Journal of Cleaner Production xxx (2016) 1e7

Contents lists available at ScienceDirect

Journal of Cleaner Production

journal homepage: www.elsevier.com/locate/jclepro

Methodology development for including environmental water

requirement in the water stress index considering the case of Thailand
Pariyapat Nilsalab a, b, Shabbir H. Gheewala a, b, *, Thapat Silalertruksa a, b
a
The Joint Graduate School of Energy and Environment (JGSEE), King Mongkut's University of Technology Thonburi (KMUTT), Bangkok, 10140, Thailand
b
Center of Excellence on Energy Technology and Environment, PERDO, Bangkok, Thailand

a r t i c l e i n f o a b s t r a c t

Article history: Environmental water requirement is important for a river or a watershed to maintain the ecosystem
Received 14 January 2016 functions and services. Thus, it needs to be an essential component in water demand calculations for
Received in revised form maintaining water resources. This study proposes to explicitly account the environmental water as a part
21 November 2016
of water withdrawal in the withdrawal to availability ratio of the water stress index. A revised equation is
Accepted 22 November 2016
Available online xxx
formulated and calibrated using Thailand as an example. This incorporation allows exibility in changing
different amounts and setting different priorities of environmental water requirement based on a local
context. The results obtained are satisfactory for explaining the actual stress situation incorporating
Keywords:
Environmental water
environmental water requirement. The method and approach proposed in this paper can also be applied
Water stress index to other countries based on their priorities for water allocation to various water demand sectors. This
Freshwater resources paper is proposed as a step towards developing a sustainable tool for assessing the impact on freshwater
Thailand resources.
2016 Elsevier Ltd. All rights reserved.

1. Introduction season, especially in the northeastern region. Some of these

Water scarcity has become an issue of concern in many coun-
tries where there are situations of insufcient water to satisfy all
demands all the year round or seasonally (IHP, 2011). This crisis
affects not only human activities but also environmental functions.
More serious is a concern about overuse of freshwater resources
that which may result from both from increasing of demands for
water or lacking of water supply. Areas where water is insufcient
to satisfy all the existing demands tend to be under water stressed
conditions. Also, if future demands for water is expected to in-
crease and future water supply may not be able to meet this
increasing demand, it will lead to put additional stress or pressure
on freshwater resources accordingly. In 2015, Thailand faced the
worst drought in 15 years and the farmers cultivating dry season
crops especially in Chao Phraya and Mae Klong watersheds were
the rst group suffering from the direct impacts of drought
(Teerawiroon, 2015). In accordance with the statistical data during
1992e2011, 51e60 provinces underwent drought in the dry
* Corresponding author. The Joint Graduate School of Energy and Environment
(JGSEE), King Mongkut's University of Technology Thonburi (KMUTT), Bangkok,
10140, Thailand.
E-mail address: shabbir_g@jgsee.kmutt.ac.th (S.H. Gheewala).
drought periods, for example in 2010, were affected by El Nin ~o
events (OAE, 2010; TMD, 2010). In particular, agricultural areas in
42 provinces were affected by drought during 2003e2007 (DWR,
2012; LDD, 2013; Mapraneat, 2014; DOAE, 2014). However, more
recently, there has been more awareness on this concern and it
has been receiving more attention with development of indicators
for measuring the stress or impact on freshwater resources from
freshwater use. The water stress index (WSI) developed by Pster
et al. (2009) has been used as it is one of the more accepted
methods and widely applied in many studies, including Thailand
(Gheewala et al., 2013, 2014). Several studies related to individual
food product such as milk and dairy farming (De Boer et al., 2013;
Zonderland-Thomassen and Ledgrad, 2012), tea and margarine
(Jefferies et al., 2012), wine (Quinteiro et al., 2014), tomato (Page
et al., 2011, 2012), fruits and vegetables (Stoessel et al., 2012) or
an agricultural crop such as cassava, maize, corn (GIZ, 2013;
Jeswani and Azapagic, 2011) have used this method. All these
case studies revealed that cultivation is the most water intensive
process of food and agricultural crop production. As a result, a
nexus of food, feed, and fuel crops such as rice, cotton, maize and
sugarcane is taken into consideration by studies of Pster et al.
(2011), Nunez et al. (2012) and Gheewala et al. (2013). Other
than these applications, the WSI has also been applied for the

http://dx.doi.org/10.1016/j.jclepro.2016.11.130
0959-6526/ 2016 Elsevier Ltd. All rights reserved.

Please cite this article in press as: Nilsalab, P., et al., Methodology development for including environmental water requirement in the water
stress index considering the case of Thailand, Journal of Cleaner Production (2016), http://dx.doi.org/10.1016/j.jclepro.2016.11.130
2 P. Nilsalab et al. / Journal of Cleaner Production xxx (2016) 1e7
assessing environmental impacts of water supply alternatives and
water use of a particular area such as Segura Basin in the southeast
of Spain (Uche et al., 2014, 2015). In addition to the studies on the
WSI application, studies have also been conducted on the method
development (Ridoutt and Pster, 2013; Hospido et al., 2013). This
index is derived from quantitative thresholds with reference to
reduced water availability from withdrawal. It is also accounted as
a water scarcity index based on the ISO 14046 (Boulay et al., 2014)
and can be used as a characterization factor for a water depriva-
tion impact based on life cycle assessment (LCA) methodology, an
decision making tool to assess environmental sustainability.
Basically not only humans need water for satisfying their de-
mands, but water is also required by the environment, especially a
river or a watershed system, to maintain its functions and
ecosystem services. The environmental water has become an
important issue at every level: global, regional, national and local,
in order to maintain the values and benets of a river or watershed.
This environmental water is affected by overuse of freshwater re-
sources as well as water scarcity (Navarro-Ortega et al., 2015) and it
requires a concern at different scale of analysis from large to small
(Smakhtin, 2002). The environmental water requirement (EWR)
appears to have been concerned at a global scale when the study of
Smakhtin et al. (2004) suggested that EWR needs to be addressed
seriously in water resources management and development. In
addition, in some countries such as the United States (some parts),
Australia, South Africa, India, Thailand, etc., a concern of EWR has
been taken into consideration in policy and management contexts
to protect their natural water resources (Davis et al., 2001; RID,
2011a; Venot et al., 2008). Hence environmental water legislation
is regulated to ensure that available freshwater must remain in a
river (Pimentel et al., 2004). As a result, the total demands for water,
including for humans and the environment, must be accounted in
order to ensure that the impact on freshwater resources are equi-
tably and consistently assessed. In the case of Thailand, allocating
water to all users and setting priorities for water allocation are
under the Royal Irrigation Department (RID) which considers EWR
as one of the water demand sectors (RID, 2011a).
The concern of environmental water is implicitly included in the
WSI metric of Pster et al. (2009) as it is accommodated in the
values of the water stress thresholds. However the EWR term is not
taken into account separately and explicitly in this method. Thus, it
is the same for the case of the country WSI for Thailand which
followed Pster's method (Gheewala et al., 2013; 2014). As the
regulation of EWR dened by RID varies based on the environ-
mental conditions, using constant threshold values as in Pster's
method may tend to take a conservative approach and sometimes
lead to overestimation of water stress. Consequently this study
attempts to emphasize the importance of the EWR by explicitly
accounting the EWR in the country WSI. This revised approach will
help ensure condence in how much freshwater can be extracted
sustainably without causing damage to the ecological system.
Accordingly, this approach can be used as decision support in
evaluating cleaner production.
2. Methodology
2.1. The water stress index of Pster et al. (2009)
The WSI of Pster et al. (2009) is developed based on a logistic
function in relation to a ratio of water withdrawal to water avail-
ability (WTA) as showed in Equation (1). Four signicant sectors of
water consumers including household, industry, agriculture, and
livestock contribute to water withdrawal, and the available fresh-
water to supply these demands is considered based on the hydro-
logical boundaries.
Please cite this article in press as: Nilsalab, P., et al., Methodology development for including environmental water requirement in the water
stress index considering the case of Thailand, Journal of Cleaner Production (2016), http://dx.doi.org/10.1016/j.jclepro.2016.11.130
1
WSI
 (1)
e6:4WTA
* 1
1 0:01 1
where; WTA* is dened by ow regulation by dams (strong p regu-
lation ow, SRF) and without dams (non-SRF); WTA* VF  WTA
(for SRF) and WTA* VF  WTA (for non-SRF).
The WTA* is established in order to address variation of water
availability affecting water stress level. Thus precipitation variation
(variation factor, VF) and ow regulation are taken into account in
Pster's method. The ow regulated by dam leads to a decrease in
the inuence of VF while this inuence is fully accounted for in the
unregulated ow. Monthly average and annual precipitation are
assumed to follow a log-normal distribution; therefore, VF is
quantied from the arithmetic standard deviations of the log-
p
transformed values of monthly

and
2
annual
 2
rainfall as expressed
in this equation: VF e lnSmonth lnSyear .
Water stress levels of Pster's method are dened based on
threshold values of the WTA ratio ranging from 0.2 to 0.6. This
range indicates a level of water stress situation from moderate to
severe and the critical threshold is determined at 0.4. Values above
this critical value indicate a severe water stress situation and an
unhealthy freshwater ecosystem. The range of water stress
thresholds is established based on expert judgement and literature
with a concern on freshwater ecosystem (Alcamo et al., 2000, 2003;
Pster et al., 2009). Accordingly it can be inferred that lesser than
60% of water availability at the critical threshold cannot meet a
good or fair condition of freshwater ecosystem. Although 60% of
available water cannot be concluded as the total EWR, it can imply
that EWR is implicitly included in the WSI via the thresholds
varying from 0.2 to 0.6.
The logistic function is applied to convert the threshold values of
WTA ratio into a continuous value of WSI which ranges from 0.09 to
0.91. Thus the inection point of the logistic curve, which is at 0.5, is
equivalent to the critical threshold at 0.4 of WTA ratio. Classica-
tion of WSI in relation to WTA is tabulated in Table 1.
In the revised approach, the national standard of EWR is ob-
tained from the RID. The EWR for 25 watersheds of Thailand is
assigned based on a minimum ow of a long-term runoff in order to
maintain a certain water volume/ow for ecosystem health. Ac-
cording to the priority given by the RID, the EWR is taken into ac-
count by incorporating it in water withdrawal. To reect this
incorporation, the thresholds are redened in relation to the EWR
by keeping the same structure of the WSI function.
2.2. Environmental water requirement
The important function and purpose of EWR is to maintain a
river system in desired environmental conditions for ecosystem
services and human livelihoods. Therefore denition of EWR can
be described in terms of a certain quantity and/or quality in
relation to a time scale (Forslund et al., 2009; Boelee, 2011).
Accordingly EWR can be used as indicators of not only aquatic
ecosystem health but also the impact on freshwater resources.
However there are various factors inuencing functions and
ecosystem services of a river or a watershed, stream ow and
Table 1
Water stress index (WSI) classication (Pster et al., 2009).
WTA WSI classication Category
<0.2 WSIp < 0.09 Low water stress
0.2 < WTA < 0.4 0.09  WSIp < 0.5 Moderate
0.4 < WTA < 0.6 0.5 < WSIp  0.91 Severe
>0.6 WSIp > 0.91 Extreme
 P. Nilsalab et al. / Journal of Cleaner Production xxx (2016) 1e7 3

ecological conditions play a signicant role in methodology Table 2

development for determining EWR. Many methods have been Environmental water requirement of 25 watersheds in Thailand (RID, 2011b).

introduced and published based on basic to complex approaches Watershed Million m3/month Million m3/year m3/sec
including hydrological (xed annual low ow thresholds by indi- Salawin 153.0 1838.0 59.1
cating ecological degradation levels), hydraulic and habitant Kok 58.2 698.4 22.5
(relationship between a measured data of river cross sections and Ping 139.3 1671.9 52.0
an aquatic species data obtained from a specic model simula- Wang 14.4 173.4 5.4
Yom 16.7 200.6 6.4
tion), and holistic (a combination of hydrological, hydraulic,
Nan 97.7 1172.4 36.5
habitat, and expert judgment for data simulation) assessments of Khong 280.1 3361.0 108.4
river system (Acreman and Dunbar, 2004; Alcamo et al., 2007; Do ll Chi 157.1 1884.7 63.3
et al., 2009; Pastor et al., 2014; Smakhtin et al., 2004). The simplest Mun 115.2 1382.6 47.4
Chao Phraya 105.0 1261.4 40.0
method is based simply only on the hydrological aspect of stream
Sakae Krang 7.4 89.1 2.9
ow; while, the most complex (holistic) method considers all the Pasak 74.6 894.8 29.0
aspects (Acreman and Dunbar, 2004; Pastor et al., 2014). Addi- Thachin 24.0 228.2 9.0
tionally, to collect data related to these three aspects requires Mae Klong 155.5 1865.6 64.3
expertise in each eld, time, and funding. To support decision Petchaburi 38.6 463.6 16.0
West Coast Gulf 26.5 318.2 11.0
makers with a concern on environmental water, Acreman and
Prachin Buri 9.8 117.4 4.1
Dunbar (2004) divide this concern into four levels; national Bang Pakong 19.8 238.0 7.4
assessment, watershed planning, impact assessment, and river Thole Sap 21.8 262.2 8.2
restoration. Therefore, to provide a general safeguard of EWR with East Coast Gulf 90.6 1087.8 33.8
a concern at national assessment, a simple and rapid approach Peninsula East Coast 372.2 4466.7 153.9
Tapi 316.5 3797.5 118.2
(the hydrological method) is suitable to apply. Thale sap Songkhla 61.7 740.9 23.1
The functions of EWR in the context of Thailand include main- Pattani 127.3 1527.1 47.5
taining aquatic ecosystems, water level for transport, pollutant Peninsula West Coast 317.3 3807.3 120.7
loads (dilution or push), sheries (sh cage culture) and seawater
intrusion, depending on location (RID, 2011a). Hence different
amounts of environmental water are required for different func- 3. Results and discussion
tions and periods of time. Thailand, an agricultural country, is
divided into 25 watersheds based on hydrological boundaries. Incorporation of the EWR in the WSI is proposed for exibility in
Around 50% of total land (2010e2013) is utilized for agriculture; changing the EWR based on a local context and applied to the case
therefore, this is the most water intensive sector accounting for 65% of Thailand as an illustration. Thus the incorporation of EWR is
of the total water demand. Water for maintaining river and its considered based on the country's regulation for EWR while
ecosystem is the second largest use of freshwater at about 18% of maintaining the original WSI function and classication. Accord-
the total water demand, followed by domestic, industry, and live- ingly, the new equation of WSI with respect to EWR is proposed for
stock respectively (RID, 2014a). In addition, water resources con- the case of Thailand.
servation and management plans from RID have been developed on
a priority basis. The priority with a water use target is allocated to 3.1. Including of environmental water requirement (EWR) in the
the agricultural, ecological, household, and industrial sectors at water stress index (WSI)
43%, 40%, 14% and 2%, respectively (RID, 2014b). The EWR is given
the second-highest priority after the requirement for agricultural The relationship between withdrawn and available water is a
purposes. focus in order to put emphasis on the importance of environmental
The country standard of EWR is quantied based on the hy- water with reference to the issue of water scarcity. As a result, the
drological method and watershed boundary in order to maintain a WTA ratio plays an important role in explicitly including the EWR
certain amount/level of water for river function and its ecosystem in the WSI.
services. Thailand is divided into 25 watersheds; therefore, EWR is According to the importance and priority of EWR in Thailand,
assigned for each watershed based on its long-term runoff (RID, the environmental water is considered and allocated based on a
2011a). Most of EWR is regulated at a minimum monthly ow, human perspective. Therefore our study proposes to bring out the
except for three watersheds, Yom, Pattani and Peninsula West Coast term of EWR by explicitly adding it the water withdrawal part of
which are indicated at 90 percent of time-specic ow (Q90). EWR WTA ratio as expressed in Equation (2);
of 25 watersheds in Table 2 are obtained from RID and estimated
based on a 50-year runoff record (1958e2008). WTAe WW EWR=WA (2)
The minimum ow or Q90 can be obtained from constructing a
ow duration curve (FDC) of long-term data. This curve represents where; WTAe stands for the ratio of withdrawal to availability
the magnitude and frequency of river ows. It is constructed from where withdrawal includes EWR.
ranked ows (sorted from the largest to smallest) and their ex-
ceedance probability during the period of interest (percentage of WW refers to four sectoral water withdrawals (household, in-
time) in order to display a range of cumulative frequency distri- dustry, agriculture, and livestock).
butions of river ow. The FDC of Ping river, the main stream of Ping EWR is environmental water requirement specied as one of
watershed based on 10 years of record, is demonstrated as an sectoral water withdrawals.
example in Fig. 1. EWR of Ping is dened at the percentage of time WA is a total of water availability.
where the minimum ow (not equal to zero) is achievable during a
given period; therefore, the EWR in this case is equal to Q100 at The WTAe term used in this study is to make a clear difference
42 m3/sec (RID, 2011a; 2011b). from the WTA referred in the Pster's method. It is clear from
Equation (2) that WTAe will always have a value greater than WTA.
Besides if there is no water withdrawal, WTAe can still be quantied

Please cite this article in press as: Nilsalab, P., et al., Methodology development for including environmental water requirement in the water
stress index considering the case of Thailand, Journal of Cleaner Production (2016), http://dx.doi.org/10.1016/j.jclepro.2016.11.130
4 P. Nilsalab et al. / Journal of Cleaner Production xxx (2016) 1e7

Fig. 1. Flow duration curve of Ping River (2000e2009) (RID, 2011a).

from the term of (EWR/WA).
Although this approach will allow the WSI to be more exible in
changing different amount of EWR, adjustment of the thresholds is
also required in order to reect changes in the WTA ratio. Incor-
poration of EWR in the water withdrawal leads to re-interpreting
the critical threshold of water stress. The denition of critical
threshold for the WTAe ratio means that a stress situation is going
to take place when the total of water withdrawal (including EWR) is
greater than the total water availability. This situation can imply
that among all water users may face different water stress situation
and to satisfy all demands for water with some additional water
resources may result in further depletion and deterioration of
freshwater resources. Thus the WTAe ratio equal to 1 is interpreted
as the critical threshold instead of 0.4 dened in the Pster's
method.
Accordingly the original WSI equation (Equation (1)) is adjusted
to the new critical threshold and the result reveals that the critical
value at 1 is not equivalent to the inection point at 0.5 of WSI curve
(the dotted line with cross makers in Fig. 2). As a result, to keep the
same structure of the WSI function using the adjusted value of VF,
the constant in Equation (1) is required to be modied with respect
to the new critical threshold. The median of a set of VF values for
Thailand is obtained from Gheewala et al. (2013, 2014). The new
constant is computed by using the country data for VF and main-
taining the new critical value equivalent at 0.5 of WSI. This leads to
shift the revised WSI curve of Thailand outward (to the right);
however, the classication of WSI remains the same (the dashed
line with triangle makers in Fig. 2). Consequently the revised WSI
equation is expressed in Equation (3) as WSIe which is the same as
WSI but using WTAe (thus including EWR in WTA).
1
WSIe *

1  (3)
1 e1:82WTAe 0:01 1
To keep the same WSI classication as shown in Table 1, the

Fig. 2. Relationship between withdrawal to availability (WTA) and water stress index (WSI) based on the original method of Pster et al. (2009) and the revised WSI in relation to
environmental water requirement (EWR).

Please cite this article in press as: Nilsalab, P., et al., Methodology development for including environmental water requirement in the water
stress index considering the case of Thailand, Journal of Cleaner Production (2016), http://dx.doi.org/10.1016/j.jclepro.2016.11.130
 P. Nilsalab et al. / Journal of Cleaner Production xxx (2016) 1e7 5

corresponding results of WSI at 0.09 and 0.91 can also be used to Table 4
indicate the minimum and maximum values of water stress Thailand's water stress index with respect to environmental
water requirement.
threshold as the new classication which is presented in Table 3.
However the meaning of moderate to severe stress is different from Watersheds WSIe
Pster's method (Table 1), the moderate level of the revised method Salawin 0.016
(Table 3) as well as the value of WTAe less than 1 is interpreted in Kok 0.017
terms of an opportunity to face a stress situation. Whereas the Ping 0.015
Wang 0.013
severe level (Table 3) or the WTAe results equal or exceed 1 is
Yom 0.016
implied when a water stressed situation initially occurs. Nan 0.013
As the precipitation data from Thailand is utilized in calculating Khong 0.013
the constant (appearing as the exponent of e) in the WSIe equa- Chi 0.039
Mun 0.076
tion [i.e. Equation (3)] developed in this study, it cannot be directly
Chao Phraya 0.037
used for other countries. However, the same procedure developed Sakae Krang 0.015
in this study can directly be used to develop the constant in the Pasak 0.020
WSIe equation for any country by using the precipitation data of Thachin 0.030
that country. For the new equation, the EWR can be varied Mae Klong 0.015
Petchaburi 0.018
depending on the local context which is not be possible using the
West Coast Gulf 0.026
Pster's method as for the latter, the EWR is implicit and cannot be Prachin Buri 0.012
changed. Bang Prakong 0.014
Thole Sap 0.013
East-Coast Gulf 0.013
3.2. Adapting EWR incorporation in the WSI to Thailand context Peninsula-East coast 0.024
Tapi 0.031
Though the basis of the revised method remains the same as the Thale sap Songkhla 0.013
original one, the explicit inclusion of EWR allows exibility in Pattani 0.027
Peninsula-West coast 0.015
changing its value based on the local environmental context. Thus
the WSIe equation (Equation (3)) with EWR based on minimum
ow (the country standards for 25 watersheds in Table 2 by RID) is
applied for determining the WSI of Thailand. Annual freshwater WSI results. As a result, another implication is also that an appro-
availability is obtained from the studies of Gheewala et al. (2013, priate selection of the estimation method for EWR is essential for
2014). Withdrawn water for household, industry, livestock and the method to effectively simulate the actually observed situation
agriculture is calculated based on the national water demand in the eld.
standards provided by RID (RID, 2011a). This corresponds to the In reality, different amount of environmental water is required
newly updated consensus on a water scarcity indicator for LCA- depending on temporal and spatial variations to maintain river
based water footprint from the Water Use in LCA Working Group function and its ecosystem services especially the regulatory ser-
(WULCA). The scarcity indicator is recommended to be developed vices to keep river's ecosystem conditions and natural protection.
based on a ratio of demand to availability (Pster et al., 2015). For example, the Bang Pakong, Mae Klong, Thachin, and Chao
The results presented in Table 4 reveal that all the watersheds in Phraya rivers have struggled with a seawater intrusion problem
Thailand have a low opportunity to face a stress situation. The (RID, 2015) while Ping has faced a problem of water quality dete-
highest result is found in the Mun watershed; therefore, this rioration due to the discharge of municipal wastewater and sh
watershed has the potential to lead to a moderate stress situation in cage culture (Royal Irrigation Department Ofce 1, 2014). These
the future. The relation between the results of WSIe and the country problems get worse as the period or dry season is longer. Also, the
WSI determined by Gheewala et al. (2013, 2014) is illustrated in problem of seawater intrusion has normally occurred in dry season
Fig. 3. The results of WSIe are smaller than the country WSI (can be especially in late January up to March (HAII, 2014). To mitigate the
noticed from Fig. 3) because the amounts of EWR regulated by RID serious problem of seawater intrusion in 2007, around 197 million
range from 1% to 19% of available water. These small amounts of m3 water is required for Bang Pakong to slow down the seawater
EWR lead to none of the watersheds facing a stress situation intrusion for 60 days or about 98 million m3/month (ONEP, 2007).
detected in this study. Although EWR implicitly associated with the Recorded data of Thachin, Mae Klong, and Chao Phraya watersheds
amount of water available from 40% to 80% can implied from the range between 130 and 168, 181e233, and 233e246 million m3 of
threshold values of water stress in Pster's method, it is likely that monthly environmental water, respectively (RID, 2015). Regarding
water stress levels are being overestimated for the case of Thailand the RID standard of EWR, only the EWR of Mae Klong watershed
in the study of Gheewala et al. (2013, 2014). To include the EWR meets the requirement for mitigating the seawater intrusion.
explicitly as proposed in this study can help reduce this over- Furthermore the shares of EWR for Chao Phraya and Thachin wa-
estimation by applying different ranges of EWR. In addition, if the tersheds have increased to 48% and 61% of total water allocation to
amount of EWR is small, it will not signicantly affect the results of account for seawater intrusion (RID, 2015).
WSI. On the other hand, the large amount of EWR will reect on the Consequently, using the WSIe equation (Equation (3)) with EWR

Table 3
Classication of water stress index with respect to environmental water requirement (WSIe).

WTAe WSIe Classication Category

WTAe < 0.2 WSIe < 0.09 No stress

0.2 < WTAe < 0.5 0.02 < WSIe < 0.09 Low (opportunity to face a stress situation)
0.5 < WTAe < 1 0.09 < WSIe < 0.5 Moderate (opportunity to face a stress situation)
1  WTAe < 1.5 0.5  WSIe < 0.91 Severe (facing a stress situation)
WTAe > 1.5 WSIe > 0.91 Extreme (under a stress situation)

Please cite this article in press as: Nilsalab, P., et al., Methodology development for including environmental water requirement in the water
stress index considering the case of Thailand, Journal of Cleaner Production (2016), http://dx.doi.org/10.1016/j.jclepro.2016.11.130
6 P. Nilsalab et al. / Journal of Cleaner Production xxx (2016) 1e7
Fig. 3. Relationship between withdrawal to availability (WTA) and water stress index (WSI) for the case of Thailand based on the original method of Pster et al. (2009) and the
revised WSI (WSIe) in relation to environmental water requirement (EWR).
based on minimum ow (the RID standard) does not capture the
real situation of EWR such as preventing the seawater intrusion in
Chao Phraya and Thachin watersheds. This is because the EWR
being regulated using only the minimum ow approach may not
correspond with reality as it may tend to underestimate results,
especially during the dry season. Thus improving or modifying the
estimation method for the case of Thailand, particularly based on a
monthly basis would be recommended. Accordingly, the situation
can be remedied by using a different EWR based on preventing
seawater intrusion because the EWR is explicit in the WSIe equation
and can be changed. However, it would not be possible using
Pster's equation where the EWR is implicit and thus, constant. As a
result, there is a chance that the Chao Phraya and Thachin water-
sheds may shift to moderate stress situation. In addition, some
areas facing a water stress especially during the dry season may be
visible in the WSI with respect to EWR. Furthermore, the WSI with
respect to EWR based on a time step of a month could provide a
much better interpretation than a time step of a year.
4. Conclusion
Explicit incorporation of the EWR into the WSI is proposed in
order to have the option to change the EWR based on a local
context. This incorporation leads to not only dening new critical
thresholds of water stress but also reinterpreting the new results of
WSI with reference to EWR.
In the case of Thailand, Pster's method is modied based on the
new water stress thresholds using the country data on precipitation
by maintaining the same function and classication of original WSI.
The EWR is taken into account as a part of water withdrawal based
on Thailand's water allocation plan. Priority for water allocation
plays a signicant role in the context of Thailand. In case of water
competition, the water demand sector having low priority has more
potential to face a water stress situation than the higher priority
sector. Hence incorporation of EWR in water withdrawal allows
different priority setting for different areas. The range of EWR from
1% to 19% that is used in Thailand may not reect the real situation;
therefore, using different percentage factors in relation to seasonal
variation of river ow would be more suitable for estimating the
Please cite this article in press as: Nilsalab, P., et al., Methodology development for including environmental water requirement in the water
stress index considering the case of Thailand, Journal of Cleaner Production (2016), http://dx.doi.org/10.1016/j.jclepro.2016.11.130
annual and monthly EWR.
In summary, the implicit EWR in Pster's method may not al-
ways match with the range of EWR regulated or estimated in each
country due to different functions and services of EWR. Hence, this
can lead to an over or under estimation of a water stress situation.
The methodology proposed in this study can help address this issue
as shown for the case of Thailand. The method is general and can be
used to calculate the WSIe equation for any country using that
country's precipitation data.
Acknowledgements
Financial support by the Joint Graduate School of Energy and
Environment (JGSEE) and the Thailand Research Fund under the
Royal Golden Jubilee Ph.D. program (Grant PHD/0028/2555) is
gratefully acknowledged. This research has been carried out under
the project Research Network for LCA and Policy on Food, Fuel and
Climate Change supported by the National Science and Technology
Development Agency, Thailand.
References
Alcamo, J., Henrichs, T., Rosch, T., 2000. World Water in 2025- Global Modeling and
Scenario Analysis for the World Commission on Water for the 21st Century.
Report No 2, the Kassel World Water Series. Center for Environmental Systems
Research, University of Kassel, Germany.
Alcamo, J., Doll, P., Henrichs, T., Kaspar, F., Lehner, B., Ro sch, T., Siebert, S., 2003.
Global estimates of water withdrawals and availability under current and future
business-as-usual conditions. Hydrol. Sci. 48 (3), 339e348.
rke, M., M
Alcamo, J., Flo arker, M., 2007. Future long-term changes in global water
resources driven by socio-economic and climate changes. Hydrol. Sci. J. Sci.
Hydrol. 52 (2), 247e275.
Acreman, M., Dunbar, M.J., 2004. Dening environmental river ow requirement-a
review. Hydrol. Earth Syst. Sci. 8 (5), 861e876.
Boelee, E., 2011. Ecosystems for Water and Food Security. United Nations Environ-
ment Programme; Colombo: International Water Management Institute, Nai-
robi. http://www.unep.org/pdf/DEPI-ECOSYSTEMS-FOOD-SECUR.pdf.
Boulay, A.-M., Bare, J., Benini, L., Berger, M., Klemmayer, I., Lathuilliere, M., Loubet, P.,
Manzardo, A., Margni, M., Nunez, M., Ridoutt, B., Worbe, S., Pster, S., 2014.
Building consensus on a generic water scarcity indicator for LCA-based water
footprint: preliminary results from WULCA. In: 9th International Conference
LCA of Food San Francisco, USA, October, 8e10.
Davis, J.A., Froend, R.H., Hamilton, D.P., Horwitz, P., McComb, A.J., Oldham, C.E.,
Thomas, D., 2001. Environmental Water Requirements to Maintain Wetlands of
 P. Nilsalab et al. / Journal of Cleaner Production xxx (2016) 1e7
National and International Importance. Environmental Flows Initiative Tech-
nical Report Number 1. Commonwealth of Australia, Canberra.
De Boer, I.J.M., Hoving, I.E., Vellinga, T.V., Van de Ven, G.W.J., Leffelaar, P.A.,
Gerber, P.J., 2013. Assessing environmental impacts associated with freshwater
consumption along the life cycle of animal products: the case of Dutch mill
production in Noord-Brabant. Int. J. Life Cycle Assess. 18, 193e203.
Do ll, P., Fiedler, K., Zhang, J., 2009. Global-scale analysis of river ow alterations due
to water withdrawals and reservoirs. Hydrol. Earth Syst. Sci. 13, 2413e2432.
DOAE, 2014. Drought Report (2003e2007). Department of Agricultural Extension,
Bangkok. http://www.agriinfo.doae.go.th/5year/diaster/46-50/drought46-50.
pdf (Accessed 11 November 2014).
DWR, 2012. Drought Management 2012. Department of Water Resources, Bangkok.
http://www.dwr.go.th/pailang/drought.pdf (Accessed 4 November 2014).
Forslund, A., Reno f
alt, B.M., Barchesi, S., Cross, K., Davidson, S., Farrell, T.,
Korsgaard, L., Krchnak, K., McClain, M., Meijer, K., Smith, M., 2009. Securing
Water for Ecosystems and Human Well-being: the Importance of Environ-
mental Flows. Swedish Water House Report 24. Stockholm International Water
Institute, Sweden.
Gheewala, S.H., Silalertusksa, T., Nilsalab, P., Mungkung, R., Perret, S.R.,
Chaiyawannakarn, N., 2013. Implications of the biofuels policy mandate in
Thailand on water: the case of bioethanol. Bioresour. Technol. 150, 457e465.
Gheewala, S.H., Silalertusksa, T., Nilsalab, P., Mungkung, R., Perret, S.R.,
Chaiyawannakarn, N., 2014. Water footprint and impact of water consumption
for food, feed, fuel crops production in Thailand. Water 6, 1698e1718.
GIZ, 2013. The Ecological Footprint of Cassava and Maize Post-harvest-losses in
Nigeria: a Life Cycle Assessment. the German Federal Ministry for Economic
Cooperation and Development (BMZ), Deutsche Gesellschaft fur Internationale
Zusammenarbeit (GIZ), Bangkok, Thailand.
HAII, 2014. Drought 2013/2014. Hydro and Agro Informatics Institute. http://www.
thaiwater.net/current/drought57/body.html (accessed 30 June 2015).
Hospido, A., Nunez, M., Anton, A., 2013. Irrigation mix: how to include water
sources when assessing freshwater consumption impacts associated to crops.
Int. J. Life Cycle Assess. 18 (4), 881e890.
IHP, 2011. The Impact of Global Change on Water Resources: the Response of
UNESCO's International Hydrological Programme. International Hydrological
Programme. http://unesdoc.unesco.org/images/0019/001922/192216e.pdf
(accessed 21 October 2016).
Jefferies, D., Mun ~ oz, I., Hodges, J., King, V.J., Aldaya, M., Ercin, A.E., Mil
a, i, Canals, L.,
Hoekstra, A.Y., 2012. Water Footprint and Life Cycle Assessment as approaches
to assess potential impacts of products on water consumption. Key learning
points from pilot studies on tea and margarine. J. Clean. Prod. 33, 155e166.
Jeswani, H.K., Azapagic, A., 2011. Water footprint: methodologies and a case study
for assessing the impacts of water use. J. Clean. Prod. 19, 1288e1299.
LDD, 2013. Repetitive Drought Areas in Thailand 2013. Land Development Depart-
ment, Ministry of Agriculture and Cooperatives, Bangkok, Thailand.
Navarro-Ortega, A., Acun ~ a, V., Bellin, A., Burek, P., Cassiani, G., Choukr-Allah, R.,
Dole dec, S., Elosegi, A., Ferrari, F., Ginebreda, A., Grathwohl, P., Jones, C.,
Rault, P.K., Kasper, K., Koundouri, P., Ludwig, R.P., Merz, R., Milacic, R., Mun ~ oz, I.,
Nikulin, G., Paniconi, C., Paunovi c, M., Petrovic, M., Sabater, L., Sabaterb, S.,
Skoulikidis, N.Th., Slob, A., Teutsch, G., Voulvoulis, N., Barcelo , D., 2015. Man-
aging the effects of multiple stressors on aquatic ecosystems under water
scarcity. The GLOBAQUA project. Sci. Total Environ. 503-504, 3e9. http://
dx.doi.org/10.1016/j.scitotenv.2014.06.081.
Mapraneat, V., 2014. Drought conditions and management strategies in Thailand.
In: The 3rd Regional Workshop for Asia-Pacic: Capacity Development to
Support National Drought Management Policies, May 6e9, Hanoi, Viet Nam.
Nunez, M., Pster, S., Anton, A., Munoz, P., Hellweg, S., Koehler, A., Rieradevall, J.,
2012. Assessing the environmental impact of water consumption by energy
crops grown in Spain. J. Ind. Ecol. 17 (1), 90e102.
OAE, 2010. Agricultural Economics Data 2010. Ofce of Agricultural Economics,
Bangkok. http://www.oae.go.th/download/download_Journal/fundamation-
2553.PDF (Accessed 11 July 2013).
ONEP, 2007. Ecosystem Management in the Bang Pakong Basin. Ofce of Natural
Resources and Environmental Policy and Planning, Bangkok. http://www.onep.
go.th/index.php?optioncom_content&viewarticle&id96&Itemid196 Venot, J.-P., Sharma, B.R., Rao, K.V.G.K., 2008. The Lower Krishna Basin Trajectory:
(accessed 13 May 2015).
Page, G., Ridoutt, B., Bellotti, B., 2011. Fresh tomato production for the Sydney
market: an evaluation of options to reduce freshwater scarcity from agricultural
water use. Agric. Water Manag. 100 (1), 18e24.
Page, G., Ridoutt, B.G., Bellotti, B., 2012. Carbon and water footprint tradeoffs in
fresh tomato production. J. Clean. Prod. 32, 219e226.
Please cite this article in press as: Nilsalab, P., et al., Methodology development for including environmental water requirement in the water
stress index considering the case of Thailand, Journal of Cleaner Production (2016), http://dx.doi.org/10.1016/j.jclepro.2016.11.130
7
Pastor, A.V., Ludwig, F., Biemans, H., Hoff, H., Kabat, P., 2014. Accounting for envi-
ronmental ow requirements in global water assessment. Hydrol. Earth Syst.
Sci. 18, 5041e5059. http://dx.doi.org/10.5194/hess-18-5041-2014.
Pster, S., Koehler, A., Hellweg, S., 2009. Assessing the environmental impacts of
freshwater consumption in LCA. Environ. Sci. Technol. 43 (11), 4098e4104.
Pster, S., Bayer, P., Koehler, A., Hellweg, S., 2011. Environmental impacts of water
use in global crop production: hotspots and trade-offs with land use. Environ.
Sci. Technol. 45, 5761e5768.
Pster, S., Bare, J., Benini, L., Berger, M., Bulle, C., Lathuilliere, M., Manzardo, A.,
Margni, M., Motoshita, M., Nunez, M., Pastor, A., Ridoutt, B., Worbe, S.,
Boulay, A.-M., 2015. Outcome of WULCA harmonization activities: recom-
mended characterization factors for water footprinting. In: International Con-
ference on Life Cycle Assessment as Reference Methodology for Assessing
Supply Chains and Supporting Global Sustainability Challenges, LCA for
Feeding the Planet and Energy for Life, Stresa, Italy, October 7.
Pimentel, D., Berger, B., Filiberto, D., Newton, M., Wolfe, B., Karabinakis, E., Clark, S.,
Poon, E., Abbett, E., Nandaopal, S., 2004. Water Resources, Agriculture, and the
Environment. New York State College of Agriculture and Life Sciences, Cornell
University, Ithaca (NY). Report 04e1.
Quinteiro, P., Cla udia Dias, A., Pina, L., Neto, B., Ridoutt, B.G., Arroja, L., 2014.
Addressing the freshwater use of a Portuguese wine (vinho verde) using
different LCA methods. J. Clean. Prod. 68, 46e55.
RID, 2011a. Work Manual No.8/16: Assessing Water Consumption by Sectors [in
Thai]. Royal Irrigation Department. http://ridceo.rid.go.th/buriram/download/
manual-08.pdf (accessed 17 October 2012).
RID, 2011b. Flow Duration Curve. Irrigation Water Division, Ofce of Hydrology and
Water Management, Royal Irrigation Department.
RID, 2014a. Annual Report 2014. Royal Irrigation Department. http://www.rid.go.th/
2009/_data/docs/AnnualRID57.pdf (accessed 13 May 2015).
RID, 2014b. Water Allocation Plan for 2014/2015. Royal Irrigation Department.
http://water.rid.go.th/hydhome/document/2557/Plan_2557-2558.pdf (accessed
13 May 2015).
RID, 2015. RID Response Plan for Tidal Bore to Prevent Seawater Intrusion and
Maintain Raw Water Quality for Tap Water. Royal Irrigation Department. http://
water.rid.go.th/news/news_58_032.htm (accessed 13 May 2015).
Ridoutt, B.G., Pster, S., 2013. A new water footprint calculation method integrating
consumptive and degradative water use into a single stand-alone weighted
indicator. Int. J. Life Cycle Assess. 8 (1), 204e207. http://dx.doi.org/10.1007/
s11367-012-0458-z.
Royal Irrigation Department Ofce 1, 2014. Prevention and Mitigation Plan for
Drought 2014/2015: Regional Irrigation Ofce 1. http://irrigation.rid.go.th/rid1/
waterrid2/data/om-rid1-19-11-2014.pdf (accessed 13 May 2015).
Smakhtin, V.U., 2002. Environmental Water Needs and Impacts of Irrigated Agri-
culture in River Basins: a Framework for a New Research Program. Working
Paper 42. International Water Management Institute, Colombo, Sri Lanka.
Smakhtin, V., Revenga, C., Do ll, P., 2004. Taking into Account Environmental Water
Requirements in Global-scale Water Resources Assessments. Comprehensive
Assessment Research Report 2. Comprehensive Assessment Secretariat,
Colombo, Sri Lanka.
Stoessel, F., Juraske, R., Pster, S., Hellweg, S., 2012. Life cycle inventory and carbon
and water foodprint of fruits and vegetables: application to a Swiss retailer.
Environ. Sci. Technol. 46 (6), 3253e3262.
Teerawiroon, S., 2015. Sustainability Resolutions on Drought. The Secretariat of the
House of Representatives. http://library2.parliament.go.th/ebook/content-
issue/2558/hi2558-027.pdf (accessed 21 October 2016).
TMD, 2010. El Nin ~ o Event: 2009e2010. Thai Meteorological Department, Bangkok.
http://tmd.go.th/programs%5Cuploads%5Cphenomena%5CSep-09.pdf (Accessed
5 November 2014).
Uche, J., Martnez-Gracia, A., Carmona, U., 2014. Life cycle assessment of the supply
and use of water in the Segura Basin. Int. J. Life Cycle Assess. 19 (3), 688e704.
http://dx.doi.org/10.1007/s11367-013-0677-y.
Uche, J., Martnez-Gracia, A., Crez, F., Carmona, U., 2015. Environmental impact of
water supply and water use in a Mediterranean water stressed region. J. Clean.
Prod. 88, 196e204.
Relationships between Basin Development and Downstream Environmental
Degradation. International Water Management Institute, Colombo, Sri Lanka,
p. 36 (IWMI Research Report 125).
Zonderland-Thomassen, M.A., Ledgrad, S.F., 2012. Water footprinting-A comparison
of methods using New Zealand dairy farming as a case study. Agric. Syst. 110,
30e40.