ISSN: 1683-1489
Mekong River Commission
Diagnostic study of water quality in
the Lower Mekong Basin
MRC Technical Paper
No. 15
March 2007
Meeting the Needs, Keeping the Balance

MekongRiverCommission
Diagnostic study of water quality in the
Lower Mekong Basin
MRC Technical Paper
No. 15
March 2007
Published in Vientiane, Lao PDR in March 2007 by the Mekong River Commission
Cite this document as:
MRC (2007) Diagnostic study of water quality in the Lower Mekong Basin. MRC Technical
Paper No. 15, Mekong River Commission, Vientiane. 57pp.
The opinions and interpretation expressed within are those of the authors and do not necessarily
reflect the views of the Mekong River Commission.
Editors: Dr. Edwin Ongley, Dr. Martine Allard and Dr. Tim Burnhill
Series Editor: Dr. Tim Burnhill
© Mekong River Commission
184 Fa Ngoum Road, Unit 18, Ban Sithane Neua, Sikhottabong District,
Vientiane 01000, Lao PDR
Telephone: (856-21) 263 263 Facsimile: (856-21) 263 264
E-mail: mrcs@mrcmekong.org
Website: www.mrcmekong.org
ii
Table of Contents
Summary
ix
1. Introduction
1
The Mekong River basin
1
Potential sources of pollution
2
Upper Mekong Basin
2
Lower Mekong Basin
3
MRC water-quality monitoring programme
5
Water-quality and MRC's Water Utilisation Programme
7
Diagnostic study framework
7
Main activities
7
Study concept and limitations
8
2. Definition of Priority Topics and Areas
9
Methodology
9
Results
13
3. Assessment of MRC Water-quality Database
15
Methodology
15
Transported loadings
15
Water-quality assessment
16
Results and discussion
18
Chemical loads
18
Water-quality assessment
19
MRC water-quality database limitations
21
4. 2003 and 2004 Field Campaigns
23
Sampling sites
23
Sampling programme
24
Selection of matrices, analyses and sample collection
24
2003 field campaign
25
2004 field campaign
25
Analytical methodology
28
Laboratories
28
Sample transportation
28
Sampling and analytical methodologies
28
Data treatment and processing
30
5. Results and Discussion
35
Water
35
Major ions
35
Salt contamination from the Khorat Plateau
36
Nutrients
36

iii
Sediments
37
Total heavy metals concentrations in sediments
37
Heavy metal concentrations in sediments and their eco-toxicity potential
38
Dioxins and furans in sediments
43
Other parameters analysed in sediments
44
Bioassays
46
Results from the 1st campaign (2003)
46
Results from the 2nd campaign (2004)
47
6. Conclusions and Recommendations
49
Synthesis
49
Multi-criteria analysis
49
Main recommendations
51
Benchmark sites
51
WQMN programme
51
WUP water-quality modelling and management
53
7. References
55
iv
Acknowledgements
This study was funded by the French Government as part of its co-financing of the MRC
Water Utilisation Programme, and was contracted to the French firm BURGÉAP. BURGÉAP
implemented this work in association with CEMAGREF1 and with input from several other
European research organizations. The field work was carried out with assistance from the four
national laboratories that participate in the water-quality monitoring activities of the MRC. This
technical paper is based largely on the final report submitted to the MRC by BURGÉAP and has
been edited to meet the style required of MRC publications. In this regard the editors wish to
thank Christophe Pateron of BURGÉAP for his assistance.
The MRC also wishes to thank the National Mekong Committees of Cambodia, Lao PDR,
Thailand and Viet Nam for their cooperation and support as well as the many regional experts
who assisted in the conduct of the field campaigns.
1Agriculture & Environmental Research Institute (a French public research centre).

v
vi
Abbreviations and Acronyms
AFNOR: Association Française de Normalisation
BOD5: Biological Oxygen Demand (5 days)
BTEX: Benzene, Toluene, Ethylbenzene and Xylene
CN:
Cyanide
COD: Chemical Oxygen Demand
DAIpo: Diatom Assemblage Index to organic water pollution
VOCs: Volatile Organic Compounds
DO:
Dissolved Oxygen
GEF:
Global Environment Facility
IBD:
Indice Biologique Diatomées = Biotic Diatom Index
IPS:
Indice de Polluosensitivité spécifique = Specific polluo-sensitivity index
ISQG: Interim Sediment Quality Guideline
I-TEQ: International Toxicity Equivalents
LMB: Lower Mekong Basin
MRC: Mekong River Commission
NMCs: National Mekong Committees
OCDD: Octadichlorodibenzodioxin
OCDF: Octadichlorodibenzofuran
PAHs: Polycyclic Aromatic Hydrocarbons
PCBs: Polychlorinated Biphenyls
PCD:
Pollution Control Department
PCDDs: Polychlorinateddibenzodioxins
PCDFs: Polychlorinated dibenzofurans
PEC:
Probable Effect Concentration
PEL:
Probable Effect Level
QA/QC: Quality Assessment / Quality Control
TEC:
Threshold Effect Concentration
TEF:
Toxicity Equivalent Factor
TEL:
Threshold Effect Level
TSS:
Total Suspended Solids
USEPA: United States Environmental Protection Agency
WQDS: Water Quality Diagnostic Study
WQMN: Water Quality Monitoring Network
WUP: Water Utilisation Programme

vii
viii
Summary
Water-quality monitoring in the Lower Mekong Basin has been carried out at approximately
100 stations (in Lao PDR, Thailand and Viet Nam since 1985, and since 1993 in Cambodia) by
national laboratories coordinated by the Mekong River Commission (MRC). This programme
uses conventional physico-chemical measurements typical of such programmes world-
wide. Because there is little data on environmental contaminants in the Mekong River and
its tributaries, the GEF/MRC Water Utilization Programme (WUP) commissioned a major
diagnostic study of water quality in the Lower Mekong Basin.
The study was carried out in two phases, with field campaigns in 2003 and 2004. Twenty-two
sites were sampled in 2003, and, on the basis of results from that year's survey, 16 sites were
selected for sampling in 2004. The field campaigns were undertaken during the dry season in
both years. Samples of river water and river-bed sediments were analysed for a wide range
of conventional parameters, and for toxic micro-pollutants, including persistent and bio-
accumulating organic pollutants such as pesticides, PAHs, PCBs, dioxins and furans. Sediment
was included as many of the persistent toxic compounds are known to accumulate in this
substrate. Because concentrations of particular chemicals are not explicitly linked to ecological
health, a bioassay test was conducted at selected sites in both years to assess presence/absence
of toxicity. The data from 2003 demonstrated that the conventional water-quality data collected
through the MRC water-quality programme is of satisfactory reliability, therefore these
conventional parameters were not analysed in 2004.
The study establishes current baseline conditions for environmental contaminants in the lower
Mekong River and its major tributaries. Concentrations of metals in water and sediment are
mainly below any level of concern. Industrial contaminants and pesticides in water are all less
than the detection limit and less than published criteria (where available) for biological effects.
The environmental effect of pesticides on sediments cannot be determined as the detection
limits available in this study were too high for most pesticides. On analysis, samples from
several sites gave a positive toxic response to the bioassay test organism, however measured
chemistry was almost always lower than published threshold effects levels. A few sites had
levels of some compounds that are higher than other sites (but lower than threshold effects
levels) and deserve additional attention, both in terms of defining more precisely the nature and
extent of contamination, and to determine if these pose any downstream and/or trans-boundary
risk. The stretch of the Mekong River where it leaves China and enters Lao PDR is problematic
insofar as toxicity was recorded in both years, however this is not correlated with measured
chemistry.
KEY WORDS: Mekong; water-quality; toxicity; environmental contaminants; trans-boundary
issues.

ix
Diagnostic study of water quality in the Lower Mekong Basin
China
Myanmar
Viet Nam
Lao PDR
Thailand
N
Cambodia
0
300 Km
Upper Mekong Basin
Lower Mekong Basin
Figure 1. The Mekong River Basin
40,000
Kratie
30,000
Se Kong, Se San and Sre Pok
Pakse
Se Bang Fai, Se Bang Hieng,
ge (cumecs)
Se Done, Nam Mun and
Nam Chi
Mukdahan
20,000
y dischar
th
Nam Ngum, Nam Theun
and Nam Hinboun
Vientiane
Luang Prabang
10,000
ean mon
Nam Ou and
M
Nam Mae Kok
Yunnan component
0
Jan
Feb Mar Apr May Jun
Jul
Aug Sep
Oct Nov Dec
Month
Figure 2. Mean monthly discharge at selected locations showing the main
tributaries in each reach. Arrows indicate from upstream to
downstream. The period of record is 1960-2000 (MRC, 2004)
1. Introduction
The water resources of the Mekong River provide livelihoods for most of the 60 million
people who live in the Lower Mekong Basin. These livelihoods to a large extent depend on
the environmental health of the Mekong River and its tributaries remaining in good condition.
Water quality is a key determinant of environmental health. The Mekong River Commission
has monitored the water quality of most of the river since the mid 1980s (monitoring of the
Cambodian stretch of the Mekong only began in 1993). The parameters the MRC monitors are
the conventional physico-chemical measures that are employed by similar programmes world-
wide. Because there are little data on environmental contaminants in the Mekong River and its
tributaries, the MRC Water Utilization Programme (WUP) commissioned a major diagnostic
study of water quality in the Lower Mekong Basin. This report documents the results of the
study, which included additional data from field sampling campaigns undertaken during 2003
and 2004.
The Mekong River Basin
The Mekong River is the longest river in southeast Asia, the 12th longest in the world, and the
10th largest by discharge (Dai and Trenberth, 2002). It rises in the Tibetan Plateau and flows
southward through China, Myanmar, Lao PDR, Thailand, Cambodia and Viet Nam where
it discharges into the South China Sea (Figure 1). The river's basin, which has an area of
795,000 km2, is functionally divided into the Upper Basin--which flows southwards through
China (where it is called the Lancang River), and the Lower Basin--which includes Lao PDR,
Thailand, Cambodia and Viet Nam (Figure 1). The river forms the boundary between Lao PDR
and Myanmar in the transition zone between the Upper and Lower basins. The Mekong River
Basin Diagnostic Study (MRC, 1997) and the State of the Basin Report (MRC, 2003) provide
further information on the basin, its water-related resources, and its inhabitants.
Since 1957, Lao PDR, Thailand, Cambodia and Viet Nam (the `riparian countries') have
cooperated in management of the Lower Mekong Basin through the `Committee for the
Coordination of Investigations of the Lower Mekong Basin' or the Mekong Committee
(a forerunner of the Mekong River Commission) under a statute endorsed by the United
Nations. In April 1995, these four countries signed the `Agreement on the Cooperation
for the Sustainable Development of the Mekong River Basin' (the Mekong Agreement)
which empowers the Mekong River Commission (MRC) and its secretariat (MRCS). Under
this agreement water-quality is specific to Article 3 (Environmental Protection), Article 7
(Prevention and Cessation of Harmful Effects), and to Article 10 (Emergency Situations). The
programmes of MRC, including the water-quality programme, apply only to the four riparian
countries. Neither China nor Myanmar are signatory to the Mekong Agreement, although
both have observer status. The MRC operates its programmes through the National Mekong
Committee (NMC) of each member country.
The annual monsoon cycle is the predominant factor controlling the hydrology of the Mekong
River and its tributaries. The cycle has a wet season from June to November and a dry season

1
Diagnostic study of water quality in the Lower Mekong Basin
from December to May (see Figure 2 for the mean annual hydrograph). China contributes only
16% of the mean annual discharge, whereas Lao PDR contributes some 35% and up to 60% of
the flow during the wet season (MRC, 2005). In contrast, China is the provenance of 50% of the
sediments that the Mekong discharges into the South China Sea (MRC, 2005).
Concerns about the construction of dams in the basin led the World Bank/MRC Water
Utilisation Project to model the effects of dam development scenarios on the hydrology of the
basin (World Bank, 2004). This study concluded that `the overall character of the hydrograph
is maintained', that `high-flows are marginally reduced, but within the historically observed
range', and that `low-flows are significantly increased and are higher than the historically
observed range'.
Downstream of Kratie (Cambodia), the hydrology of the Lower Mekong Basin is particularly
complex because of the extremely low gradients. During the dry season, the Mekong flows into
the South China Sea through the Mekong and Bassac distributary channels and the Mekong
Delta. Salinities of up to 1 g/l can extend 70 km upstream from the river mouth and tidal
influences are noticed as far upstream Phnom Penh. During the rainy season, a sizeable portion
of the Mekong's water flows `upstream' in the Tonle Sap river and into the Great Lake of
Cambodia, causing the area of the lake to expand up to six-fold and creating extensive wetlands
around the entire water body. The water drains out of the Great Lake back through the Tonle Sap
and into the Mekong system during the dry season, thereby adding to low-flow discharges in the
region downstream of Phnom Penh. This hydrological pattern makes hydrological and water-
quality monitoring and interpretation difficult, especially in mainstream stations below Kratie.
Reverse flows occur daily during the tidal cycle at delta stations in Viet Nam and during wet
season reverse-flow in the Tonle Sap river.
The Great Lake­Tonle Sap system of Cambodia is a unique lacustrine/wetland complex. The
MRC has monitored the water quality of lake since 1993 and the WUP has undertaken a special
study on nutrient and sediment fluxes. However, there has been no systematic or substantial
scientific study of the nutrient dynamics of the system. As a result, it is not known with certainty
if the lake is N or P limited and whether nutrient loadings from the surrounding land are
transported through the wetlands into this shallow lake, or if these loads are consumed within
the wetlands. It is known, however, that there is extensive anoxia in the wetlands surrounding
the lake and that this probably results from bacteria consuming oxygen during the decay of
organic matter. Despite this anoxia, the wetlands are enormously productive and fish species
have adapted to these conditions.
Potential sources of pollution
Upper Mekong Basin
Hydropower stations
Two hydropower stations have been built on the on the mainstream (Manwan and Dachaoshan)
and six more hydropower stations are planned for development in the next 20 years, including
the Xiaowan site that is under construction. The Xiaowan dam, located 550 km upstream of the
2
Diagnostic study of water quality in the Lower Mekong Basin
China/Lao PDR border, will be the highest hydroelectric dam in China after the Three Gorges
Dam on the Yangtze River (CERN, 2002). A Thai-Chinese consortium will build two additional
dams on the Lancang in Yunnan province (construction of the Jinhong station was scheduled to
begin in 2005 and the Nuozhadu in 2006).
While Chinese sources claim that these dams will have no impact on water quality in the Lower
Mekong Basin, there is the danger of release of anoxic bottom waters from reservoirs, which
happens at many dams worldwide. Also, according to the China Daily (2002), a 2.7oC reduction
in temperature in water is anticipated at the Xiaowan dam site, although this will be mitigated
downstream.
Industrial pollution
In 2000, the provincial government of Yunnan Province in China (located immediately upstream
of the China/Lao PDR border) inspected 1042 industrial enterprises. This resulted in the forced
closures of four plants (CIIS, 2002). Since 1986, the Simao Paper Plant and the Lanping Lead-
Zinc Mine have been built on the banks of the Lancang (Mekong) River.
Nevertheless, the rapid development of the Lancang basin in China and increasing pollution
in Chinese rivers (Ongley & Wang, 2004; Xinhua News Agency, 2005) raise concerns about a
deterioration quality of the water arriving in the Lower Mekong basin from China in the future.
However, Chinese news sources (e.g. CIIS, 2002) frequently claim that the water of the Lancang
meets international drinking water standards (for those parameters for which Chinese agencies
routinely monitor). Nevertheless, ecotoxicological assessment of a site on the Lao side of the
border with China, which is discussed later in this report, suggests that there is some toxicity in
this stretch that needs further investigation.
Lower Mekong Basin
There are only a few sources that could potentially pollute the mainstream in the Lower Mekong
Basin. In Thailand, salt leaching from halite deposits underlying the Khorat Plateau--part of the
Nam Mun catchment--is a problem, but this is an entirely natural phenomenon. However, even
this is diluted to the point where there is no visible change in water-quality in the mainstream
downstream of the confluence between the Nam Mun and the Mekong River. There are no data
that suggest that irrigated agriculture or the limited industrial areas in Thailand within the Lower
Mekong Basin are significant contributors of pollution to the mainstream of the river.
The two largest urban areas (Vientiane in Lao PDR, and Phnom Penh in Cambodia) are of
concern as they lie on the banks of the Mekong. Currently, Vientiane, a city of around 500,000
inhabitants, discharges its municipal sewage into the That Luang wetland--a wetland that
discharges into the Mekong River downstream of the city This discharge is small and poses little
immediate risk to the Mekong mainstream. However, development of Vientiane with substantial
land reclamation in the That Luang wetland for urban purposes is a concern, and may pose
greater threats to the Mekong mainstream in the future if it is not managed properly.
Phnom Penh, a city of approximately 1.7 million inhabitants, also discharges much of its urban
sewage into a series of wetlands that drain into the Bassac River--a distributary of the Mekong.

3
Diagnostic study of water quality in the Lower Mekong Basin
In addition, some industrial and municipal waste products, as well as storm-water runoff,
discharge directly into the Tonle Sap river--a tributary of the Mekong. While here there is some
localised industrial-pollution, it is unclear whether this poses any significant risk either locally
or further downstream. Likewise, we do not know the scale of the risk of pollution caused
by the sewage coming from many stilt-houses built along the banks of the river and from the
floating villages on the Great Lake.
Tidal factors influence the large cities on the Mekong Delta, such as Tan Chau and Chau Doc
--located by the Mekong River and the Bassac respectively. The river-pollution recorded at
these locations is probably attributable to local sources; however, there is no definitive work on
trans-boundary transport of pollutants to rule out upstream sources (Hart et al., 2001). However,
the extensive large-scale caged aquaculture farming that lines the banks of the Mekong River
and the Bassac river downstream of the Cambodia/Viet Nam border is also a possible source of
these pollutants. The in-stream caged fish culture, which occurs throughout much of the lower
basin, is not on such a large scale and is unlikely to contribute significant pollutants.
There is very limited research or other data on organic contaminants or on non-point sources of
pollutants in the Mekong River basin. Information presented at the 2nd Asia Pacific International
Conference on Pollutants Analysis and Control, indicates that there is little evidence in the basin
of persistent organic-pollutants, even in locations where there is known to have been high levels
of use, for example the extensive deployment Agent Orange during the American War, and the
intensive application of agricultural pesticides in parts of Thailand.
Recent work by Agusa et al. (2005) in Cambodia--a country where fish is the main source of
dietary protein--indicates that mercury in some species of freshwater and marine fish is above
dietary guidelines. However, their work does not imply large-scale mercury contamination in
the freshwater system, although there is anecdotal evidence of mercury usage in the extraction
of deposits of placer-gold upstream of the Great Lake and in the mainstream in Lao PDR, and
possibly in some of the Mekong's tributaries such as the Se Kong and Sre Pok rivers.
Other non-point sources include rapid expansion of caged-fish culture throughout the Mekong
and its tributaries, discharge of human wastes from vessels plying the Mekong, especially tour
boats in the middle reaches upstream from Luang Prabang (Lao PDR), and accidental spills
from river boat traffic. The recent hydraulic works (to enhance barge traffic) on the stretch of the
Mekong between China and Lao PDR may increase the chance of spillage.
In summary, existing data suggest the quality of the water in mainstream is good, even though
the concentrations of suspended sediment are high--especially during the wet season.
While there are valid concerns over specific threats arising from current and future land use
in the basin, these are based largely on anecdotal evidence. There is little well-documented
evidence about specific threats to water quality coming from outside or within the basin. An
exception is the Delta region; here good-quality data raises concerns over problems arising
from acidification, salinity and organic pollution. However, the extent to which trans-boundary
flux of chemicals, caused by local pollution, exacerbates these problems is not clear. Hydraulic
conditions, specifically low-flow, may be responsible.
4
Diagnostic study of water quality in the Lower Mekong Basin
This lack of detailed trans-boundary information, even for issues such as salinity, is a feature of
virtually all documents about water quality in the LMB. Also, there has never been any attempt
to define baseline conditions for more complex issues, such as sediment quality or persistent
bio-accumulation of toxic substances, as a means of establishing criteria against which to
measure future change.
MRC water-quality monitoring programme
Until the early 1980s there was little systematic ambient water-quality monitoring in any of the
countries, with the exception of Thailand. In 1985, the MRC, in response to riparian concerns
over potential water pollution and its trans-boundary implications, began a water-quality
monitoring programme in Lao PDR, Viet Nam and Thailand, with assistance from the Swedish
government. Cambodia joined the programme in 1993.
The programme monitors approximately 100 permanent stations on the mainstream and
important tributaries of the Mekong River (Figure 3). The monitoring involves sampling the
mid-stream, or the thalweg at locations where the river is very wide, on a monthly basis. The
sampling programme tests only river water, not sediment or other substrates.
Table 1. List of parameters measured in the MRC water-quality monitoring programme
Temperature
Na
Sulphate
PO -3
Fe
4
Conductivity
K
Alkalinity
Total P
TSS (mg/l)
Ca
NO
Si
2-3
pH
Mg
NH +
COD
4
Mn
DO
Cl
Total N
Al
Note: NH is the sum of NH + NH + however at neutral pH values most of the ammonia in river water is in
4
3
4 ,
the form on NH +.
4
Stations in some countries collected other parameters, such as total coliform, metals and some
pesticides. The MRC acts as a facilitating agency it does not have its own laboratories but rather
provides technical guidance to the four member countries.
The MRC's water-quality monitoring programme provides good information on the status and
the trends of selected parameters. While it can generate diagnostic and prescriptive information
on certain kinds of threats, it cannot deal comprehensively with the impacts (real or anticipated)
of present and proposed issues of land use. This type of traditional monitoring focuses mainly
on water chemistry and says nothing of ecological effects that are now the primary determinant
of `environment effects' in modern water-quality programmes. In addition, it is now known
that it is better to measure many anthropogenic toxicants on solids rather than in the aqueous
medium, however, measuring toxicants in sediments is a challenge even for developed
countries.
This study provides an interpretation of the large set of data compiled during the years of
monitoring. There has been no previous attempt to define downstream loads of common
chemicals, or to determine input loadings from China and output loadings into the delta
region and to the South China Sea. Furthermore, there has been no attempt to define `natural'

5
Diagnostic study of water quality in the Lower Mekong Basin
MRC water-quality monitoring
network in the Lower Mekong Basin
Primary stations
0
50 100
200
300 km
Secondary stations
Figure 3. MRC water-quality monitoring network. The map shows the sampling stations in
2004. While this is broadly the same as the network of stations monitored from
1985 to 2003, there are some differences, mainly in the tributary stations. Also from
2004 onwards, the network was divided into primary stations (basin-wide or trans-
boundary significance) and secondary stations (national or local significance).
6
Diagnostic study of water quality in the Lower Mekong Basin
background levels (which are probably not useful) or current `baseline' levels (which are very
useful) as yardsticks against which to assess future changes in water quality. This diagnostic
study addresses these concerns.
The MRC is very aware that the conventional data collected under its water-quality monitoring
programme, while useful for conventional issues such as organic pollution, eutrophication and
salinity, does not allow the identification of `hot-spots' where inorganic or organic contaminants
may be present or the level of risk that may be associated with contaminants. Nor does it help
to identify benchmark sites that future studies can use as a baseline from which to evaluate
any changes in water and/or sediment-quality. The lack of reliable information on the status of
contaminants in the Mekong system has been of considerable concern to riparian countries.
Water-quality and MRC's Water Utilisation Programme
The Water Utilization Programme (WUP) is a GEF-funded project for the Lower Mekong
River Basin, implemented by the World Bank and executed by the MRC. The WUP addresses
the main objective of the MRC, namely helping trans-boundary basin management of water
resources, according to the 1995 Mekong Agreement. Accordingly, one of WUP's chief
activities is to develop a set of management `proceedures' concerning water-quality. While
doing this the WUP identified the poor state of knowledge of water-quality concerning
specific contaminants such as pesticides and industrial pollutants within the basin. Further,
WUP specified that the knowledge of water-quality status, trends, and related natural and
anthropogenic causes and impacts, present and future, needed clarification before developing
water-quality modelling tools.
This diagnostic study aims to provide the WUP and other MRC programmes, with primary
information on the status and the trends of important pollutants as a basis for subsequent
modelling and development, and to identify potential trans-boundary water-quality issues.
The study also fills an important gap in a parallel activity the MRC is undertaking, that is, a
major revision of its water-quality monitoring programme. The revision includes all aspects of
the programme; however, this study will allow MRC to (i) develop a rational and more cost-
effective sampling of contaminants and (ii) to include hot-spots and/or benchmark sites as part
of the revised monitoring programme.
Diagnostic study framework
Main activities
The study involved four main activities:
· Activity 1: define priority topics, and potential areas within the basin, as a basis for
planning and executing the diagnostic study.

7
Diagnostic study of water quality in the Lower Mekong Basin
· Activity 2: provide a comprehensive assessment of water-quality according to the data
held in the MRC water-quality database, and to determine the extent to which the MRC
database can be used.
· Activity 3: design an innovative, advanced, programme of field and laboratory
investigations that will complement the MRC water-quality monitoring network
activities, by providing benchmark information at key sites, and lead to information that
demonstrates the importance, or lack thereof, of real or perceived trans-boundary and
basin-wide issues.
· Activity 4: implement a field and laboratory programme to obtain contrasting seasonal
data as a basis for determining the status of water-quality relative to basin-wide and trans-
boundary issues.
Study concept and limitations
The diagnostic study was designed to progress in a series of discrete steps, each of which
was dependent on the outcome the previous step or steps. For example, the field programme
(Activity 3) could not be designed until the outcome of Activities 1 and 2 was known.
The study used a `broad brush' approach that allowed sampling of the full range of possible
water-quality issues (i.e. the full range of contaminants) across the whole basin. This data
is particularly relevant to so-called trans-boundary issues. However, achieving a broad
geographical coverage meant sacrificing multiple analyses in the river cross-section and through
time, with a consequential reduction of statistical confidence.
8
2. Definition of Priority Topics and Areas
The principal objective of the diagnostic study was to identify indicators of water-quality, and,
if possible, trends in these indicators, that signify basin-wide or trans-boundary threats to the
ecological health of the river.
There have been numerous efforts over the years, both under WUP and through MRC's water-
quality monitoring programme, to identify significant basin-wide and trans-boundary issues.
However, while the MRC member countries have consistently identified important issues of
water-quality, these tend to be lists of generic concerns (e.g. `pesticides') based on anecdotal
evidence and with imprecise geographical locations. This is not surprising given the lack of hard
data on the changes in water-quality conditions caused by development and other anthropogenic
activities. Therefore, an essential first step in the study was to examine the historical database
to identify patterns and trends in the water quality of the basin as a basis for establishing the
sampling programme to be carried out later in the study.
Methodology
The first step in the diagnostic study was an extensive review of published literature and
secondary material, such as the TEAM study, to delineate and prioritise basin-wide and trans-
boundary issues and their priority.1 This information helped establish the priorities of the
subsequent field-sampling and analytical programme.
According to criteria used MRC's Watershed Classification Project, there are 103 sub-basins in
the Lower Mekong Basin. The MRC's Basin Development Programme (BDP) groups these sub-
basins into ten hydro-ecological regions, or sub-areas, based on hydrology, land use and land
cover, river slope, etc (Figure 4). These subdivisions provided the geographic framework for the
literature-based analysis.
Some of the sub-basins are trans-jurisdictional, that is the catchment spreads across international
borders before the tributary rivers reach the Mekong (Figure 4).
· Rivers in Nam Mae Kok and Nam Mae Kham sub-basins (Sub-area 1 - Chaing Rai) that
rise in Myanmar and pass through Thailand before entering the Mekong.
· Rivers in the Strung Mongkol Borey sub-basin (Sub-area 9 - Tonle Sap) that rise in
northeast Thailand and flow south-eastwards into the Great Lake.
· The Se San and Sre Pok rivers (Sub-area 7 - Se San/Sre Pok/Se Kong), which rise in
central Highlands of Viet Nam and flow westwards through Cambodia.
· The Se Kong river (Sub-area 7 - Se San/Sre Pok/Se Kong), which rises in Viet Nam and
flows through Lao PDR before entering Cambodia.
1 TEAM investigated current and emerging issues of water quality in the Mekong portion of Thailand (TEAM 2001).

9
Diagnostic study of water quality in the Lower Mekong Basin
BDP Sub-area
1. Northern Laos
2. Chaing Rai
3. Nong Khai/Songkhram
Nam Mae
4. Central Laos
Kham
Nam Mae
5. Mun/Chi
Kok
6. Southern Laos
7. Se San/Sre Pok/Se Kong
8. Kratie
9. Tonle Sap
10. Delta
Se Kong
Se San
St. Mongkol
Borey
Sre Pok
Khampong Cham
Phnom Penh
Takeo
Sub-areas and sub-basins in the
Delta
Lower Mekong Basin
Sub-basins which are
TJ
trans-jurisdictional
0
50 100
200
300 km
Figure 4. Hydrographic subdivision of the Lower Mekong Basin. Sub-basins that extend
across international borders and therefore are trans-jurisdictional are named. (Note
also the parts of the Nam Mae Kok and Nam Mae Kham --Sub-area 2 - Chaing
Rai--that lie within Myanmar are not included in the Lower Mekong Basin despite
the fact that rivers in these catchments flow in the Mekong within the limits of the
basin.) The red `dots' are locations referred to in the text.
10
Diagnostic study of water quality in the Lower Mekong Basin
In addition to these major tributaries, the Bassac river, the Mekong's major distributary, splits
from the mainstream just south of Phonm Penh, Cambodia, before entering the Mekong Delta
(Sub-area 10 - Delta). The same is true of a number of the Mekong's other distributaries,
including the Tonle Touch river system, which splits from the Mekong even further upstream
near Khampong Cham (Figure 4).
Environmental stresses related to land-use, development activities and historical events (such
as the use of Agent Orange during the American War) that may have implications for water-
quality were noted (see Figure 5 for a detailed list of these stressors). The possible impact of
each stressor in each sub-basin was then assessed. In those instances where hard data were not
available, which was the majority of cases, subjective estimates of the potential impact were
made.
The probable severity of the impact of each stressor was ranked on a scale where H = high
impact, M = medium impact, L = low impact, O = not relevant in the sub-basin and ? = lack of
information. Figure 5 shows examples of this analysis from two sub-basins.
Cpa: 06/09/02
78
name of river
PREK THNOT
Cpa: 06/09/02
33
name of river
SE KONG
Aau
or name of lake
Aau
or name of lake
CAMBODIA
name of main tributary
LAOS
name of main tributary
n° of main watershed
VIETNAM
n° of main watershed
disharge contribution :
to main tributary
disharge contribution (m3/s) :
to main tributary
to Mekong river RIGHT
max: 3290 min: 33.6 mean: 218
to Mekong river LEFT
station #: 430105
H=high
? H M L O Development issue or land use activities
H=high
? H M L O Development issue or land use activities
M=medium
M=medium
L=low
L=low
0=out of prupose
X hydro-electric powerplant
0=out of prupose
X
hydro-electric powerplant
?=unknown
X dams which may af ect water quality
?=unknown
X
dams which may af ect water quality
X weirs and dikes
X weirs and dikes
X
Forest
X
Forest
X
Agriculture
X
Agriculture
X
ir igation water supply
X
ir igation water supply
X
rice field
X
rice field
X
arable land
X
arable land
X
ir igation project
X
X ir igation project
X
fish pond culture
X
fish pond culture
X
fishery production
X
fishery production
X
population density
X
population density
X
set lement area
X set lement area
X
agro-industrial plants
X
agro-industrial plants
X
food processing factories
food processing factories
X
economic (industrial) development
X
economic (industrial) development
X
economic development project
X
economic development project
industrial waste accidently discharged in
industrial waste accidently discharged in
X
reservoir or river
X
reservoir or river
X
aquatic life
X
aquatic life
X
recreation
X recreation
X
navigation
X navigation
X
flood- free land devlpmnt
X flood- free land devlpmnt
X drinking water supply
X
drinking water supply
water diversion from Mekong river (Kong-Chi-
water diversion from Mekong river (Kong-Chi-
X Mun project
X Mun project

Figure 5. Examples of the analysis of stressor levels in two sub-basins (No. 78 - Prek Thnot ,Cambodia
and No. 33 - Se Kong, Lao PDR/Viet Nam)

11
Diagnostic study of water quality in the Lower Mekong Basin
Nam Mae
KokNam Mae
Ing
Nam Ngum
Nam Chi
Se Bang Hieng
Nam Mun
Se Kong
St. Mongkol
Borey
Sre Pok
St. Baribo
Prek Thnot
Takeo
Sub-basins in the Lower Mekong
Delta
Basin with medium or high levels of
one or more stressor
0
50 100
200
300 km
Sub-basin with elevated levels
of stressors
Figure 6. Sub-basins with medium or high levels of one or more stressor
12
Diagnostic study of water quality in the Lower Mekong Basin
Results
Thirteen of the 103 sub-basins have medium or high levels of one or more stressor (Figure 6 and
Table 2). These were chosen as the priority locations for sampling and analysis under Activities
3 and 4 of the study (see Chapter 4).
Table 2. Sub-basins with medium or high levels of one or more stressor
Sub-area
Sub-basin
Country
Agriculture
Fishery
Population
Economic
Industrial
Development
pressure
development
waste
projects
Chaing Rai
Nam Mae Thailand/
X
X
X
X
X
Kok
Myanmar
Nam Mae Thailand
X
X
X
Ing
Central Laos
Nam
Lao PDR
X
X
X
X
Ngum
Se Bang
Lao PDR
X
X
Hieng
Se San/Sre
Se Kong
Viet Nam/
X
X
Pok/Se Kong
Lao PDR/
Cambodia
Sre Pok
Viet Nam/
X
X
X
Cambodia
Mun/Chi
Nam Chi
Thailand
X
X
X
X
X
Nam Mun
Thailand
X
X
X
X
X
X
Tonle Sap
Stung
Thailand/
X
X
X
Mongkol Cambodia
Borey
Stung
Cambodia
X
X
X
Baribo
Prek Thnet Cambodia
X
X
X
X
X
Delta
Takeo
Cambodia
X
X
X
X
Delta
Viet Nam
X
X
X
X
X
X

13
Diagnostic study of water quality in the Lower Mekong Basin
14
3. Assessment of MRC Water-quality Database
At the inception of the diagnostic study there had been no comprehensive analysis of the water-
quality data held by MRC from its routine Water Quality Monitoring Network (WQMN). In part
this was because, until 2001, a consolidated and verified database was not in place. As part of
a programme of modernisation of the entire water-quality activity, the data held by MRC since
1985 were reviewed, `nonsense' values removed, outliers flagged, and the data evaluated against
a variety of reliability criteria such as ion balance, etc. The MRC made the resulting `verified'
database available to the study as a basis to determine:
· the status and trends in time and space of important parameters, including nutrients and
chloride;
· transported loadings of these parameters;
· weaknesses in the data-sets;
· priority areas of pollution to use as potential benchmark locations for sampling in
Activity 3.
Methodology
Transported loadings
To assess transported loadings, data was chosen from only those stations where MRC had
recorded at least five continuous years of discharge data (within the period 1985-20001). Of the
98 stations in the network, only 20 meet this criterion. Six of the 20 selected stations are located
on the Mekong, four on main tributaries (Nam Chi and Nam Mun rivers), and the remaining
10 on tributaries with smaller drainage areas. Five stations are in Lao PDR and 15 in Thailand,
none were in Cambodia or Viet Nam.
Chloride, nitrate (NO ­N) and phosphate (PO ­P) loadings were calculated for each of these
3,2
4
stations.
The chemical load, L (in kg/d or in t/d), is the product of the daily mean flow, q (m3/s), and the
concentration value at the day of the sampling, C (in mg/l), multiplied by the time dimension
factor:

L (kg/d) = q (m3/s) x C (mg/l) x 86.4
Where: L = Load, q = discharge and C = Concentration
The mean monthly load per year in kg/d or in t/d is then calculated from the 12 monthly loads
obtained per year.
1 Discharge data are from the MRC hydrometric station database.

15
Diagnostic study of water quality in the Lower Mekong Basin
It is assumed that sampling once a month is representative for the month and that the discharge
measurements are accurate. The assumption of accuracy2 of the measured data is an important
consideration insofar as loads are the product of concentration multiplied by discharge and,
therefore, even small errors in measured values can have a major impact on the calculated load.
Because it is well known in science that there is variance, sometimes substantial variance, in
the measured water-quality and water-quantity data, load values are prone to substantial and
inevitable uncertainty.
Water-quality assessment
Water-quality assessment was undertaken on samples from 11 of the 20 stations (Table 3)
mentioned above. Of these, six were located on the Mekong mainstream and five on major
tributaries. All were situated in sub-basins classed as `priority' following the literature review
described in Chapter 2.
Table 3. Stations selected for Mekong River water-quality assessment
Station Code
Station Name
Country
Water Body
H010501
Chiang Saen
Thailand
Mekong
H011201
Luang Prabang
Lao PDR
Mekong
H011901
Vientiane
Lao PDR
Mekong
H013101
Nakhon Phanom
Thailand
Mekong
H013801
Khong Chiam
Thailand
Mekong
H013901
Pakse
Lao PDR
Mekong
H050104
Chiang Rai
Thailand
Nam Mae Kok
H350101
Ban Keng Done
Lao PDR
Se Bang Hieng
H380103
Ubon
Thailand
Nam Mun
H380134
Rasisalai
Thailand
Nam Mun
H370104
Yasothon
Thailand
Nam Chi
There are no water-quality standards, objectives or guidelines that are specific to the Lower
Mekong Basin. Therefore, this analysis uses a methodology developed by the French water-
basin agencies for the classification of water bodies (SEQ-Eau, 1999, see Table 4). SEQ-Eau
define five classes of water-quality based on impairments to ecological health, to drinking
water and to recreational activities. The standards, parameters and thresholds defined by these
agencies were used in this study. Parameters of similar nature or having similar effects on
water bodies are combined into eight groups: (i) organic and other matter that can be oxidised,
(ii) nitrogenous matter, (iii) nitrates, (iv) phosphorous matter, (v) suspended matter, (vi)
temperature, (vii) mineralisation and (viii) acidification.
However, the French methodology was developed for European rivers and, as a result, some
of the threshold values it uses are not appropriate for the Mekong River system. Two problems
concern the use of temperature (T) and total suspended solids (TSS) parameters. The rivers in
the Lower Mekong Basin (because of its climate, physiography, geology, and land-use) have
much higher natural water temperatures and TSS concentrations than any European river.
2 There is a substantial body of literature on the subject of load estimates, load algorithms, and uncertainty in these calculations.
16
Diagnostic study of water quality in the Lower Mekong Basin
Table 4. Parameters and thresholds used to classify the quality of water in the Mekong River system
(after SEQ-Eau, 1999)
Class of water quality1
Very good
Good
Fair
Bad
Very bad
Organic matter and other matter that can be oxidised
DO (mg/l)
8
6
4
3
<3
COD (mg/l O )
5
7
10
12
>12
2
NH + (mg/l NH )
0.5
1.5
2.8
4
>4
4
4
Nitrogenous matter
NH + (mg/l NH )
0.1
0.5
2
5
>5
4
4
Nitrates
NO (mg/l NO )
2
10
25
50
>50
3
3
Phosphorous matter
TP (mg/l)
0.05
0.2
0.5
1
>1
PO 3- (mg/l PO )
0.1
0.5
1
2
>2
4
4
Suspended matter
TSS (mg/l)
5
25
38
50
>50
Temperature
T (°C)
21.5
23.5
25
28
>28
Mineralization
Conductivity (µS/cm)
2500
3000
3500
4000
>4000
Acidification
pH Min
6.5
6.0
5.5
4.5

Max
8.2
8.5
9.0
10
For this reason, temperature and suspended TSS concentrations are not given in the tabulation
of results included in the following section (Table 5). Another problem is that this assessment
methodology does not include other important parameters such as endemic tropical pathogens
and parasites.3 However, despite these concerns, the other standard parameters used in this
methodology provide a synopsis of the overall water-quality of the Mekong River system.
A colour-coded classification of each parameter was designed using these SEQ-Eau (Table 4).
The classification is based on the principle of the `disqualifying parameter' as follows:
1. Measurements of each parameter were assigned a colour according to the criteria in Table 4.
This was done for each sampling date at each station.
2. Where several parameters are grouped together, such as organic matter (see Table 4), the
group was assigned the colour rating of the worst parameter. For example, if DO and COD
were rated green (good) and NH + was rated orange (bad) the organic matter group was
4
given a overall orange rating.
3. For the period of record, each parameter group was assigned the worst colour recorded,
provided that colour was present in 10% of the data set for that year.
3 These are not included in the MRC monitoring programme.

17
Diagnostic study of water quality in the Lower Mekong Basin
Results and discussion
The major points arising from the review of the data from the MRC's water-quality monitoring
network are summarised below.
Chemical loads
Chloride
Mekong stations:
· The average daily load of 1000­2000 t/d at upstream stations increases to about 5000 t/d
at Pakse.
· Pakse is the only mainstream station where the chloride load shows an increase over the
past years.
· The minimum and maximum chloride loadings in the Mekong River vary significantly
from year to year. However, this may be an artefact of the methodology and/or the
assumption that a single monthly sample is indicative of the monthly concentration.
Tributary stations:
· Chloride load from the Nam Mun river is 1000­2000 t/d and contributes 20­40% of the
loading at Pakse.
Nitrate
Mekong stations:
· Average daily loads of nitrate, calculated on yearly basis at each station, are stable over
the 1985­2000 period.
· At upstream stations (Chiang Saen, Luang Prabang and Vientiane) average loads are
under 100 t/d.
· At the next downstream stations (Nakhon Phanom and Kong Chiam), average loads can
reach 150­400 t/d--maximum values of more than 1000 t/d are observed in several years.
· At Pakse, the average loads decrease to 100­150 t/d.
Tributary stations:
· The contribution from the Nam Mun river is estimated at 10­30 t/d.
18
Diagnostic study of water quality in the Lower Mekong Basin
Phosphate
Mekong stations:
· Average daily loads of phosphate are also stable over the period 1985­2000, however,
concentrations of the ion increase significantly from upstream stations to those further
downstream.
· At Chiang Saen, the average daily load fluctuates in the range 3000 to 6000 kg/d.
· At the next four stations (Luang Prabang, Vientiane, Nakhon Phanom and Kong Chiam),
the concentration increases to 5000­20,000 kg/d.
· At Pakse, the calculated loads fluctuates in the range 10,000 to 50,000 kg/d.
Tributary stations:
· In most of the data-sets, the records for the wet season are incomplete. However, the
available data shows the Nam Mun river makes only a small contribution (in the range
500 to 1000 kg/d) to phosphate concentrations in the mainstream.
Water-quality assessment
Table 5 gives the results of the water-quality assessment for six of the groups of parameters
(excluding Temperature and Total Suspended Solids) over the 1985­2000 period.
Table 5. Water quality at 11 localities on the Mekong and its major tributaries (1985-2000)
Parameters
Organic matter Nitrogenous
Nitrates
Phosphorous Mineralisation Acidification
(grouped)
matter
matter
Chiang Saen
(Mekong)
Luang Prabang
(Mekong)
Vientiane
(Mekong)
N. Phanom
(Mekong)
K. Chiam
(Mekong)
Pakse (Mekong)
Chiang Rae
(Nam Mae Kok)
Ban K. Done
(Se Bang Hieng)
Rasisalai
(Nam Mun)
Yasothon
(Nam Chi)
Ubon
(Nam Mun)
Note; Red ­ bad, yellow ­ fair, green ­ good, blue ­ very good--according to the French water-quality classification system.

19
Diagnostic study of water quality in the Lower Mekong Basin
Mekong river:
In general, as can be observed from Table 5, water quality at the mainstream stations during the
period 1985 to 2000 was generally `good' or `very good'. Other interesting observations arising
from the analysis of the data set are:
· There is no degradation of the water quality between upstream and downstream
stations with the exception of Vientiane where, mainly during the rainy season, low
concentrations of DO, higher conductivity and lower pH have been observed.
· Mineralisation in the river decreases from Chiang Saen to Pakse. The fall in conductivity
from 2402 to 1873 µS/cm is caused by dilution.
· Temperature increases from upstream to downstream stations, with average values of
23.4°C at Chiang Saen to 26.9°C at Pakse.
· High and increasing TSS concentrations are observed between upstream stations and
Vientiane (where they reach an average of 400 mg/l). Downstream of Vientiane the
average concentration of TSS drops to 200 mg/l.
· TSS concentration decreases over time, with a notable drop in 1992 (Figure 7)
corresponding to the construction of a new dam in the upper-part of the basin.
3500
3000
s
2500
l
i
d
o
e
d S
2000
d
/
l
)
e
n
g
s
p
(
m
u
1500
t
a
l S
T
o
1000
500
0
1985
1990
1995
2000
Date
Figure 7. TSS concentrations over time at the Luang Prabang station
20
Diagnostic study of water quality in the Lower Mekong Basin
Main tributaries:
· In general, water quality (notably the levels of nitrogenous matter, nitrates, phosphorous
matter and acidification) at main tributary stations are generally `good' or `very good'
(Table 5).
· The high levels of mineralisation in the Nam Chi and Nam Mun tributaries, where
average conductivity values range between 2220 and 5200 µS/cm, is a concern. The high
levels of salt in these tributaries come from natural (i.e., deposits of rock salt--halite)
and anthropogenic (i.e., irrigated agriculture) sources in the Khorat Plateau. However,
by Pakse the concentration of salt ions is diluted, and here conductivity is no longer a
significant issue.
· In the Se Bang Hieng, Nam Mun and Nam Chi, the concentration of dissolved oxygen
is relatively low and as a result the measure of `organic matter' ranks as only `fair'. The
higher levels of organic matter in the Se Bang Hieng come from agricultural, forestry and
industrial activities in the river's catchment. The high levels of organic matter in the Nam
Mun and Nam Chi rivers are the result of intensive agriculture on the Khorat Plateau.
· Average concentrations of TSS in the tributaries are lower, or much lower, than in the
mainstream stations downstream of Vientiane.
MRC water-quality database limitations
· Sampling and site location: The MRC set up water-quality sampling stations at existing
gauging stations. These are not necessarily the best sites for assessing water quality and
pollution threats because, sampling frequency is monthly and there is no distinction
(e.g., sampling protocol) between the dry and wet season. Stations located on tributaries
near their confluence with the Mekong suffer backwater and reverse flow effects during
high-water periods, which makes determining causes and effects difficult.
· Parameters and the media sampled: The MRC's database contains mainly the basic
physico-chemical parameters of the river water. In addition, sediments are analysed for
only TSS; biological parameters are not measured, there is little data on pesticide or
industrial pollutants and no toxicity measurements.
· Data quality and gaps: The records from many locations are incomplete. However, the
data that exists is reliable.

21
Diagnostic study of water quality in the Lower Mekong Basin
BDP Sub-area
1. Northern Laos
2. Chaing Rai
LS3
3. Nong Khai/Songkhram
TP2
4. Central Laos
TS1
LS4
5. Mun/Chi
6. Southern Laos
7. Se San/Sre Pok/Se Kong
LS6
8. Kratie
9. Tonle Sap
LS5
10. Delta
TS7
LS8
TS9
TS10
TS11
LP12
CS13 CS23
CS16
CS14
CP15
CP17
CS18
CS19
VP20
VS21
Sampling sites in the 2003 and 2004
VS22
Field Campaigns
2003 & 2004 primary stations
2003 & 2004 secondary stations
0
50 100
200
300 km
Sampled in 2003 only
Figure 8. Sampling stations in the 2003 and 2004 field campaigns
22
4. 2003 and 2004 Field Campaigns
The programme involved two campaigns--during 2003 and 2004. Samples were taken towards
the end of dry season (i.e., March or April) to maximise the ability to observe point sources and
to minimise the effect of dilution and runoff in the wet season. Sampling during the dry season
also eliminates the effects of non-point sources of pollution. The main objectives of the 2003
campaign were to obtain an overview of water-quality throughout the Lower Mekong Basin (in
the Mekong and its tributaries) and to identify priority sites for further investigation. The second
(2004) campaign concentrated on the stations on the Mekong mainstream and the priority sites
that had been identified during the first campaign.
The programme for the 2003 campaign involved selecting the parameters to analyse, the
sampling sites, sampling times and frequencies and those laboratories that would undertake the
analytical work. In order to get the broadest picture of the level of pollution in the Mekong and
its tributaries, the maximum number of parameters were analysed in the first campaign. The
information gained during the 2003 campaign was then used to modify the design of the 2004
campaign.
Sampling sites
The 2003 campaign involved 22 sampling stations. The selection of stations was based on the
results of the review of literature and other material (Chapter 2) and the analysis of the MRC
water-quality data (Chapter 3). These stations are recognised as potential benchmark sites--that
is, sites that are representative of a larger geographical area and that provide a reference against
which future changes in water quality can be measured.
All the sites were located within the 13 `priority sub-basins' identified during `Activity 1' of the
study. Nine of the sites were on the mainstream of the Mekong and 14 were on major tributaries
(Figure 8).
Seven of the 22 sites were dropped from the 2004 campaign because no evidence of pollution
was found at them during the 2003 campaign. A new site, station CS23 (Se San) was added to
the 2004 campaign.
For the following reasons the stations below were designated as `main sites':
· TP2 (Chiang Saen): the most upstream station between Thailand and Lao PDR;
· LP12 (Pakse): trans-boundary contamination from the Khorat Plateau in Thailand;
· CP15 (Kratie): downstream from three trans-boundary catchments (Se Kong: Lao PDR/
Cambodia and Sre Pok and Se San: Viet Nam/Cambodia);
· CP17 (on the Tonle Sap river): representative of the `transition' hydro-ecoregion, an
area well-known for the phenomena of `reverse flow' and susceptible to trans-boundary

23
Diagnostic study of water quality in the Lower Mekong Basin
pollution from the catchments close to the Great Lake and from the upper part of the
Lower Mekong Basin during the wet season;
· VP20 (Tan Chau): a few kilometres from the Cambodia/Viet Nam border.
The other stations are considered as secondary. They were selected mostly to address local
issues that had been identified by the riparian participants. The distribution of these sampling
sites among the four riparian countries is shown in Table 6; most of these stations, with the
exception of six new sites, are part of the MRC Water Quality Monitoring Network.
Table 6. Sampling stations in the 2003 and 2004 campaigns
Country
Watershed
Station name
Code
Station
Field
Number
campaign
Cambodia Mainstream
Kratie
CP15
H014901
2003, 2004
St Baribo
Prek Kdam (Tonle Sap river) CP17
H020102
2003, 2004
Se Kong
Se Kong
CS13
NEW 2003
2003, 2004
Takeo
Koh Khel (Bassac)
CS18
H033402
2003, 2004
Sre Pok
Sre Pok
CS14
NEW 2003
2003, 2004
Upstream Great Lake Bak Prea
CS16
NEW 2003
2003, 2004
Mainstream
Neak Leang
CS19
H019806
2003
Se San
Se San
CS23
NEW 2004
2004
Lao PDR Mainstream
Pakse
LP12
H013901
2003, 2004
Mainstream
Lao/China Border
LS3
NEW 2003
2003, 2004
Nam Ngum
Thangone
LS6
H23102
2003
Mainstream
Luang Prabang
LS4
H011201
2003, 2004
Mainstream
Vientiane
LS5
H011901
2003, 2004
Se Bang Hieng
Ban Keng Done
LS8
H350101
2003, 2004
Thailand Mainstream
Chiang Saen
TP2
H010501
2003, 2004
Nam Mae Kok
Chiang Rae
TS1
H050104
2003
Mainstream
Nakhon Phanom
TS7
H013101
2003
Nam Chi
Yasothon
TS9
H370104
2003
Nam Mun
Rasisalai
TS10
H380134
2003
Nam Mun
Khong Chiam
TS11
H013801
2003, 2004
Viet Nam Mainstream
Tan Chau
VP20
H019803
2003, 2004
Plain of Reeds
My An
VS22
NEW 2003
2003
Bassac
Chau Doc
VS21
H039801
2003, 2004
Sampling programme
Selection of matrices, analyses and sample collection
The assessment of water quality and pollution in the Mekong River system requires multi-media
sampling--water, sediment and biota.
Water. Water analyses gives basic information on pollution in terms of dissolved and non-
dissolved pollutants (e.g., organic matter, major ions, nutrients, pesticides, PAHs and PCBs).
Microbiological measures of pollution (e.g., coliforms and streptococcus) were not recorded
because these factors vary through time and single values (as would have been recorded during
the 2003 campaign) are not very useful, and can be misleading.
24
Diagnostic study of water quality in the Lower Mekong Basin
Organochlorine pesticides, a wide range of industrial contaminants (e.g. PCBs, PAHs, dioxins
and furans), and most metals in neutral pH environments, have low solubility and are mainly
associated in the environment with sediments and biological tissues. These contaminants enter
the food chain partly through the ingestion of fine particles by filter feeders and become more
concentrated upwards through the food chain. They can reach high enough concentrations
in fish and other organisms to cause a variety of problems such as toxicity and/or endocrine
disruption.
Sediments. Bottom sediments represent accumulation over long periods of time (weeks to
months) and are relatively easy to sample. However, it is recognised that they tend to produce
conservative values for sediment-associated chemistry because biological processing of these
compounds by micro-organisms can cause rapid decontamination of the sediments.
The concentration of contaminants in bottom sediments (and their toxicity) are influenced
by several in situ variables such as deposition rates, grain size, organic content and the
geochemistry of the adsorbent coatings on fine particles. Bulk samples of bottom sediment,
uncorrected for grain size, can be used for quick assessments of the levels of contaminants in
bottom sediments at a given site. However, inter-site comparison of uncorrected samples is
unreliable because of variations in grain size, organic content, etc. Whether or not one does bulk
analysis or analysis on a sample that is corrected for factors such as grain size effects, depends
very much on the nature of the questions being asked. In this study the 250 m fraction1 of
freeze-dried and sieved samples was analysed. This fraction contains almost all the contaminant
chemistry. Comparisons between sites, in terms of their toxicity, can be improved using
bioassays of the sediment samples.
Biota. In terms of biota, the original focus was on fish with analyses of bio-markers, pesticides,
antibiotics, PCBs, dioxins and furans. Sampling fish, however, proved difficult, partly because
sedentary species representative of the river fauna (rather than that of wetland habitats) are
not ubiquitous and partly because of the time constraints imposed by the schedule of the 2003
field campaign. Therefore, it was decided not to sample fish during the 2003 campaign and to
focus instead mainly on water and sediment chemistry, while adding an invertebrate bioassay
(Hyalella azteca).
Analyses conducted on the different matrices are presented in Table 7.
2003 field campaign
The sampling programme and analyses was carried out at each of the 22 sampling stations
noted in Figure 8 and Table 6. Only one sample of each substrate at each site was collected and
analysed (Table 8).
1 There are a variety of methods for standardization of sediment samples for contaminant analysis in the literature however there is
no universally accepted method.

25
Diagnostic study of water quality in the Lower Mekong Basin
2004 field campaign
Seven of the 22 stations sampled in the 2003 campaign (Figure 6) were dropped from the 2004
campaign because there was no evidence of pollution at these sites. At the request of Cambodia,
station CS23 was added in the 2004 field campaign, as it is downstream of the area extensively
sprayed with Agent Orange during the American War.
Water analyses at most sites were not repeated as i) major concentrations of ions were
similar to those analysed routinely by MRC over the years; ii) nutrients analyses revealed
significant quality control problems by the laboratory contractor; and iii) the results of some
other parameters were below the level of detection. In general, as the routine water chemistry
was very similar to the data already held by MRC, the focus of the analyses was changed to
contaminants, with an emphasis on sediments.
Two new parameters were added: (i) Total Organic Carbon (TOC) and (ii) cyanide in water and
sediment. The latter was added to determine if the toxin played some role in the high mortality
rate of test organisms used in bioassays of sediments from the station at the Lao/China border
(LS3) and to address mining activities in the catchments at the Luang Prabang (LS4) and Ban
Keng Done (LS8) stations.
A total of 16 stations were sampled. At five stations, where higher levels of contaminants were
found during the 2003 campaign, two samples were taken (about 500 m apart) at each site and
analysed separately. At the other sampling stations, only one sample was collected (Table 8).
The complete sampling and analytical scheme is presented in Tables 7 and 8.
Table 7. Analyses conducted on samples collected during the 2003 and 2004 campaigns
Water
Sediments
Routine Parameters
W1
BS2
TSS, pH, Conductivity, Ca, Mg, Na,
Bioassays with Hyalella azteca
K, Fe, Cu, Zn, Mn, Al, CO , HCO ,
3
3
Cl, SO4, NO , NO , PO , COD
3
2
4
Parameters linked to
W2
P2
industry
BTEX, COHV, PAHs, Total
BTEX, COHV, PAHs, Total
Hydrocarbons, Heavy Metals
Hydrocarbons, Heavy Metals
Parameters linked to
W4
P1
agriculture
Pesticides: Organochloride,
Pesticides: Organochloride,
organophosphorous and triazines
organophosphorous and triazines
P3
PCBs
P4
Dioxins and furans
Parameters possibly linked
CNw
CNs
to Hyalella toxicity and
Cyanide in water
Cyanide in sediments
mining activities
26
Diagnostic study of water quality in the Lower Mekong Basin
Table 8: Sampling programme undertaken during the 2003 and 2004 campaigns--number of
samples collected and analysed by station
CNs
0
0
0
0
0
0
0
0
2
1
0
1
0
0
0
0
0
0
0
0
0
0
0
0
2
1
0
1
0
0
0
0
Analyses
CNw
T
otal
New
T
OC:
2
2
1
1
1
1
1
1
2
1
1
0
1
1
2
2
T
OC
ganophosphate,
BS2
(1)
1
0
1
0
1
0
1
2
1
0
0
1
1
1
1
P4
2
2
1
1
1
1
(1)
1
2
1
1
0
1
1
2
2
AIGN (March 2004)
P3
2
2
1
1
1
1
1
1
2
1
1
0
1
1
2
2
bioassays on sediments;
Sediments
P2
2
2
1
1
1
Station not sampled in 2004
1
1
1
2
Station not sampled in 2004
1
1
1
1
Station not sampled in 2004
Station not sampled in 2004
Station not sampled in 2004
Station not sampled in 2004
1
2
Station not sampled in 2004
2
FIELD CAMPnd
2
P1
2
2
0
1
0
1
0
1
2
1
1
0
1
1
2
2
Hyalella azteca
W4
0
0
0
0
0
0
0
0
2
0
0
0
0
0
0
0
ater
ganochloride and triazines); P1: Pesticide (or
W
W2
0
0
0
0
0
0
0
0
2
0
0
1
0
0
0
0
BS2
1
0
0
0
0
0
(1)
1
1
0
0
0
0
1
0
0
0
0
1
1
0
1
P4
1
1
0
0
1
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
ganophosphate, or
P3
1
1
0
1
0
0
1
1
1
0
0
0
0
1
0
0
1
1
1
1
0
1
Sediments
P2
1
1
0
1
1
0
1
1
1
0
0
0
1
1
0
0
0
0
1
1
0
1
W4: Pesticide (or
AIGN (April 2003)
P1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
W4
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
Station not sampled in 2003
FIELD CAMPst
1
ater
1
1
0
1
0
0
0
1
1
0
0
0
1
1
0
0
0
0
1
1
0
1
W
W2
W1
1
1
1
1
1
1
0
1
1
0
1
1
1
1
1
1
1
1
1
1
0
1
1
AHs, total hydrocarbons, heavy metals; P3: PCBs; P4: Dioxins and furans; BS2:
, P
CODE
CP15
CP17
CS13
CS18
CS14
CS16
CS19
CS23
LP12
LS3
LS6
LS4
LS5
LS8
TP2
TS1
TS7
TS9
TS10
TS1
VP20
VS22
VS21
AHs, total hydrocarbons, heavy metals;
, P
onle Sap)
W2: BTEX, COHV
A
TION NAME
An
ientiane
ST
Kratie
Prek Kdam (T
Se Kong
Koh Khel (Bassac)
Sre Pok
Bak Prea
Neak Leang
Se San
Pakse
Lao/China border
Thangone
Luang Prabang
V
Ban Keng Done
Chiang Saen
Chiang Rae
Nakhon Phanom
Y
asothon
Rasisalai
Khong Chiam
T
an Chau
My
Chau Doc
Y
ganochloride and triazines); P2: BTEX, COHV ganic carbon; CNw: Cyanide in water; CNs: Cyanide in sediments; (1): Sample taken but not analysed
BODIA
NAM


W1: Routine parameters; or or
ET
COUNTR
CAM
LAO PDR
THAILAND
VI
Note:

27
Diagnostic study of water quality in the Lower Mekong Basin
Analytical methodology
Laboratories
Four laboratories carried out the analytical work:
· The Pollution Control Department Laboratory (Thailand) carried the analytical work for
the routine parameters in water (W1 samples); most of these parameters are unstable and
needed to be analysed as quickly as possible.
· CEMAGREF (France)--bioassays.
· CARSO, (France)--pesticides in water and sediments.
· LEM COFRAC, (Savenne, France)--all other parameters in water or sediments.
Sample transportation
Water and sediment samples were packed in cool boxes (covered with ice wherever available)
and shipped by Federal Express to the destination laboratories. Sample conservation and
transportation conditions were not always optimal due to the lack of ice in some areas,
transportation delays relating to customs procedures, and other factors.
Sampling and analytical methodologies
The sampling and analytical methodologies are briefly summarised in the next three sub-
sections.
Water sampling and analysis
Water samples were taken from the middle section of the river, at a depth of between 20 to
30 cm below the surface.
Sediment sampling and analysis
At each site, a sediment grab was used to take samples. Samples were recovered at three
different places forming a triangle in order to integrate the spatial variability of the area
(Figure 9). The sediment samples from these three areas were then mixed together and stored in
jars.
For volatile parameters (BTEX, COHV, etc.), analyses were performed on bulk sediments. For
all other parameters, a physico-chemical processing was undertaken (lyophilization, screening
to 250 µm, acid mineralization, etc.) before analysis in order to have homogeneous samples that
are representative of the original sample.
28
Diagnostic study of water quality in the Lower Mekong Basin
S1
River bank
100 m
100 m
Sampling area
S2
S3
100 m
S1
Sampling site
Figure 9. Sediment sampling protocol
Bioassays
In addition to chemical analyses, sediment bioassays were conducted on sediments sampled
from selected stations (Table 8) to provide information on the toxicity of the sediments. The
bioassays were performed using a crustacean amphipod, Hyalella azteca. The measured end
points were survival, for acute toxicity, and length, for growth inhibition.
Bioassays from 2003 campaign were conducted according to the draft standards proposed by the
Association Française de Normalisation (AFNOR). For the 2004 campaign, the standards used
were according to AFNOR Standard T 90.338.1 (AFNOR, 2002). The main difference between
the draft and final standard is a small difference in age of the organism at the start of the test
(2­9 days for the draft standard, and 4­12 days for the final standard). These two tests produced
comparable results, and it was assumed that small difference in the initial age would not induce
differences in sensitivity. There was no independent test of this assumption; however, according
to ASTM, `the sensitivity of H. azteca appears to be relatively similar up to at least 24­26 day
old organisms (Collyard et al., 1994)'.
The environmental conditions in the AFNOR standard are similar to those in the ASTM
standard (ASTM, 2000) except for the test duration2 and the age of Hyalella at the beginning of
the test3. It is also similar to the USEPA standard (USEPA, 2000), except that the composition of
the control sediments4 are different.
The following set of conditions were used, in accordance with the draft or published AFNOR
standards:
· From their arrival and until the bioassays were performed, the sediment samples were
stored at 7°C.
2 Test duration of 10 (ASTM) and 14 (AFNOR) days.
3 Age of 7 to 14 days (ASTM) and 4 to 12 days (AFNOR) at start of the test.
4 In the US EPA standard, the sediment control is not precisely specified, but it has to allow for the survival, the growth, and the
reproduction of a variety of benthic invertebrates; in the AFNOR standard, the composition of the sediment control is specified
(mixture of sand and organic matter).

29
Diagnostic study of water quality in the Lower Mekong Basin
· In the instance of the 2004 field campaign, sediments arrived at the CEMAGREF
laboratory in two batches with a ten-day interval in between; bioassays were carried out
in three series over three different time-periods: once on the first batches of sediments
(samples from Thailand and Lao PDR) and twice on samples from the second batch (from
Cambodia and Viet Nam). The first series of bioassays on the second batch was rejected
since the validity criteria were not respected and a new series was assayed.
· Two control tests were undertaken 10 days before the start of the assays using silica sand
(Fontainebleau sand, 150­210 µm diameter) enriched with organic matter (fish food
Tetramin®: 4 g for 2 l of sand).
· Five replicates were done with each sediment sample; 10 organisms were used in each
replicate.
· Hyalella were raised in the laboratory. Their age at the beginning of the assay ranged
from 2 to 9 days for samples from the 2003 campaign, and 5 to 10 days for the 2004
campaign.
· The tests were carried out with continuous water renewal, using four times the volume of
the water column per day.
· The test temperature was 23° ± 1°C.
· Physico-chemical parameters (pH, conductivity, temperature, nitrites and ammonia)
were measured at least four times during each series of bioassays; no abnormalities in
the measurements occurred during the two valid test series as the ammonia concentration
remained below 9.7 mg/l5, which is considered as acceptable (Whiteman et al., 1996).
· Minimum survival rate in the control is 70%.
· The survival and length measurements were recorded only at the end of the test (day 14):
The organisms were counted, frozen and measured after thawing. The distance between
the base of the first antenna and the extremity of the abdomen (top of the third uropod)
was measured to ± 0.1 mm using a stereo-microscope at x10 magnification.
Data treatment and processing
Heavy metal toxicants
· Total metal concentrations: The total concentration at each station was calculated by
summing all the heavy metal concentrations (values the under detection limit were
considered as zero) at that station.
· Threshold effect concentration (TEC) and probable effect concentration (PEC): The
TEC and PEC guidelines established by MacDonald et al. (2000) were used to assess
the exposure of benthic organism to metal toxicants in sediment and to provide an
estimation of the toxicity of the sediments (Table 9). The guidelines provide a `threshold
5 LC 96h : 9.7mg/l total nitrogen compound (NH +NH ).
50
4
3
30
Diagnostic study of water quality in the Lower Mekong Basin
effect concentration' (TEC), below which toxic effects are unlikely, and a `probable
effect concentration' (PEC) above which toxic effects are highly likely to occur. These
guidelines, although developed in a context rather different than that prevailing in the
Lower Mekong Basin, seem appropriate as relative indicators for assessing potentially
toxic sediments.

Table 9. Threshold level effect concentration (TEC)
and probable effect concentrations (PEC)
guideline values for heavy metals established
by MacDonald et al., (2000)
Parameters
TEC (mg/kg)
PEC (mg/kg)
Arsenic
9.79
33.00
Cadmium
0.99
4.98
Chromium
43.40
111.00
Copper
31.60
149.00
Nickel
22.70
48.60
Lead
35.80
128.00
Zinc
121.00
459.00
Mercury
0.18
1.06
Note: concentrations are expressed in terms of total concentrations of each
metal
· Threshold Effect Level (TEL) and Probable Effect Level (PEL): TEL and PEL are interim
guidelines of sediment-quality developed by the Canadian Council of Ministers of the
Environment (CCME, 1999-2002) for assessing the potential toxicity of sediment-
related PCBs and PAHs. They are different from the TEC and PEC values given by
MacDonald et al. (2000) as they are calculated differently, but they have similar
significance.
· Metal contamination assessment: The evaluation of overall contamination of metals and
their potential toxicity to benthic organisms was assessed and interpreted using the `mean
quotient' method.
This a parametric method that accounts to a certain extent for additive effects. To obtain
the mean quotient at each station for each year, the concentration of individual metals
is divided by its PEC; all individual quotients at each site/year are then summed, and
the resulting number is divided by the number of measured parameters. A significant
increase in the toxicity incidence occurs at mean quotient values above 0.1. Where the
mean quotient is above 0.5, about 80% of the samples are considered toxic according to
MacDonald et al. (2000).
The mean quotient method was used to interpret the data in the Sediment Results section
because it allowed identification of those metals that contributed most to the score and
the simplified scoring approach, which is much less discriminatory, for the multi-criteria
analysis.

31
Diagnostic study of water quality in the Lower Mekong Basin
Dioxins and furans in sediment
Toxicity Equivalent Index (I-TEQ): Of the 210 dioxins and furans, only 17 are recognised as
toxic. These 17 dioxins and furans have a toxicity ranging from 1 to 0.0001. The estimation
of the toxicity of a sediment sample is derived from the quantitative measurement of these
17 toxic congeners to which is applied the respective toxicity equivalent factor (I-TEF). The
concentrations of the compounds multiplied by their respective TEF are then summed to obtain
a Toxicity Equivalent Index (I-TEQ). Determining I-TEQs for sediment samples will only
indicate a potential problem; the actual exposure and toxicity of any species cannot be inferred
from the I-TEQs. The I-TEFs initially proposed by Safe and Phil (1990) were used to calculate
I-TEQs of the sampled sediments. Based on an interim sediment-quality guideline (ISQG) of
0.85 pg/g proposed in the Canadian Environmental Quality Guidelines (CCME, 1999-2002) a
I-TEQ value of 0.1 pg/g was used as background value for the Mekong River Basin sediments6.
This is far below the 21.5 pg/g Probable Effect Level (PEL) level proposed by CCME.
Polycyclic Aromatic Hydrocarbons (PAHs) in sediments
TEL and PEL values are provided for most of the individual toxic PAHs in the Interim Sediment
Quality Guidelines (CCME, 1999­2002). TEC and PEC values of 1,610 g/g and 22,800 g/g
respectively, for PAHs in sediments are from MacDonald et al. (2000) and are used in this study
as PAHs were detected at only three stations, and at low values.
Bioassays
Statistical Tests: In results from the 2003 field campaign, significant differences from the control
(p<0.05) were found by hypothesis tests. Dunnett's test was used--following verification using
the Shapiro-Wilk's tests for normality, and the Hartley's test for homogeneity of variance.
Calculations were performed using TOXSTAT 3.0 software (Gulley et al., 1989). Significant
differences in the results of the 2004 field campaign were tested using the Bonferroni t-test
(p<0.05) 7.
Multi-criteria analysis
Data from the 2003 and 2004 field campaigns and results from the bioassays were used to
perform a multi-criteria analysis. This provides a more synthetic and comprehensive picture
of the water-quality of the Mekong system. Parameters for which quantifiable results had been
obtained were selected for this multi-criteria analysis. These included heavy metals and arsenic,
PAHs, dioxins and furans and H. azteca bioassays. For each parameter a scoring method similar
to that used for heavy metals was used to enable an overall comparison of the stations:
· Scoring method for heavy metals and arsenic: (for details see the section on heavy metals
presented earlier in this chapter).
· Scoring method for Total PAHs: For each station, if the total PAHs concentration is below
the TEC value, the score is 0; if it ranges between the TEC and PEC values, the score is 1,
6 The interpretation of the results would change somewhat had a value of 0.85 pg/g been chosen rather than 0.1 pg/g. This has an
impact on the subsequent interpretation of the results, as noted below.
7 These values are applied to the total PAHs in this study.
32
Diagnostic study of water quality in the Lower Mekong Basin
and if it is equal or above PEC value, the score is 2. The higher the score the more likely
the toxicity to benthic organisms.
· Scoring method for Dioxin/Furans (PCDD/PCDFs): A score of 1 is given when the
threshold of 0.1 pg/g used by BURGÉAP (2005) is exceeded, and a score of 2 when the
CCME (1999-2002) Interim Sediment Quality Guidelines of 0.85 pg/g is exceeded.
· Scoring method for H. azteca: According to the hazard ranking system elaborated for
the French Ministry of Transport (Babut et al., 2004), a score 0 is attributed when both
survival and growth are similar to that found in the control (10%), a score of 2 is given
when either the survival or growth deviated from the control by more than 50%, and
otherwise a score of `1' is assigned. This approach to scoring toxicity test results therefore
accounts for two end-points, encompassing a wider range of toxic stress causes.
Table 10. Threshold values used in the multi-criteria analysis
Scores and Thresholds
Parameters
0
1
2
Heavy metals and
TEC < Concentration <
Concentration < TEC
Concentration > PEC
arsenic
PEC
TEC < Concentration <
Total PAHs
Concentration < TEC
Concentration > PEC
PEC
0.1 pg/g < I-TEQ < 0.85
Dioxins and Furans
I-TEQ < 0.1 pg/g
I-TEQ > 0.85 pg/g
pg/g
Mortality or growth 10% 10%< mortality or growth Mortality or growth >50%
H. azteca
of control
50% of control
of control
For each station, the sum of all the scores (for each parameter) is divided by the number of
parameters used in order to standardise the stations scores, as not all the stations have values for
all the parameters.

33
Diagnostic study of water quality in the Lower Mekong Basin
34
5. Results and Discussion
Water
The results outlined below are mostly from the 2003 campaign. As noted above, in 2004, water
samples were collected and analysed at only two stations (Lao/China border--LS3, and Ban
Keng Done--LS8) for industrial pollutants and pesticides.
Major ions
The results from the 2003 campaign show that the Mekong River system has lower
concentrations of mineral ions than some other major rivers of the world (Figure 10).
160
Mekong (2003)
140
Mississippi (1965/67)
120
Saint Laurent (1968)
n
100
Rhine (1971/72)
t
r
a
t
i
o
/
l
)
g
80
c
e
n
(
m
n
60
o
C
40
20
0
Cl-
SO 2-
Mg2+
Ca2+
Na+
K+
CaCO
4
3
Figure 10. Comparison of the chemical profile of the Mekong with other rivers
120
Cl-
SO 2-
4
100
Mg2+
Ca2+
Na+
80
K+
n
CaCO3
t
r
a
t
i
o
/
l
)
g 60
c
e
n
(
m
n
o
C
40
20
0
China border C. Saen L. Prabang Vientiane N. Phanom
Pakse
Kratie
Tan Chau
LS3
TP2
LS4
LS5
TS7
LP12
CP15
VP20
Figure 11. The major ion profiles in the Mekong from upstream (left) to downstream (right)

35
Diagnostic study of water quality in the Lower Mekong Basin
The concentrations of some major ions (e.g., CaCO , Ca2+, and SO 2-) show a tendency to
3
4
decrease from upstream to downstream stations along the Mekong River (Figure 11). These
results have to be taken cautiously, as only one sample was analysed at each station.
Salt contamination from the Khorat Plateau
On the Nam Mun river, high concentrations of chloride and sodium ions (302 and 177 mg/l
respectively) were recorded at the upstream station of Rasisalai (TS10). The concentrations
downstream at Khong Chiam (TS11) are significantly lower (Cl- = 22.3 mg/l and Na+ =
12 mg/ l). This reduction results from dilution caused by the discharge of the Nam Chi river
at Yasothon station (TS9) (Figure 12). These results were observed during the dry season; it is
probable that a similar picture would emerge during wet season flow.
At Pakse (LP12), downstream of the confluence of the Nam Mun and the Mekong River,
chloride and sodium concentrations are back to normal with levels (around 9.6 and 8
mg/l respectively). This demonstrates that, as Hart et al. (2001) have already noted, salt
contamination from the Khorat Plateau carried by the Nam Mun has very little impact on the
salinity of the Mekong river.
350
Cl-
SO 2-
300
4
Mg2+
Ca2+
250
Na+
n
K+
200
t
r
a
t
i
o
/
l
)
g
c
e
n
(
m
n
150
o
C
100
50
0
Mun River
Nam Chi River
Mun River
Rasisalai TS10 (T)
Yasothon TS9 (T)
Khong Chiam TS11 (T)
Figure 12. Major ion profiles of tributaries on the Khorat Plateau
Nutrients
Concentrations of nutrients were low despite some evidence of eutrophication (at Koh Khel--
CS18) and anthropogenic pollution (at Bak Prea--CS16 and Prek Kdam--CP17) recorded
during the 2003 field campaign. These higher levels were generally found at sites located in
areas with high population density and/or poor sewage facilities. However, due to poor quality
control in the contracted laboratory these data are not reliable.
Industrial contaminants and pesticides
The water samples analysed from both campaigns were below the detection limits for industrial
contaminants and pesticides.
36
Diagnostic study of water quality in the Lower Mekong Basin
Sediments
Total heavy metals concentrations in sediments
2003 campaign
200
a
tion g)
tr
en
100
(mg/k
onc
C
0
LS3
TP2
LS8
TS11
LP12
CS14
CS15
CP17
CS18
CS19
VP20
VS21
(M)
(M)
(T)
(T)
(M)
(T)
(M)
(T)
(M*)
(M)
(M)
(M*)

Total heavy metals concentration
(M)
Mekong
(T)
Tributary
(M*)
Bassac
Figure 13. Total heavy metals concentrations recorded during the 2003 campaign
The concentrations of heavy metals recorded during the 2003 campaign are shown in Figure 13.
The results indicate that, in the Mekong river, the sites at the Lao/China border (LS3) and Neak
Leang (CS19) have elevated levels of heavy metals. The station at Neak Laeng may be affected
by the Tonle Sap river, as the Prek Kdam (CP17) station, which is also on the Tonle Sap, shows
the highest level of total heavy metals. The second highest concentration of total heavy metals is
observed at Chau Doc (VS21), on the Bassac River. Koh Khel (CS18), also on the Bassac, has
higher levels of heavy metals. Both sites are downstream of Phnom Penh.
2004 campaign
160
a
tion
tr
en
120
onc
g)
80
(mg/k
v
y metals c
40
T
otal hea
0
LS3
TP2
LS4
LS5
LP12
CS13
CP15
CS19
VP20
(M)1
(M)
(M)
(M)
(M)
(M)
(M)1
(M)
(M)1

2003 Campaign
(M)
Mekong

2004 Campaign
(M)1
Average of values
where two samples
were collected in 2004
Figure 14. Total heavy metals concentrations recorded during the 2003 and 2004 campaigns

37
Diagnostic study of water quality in the Lower Mekong Basin
The total heavy metal concentrations found at Kratie (CP15) and Tan Chau (VP20) were higher
in 2004 than in 2003 (Figure 14). The other four stations that were sampled during both years
have similar concentrations.
In summary, based on total heavy metals concentration results, within-site and between-site
variations cannot be explained on the basis of the one or two samples collected at each site once
a year during each campaign. However, at present, it can be reasonably inferred that the stations
with higher total heavy metal concentrations are located in areas with significant boat traffic
and/or with high population densities:
· Lao/China border (LS3): Commercial shipping highway between China and Lao PDR;
· Luang Prabang (LS4): Large volume of tourist-boat traffic;
· Vientiane (LS5): High population density;
· Kratie (CP15): Large volume of boat traffic for tourism and transportation;
· Prek Kdam - Tonle Sap river (CP17): Boat traffic and some industrial activities;
· Koh Khel, Bassac (CS18) and Neak Leang, Mekong (CS19): Downstream of Phnom
Penh;
· Tan Chau, Mekong (VP20): High population density in the Mekong Delta;
· Chau Doc, Bassac (VS21): High population density in the Mekong Delta.
Heavy metal concentrations in sediments and their eco-toxicity potential
In order to assess the potential toxicity of the sediments collected during both campaigns, total
heavy metal concentrations were compared against threshold values for TEC and PEC; these
follow MacDonald et al. (2000).
Arsenic
In 2003, only the sediment from the Lao/China border (LS3) showed a concentration above the
TEC threshold (9.79 mg/kg). In 2004, six stations registered levels above this threshold --Lao
/China border (LS3), Luang Prabang (LS4), Vientiane (LS5), Prek Kdam (CP17), Tan Chau
(VP20) and Chau Doc (VS21). However, concentrations at three of these sites only slightly
exceeded the TEC threshold. Higher arsenic concentrations were observed at the three others:
Lao/China border, Vientiane and Prek Kdam stations (Figure 15). However, at none of the sites
did the levels of arsenic approach the PEC threshold.
It is important to note that the observed differences in arsenic concentrations between 2003 and
2004 exceed the analytical uncertainty for arsenic (which is around 22%). These differences
could be attributed to a number of factors including sampling variation, heterogeneity at the
sites, or a real increase in concentrations of arsenic. Unfortunately, the limited data set does not
allow resolution of which of these factors is the cause.
38
Diagnostic study of water quality in the Lower Mekong Basin
It should also be noted that, as in other countries in S.E. Asia, arsenic is a naturally occurring
mineral in the countries of the Lower Mekong Basin.
35
PEC threshold
30
25
n
) 20
t
r
a
t
i
o
/
kg
g
c
e
n
(
m 15
n
o
C
10
TEC threshold
5
Detection limit 2003
Detection limit 2004
0
)
1
)
)
)
T
)
)
T
)
T
)
)
)
1
*
)
)
)
1
M
M
M
M
M
M
M
T
)
1
M
M
M
*
)
1
M
L
S
8 (
T
S
1
1
(
T
)
S
2
3 (
S
1
4 (
S
1
3 (
1
7 (
S
1
9 (
L
S
3 (
T
P
2 (
L
S
4 (
L
S
5 (
L
P
1
2 (
C
C
C
1
5 (
S
1
8 (
C
CP
CP
C
P
2
0 (
V
S
2
1 (
V

As 2003 Campaign
(M)
Mekong

As 2004 Campaign
(M*)
Bassac
(T)
Tributary
()1
Average of values where
two samples were collected
in 2004
Figure 15. Concentrations of arsenic recorded during the 2003 and 2004 campaigns
Cadmium
No analyses in either campaign detected cadmium. The analytical detection limit of cadmium
is 1 mg/kg, which is very close to the TEC threshold of 0.99 mg/kg. It is assumed that all the
sediment samples at all sites are unlikely to contain toxic levels of this metal.
Chromium
PEC threshold
100
80
n )
t
r
a
t
i
o
60
/
kg
g
c
e
n
n (m
TEC threshold
o
40
C
20
0
)
1
)
)
)
)
)
)
(
T
)
(
T
)
(
T
)
)
1
*
)
)
1
*
)
1
(
M
(
M
(
M
(
M
(
M
(
M
(
M
(
T
)
1
(
M
(
M
(
M
L
S
8
S
2
3
S
1
4
(
M
L
S
3
T
P
2
L
S
4
L
S
5
T
S
1
1
(
T
)
L
P
1
2
C
C
S
1
3
S
1
9
C
P
1
5
P
1
7
S
1
8
C
P
2
0
C
C
C
V
S
2
1
V

Cr 2003 Campaign
(M)
Mekong

Cr 2004 Campaign
(M*)
Bassac
(T)
Tributary
()1
Average of values where
two samples were collected
in 2004
Figure 16. Concentrations of chromium recorded during the 2003 and 2004 campaigns
Chromium concentrations were slightly above the TEC threshold at only one site, (Sre Pok--
CS14) and during one year (2004) (Figure 16). At this site in 2003, the concentration was

39
Diagnostic study of water quality in the Lower Mekong Basin
slightly below the same threshold. The 23% difference in concentration observed between the
two campaigns is higher than the analytical uncertainty for chromium, which is usually around
13%. This difference is attributed to sample variation and not to analytical variability.
Copper
TEC threshold
30
n ) 20
t
r
a
t
i
o
/
kg
g
c
e
n
n (m
o
C
10
0
)
1
)
)
)
T
)
)
T
)
T
)
)
)
1
*
)
)
)
1
M
M
M
M
M
M
M
T
)
1
M
M
M
*
)
1
M
L
S
8 (
T
S
1
1
(
T
)
S
2
3 (
S
1
4 (
S
1
3 (
1
7 (
S
1
9 (
L
S
3 (
T
P
2 (
L
S
4 (
L
S
5 (
L
P
1
2 (
C
C
C
1
5 (
S
1
8 (
C
CP
CP
C
P
2
0 (
V
S
2
1 (
V

Cu 2003 Campaign
(M)
Mekong

Cu 2004 Campaign
(M*)
Bassac
(T)
Tributary
()1
Average of values where
two samples were collected
in 2004
Figure 17. Concentrations of copper recorded during the 2003 and 2004 campaigns
The concentrations of copper measured in the sediments samples in both campaigns were all
below the TEC threshold (Figure 17). Benthic organisms are unlikely to suffer from toxicity
caused by the concentrations of this metal recorded in these sediments.
Nickel
35
30
25
n
TEC threshold
)
20
Detection limit 2004
t
r
a
t
i
o
/
kg
g
c
e
n
n (m
15
o
C
10
Detection limit 2003
5
0
)
1
)
)
)
T
)
)
T
)
T
)
)
)
1
*
)
)
)
1
M
M
M
M
M
M
M
T
)
1
M
M
M
*
)
1
M
L
S
8 (
T
S
1
1
(
T
)
S
2
3 (
S
1
4 (
S
1
3 (
1
7 (
S
1
9 (
L
S
3 (
T
P
2 (
L
S
4 (
L
S
5 (
L
P
1
2 (
C
C
C
1
5 (
S
1
8 (
C
CP
CP
C
P
2
0 (
V
S
2
1 (
V

Ni 2003 Campaign
(M)
Mekong

Ni 2004 Campaign
(M*)
Bassac
(T)
Tributary
()1
Average of values where
two samples were collected
in 2004
Figure 18. Concentrations of nickel recorded during the 2003 and 2004 campaigns
During both campaigns, the concentration of nickel was well below the TEC threshold
(Figure 18). In 2004, the concentrations of nickel at most of the sites was below the detection
40
Diagnostic study of water quality in the Lower Mekong Basin
limit used that year. Due to modifications in the analytical protocol, the detection limit used
in 2004 was higher than that used in 2003. However, as both detection limits were below the
TEC threshold, it is unlikely that nickel in the sediments could have toxic effects on benthic
organisms.
Lead
40
TEC threshold
30
n )
t
r
a
t
i
o
/
kg
g
20
c
e
n
n (m
o
C
10
Detection limit
0
)
1
)
)
)
T
)
)
T
)
T
)
)
)
1
*
)
)
)
1
M
M
M
M
M
M
M
T
)
1
M
M
M
*
)
1
M
L
S
8 (
T
S
1
1
(
T
)
S
2
3 (
S
1
4 (
S
1
3 (
1
7 (
S
1
9 (
L
S
3 (
T
P
2 (
L
S
4 (
L
S
5 (
L
P
1
2 (
C
C
C
1
5 (
S
1
8 (
C
CP
CP
C
P
2
0 (
V
S
2
1 (
V

Pb 2003 Campaign
(M)
Mekong

Pb 2004 Campaign
(M*)
Bassac
(T)
Tributary
()1
Average of values where
two samples were collected
in 2004
Figure 19. Concentrations of lead recorded during the 2003 and 2004 campaigns
All of the lead concentrations were below the TEC threshold value (Figure 19). Sediments
from stations CS23 (Se San) and CP17 (Prek Kdam) have higher concentrations than do the
other sites. The analytical uncertainty of lead analysis is low (approximately 14%), therefore
these higher values are not the result of analytical variability, but because of some other natural
phenomenon.
Zinc
TEC threshold
n )
t
r
a
t
i
o
/
kg
g
c
e
n
n (m
o
C
)
1
)
)
)
T
)
)
T
)
T
)
)
)
1
*
)
)
)
1
M
M
M
M
M
M
M
T
)
1
M
M
M
*
)
1
M
L
S
8 (
T
S
1
1
(
T
)
S
2
3 (
S
1
4 (
S
1
3 (
1
7 (
S
1
9 (
L
S
3 (
T
P
2 (
L
S
4 (
L
S
5 (
L
P
1
2 (
C
C
C
1
5 (
S
1
8 (
C
CP
CP
C
P
2
0 (
V
S
2
1 (
V

Zn 2003 Campaign
(M)
Mekong

Zn 2004 Campaign
(M*)
Bassac
(T)
Tributary
()1
Average of values where
two samples were collected
in 2004
Figure 20. Concentrations of zinc recorded during the 2003 and 2004 campaigns

41
Diagnostic study of water quality in the Lower Mekong Basin
Concentrations of zinc in the Mekong River and tributaries during both the 2003 and 2004
campaigns were below the TEC threshold (Figure 20). The highest levels are observed at Prek
Kdam (CP17). This difference exceeds analytical uncertainty for zinc (approximately 10%)
therefore the difference at CP17 is due to some other cause.
Mercury
300
250
n
200
)
TEC threshold
t
r
a
t
i
o
/
kg 150
g
c
e
n
n (u
o
C
100
Detection limit
50
0
)
1
)
)
)
T
)
)
T
)
T
)
)
)
*
)
)
)
*
)
M
M
M
M
M
M
M
T
)
1
M
M
M
M
L
S
8 (
T
S
1
1
(
T
)
S
2
3 (
S
1
4 (
S
1
3 (
1
5 (
1
7 (
S
1
9 (
L
S
3 (
T
P
2 (
L
S
4 (
L
S
5 (
L
P
1
2 (
C
C
C
CP
P
2
0 (
CP
S
1
8 (
S
2
1 (
C
C
V
V

Hg 2003 Campaign
(M)
Mekong

Hg 2004 Campaign
(M*)
Bassac
(T)
Tributary
()1
Average of values where
two samples were collected
in 2004
Figure 21. Concentrations of mercury recorded during the 2003 and 2004 campaigns.
Concentrations of mercury at most of the stations sampled in 2003 and 2004 were below the
detection limit of 100µg/kg, which is considered high (Figure 21). Mercury was detected
at six stations; the TEC threshold was exceeded at three stations: Prek Kdam (CP17) with
concentrations 272 µg/kg (average of upstream and downstream stations) in 2004; Neak
Leang (CS19) with concentrations in 2004 of around 200 µg/kg; and Kratie (CP15), where Hg
concentrations only just exceeded the threshold of 180 µg/kg.
Summary of arsenic and metal contamination in sediments
0.3
t
0.2
t
i
e
n
o
u
e
a
n q
0.1
Specific attention limit
M
0
)
)
)
)
)
)
)
)
)
M
M
M
M
T
)
T
)
M
T
)
T
)
M
M
T
)
*
)
*
)
M
M
M
M
L
S
3 (
T
P
2 (
L
S
4 (
L
S
5 (
L
S
8 (
T
S
1
1 (
S
2
3 (
S
1
4 (
P
1
7 (
L
P
1
2 (
C
C
S
1
3 (
P
1
5 (
S
1
9 (
C
C
C
S
1
8 (
P
2
0 (
S
2
1 (
C
C
V
V

2003 Campaign
(M)
Mekong

2004 Campaign (up stream)
(M*)
Bassac

2004 Campaign (down stream)
(T)
Tributary
Figure 22. Mean quotient of arsenic and heavy metals recorded during the 2003 and 2004 campaigns.
42
Diagnostic study of water quality in the Lower Mekong Basin
The overall metal contamination and potential toxicity to benthic organisms is assessed
using the `mean quotient' approach which accounts to a certain extent for additive effects. A
significant increase in the toxicity incidence occurs when values are above 0.1 (specific attention
limit), at values above 0.5 it is estimated that 80% of the samples are toxic (MacDonald et al.,
2000).
Using this approach, it can be concluded that metal contamination occurs at low to moderate
levels in the sediments at several sites (Figure 22). No sites exceeded the 0.5 value. Of the
stations having data from both campaigns, four stations recorded measurements above the
specific attention limit (0.1) in both years:
· Lao/China border (LS3): The metals contributing most to these results were arsenic,
chromium and nickel;
· Prek Kdam (CP17): The main contributions to the mean quotient stem from arsenic,
chromium, mercury and lead, along with copper and zinc to a lesser extent;
· Chau Doc (VS21): The metals contributing most to these results were copper, nickel, and
arsenic;
· Sre Pok (CS14): Chromium was the metal that contributed most to this result. This would
appear to be inconsistent insofar that there is little reason to believe there should be
metals in sediments in this relatively undeveloped trans-boundary tributary. On the Viet
Nam side the Yok Don National Park lies at the international boundary.
Other stations are above this limit, but are not confirmed as they have only one year's data set.
Dioxins and furans in sediments
1.0
2003
2004 (up)
0.8
2004 (down)
)
Mean
/
g
0.6
g
p
0.4
I
-
T
E
Q (
0.2
0.0
LS3 (M) CS14 (T) CP15 (M) CP17 (T) VP20 (M) VS21 (M*)
(M) = Mekong (M*) = Bassac (T) = Tributary
Figure 23. Toxicity Equivalent Index (I-TEQ) variability for dioxins and furans (mean values and
standard deviations at stations where samples were collected upstream and downstream of the
station in 2004).

43
Diagnostic study of water quality in the Lower Mekong Basin
Four stations were sampled and analysed in 2003 and 14 in 2004. In the second campaign,
sediments were sampled up- and downstream of some of the sites to provide some indication of
spatial variance of polychlorodibenzodioxins (PCDDs) and polychlorodibenzofurans (PCDFs).
The results of PCDDs/PCDFs concentrations in sediments are presented in terms of their
Toxicity Equivalent Index (I-TEQ).
At some stations the difference in I-TEQ between the up- and downstream samples is not large
(Figure 23). However, at Prek Kdam (CP17) and Kratie (CP15) there is approximately a four
fold difference in I-TEQ concentrations. As the analytical uncertainty for dioxins and furans
analysis is around 15%, the observed differences are likely the result of spatial variation.
The I-TEQ values obtained at all the sampled situations in both years are presented in Figure 24
using a threshold of 0.1 pg/g.
0.7
Campaign 2003
0.6
Campaign 2004
0.5
)
/
g
g
0.4
p
0.3
I
-
T
E
Q (
0.2
0.1
Specific attention limit
0.0
)
1
)
)
)
)
T
)
T
)
)
1
*
)
)
)
1
M
M
M
M
M
M
T
)
1
M
M
M
*
)
1
M
T
S
1
1
(
T
)
S
2
3 (
S
1
4 (
1
7 (
S
1
9 (
L
S
3 (
T
P
2 (
L
S
4 (
L
S
5 (
L
P
1
2 (
C
C
1
5 (
S
1
8 (
C
CP
CP
C
P
2
0 (
V
S
2
1 (
V

2003 Campaign
(M)
Mekong

2004 Campaign
(M*)
Bassac
(T)
Tributary
()1
Average of values where
two samples were collected
in 2004
Figure 24. Toxicity Equivalent Index (I-TEQ) values of dioxins and furans
recorded during the 2003 and 2004 campaigns.
The six sites with I-TEQ values above 0.1 pg/g are Khong Chiam (TS11), Prek Kdam (CP17),
Koh Khel (CS18), Neak Leang (CS19), Tan Chau (VP20) and Chau Doc (VS21). However, this
value is lower than the thresholds (0.85 pg/g) in sediment-quality provided by the Canadian
guidelines. If the higher value is used, all stations fall below levels that may cause concern.
Nevertheless, the higher values observed at Prek Kdam (CP17) merit further investigation. The
values downstream from Prek Kdam (in Cambodia) into Viet Nam may indicate transport of
some toxins across the border, or they may be generated domestically within Viet Nam.
Other parameters analysed in sediments
Pesticides
The concentrations of all pesticides in sediments sampled during both field campaigns are all
below the 20 and 10 µg/kg detection limits used in 2003 and 2004 respectively. These detection
limits are, however, well above the equivalent TEC (0.6 to 2.85 µg/kg) and PEC values (1.3
44
Diagnostic study of water quality in the Lower Mekong Basin
to 1.5 µg/kg). As a result, it is not possible to determine if the sediments are contaminated by
pesticides.
Total hydrocarbons
Most of the sites sampled contained concentrations of total hydrocarbons below the detection
limit of 10 mg/kg. Figure 25 gives the concentrations of hydrocarbons for the sites with
measures above this limit.
80
70
60
n ) 50
t
r
a
t
i
o
/
kg 40
g
c
e
n
n (m 30
o
C
20
10
0
)
)
)
)
)
M
T
)
M
T
)
M
T
)
*
)
*
)
M
M
M
M
L
S
3 (
L
S
8 (
T
P
2 (
T
S
1
1
(
T
)
S
1
4 (
C
P
1
5 (
P
1
7 (
S
1
9 (
C
C
S
1
8 (
P
2
0 (
S
2
1 (
C
C
V
V
Mean
(M)
Mekong
(M*)
Bassac
(T)
Tributary
Figure 25. Total hydrocarbons concentrations recorded during the 2003 and 2004 campaigns
The highest concentration of total hydrocarbons was measured at Koh Khel (CS18) at 67 mg/kg.
Currently there are no ecotoxicity guidelines for total hydrocarbons in sediments. Therefore, it
is not possible to say if these levels are significant or otherwise.
Total cyanides
In 2004, sediment samples were analysed for total cyanides at four stations: the Lao/China
border (LS3), Luang Prabang (LS4), Ban Keng Done (LS8) and Kratie (CP15) to determine the
influence of mining activities upstream of these sites. Results were all below the detection limit
of 0.5 mg/kg.
There are no TEC values available for cyanides. Persaud et al. (1992), however, suggest that
concentrations below 0.1 mg/kg are acceptable; USEPA (1977) used a benchmark of 0.25 mg/kg
for classifying heavily polluted sediments. Unfortunately, the detection limit used in this study is
higher than both those values. The presence of cyanides at the levels of concern cannot therefore
be evaluated.
PAHs
Total PAHs concentrations were obtained by summing all the results of the different PAHs; `non
-detections' were assigned a zero value.

45
Diagnostic study of water quality in the Lower Mekong Basin
Total PAHs levels could be calculated for only three stations: Kratie (CP15), Prek Kdam (CP17)
and Koh Khel (CS18), with total PAHs concentrations of 190, 55 and 50 µg/kg respectively.
At Kratie, for example, naphthalene, phenanthrene and benzo(b)fluoranthene were measured
at levels of 80, 60 and 50 µg/kg respectively. Naphthalene and phenanthrene concentrations
exceed the interim sediment quality guidelines1 established by CCME (1999-2002).
PCBs
Seven toxic PCBs congeners were analysed in sediments. The concentrations of all were below
the detection limit of 10 µg/kg.
The TEL value for total PCBs developed by CCME (1999­2002) is 34.1 µg/kg. The seven
PCBs congeners are only a fraction (estimated at perhaps 20­25%) of the total PCBs, therefore
it was not possible to draw a conclusion on their eco-toxicological significance. However, as
all measured values are less than the detection limit, the risk of harmful effects from PCBs in
sediments is low.
BTEX
The BTEX analyses on all the sediments in both campaigns were below the detection limit
of 0.05 mg/kg. BTEX guidelines for sediments were not available. However, for the purpose
of comparison, in Canada (Canadian environmental quality guidelines) the maximum
allowable limit in agricultural soils is 0.05 mg/kg for benzene and 0.1 mg/Kg each for toluene,
ethylbenzene and xylene. Based on this evidence, we conclude that BTEX is not significant in
Mekong river sediments.
Bioassays
Results from the 1st campaign (2003)
H. azteca bioassays were performed on sediment samples from eight stations sampled
during 2003. One sample, from Neak Leang (CS19), was broken during transportation and
not analysed. The sediment sample from Kratie (CP15) was broken in transport and was not
analysed. Bioassays were performed in two series of tests, which were carried out at different
times (Table 11). The first series, included samples from Lao PDR, Thailand, and one of the
2003 samples from the Lao/China border. The second series, that was performed on samples
from Cambodia, Viet Nam and the other 2003 sample from the Lao/China border, had to be re-
assayed as the validity criteria in the standard protocol were not respected.
The survival rate of H. Azteca exposed to sediment samples from three stations was reduced
significantly. High mortality was associated with the sediments from the Lao/China Border
(LS3) (for which the bioassay was repeated since the mortality was exceptionally high) and to a
lesser extent from Tan Chau (VP20) and Kratie (CP15).
1 Threshold Effect Levels: Naphtalene: 34.6 µg/g; phenanthrene: 41.9 µg/g (CCME, 1999-2002).
46
Diagnostic study of water quality in the Lower Mekong Basin
A slight but not statistically significant increase, in the length of H. azteca was recorded at some
stations (TS11, CP15, VS21 and VP20), possibly because of greater amount of food (organic
carbon) naturally present in the field sediment than in the control sediments.
Results from the 2nd campaign (2004)
Sediment samples from 11 stations were collected for the Hyalella azteca bioassay. Two
samples (one upstream and one downstream stream) from the station at the China/Lao border
(LS3). These were analysed separately.
Some samples from both test series had significantly lower survival rates. The stations at the
Lao/China border (LS3) downstream, Luang Prabang (LS4), Koh Khel (CS18), Neak Leang
(CS19), Tan Chau (VP20) and Chau Doc (VS 21) had significantly lower survival rate than the
control (Table 11). Some mortality also occurred in the control samples in both series, however,
the mean survival rates of 90% and 78% are acceptable, as the minimum survival required by
the standard protocol is 70%.
In the 2004 series, a significant length increase was measured for samples from the following
sites: Koh Khel (CS18), Neak Leang (CS19) and Chau Doc (VS21). Chronic toxicity usually
leads to a growth decrease, however in this case, the organic content (i.e., food for Hyalella) is
probably greater in the sediments from the sites than in the artificial control sample.
Table 11. Survival rates (mean and standard deviation) of Hyalella azteca after 14 days exposure to
sediment samples from the 2003 and 2004 campaigns
2003
2004
Test series Code
Mean Survival
Standard
Mean Survival
Standard
(%)
deviation
(%)
deviation
Si1 (control)
96.0
5.5
Si1 (control)
90.0
7.1
LS3 (1) (Lao/China border)
30.0*
41.2
LS3 (upstream)
80.0
7.1
I
LS3 (downstream)
64.0
26.1
TP2 (Chiang Sean)
90.0
14.1
90.0
12.2
LS4 (Luang Prabang)
54.0
11.4
TS11 (Khong Chiam)
98.0
4.5
72.0
16.4
LP 12 (Pakse)
98.0
4.5
95.0
5.8
Si2 (control)
96.0
8.9
Si3 (control)
78.0
8.4
LS3 (2) (Lao/China border)
0.0*
CP15 (Kratie)
82.0*
16.4
II
CP17 (Prek Kdam)
77.5
15.0
CS18 (Koh Khel)
55.0
5.8
CS19 (Neak Leang)
45.0
12.9
VS21 (Chau Doc)
96.0
5.5
55.0
12.9
VP20 (Tan Chau)
78.0*
1.1
52.5
12.6
Note: * = Significant difference from the control (Dunnett's test, p < 0.05) in 2003.
= Significant difference from the control (Bonferroni t-test, p < 0.05) in 2004.

47
Diagnostic study of water quality in the Lower Mekong Basin
Summary of bioassay results
Of the seven sites where sediment toxicity was tested in both 2003 and 2004, five had similar
effects on the survival rate of H. azteca. Two sites had diverging results: i) one of the three
sediment samples from the Lao/China border (2004 upstream) did not show a significant effect
while the other two samples did; ii) samples from Chau Doc showed an effect in 2004, but not
in 2003.
The observed differences between the two years can not be attributed to the minor changes
in the methodology (i.e., the age of the organisms at the beginning of the bioassay), since
according to ASTM (2000), the sensitivity of H. azteca appears to be relatively similar up to at
least 24 to 26 days old. The range of ages at the end of the 14-day bioassays in 2003 and 2004
was 16 to 24 days. The differences are probably because the sediments at these stations were not
homogenous.
Overall, out of 18 tested sediments sampled at 12 different sites, 9 sediment samples were toxic
to H. azteca, inducing a significant decrease of its survival when compared to the control. The
seven stations and number of toxicity occurrences are:
· Lao/China border (LS3): two (2003, 2004 downstream), out of three samples;
· Luang Prabang (LS4): one sample (2004);
· Kratie (CP15): one sample (2003);
· Koh Khel (CS18): one sample (2004);
· Neak Leang (CS19): one sample (2004);
· Tan Chau (VP20): two samples (2003, 2004);
· Chau Doc (VS21): one (2004), out of two samples.
The results from sites LS3 and VS21 are not conclusive as one sample from each site showed no
toxicity response. The differences are likely due to spatial variations at the sampling sites--this
may also occur at other sites. Since only a few samples were analysed at each of the seven sites,
confirmation of the toxicity potential will require further bioassays. A further consideration is
that environmental bioassays are normally carried out as a `battery of tests' using at least three
trophic levels (e.g. three from bacteria, algae, invertebrates or fishes) as a single bioassay is not
equally responsive to all types of contaminants. The results are usually then pooled to derive a
composite ecotoxicological value (e.g. Costan et al., 1993).
A significant increase in H. azteca length also occurred in samples from three of these sites
(CP18, CP19, and VS21). All are located in, or downstream from, areas with high population
densities. The increase in length is probably explained by a higher nutritional value in sediments
that are enriched by organic matter from waste water and runoff.
48
6. Conclusions and Recommendations
Synthesis
The assessment of the 2003 and 2004 data was based on contaminants recorded at a limited
number of sites with a limited number of samples. This provides a useful, but preliminary,
picture of the quality of the river's water and sediments and the potential risks of contamination.
Moreover, some of the interpretation tools that were used (e.g., water classification thresholds,
contaminant guidelines, threshold effect and probable effect concentrations (TEC and PEC),
etc.) were developed in different contexts or for different purposes, and are not necessarily well
adapted to the Lower Mekong Basin.
For these reasons, and in order to synthesise the data in a more easily understood form, a multi-
criteria approach was employed. This gives a more integrated perspective of the toxicity in the
Mekong River. It will allow managers of water-resources to identify potential threats based on a
scoring approach an can be used to identify those sites that require more work to raise the level
of confidence through more robust statistical techniques. However, while this method provides a
valuable insight the results should be treated with some caution because the number of samples
used in the analysis was not consistent across all of the sites. Furthermore, the values represent a
single time in each sampling campaign and more work is require to establish if these values are
representative of longer periods of time or of a larger special area than the immediate environs
of the sampling sites.
Multi-criteria analysis
The multi-criteria analysis uses a `scoring system' in which rankings are assigned based on
actual or inferred levels of impacts on the environment.
Only those parameters for which quantifiable results were available were used in the analysis
(i.e., heavy metals and arsenic; pH, PCDD/PCDF, and Hyalella azteca bioassays). For each
parameter, a similar scoring method was used to that used for heavy metals (see page 31). This
allowed the derivation of an overall comparison of the stations (see Table 10 for the threshold
for scores: 0, 1 and 2).
Table 12 gives the scores for the four parameters--where data were available. The last
column to the right gives the standardised score (i.e., the total score divided by the number
of parameters where a score could be allocated), in order to take into account differences
in numbers of parameters at each station. For ease of examination and comparison, the
standardised scores are presented in Figure 26.
The results from the multi-criteria analysis show that the Neak Leang (CS19 in 2004) and Chau
Doc (VS21-upstream in 2004) are the most impacted stations. These stations are located near or
downstream from heavily populated areas.

49
Diagnostic study of water quality in the Lower Mekong Basin
Table 12. Multi-criteria analysis for the Mekong and main tributaries in 2003 and 2004
Station
Year
Metals and
PAHs
PCDDs &
Tox-HYA
Total
Number of
Standardised
arsenic
PCDFs
counts
score
LS3
2003
1
0
2
3
3
1.00
LS3 (up)
2004
1
0
1
2
3
0.67
LS3 (down)
2004
1
0
1
2
3
0.67
LS4
2004
1
0
1
2
3
0.67
LS5
2004
1
0
1
2
0.50
LS8
2003
0
0
1
0.00
LS8
2004
0
0
1
0.00
LP12
2003
0
0
0
2
0.00
LP12
2004
0
0
1
1
3
0.33
TP2
2003
0
0
0
2
0.00
TP2
2004
0
0
1
1
3
0.33
TS11
2003
0
1
1
2
0.50
TS11
2004
0
1
1
2
3
0.67
CS13
2004
0
0
1
0.00
CS14
2003
0
0
0
2
0.00
CS14
2004
1
1
2
2
1.00
CP15
2003
0
0
1
1
3
0.33
CP15 (up)
2004
1
0
1
2
3
0.67
CP15 (down)
2004
0
0
0
2
0.00
CP17
2003
1
1
1
1.00
CP17(up)
2004
2
0
1
1
4
4
1.00
CP17 (down)
2004
2
0
1
3
3
1.00
CS18
2003
0
0
1
0.00
CS18
2004
0
0
1
2
3
4
0.75
CS19
2003
0
0
1
0.00
CS19
2004
1
1
2
4
3
1.33
CS23
2004
0
0
0
2
0.00
VP2O
2003
0
1
1
2
0.50
VP2O (up)
2004
0
1
1
2
3
0.67
VP2O (down)
2004
1
1
2
2
1.00
VS21
2003
0
1
1
2
0.50
VS21 (up)
2004
1
1
2
4
3
1.33
VS21 (down)
2004
0
1
1
2
0.50
1.5
1.2
ers
0.9
amet
e/# of par
or 0.6
T
otal Sc
0.3
0.0
)
)
)
)
)
)
)
)
)
)
p
n
p
n
p
n
p
n
p
n
L
S
3
u
w
S
1
3
S
1
4
S
1
4
P
1
5
u
w
P
1
7
w
S
1
8
S
1
8
S
1
9
S
1
9
S
2
3
u
o
L
P
1
2
L
P
1
2
L
S
8
L
S
8
L
S
5
L
S
4
T
S
1
1
T
S
1
1
T
P
2
T
P
2
P
2
O
w
S
2
1
u
w
C
C
C
C
o
C
o
C
C
C
C
C
V
o
V
o
d
d
d
d
d
L
S
3 (
P
1
5 (
P
1
7
(
u
S
2
1 (
C
C
P
2
O (
V
V
L
S
3 (
P
1
5 (
P
1
7 (
S
2
1 (
P
2
O (
C
C
V
V
Station (2003 campaign)
Station (2004 campaign)
Figure 26. Multi-criteria analysis for the Mekong mainstream and main tributary stations
50
Diagnostic study of water quality in the Lower Mekong Basin
Prek Kdam (CP17) and the Lao/China Border (LS3) also have high scores. These stations
were identified at the end of the 2003 field campaign as the most polluted sites--CP17 for its
highest concentrations of total heavy metals and dioxins and furans; LS3 for the highest toxicity
observed in the bioassays. However, further investigation is required because of the graet
variability of the reults.
This classification is based on only two field campaigns, with varying numbers of parameters
measured at each station. Some sites, in a given year, have only been analysed/scored for one
type of parameter, while only two stations were analysed/scored for all four types of parameters.
Main recommendations
Benchmark sites
One of the major requirements of this study was to provide data with which to asses the extent
of pollution in the Mekong River originating from pesticides and industrial pollutants. This
study shows that the Lower Mekong Basin is relatively unpolluted with industrial organic
pollutants and metals. However, the data for pesticides are inconclusive because of poor
detection limits are available at this time.
Some `hot-spots' identified in this study appear to link to local sources of pollution. However,
it is unknown whether elevated levels of pollutants at sites downstream from Phnom Penh
are local factors or the result of trans-boundary transport. The study also demonstrated the
important role the Mekong River system plays in diluting pollution, such as salinity.
Based on these results, benchmark stations should be located at the six following sites:
· Lao/China border (LS3) on the Mekong River;
· Vientiane (LS5) on the Mekong River;
· Prek Kdam (CP17) on the Tonle Sap;
· Neak Leang (CS19) on the Mekong River;
· Tan Chau (VP2) on the Mekong River; and
· Chau Doc (VS21) on the Bassac.
WQMN programme
This study has provided information about which parameters are most applicable to assess the
status of the Mekong River regarding pesticides and industrial pollutants:
· Sediments: proved to be very useful and practicable, however, some of the analysed
parameters gave inconclusive results. Nevertheless, the following parameters are
particularly useful:

51
Diagnostic study of water quality in the Lower Mekong Basin
- Heavy metals;
- PAHs;
- Organochlorine, organophosphate and triazine pesticides with a detection limit lower
than 10 µg/kg;
- Dioxins and furans (e.g., by using the UNEP `toolkit' for PCDD/PCDFs would also
allow data comparison within the region).
Moreover, regarding future monitoring programmes, if sediment pollution is to be
correctly assessed, replicates, grain-size distribution and total organic carbon analyses
have to be included for each site in order to be able to properly compare the results at
individual sites and between sites. These parameters need not need be analysed every
month, once a year would be enough.
· Bioassays also proved to be useful indicators of environmental toxicology. In future
two types of bioassay tools should be considered. At sites such as the China/Lao border,
specialised test procedures such as TIE1 (Toxicity Identification Evaluation) can be used
to determine what specific chemistry is causing toxic effects. When this is known the
source can usually be identified.
A second approach is the `battery of tests' approach; this is more commonly used for
environmental effects assessment. This will require selecting appropriate species that
represent different trophic levels of the Mekong system.
Although not reported here, the use of diatoms has potential for assessing ecological
health. The diatom index can be developed for organic pollution and abnormal forms/
diversity used for toxic pollution. This needs development work for the Mekong.
Finally, analysis of biotic substrates (tissue, bile, liver, etc.) that were planned in this
study but not implemented for logistical reasons. Representative species should also be
included to assess the impact of toxic pollutants on the food chain and the potential risk
for human health. This could usefully include sentinel species that accumulate organic
and inorganic pollutants.
· Water quality data held by MRC appears to be very similar to that collected in this study.
This suggests that the MRC database is adequate for interpretation of routine water
parameters. However, analysis of the MRC data and the experience with the regional
laboratory contracted to do nutrients and other basic parameters, suggests that greater
effort needs to be directed to quality assurance and quality control to ensure that MRC
data reach and maintain a high level of reliability.
· Regional analytical capacity: The capacity of regional laboratories limits the use of some
types of advanced analyses. However, MRC should undertake an evaluation of local
1 See, for example, Birkholz et al. 2002.
52
Diagnostic study of water quality in the Lower Mekong Basin
capacities to determine which types of contaminant analyses can be performed reliably
within the region.

53
Diagnostic study of water quality in the Lower Mekong Basin
54
7. References
AFNOR (2002) Projet de norme - Qualité de l'eau ­ Détermination de la toxicité des sédiments
d'eau vis à vis de Hyalella azteca ­ partie 1 : sédiments naturels. Document T/ 95E N348 du
29/10/2002
Agusa, T., Kunito, T., Iwata, H., Monirith, I., Tana, T.S., and Subramanian, A., and Tanabe, S.,
(2005) Mercury contamination in human hair and fish from Cambodia: levels, specific
accumulation and risk assessment. Environmental Pollution 134, 79­86.
ASTM (2000) Test method for measuring the toxicity of sediment-associated contaminants with
freshwater invertebrates ­ ASTM report E1706-00, Harbor Drive, USA, 117p.
Babut, M., Perrodin, Y., Bedell, J.P., Clément, B., Cosnier, S., Corriger, B., Delmas, H.,
Delolme, C., Devaux, A., Miège, C., Péry, A., Roulier, J.-L. and B. Vollat (2004)
Méthodologie d'évaluation écotoxicologique de matériaux de dragage: tests de la démarche
et essais d'optimisation. CETMEF, VNF, 90p.
Birkholz, D.A., Headley, J.V., Ongley, E.D., Goudey, S. and I. Johnson (2002) Toxicity
assessment and identification for protection of natural waters. Canadian Water Resources
Journal 27(1), 85-103.
BURGÉAP (2005) Transboundary and Basin-wide water quality issues in the Lower Mekong
River Basin. Final Report by BURGÉAP, consultants to the Mekong River Commission.
CCME (1999-2002) Canadian Environmental Quality Guidelines for the Protection of Aquatic
Life. Canadian Council of Ministers of the Environment.
CERN (China Educational and Research Network) (2002) Work starts on Lancang River Power
Station. http://www.edu.cn/20020125/3018726.htm
CIIS (China Internet Information Services) (2002) No Pollution to Mekong River. http://us.tom.
com/english/543.htm
Collyard, S.A, Ankley, G.T, Hoke, R.A. and T Goldenstein. (1994) Influence of age on the
relative sensitivity of Hyallela azteca to diazinon, alkylphenol ethoxylates, copper, cadmium
and zinc. Archives of environmental Contamination and Toxicology 26: 110-113.
Costan, et al., (1993). Potential Ecotoxic Effects Probe (PEEP): A Novel Index to Assess and
Compare the Toxic Potential of Industrial Effluents. Environmental Toxicology and Water
Quality 8, 115-140.
Dai, A. and K.E. Trenberth (2002) Estimates of Freshwater Discharge from Continents:
Latitudinal and Seasonal Variations. Journal of Hydrometeorology 3(6), 660 ­687.
Gulley et al. (1989) TOXSTAT 3.0 Statistical software package instruction manual.

55
Diagnostic study of water quality in the Lower Mekong Basin
Hart, B.T., Jones, M.J., and G. Pistone (2001) Transboundary Water Quality Issues in the
Mekong River Basin. Final Report (unpublished), Mekong River Commission.
MacDonald, D.D., Ingersoll, C.G. and T.A. Berger (2000) Development and evaluation
of consensus-based sediment quality guidelines for freshwater ecosystems. Archives of
Environmental Contamination and Toxicology 39, 20-31.
MRC (1997) Mekong River Basin Diagnostic Study. Mekong River Commission and United
Nations Environment Programme, Bangkok.
MRC (2003) State of the Basin Report. Mekong River Commission, Phnom Penh.
MRC (2005) Overview of the Hydrology of the Mekong Basin. Mekong River Commission,
Vientiane.
Ongley, E.D., Birkholz, D.A., Carey, J. and M.R. Samoiloff (1988) Is water a relevant sampling
medium for toxic chemicals?: An alternative toxic chemical sensing strategy. J. Env. Quality
17, 391-401.
Persaud, D., Jaagumaji, R., and A. Hayton (1992) Guidelines for the protection and
management of aquatic sediment quality in Ontario. Ontario Ministry of the Environment,
Water Resources Branch, Toronto.
Safe, S. and D. Phil, (1990) Polychlorinated biphenyls (PCBs) dibenzo-p-Dioxins (PCDDs),
dibenzofurans (PCDFs), and related compounds: environmental and mechanistic
considerations which support the development of toxic equivalency factors (TEFs). Critical
Reviews in Toxicology 21, 51-87.
SEQ-Eau (1999) Système d'évaluation de la qualité de l'eau des cours d'eau - Rapport de
présentation ; Les Études des Agences de l'eau N°64, January 1999.
TEAM (2001) Water quality diagnostic of the Thai portion of the Mekong River basin. Final
Report by TEAM Consultants to the Mekong River Commission.
USEPA (1977) Guidelines for the Classification of Great Lakes Harbour Sediment. USEPA,
Region V, Chicago.
USEPA (2000) Methods for measuring the toxicity and bioaccumulation of sediment-associated
contaminants with freshwater invertebrates 2nd edn. Office of research and development
Mid-continent ecology division USEPA, Duluth.
Whiteman, F.W., Ankley, G.T., Khal, M.D., Rau, D.M., and M.D. Balcer (1966) Evaluation
of interstitial water as a route of exposure for ammonia in sediment tests with benthic
macroinvertebrates. Environmental Toxicology and Chemistry 15:794-801.
World Bank (2004) Modelling Observations on Development Scenarios in the Lower Mekong
Basin. World Bank Report. Vientiane.
Xinhua New Agency (2005) Water Pollution Worsening: SEPA. http://www.china.org.cn.
56
Diagnostic study of water quality in the Lower Mekong Basin

57
For further information please contact
Mekong River Commission
P.O. Box 6101, Vientiane 01000, Lao PDR.
Telephone: (856) (21) 263 263 Facsimile: (856) (21) 263 264
Email: mrcs@mrcmekong.org
Website: www.mrcmekong.org

Document Outline