of irrigation facilities and dam reservoirs in the Basin were studied by means
of satellite remote sensing images and field investigations. The analysis is focused
on the distribution of reservoirs on the mainstream and tributaries, the distribution
of irrigation projects in the basin, the scale of irrigation districts, the proportion
of irrigation area, and other hydraulic engineering parameters closely related to
drought relief.
• Through the analysis of relevant data, in combination with expert consultation
and exchange, and the status of non-structural measures for flood prevention and
drought relief in the Mekong River Basin countries was studied, including: the
flood prevention and drought relief organization system of Mekong countries, the
monitoring, forecast and early warning of Mekong River by the Mekong River
Commission and countries in the basin, and situation of flood prevention and
drought relief in of the Mekong River Basin countries, etc.
• Based on the analysis of flood prevention structural measures, and non-structural
measures in the Mekong River Basin countries, and the further analysis of the
flood prevention and disaster reduction organization system, the embankment
status, the reservoir storage capacity, the river channel discharge capacity and the
flood storage capacity of flood storage areas in these countries, the flood prevention
capacity of these countries was analyzed. The flood prevention status of the basin
from the perspective of the whole river basin was analyzed, and the experience of
basin-level flood prevention management of transnational rivers with the example
112 Z. Ding et al.

of Rhine River Basin was introduced, to provide reference for the implementation
of basin-level flood prevention management in the Mekong River Basin.
• Based on structural and non-structural measures for drought relief in the Basin, in
combination with economic strength of the Mekong River countries, and taking
the size and distribution of the reservoir irrigation area, the proportion of irri-
gation area, and per capita GDP as indicators, the drought relief capability of
these countries was comprehensively investigated and assessed from the aspects
of water project capacity, economic strength, and drought prevention and dis-
aster reduction response capability. And furthermore, through the specific case
study, the comprehensive drought relief of the Mekong River Basin from a whole
basin perspective to tap the greater drought relief potential within the basin was
analyzed.

5.2 Structural Measures for Flood Prevention

5.2.1 Cambodia

The flooding in Cambodia is mainly affected by its own topography and meteoro-
logical conditions, as well as the upper reaches of the main stream of the Mekong
River, and the three rivers of Se Kong, Se San and Sre Pok. In recent years, with the
increase of upstream irrigation water consumption and the construction of reservoir
projects, the threat of flooding in the floodplain of the Mekong River in Cambodia
and the surrounding area of Tonle Sap has been reduced. As a result, the Cambodian
government has increased its development in these areas, and these development
areas are still threatened by floods. In addition, the reservoir built upstream of the
tributary also has the risk of dam break, resulting in flooding in Cambodia. For exam-
ple, on July 24, 2018, the dam of the Xe Pian-Xe Namnoy reservoir in the province of
Api, Laos, caused extensive flooding in the Se Kong river. The dam-breaking flood
affected the territory of Cambodia in the lower reaches of the Se Kong River. At the
Siem Pang station on the Se Kong River in Cambodia, the water level was 8.30 m at
3 p.m. on July 22, and 12.47 m at 2 p.m. on July 27, an increase of 4.08 m.1
(1) Reservoirs
Along with economic and social development, Cambodia has built 2 reservoirs, one
of which is O Chum 2 with the storage capacity of 120,000 m3 , the other one is
Lower Se San 2 with the storage capacity of 2.715 km3 , and plans to build another
12 reservoirs (Fig. 5.1), with a gross storage capacity of around 16 km3 . Among the
12 planned reservoirs, 5 reservoirs are located in the mountainous region on the upper
Tonle Sap Lake, with a gross storage capacity of 5.03 km3 ; 5 reservoirs are located

1 MRC monitors the flooding situation in Southern Laos closely, http://www.mrcmekong.org/news-

and-events/news/mrc-monitors-the-flooding-situation-in-southern-laos-closely/. 25th July 2018.

5 Overview of Measures and Assessment of Capacity … 113

Fig. 5.1 Distribution of reservoirs in Mekong River Basin in Cambodia

in the eastern mountainous region, with a gross storage capacity of about 9 km3 ;
and the rest 2 reservoirs are located on the mainstream Mekong River, with a gross
storage capacity of 2.07 km3 . Informations of the reservoirs are listed in Table 5.1.
The distribution see Fig. 5.1 (labeled with ID).
(2) Levee projects
In Cambodia, river channels are basically in a natural state without levees or other
flood prevention projects on both banks; instead, there are some urban bank protection
works in only Stung Treng, Kratie, Kompong Cham, Phnom Penh and other cities.
On the right bank of the Mekong River in Cambodia is the catchment basin of the
Tonle Sap Lake, which not only takes in, regulates and stores floodwater (about 57%
of lake water) of the Mekong River and also collects water (about 30% of lake water)
from surrounding tributaries, with a gross catchment area of 85,850 km2 (including
the area of the lake region). Residents inhabit along the lake. About 1.5 million people
live on fishing, collection of wild animals and plants in wet land or cultivation of
few cereals in floating houses or hamlets near the lake. Near Stung Treng City on the
left bank of the Mekong River in Cambodia, three big tributaries flow together into
the Mekong River, and they are the Se Kong River that originates from Lao PDR
and the Se San River and the Sre Pok River with their sources in Viet Nam. Gross
catchment area of the three rivers is 78,645 km2 or about 13% of the drainage area
114 Z. Ding et al.

Table 5.1 Reservoirs in Cambodia

ID Name River Status Storage (106 Drainage area
m3 ) (km2 )
C001 O Chum 2 O Chum Existing <10 45
C002 Lower Se San2 Se San Existing 2715 49,200
C003 Battambang 1 Stung Sangker Planned 1000–3000 2135
C004 Battambang 2 Stung Sangker Planned 100–1000 120
C005 Sambor Mekong Planned 1000–3000 646,000
C006 Stung Treng Mekong Planned 10–100 635,000
C007 Pursat 1 Stung Pursat Planned 100–1000 1100
C008 Pursat 2 Stung Pursat Planned 100–1000 2080
C009 Lower Se San 3 Se San Planned >3000 15,600
C010 Prek Liang 1 Prek Liang Planned 100–1000 883
C011 Prek Liang 2 Prek Liang Planned 100–1000 595
C012 Lower Sre Pok 3 Sre Pok Planned >3000 26,200
C013 Lower Sre Pok 4 Sre Pok Planned 1000–3000 13,800
C014 Stung Sen Stung Sen Planned 1000–3000 10,540

Fig. 5.2 Water level-area-volume curve of Tonle Sap Lake

5 Overview of Measures and Assessment of Capacity … 115

of the Mekong River. Under the influence of the Annam Mountainous Region, these
rivers contribute to about 17% of the average runoff of the basin. Their afflux into
the mainstream gathers in August-October. After joining the runoffs from upstream,
the water pushes up the flood peak of the mainstream, submerges the flood plains of
the Mekong River in Cambodia and further enhances the backflow of the Tonle Sap
River, with an important hydrological function in the Tonle Sap Lake’s regulation
of flood in the mainstream Mekong River. Main hydrological stations on Mekong
mainstream in Cambodia include Stung Treng and Kratie. The warning water level
and flood water level are 47.49 m and 48.79 m at Stung Treng station; that of Kratie
station are 20.92 m and 21.92 m respectively. The Mekong mainstream becomes
wider in the Kratie section.
(3) Flood storage and detention areas
With a special geographical location, Cambodia becomes an important flood storage
and detention region in the Mekong River Basin that plays an important role in
reducing the flood peak of the mainstream Mekong River, storing and detaining
superfluous floodwater of the Mekong River and relieving flood inundation in the
lower reaches. From mid-May to early June, the Lancang-Mekong River Basin will
usually enter into the wet season under the influence of the East Asian Monsoon
and Southwest Monsoon. In mid-July, the flood of the Lancang River will reach the
peak. While a huge amount of floodwater rushing through the Khone Falls into the
flat river channels in the lower Mekong River in Kratie, water level of the Chaktomuk
intersection between the mainstream Mekong River and the Tonle Sap River will rise
rapidly, outflow of the Tonle Sap River will be inverted into inflow, and flood of the
Mekong River will flow into the Tonle Sap Lake. Upon mid-August, the East Asian
Monsoon of western China weakens, but the growing southwest monsoon brings
forth a huge rainfall as blocked by the Annam Mountain; in combination with the
Pacific Ocean’s tropical storm and typhoon in September, rainfall gets intensified in
the Mekong River Basin, bringing more floodwater to the Tonle Sap Lake. In the
meantime, floodwater below Kratie also overflows into the wetland region of the
flood plain of the Mekong River. In late September, backflow of the Tonle Sap Lake
gets to the peak and after the short-term high water level, the Tonle Sap Lake will get
back to normal direction of flow into the Mekong River in the beginning of October,
along with the weakening of the monsoon. From July to September, the storage and
detention of floodwater in the Tonle Sap Lake and the flood plains of Cambodia not
only can relieve life and property loss of residents in the lower reaches, but also plays
a positive and important role in maintaining the diversity of ecosystems surrounding
the Tonle Sap Lake and assuring water replenishment in the dry season in the delta
region in the Mekong River. The scale of floodwater detained in the Tonle Sap Lake
and the flood plains rests with the intensity of tropical storm, the rainfall in the Annam
Mountainous Region and tributary replenishment in Lao PDR.
As for the Tonle Sap Lake, water area increases from 2215 km2 on average in
dry season to 13,258 km2 in flood season, water depth rises from 1 m to 6–10 m,
and water volume increases from 1.5 to 70 km3 . According to MRC’s research, 57%
of water replenishment of the Tonle Sap Lake is derived from the Mekong River,
116 Z. Ding et al.

30% from tributaries of the lake and 13% from surface rainfall. Annual inflow to the
Tonle Sap Lake ranges from 44.1 km3 (1998, exceptionally dry year) to 106.5 km3
(2000, wet year), with an average of 79 km3 . After the flood peak, water of Tonle Sap
Lake flows back to the Mekong River, with 87% flowing through Tonle Sap River
to Mekong River, 1% becoming overland flow and 12% evaporated on lake surface.
Outflow ranges between 43.5 km3 (1998) and 104.8 km3 (2000), with an average of
78.6 km3 . Figure 5.2 is the water level-area-volume curve of Tonle Sap Lake.
In the flood plains of the mainstream Mekong River in Cambodia, floodwater
overflows the bank in flood season, making the river as wide as 50 km at maximum.
Usually, overflowing floodwater stays for weeks in the flood plains, which play an
important role in storing and detaining the floodwater of the mainstream Mekong
River, similar with the impounding regions of Tonle Sap Lake. In wet year (e.g.
2000), the flood plains’ water storage and retention is equivalent to that of Tonle Sap
Lake. In dry year (e.g. 1998), it is close to one third of the water storage and detention
of Tonle Sap Lake. On average, the former is equivalent to 1/2 of the latter.

5.2.2 Lao PDR

With the influence of the terrain of Annan Mountain, the basin area of Mekong
River in Laos which is on the left bank of the Mekong River accounts for 25% of
the total area of the River, but the annual average runoff accounts for 35% of the
total. The Nam Tha and Nam Ou rivers between Chiang Saen and Luang Prabang
contribute 6%; and there was no large tributary between Luang Prabang and Vientiane
in Laos on the left bank of the Mekong River, so the runoff only accounts for 1%;
the Nam Ngum, Nam Theun, Nam Hinboun and Se Bang Fai between Vientiane and
Savannakhet contribute 22% of the annual average total runoff of the Mekong River;
the remaining 6% is contributed by some small tributaries of the Mekong River in
Laos below Mukdahan, including the upper reaches of the Se Kong River that flows
into Cambodia.

(1) Reservoirs

There are 100 reservoirs existing, under construction or planned, with a gross storage
capacity of 58.63 km3 , in the Mekong River Basin, including the mainstream Mekong
River, in Lao PDR, and their locations are shown in Fig. 5.3. In specific, 14 reservoirs
have been constructed, with a gross storage capacity of 11.74 km3 ; 27 reservoirs are
under construction, with a gross storage capacity of 17.45 km3 ; 59 reservoirs are
planned, with a gross storage capacity of 29.43 km3 ; 9 reservoirs are located on the
mainstream Mekong River, with a gross storage capacity of 3.16 km3 , including 2
reservoirs under construction—Xayabury and Don Sahong whose storage capacity
is 0.225 km3 and 0.115 km3 respectively—and 7 reservoirs planned. Information
and distribution of these reservoirs see Table 5.2 and Fig. 5.3. These reservoirs are
mainly located in Nam Ou, Nam Kan, Nam Ngum basin and upper Se Kong.
5 Overview of Measures and Assessment of Capacity … 117

Fig. 5.3 Reservoirs in Mekong River Basin in Lao PDR

(2) Levee projects

Levees on the mainstream Mekong River in Lao PDR include 7.4 km urban revetment
in Vientiane, 2.5 km in Pakse and 5 km in Champasak. There are gates and drainage
pumping stations for flood prevention in Vientiane, Pakse, Thakhek, Savannakhet,
Champasak and other cities, but most reaches along the mainstream Mekong River
in Lao PDR only have natural banks, and the estuary of tributaries of Mekong River
is basically in a natural state, liable to the influence of high-level flood and back-
water effect of Mekong River. Because rainfall is frequent in mountainous areas of
Lao PDR, minor basin management is carried out and flash flood warning systems
118 Z. Ding et al.

Table 5.2 Reservoirs in Lao PDR

ID Name River Status Storage (106 Drainage
m3 ) area (km2 )
L001 Nam Ngum 1 Nam Ngum Existing >3000 8460
L002 Nam Dong Nam Don Existing <10 4
L003 Xelabam Se Done Existing <10 6360
L004 Xeset 1 Se Set Existing <10 485
L005 Theun-Hinboun Nam Existing 10–100 8927
Theun,
Nam Hin
Boun
L006 Houayho Houayho, Existing 100–1000 192
Se Kong
L007 Nam Leuk Nam Leuk, Existing 100–1000 274
Nam Ngum
L008 Nam Mang 3 Nam Mang, Existing 10–100 82
Nam Ngum
L009 Nam Ko Nam Ko Existing <10 223
L010 Nam Ngay Nam Ngay Existing <10 315
L011 Nam Theun 2 Nam Construction >3000 4013
Theun, Se
Bang Fai
L012 Xekaman 3 Houayho, Construction 100–1000 712
Se Kong
L013 Xeset 2 Se Set Construction <10 392
L014 Nam Ngum 2 Nam Ngum Existing 1000–3000 5640
L015 Nam Lik 2 Nam Lik Existing 100–1000 1993
L016 Nam Ngum 5 Nam Ngum Construction 100–1000 483
L017 Xekaman 1 Se Kaman Construction 1000–3000 3580
L018 Xekaman-Sanxay Se Kaman Construction <10 3740
L019 Theun-Hinboun Nam Theun Existing 10–100 8937
expansion
L020 Theun-Hinboun exp. Nam Theun Existing 1000–3000 2942
(NG8)
L021 Nam Ngum 3 Nam Ngum Construction 100–1000 3888
L022 Nam Theun 1 Nam Theun Construction 1000–3000 14,070
L023 NamNgiep 1 Nam Nhiep Construction 1000–3000 3700
L024 Nam Nam Nhiep Construction <10 3750
Ngiep-regulating
dam
L025 Nam Tha 1 Nam Tha Construction 100–1000 8990
L026 Nam Long Nam Ma Construction <10 156
(continued)
5 Overview of Measures and Assessment of Capacity … 119

Table 5.2 (continued)

ID Name River Status Storage (106 Drainage
m3 ) area (km2 )
L027 Xepian-Xenamnoy Se Pian, Se Construction 100–1000 820
Nam Noy
L028 Xe Katam Se Nam Construction 100–1000 263
Noy
L029 Xekong 4 Se Kong Construction >3000 5400
L030 Nam Kong 1 Nam Kong Construction 100–1000 1250
L031 Xe Kong 3up Se Kong Planned 10–100 5882
L032 Xe Kong 3d Se Kong Planned 100–1000 9700
L033 Xe Kong 5 Se Kong Planned 1000–3000 2615
L034 Don Sahong Mekong Construction 100–1000 553,000
L035 Nam Ou 1 Nam Ou Construction 10–100 25,979
L036 Nam Ou 2 Nam Ou Construction <10 22,568
L037 Nam Ou 3 Nam Ou Construction 10–100 19,774
L038 Nam Ou 4 Nam Ou Construction <10 11,799
L039 Nam Ou 5 Nam Ou Construction 10–100 10,371
L040 Nam Ou 6 Nam Ou Construction 100–1000 5527
L041 Nam Ou 7 Nam Ou Construction 1000–3000 3477
L042 Nam Lik 1 Nam Lik Construction <10 5050
L043 Nam San 3 Nam San Construction 100–1000 155
L044 Nam Pha Nam Pha Planned 1000–3000 2837
L045 Nam suang 1 Nam Suang Planned 10–100 5755
L046 Nam Suang 2 Nam Suang Planned 1000–3000 5195
L047 Nam Nga Nam Ou Planned 1000–3000 2477
L048 Nam Beng Nam Beng Planned 10–100 1908
L049 Nam Feuang 1 Nam Planned 10–100 714
Feuang
L050 Nam Feuang 2 Nam Planned <10 374
Feuang
L051 Nam Feuang 3 Nam Planned <10 184
Feuang
L052 Pakbeng Mekong Planned 100–1000 218,000
L053 Luangprabang Mekong Planned 100–1000 230,000
L054 Xayabury Mekong Existing 100–1000 272,000
L055 Paklay Mekong Planned 100–1000 283,000
L056 Sanakham Mekong Planned 100–1000 292,000
L057 Sangthong-Pakchom Mekong Planned 100–1000 295,500
L058 Ban Kum Mekong Planned 100–1000 418,400
(continued)
120 Z. Ding et al.

Table 5.2 (continued)

ID Name River Status Storage (106 Drainage
m3 ) area (km2 )
L059 Latsua Mekong Planned 100–1000 550,000
L060 Xe Pon 3 Se Bang Planned 100–1000 464
Hieng
L061 Xe Kaman 2A Se Kaman Planned <10 1970
L062 Xe Kaman 2B Se Kaman Planned 100–1000 1740
L063 Xe Kaman 4A Se Kaman Planned 10–100 265
L064 Xe Kaman 4B Se Kaman Planned 10–100 192
L065 Dak E Mule Se Kong Planned 100–1000 127
L066 Nam Khan 1 Nam Khan Planned 100–1000 7300
L067 Nam Khan 2 Nam Khan Planned 100–1000 5221
L068 Nam Khan 3 Nam Khan Planned 100–1000 3392
L069 Nam Ngum 4A Nam Ngum Planned 100–1000 1725
L070 Nam Ngum 4B Nam Ngum Planned <10 1818
L071 Nam Ngum, Lower Nam Ngum Planned 100–1000 16,900
dam
L072 Nam Pay Nam Ngum Planned 10–100 157
L073 Nam Mang 1 Nam Mang Planned 100–1000 577
L074 Nam Pouy Nam Phuai Planned 100–1000 1680
L075 Nam Poun Nam Pung Planned 100–1000 1700
L076 Nam Ngao Nam Ou Planned 100–1000 157
L077 Nam Chian Nam Nhiep Planned <10 271
L078 Nam Ngieu Nam Nhiep Planned 10–100 309
L079 Nam Pot Nam Nhiep Planned 10–100 60
L080 Nam San 3B Nam San Planned 10–100 257
L081 Nam San 2 Nam San Planned 1000–3000 1429
L082 Nam Pok Nam Ou Planned <10 466
L083 Nam Phak Nam Ou Planned <10 594
L084 Nam Hinboun 1 Nam Hin Planned 1000–3000 1380
Boun
L085 Nam Hinboun 2 Nam Hin Planned 10–100 31
Boun
L086 Xe Bang Fai Se Bang Fai Planned <10 6350
L087 Xe Neua Se Bang Fai Planned 100–1000 915
L088 Nam Theun 4 Nam Theun Planned 100–1000 660
L089 Nam Mouan Nam Theun Planned 1000–3000 1652
L090 Xe Bang Hieng 2 Se Bang Planned 100–1000 669
Hieng
(continued)
5 Overview of Measures and Assessment of Capacity … 121

Table 5.2 (continued)

ID Name River Status Storage (106 Drainage
m3 ) area (km2 )
L091 Xedon 2 Se Done Planned 1000–3000 4090
L092 Xe Set 3 Se Done Planned <10 127
L093 Xe Bang Nouan Se Bang Planned 1000–3000 474
Nouan
L094 Xe Lanong 1 Se Bang Planned 100–1000 1415
Hieng
L095 Xe Lanong 2 Se Bang Planned 10–100 339
Hieng
L096 Nam Phak Houay Planned 10–100 80
Namphak
L097 Xe Nam Noy 5 Se Kong Planned <10 60
L098 Houay Lamphan Se Kong Planned 100–1000 140
L099 Nam Kong 2 Se Kong Planned 100–1000 860
L100 Xe Xou Se Kong Planned 1000–3000 1273

are built also in some mountainous regions. Lao PDR has 6 hydrological stations—
Luang Prabang, Vientiane, Paksane, Thakhe, Savannakhet and Pakse—on the main-
stream Mekong River, and the corresponding locations are demonstrated in Fig. 5.3.
The warning water level and flood water level of Luang Prabang is 284.695 m and
285.195 m respectively; that of Vientiane is 168.36 and 170.54 m; that of Paksane
is 155.625 and 156.625 m; that of Thakhek is 142.629 m and 143.629 m; that of
Savannakhet is 137.022 and 138.022 m; that of Pakse is 97.49 and 98.49 m.

5.2.3 Myanmar

The Mekong River Basin in Myanmar is made up of the transboundary river between
Lao PDR and Myanmar and some tributaries in mountainous regions, involving a
drainage area of 28,600 km2 and an average annual runoff of 17.63 km3 , and flood-
water there is largely sourced from the Lancang River and flash floods of surrounding
tributaries (see Fig. 5.4). According to Myanmar experts, as it mainly covers moun-
tainous regions, resident population is small, and river channels along the mainstream
Mekong River are in a natural state without such infrastructure as flood prevention
projects or such water monitoring facilities as hydrological stations. There are minor
basin management projects in mountainous regions that can prevent flash flood to
some extent.
122 Z. Ding et al.

Fig. 5.4 Mekong River Basin in Myanmar

5.2.4 Thailand

Thailand’s floods in the Mekong River Basin are mainly affected by the high water
level of the main stream of the Mekong River, causing flooding in the low-lying
areas along the Mekong River or the flooding in the tributaries of Thailand because
of the backflow of the high water level of Mekong River. In the northern part of
Thailand, it is mainly the two larger tributaries of Mae Nam Kok and Mae Nam Ing;
in the dish-like highland area in northeastern Thailand, the main tributaries are Huai
Luang, Nam Songkhram and Nam Mun. These tributaries are susceptible to high
flood levels in the main stream of the Mekong River without gates and pumps.

(1) Reservoirs

There are 7 reservoirs (Fig. 5.5), with a gross storage capacity of 3.57 km3 , in the
Mekong River Basin in Thailand. In specific, 7 reservoirs are built up and 1 reservoir
is under construction. Information of these reservoirs see Table 5.3 and Fig. 5.5.

(2) Levee projects

According to field investigation, consultation and remote sensing interpretation, there

are revetment bank along urban river sections on the Thailand side of Mekong River,
5 Overview of Measures and Assessment of Capacity … 123

Fig. 5.5 Reservoirs in Mekong River Basin in Thailand

Table 5.3 Reservoirs in Thailand

ID Name River Status Storage (106 Drainage area
m3 ) (km2 )
T001 Chulabhorn Nam Phrom Existing 100–1000 545
T002 Huai Kum Nam Phrom Existing 10–100 282
T003 Nam Pung Nam Pung Existing 100–1000 296
T004 Pak Mun Nam Mun Existing 100–1000 117,000
T005 Sirindhorn Lam Dom Noi Existing 1000–3000 2097
T006 Ubol Ratana Nam Pong Existing 1000–3000 12,104
T007 Lam Ta Khong Lam Ta Kong Existing 100–1000 1430
P.S.

with a total length of around 300 km. Levees formed by road could be found in some
low-lying sections. But low-lying sections are still under the impact of the high-level
flood of the Mekong River (Fig. 5.6). In some key cities and towns, revetment projects
(Fig. 5.7) that can withstand high flood levels are built with high quality.
124 Z. Ding et al.

Fig. 5.6 A low-lying section along Mekong River liable to flood inundation

Fig. 5.7 A revetment project of Mekong River in Thailand

Thailand’s hydrological monitoring stations on the mainstream Mekong River are

Chiang Saen, Chiang Khan, Nongkhai (see Fig. 5.8), Nakhon Phanom, Mukdahan
and Khong Chiam. The warning water level and flood water level of Chiang Saen is
368.61 m and 369.91 m respectively; that of Chiang Khan is 208.618 and 210.118 m;
that of Nongkhai is 165.048 and 165.848 m; that of Nakhon Phanom is 142.461 and
143.021 m; that of Mukdahan is 136.219 and 136.719 m; that of Khong Chiam is
102.53 and 103.53 m.
5 Overview of Measures and Assessment of Capacity … 125

Fig. 5.8 Nong Khai station

(3) Sluices and pumps

In the disk-shaped highlands in the northeastern Thailand within the Mekong River
Basin, high-level backflow of the Mekong River may be easily caused, flood drainage
may be impeded and inland inundation may be incurred where there are no sluices
and other flood prevention facilities once sudden rainstorm takes place in the high-
water-level season of the Mekong River, though flood inundation is less probable.
An example is the Huai Luang regulation project (Fig. 5.9). Irrigation department,
responsible for the regulation of the sluice, closes the sluice to raise water level and
provide irrigation water in the region when irrigation water is needed; and closes the
sluice and discharges floodwater using water pumps to the Mekong River when flood
takes place in the Mekong River and water level of the sluice is high (Figs. 5.10 and
5.11). Presently, the project is planned further (Fig. 5.12) to draw water through multi-
stage pumping station from Mekong River for irrigation in dry season, and discharge
floodwater also through the multi-stage pumping station from the irrigation region
to the Mekong River in flood season.

Fig. 5.9 Huai Luang

regulation sluice
126 Z. Ding et al.

Fig. 5.10 Flood drainage

pump at Huai Luang Sluice

Fig. 5.11 Flood drainage

pipe at Huai Luang

Seen from the satellite image, at least 5 tributaries in Khorat highland of northeast-
ern Thailand are installed with sluices, pumps and other projects, and their locations
are shown in Fig. 5.13. These sluices aim to provide irrigation water and also prevent
flood.
5 Overview of Measures and Assessment of Capacity … 127

Fig. 5.12 Huai Luang Sluice multi-stage pumping station program

Fig. 5.13 Sluices and gates in Khorat highland of northeastern Thailand

128 Z. Ding et al.

5.2.5 Viet Nam

Mekong River Basin in Viet Nam mainly include Viet Nam the Mekong Delta and
the mountain areas at upper reaches of the Se San and Sre Pok in the Central High-
lands. The floods in the Mekong Delta region mainly come from the upstream of the
mainstream Mekong River and its tributaries, or the sea water encroachment. When
the flood in floodplain around the main stream of the Mekong River in Cambodia
and the Tonle Sap Lake excesses their storage capacity, there might be flooding in
the downstream delta. The threat in the mountainous areas at upper reaches of the
Se San and Sre Pok in the Central Highlands mainly is the mountain torrent.

(1) Reservoirs

In Viet Nam, there are 15 reservoirs, with a gross storage capacity of 3.16 km3 in
the mountainous regions of upper Se San River and Sre Pok River in the Central
Highlands. According to our acknowledgement, these reservoirs are designed for
power generation and on small and medium scales, and their locations are shown in
Fig. 5.14. In specific, 10 reservoirs have been built up, with a gross storage capacity
of 2.59 km3 ; 3 reservoirs are under construction, with a gross storage capacity of
0.15 km3 ; and 2 reservoirs are planned, with a gross storage capacity of 0.41 km3 .
Information of there reservoirs see Table 5.4 and Fig. 5.14.

Fig. 5.14 Reservoirs in Mekong River Basin in Viet Nam

5 Overview of Measures and Assessment of Capacity … 129

Table 5.4 Reservoirs in Viet Nam in Mekong River Basin

ID Name River Status Storage (106 Drainage area
m3 ) (km2 )
V001 Upper Kontum Se San, Dak Construction 100–1000 350
Bla, Dak Nghe
V002 Plei Krong Se San, Krong Existing 100–1000 3216
Poko
V003 Yali Se San Existing 100–1000 7455
V004 Se San 3 Se San Existing <10 7788
V005 Se San 3A Se San Existing <10 8084
V006 Se San 4 Se San Existing 100–1000 9326
V007 Se San 4A Se San Existing <10 9368
V008 Duc Xuyen Sre Pok, Ea Planned 100–1000 1100
Krong Kno
V009 Buon Tua Srah Sre Pok, Ea Existing 100–1000 2930
Krong Kno
V010 Buon Kuop Sre Pok Construction 10–100 7980
V011 Dray Hlinh 2 Sre Pok Existing <10 8880
V012 Sre Pok 3 Sre Pok Existing 10–100 9410
V013 Sre Pok 4 Sre Pok Construction 10–100 9568
V014 Dray Hlinh 1 Sre Pok Existing <10 8880
V015 Sre Pok 4A Sre Pok Planned <10 9568

(2) Levee projects

In the Mekong delta, besides revetments in urban region, most of the river sections
do not have high quality protection projects (Fig. 5.15). During the 2017 flood, large-
scale bank collapse and erosion took place (Fig. 5.16), in which disaster was avoided
through effective prevention and evacuation of people. In the upper Mekong Delta,
paddy that relies on fresh water irrigation is the major crop. In dry season, the region
is faced with a shortage of fresh water because little water comes from upper Mekong

Fig. 5.15 Natural bank in

the Mekong delta
130 Z. Ding et al.

Fig. 5.16 Bank erosion in

the Mekong delta

River and intrusion of downstream saltwater. Besides, gates used to block saltwater
are mainly located on internal small channels and big channels are not installed with
gates, so large-range salinity intrusion takes place during high tide. The lower region
is basically exempted from the influence of salinity intrusion because it mainly relies
on aquaculture that is resilient to saltwater.

5.2.6 Mainstream Flood Prevention

(1) From the Chinese border to Chiang Saen, Thailand

From the Chinese border to Chiang Saen of Thailand, it is mainly the boundary river
between Laos, Myanmar and Thailand. This river section is dominated by a mountain
valley-type river channel. The river section is trapezoidal. When the water level is
low, the river is about 5–6 m deep. In the wet season, the river water level will rise
about 10 m, which is almost bankfull. There is a small population around this river
section and the runoff mainly comes from Lancang River in China and several small
tributaries in Thailand, Laos and Myanmar.

(2) From Chiang Saen to Vientiane

The river channel from Chiang Saen to Vientiane becomes broader, and there are
terraced river floodplains and crops fields. There are large population-concentrated
cities such as Chiang Saen, Chiang Khan and Nongkhai on Thai side; and on Laotian
side there are cities with large population such as Luang Prabang and Vientiane. In
these urban areas where the population is concentrated, there are generally revetments
that can withstand higher water level. The farmland on both sides of the river is mostly
natural banks and is vulnerable to flooding. In particular, there are two large tributaries
of Mae Nam Kok and Mae Nam Ing along the river section of Chiang Saen on Thai
side; and on the Laos side near Luang Prabang, there are several large tributaries
such as Nam Ou, Nam Soung, and Nam Khan that originate from Annan mountain
area. These tributary floods could easily encounter floods from the Lancang River,
resulting in flooding on both sides of the river and causing outer flooding and inner
water logging to cities.
5 Overview of Measures and Assessment of Capacity … 131

(3) Vientiane to the south border of Laos

Vientiane to the south border of Laos, it is mainly boundary river between Thailand
and Laos, and the lower half is the inner river section of Laos. This section basically
has similar topography of the previous one, and there are more cities with concen-
trated population on both sides. There are cities such as Nakhon Phanom, Mukdahan
and Khong Chiam in Thailand; Thakhek, Savannakhet and Pakse in Laos. In terms
of tributary, there are Huai Luang, Nam Songkhram and Nam Mun on the right bank
of Thailand and Nam Ngum, Se Bang Fai, Se Bang Hieng and Se Done originating
from the Annan mountainous area on the left side of Laos. There are sluices and
pumps at the junction of mainstream Mekong River and Huai Luang River on the
Thai side. And on the Nam Mun tributary there is a large dam project (Puk Mun
Reservoir, with a storage capacity of 125 million m3 ). These projects are significant
in flood prevention and drainage in the tributary basin on the Thai side. While the
tributaries lacking in sluices and dams on Thai side and the incoming tributaries from
Laotian side are susceptible to flood from mainstream Mekong River. The floods in
farmland and cities on both sides of this river section are mainly from the tributary in
the upper reaches and left Annan mountainous areas; and there might be also small
amount of flood from tributaries on the right bank in Thailand.
(4) Within Cambodia
The Mekong River is mainly inland rivers in Cambodia. The floodplain of Mekong
River and the Tonle Sap Lake in Cambodia stores and detains the flood. Without
flooding the downstream, the Mekong River floodplain and Tonle Sap totally stores
about 100 km3 of flood water. The flood storage capacity of the Mekong River
floodplain and the Tonle Sap Lake plays an important role in flood regulation and
replenishment in Mekong Delta. The Tonle Sap Lake on the right bank not only
detains floods in the upper reaches of the Lake District in the mountain area, but
also stores floods from the Mekong River during the rainy season and replenishes it
to the Mekong River during the dry season. On the left bank there are three major
tributaries of Se Kong, Se San, and Sre Pok, which originate from Laos and the
Annan Mountains of Viet Nam. The main large population-concentrated cities along
the Mekong River in Cambodia include StungTreng, Kratie, Kompong Cham and
Phnom Penh. These cities are susceptible to the cumulated impact of floods from the
upper reaches of the Mekong River and the three tributaries of Se Kong, Se San and
Sre Pok. Except for the cities with concentrated population, it is mostly floodplain
on both sides of Mekong River where natural fishery is the major economy and the
flood hazard is not serious. But now some areas have been cultivated. Once there is
a major flood, the damage will be severe.
(5) Delta
The delta of the Mekong River is mainly in Viet Nam. From the perspective of the
basin, this area is most vulnerable to flooding. But with the immense storage and
detention by upstream floodplain and Tonle Sap Lake in neighboring Cambodia, it
can withstand most floods provide a steady supply of fresh water during the dry
132 Z. Ding et al.

season. However, with the development of floodplains in Cambodia, this function

may be affected to some extent. And the backflow caused by high water lever could
worsen the impact. The delta is mainly dominated by agriculture, and near the coastal
areas the fishery. The cities with large populations are Chau Doc, Tan Chau, Long
Xuyen, Cao Lamh, Cao Tho and Vinh Long, etc. These cities are all located on the
banks of mainstream Mekong River and Bassac River. In the event of a major flood
or a high tide, the entire Mekong Delta may be flooded.
(6) Discharge Capacity of Sections on Mainstream Mekong River
The Mekong River Commission has played an important role in monitoring, fore-
casting and early warning of water conditions in the Mekong River Basin. Table 5.5
shows the discharge capability of the main hydrological monitoring stations on main-
stream Mekong River based on the statistics of the Mekong River Commission. The
flood discharge is calculated according to the relationship between water level and
discharge volume based on the cross section at each station (see Sect. 5.2). The capac-
ity of reservoirs built or under construction along Mekong River in Thailand, Laos,
Cambodia and Viet Nam is 34.35 km3 , accounting for 8.59% of the Mekong River’s
average annual total runoff (400 km3 ). The planned reservoirs capacity of Laos is
29.65 km3 , that of Cambodia 18.89 km3 , and Viet Nam 0.54 km3 . After all these
planned reservoirs have been built, the total reservoir capacity on Mekong River is
83.43 km3 , accounting for 20.86% of Mekong River’s average annual total runoff.
Under the current conditions, even though these reservoirs are full, the Mekong River
itself has more than 90% of runoff floods on a multi-year average; under the planning
conditions, it has nearly 80% of runoff floods on a multi-year average. In addition,
most of the reservoirs on Mekong River are mainly for power generation and irriga-
tion, and have not been considered for their flood prevention functions. Therefore,
the flood prevention situation in the countries along Mekong River has not changed
fundamentally.

5.3 Structural Measures for Drought Relief

Improving the water resources support capacity during drought through building
irrigation systems like reservoirs, sluices, canals and increasing area of irrigation
regions is a major structural measure for drought relief in Mekong River Basin
countries. The data used in this section was based on multi-source materials, including
MRC publications, web-site information, as well as on-site investigations.
As far as the entire river basin is concerned, nearly 6755 irrigation projects have
been put into use, covering an area of 4.78 million hm2 . In terms of spatial distri-
bution, the irrigation area is mainly in the Delta of Mekong River, the northern and
northeastern parts of Thailand, the southern part of Cambodia and the upper reaches
of the Sre Pok basin in Viet Nam. In terms of seasonal distribution, current irrigation
projects are mainly used to cope with the drought that occurs during the wet season
rice planting. Specifically, wet-season rice accounts for the largest area of irrigation,
5 Overview of Measures and Assessment of Capacity … 133

Table 5.5 Discharge capacity of main hydrological monitoring stations on mainstream Mekong
River (The correlation between water lever and runoff below Kratie is unstable, and the
corresponding runoff of certain flood water level is unavailable)
No. Name of Reach Country Warning Flood water Flood
station water level level (m) discharge
(m) (m3 /s)
1 Chiang Saen Upper Thailand 11.5 12.8 17,610
Mekong
River
2 Luang Upper Lao PDR 17.5 18.0 18,885
Prabang Mekong
River
3 Chiang Upper Thailand 14.5 16.0 21,497
Khan Mekong
River
4 Vientiane Middle Lao PDR 11.5 12.5 22,772
Mekong
River
5 Nongkhai Middle Thailand 11.4 12.2 18,658
Mekong
River
6 Paksane Middle Lao PDR 13.5 14.5 23,800
Mekong
River
7 Nakhon Middle Thailand 11.5 12.0 30,470
Phanom Mekong
River
8 Thakhek Middle Lao PDR 13.0 14.0 38,521
Mekong
River
9 Mukdahan Middle Thailand 12.0 12.5 31,811
Mekong
River
10 Savannakhet Middle Lao PDR 12.0 13.0 37,610
Mekong
River
11 Khong Middle Thailand 13.5 14.5 33,475
Chiam Mekong
River
12 Pakse Middle Lao PDR 11.0 12.0 38,899
Mekong
River
13 StungTreng Lower Cambodia 10.7 12.0 66,018
Mekong
River
(continued)
134 Z. Ding et al.

Table 5.5 (continued)

No. Name of Reach Country Warning Flood water Flood
station water level level (m) discharge
(m) (m3 /s)
14 Kratie Lower Cambodia 22.0 23.0 58,540
Mekong
River
15 Kompong Lower Cambodia 15.2 16.2 –
Cham Mekong
River
16 Chaktomuk Bassac Cambodia 12.0 15.0 –
River
17 Prek Kdam Tonle Sap Cambodia 9.5 10.0 –
River
18 Tan Chau Lower Viet Nam 3.5 4.5 –
Mekong
River
19 Chau Doc Bassac Viet Nam 3.0 4.0 –
River

up to 87%; dry-season rice accounts for 31%; third-season rice, which is mainly
grown in southern Viet Nam and a small part of Cambodia, accounts for about 37%;
the non-rice fields account for about 15% of the irrigated area. At present, there are
over 1300 dams/reservoirs for irrigation in the basin, of which most are Earth dam.
About half of the dams/reservoirs are located in the Highlands of Viet Nam.
The following is an analysis of the specific structural measures of countries in the
basin.

5.3.1 Cambodia

Cambodia has built up 2047 irrigation projects, with a gross irrigation area of 479,762
ha in wet and dry season, and average irrigation area per project is 234 ha. In specific,
up to 2011, the number of large-scale systems whose irrigation area exceeds 5000 ha
amounts to 33, the number of medium-scale systems whose irrigation area ranges
between 200 and 5000 ha is 955, and others are small-scale systems whose irrigation
area is smaller than 200 ha. In terms of spatial distribution, irrigation structural
measures are mainly located in the southern region, especially in Phnom Penh and
surrounding Tonle Sap Lake, and their irrigation area takes up 10–30% of local area
(district/county), which indicates water use of above regions in dry period can be
guaranteed effectively. Relatively speaking, there is almost no irrigation project in
other regions, especially in the northeastern region, northwestern region and other
remote regions. From the temporal perspective, Cambodia has three harvests of paddy
a year. Current projects mainly provide irrigation water for paddy in dry season and
5 Overview of Measures and Assessment of Capacity … 135

in wet season (either around 50%), and little irrigation water is provided to the third-
season paddy (between dry season and wet season) and other crops (except paddy). In
future 10 years, Cambodia plans to increase nationwide irrigation area to 774,000 ha
by 2030.

5.3.2 Lao PDR

Lao PDR has built up 3094 irrigation projects, with an irrigation area of 225,446 ha
in wet season. Because most irrigation measures are of small scale, average irrigation
area per project in Lao PDR is merely 113 ha. The irrigation area in Lao PDR is much
smaller than that in other countries. On the one hand, land suitable for irrigation is
limited because of the long, narrow and steep terrain; on the other hand, irrigation
input was small in history. In terms of spatial distribution, existing irrigation projects
gather in the corridor zones along tributaries of Mekong River and lower flood plains,
and mainly aim to irrigate paddy fields. For instance, pumping stations, gates for flow
control and 16-km-long conveyance canals are built in Vientiane, Pakse, Thakhek,
Savannakhet, Champasak and other cities along the Mekong River. In Lao PDR,
irrigation area takes up less than 3% of local area (district/county) and even less
than 1% in most regions. Nevertheless, Lao PDR is exerting efforts to develop its
hydropower potential and improve water resources project guarantee capacity. On
the current basis, Lao PDR plans to build more irrigation projects for dry season rice
and non-rice crops irrigation on the plateau region and increase national irrigation
area to 213,000 ha by 2030, which will promote engineering support capability for
drought relief.

5.3.3 Myanmar

Myanmar is the largest country in the Mekong River, but it is less dependent on the
Mekong River because only 2% of its territorial area is located here. Though the
area situated in the Mekong Basin is remote region without development, structural
measures in other regions of Myanmar is not underdeveloped. In specific, famous
dams include North Nawin, South Nawin, Phyu Chaung and Wegyi. In addition,
Myanmar has constructed 207 irrigation pipelines covering an area of 250,000 ha.
Construction and management of these water resources projects provides effective
water resources guarantee and policy support for coping with drought disasters.
136 Z. Ding et al.

5.3.4 Thailand

About 36% of Thailand’s territorial area, covering 166,000 km2 and mainly located in
the northeast, fall within the Mekong River Basin. On Mekong tributaries in Thailand,
256 dams/reservoirs have been built up for irrigation, with accessory constructions
of sluices, pumps and canals, forming a complete set of water division-irrigation
system. Take Huai Luang irrigation project as example (Fig. 5.17). It was buit up
in 1994, with two-way sluice of three 11 m-wide channels, the maximum drainage
capacity is about 990 m3 /s. The irrigation area around the project is about 92,911 ha,
of which the low-lying area is 42,227 ha, accounting for about 45%; the remaining
55% (50,684 ha) is the high land. The main crop type there is rice, with an area of
2.02 million ha, makes up 74% of the total area. The remaining 26% is covered by
various commercial crops. Along the canals, there are sluices and gates, water will
be pumped from mainstream Mekong to the reservoir when drought happens, and
then gate will be closed to raise the water level for irrigation in this region. Thailand
has built up 134 irrigation facilities in the Mekong River Basin. Thailand ranks only
second to Viet Nam by an irrigation area of 0.904 million ha in wet season, the share
of irrigation area (in area of local district/county) is close to or larger than 10% in most
regions. For spatial distribution, irrigation projects and irrigated regions are largely
located in corridor zones along rivers and flood plains in northeastern Thailand in
the Mekong River Basin. From the temporal perspective, wet-season paddy takes the
largest share of irrigation area, nearly 95%; and the share of irrigation of dry-season
paddy is the smallest one, less than 3%. Because of low runoff, high groundwater

Fig. 5.17 Huai Luang

irrigation project in Thailand
5 Overview of Measures and Assessment of Capacity … 137

level and soil salinization, the irrigation of dry-season paddy is restricted. Thailand
is planning to build more irrigation projects in the northeastern region in the Mekong
River Basin, with the aim to further advance the irrigation area and irrigated raio.

5.3.5 Viet Nam

Viet Nam has built up 670 dams/reservoirs for irrigation in the Highlands, 643 of
which were identified as Earth dam. Viet Nam currently has 1360 irrigation projects
in the Highlands, which could cover 47,001 ha; and 107 irrigation projects in the
Delta region, which could provide benefit for 1.99 million ha. There are about 5625
earth canals with the total length of 13,410 and 23,110 km for the main and sec-
ondary canals, respectively, see Fig. 5.18. It indicates that nearly half the crop land
in the Mekong Delta can be irrigated effectively. Similar with Cambodia, Viet Nam
has three harvests of paddy a year—wet-season paddy, dry-season paddy and third-
season paddy between them. At present, Viet Nam’s irrigation projects mainly pro-
vide irrigation water to wet-season paddy and third-season paddy, and little irrigation
water is provided to dry-season paddy and other crops. Viet Nam plans to build more
small-scale irrigation projects in the eastern highland region. These irrigation projects
will be used mainly to guarantee water supply for crops in non-paddy regions.
In Mekong Delta, saltwater intrusion increases the salinity of fresh water in the
Mekong River, due to which the water can be hardly used for irrigation and “non-
drought water deficit” is caused. Presently, Viet Nam has built a series of saltwater
obstruction projects on small tributaries (e.g. Cai Oanh Tide Gate in Fig. 5.19),
for the purpose of controlling the connection and exchange between inland rivers
and saltwater and eliminating the influence of saltwater intrusion through structural
approaches. Nevertheless, there is almost no tide gate on big tributaries, so the main-
stream is still not exempted from the influence of saltwater intrusion. Besides, Viet
Nam strengthens the coordination with upstream countries, and relieves drought
influence through increasing water from the upstream.

5.3.6 Drought Relief in the Basin

The asymmetry of the upstream and downstream positions of countries along the
Mekong River determines that these countries are not independent individuals, but
are bonded by natural hydrological links. In order to meet the needs of drought
control, upstream countries can usually transfer water from the Mekong River to
tributaries for irrigation by constructing a large number of irrigation projects. This
negative effect of “upstream mentality” giving priority to their own needs may lead to
few water available in downstream countries, which in turn undermines their drought
control capability. For example, in 2016, a large-scale meteorological and agricultural
drought occurred in the Mekong River Basin. The Chinese government provided
138 Z. Ding et al.

Fig. 5.18 Canal systems in the Mekong delta in Viet Nam (MRC 2005. Overview of the Hydrology
of the Mekong Basin. Mekong River Commission, p 54)

emergency water supplement to the Mekong River by increasing the outflow of the
Jinghong Reservoir. The research shows that during the emergency supply period, the
net contribution of the discharge amount of Jinghong station to the total runoff of each
hydrological station decreases from upstream to downstream. During this period, the
runoff from upstream in Chiang Saen, Nongkhai and StungTreng stations respectively
accounted for 99, 92 and 58% of the total runoff, and the net contribution rates of
emergency water supply were 44%, 38% and 22% respectively. For the StungTreng
station located downstream, the net contribution rate of emergency water supply
is lower than the upstream stations, due to the water intake and use by countries
in the middle and upper reaches, the consumption of water along the way and the
inflow of tributary water. This indicates that only depending on the water supplement
from upstream Jinghong station has a limited impact on drought relief downstream.
5 Overview of Measures and Assessment of Capacity … 139

Fig. 5.19 Cai Oanh Tide Gate in the delta region of Viet Nam

Therefore, coordination and communication between countries along the Mekong

River should be further strengthened, and scientific planning should be carried out
from the whole basin level to further explore the greater potential for drought relief.

5.4 Non-structural Measures for Flood Prevention

and Drought Mitigation

5.4.1 Mekong River Commission

As a cooperative organization for flood management of the Mekong River for coun-
tries in the basin, the Mekong River Commission (MRC) is responsible for collecting
rainfall and hydrological data submitted by each country and forecasting floods of
main rivers in the basin. In the flood season, between June and October each year, the
Regional Flood and Drought Management Center (RFDMC) subordinate to MRC
will release flood forecasts and alerts every day. Water levels of 22 predictive points
in the Mekong River Basin are forecasted according to the data of 146 hydrological
and meteorological stations of all countries in the basin. The RFDMC sends the daily
bulletin to MRC, NOGs, media and the public. Figure 5.20 shows the water level
forecast results of rivers in the basin for next 5 days. In Figs. 5.21, 5.22 and 5.23,
Nongkhai River is taken as an example for an introduction to the measurement, fore-
cast and comparison of its water level. Figures 5.24, 5.25, 5.26, 5.27, 5.28 and 5.29
140 Z. Ding et al.

Fig. 5.20 Flood forecasting of MRC [http://www.mrcmekong.org (accessed on 18 July 2019)]

Fig. 5.21 One cross section and water level of Nongkhai River [http://ffw.mrcmekong.org/stations.
php?StCode=NON&StName=Nongkhai (accessed on 18 July 2019)]
5 Overview of Measures and Assessment of Capacity … 141

Fig. 5.22 Observed water level in the past 7 days and predicted water level in the next 3 days
of Nongkhai River [http://ffw.mrcmekong.org/stations.php?StCode=NON&StName=Nongkhai
(accessed on 18 July 2019)]

Fig. 5.23 Water level of Nongkhai station in flood season in different years [http://ffw.mrcmekong.
org/stations.php?StCode=NON&StName=Nongkhai (accessed on 18 July 2019)]
142 Z. Ding et al.

Fig. 5.24 Early warning of flash floods in the next 1 h [http://ffw.mrcmekong.org/ffg.php (accessed
on 18 July 2019)]

are information diagrams about flash flood warning and forecast, near-real time soil
moisture and rainfall released by MRC.
In Fig. 5.20, blue represents normal water level, yellow means warning water level
is exceeded, and red means flood level is exceeded. In specific, it forecasts the water
level will reach the flood level, namely entering into the warning state, in 3 days, and
the flood level is determined by each member country.
In Fig. 5.23, blue line represents the measured water level, orange line represents
the forecasted water level, red line is the flood level, yellow line is the warning level,
and black line is the average annual level.
5 Overview of Measures and Assessment of Capacity … 143

Fig. 5.25 Early warning of flash floods in the next 3 h [http://ffw.mrcmekong.org/ffg.php (accessed
on 18 July 2019)]

5.4.2 Thailand

Thailand has a complete hydrological and meteorological monitoring, forecasting

and warning system for the Mekong River Basin. There are 11 hydrological moni-
toring stations in the Mekong River Basin in Thailand. These hydrological stations
are installed with the corresponding remote measuring system and video monitoring
system, in which national hydrological monitoring technicians assure data can be
sent to National Information System (NIS) accurately and in time. In the meantime,
hydrological monitoring data can be transmitted in real time to MRC’s server for data
analysis, quality control and forecasting. Important stations forecast, for 3–5 days,
144 Z. Ding et al.

Fig. 5.26 Early warning of flash floods in the next 6 h [http://ffw.mrcmekong.org/ffg.php (accessed
on 18 July 2019)]

water level and runoff every day and then, relevant information is released by NIS
and MRC respectively to the general public and relevant departments. Meanwhile,
the flash flood and rainstorm warning system is also established for mountainous
regions, which can provide audible and visual alarm against rainfall, with green,
yellow and red light alerts as well as tweeter against rainfall at different levels. After
the alert is issued, villagers of the village where the alerting station is located will get
evacuated to nearby highland and resettlement areas under the help of volunteers.
5 Overview of Measures and Assessment of Capacity … 145

Fig. 5.27 Soil moisture in the past 6 h [http://ffw.mrcmekong.org/ffg.php (accessed on 18 July

2019)]

5.4.3 Viet Nam

The Mekong Delta plain in Viet Nam is low-lying without dyke or any other protective
project. Even if there is any protective project, collapse banks are protected and
reinforced, which are basically unable to withstand high flood level and are liable
to high-level flooding of the Mekong River. In this region are some hydrological
stations guarded by specially-assigned people, which can forecast automatically and
send forecasting and warnings to MRC. National Mekong Committee of Viet Nam
establishes a hydrodynamic model in the delta region that can forecast flood and send
the forecasting to MRC for sharing. In the delta region, monitoring and forecasting of
146 Z. Ding et al.

Fig. 5.28 Rainfall forecast in the next 24 h [http://ffw.mrcmekong.org/ffg.php (accessed on 18 July

2019)]

water regime is very important. When a flood takes place, monitoring and forecasting
water level can early warn or transfer people in regions that may be flooded (Fig. 5.30).
In Viet Nam, flash flood becomes more frequent because of climate change in
recent years. On average, 2–4 flash floods take place in the flood season every year.
It usually occurs frequently in a place and features abruptness and great life and
property losses in a small scope. At present, flash flood is hard to forecast, and
prevention is carried out through designation of high risk regions and establishment
of warning systems.
5 Overview of Measures and Assessment of Capacity … 147

Fig. 5.29 Rainfall distribution in the past 1 h [http://ffw.mrcmekong.org/ffg.php (accessed on 18

July 2019)]

5.5 Disaster Mitigation Management

5.5.1 Flood Prevention and Drought Mitigation System

(1) Cambodia

Cambodia’s disaster management body is the National Committee for Disaster Man-
agement (NCDM) founded in 1995. All departments and organizations should coop-
erate closely with NCDM and jointly provide disaster relief in emergency circum-
stances. As an executive unit, NCDM Secretariat provides disaster management
148 Z. Ding et al.

Fig. 5.30 Transmission and release of hydrological monitoring data in Viet Nam (Tong Thi Lien.
Develop an automatic water resources monitoring system for water management in Ba river basin.
Technical exchange meeting report in August 2018, in Beijing)

suggestions to the government. Disaster management councils are also established

at provincial/municipal and regional levels.
NCDM is tasked with nationwide disaster management. Cambodia needs to
improve the capability of coping with disasters while further developing economy.
Meanwhile, because flood is the major natural disaster in the country, laying empha-
sis on the cooperation with Mekong River Basin countries is an indispensable disaster
reduction measure (Fig. 5.31).
(2) Lao PDR
In Lao PDR, Order No. 158 (1999) of the Prime Minister establishes disaster manage-
ment committees at national, provincial and regional levels, providing a foundation
for the development of disaster management policy. Decree No. 97 of the National
Committee for Disaster Management stipulates the roles and duties of its units.
Lao PDR has established Flood Prevention and disaster reduction systems from
central government to towns and villages, with attention paid to the organization
and coordination in flood prevention and mitigation. The coordination mechanism
is in the charge of disaster management committees from national to basic levels.
Although there is not specialized disaster management and mitigation law, duties
of member units of the National Committee for Disaster Management are stated
in relevant laws. In recent years, Lao PDR has started to attach importance to inte-
grate city planning and waterlogging and implement uniform disaster relief measures
(Fig. 5.32).
5 Overview of Measures and Assessment of Capacity … 149

Fig. 5.31 Flood prevention and disaster mitigation system in Cambodia (https://www.adrc.asia/
img_map/116_org.gif)

(3) Myanmar

In Myanmar, the major disaster management body is the Central Committee for
Disaster Prevention and Relief (CCDPR). CCDPR was founded according to the
guideline of the Safety and Management Committee of the National Peaceful Devel-
opment Parliament, with the aim to implement disaster prevention and reduction
measures more efficiently and led by the Minister of Internal Affairs and Religious
Affairs. Subordinate to the Central Committee for Disaster Prevention and Relief
are ten sub-committees on information and education, emergency communication,
search and rescue, loss information and rescue, loss assessment and etc.
In the flood prevention and mitigation field, the Relief and Resettlement Depart-
ment has cooperation with Department of Health, Department of Meteorology and
Hydrology, Fire Services Department, Department of Human Settlement and Hous-
ing Development, Department of Irrigation and Myanmar Red Cross Society. The
Department of Irrigation repairs and reinforces dams and watertight barriers in the
delta region where flood often occurs, and builds cofferdams in several provinces
liable to hurricane and storm surge. Once a disaster takes place, the Department of
Meteorology and Hydrology will release to the general public information and alerts
about hurricane, flood, super rainfall and river level.
In Myanmar, every stage of flood disasters is in the charge of specialized organi-
zation or department. In the flood prevention and disaster reduction field, the country
150 Z. Ding et al.

Fig. 5.32 Flood prevention and disaster mitigation system in Lao PDR (https://www.adrc.asia/
img_map/418_org.gif)

attaches importance to the transmission of flood prevention information to the general

public as well as the construction of structural facilities (Fig. 5.33).

(4) Thailand

Thailand promulgated the Disaster Prevention and Mitigation Act in 2007 and
brought it into force on November 6 the same year. The country’s flood disaster
management system is founded on the basis of this act. It has 5 distinguishing fea-
tures: 3 major decision-making and planning bodies are established at national level,
provincial level and of Bangkok City; Prime Minister or appointed Vice Prime Min-
ister serves as Commander at national level; the Department of Disaster Prevention
and Mitigation (DDPM) is authorized to be the core department for flood disas-
ter management; local governments are authorized to be responsible for local flood
disaster management according to provincial-level planning (Fig. 5.34).
Presently, Thailand’s national overall plan for disaster prevention and mitigation
is divided into three parts. Part I includes flood situation and management system,
with description of the situation of flood disasters, management system, planning
objectives and emergency work for coping with flood disasters. Part II sets forth
5 Overview of Measures and Assessment of Capacity … 151

Fig. 5.33 Flood prevention and disaster mitigation system in Myanmar (https://www.adrc.asia/
img_map/104_org.gif)

standard operation procedures. Part III discusses national safety issues related to
flood prevention and mitigation.
The Thailand National Committee on Disaster Prevention and Mitigation
(NCDPM) is a major department responsible for formulating policy proposals,
mainly including developing national plans for disaster prevention and mitigation
and arranging the disaster prevention and mitigation abilities of central government,
local governments and other related private sectors. NCDPM is led by Prime Minister
and consists of 34 members from departments, institutions and organizations related
to disaster management. At national level, Prime Minister or appointed Vice Prime
Minister serves as President of NCDPM, responsible for formulating disaster risk
management policies. At local level, provincial committees on disaster prevention
and mitigation are established.
(5) Viet Nam
At present, Viet Nam is striving to probe into the legislation of flood disaster man-
agement and gradually establish a set of legal system. Relevant legal documents
formulated and issued in recent years include the Dyke Management Law, Law on
Water Resources, Forest Protection and Development Law, Environmental Protection
Law, Land Law, Ordinance on Dyke Management, Ordinance on Flood and Storm
Control, Ordinance on Development and Protection of Water Resources Structure
and Ordinance on Structure of Hydrometeorological Survey.
The Central Committee for Storm and Flood Prevention (CCSFC) is a major
institution that formulates flood management regulations and mitigation measures
152 Z. Ding et al.

Fig. 5.34 Flood prevention and disaster mitigation system in Thailand (https://www.adrc.asia/img_
map/764_org.gif)

of Viet Nam. At the central level, CCSFC is responsible for coordination of disaster
relief. CCSFC Secretary is appointed by the Dyke Management and Flood Preven-
tion Department of the Ministry of Agriculture and Rural Development. CCSFC
is responsible for formulating flood regulations and mitigation measures, and local
emergency work is coordinated by storm and flood mitigation committees at provin-
cial level. As the highest coordinating body, CCSFC is tasked with nationwide flood
prevention and disaster reduction. It is the primary unit that implements relevant
policies.
For CCSFC, the most import task is the upgrading and maintenance of the dam
system including river dams as long as 5000 km and seaside dams as long as 3000 km.
Under the help of the United Nations Development Programme (UNDP) and the
World Food Programme (WFP), Viet Nam is restoring and reinforcing the dykes
on northern and central coasts. In addition, Viet Namese authorities have set about
5 Overview of Measures and Assessment of Capacity … 153

Fig. 5.35 Flood prevention and disaster mitigation system in Viet Nam (https://www.adrc.asia/
img_map/704_org.gif)

reinforcing dykes in the Red River Delta Region, and have started to prepare and
improve flood discharge plans of important reaches (Fig. 5.35).
Along with national economic development, Viet Nam is improving its flood dis-
aster coping mechanism in dykes, forest protection and development, water resources
utilization and other links. In the meantime, Viet Nam also lays emphasis on improv-
ing communities’ risk prevention awareness. It develops and implements a string
of social and economic development projects with the focus on flood management,
such as water conservation forests and protective forest belts. Besides, more attention
is also paid to the all-round management of flood and drought in such projects as
reservoir construction for coping with flood and drought, the project of “Living with
Flood” and reinforcement of dykes.

5.5.2 Emergency Response to Floods and Droughts

(1) Cambodia
Cambodia recognizes it’s urgent to develop integrated disaster management strate-
gies after going through the recent destructive flood and realizing similar events
may continue to take place in future. At present, Cambodia’s comprehensive dis-
aster management strategies include “flood recovery plan” and the local project of
rural disaster management “community independence and flood mitigation plan”.
The flood recovery plan aims to recover the socio-economic structure and in the
meantime, to indirectly support production and income recovery in rural areas. The
“community independence and flood mitigation plan” is a technological assistance
project that helps the Cambodian government improve community independence
during periodic floods. Four tasks are included: preparedness for and emergency
154 Z. Ding et al.

response to disasters, disaster management information system, public awareness

and warning, and coordination and cooperation.
(2) Lao PDR

In Lao PDR, national flood prevention and emergency response management is

under the charge of the National Disaster Management Office, which coordinates
and promotes relevant ministries, commissions and provincial governments to carry
out emergency response to flood disasters. In the country, emergency response to
flood disasters is usually implemented by the Ministry of Public Security and the
Science Technology and Environment Agency. Other ministries, commissions and
provincial-level governments are responsible for managing the emergency response
to floods in their respective regions in cooperation with the Ministry of Public Secu-
rity, Science Technology and Environment Agency, Lao PDR’s Red Cross Society,
Ministry of Agriculture and Forestry and Ministry of Health. In recent years, Lao
PDR has started to lay emphasis on integrating city planning and construction with
urban flood disasters and implementing uniform measures of emergency response to
flood disasters.

(3) Myanmar
Fully recognizing the importance of international cooperation in flood prevention
and mitigation, training and experience exchange, the Myanmar government often
sends officials to participate in training courses, study classes and relevant meetings
in foreign countries. On the other hand, it holds study classes and workshops through
cooperation with or sponsorship of international and regional organizations such as
the UN Economic and Social Commission for Asia and the Pacific (ESCAP), World
Meteorological Organization (WMO) and UN Development Programme (UNDP).
(4) Thailand

Thailand Department of Disaster Prevention and Mitigation (DDPM) is a major

disaster management institution founded in 2002 and subordinate to Ministry of
Interior. It is responsible for the coordination of departments related to disaster pre-
vention and mitigation. According to law, Minister of Interior should be Commander
in Chief in emergency circumstances of disasters, especially large-scale disasters.
DDPM Director General is NCDPM Secretary General. When a disaster takes place,
relevant officials should be on-site commander.
Related decision-making organizations, at national level, provincial level and of
Bangkok, are led by Prime Minister or appointed Deputy Prime Minister, provin-
cial governor and Mayor of Bangkok respectively. Decision-making organization at
each level includes the corresponding disaster prevention and mitigation committee.
Director at provincial level can ask for support from neighboring provinces according
to disaster condition. Regional director can ask emergency teams, civil defense vol-
unteers and even NGOs to participate in disaster relief. Local emergency units, such
as civil defense centers, should respond rapidly to disasters; disasters with which
emergency units are unable to cope should be dealt with by competent units, such as
5 Overview of Measures and Assessment of Capacity … 155

civil defense centers and police and fire stations; backup teams should be responsi-
ble for manpower, equipment and technology support according to the requirements
of emergency units or competent units, and all government organs, NGOs, regional
civil defense centers and armed forces are backup units.
Subordinate to DDPM, the Disaster Prevention and Mitigation Academy (DPMA)
carries out flood disaster management trainings oriented to practitioners, managers
and local government officials. DPMA is founded by the Ministry of Interior. It has
become the most important flood management education base, and carries out flood
prevention and mitigation education actively in elementary schools.

(5) Viet Nam

Viet Nam attaches great importance to public involvement and use of resources of
international organizations. When a disaster takes place, government, labor unions,
youth and women associations and other social organizations will organize donation
activities proactively to help victims in affected areas. In disaster relief and recovery,
emphasis is laid on mutual assistance. Public awareness is improved, especially
through media, so that officials at basic level and from various departments and local
governments receive a good training. As a result, awareness of government officials
and the general public is improved.
The government provides preferences to flood disaster projects every year, and
continues to increase disaster mitigation spending. Priority is given to flood preven-
tion and mitigation plans and programs, such as afforestation project, dyke renovation
project, reservoir project, landslide project, “Living with Flood” project and etc.

5.6 Assessment of Flood Prevention and Drought Relief

Capacity

5.6.1 Cambodia

Because flood is a major disaster in the country, Cambodia has a relatively complete
organizational system of flood mitigation, with disaster management departments at
national, provincial and community levels. Non-structural measures for emergency
response to flood are implemented through some “flood management programs” and
“disaster mitigation programs”. Limited to economic and other factors, however,
apart from part of the revetment works in cities such as Phnom Penh, Kratie and
Kampong Cham, there is no flood prevention facility established actively in most
regions. Though 14 reservoirs are planned with a gross storage capacity of 18.89 km3 ,
only one has been buit up, with a storage capacity of 120,000 m3 . Based on the
population and planned storage capacity of Cambodia within Mekong River Basin,
the per capita reservoir storage capacity will be 1629 m3 /people, a little higher than
156 Z. Ding et al.

the world average level of 1000 m3 /people.2 But the current per capita reservoir
storage capacity is only 0.01 m3 /people, much lower that of the world average.
Residents surrounding the Tonle Sap Lake and flood plains of Mekong River live on
fishery and catch and are accustomed to flood. On the contrary, flood reduction is
regarded as damage to their life. Rainfall is small and flash flood is less destructive in
the mountainous regions in upper Tonle Sap Lake in western Cambodia, but eastern
mountainous regions are liable to floods from upper mountainous regions in Lao
PDR and Viet Nam. In the meantime, the provision of huge capacity of flood storage
and detention in the Tonle Sap Lake and flood plains (Tonle Sap lake of 70 km3 ,
floodplains of 35 km3 ) of Mekong River reduces the flood prevention pressure of the
delta region in the lower reaches in wet season and also provides continuous water
replenishment in dry season. In recent years, along with the improvement of flood
prevention capacity in upper regions and the increase of irrigation water diversion,
flood level of the Tonle Sap Lake and flood plains of Mekong River in Cambodia
has been no longer as high as in history (e.g. 2000), so some regions in the flood
plains of Mekong River, especially near Phnom Penh, have been developed, such
as reclamation and banking for infrastructure construction, which reduces the flood
storage capacity of the flooded area to some extent. But the development should be
a natural balance process between water and human as long as it is conducted within
a scope.
For extremely uneven spatial distribution of irrigation engineering system, the
drought relief capacity of structural measures distinguished much among regions
within Cambodia. Presently, irrigation coverage rate exceeds 10% universally in the
region centering on capital Phnom Penh and Tonle Sap Lake, enabling the region with
strong drought relief capacity. However, there is no irrigation project in northeastern,
northwestern regions with no irrigation area. Though one reservoir of 120,000 m3
storage capacity has been buit, the irrigation function of it is very limited. From
the perspective of financial support, the per capita GDP of Cambodia is 1270 USD,
ranking 158 in the world, much less than the world average. Comparatively limited
financial support constrains the government input to drought relief. Based on the
above two aspects, the drougt relief capacity of Cambodia is not outstanding, and
there are great differences between different regions.

5.6.2 Lao PDR

Lao PDR has established a flood prevention and mitigation system from central gov-
ernment to towns and villages. Lao PDR National Disaster Management Office is
responsible for organizing and coordinating emergency management of disasters and
enabling emergency response to flood. In Lao PDR, the flood prevention engineer-
ing system is weak on the whole, and banks along the Mekong River, except some
big towns and cities with revetment, are basically in a natural state. The revetment

2 Hydropower and Dams, World Atlas, 2008.

5 Overview of Measures and Assessment of Capacity … 157

account for less than 1% of the whole bank. Therefore, riparian regions are easily
flooded once water level of Mekong River is high. Except for cities like Vientiane,
Pakse, Savannakhet, Champasak are installed with sluices and pump stations for
flood prevention, tributaries of Mekong River in Lao PDR are merely installed with
flood prevention project, so backflow effect occurs easily in case of high water level
in Mekong River and floodwater may affect large areas in upper tributaries. In most
mountainous regions, rainfall runoff is frequently and flash flood takes place very eas-
ily, but the flash flood warning construction gradually conducted in recent years can
relieve the damage of flash flood to some extent. The existing, under construction and
planned reservoirs in Lao PDR have a gross storage capacity of 58.62 km3 . Accord-
ing to the population and planned (including the existing and under construction)
reservoirs of Lao PDR within Mekong River basin, the per capita storage capacity
is 11,060 m3 /people, much higher than the world average (1000 m3 /people), after
Norway who ranking number two in the world (14,000 m3 /people, see Fig. 5.36);
the per capita storage capacity based on the existing and under construction projects
is 5507 m3 /people, which is 5.5 times of the world average, and close to than of
German (see Fig. 5.36). Therefore, construction of these reservoirs could control
some floods. However, these reservoirs are not designed for flood prevention, so
floodwater can be only discharged instead of being stored. After the collapse of the
Xe Pian-Xe Namnoy Reservoir in 2018, adjustment of policy may help to promote
the flood prevention function of reservoirs.
Per capita GDP of Lao PDR in 2016 is 2339 USD, ranking number 138 in the
world, much less than the world average. With limited investment to water conser-
vancy projects in history, Lao PDR has only 2000 plus irrigation projects in total,
Per capita reservoir capacity(m^3/person)

Venezuela

Germany

Argentina

NewZealand
Canada

Norway

Iceland

Australia

America

Brazil

China

Fig. 5.36 Per capita reservoir storage capacity of typical countries in the world (http://www.
chincold.org.cn/chincold/upload/news/lin200911254534542.pdf)
158 Z. Ding et al.

due to which the irrigation percentage is less than 1% in nearly 95% territorial area
of Lao PDR. Meanwhile, most of these irrigation projects are of a small size, less
than 200 ha, and located in riparian area along tributaries and in small floodplains of
lower reaches of Mekong River. At present, 10 reservoirs have been built on Mekong
river tributaries in Lao PDR, but they are used mainly for power generation, and
the contribution to irrigation is negligible. Overall, the irrigation projects are with
limited quality and uneven distribution, as a result, the water supply per unit area of
cultivated land and the irrigation percentage are low, which affects the water con-
servancy guarantee ability of drought resistance. It should be noted that Lao PDR is
exerting great efforts to invest in water conservancy projects, and will build another
2768 irrigation projects and increase irrigation area by 240% for non-paddy fields
and 460% for dry-season paddy by 2030 and thereby, improve the drought resistance
capacity.

5.6.3 Myanmar

In Myanmar, the Central Committee for Disaster Prevention and Relief is a major
body of disaster prevention and mitigation that has multiple sub-committees of infor-
mation and education, emergency communication, search, loss information and res-
cue, loss assessment and etc., and has cooperation in emergency rescue and response
for disaster prevention and mitigation with Department of Health, Department of
Meteorology and Hydrology, Fire Services Department, Department of Human Set-
tlement and Housing Development, Department of Irrigation and Myanmar Red
Cross Society. Myanmar’s flood prevention engineering systems, such as dykes,
reservoirs and gate dams, are largely located on the Irrawaddy River, Salween River,
Chindwin River, Sittaung River and other major rivers. Flood prevention capacity
of these projects is relatively good. Few flood prevention projects are located in the
Mekong River Basin. But influence is not great because the population in the region
is small and there is no prominent flood problem.
Per capita GDP of Myanmar is 1196 USD, which is the lowest among Mekong
countries, the contribution of agriculture to GDP is about 1/4. To cope with drought,
flood and other extreme disasters, the Ministry of Agriculture, Livestock and Irriga-
tion of Myanmar established Irrigation and Water Utilization Management Depart-
ment (IWUMD) responsible for managing nationwide irrigation and flood prevention
facilities and maintaining the operation of irrigation systems and flood prevention
dykes across the country. At present, about 1.58 million ha of arable land in Myanmar
is irrigated effectively, and about 0.25 million ha of land is installed with irrigation
canals. Due to limited financial support, the existing projects are mainly located in
certain paddy land, the coverage is limited. In sum, despite the unified and coordi-
nated official management, the relatively limited water conservancy project support
capacity has limited the drought resistance of Myanmar to some extent.
5 Overview of Measures and Assessment of Capacity … 159

5.6.4 Thailand

Thailand has a complete meteorological and water level monitoring and warning
system that can monitor rainfall and water regime in real time, provide relevant
forecasting, and warning against flash flood. Its disaster mitigation and relief system
is also sound, covering organizational systems of emergency response at national,
provincial, regional and response team levels. Most tributaries (at least 5) on Korat
plateau are installed with sluices and dams, having regulation capacity of tributary
flood and Mekong flood. The left tributaries without sluices are apt to influence
of high water level of Mekong River. Urban areas along Mekong River are mostly
protected by revetments, 300 km out of the total length (955 km) of the Mekong
mainstream in Thailand are revetment or embankment works, which could withstand
the high water level of Mekong River. But the natural low banks are still apt to
flooding during the high water level period of Mekong River. Thailand has not set up
a specific agency to in charge of the national flood defence coordination, resulting
in weak national coordination ability. The gross storage capacity of the existing
and under construction reservoir in Thailand is 3.57 km3 . The per capita reservoir
storage capacity is 155 m3 /people, lower than the world average of 1000 m3 /people.
For flood prevention capacity is not designed for these reservoirs, they could not play
an important role in flood detention and regulation.
Per capita GDP of Thailand in 2016 is 5980 USD, ranking the top among Mekong
countries. With strong financial support, Thailand has established a full-coverage and
relatively complete hydrological and meteorological monitoring and forecasting sys-
tem, including 11 major hydrological stations on mainstream Mekong River, which
is used to get real-time water (e.g. water level, runoff, etc.) information, diagnoses
occurrence of flood and drought in time, and provide scientific decision-making
support for emergency dispatching of irrigation water. With regard to engineering
support, Thailand ranks top in countries along Mekong River by the quantity of irri-
gation projects, thanks to which nearly 14% of northeastern Thailand (in Mekong
River Basin) is covered with reservoirs, weirs, channels and other water conservancy
facilities. So dense irrigation projects assure the share of irrigation area in northeast-
ern Thailand exceeds 10%, and provide reliable emergency water sources for local
crops’ production. On the whole, Thailand is at a high level and has a good capability
of drought resistance because of good capacity in drought prevention and mitigation,
powerful economic support capacity and water resources guarantee capacity. Nev-
ertheless, it should be pointed out that current irrigation aims mainly at wet-season
paddy, and the irrigation guarantee capacity for dry-season paddy remains insuffi-
cient. As a matter of fact, drought is the commonest natural disaster in the dry season
in Thailand that seriously influences the growth of dry-season paddy and other crops.
Therefore, improving the irrigation guarantee capacity in dry season is critical for
improving drought resistance level and capacity of Thailand.
160 Z. Ding et al.

5.6.5 Viet Nam

Viet Nam has a sound legal system for disaster prevention and mitigation. Disas-
ter mitigation committees at national, provincial, regional and community levels
were established. And the capability of emergency response to disasters is improved
through improving public awareness and personnel trainings and drills. Mekong
Delta is a major food production region and aquaculture region of Viet Nam that is
liable to the influence of flood, drought and other disasters. With a low-lying terrain
and few dykes and other flood prevention projects, the region is definitely liable to
long-term and large-scope flooding when water level of Mekong River is high. In
recent year, Viet Nam has avoided flooding through migration or lifting residential
housing foundation, which is also an active preventive measure. In addition, if reser-
voirs on upper Mekong River are all capable of flood prevention, flooding of the
delta region may be relieved to some extent when regulation is conducted well and
floodwater is stored normally in Tonle Sap lake and floodplains along Mekong River
in Cambodia. There are more than 10 reservoirs on Se San River and Ea Krong River
in Central Highlands on tributaries of Mekong River, with a gross storage capacity of
3.16 km3 . The per capita reservoirs storage capacity of Viet Nam within the Mekong
River basin is 152 m3 /people, lower than the world average of 1000 m3 /people. Nev-
ertheless, these reservoirs are desined mainly for power generation without flood
prevention capacity, could not play a role of flood storage and regulation.
Mekong Delta is “a land flowing with milk and honey” in Viet Nam, where drought
is a major natural disaster. Per capita GDP of Viet Nam in 2016 is 3179 USD, after
Thainland. Besides, the Viet Namese government establishes a uniformly coordinated
drought resistance and mitigation system, which assures the scientific formulation
and implementation of flood prevention measures. In the region, because of the con-
struction of a set of large irrigation facilities, Viet Nam ranks top in countries along
Mekong River by irrigation area (nearly 2 million ha or 50%). It means nearly half the
paddy fields in the region can receive water replenishment through irrigation. This
region is faced with typical drought caused by the shortage of freshwater because
saltwater intrusion leads to soil salinization and water salinity rises which cannot
be used for irrigation. Hence, Viet Nam prevents saltwater from flowing into inland
rivers through building tide gates and thereby, withstands freshwater shortage-caused
drought. Overall, Viet Nam has good resistance ability against drought to make up
the locational weakness of Mekong Delta. It is notable that due to limited finan-
cial input, the existing tide gates are mainly located on small tributaries, making
the mainstream and riparian area apt to salinity intrusion. Better resolvement of
salinity intrusion problem is critical for further improving the drought resilience of
Viet Nam.
5 Overview of Measures and Assessment of Capacity … 161

5.6.6 Overall Situation at the Basin Level

In general, the overall flood prevention capacity of the Mekong River Basin is weak.
The Tonle Sap Lake and the Mekong River floodplain in Cambodia provide large
flood prevention capacity. Due to the insufficient flood prevention engineering sys-
tem, it is inevitable that it will be exposed to flood damage. Lao PDR is mostly hilly
and mountainous. Due to the weak flood prevention works along Mekong River, it is
more vulnerable to floods from the mainstream. Myanmar is in the upper mountainous
area in the Mekong River Basin. Although there is no flood prevention engineering
system such as levees on mainstream Mekong River, the flood hazard is relatively
small due to the scarcity of riparian population. Despite the high terrain, and some
flood prevention projects such as revetment, sluice and pump, in the case of regional
heavy rain and the high water level of Mekong River, it will still suffer serious dam-
age. Viet Nam is at a disadvantage in flood prevention since it is in the delta area. In
addition, its flood prevention works such as embankment are weak. So it is easy to
suffer from serious flooding when floods from the upper reaches of Mekong River
encounter the high tides.
The situation in the Mekong River Basin countries varies greatly. They are afraid
of the flood of the Mekong River, while but they need the water from it. With-
out an effective coordination and cooperation mechanism, flood prevention at the
basin level is hard to be effective. The flood management cooperation mechanism
adopted by some transboundary rivers in the world may be worth learning, for exam-
ple the Rhine River in Europe. It originates from the northern foothills of the Alps
in Switzerland, flows through six countries including Switzerland, Liechtenstein,
Austria, France, Germany and the Netherlands, and goes into the North Sea near
Rotterdam in the Netherlands, with a total length of about 1232 km. At the interna-
tional level, an institutionalized cooperation mechanism has been established for the
Rhine River and ambitious flood prevention targets set (Rhine Action Plan: ICPR,
1998) to provide framework for joint flood management in the whole basin. Flood
risk management measures for the whole basin include flood risk zoning, coordina-
tion mechanisms between countries, and economic compensation schemes between
upstream and downstream regions. The results of many years of operation show that
the cooperation mechanism and framework of the Rhine River, a transboundary river,
has been successful.
The Lancang-Mekong River flows through China, Myanmar, Laos, Thailand,
Cambodia and Viet Nam and goes into the South China Sea in the Delta of Viet
Nam. The total length of the Lancang-Mekong River is about 4880 km, which is
much longer than that of the Rhine River. Since it flows through many countries,
it also faces the problem of flood prevention cooperation and coordination among
countries. Therefore, the successful and advanced flood management experience of
similar transnational rivers such as the Rhine could be learned from and establish a
flood prevention cooperation mechanism and framework at the basin level. The flood
prevention structural facilities in the basin could be strengthened, other activities
like improving the flood prevention capacity, carrying out flood risk zoning and
162 Z. Ding et al.

assessment, developing flood detention and control functions of reservoirs in the

basin, coordinating flood regulation between upstream and downstream countries to
make full use of flood resources, giving full play to their respective regional and
geographical advantages, and appropriately adopting economic adjustment levers
could be taken to achieve unified flood management of the whole basin.

5.6.7 Joint Efforts to Cope with Flood and Drought

For all regions of the world, water security is a prerequisite for economic and political
stability in any country and region. In order to achieve water security, countries in the
basin must establish mutual trust and cooperation, have a unified understanding of
each other’s rights and obligations, and build a platform for efficient cooperation in
science and technology, information sharing and basin management.3 The Lancang-
Mekong River Basin countries are exploring the establishment of a whole basin
cooperation mechanism and have carried out a lot of cooperation in the field of
disaster prevention and relief.
In 1995, Laos, Thailand, Cambodia and Viet Nam jointly established the Mekong
River Commission (MRC). The MRC and its member countries have played an
important role in the sustainable use of water resources in the Mekong River Basin,
especially the Regional Flood and Drought Management Center (RFDMC) for fore-
casting and warning the floods of the Mekong River, which is of great significance
for flood prevention in the basin. China and Myanmar have maintained close cooper-
ation with the four downstream countries through dialogue mechanism of the MRC.
Since 2002, China has regularly submitted to the MRC the hydrological data of the
Lancang River in flood season and the emergency hydrological information during
the dry season. In 2010, in order to help downstream countries cope with the extreme
drought, the Ministry of Water Resources of China provided hydrological informa-
tion to the MRC for emergency response. Later, when encountering various special
climates such as typhoons, the Chinese side repeatedly informed the MRC the reg-
ulation information of the upstream reservoir. The MRC and its member countries
highly valued the support from China and considered the hydrological information
provided by China is of great significance to the flood prevention and drought relief
work of the lower reaches of the Mekong River.
Information sharing and technical cooperation have also been carried out among
the countries in the Lancang-Mekong River basin. For example, China has signed
memorandum of understanding on water resources cooperation or agreement on
reporting flood information with Laos (2014), Thailand (2014 and 2016), Cambodia
(2013), and Viet Nam (2002). In recent years, the Chinese side invites more than
100 water conservancy officials and experts from downstream countries and MRC
to visit China every year, to visit the reservoirs in the Lancang River, the Three

3 Zhong et al. Rivers and reciprocity: perceptions and policy on international watercourses, Available

Online 29 February 2016, wp2016229; 10.2166/wp.2016.229.

5 Overview of Measures and Assessment of Capacity … 163

Gorges Project, the South-to-North Water Diversion Project, and the water conser-
vancy projects in Beijing, Shanghai, Guangzhou, and to share with them Chinese
experience. At the same time, the Chinese side has also sent several delegations to
the Mekong River Basin to carry out technical exchanges and understand the concerns
of downstream countries.
In the case of major disasters, the countries in the Lancang-Mekong basin actively
carried out disaster alleviation cooperation and jointly cope with the challenges of
floods and droughts. For example, the Chinese government implemented emergency
water supplement in response to the major drought in the Lancang-Mekong River
Basin in 2016 caused by the El Nino phenomenon. According to the joint assessment
of emergency water supplement effect carried out by Chinese side and Mekong River
Commission, the emergency water supplement from the Lancang River increased the
amount of water flowing into the Mekong River by about 1000 m3 /s, which raised
the water level along the main stream of Mekong River by 0.18–1.53 m. A letter from
the Mekong River Commission Secretariat CEO Dr. Pham Tuan Phan to the Minister
of Water Resources of China stated: “the Secretariat views this decision (emergency
water supply) as a kind and considerate gesture exhibited by a good neighbor and
friend to the Mekong countries. This has also shown China’s sincerity in the cooper-
ation with the downstream countries, especially within the context where China itself
was also suffering from drought condition, which has affected its household water
supply and agricultural production. The Secretariat would like to extend its profound
and sincere thanks to the People’s Republic of China for its consideration and spe-
cial arrangement of such a relief measure in the interest of downstream countries.”
When severe flooding occurred in Thailand in 2011 and Myanmar in 2015, the Chi-
nese government was invited to send a flood prevention consulting expert group to go
to the disaster area, carry out disaster reduction consultation and present a high-level
consultation report, which was highly recognized by the local governments.
Since the establishment of the Joint Working Group on Lancang-Mekong Water
Resources Cooperation, the work of flood prevention and disaster mitigation in the
basin has been highly concerned. In the “Five-Year Action Plan for Lancang-Mekong
Water Resources Cooperation” drafted by the six countries in the basin, addressing
global climate change and disaster prevention and mitigation is an important part.
It proposes to strengthen cooperation between the six countries in preventing floods
and droughts, improve the structural facilities system, monitoring and early warning
levels, and emergency support mechanisms, raise public awareness of flood and
drought disasters, enhance the ability to prevent floods and droughts, reduce flood
and drought losses, to ensure the sustainable economic development in this region.

5.7 Summary

This chapter investigates and analyzes flood prevention and drought relief measures
in the Mekong River Basin countries, and evaluates their respective flood prevention
and drought resistance capabilities. Related conclusions are as follows:
164 Z. Ding et al.

(1) Flood prevention structural measures in the Mekong River Basin countries are
insufficient. Thailand has some urban revetment projects in the main urban
sections of mainstream Mekong River. And there are sluices, dams (pumps)
on Mekong River tributaries in Khorat. These projects can to a certain extent
withstand the high water level flood from mainstream Mekong River. Laos also
has some revetment works in the main urban sections of mainstream Mekong
River, but most of the river banks and tributaries are in the original natural
state and are susceptible to high water level floods from mainstream Mekong
River. Due to its location in alpine valley, small range of river section and
sparse population in the area, there is basically no flood prevention project on
mainstream Mekong River in Myanmar. In Cambodia, except for a few large
cities such as Phnom Penh with river revetment projects, there is basically no
flood prevention project in other areas. Viet Nam has some revetment measures
in the Mekong Delta. There are also some tidal gates and other facilities in the
offshore area. However, these structural measures are weak and cannot withstand
the high water level impact of the Mekong River.
(2) The Mekong River Basin countries have established their own disaster preven-
tion and mitigation organizations, covering disaster prevention and mitigation
departments at different levels, from the state to the province, city, and village.
(The actual emergency response situation of each country at the time of the
disaster is not well documented). Under the organization and coordination of
the Mekong River Commission, countries in the Mekong River Basin (except
for Myanmar, which is not a member country) have achieved good results in
monitoring and forecasting the water conditions of Mekong River, and have a
crucial role in the flood forecasting of Mekong River. However, there are no spe-
cial departments for coordinating flood prevention. The flood prevention tasks
are the responsibility of each region and relevant departments. For example, the
irrigation department in Thailand is responsible for flood prevention in its own
irrigation area. The reservoir is subordinate to power department or irrigation
department. There is no consideration of flood prevention function by the reser-
voirs. The upstream flood regulation may cause downstream disasters, which
means the flood prevention cannot be comprehensively coordinated.
(3) The overall flood prevention capacity of the Mekong River Basin countries is
to be improved, and their geographical location and flood prevention status are
different. From the perspective of the river basin, the countries along the Mekong
River need to work and coordinate together to achieve the flood prevention effect
at the basin level. For example, the mainstream Mekong River and tributary
reservoirs shall be set with certain flood prevention capacity to regulate floods.
The floodplain of mainstream Mekong River and Tonle Sap Lake in Cambodia
have the most suitable locations and capacity for flood storage and detention.
This area plays a very important role in ensuring the flood prevention security
of the Mekong Delta.
5 Overview of Measures and Assessment of Capacity … 165

(4) As far as the national level is concerned, there are certain differences in the
national economic support capacity and water conservancy project support capa-
bilities among the five countries in the Mekong River Basin, resulting in different
drought relief capabilities. In the basin, the per capita GDP of Thailand and Viet
Nam is more prominent, which has promoted the economic investment of the two
countries in establishing emergency response systems and engineering projects
for drought and disaster relief. Specifically, a series of large and medium-sized
irrigation systems and projects have been built on Mekong River in Viet Nam
and Thailand, forming a high-coverage irrigation network and making the irri-
gated area accounting for 10% or even 30% of local land, which means it to
some extent possesses the ability of drought relief. The distribution of irrigation
projects and irrigation districts in Cambodia is uneven, mostly concentrated in
the southern areas such as the capita Phnom Penh and Tonle Sap, and there
are almost no irrigation areas in other areas, resulting in significant regional
differences in drought relief capacity of water conservancy structural measures.
Lao PDR and Myanmar are two countries with relatively weak economies in
the basin, which limit their investment in drought relief structural measures. At
present, in the basin countries, the number of irrigation projects and the size of
irrigation districts in Laos are behind the others. It has the lowest proportion of
irrigation area (less than 1%). It should be noted that Lao PDR is expanding
investment and development of hydropower projects, increasing the number of
irrigation projects and the scale of irrigation districts, and has great potential in
improving national drought relief capacity.
(5) At the basin level, in the process of drought relief, the upstream countries of
Mekong River Basin directly draw water from Mekong River to the tributaries
through diversion irrigation project for water use of local rice fields. Currently,
there is a lack of coordination of emergency water between upstream and down-
stream countries. Although the water supply from Lancang River reservoir may
relieve the hydrological drought of mainstream Mekong River to a certain extent,
it has limited influence over the drought relief in Mekong Delta. Therefore,
the basin countries should further strengthen coordination and communication,
carry out scientific deployment and management of emergency water sources,
and tap the drought relief potential of the whole basin.
(6) Working together to deal with floods and droughts has become the consensus
of the countries in the basin. While building their own flood prevention and
drought relief systems, all countries actively carry out international coopera-
tion on information sharing, technical exchanges and assistance to jointly solve
the problem of flood and drought in the basin. From the establishment of the
Mekong River Commission to setting up of Lancang-Mekong Water Resources
Cooperation mechanism, the platform for flood prevention and drought relief
cooperation in the basin has become increasingly diversified and broader. The
joint efforts of the countries in the basin have provided new opportunities for
coordinating and coping with floods and droughts from the perspective of the
whole basin.
166 Z. Ding et al.

Dr. Ding’s research primarily focuses on hydro-hydraulic modeling, flood risk management, flood
disaster assessment and water resources application of remote sensing and GIS. He was the Pro-
gram Manager of important national and provincial level projects. The number of investigated
or participated projects and programs exceeds 50. He has written 2 technology standards (or
specifications) and 5 books, has published over 40 papers.

Dr. Zhang’s research primarily focuses on drought monitoring and forecasting, hydrometeorol-
ogy, and climate change. He has developed an operational drought monitoring and forecasting sys-
tem over China and the Mekong River basin, respectively, in support of drought early-warning. To
date, he has been funded as the director of Chinese Postdoctoral Science Foundation, and partici-
pated in more than 10 national-level funding projects. He has published more than 15 papers in the
famous high-impact journals like Nature Climate Change, and also been involved in 3 published
books like the AGU Geophysical Monograph Series.

Dr. Han’s research primarily focuses on flood risk management theory, flood risk analysis. He was
the program manager of important national and provincial level projects. The number of investi-
gated or participated projects and programs exceeds 30. He has written 2 books, has published
over 30 papers.

Dr. Liu’s research primarily focuses on hydrological processes and modeling, flood management,
and transboundary river water cooperation. She has over 5 years’ research experiences in trans-
boundary water cooperation of China. She was a participant of various Lancang-Mekong water
resources cooperation activities. She is an editor of a joint publication by MWR of China and
MRC in 2016. She is Program Managers of 3 international cooperation projects with Mekong
River countries. She is author of over 10 journal papers and 1 book.
Chapter 6
Main Findings and Recommendations

Hui Liu and Xiangpeng Mu

Abstract Based on the investigation and analysis of the losses and basic
characteristics of flood and drought disasters and development status of flood pre-
vention and drought relief projects in the Mekong River Basin countries, and eval-
uation of the flood prevention and drought relief capabilities of the countries in the
basin, the main conclusions are summarized in this chapter from the angle of cause,
flood, drought and management. To improve the flood prevention and drought relief
capability of the basin, suggestions are proposed from the following three aspects.
(1) Exploit potentialities and enhance the ability of countries to cope with natural
disasters; (2) Make overall plans and coordinate to enhance the disaster mitigation
ability from the whole-basin perspective; (3) Strengthen cooperation and carry out
technical exchanges and mechanism building.

Lancang-Mekong River is a trans-boundary river that flows through the most South-
east Asian countries and has produced splendid social culture and rich ecosystems
along the coasts of China, Myanmar, Lao PDR, Thailand, Cambodia and Viet Nam.
It is a natural link among six countries along the basin. Although the scale of impact
of weather system is often smaller than that of the Lancang-Mekong River Basin,
the geographical proximity and the natural hydraulic links of the Lancang-Mekong
River Basin make the upstream and downstream and the left and right bank countries
closely linked with the disaster situation, thus making the six countries in the basin
face common natural disasters and water resources challenges.
Based on the flood and drought problems faced by the five countries in the Mekong
River Basin, this assessment report investigated and analyzed the losses and basic
characteristics of flood and drought disasters and development status of flood preven-
tion and drought relief projects in the Mekong River Basin countries, and evaluated

H. Liu (B) · X. Mu
China Institute of Water Resources and Hydropower Research, A-1 Fuxing Road,
Haidian District, Beijing 100038, China
e-mail: [email protected]
X. Mu
e-mail: [email protected]

© China Water & Power Press and Springer Nature Singapore Pte Ltd. 2020 167
H. Liu (ed.), Flood Prevention and Drought Relief in Mekong River Basin,
Springer Tracts in Civil Engineering,
https://doi.org/10.1007/978-981-15-2006-8_6
168 H. Liu and X. Mu

the flood prevention and drought relief capabilities of the countries in the basin. The
main conclusions are as follows:

Cause

(1) Affected by topographic features, southwest monsoon and tropical cyclone, the
Mekong River Basin has formed the spatial distribution characteristics of water
resources with increasing rainfall from west to east, as well as heterogeneous
flood and drought challenges.
(2) Flood in this region is usually caused by tropical cyclone and southwest mon-
soon, which brings heavy or lasting rainfall. Drought is usually directly related
with rainfall deficit and high temperature, some extreme droughts are related
with El Nino. Fresh water deficit in the delta area is a complex result rooted in
strong tide, rainfall deficit and low runoff from upstream.

Flood

(3) Flood is the main form of natural disaster threatening the Mekong River Basin
countries. The number of deaths caused by flood is higher in the basin, with
Cambodia and Cuu Long River Delta in Viet Nam accounting for the largest
proportion of deaths and northeastern Thailand and Lao PDR accounting for
a relatively small proportion. Floods have a significant impact on agriculture.
From the regional distribution of affected agriculture, Cambodia, Cuu Long
River Delta in Viet Nam and northeastern Thailand are the main affected areas.
From the perspective of economic loss in Mekong River Basin of the coun-
tries caused by floods, Cambodia has the largest amount of losses, followed by
Thailand and Viet Nam, and Lao PDR with the fewest losses.
(4) There are two main types of flood in the Mekong River Basin. For the riverine
flood, it could be predicted and effectively prevented from thereby damages
with proper measures. For the flash flood, it is usually local and hard to predict.
It is important for the local agencies to develop monitoring and early warning
system, as well as carry out immediate relief and aid.
(5) Based on yearly flood peak analysis of 1985–2016, it shows that the maximum
flood peak happened in 2002 and 2008 at the upper reach of Mekong River
Basin (Chiang Saen, Luang Prabang and Nong Khai), with an extreme peak of
around 20,000 cumec; while that of the middle reach occurred in 1991, 1996,
1997, 2000, 2001 and 2011, with an extreme peak of 70,000 cumec at Stung
Treng.
(6) By defining the onset and withdrawl date of flood season, the annual flood
volume and duration at mainstream stations along Mekong River was calculated
and analyzed. The flood volume shows an increasing trend from upper to lower
reaches, with 54 km3 at Chiang Saen and 306 km3 at Stung Treng. The flood
duration distinguishes among years and stations, with average ranging from
128 days to 135 days.
6 Main Findings and Recommendations 169

(7) Based on flood routing analysis in the wet season of 2000, it is found that the
time for Mekong River mainstream flood peak propagting from Chiang Saen
station to Kratie station is about 14–15 days. As to damaging flood composition,
the gauged flood data at Stung Treng station and simulation analysis results
show that the contribution rate of Mun River, Sekong River and Nam Ngum to
Stung Treng flood volume is the highest, 15.9%, 8.6% and 6.0% respectively;
the contribution rates of Nong Khai-Nakhon Phanom, Pakse-Stung Treng and
Nakhon Phanom-Pakse are higher than other regions, 25.0%, 21.3% and 19.6%
respectively.
(8) The flood-prevention structural system in the Mekong River Basin countries
is to be improved. Although most of the major cities along the coast have
embankment or revetment works, most of the reaches can hardly withstand the
impacts of high water level floods in the Mekong River. There are few large
reservoirs in the basin. In Cambodia, the floodplains in the mainstream Mekong
River and Tonle Sap Lake are the naturally formed locations in the basin and
flood detention areas with the most suitable capacity. The existence of the flood
detention area plays an important role in ensuring the flood prevention safety
of the Mekong Delta.

Drought

(9) Drought has a large scope of influence, lasts for a long time and causes huge
economic losses. Compared with floods, droughts occur less frequently, but
cause huge losses and the total losses show a significant increase trend. From
the spatial distribution of drought losses, drought losses in northeastern Thai-
land, Cambodia and Viet Nam are higher, while drought losses in Lao PDR
are relatively low.
(10) The results of meteorological drought analysis show that the drought severity in
northeastern Thailand, most of Cambodia and Myanmar has increased over the
past half century, especially in some parts of northeastern Thailand; in most
parts of the basin, the frequency of meteorological drought is close to 25%
due to low rainfall, especially in northeastern Thailand and Cambodia; and
Cambodia and Viet Nam Delta in the lower reaches are more liable to severe
and exceptional droughts. The results of hydrological drought analysis show
that there is no obvious drought at Chiang Saen station and Mukdahan station
after 2005; while the downstream Stung Treng station experienced more severe
droughts in 2010 and 2016 (SRI approaching −2). By analyzing the causes of
typical drought events, it is found that the main reason for the drought in the
Mekong River Basin is the extraordinarily less rainfall (compared with that in
the same period in history) caused by El Nino and anomalies of atmosphere-
ocean circulation system.
(11) Because of the differences between the national economic supporting capacity
and development of water conservancy projects, the drought resistance abil-
ity of different countries is different. Among them, Viet Nam and Thailand
170 H. Liu and X. Mu

have built high-density, large-scale irrigation projects and irrigation network,

providing a reliable engineering guarantee for the development of drought
resistance. The distribution of irrigated areas is uneven in Cambodia and the
ability of drought resistance in different regions is different. The number of irri-
gation projects and irrigation area in Lao PDR is relatively small and drought
resistance ability is slightly inadequate.

Management

(12) The Mekong River Basin countries have established their own organization
system for disaster prevention and reduction, covering different levels of dis-
aster prevention and reduction departments from the state to provinces, cities
and villages. Under the organization and coordination of the Mekong River
Commission, the countries in the basin have carried out better monitoring,
forecasting and early warning of the water regime of the mainstream Mekong
River. However, all countries, except Thailand, do not have specific flood pre-
vention departments, so it is difficult to carry out unified flood prevention
command, regulation and decision-making. At the basin level, with regard to
emergency water for drought resistance, the countries are still in the state of
giving priority to their own needs at present. Although there is joint drought
relief cooperation among countries in the basin, the scientific allocation and
management of water resources have not been carried out from the perspective
of the whole basin.
(13) Working together to deal with flood and drought disasters has become the con-
sensus of all countries in the basin and joint efforts have provided a new oppor-
tunity for overall and coordinated response to flood and drought disasters from
the perspective of the whole basin. While building their own flood prevention
and drought relief systems, all countries actively carry out international coop-
eration in information sharing, technical exchange and assistance to jointly
solve the flood and drought problems in the basin. From the establishment of
the Mekong River Commission to the establishment of the Lancang-Mekong
Water Resources Cooperation, the platform for basinwide flood prevention and
drought relief cooperation has become increasingly rich and broad.
Recommendations

(1) Exploit potentialities and enhance the ability of countries to cope with natural
disasters.

The Mekong River Basin countries have established organization system for disaster
prevention and reduction to deal with flood and drought disasters. Although there
is a significant correlation between the national flood prevention and drought relief
capacity and economic and social development level, the consensus of the whole
basin on improving the flood prevention and drought relief structural measures is
also very important. Based on the viewpoint of engineering, Thailand and Viet Nam
6 Main Findings and Recommendations 171

have taken adequate measures to combat drought. Lao PDR has more reservoirs
established, under construction and planning. Cambodian drought-resistant potential
needs to be tapped and developed.
(2) Make overall plans and coordinate to enhance the disaster mitigation ability
from the whole-basin perspective.
The reservoirs built in the Mekong River Basin have a capacity of more than 20
billion m3 . It is suggested that the possible positive role of these reservoirs in flood
prevention and drought relief in all countries and the whole basin should be studied
in depth, and the impacts of flood prevention and drought relief projects on the water
disasters in the downstream regions or countries should be deeply analyzed while
they alleviate the local disasters to lay a foundation for the overall coordination of
the basin.
(3) Strengthen cooperation and carry out technical exchanges and mechanism
building.
The establishment of Lancang-Mekong Cooperation Mechanism provides a new
platform for China and five other countries to exchange deeply and share their expe-
rience in water resources management and jointly address the challenges of climate
change. The theme spirit of “Shared River, Shared Future” is China’s appeal for
cooperation in the whole basin, and also expresses a profound understanding of the
Lancang-Mekong water resources cooperation. In addition to the existing bilateral
cooperation and dialogue and cooperation mechanisms, it is suggested that the six
countries in the basin should strengthen the technical exchanges and cooperation in
flood prevention and drought relief under the Lancang-Mekong Cooperation Mecha-
nism and study the possibility of establishment of a coordinated mechanism for flood
prevention and drought relief in the whole basin to work together for the sustainable
development of the basin and equitable and rational utilization of water resources.
(4) Proposed cooperation projects and activities
Capacity building on water modelling tools and disaster mitigation related tech-
nologies are recommended to be carried out by all 6 countries under LMC
mechanism.
Limitations of this assessment report
Due to the limitation of data and research time, the statistical data of disaster losses
in this project may not be comprehensive and accurate; the accuracy of flood compo-
sition analysis is likely to be affected because it is based on statistical analysis rather
than observed hydrological data, and the contribution rate of tributaries and regions
to floods is based on overall statistical analysis and is not analyzed in detail according
to different types of floods; some of the flood prevention and drought relief projects
are based on manual interpretation of remote-sensing images and may be missed or
inaccurate. In conclusion, this research analyzed the main flood and drought char-
acteristics of the Mekong River Basin and qualitatively assessed the overall disaster
prevention and mitigation capacity of the basin.
172 H. Liu and X. Mu

Dr. Liu’s research primarily focuses on hydrological processes and modeling, flood management,
and transboundary river water cooperation. She has over 5 years’ research experiences in trans-
boundary water cooperation of China. She was a participant of various Lancang-Mekong water
resources cooperation activities. She is an editor of a joint publication by MWR of China and
MRC in 2016. She is Program Managers of 3 international cooperation projects with Mekong
River countries. She is author of over 10 journal papers and 1 book.

Dr. Mu’s research primarily focuses on water resources management, river ecosystem restoration
and hydraulic computation. He has over 10 years’ research experiences in relative national level
projects. He is author of over 40 journal papers and 4 books, as well as dozens of patents for
inventions.
Appendix
Introduction of THREW Model

Fuqiang Tian and Shiyu Hou

I. Principle
Only the principle of mass conservation is taken into account in conceptual hydro-
logic models, physical hydrologic models are evidently more advantageous because
the principles of energy conservation and mass conservation are both taken into
account, underlying surface conditions and meteorological factors are utilized fully
and more detailed distributed modeling is conducted, and the physical significance
of hydrologic model parameters are improved. Nevertheless, because underlying
surface is highly changeable, spatial discretization leads to an excess of calculation
elements and over-parameterization, and different parameter schemes can reach the
same simulation result, namely the phenomenon of “same effect of different param-
eters”. Physical hydrologic models are also faced with the problems that underlying
surface conditions and meteorological conditions are hard to access, calculate and
verify in data-insufficient areas. The fundamental reason lies in the mismatching
between the applicable scale of the mathematical physical equation of hydrologic
modeling and the model’s applied scale. For instance, point scale or representa-
tive unit volume scale is applicable to Richards equation, Boussinesq equation and
Saint-Venant equation, but the parameters, initial conditions and boundary conditions
needed cannot be observed using current hydrologic methods, so parameters can be
figured out only through calibration, which is bound to cause over-parameterization.
Reggiani proposed a hydrologic simulation method based on representative ele-
mentary watershed (REW) in 1998. In this method, a basin is set apart into REW and
then, divided into several functional sub-regions. Subsequently, mass, momentum
and energy conservation and equilibrium equations on the micro scale are estab-
lished in the functional sub-region and after homogenization, a system of ordinary

F. Tian (B) · S. Hou

Tsinghua University, New Hydraulic Building, Beijing 100050, China
e-mail: [email protected]
S. Hou
e-mail: [email protected]

© China Water & Power Press and Springer Nature Singapore Pte Ltd. 2020 173
H. Liu (ed.), Flood Prevention and Drought Relief in Mekong River Basin,
Springer Tracts in Civil Engineering,
https://doi.org/10.1007/978-981-15-2006-8
174 Appendix: Introduction of THREW Model

differential equations on the REW’s scale is established. Because the equation system
is unclosed, ordinary differential equations of geometrical relationship and constitu-
tive relationship are added to make the equation system closed. However, the REW
concept Reggiani et al. proposed was not clearly and scientifically defined. In the
model, there is no ice or snow module, vegetation is not taken into account, simula-
tion effect of high mountains, cold regions and other regions with a big ratio of snow
and ice is unsatisfactory, and evaporation and transpiration cannot be distinguished
from each other.
Therefore, Tian Fuqiang et al. redefined REW in a rigorous manner by dividing
REW into surface and subsurface layers and dividing subsurface layer into 2 sub-
regions and surface layer into 6 sub-regions. The subsurface layer is divided into
saturated and unsaturated zones by groundwater level. The saturated zones are below
the groundwater level and unsaturated zones are above the groundwater level and
below the surface. In order to consider the freezing and thawing of water in soil,
saturated and unsaturated zones, besides soil, liquid water and gas, should contain
ice phase substances.
In the surface layer, the most notable feature is that the stream network and its
adjacent hillsides form an interconnected system. From the point of view of hydro-
logical simulation, the hillsides and stream network are the two basic components
of the basin.
(1) The stream network is further divided into main channel and sub-stream net-
work. In order to maintain the geometrical invariability of REW, the lakes,
reservoirs, streams and channels on the sub-REW scale are included in the
sub-stream network.
(2) The hillside is the main area of precipitation redistribution and transportation,
which is divided into four zones: bare soil zone, vegetated zone, snow covered
zone and glacier covered zone.
Finally, the REW is divided into surface and subsurface layers. The subsurface
layer is further divided into 2 sub-regions and surface layer is further divided into
6 sub-regions, which can reflect four typical underlying surface types: bare soil,
vegetation, snow, glacier, as shown in Fig. A.1.
In the horizontal direction, the 6 sub-regions of the surface layer form a complete
coverage of the surface. In the vertical direction, bare soil zone, vegetated zone, snow
covered zone and glacier covered zone are located at the unsaturated zone; according
to the relationship with the groundwater level, other two surface sub-regions (sub-
stream network and main channel) may be located at the saturated zone, or at the
unsaturated zone. According to the availability of data and purpose of application,
a specific spatial scale is selected to set apart the basin into REW, thus making
the attributes of each REW including soil, vegetation, snow and glacier uniquely
determined.
In the surface sub-regions, surface runoff is generated when the precipitation
intensity exceeds infiltration capacity. In unsaturated zone, soil heterogeneity can
result in subsurface flow. For example, if the unsaturated zone is divided into upper
and lower layers and the upper layer’s infiltration capacity is greater than that of the
Appendix: Introduction of THREW Model 175

Fig. A.1 REW zoning in THREW model

lower layer, the subsurface flow may take place at the interface of two layers of soil.
In order to simplify the calculation, the heterogeneity is not directly considered in the
definition of REW and the subsurface flow and preferential flow can be expressed
by material exchange items and corresponding constitutive relationship between
unsaturated zone and adjacent REW as well as outside basin.
In order to consider evapotraspiration, the surface sub-regions contain vapor phase
substances. The soil pores in unsaturated zone contain air and water vapor, which are
considered as gases and defined as a phase substance. Similar phenomena are found
in the snow covered zone. The water storage in the bare soil zone can be regarded as
depression detention water and water storage in the vegetated zone can represent the
sum of the water quantity of vegetation interception and depression detention. The
runoff from surface sub-regions flows into the sub-stream network, and then into the
main channel.
II. Basic Equations
In the THREW model, basin is divided into sub-systems at REW level, sub-region
level and phase material level. It leads to the general form of conservation laws on
REW’s scale.
(1) Equation of mass conservation on REW’s scale

d  j j j j
L=1...Nk ,i=j
  j
ρα εα y ω = eαjP + eαβ (A.1)
dt P=EXT ,L,T ,B,i β=α
176 Appendix: Introduction of THREW Model

(2) Equation of momentum conservation on REW’s scale

    L=1...Nk ,i=j
  j
j j j j d j j j j j j
ρα εα y ω vα = gα ρα εα y ω + TαjP + Tαβ (A.2)
dt P=EXT ,L,T ,B,i β=α

(3) Equation of thermal balance on REW’s scale

L=1...Nk ,i=j
j
d θα j j
  j
(εαj yj ωj cαj ) = hα ρα εαj yj ωj + QαjP + Qαβ (A.3)
dt P=EXT ,L,T ,B,i β=α

(4) Entropy equation on REW’s scale

  j L=1...NK ,i=j
  j
j j j j d ηα j j j j j j j j j
ρα εα y ω = bα ρα εα y ω + Lα εα y ω + FαjP + Fαβ
dt P=EXT ,B,T ,L,i β=α
(A.4)

III. Constitutive Relationship

The system of said conservation equations is indeterminate for the number of vari-
ables is bigger than that of equations, and other relational expressions should be added
for solving the equations. Therefore, geometrical and constitutive relationships must
be added to make the equation system closed.
In hydrology, the linear relationship, namely a constitutive relationship, of Darcy’s
law is true in a given scope of flow velocity. However, constitutive relationship is
less accurate than kinematic formula. In THREW model, formulas are established
on REW’s scale, and traditional constitutive relationship on micro scale, such as
Chezy formula, Darcy’s law and soil moisture characteristic curve are no longer
applicable. Up-scaling of constitutive relationship is mainly realized with statistical
physics method.
In the model, equations are further simplified, and momentum conservation equa-
tions of different sub-regions are closed. Then, mass exchange term, vegetation inter-
ception and other constitutive relationships and the 1 area ratio of saturated zone and
other geometrical relationship are established to finish the establishment of the closed
system of equations.
IV. Model Application
THREW has been applied in the basins with different climatic and hydrological
conditions, including the Lancang River, including humid region, semi-arid region,
alpine valley region and so on. Table A.1 lists some of the THREW model-applied
basins.
Appendix: Introduction of THREW Model 177

Table A.1 List of THREW model-applied basins

Country Basins Basin characteristics Published papers or
THREW
model-applied
projects
China Lancang River 160,000 km2 , alpine Applied in fund
and gorge region projects research
China Tailan River 13,240,000 km2 , He et al. [1], HESS
Tianshan region,
glacier
China Three Gorges 56,000 km2 , monsoon, Applied in the
complex terrain entrusted projects of
Three Gorges
Corporation
China Ya-lung River 128,000 km2 , alpine Applied in fund
and gorge region projects research
China Hanjiang River Basin 95,000 km2 , humid Sun et al. [2], J.
mountain basin Hydrol.
China Urumqi River 289,000 km2 , Mou et al. [3], HESS
permafrost region
China Chabagou River 1,860,000 km2 , arid Applied in fund
Loess Plateau projects research
China Weihe River 25,000 km2 , arid Liu et al. [4], HSJ
Loess Plateau
China Jinghe River 4500 km2 , arid Loess Applied in fund major
Plateau projects research
China Chaohe River 4808 km2 , Applied in fund major
semi-humid projects research
earth-rock mountain
region
China Baihe River 8603 km2 , Applied in fund major
semi-humid projects research
earth-rock mountain
region
China–India Yarlung 520,000 km2 , upper Applied in the
Zangbo-Brahmaputra ice and snow, the entrusted projects of
River lower monsoon Lancang River
Corporation
China-Kazakhstan Yili River 90,000 km2 , Applied in the
trans-boundary water, entrusted researches
drought of China Renewable
Energy Engineering
Institute
(continued)
178 Appendix: Introduction of THREW Model

Table A.1 (continued)

Country Basins Basin characteristics Published papers or
THREW
model-applied
projects
China-Kazakhstan Emin River 20,000 km2 Applied in the
entrusted researches
of China Renewable
Energy Engineering
Institute
China-Kazakhstan Eerqisi River 89,000km2 , Applied in the
trans-boundary water, entrusted researches
drought of China Renewable
Energy Engineering
Institute
The United States Sangamon 3150 km2 , agricultural Liu et al. [5], IAHS
watershed Publ.
The United States Reynolds 238.9 km2 , mountain Applied in fund
area, snow projects research
The United States Salt Creek 49.9 km2 , urban Applied in the project
watershed research of American
Foundation
The United States Blue River 1233 km2 , temperate Li et al. [6], J. Hydrol.
continental climate
The United States Illinois river 2484 km2 , temperate Li et al. [6], J. Hydrol.
continental climate
Lao PDR Nam Ou River 3684 km2 , tropical Applied in the
monsoon climate entrusted researches of
Kunming Engineering
Corporation Limited
Austria Lienz 1198 km2 , Alps He et al. [7], HESS

References

1. He ZH, Tian FQ, Gupta HV et al (2015) Diagnostic calibration of a hydrological model in a

mountain area by hydrograph partitioning [J]. Hydrol Earth Syst Sci 19(4):1807–1826
2. Sun Y, Tian F, Yang L et al (2014) Exploring the spatial variability of contributions from climate
variation and change in catchment properties to streamflow decrease in a mesoscale basin by
three different methods [J]. J Hydr 508:170–180
3. Mou LQ, Tian FQ, Hu HP (2009) Artificial neural network model of runoff prediction in high
and cold mountainous regions: a case study in the source drainage area of Urumqi river [J]. J
Hydr Eng 1:64–69 (in Chinese)
4. Liu DF, Tian FQ, Hu HC et al (2012) The role of run-on for overland flow and the characteristics
of runoff generation in the Loess Plateau, China [J]. Hyd Sci J 57(6):1–11
5. Liu DF, Tian FQ, Hu HP (2009) Sediment simulation at Upper Sangamon river basin using the
THREW model [J]. IAHS-AISH Publication 335:187–195
Appendix: Introduction of THREW Model 179

6. Li HY, Sivapalan M, Tian FQ (2012) Comparative diagnostic analysis of runoff generation

processes in Oklahoma DMIP2 basins: the Blue river and the Illinois river [J]. J Hyd 418:90–109
7. He ZH, Parajka J, Tian FQ, Bloschl G (2014) Estimating degree-day factors from MODIS for
snowmelt runoff modeling [J]. Hydrol Earth Syst Sci 18(12):4773–4789

Dr. Tian’s research primarily focuses on hydrological processes and modeling, flood manage-
ment, agricultural water management, and transboundary river water management. He is editor of
two journals of Hydrology and Earth System Sciences and Journal of Hydrology. He is the Chair
of Panta Rhei Initiative (the IAHS Scientific Decade 2013–2022) for the fourth biennium (2019–
2020), and also the chair of Biofuel Working Group of ICID from 2015 to the present. He has
published over 150 journal papers, 3 books and 1 textbook.

Ms. Hou’s research primarily focuses on hydrological processes, hydrological predictions and
modelling, and flood management. She is in Department of Hydraulic Engineering of Tsinghua
University. She has published over 1 journal papers.