Water balance analysis for the Tonle Sap Lake–floodplain system

Abstract

The Tonle Sap Lake of Cambodia is the largest freshwater body of Southeast Asia, forming an important part of the Mekong River system. The lake has an extremely productive ecosystem and operates as a natural floodwater reservoir for the lower Mekong Basin, offering flood protection and assuring the dry season flow to the Mekong Delta. In light of the accelerating pace of water resources development within the Mekong Basin and the anticipation of potentially significant hydrological impacts, it is critical to understand the overall hydrologic regime of Tonle Sap Lake. We present here a detailed water balance model based on observed data of discharges from the lake's tributaries, discharge between Mekong and the lake through the Tonle Sap River, precipitation, and evaporation. The overland flow between the Mekong and lake was modelled with the EIA 3D hydrodynamic model. We found that majority (53.5%) of the water originates from the Mekong mainstream, but the lake's tributaries also play an important role contributing 34% of the annual flow, while 12.5% is derived from precipitation. The water level in the lake is mainly controlled by the water level in the Mekong mainstream. The Tonle Sap system is hence very vulnerable, from a water quantity point of view, to possible changes in the Mekong mainstream and thus, development activities in the whole Mekong basin. From a biogeochemical point of view, the possible changes in the lake's own catchment are equally important, together with the changes in the whole Mekong Basin. Based on our findings, we recommend of continuing the monitoring programmes in lake's tributaries and urgently starting of groundwater measurement campaign within the floodplain, and including the groundwater modelling to be part of the hydrodynamic models applied for the lake.

Full text

Water balance analysis for the Tonle Sap Lake–floodplain system
M. Kummu,
1
*S. Tes,
2
S. Yin,
2
P. Adamson,
3
J. Józsa,
4,†
J. Koponen,
5
J. Richey
6
and J. Sarkkula
7
1
Water and Development Research Group, Aalto University, Finland
2
Department of Hydrology and River Works, Ministry of Water Resources and Meteorology, Phnom Penh, Cambodia
3
School of Mathematical SciencesAdelaide University, South, Australia
4
Department of Hydraulic and Water Resources Engineering, Budapest University of Technology and Economics, Hungary
5
Environmental Impact Assessment Centre of Finland, EIA Ltd., Espoo, Finland
6
School of Oceanography, University of Washington, Seattle, WA, 98195, USA
7
SYKE, Finland Environment Institute, Helsinki, Finland
Abstract:
The Tonle Sap Lake of Cambodia is the largest freshwater body of Southeast Asia, forming an important part of the Mekong
River system. The lake has an extremely productive ecosystem and operates as a natural floodwater reservoir for the lower
Mekong Basin, offering flood protection and assuring the dry season flow to the Mekong Delta. In light of the accelerating pace
of water resources development within the Mekong Basin and the anticipation of potentially significant hydrological impacts, it
is critical to understand the overall hydrologic regime of Tonle Sap Lake. We present here a detailed water balance model based
on observed data of discharges from the lake’s tributaries, discharge between Mekong and the lake through the Tonle Sap River,
precipitation, and evaporation. The overland flow between the Mekong and lake was modelled with the EIA 3D hydrodynamic
model. We found that majority (53.5%) of the water originates from the Mekong mainstream, but the lake’s tributaries also play
an important role contributing 34% of the annual flow, while 12.5% is derived from precipitation. The water level in the lake is
mainly controlled by the water level in the Mekong mainstream. The Tonle Sap system is hence very vulnerable, from a water
quantity point of view, to possible changes in the Mekong mainstream and thus, development activities in the whole Mekong
basin. From a biogeochemical point of view, the possible changes in the lake’s own catchment are equally important, together
with the changes in the whole Mekong Basin. Based on our findings, we recommend of continuing the monitoring programmes
in lake’s tributaries and urgently starting of groundwater measurement campaign within the floodplain, and including the
groundwater modelling to be part of the hydrodynamic models applied for the lake. Copyright © 2013 John Wiley & Sons, Ltd.
Supporting information may be found in the online version of this article.
KEY WORDS flood pulse; water balance model; floodplain; hydrology; Tonle Sap Lake; Mekong
Received 11 June 2012; Accepted 11 January 2013
INTRODUCTION
Floodplains are one of the most important parts of aquatic
ecosystems. They provide important services for human-
kind, as they are rich in biodiversity and highly productive
in terms of fishes and other aquatic animals (Finlayson and
Spiers, 1999; Millennium Ecosystem Assessment, 2005).
Maintenance of the hydrological regime of a floodplain and
its natural variability is necessary to sustain its ecological
and biodiversity characteristics (Millennium Ecosystem
Assessment, 2005).
The most extensive wetland habitats of the Mekong River
BasinarelocatedintheheartofCambodia,theTonleSap
Lake (Figure 1). The lake and its surrounding floodplains
form the largest freshwater body in Southeast Asia. The lake
is reported to be very productive (Rainboth, 1996; Sverdrup-
Jensen, 2002), driven by an annual mono-modal flood pulse
(Junk et al., 2006; Lamberts, 2006). The lake functions as a
natural floodwater reservoir for the Mekong system during
the dry season (November–April), when approximately half
of the discharge to the Mekong Delta in Vietnam originates
from the lake (Fuji et al., 2003). The ecosystem is driven by a
flood pulse regime, supporting a fishery and aquaculture that
provides approximately up to 80% of the protein consump-
tion of Cambodia (e.g. Ahmed et al., 1998; Hortle, 2007).
Although there are over 70 million people living in the
Mekong Basin (Pech and Sunada, 2008) and more than 30
existing large dams (Mekong River Commission, 2009), the
Mekong is still considered to be one of the few large river
basins in which the flow regime has remained in rather natural
conditions (Mekong River Commission, 2005b). The
Mekong is, however, facing rapid development, including
deforestation (FAO, 2006; Shi, 2008), large irrigation
schemes (Hori, 2000; Kummu et al., 2009), and construction
of hydropower dams with large reservoirs (e.g. Hori, 2000;
Dore and Yu, 2004; King et al., 2007; Mekong River
Commission, 2008; ICEM, 2010; Kummu et al., 2010;
Grumbine and Xu, 2011).
Recent hydrological impact assessment studies
(Adamson, 2001; ADB, 2004; World Bank, 2004; ICEM,
2010; Mekong River Commission, 2010; Lauri et al.,
2012; Räsänen et al., 2012) conclude that due to the
planned development, the dry season water levels would
rise, and wet season water levels would become lower,
relative to current conditions. The magnitude of predicted
*Correspondence to: M. Kummu, Water and Development Research
Group, Aalto University, Finland.
E-mail: matti.kummu@iki.fi
†
Present address: MTA-BME Water Management Research Group,
Hungary
HYDROLOGICAL PROCESSES
Hydrol. Process. 28, 1722–1733 (2014)
Published online 14 February 2013 in Wiley Online Library
(wileyonlinelibrary.com) DOI: 10.1002/hyp.9718
Copyright © 2013 John Wiley & Sons, Ltd.

change differs, however, between the studies (Johnston
and Kummu, 2012; Lauri et al., 2012). The predicted
climate change adds an extra uncertainty on the future
simulations (Keskinen et al., 2010; Kingston et al., 2011;
Lauri et al., 2012), as the General Circulation Models
disagree with the climate change impacts on Southeast
Asian monsoon regime (Ashfaq et al., 2009). The flow
alterations in the Mekong mainstream would directly
impact the flood pulse of Tonle Sap Lake (Kummu and
Sarkkula, 2008), because the water level in the lake is
controlled by the water level in the Mekong mainstream
(Kummu et al., 2008). The resulting lower flood peak and
higher dry season water levels would have direct impacts
on the lake’s ecosystem productivity: the smaller the area
that becomes flooded, the smaller the area between
‘aquatic/terrestrial transition zones’, and thus the potential
transfer of floodplain terrestrial organic matter and energy
into the aquatic phase would be also smaller (Junk, 1997;
Lamberts, 2008; Lamberts and Koponen, 2008).
The knowledge of the hydrology of the Tonle Sap
system and its hydrodynamic relationships with the
Mekong mainstream has increased rapidly during the last
years (Fuji et al., 2003; MRCS/WUP-FIN, 2003; MRCS/
WUP-FIN, 2007; Inomata and Fukami, 2008; Kummu
and Sarkkula, 2008). There are still, however, information
gaps. First, sufficient data have not been available for all
the water balance components of the lake, most
particularly discharge and volumetric flow data –largely
the result of the complexity of the regional relationships
of water level, discharge, and flow direction. The
previous studies (Fuji et al., 2003; Inomata and Fukami,
2008) have analysed only part of the water balance
components impacting on lake’s hydrology and thus, the
whole picture of the lake water balance is still missing.
Second, only rather recently have hydrodynamic models
of the system been developed with an order of accuracy
and detail sufficient to provide the required insights on,
for example, the overland flow from the Mekong
mainstream to the lake’sfloodplain (e.g. Fuji et al.,
2003; MRCS/WUP-FIN, 2003; MRCS/WUP-FIN, 2007;
Hai et al., 2008; Västilä et al., 2010).
In light of the accelerating pace of water resources
development within the Mekong Basin and the anticipation
of potentially significant hydrological impacts, it is critical to
understand the overall hydrologic regime of Tonle Sap Lake.
The principal objective of this study is to provide a detailed
estimate of the water balance of the Tonle Sap system. We
focus on two research questions motivated by the existing
information gaps of understanding the lake’s hydrology:
a) What is the timing and relative proportion of water
supplied to Tonle Sap Lake during rising water, by the
Mekong River, via both the Tonle Sap River and
overland flooding, relative to tributaries directly to the
lake, and direct precipitation?
b) What is the relative percent of water supplied during
falling water to the Delta, from the draining Tonle Sap
Lake relative to the Mekong mainstream? This is not only
an important hydrologic question, but a biogeochemical
one, as the water chemistry of the two sources differs
dramatically (Ellis et al., 2012).
TONLE SAP LAKE
The Tonle Sap Lake is a sub-catchment of the Mekong,
the largest river in Southeast Asia. The catchment is
located principally in Cambodia, with 5% in Thailand
(Figure 1). The total area of the Tonle Sap catchment is
Sen
Sreng
Sisophon
Chinit
Baribor
Mongol Borey
Pursat
Dauntri
Siem Reap
Staung
Chikreng
Sangker
PHNOM PENH
Kralanh
Sisophon
Mongol Borey
Battambang
Maung
Russey
Bac
Trakoun
Baribor Kg. Thmar
Kg. Thom
Kg. Chen
Kg. Kdei
Tonle Sap
Lake
Mekong
Bassac
Mekong
Tonle Sap River
Kratie
Untac
Bridge
100
km
Thailand
Cambodia
Kg. Luong
Bac Prea
Prek Kdam
1a
1b
1c
23
4
5
6
Prec. and evap.
WL (Lake, TS River)
WL, Q (Tonle Sap basin)
Permament water
Floodplain
Country border
Figure 1. Tonle Sap Basin and its sub-catchments with the main rivers (white lines are catchment borders and blue ones are rivers), and floodplain,
including the locations of the water balance calculation elements, and hydrological and precipitation measurement sites. Water balance calculation
elements:1a–1c: Water level measurements; 2: Tonle Sap River discharge at Prek Kdam; 3: Overland flow; 4: Flow from tributaries; 5: Precipitation to
open water; and 6: Evaporation from open water
1723TONLE SAP WATER BALANCE
Copyright © 2013 John Wiley & Sons, Ltd. Hydrol. Process. 28, 1722–1733 (2014)

85,790 km
2
being approximately 11% of the total area
of the Mekong basin (Mekong River Commission, 2003).
The catchment size includes the permanent lake area of
around 2400 km
2
, being the area of the lake–floodplain
system during driest month with water level of 1.44 m
(Kummu and Sarkkula, 2008). The average maximum
floodplain size is 10,800 km
2
(excluding the permanent lake
area) with water level of 9.09 m (Kummu and Sarkkula,
2008). The interannual variation of the floodplain size is,
however, rather large ranging from 7190 km
2
to 12,720 km
2
(within the time period of 1997–2005). The volume of
the system varies from 1.8 km
3
during the driest month to
58.3 km
3
(with variation of 31.1 –73.9 km
3
)duringthepeak
water level (Kummu and Sarkkula, 2008).
The floodplain can be divided into five habitat groups
(after Arias et al., 2012): (1) Open water (under water 12
months a year); (2) Gallery forest (annual flood duration
of 9 months); (3) Seasonally flooded habitats, dominated
by shrublands and grasslands (flooded 5–8 months); (4)
transitional habitats, dominated by abandoned agricultural
fields, receding rice/floating rice, and lowland grasslands
(flooded 1–5 months); and (5) Rainfed habitats, consist-
ing mainly of wet season rice fields and village crops
(flood duration less than 1 month).
The hydrology of the Tonle Sap Lake is driven by the
monsoonal flood regime of the Mekong. The Tonle Sap
River, which flows from the southeastern end of Tonle
Sap Lake (Figure 1), joins the Mekong River at the
Chaktomuk confluence, in the vicinity of Phnom Penh.
After the confluence, the river immediately splits into the
smaller Bassac River and the larger Mekong River
(Figure 1). In the wet season, from May to September,
flooding and the associated water level increase in the
Mekong River causes the Tonle Sap River to change flow
direction and flow towards the northwest (upstream) into
Tonle Sap Lake. The 11 principal tributaries that drain
directly into the lake provide another important source of
incoming water (Figure 1). Floodwater enters into the
lake via overland flow across the floodplain.
ANALYTICAL FRAMEWORK
To examine the relative magnitude and timing of each of
the flows entering and leaving the lake, we developed a
water balance model for the lake–floodplain system. The
components of the water balance model for this lake –
floodplain system include the (1) Tonle Sap River –
Mekong influence, (2) Tonle Sap Lake’s own tributaries,
(3) Rainfall, (4) Evaporation, and (5) Overland flow via
floodplain linked together, as follows:
dV=dt ¼QTSR þQTRIB þQOVR þQPREC QEVAP (1)
Where
Vis lake volume derived from a digital bathymetric
model (DBM) and lake water level (WL
KL
)
Q
TSR
is inflow to (+) and outflow from () the lake via
the Tonle Sap River as observed at Prek Kdam,
Q
TRIB
is inflow from the 11 main tributaries of the lake
Q
OVR
is overland flow from the Mekong mainstream
to the Tonle Sap system via the floodplain,
Q
PREC
is rainfall on the lake itself, observed at two
sites close to the lake (Figure 1), and
Q
EVAP
is evaporation data, observed at two sites
(Figure 1).
The water balance was computed at a daily time step
over the eight years, for the period May/1997 –April/2005.
While relatively short, this sequence may be considered to
be representative of the longer term range of hydrological
conditions. The record covers the extreme drought flow of
year 1998, the year 2000 high flood episode, and various
intermediate flood years. The analysis is based on a
‘hydrological year’(from 1
st
of May to 30
th
of April next
year) rather than a calendar year, defined for the lake by
Kummu and Sarkkula (2008).
It is important to note that our model does not include
the interaction of groundwater and floodplain due to the
absence of detail data of these processes from the
floodplain. There exist various groundwater studies in
Cambodia, as well summarized by Landon (2011), but
these are mostly located outside the floodplains with only
few exceptions (Kazama et al., 2007; May et al., 2011;
Burnett et al., 2012). Moreover, our hydrodynamic model
applied to the lake does not have a groundwater component
that could be coupled with the flood modelling, and thus
enable simulation of flood –groundwater interconnections.
Kazama et al. (2007) and May et al. (2011) found that
flooding around the Mekong mainstem in Cambodia
(Southeast from our study area) plays an important role
in recharging groundwater. The findings of Burnett et al.
(2012), who studied the groundwater discharge along the
northern section of the Tonle Sap Lake based on radon
analysis from the surface waters and groundwater wells,
support these findings. Burnett et al. (2012) estimated that
groundwater discharge to the lake during the recession
phase of year 2009 was around 4–8km
3
, or about 13–27%
of average tributary inflows and 5–10% of average total
inflow (see Section on Results of the Water Balance
Analysis). They did not, however, estimate how much
water infiltrated from the floodplain to groundwater. Both
of these studies (May et al., 2011; Burnett et al., 2012)
highlight the importance of including the groundwater to
the water balance calculations, and thus this should be high
in priority in the future research.
DBM
A DBM of the lake and its floodplain was derived from
three spatial data sets (Table I; Figure S1 in the Online
Supplement): (1) a hydrographic survey (Mekong River
Commission, 1999) was used to compute the contours for
the dry season lake area and Tonle Sap River; (2) the
Certeza survey map (1964) was used for the Tonle Sap
Floodplain; and (3) the Shuttle Radar Topography
Mission data (Farr et al., 2007) was used for the
surrounding areas to complete the DBM. GIS datasets,
used for mapping and flood analyses, included river, lake
1724 M. KUMMU ET AL.
Copyright © 2013 John Wiley & Sons, Ltd. Hydrol. Process. 28, 1722–1733 (2014)

and inundated area layers, the Tonle Sap Basin sub-
catchments, and the locations of hydrological measure-
ment stations (Mekong River Commission, 2005a).
The DBM was used to calculate the daily lake area
[A,km
2
] and lake volume [V,km
3
] as a function of water level
in Kampong Luong [WL
KL
,m].TheDBMwasalsousedfor
the flooded area calculations for each catchment (see below).
Water level –Area –Volume relationship
The relationships between water level (WL), inundated
area (A), and volume (V) were calculated using the DBM
of the lake and its floodplain (see Figure S2 in the Online
Supplement). The open water area and volume of the lake
varies as a direct function of water level:
Akm
2

¼5:5701 WLKL3þ1:374 WLKL 2
þ470:29 WLKL þ1680:2;r2>:99
(2)
Vkm
3

¼0:7307 WLKL2þ0:3554 WLKL
þ0:9127;r2>:99
(3)
The average water level over the sampling period at
Kampong Luong varies from (daily minimum) 1.32 m
above the mean sea level (AMSL) to its peak flood of
9.17 m AMSL (Figure 2, Table II). At the same time, the
open water surface area varies from 2215 km
2
to 13,260
km
2
and the volume from 1.6 km
3
to 59.7 km
3
.
Tonle Sap River: inflow and outflow (Q
TSR
)
As a result of the complex hydraulic interrelationships
between river, lake, and floodplains, the rating curve for
Tonle Sap River at the Prek Kdam hydrometric gauging site
is particularly complicated. Separate rating curves were
required for the inflow and outflow phases of the annual
hydrological cycle (Forsius, personal communication, 2006).
Table I. List of data used for the water balance calculations. The location of the weather and water levels stations are presented in Figure 1. The bathymetry data is illustrated in Figure S1 of the
Online Supplements. Note: SRTM stands for Shuttle Radar Topography Mission;MoWRaM stands for Ministry of Water Resources and Meteorology
Data Year Source Form of data Notes
BATHYMETRY Lake proper 1999 Mekong River Commission (1999) Vector
Lake’sfloodplain 1963 Certeza Surveying (1964) Vector The survey results were validated during WUP-FIN project (MRCS/
WUP-FIN, 2003)
Surrounding areas 2003 SRTM data (Farr et al., 2007) Raster Original resolution 90m
CLIMATE Precipitation 1997–2005 MoWRaM of Cambodia Daily timeseries Averaged data from two weather stations: Siem Reap and Prek Kdam
Evaporation 1997–2005 MoWRaM of Cambodia Daily timeseries Data from Siem Reap weather station
WL&Q WL
PP
1997–2005 Mekong River Commission (2007) Daily timeseries
WL
PK
1997–2005 Mekong River Commission (2007) Daily timeseries
WL
KL
1997–2005 Mekong River Commission (2007) Daily timeseries
Tributary discharges 1997–2005 MoWRaM of Cambodia Daily timeseries Discharges are based on rating curves (see Table S2 in the Online
Supplements)
0
2
4
6
8
10
12
14
0
2
4
6
8
10
12
14
May
Jun
Jul
Aug
Sep
Oct
Nov
Dec
Jan
Feb
Mar
Apr
Water level [m], AMSL in Hatien
Rising flood Receding flood
area
water level
Area [103 km2]
Figure 2. Monthly average water level of Tonle Sap Lake at Kampong
Luong (continuous line –right vertical axis) and inundated area (circles –
left vertical axis). The observed minimum and maximum monthly average
water levels have been illustrated with dotted bars
1725TONLE SAP WATER BALANCE
Copyright © 2013 John Wiley & Sons, Ltd. Hydrol. Process. 28, 1722–1733 (2014)

Inflow to the Tonle Sap Lake takes place between early
May and mid-September, with relatively little interannual
variation. The inflow rating curve (Equation (5); Figure
S2 in the Online Supplement) was used when the water
level is lower in the lake (measured at the Kampong
Luong site, Figure 1) than in Phnom Penh (at the Phnom
Penh Port gauge). A total of 31 field discharge
measurements within the period of 1998–2006 were used
to generate the inflow phase curve. The rating curve is
based on the index area in Prek Kdam (calculated from
the water level: WL
PK
1.2
) and the water level difference
between Kampong Luong and Phnom Penh Port
hydrometric gauges (dH =WL
PP
–WL
KL
):
Fin ¼WLPK
ðÞ
1:2WLPP WLKL
jj
ðÞ
0:5(4)
Where WL is the water level of the lake, in meters
AMSL in Hatien, Vietnam), for Kampong Luong (WL
KL
),
Prek Kdam (WL
PK
), and Phnom Penh Port (WL
PP
).
The least squares functional approximation to the
inflow rating curve is:
QTSR;in ¼15:0467 Fin2þ859:839
Fin 782:264;r2¼:95
(5)
Outflow from the Tonle Sap Lake takes place between
mid-September and early May, again with moderate
interannual variation. The outflow rating curve (Equation
(7); Figure S2B in the Online Supplement) is used when the
lake water level is higher than at the water level in Phnom
Penh Port. In this case, 22 field discharge measurements
were available to generate the relationship; based on the
index area in Prek Kdam (WL
PK
1.2
) and the water level
difference between Kampong Luong and Phnom Penh Port
(dH =WL
PP
–WL
KL
), giving:
Fout ¼WLPK
ðÞ
1:2WLPP WLKL
jj
ðÞ
0:5(6)
The final outflow curve is the least squares approximation:
QTSR;out ¼8:784 Fout 2þ434:465
Fout þ167:151;r2¼:98
(7)
The discharge in Prek Kdam was then calculated
(Figure 3A) based on the rating curves presented
above. The peak inflow discharge was on average 6000
m3/s occurring in August, while the peak outflow
occurred in late October –early November with a
discharge of 8000 m3/s.
For the discharge calculations in the Chaktomuk
confluence, where the Tonle Sap River joins the Mekong
mainstem, we used 70 Acoustic Doubler Current Profiler
discharge measurements. These measurements were
conducted during the MRCS/TSLV (2004) and MRCS/
WUP-FIN (2007) projects between years 2002 and 2007.
We used the data to calculate the monthly average
discharges for Tonle Sap River and Mekong mainstem
discharge components and finally the share of Tonle Sap
river discharge entering to the Mekong Delta.
Tonle Sap Basin tributary inputs Q
TRIB
There are 11 primary tributaries draining into Tonle
Sap Lake (Figure 1). Due to the extensive seasonal
inundation of substantial areas of most of these catch-
ments, the area contributing to the runoff changes
seasonally (Figure 4). The total area of the Tonle Sap
catchments, discounting the area that remains inundated
during the dry season, was 83,010 km
2
from which the
flows of 48,680 km
2
or just under 60% are gauged. For
computational purposes, the basin can be divided into
gauged, ungauged, and inundated areas (see Figure 4).
For each tributary the discharge of the gauged areas
were calculated from the observed water levels using
rating curves developed for each station (see Table S2 in
the Online Supplement).
The ungauged part of the discharge for the i
th
catchment
was then calculated as:
Qitotal ¼Aitotal =Aigauged

Qi(8)
Where
Q
i_total
= total discharge of tributary i[m3/s]
A
i_total
= total catchment area [km
2
]
A
i_gauged
= observed catchment area [km
2
]
Q
i
= discharge of tributary i[m3/s]
This procedure for estimating the ungauged runoff is
supported by the fact that at the time when error would be
expected to be potentially high, during the flood season,
the ungauged areas that are not inundated are the smallest.
It is also prescribed by the lack of spatial rainfall data
coverage that would enable a detailed-enough modelling
procedure to be developed for the purpose. Part of the
each catchment area (Table S1 in the Online Supplement)
is inundated each year. Thus, the hydrological status of
the area shifts from terrain to lake that needs to be taken
into account in the analysis in a way that the flooded area
is not included in the tributary discharge calculations. A
rating curve has been developed to describe the inundated
area versus water level separately for each catchment (see
Table S3 in the Online Supplement) by using the DBM
Table II. Water level, area, and volume statistics of Tonle Sap
Lake for years 1997–2005. Years shown are hydrological years
(01/05 of year indicated –30/04 of next year)
Annual minimum Annual maximum
wl area volume wl area volume amplitude
Year mkm
2
km
3
mkm
2
km
3
m
1997 1.40 2307 1.7 9.33 13,540 61.5 7.9
1998 1.24 2120 1.4 6.86 9640 33.0 5.6
1999 1.38 2284 1.6 8.97 12,950 56.8 7.6
2000 1.48 2402 1.8 10.36 15,280 76.1 8.9
2001 1.42 2331 1.7 9.89 14,480 69.2 8.5
2002 1.19 2061 1.3 10.10 14,830 72.2 8.9
2003 1.27 2155 1.5 8.26 11,810 48.1 7.0
2004 1.25 2131 1.4 9.20 13,330 59.8 8.0
2005 1.26 2143 1.5 9.29 13,480 61.0 8.0
avg.1.32 2215 1.6 9.17 13,260 59.7 7.8
1726 M. KUMMU ET AL.
Copyright © 2013 John Wiley & Sons, Ltd. Hydrol. Process. 28, 1722–1733 (2014)

and spatial analysis tools of ArcGIS 9.2 software. The
example of Staung Catchment is presented in Figure 4,
where the flooded area is calculated for selected water levels.
The inundated areas at the peak of the year 2000 flood
event, when lake levels reached 10.34 m AMSL, are presented
in Table S1 of the Online Supplement. The Sangker catchment
has the largest flooded area of 1957 km
2
, being 32% of the
whole catchment. The total area of the lake’sfloodplain
inundation, in the catchment areas, during the year 2000 flood
was 12,140 km
2
(Table S1 in the Online Supplement). Adding
the dry season lake area of 2350 km
2
gives a total area of open
water of 14,500 km
2
at the peak of the event, this figure being
equivalent to over 8% of the national area of Cambodia.
The final daily discharge for each catchment, account-
ing for the flooded area was calculated on the basis of the
following equation
QiTRIB ¼Aitotal Aiflood

=Aitotal

Qitotal (9)
Where
Q
i_TRIB
=final discharge for tributary i[m
3
/s]
Q
i_total
= total discharge for tributary i[m
3
/s] from
Equation (4).
A
i_total
= total catchment area [km
2
]
A
i_flood
=flooded area of i
th
catchment [km
2
] based on
the daily water level and rating curves
The results are summarized as monthly average flow
volume (10
6
m
3
) in Figure 3B. The annual average total
volume of runoff from all Tonle Sap system tributaries is
24.5 km
3
of which the Sen Tributary contributes to the
largest portion with 32%. The Chinit, Sanker, Pursat, and
Baribor each provide 10% or more. Elsewhere, the
individual tributary flow contributions are a minor
proportion of the whole, with mean annual flow volumes
below 1 km
3
.
The averagerunoff for the Tonle Sap Basin is 310 mm/yr.
However, the annual average runoff varies greatly from 133
mm/yr in Sisophon to 486 mm/yr in Chinit. The annual
variation is also remarkable: the driest year (1998) and
wettest year (2000) had an average runoff over the Tonle
Sap catchment of 155 mm/yr and 515 mm/yr, respectively.
Approximately 81% of the total annual discharge occurs
during the wet season (May –October) as can be seen from
Figure 3B. The peak discharge takes place in October, when
in average 26.6% of the total annual discharge occurs.
Rainfall and evaporation
Daily measurements from two precipitation stations, at
Siem Reap and Prek Kdam (see locations in Figure 1),
were used for the water balance calculations. The data
were averaged between the two sites for the purpose of
this study.
-8000
-4000
0
4000
8000
B:A:
D:
C:
May
Jun
Jul
Aug
Sep
Oct
Nov
Dec
Jan
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
Oct
Nov
Dec
Jan
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
Oct
Nov
Dec
Jan
Feb
Mar
Apr
Average discharge at Prek Kdam [m3/s]
0
5
10
Average WLKL [m], AMSL
QTSR WLKL
0
200
400
600
800
May
Jun
Jul
Aug
Sep
Oct
Nov
Dec
Jan
Feb
Mar
Apr
Average monthly tributary discharge [ m3/s]
Average monthly overland flow [m3/s]
Chinit
Sen
Staung
Chikreng
Sreng
Sisophon
Mongol
Sanker
Pursat
Boribo
Dauntri
-800
-400
0
400
0
50
100
150
200
250
0
50
100
150
200
250
Evaporation (mm/month)Precipitation (mm/month)
Prec
Evap
Figure 3. Measured values for water balance components: A. Tonle Sap river daily average discharge at Prek Kdam plotted with the water level in the
lake (1997–2004) at Kampong Luong (WL
KL
). B. Average monthly discharge of tributaries in 10
6
m
3
. C. Monthly average precipitation and evaporation,
average of Siem Reap (SR) and Prek Kdam (PK) stations for precipitation and SR station for evaporation. D. Overland flow. Note: the scales of y-axes
vary between the figures.Positive discharge values correspond to inflow into the lake and negative outflow from the lake
1727TONLE SAP WATER BALANCE
Copyright © 2013 John Wiley & Sons, Ltd. Hydrol. Process. 28, 1722–1733 (2014)

The evaporation data used to estimate the open water
evaporation of the lake and floodplain was based on the
daily pan evaporation measurements from a weather
station at Siem Reap. Daily pan evaporation measure-
ments were available for three years (1998–2000). For the
years 2001–2005, evaporation was estimated from
measured daily minimum temperature, daily maximum
temperature, and relative humidity, with a multivariate
correlation between these three variables and measured
pan evaporation for years 1998–2000 (p<0.000). The
correlation coefficients were then used to calculate the
evaporation for years 2001–2005. For 1997, no measure-
ments were available, and thus the daily average values
from the measured three years of data were used.
Daily rainfall over the lake area and evaporation loss
from it were simply obtained as a function of lake’s area
at a given day:
QPREC ¼precipitation A(10)
QEVAP ¼evaporation A(11)
The regional SW monsoon seasons –May to October
(wet) and November to April (dry) –mean that almost 90%
of annual rainfall is confined to 6 months of the year
(Figure 3C). While the differences in mean annual rainfall
between Siem Reap and Prek Kdam are considerable, the
working assumption, however, is that a 1290 mm/yr
regional figure is an acceptable estimate of Tonle Sap Lake
mean annual rainfall, a figure broadly in line with those
mapped elsewhere (Mekong River Commission, 2005b).
The monthly average daily evaporation (based on the
measured values over the time period of 1998–2000) varied
from 3.5 mm/day in October to 5.8 mm/day in March. The
total evaporation was approximately 1627 mm/yr.
Overland flow
Overland flow from the Mekong via the floodplains is
an important element in the Tonle Sap water balance.
Further, it is emerging as potentially the most immediate
susceptible component of human influence, primarily in
response to national road construction and road improve-
ment (e.g. Fuji et al., 2003; Mekong River Commission,
2005c). These changes are affected principally by road
and other embankments within the floodplain, which
either redirect overland flow or restrict it all together by
acting as low elevation flood bunds, which have a very
effective hydraulic impact in the flat landscape.
The quantities involved in the overland flow phase are
difficult to monitor directly. In order to estimate them for
the purpose of a robust assessment of the system water
balance, the three-dimensional (3D) EIA 3D hydro-
dynamic model was used (Koponen et al., 2005; Kummu
et al., 2006; MRCS/WUP-FIN, 2008). This model is a 3D
baroclinic multilayer model that numerically solves the
simplified Navier–Stokes equations using the implicit
finite difference method (Koponen et al., 2005). The EIA
3D model has been set up for the Lower Mekong
floodplains, including the Tonle Sap system, during the
WUP-FIN project (MRCS/WUP-FIN, 2008). The model
covers an area of 430 km * 570 km, extending from
Kratie down to the South China Sea. The grid resolution
of the model is 1 km * 1 km, and the flow is simulated in
ten vertical layers.
The EIA 3D model incorporates data on topography,
bathymetry, and land use. The model takes into account
the natural stream network of the Mekong and the large
man-made channels. The model determines flow resist-
ance based on two friction parameters: bottom friction,
which is uniform across the whole modelling area, and
vegetation drag that differs according to the land use
class.
The boundary conditions of the model include daily
discharges of the Tonle Sap tributaries, the Mekong
discharge at Kratie, and hourly sea levels at the South
China Sea. The EIA 3D model does not consider
hydrological processes in the modelling area. Overland
flow was estimated for both the inflow and outflow
phases. The model was separately run for each simulated
hydrological year. The simulations were run with time
steps of 45–2000 s depending on the simulated process.
The model application for the Lower Mekong floodplains
was calibrated by Västilä et al. (2010) with satisfactory
results. The annual average overland flow into the lake was
estimated to be 2.6 km
3
during the study period while the
average overland flow from the lake was approximately
similar magnitude, being 2.7 km
3
(Figure 3D). The inflow
overland flow results are rather well in line with the other
studies (e.g. Fuji et al., 2003; Inomata and Fukami, 2008)
while our outflow overland flows are slightly larger than
their estimates.
10.64
Ai_gauged
Ai_flood
Ai_ungauged
Ai_total
9.30
8.00
6.64
5.64
4.00
2.00
3.00
20
km
Staung catchment
Floodplain at selected levels [m] amsl
Observed part of catchment
Non-observed part of catchment
Figure 4. Definitions of the active catchment area calculation components
and flooded area for selected water levels in Staung. See location of the
Staung catchment in Figure 1
1728 M. KUMMU ET AL.
Copyright © 2013 John Wiley & Sons, Ltd. Hydrol. Process. 28, 1722–1733 (2014)

RESULTS OF THE WATER BALANCE ANALYSIS
The components of the water balance model for this lake
–floodplain system include the (1) Tonle Sap River –
Mekong influence, (2) Tonle Sap Lake’s own tributaries,
(3) Rainfall, (4) Evaporation, and (5) Overland flow via
floodplain (Figure 5), as explained in details in previous
section. Over the eight-year simulation period, the model
replicated the observed data very well (Figure 6), R
2
(observed lake volume versus modelled lake volume)
being 0.94. There were, however, seven sub-periods
when the water balance model performed less well
(circled in Figure 6B), particularly for the lower water
levels occurring towards the end of the dry season during
late April and early May.
Breaking the overall water balance into its components
shows the relative magnitude of each term, summarized
for analysed years (Table III; Figure 7A) and monthly
average water balance results (Figure 7B). On average,
53.5% of Tonle Sap Lake volume originates from the
Mekong mainstream, either via the Tonle Sap River
(50%) or overland flow (3%) (Table III). The system’s
own tributaries contribute 34%, and the balance of 12.5%
is sourced from precipitation. Interannual variation is
rather large: inflow during the eight-year study period
ranged from 51 km
3
during the dry conditions of 1998 to
109 km
3
during the 2000 high flood episode (Figure 7A).
The estimated eight-year average inflow was 83.1 km
3
(Table III).
Around 88.5% of the total annual outflow from the lake
is discharged into the Mekong mainstream via the Tonle
Sap River (84%) while the overland drainage back to the
Mekong constitutes a fraction of 3% (Figure 7A). The
open water evaporative losses are assessed at 13%. The
mean annual outflow during the study period was 81.9
km
3
, with annual volumes varying between 46 km
3
(1998) and 111 km
3
(2000) (Table III; Figure 7A).
On monthly scale, the Mekong has the highest
contribution to the inflow during four months (Jun –Sep)
while during Oct –Jan the tributaries contribute most
(Figure 7B). The large majority of the inflow occurs during
the wet season (i.e. Jun –Oct). The outflow is dominated by
the discharge to the Mekong from October to April while
during May and September the evaporation and discharge
to the Mekong had an equal share (Figure 7B). From June
to August, the only outflow occurs via evaporation
(Figure 7B).
The mean annual difference between inflow and
outflow, when calculated from absolute differences, was
5.5 km
3
(Table III), equivalent to a mass balance error of
6.6% (calculated from absolute balance values). In other
words, the assumptions made in the water balance
simulations appear to have a small positive bias, slightly
overestimating the inflows relative to the outflows. There
were, however, years also with negative balance (Table III)
meaning that our estimated outflow was larger than inflow.
It should be noted that this does not mean that the lake
would dry out during these years, as the lake water level
returns to rather stable level at the end of each dry season
(Figure 6B), but that our model underestimated the outflow
or overestimated the inflow during that year.
These results serve to underscore once again the
dominant role of the Mekong mainstream in controlling
the seasonal storage volumes and water levels of Tonle
Sap Lake, consistent with the previous studies (e.g. Fuji
et al., 2003; Inomata and Fukami, 2008). Nonetheless, the
contribution of its own tributaries to the seasonal
dynamics of the system is substantial. Any impacts upon
the hydrological or biochemical responses of these
catchments would themselves lead to complex influences
Figure 5. The flow chart of the water balance calculations
1729TONLE SAP WATER BALANCE
Copyright © 2013 John Wiley & Sons, Ltd. Hydrol. Process. 28, 1722–1733 (2014)

on the lake, as would changes in the Mekong mainstream.
However, as the water level in the lake is mainly
controlled by the water levels in the Mekong mainstream
(Kummu and Sarkkula, 2008), the changes in the
mainstream are more crucial in determining flooding
extent.
The lake, essentially functioning as a natural flood-
water reservoir, is important for the Lower Mekong Basin
and particularly for the Mekong Delta. This reduces
flooding from other areas during the wet season and
provides higher dry season flow to the delta from
September/October onwards. During the July and August,
when the discharge in the Mekong is at its highest, around
20% of the Mekong mainstream discharge enters into the
lake (Figure 8). The dry season impact is more significant.
During the period of October–March, the Tonle Sap
discharge accounts for 20–50% of the total discharge
entering to the Chaktomuk confluence and further into the
delta (Figure 8). This is well in with the findings of Fuji
et al. (2003) and Inomata and Fukami (2008).
A: Modelled lake volume - measured lake volume
-10
-5
0
5
10
May-1997 May-1998 May-1999 May-2000 May-2001 May-2002 May-2003 May-2004
B: Measured lake volume vs. modelled lake volume
0
10
20
30
40
50
60
70
80
May-2005
May-1997 May-1998 May-1999 May-2000 May-2001 May-2002 May-2003 May-2004 May-2005
Measured volume Modelled volume
Volume
[
km3
]
Volume
[
km3
]
Figure 6. Water balance model results for Tonle Sap: A. Measured lake volume subtracted from the modelled lake volume. B. Measured lake volume
derived from water levels plotted against simulated lake volume from water balance calculations
Table III. Summary of the in-flow and out-flow in km
3
for each analysed hydrological year (May 1
st
–Apr 30
th
). Tribs. refers to Tonle
Sap’s own tributaries, TSR to Tonle Sap River (i.e. water from Mekong mainstream, Overl. to overland flow, Prec. to precipitation,
Evap. to evaporation, and TOT to total. Note: negative balance means that our estimated outflow was larger than inflow;it does not
mean that the lake would dry out during these years
Inflow into the lake Outflow from the lake
Balance
Tribs. TSR Overl. Prec. TOT TSR Overl. Evap. TOT
Year km
3
/yr km
3
/yr km
3
/yr km
3
/yr km
3
/yr km
3
/yr km
3
/yr km
3
/yr km
3
/yr km
3
/yr
1997 26.4 46.5 1.7 8.5 84.5 65.9 1.7 10.0 77.6 6.4
1998 20.0 24.5 0.0 6.7 51.1 38.8 0.0 7.0 45.8 5.8
1999 38.8 29.2 0.7 11.3 80.7 74.0 0.7 11.1 85.7 5.0
2000 42.1 47.1 6.4 15.0 109.0 93.3 6.4 14.7 114.4 5.7
2001 31.9 49.8 4.4 12.4 98.3 84.2 4.4 12.2 100.8 2.4
2002 23.7 54.0 3.8 11.3 93.2 82.6 3.8 10.6 97.0 4.0
2003 20.9 38.9 0.9 8.0 69.1 51.1 0.9 8.7 60.8 8.1
2004 20.8 44.6 2.0 9.6 79.5 60.8 2.0 10.5 73.2 6.6
avg. 29.1 41.8 2.5 10.4 83.1 68.8 2.5 10.6 81.9 5.5
%of TOT 34.1%50.3%3.2%12.4%84.0%3.0%13.0%6.6%
1730 M. KUMMU ET AL.
Copyright © 2013 John Wiley & Sons, Ltd. Hydrol. Process. 28, 1722–1733 (2014)

DISCUSSION AND CONCLUSIONS
The exploratory Tonle Sap water balance analysis
presented here describes the hydrologic regime of the
lake–floodplain system and provides a meaningful
assessment of the inflow and outflow contributions by
the various hydrological and hydrometeorological com-
ponents of the system. Moreover, it demonstrates the
importance and the influence of the relationships among
water level, flow, and flow direction in a complex
hydraulic environment.
We found that more than half of the annual inflow to
the lake originates from the Mekong mainstream. Thus,
flow alterations in the mainstream would have direct
impacts on the Tonle Sap water levels and hydrology as
well. Recent research has shown that the relatively small
rises in the dry season lake water level would
permanently inundate disproportionately large areas of
floodplain (Kummu and Sarkkula, 2008), rendering it
inaccessible to floodplain vegetation and eroding the
productivity basis of the ecosystem by reducing the
inundated area, and duration and amplitude of flooding
(Lamberts, 2008). The lake extension would thus cause
permanent submersion, in essence destruction of
considerable areas of the gallery forest surrounding the
lake (Kummu and Sarkkula, 2008).
A number of priority issues have emerged that would
need to be addressed in order to progress in the hydrological
and hydrodynamic understanding of the system and
therefore better understand its vulnerability to the impacts
of regional development:
•Accurate flow rating curves must be maintained and
methodically updated. This is not only the case for the
Tonle Sap River itself, but also for the lake’s own tributary
systems.
•Extreme hydrological episodes and floods in particular
will require a combination of field observations and
complex hydraulic models in order to understand the
processes involved and therefore identify management
and mitigation strategies that will be effective.
•Simple but reliable methods are required to estimate the
ungauged hydrology of the system, which remains
significant.
•The utility of the available data needs to be maximized
not only by using it but also by verifying it and
uncovering the observed temporal and spatial patterns
and ensuring that these are accounted for in any system
simulation models that are developed.
•The overland flow component is almost impossible to
monitor continuously. The ability to accurately quantify
this fundamental hydrological element of the system
-125
-100
-75
-50
-25
0
25
50
75
100
125
1997 1998 1999
Outflow
Inflow
2000 2001 2002 2003 2004
Precipitation
Tributaries
Mekong
Evaporation
Mekong
84.5
51.1
80.7
109
98.3 93.2
69.1
79.5
76.3
45.5
85.1
111
99.0 96.4
59.6
71.8
12.5%
34.0%
52.5%
88.5%
11.5%
Volume (km3)
Outflow
Inflow
Precipitation
Tributaries
Mekong
Evaporation
Mekong
-25
-20
-15
-10
-5
0
5
10
15
20
25
Jan Feb Mar AprMay Jun Jul Aug Sep Oct Nov Dec
A:
B:
Volume (km3)
Figure 7. Water balance model results for Tonle Sap:A. Annual (hydrological year) water balancesfor Tonle Sap Lake sorted by water balance components. B.
Monthly average water balance for Tonle Sap Lake. Note: the Mekong part includes both,the discharge via Tonle Sap River and overland flow
1731TONLE SAP WATER BALANCE
Copyright © 2013 John Wiley & Sons, Ltd. Hydrol. Process. 28, 1722–1733 (2014)

hydrodynamics lies with large highly detailed topographic
information combined with fine-scale hydrodynamic
modelling. Thus, intensive survey campaigns are needed
in order to obtain the detailed topographical information
and appropriate scale hydrodynamic modelling.
•As stated in Section on Analytical Framework, our model
does not include groundwater infiltration due to the
absence of detail data of these processes in the lake and its
floodplain. This might partly explain the observed small
positive bias in the water balance simulations (i.e. slightly
overestimating the inflows relative to the outflows). We
thus highlight the importance of starting a continuous
groundwater measurement campaign within the Tonle
Sap floodplains and conducting a detail modelling
exercise on floodplain –groundwater interactions.
•The regional rainfall climate needs improved study,
particularly from the point of view of the contribution
made by rainfall upon the lake surface. The same point
also applies to open water evaporation.
ACKNOWLEDGEMENTS
The work was partly completed within the WUP-FIN
project under the Mekong River Commission. Parts of the
work was also funded by the Maa- ja vesitekniikan tuki ry,
while M. Kummu was partly funded by the Academy of
Finland (projects 111672 and 133748) and postdoctoral
funds of Aalto University. The authors are grateful to the
WUP-FIN team, particularly to Marko Keskinen, Hannu
Lauri, and Markku Virtanen. The support of the Mekong
River Commission is greatly acknowledged; particularly,
the help of John Forsius in developing the Prek Kdam
rating curve is very much appreciated. Our colleagues at
Aalto University, particularly Marie Thouvenot-Korppoo
and Prof. Olli Varis, are acknowledged for their valuable
support. The constructive and thoughtful comments of the
two anonymous reviewers are equally highly appreciated.
REFERENCES
Adamson PT. 2001. Hydrological perspectives on the Lower Mekong
Basin - the potential impacts of hydropower developments in Yunnan
on the downstream flow regime. International Water Power and Dam
Construction March 2001: 16–21.
ADB. 2004. Cumulative impact analysis and Nam Theun 2 contributions.
Final report. Prepared by NORPLAN and EcoLao for Asian
Development Bank. 143 pp.
Ahmed M, Hap N, Ly V, Tiongco M. 1998. Socioeconomic assessment of
freshwater capture fisheries of Cambodia. A report on a household
survey. MRC/DoF/Danida Project for the Management of the
Freshwater Capture Fisheries of Cambodia. Mekong River Commis-
sion, Phnom Penh, Cambodia. 186p.
Arias ME, Cochrane TA, Piman T, Kummu M, Caruso BS, Killeen TJ.
2012. Quantifying changes in flooding and habitats in the Tonle Sap
Lake (Cambodia) caused by water infrastructure development and
climate change in the Mekong Basin. Journal of Environmental
Management 112:53–66.
Ashfaq M, Shi Y, Tung WW, Trapp RJ, Gao XJ, Pal JS, Diffenbaugh NS.
2009. Suppression of south Asian summer monsoon precipitation in the
21st century. Geophysical Research Letters 36: L01704.
Burnett WC, Peterson RN, Chanyotha S, Wattayakorn G, Ryan B. 2012.
Using high-resolution in situ radon measurements to determine
groundwater discharge at a remote location: Tonle Sap Lake,
Cambodia. Journal of Radioanalytical and Nuclear Chemistry 1–7.
Certeza Surveying. 1964. Final report to Committee for Coordination of
Investigations of the Lower Mekong Basin. Quezon City, The Philippines.
Dore J, Yu X. 2004. Yunnan hydropower expansion - Update on China’s
energy industry reforms & the Nu, Lancang & Jinsha hydropower
dams. Chiang Mai University’s Unit for Social and Environmental
Research & Green Watershed, Kunming: PR of China; 38.
Ellis EE, Keil RG, Ingalls AE, Richey JE, Alin SR. 2012. Seasonal
variability in the sources of particulate organic matter of the Mekong
River as discerned by elemental and lignin analyses. Journal of
Geophysical Research 117: G01038.
FAO. 2006. Global Forest Resources Assessment 2005 - Progress towards
sustainable forest management In FAO Forestry Paper. Food and
Agriculture Organisation of United Nations (FAO): Rome.
Farr TG, Rosen PA, Caro E, Crippen R, Duren R, Hensley S, Kobrick M,
Paller M, Rodriguez E, Roth L, Seal D, Shaffer S, Shimada J, Umland J,
Werner M, Oskin M, Burbank D, Alsdorf D. 2007. The Shuttle Radar
May Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr
0%
-30%
-20%
-10%
10%
20%
30%
40%
50%
60%
70%
-15,000
-10,000
-5000
0
5000
10,000
15,000
20,000
25,000
Share of total inflow to Mekong Delta
30,000
35,000
TSR share of inflow to Mekong delta
Discharge: Mekong mainstream
Discharge: Tonle Sap River (TSR)
Discharge [m3/s]
Figure 8. Monthly average discharge into Chaktomuk confluence (at Phnom Penh in Figure 1), showing the discharges of the Mekong mainstream and
Tonle Sap River at Phnom Penh Port (positive discharges correspond to flow into the confluence) plotted with the relative percentual share of Tonle Sap
River flow into the confluence and further on Mekong delta
1732 M. KUMMU ET AL.
Copyright © 2013 John Wiley & Sons, Ltd. Hydrol. Process. 28, 1722–1733 (2014)

Topography Mission. Reviews of Geophysics 45: RG2004, doi:10.1029/
2005RG000183.
Finlayson CM, Spiers AG. 1999. Global review of wetland resources and
priorities for wetland inventory. Supervising Scientist Report 144 / Wetlands
International Publication 53, Supervising Scientist, Canberra, Australia;524.
Fuji H, Garsdal H, Ward P, Ishii M, Morishita K, Boivin T. 2003.
Hydrological roles of the Cambodian floodplain of the Mekong River.
International Journal of River Basin Management 1:1–14.
Grumbine RE, Xu J. 2011. Mekong Hydropower Development. Science
332: 178–179.
Hai PT, Masumoto T, Shimizu K. 2008. Development of a two-
dimensional finite element model for inundation processes in the Tonle
Sap and its environs. Hydrological Processes 22: 1329–1336.
Hori H. 2000. The Mekong: Environment and Development. United
Nations University Press: Tokyo; 398.
Hortle KG. 2007. Consumption and the yield of fish and other aquatic
animals from the Lower Mekong Basin. MRC Technical Paper No. 16,
Mekong River Commission, Vientiane. 87 pp.
ICEM. 2010. The Strategic Environmental Assessment of the Mekong
Mainstream: Final report. A report prepared for the Mekong River
Commission (MRC) by International Centre for Environmental
Management (ICEM): Vientiane, Lao PDR.
Inomata H, Fukami K. 2008. Restoration of historical hydrological data of Tonle
Sap Lake and its surrounding areas. Hydrological Processes 22: 1337–1350.
Johnston RM, Kummu M. 2012. Water resource models in the Mekong
Basin: a review. Water Resources Management 26: 429–455.
Junk W, Brown M, Campbell I, Finlayson M, Gopal B, Ramberg L,
Warner B. 2006. The comparative biodiversity of seven globally
important wetlands: a synthesis. Aquatic Sciences - Research Across
Boundaries 68: 400–414.
Junk WJ. 1997. The Central Amazon Floodplain: Ecology of a pulsing
system. Springer: Berlin; Ecological Studies 126 525.
Kazama S, Hagiwara T, Ranjan P, Sawamoto M. 2007. Evaluation of
groundwater resources in wide inundation areas of the Mekong River
basin. Journal of Hydrology 340: 233–243.
Keskinen M, Chinvanno S, Kummu M, Nuorteva P, Snidvongs A, Varis O,
Västilä K. 2010. Climate change and water resources in the Lower
Mekong River Basin: putting adaptation into the context. Journal of
Water and Climate Change 1: 103–117.
King P, Bird J, Haas L. 2007. The Current Status of Environmental
Criteria for Hydropower Development in the Mekong Region: A
Literature Compilation. Consultants Report to ADB (Asian Develop-
ment Bank), MRCS (Mekong River Commission Secretariat) and WWF
(World Wide Fund for Nature): Vientiane, Lao PDR; xv + 155.
Kingston DG, Thompson JR, Kite G. 2011. Uncertainty in climate change
projections of discharge for the Mekong River Basin. Hydrology and
Earth System Sciences 15: 1459–1471.
Koponen J, Kummu M, Sarkkula J. 2005. Modelling Tonle Sap Lake
environmental change. Verhandlungen des internationalen Vereins für
Limnologie 29: 1083–1086.
Kummu M, Lu XX, Wang JJ, Varis O. 2010. Basin-wide sediment
trapping efficiency of emerging reservoirs along the Mekong.
Geomorphology 119: 181–197.
Kummu M, Penny D, Sarkkula J, Koponen J. 2008. Sediment - Curse or
Blessing for Tonle Sap Lake? Ambio 37: 158–163.
Kummu M, Sarkkula J. 2008. Impact of the Mekong river flow alteration
on the Tonle Sap flood pulse. Ambio 37: 185–192.
Kummu M, Sarkkula J, Koponen J, Nikula J. 2006. Ecosystem
management of Tonle Sap Lake: integrated modelling approach.
International Journal of Water Resources Development 22: 497–519.
Kummu M, Varis O, Sarkkula J. 2009. Impacts of land surface changes on
regional hydrology - Mainland Southeast Asia. In Critical states:
Environmental challenges to development in monsoon Southeast Asia,
Lebel L, Snidvongs A, Chen C-TA, Daniel R (eds). Gerakbudaya:
Kuala Lumpur, Malaysia; 221–238.
Lamberts D. 2006. The Tonle Sap Lake as a productive ecosystem.
International Journal of Water Resources Development 22: 481–495.
Lamberts D. 2008. Little impact, much damage: the consequences of
Mekong River flow alterations for the Tonle Sap ecosystem. In Modern
Myths of the Mekong - A Critical Review of Water and Development
Concepts, Principles and Policies, Kummu M, Keskinen M, Varis O
(eds). Water & Development Publications - Helsinki University of
Technology: Finland; 3–18.
Lamberts D, Koponen J. 2008. Flood pulse alterations and productivity of the
Tonle Sap ecosystem: a model for impact assessment. Ambio 37: 178–184.
Landon MK. 2011. Preliminary Compilation and Review of Current
Information on Groundwater Monitoring and Resources in the Lower
Mekong River Basin, July 2011. Mekong River Commission.
Lauri H, de Moel H, Ward PJ, Räsänen TA, Keskinen M, Kummu M.
2012. Future changes in Mekong River hydrology: impact of climate
change and reservoir operation on discharge. Hydrology and Earth
System Sciences 16: 4603–4619.
May R, Jinno K, Tsutsumi A. 2011. Influence of flooding on
groundwater flow in central Cambodia. Environmental Earth Sciences
63: 151–161.
Mekong River Commission. 1999. Hydrographic atlas of 1999 for the
Tonle Sap River and Tonle Sap Lake. Mekong River Commission:
Phnom Penh, Cambodia.
Mekong River Commission. 2003. State of the Basin Report: 2003.
Mekong River Commission (MRC), Phnom Penh, Cambodia. 316 pp.
ISSN: 1728–3248.
Mekong River Commission. 2005a. Databases of Mekong River
Commission. Mekong River Commission (MRC): Vientiane, Lao
PDR. Metadata available on-line at http://www.mrcmekong.org/spatial/
data_catalog.htm.
Mekong River Commission. 2005b. Overview of the hydrology of the Mekong
Basin. Mekong River Commission, Vientiane, November 2005;82.
Mekong River Commission. 2005c. A Report on Likely Infrastructure
Developments on the Mekong Floodplain in Cambodia & their
Significance in Changing Flow Patterns; Draft, March 2005. Prepared
by Hugh Cross, Ecosphere Solutions Pty Ltd, for Mekong River
Commission: Vientiane, Lao PDR;48.
Mekong River Commission. 2007. Databases of MekongRiver Commission.
Mekong River Commission (MRC): Vientiane, Lao PDR. Metadata
available on-line at http://www.mrcmekong.org/spatial/data_catalog.htm.
Mekong River Commission. 2008. Map of the Existing, Under
Construction and Planned/Proposed Hydropower Projects in the Lower
Mekong Basin, September 2008. Map produced by the Mekong River
Commission (MRC). Available online at: http://www.mrcmekong.org/
programmes/hydropower.htm.
Mekong River Commission. 2009. Database of the Existing, Under
Construction and Planned/Proposed Hydropower Projects in the Lower
Mekong Basin. Mekong River Commission (MRC).
Mekong River Commission. 2010. Assessment of basin-wide develop-
ment scenarios –main report. Mekong River Commission (MRC),
Vientiane, Lao PDR.
Millennium Ecosystem Assessment. 2005. Ecosystems and human well-
being: wetlands and water synthesis. World Resources Institute:
Washington, DC;74.
MRCS/TSLV. 2004. Consolidation of Hydro-Meteorological Data and
Multi-functional Hydrologic Roles of Tonle Sap Lake and its Vicinities,
Phase-III Project Final Report. Mekong River Commission: Vientiane.
MRCS/WUP-FIN. 2003. Final Report. WUP-FIN Phase I - Modelling
Tonle Sap for Environmental Impact Assessment and Management
Support, Mekong River Commission and Finnish Environment Institute
Consultancy Consortium, Phnom Penh;107.
MRCS/WUP-FIN. 2007. Final Report - Part 2: research findings and way
forward. WUP-FIN Phase 2 - Hydrological, Environmental and Socio-
Economic Modelling Tools for the Lower Mekong Basin Impact
Assessment. Mekong River Commission and Finnish Environment
Institute Consultancy Consortium, Vientiane, LaoPDR. 126 pp. Available
on-line at http://www.eia.fi/wup-fin/wup-fin2/publications.htm.
MRCS/WUP-FIN. 2008. Model Report. WUP-FIN Phase II –Hydrological,
environmental and socio-economic modelling tools for the Lower Mekong
Basin impact assessment. Mekong River Commission and Finnish Environ-
ment Institute Consultancy Consortium: Vientiane, Lao PDR; 413 pp.
Pech S, Sunada K. 2008. Population growth and natural-resources
pressures in the Mekong River Basin. Ambio 37: 219–224.
Rainboth WJ. 1996. FAO species identification field guide for fishery
purposes. Fishes of the Cambodian Mekong. FAO (Food and
Agriculture Organization of the United Nations), Rome;265.
Räsänen TA, Koponen J, Lauri H, Kummu M. 2012. Downstream
hydrological impacts of hydropower development in the Upper Mekong
Basin. Water Resources Management 26: 3495–3513.
Shi W. 2008. Rubber Boom in Luang Namtha: A Transnational Perspective. In
Rural Development in Mountainous Areas of Northern Lao PDR. GTZ.
Sverdrup-Jensen S. 2002. Fisheries in the Lower Mekong Basin: Status
and perspectives. MRC Technical Paper No. 6, Mekong River
Commission, Phnom Penh. 103 pp. ISSN: 1683–1489.
Västilä K, Kummu M, Sangmanee C, Chinvanno S. 2010. Modelling
climate change impacts on the flood pulse in the Lower Mekong
floodplains. Journal of Water and Climate Change 1:67–86.
World Bank. 2004. Modelled Observations on Development Scenarios in
the Lower Mekong Basin. In Mekong Regional Water Resources
Assistance Strategy, Prepared for the World Bank with Mekong River
Commission cooperation. 142 p.
1733TONLE SAP WATER BALANCE
Copyright © 2013 John Wiley & Sons, Ltd. Hydrol. Process. 28, 1722–1733 (2014)