See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/276080697

Sustainability index for water resources planning and management. J Water

Resour Plan Manag Am Soc Civil Eng

Article  in  Journal of Water Resources Planning and Management · September 2011

DOI: 10.1061/(ASCE)WR.1943-5452.0000134

CITATIONS READS

223 1,474

3 authors:

Samuel Sandoval-Solis Daene C. Mckinney

University of California, Davis University of Texas at Austin
101 PUBLICATIONS   1,146 CITATIONS    168 PUBLICATIONS   5,572 CITATIONS   

SEE PROFILE SEE PROFILE

Pete Loucks
Cornell University
143 PUBLICATIONS   8,828 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

WATER RESOURCES SUSTAINABILITY INDEX FOR A WATER-STRESSED BASIN IN BRAZIL View project

Groundwater Management and Remediation View project

All content following this page was uploaded by Daene C. Mckinney on 15 April 2016.

The user has requested enhancement of the downloaded file.

 Sustainability Index for Water Resources
Planning and Management
S. Sandoval-Solis1; D. C. McKinney, M.ASCE2; and D. P. Loucks, M.ASCE3

Abstract: This paper presents a water resources sustainability index that makes it possible to evaluate and compare different water
management policies with respect to their sustainability. The sustainability index identifies policies that preserve or improve the desired
water management characteristics of the basin in the future. This index is based on a previous sustainability index with improvements
in its structure, scale, and content to make it more flexible and adjustable to the requirements of each water user, type of use, and basin.
The Rio Grande transboundary basin is used as a case study demonstrating the use of the index. Tailor-made sustainability indexes are defined
for water users in Mexico, the United States, the environment, and for meeting system requirements (international treaty obligations). Group
sustainability indexes are calculated to summarize the results for groups of water users of each country, the environment, and the basin as a
whole. Sustainability indexes by subbasins are calculated to identify areas of potential improvement and regions at risk. DOI: 10.1061/
(ASCE)WR.1943-5452.0000134. © 2011 American Society of Civil Engineers.
CE Database subject headings: Sustainable development; Water resources; Rio Grande; Water management.
Author keywords: Sustainability index; Sustainable policies; Adaptive capacity; Water resources; Rio Grande.

Introduction environment, now and in the future. To accomplish this goal, it

is necessary to have performance measures or indexes that allow
It has been 30 years since the concept of sustainable development the evaluation and comparison of water resources systems under
was introduced for the first time by the World Conservation Strat- different scenarios. The objective of this paper is to present a water
egy [International Union for Conservation of Nature (IUCN) 1980]. resources sustainability index that makes it possible to evaluate and
Sustainable development balances the exploitation of natural re- compare alternative management policies for water resources
sources, technology development, and institutional change to en- systems.
hance the potential to meet human needs and aspirations, now The sustainability index (SI) summarizes the performance of
and in the future [World Commission on Environment and Devel- alternative policies from the perspective of water users and the envi-
opment (WCED) 1987]. To achieve sustainability, all the compo- ronment; it is also a measure of a system’s adaptive capacity to
nents in the system must be also in balance. Loucks (1997) defined
reduce its vulnerability. If a proposed policy makes the system
sustainable water resources systems as “those systems designed and
more sustainable, the index will show that the system will have
managed to contribute fully to the objectives of society, now and in
a larger adaptive capacity. The index proposed here is a variation
the future, while maintaining their ecological, environmental and
of the sustainability index developed by Loucks (1997) with im-
hydrological integrity.” Although this concept is still valid, water
provements in its structure, scale, and content to make it more flex-
management policies that promote sustainable water resources sys-
ible and adjustable to the requirements of each basin. The SI is an
tems are becoming more difficult to identify because of environ-
mental considerations, water scarcity, and climate change. integration of performance criteria that capture the essential and
Recently, strong emphasis has been placed on the adaptive desired sustainable characteristics of the basin. The index facilitates
capacity of water resource systems, which refers to measures that comparison of policies when there are trade-offs among perfor-
reduce the vulnerability of systems to actual or expected future mance criteria.
changes. Vulnerability is the magnitude of an adverse impact on First, the performance criteria parameters used in the SI are
a system. Thus, the objective is to look for policies that reduce described. Second, the SI for individuals and water user groups
the adverse impacts of actual and expected events, and to the ex- is defined. Third, water management in the Rio Grande basin, used
tent possible, meet the water requirements for humans and the as a case study, is presented. Fourth, the SI for the current water
management policy and three adaptation policies are defined for
1
Dept. of Land, Air and Water Resources, Univ. of California, Davis, different groups of stakeholders in the Rio Grande basin. Finally,
CA 95616-8627; formerly, Center for Research in Water Resources, Univ. conclusions and recommendations are presented.
of Texas, Austin, TX 78712. E-mail: [email protected]
2
Center for Research in Water Resources, Univ. of Texas, Austin, TX
78712 (corresponding author). E-mail: [email protected] Performance Criteria
3
School of Civil and Environmental Engineering, Cornell Univ., Ithaca,
NY 14853. E-mail: [email protected] Performance criteria are used to evaluate water management pol-
Note. This manuscript was submitted on March 27, 2010; approved on
icies and enable the comparison of alternative policies. Perfor-
October 21, 2010; published online on October 26, 2010. Discussion period
open until February 1, 2012; separate discussions must be submitted for mance criteria can be simple averages, such as system storage,
individual papers. This paper is part of the Journal of Water Resources water supply, evaporation, municipal shortfalls (average deficits),
Planning and Management, Vol. 137, No. 5, September 1, 2011. ©ASCE, and outflow of water from a system (Vigerstol 2002). Probability-
ISSN 0733-9496/2011/5-381–390/$25.00. based performance criteria include time-based (annual, monthly)

JOURNAL OF WATER RESOURCES PLANNING AND MANAGEMENT © ASCE / SEPTEMBER/OCTOBER 2011 / 381
and volumetric reliability [Texas Commission on Environmental Standard Deviation
Quality (TCEQ) 2007], and resilience (Hashimoto et al. 1982).
The standard deviation of the water supply for the ith water user in
period t is
Reliability
rffihffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffii.
ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
Water demand reliability is the probability that the available water Pt¼n i
ðX " X!i Þ 2 ðn " 1Þ
t¼1 Supplied;t Supplied
supply meets the water demand during the period of simulation
σi ¼ ð5Þ
(Klemes et al. 1981; Hashimoto et al. 1982). For each time period Water demandi
t, deficits Dit are positive when the water demand X iTarget;t is more !i
where the average water supply X Supplied is
than the water supplied X iSupplied;t for the ith water user; if the water
t¼n
X
supplied is equal to water demand (X iSupplied;t ¼ X iTarget;t ), deficits !i 1
X Supplied ¼ X iSupplied;t ð6Þ
are zero (Dit ¼ 0) (Loucks 1997). n t¼1

8 This performance criterion (Hirsch 1979; Cai et al. 2002) indi-

< X iTarget;t " X iSupplied;t if X iTarget;t > X iSupplied;t
Dit ¼ ð1Þ cates the variability of the water supply when part or all of a user’s
:0 if X iTarget ¼ X iSupplied;t water demand is not supplied from controlled facilities, such as
unregulated rivers. A dimensionless standard deviation has been
defined in Eq. (5) by dividing the volumetric standard deviation
Time-based reliability Reli is considered, which is the portion of by the annual water demand, water demandi .
time that the water demand is fully supplied, the number of times
Dit ¼ 0, with respect to the number of time intervals considered Maximum Deficit
(e.g., n months or years) (McMahon et al. 2006): The maximum deficit, if deficits occur, is the worst-case annual
deficit, maxðDiAnnual Þ, for the ith water user (Moy et al. 1986).
No: of times Dit ¼ 0 A dimensionless maximum deficit is calculated by dividing
Reli ¼ ð2Þ
n the maximum annual deficit by the annual water demand,
water demandi :

Resilience maxðDiAnnual Þ
Max def i ¼ ð7Þ
Water demandi
Resilience is a system’s capacity to adapt to changing conditions
[World Health Organization (WHO) 2009]. Because climate con-
ditions are no longer steady, resilience must be considered as a sta-
tistic that assesses the flexibility of water management policies to Sustainability Index
adapt to changing conditions. According to Hashimoto et al.
(1982), resilience is the probability that a system recovers from Indexes represent aggregate measures of a combination of perfor-
a period of failure. Moy et al. (1986) used the maximum number mance measures, or in other words, an index is a “synthesis of nu-
of consecutive deficit periods prior to recovery as an alternative merous factors into one given factor” (Sainz 1989). Several indexes
definition of resilience. Resilience Resi is the probability that a suc- have been developed for environmental processes, such as the envi-
cessful period follows a failure period (the number of times Dit ¼ 0 ronmental index (Milbrink 1983), environmental stresses index
follows Dit > 0) for all failure periods (the number of times Dit > 0 (Reiquam 1972), environmental sustainability index (Esty et al.
occurred). This statistic assesses the recovery of the system once it 2005), the multiattributed environmental index (Hajkowicz 2006),
has failed: and also some indexes specifically for water resources, such as the
drought risk index (Zongxue et al. 1998), the Palmer drought
No: of times Dit ¼ 0 follows Dit > 0 severity index (Palmer 1965), water quality index (Brown et al.
Resi ¼ ð3Þ 1972), fairness (Lence et al. 1977), reversibility (Fanai and Burn
No: of times Dit > 0 occurred 1997), and consensus (Simonovic 1998).
To quantify the sustainability of water resources systems,
Loucks (1997) proposed the SI, with the objective to facilitate
Vulnerability the evaluation and comparison of water management policies.
Vulnerability is the likely value of deficits, if they occur The SI is a summary index that measures the sustainability of water
(Hashimoto et al. 1982). Essentially, vulnerability expresses the se- resources systems; it can be used to estimate the sustainability for
verity of failures. Vulnerability can be expressed as (1) the average water users and to obtain the change in sustainability by comparing
failure (Loucks and van Beek 2005); (2) the average of maximum the index among several water policies proposed. Frequently, in-
shortfalls over all continuous failure periods (Hashimoto et al. dexes are criticized because they are seen as a sum of disparate
1982; McMahon et al. 2006); and (3) the probability of exceeding items (Hopkins 1991), and sometimes in practice, people in the
a certain deficit threshold (Mendoza et al. 1997). This paper uses water sector are reluctant to use indexes (Brown et al. 1972).
the first approach, the expected value of deficits, which is the sum The SI summarizes essential performance parameters of water man-
of the deficits, Dit , divided by the deficit period, the number of times agement in a meaningful manner, rather than adding broad factors,
Dit > 0 occurred. Dimensionless vulnerability is calculated by di- and the SI has been used by the scientific community (Ray et al.
viding the average annual deficit by the annual water demand, 2010; McMahon et al. 2006; Loucks 1997).
water demandi, for the ith water user:
Sustainability by User
!P ".
t¼n
Dit No: of times Dit > 0 occurred Loucks (1997) proposed the following SI for the ith water user:
t¼0
i
Vul ¼ ð4Þ SIi ¼ Reli % Resi % ð1 " Vuli Þ ð8Þ
Water demandi

382 / JOURNAL OF WATER RESOURCES PLANNING AND MANAGEMENT © ASCE / SEPTEMBER/OCTOBER 2011
 The SI has the following properties: (1) its values vary from 0–1; Sustainability by Group
(2) if one of the performance criteria is zero, the sustainability will
To compare groups of water users, the sustainability by group (SG)
be zero also; and (3) there is an implicit weighting because the in-
was defined as a weighted average of sustainability indexes
dex gives added weight to the criteria with the worst performance.
(Loucks 1997). The SG is used to calculate the sustainabi-
The multiplicative form of the SI considers each criterion as essen-
lity for a group k with ith to jth water users belonging to this
tial and nonsubstitutable. Sagar and Najanm (1998) suggested this
group:
as the proper manner for integrating performance criteria. For in-
stance, Reiquam (1972) used the multiplicative form for the envi- i¼j∈k
X
ronmental stresses index. SGk ¼ W i % SIi ð11Þ
A variation of Loucks’ SI is proposed here, with the index i¼1∈k
defined as a geometric average of M performance criteria (C im )
where W i = relative weight for the ith water user, ranging from 0–1
for the ith water user:
and summing to one. If the SI of each user is weighted by its annual
"
M
#1=M water demand, the SG for the kth group is expressed as
Y
i
SI ¼ C im ð9Þ i¼j∈k
X Water demandi
m¼1 SGk ¼ % SIi ð12Þ
i¼1∈k
Water demandk
For instance, if the performance criteria are C i1 ¼ Reli ,
C i2 ¼ Resi , and C i3 ¼ 1 " Vuli , the SI for the ith water user is where
i¼j∈k
X
i i i i 1=3 Water demandk ¼ Water demandi ð13Þ
SI ¼ ½Rel % Res % ð1 " Vul Þ' ð10Þ
i¼1∈k

This index satisfies the properties of the SI defined by Loucks The relative importance of each variable is reflected in the
(1997), but, in addition, it has the following improvements: weights. There are many weighting options, such as (1) an arithmetic
Content—It allows the inclusion of other criteria of interest ac- average or equal-attribute-based weighting system (Slottje 1991;
cording to the necessities of each case. The SI is no longer a fixed Reiquam 1972); (2) explicit weights obtained through (a) utility
performance criteria related to water quantity; performance criteria theory analysis (Loucks et al. 1997; Von Neumann and Morgenstern
of water quality and environmental performance might be included 1974), principal components analysis, or hedonic model according
in the SI. For instance, if the total dissolved solids (TDS) of the to regression coefficients (Slottje 1991); or (b) weights defined by
water delivered to a user must be below a permitted value, the reli- consultations with experts (Gwartney et al. 1996), decision makers
ability for TDS not exceeding the desired threshold can be calcu- (Vigerstol 2002), or researcher expertise (Giorgi and Mearns 2002).
lated and included in the SI. The criteria (C im ) included in Eq. (9) Weights of Eq. (12) obtained through the annual water demand are
must have a scale from 0–1, and desirable criteria values tend to 1. used in this paper, considering that (1) the necessities of the water
Scaling and complements 1-C im can be applied prior to including users and the environment can be expressed in the water demand
any performance criteria into Eq. (10). value; (2) in interviews, authorities and water users agreed with this
Scaling—The use of the geometric average scales the values of formulation; and (3) other performance criteria of interest are func-
the SI, generating numbers that can be more practical to interpret tions of water demand value, and can be scaled (normalized) using it,
and communicate. If a certain water user has a reliability, resilience, i.e., vulnerability, maximum deficit and standard deviation. The
and vulnerability of 50% for each performance criterion, then the SI authors considered that the water necessities for water users and
calculated with the prior definition [Eq. (8)] and the proposed index the environment are expressed in their water demand. However,
[Eq. (9)] are 13% and 50%, respectively. The scaling of the SI does there are limitations when water demands have not yet been esti-
not obscure poor performance; its only purpose is to scale the val- mated, e.g., for the environment, or when the water demand provided
ues and make the index more practical and intuitive. In addition, by the authorities underestimates the water necessities for water
more than three parameters can be included in the SI; the product of users and the environment.
several factors will result in small numbers, and without scaling,
changes in the SI might be difficult to discern.
Flexibility—Several structures for the SI might be applied in the Rio Grande Basin
same basin for different groups of water users or types of use. For
instance, SI for municipal or recreational water use may include The Rio Grande basin is a transboundary basin between the United
different performance criteria than the SI for agriculture water States and Mexico [Fig. 1(a)]. Because of its geographical position,
use. Water quality and environmental performance criteria may it is one of the most stressed basins in the world [World Wildlife
be included for municipal and recreational water use, respectively, Fund (WWF) 2007], not only because of the increase in water
and the standard SI [Eq. (10)] might be appropriate for agriculture demand as a result of population and industry growth, but also
use. Sustainability does not mean the same thing for all water users, because of the natural water scarcity in the region. Extended
and the proposed index allows it to be adjusted to suit the user or periods of drought (> 10 years), coupled with overallocation of
use of water. water rights, low efficiency in irrigation systems, and international
The improvements to the SI are not merely mathematical. The agreements, make the Rio Grande basin a highly complex water
updated SI is a holistic approach to define the sustainability for resources system.
each group of water user. The structure of the index incorporates
Water Management Principles for the Rio Grande
tailor-made parameters that for some water users may be crucial to
Basin
their water management; the scaling of the index allows a more
intuitive result; and the flexibility to use different SI structures The Rio Grande basin is used to exemplify the proposed SI. The
in the same system allows the meaningful discrimination of perfor- middle and lower part of the basin is analyzed from Elephant
mance parameters for specific groups of water users. Butte dam in New Mexico to the mouth at the Gulf of Mexico

JOURNAL OF WATER RESOURCES PLANNING AND MANAGEMENT © ASCE / SEPTEMBER/OCTOBER 2011 / 383
 Fig. 1. (a) Rio Grande basin; (b) hydrography of Rio Grande basin; (c) environmental control points

[Fig. 1(b)]. Water management of the basin results from four hurricane season contribute to the delivery of treaty obligations.
aspects: (1) international agreements; (2) Mexican water policies; Because of the uncontrolled nature of the treaty deliveries, the
(3) U.S. water policies; and (4) the environment. standard deviation criterion is included in the SI to assess treaty
obligations and help identify adaptation policies that reduce the
International Agreements: Treaty of 1944 variability of deliveries, providing a more steady delivery of treaty
The 1944 treaty between United States and Mexico specifies the water by increasing low flows during drought periods and reducing
water allocation for both countries [International Boundary and spills during wet periods. The standard deviation for the treaty ob-
Water Commission (IBWC) 1944] with a primary division of ligations is calculated from the annual deliveries of the six Mexican
six tributaries originating in Mexico as one-third to the United tributaries.
States and two-thirds to Mexico. The third shall not be less than
Mexican Water Policy
431:721 million m3 =year as an average over cycles of five consecu-
tive years. Two international dams (Amistad and Falcon) are used Mexican water demands are characterized by use [Comisión
to store and manage the water for both countries, and each country Nacional del Agua (CONAGUA) 2004a]. For this research, munici-
has its own storage account in each reservoir. The treaty cycles can pal, domestic, and agricultural water users are considered, account-
expire in less than five years if the account of U.S. storage in both ing for the 99.2% of the total Mexican water demand (CONAGUA
dams is filled with water. At the end of a 5-year cycle, the IBWC 2004b). Municipal and domestic users have the highest priority,
evaluates the Mexican delivery of water to the U.S. and determines and two times their annual water demand must be stored in the res-
if the treaty obligations have been met. If there is a deficit in the ervoirs. Water allocations to agricultural users are not guaranteed,
treaty delivery, it must be paid in the following cycle (IBWC 1944). and their allocations depend on the available storage in the respec-
The sustainability index proposed for the treaty obligations is tive dams that supply them. Each October, CONAGUA (the water
authority in Mexico) determines the available reservoir storage
SITreaty ¼ ½RelTreaty % ResTreaty % ð1 " VulTreaty Þ % ð1 " σTreaty Þ'1=4 after deducting municipal allocations, evaporation, and operation
ð14Þ losses (Collado 2002). Then, a negotiation between CONAGUA
and the irrigation districts sets the agricultural water allocation
Four out of the six Mexican tributaries delivering water to the for the coming water year.
treaty are unregulated rivers (Arroyo Las Vacas, San Diego, San The sustainability index proposed for Mexican water users is
Rodrigo, and Escondido). In addition, there is no defined policy
in the other two regulated rivers (Rio Conchos and Salado) to SIMXi ¼ ½RelMXi % ResMXi % ð1 " VulMXi Þ % ð1 " Max def MXi Þ'1=4
deliver water to meet treaty obligations. In practice, only the gains ð15Þ
of the reach between the most downstream reservoir in each tri-
butary and the Rio Grande confluence are left in the river to meet The Rio Grande is a naturally water-scarce basin; extended and
the treaty obligations. Sporadically, reservoir spills during the severe periods of drought have occurred in the basin. During the

384 / JOURNAL OF WATER RESOURCES PLANNING AND MANAGEMENT © ASCE / SEPTEMBER/OCTOBER 2011
latest drought (1994–2003), Mexico was not able to deliver the SIEnvi ¼ ½RelEnvi % ResEnvi % ð1 " VulEnvi Þ % ð1 " Max def Envi Þ'1=4
treaty water to the United States in two consecutive cycles of ð17Þ
the 1944 treaty: Cycle 25 (1992–1997) and Cycle 26 (1997–2002).
To cover these deficits, extraordinary measures were taken by the
authorities, such as stopping the supply for some Mexican irriga- Simulation Model of the Rio Grande Basin
tions districts and transferring Mexican storage in the international
To illustrate the use of the new SI, several scenarios of water man-
reservoirs to the United States These decisions severely affected
agement in the Rio Grande basin are evaluated. Water resource
Mexican agriculture water users in the basin, almost extinguishing
allocation in the Rio Grande basin has been simulated using the
this activity in the lower part of the basin. Because of this, the
Water Evaluation and Planning System (WEAP) software (Danner
maximum deficit criterion is included in the SI for Mexican water
et al. 2006). The allocation logic represented in the model follows
users.
the allocation of water for Mexico (CONAGUA 2004a), Texas
U.S. Water Policy (TCEQ 2006), and the international allocation of water established
in the Convention of 1906 (IBWC 1906) and the treaty of 1944
The Texas Rio Grande water master program, represented by the (IBWC 1944). Data for naturalized flows, conveyance losses, res-
TCEQ (the water authority in Texas). regulates the U.S. water di- ervoir capacities, and evaporation, among other variables, were
version from the Amistad reservoir to the Gulf of Mexico (TCEQ provided by CONAGUA, TCEQ, and the IBWC (Danner et al.
2005) based on to the U.S. storage provided by the IBWC. Each 2006). For the United States, 100% water demand is taken as
user has an account, and water is allocated (TCEQ 2006) based on the full allocation water right established by the TCEQ in the Water
the water use (irrigation, municipal, mining, industrial, and other) Availability program (TCEQ 2006). For Mexico, 100% demand is
and the type of water right (Type A or B). Municipal and industrial taken as the volume declared by CONAGUA in 2004 (CONAGUA
users have the highest priority, and they are guaranteed an amount 2004b). Table 1 shows the water demands for each country.
for each year. Allocations to the other users are not guaranteed and Monthly use coefficients are used to account for the seasonal
depend on the water remaining in their accounts. variability for each demand. The period of analysis for the model-
The sustainability index proposed for U.S. water users is ing is 60 years, using as input the naturalized streamflows from
October 1940 to September 2000. The Rio Grande model has been
SIUSi ¼ ½RelUSi % ResUSi % ð1 " VulUSi Þ % ð1 " Max def USi Þ'1=4
calibrated and validated using a 24-year period (1976–2000) over
ð16Þ which both international reservoirs (Amistad and Falcon) were in
existence; in addition, the historic records of water diversions were
Similar to Mexico, agricultural water users in the United States available for this period. The simulation process considered the rep-
suffered shortages during the last drought, and so the maximum etition of the 60-year hydrologic period with the recent infrastruc-
deficit criterion is also included in the SI for U.S. water users. ture and demands in the basin.
Environment
Environmental flows have not been considered an integral part of Sustainability Index Use
water management in the Rio Grande. Important environmental
habitats, such as the Big Bend State and National Park in the United In complex, stressed, and shared water resources systems, such as
States, the northern Chihuahuan desert, Maderas del Carmen, the Rio Grande, it can be difficult to identify policies that improve
Ocampo, and Cañon de Santa Elena natural reserves in Mexico, water management. This section illustrates how the SI and SG can
are ecologically threatened because of the lack of environmental help identify which policies improve water management, for
water management policies. Historically, the basin has been manip- whom, where, and by how much. The SI and SG are comprehen-
ulated in an exclusive human water resource management mode sive tools integrating multiple performance measures that facilitate
(Enriquez-Coyro 1976) without consideration for the environmen- the evaluation and comparison of different water management
tal needs of the native ecosystems. policies.
Several efforts have been undertaken to determine environmen-
By Water User
tal flows needed in the basin (Sandoval-Solis and McKinney 2009).
As part of an environmental flow assessment for the Rio Conchos, To demonstrate the use of the proposed SI, two scenarios are com-
environmental flows were estimated at nine locations [Fig. 1(c)] pared systematically: a baseline scenario that represents current
(WWF 2006). A monthly variation for two conditions, maintenance water management policies in the basin and an alternative scenario.
or drought, was determined for each location. These flows were The alternative scenarios represent policies that improve the effi-
used to evaluate the performance of the environmental require- ciency of the system through water conservation measures, policies
ments. The sustainability index proposed for the environmental whose objectives are to reduce the use and/or consumption of
flows is water. In this section, alternative scenarios are analyzed in which

Table 1. Water Demands Considered in the Rio Grande WEAP Model

Water use Mexico number of demands Mexico demand (million m3 =year) U.S. number of demands U.S. demand (million m3 =year)
Municipal 21 731 23 283a
Irrigation 39 3,881 53 3,034a
Other 1 47 20 11a
Groundwater 35 1,852 21 2,840b
Total 96 6,511 120 6,168
a
Full allocation demand for U.S. water demands. The current allocation is 70% of the full allocation.
b
This value represents an upper bound on aquifer withdrawal by these water demands.

JOURNAL OF WATER RESOURCES PLANNING AND MANAGEMENT © ASCE / SEPTEMBER/OCTOBER 2011 / 385
water demand is reduced below the baseline demand because of farmers without any income for at least 1 year. Adaptive policies
water conservation measures. that promote conjunctive use of surface and groundwater, such as
In the first alternative, Scenario A, water conservation measures the groundwater banking proposed by Sandoval-Solis et al. (2011),
are implemented in irrigation district 005 Delicias (DR-005), may reduce the risk of high deficits in DR-005.
the biggest water user on the Mexican side with a demand of Because there is no policy to allocate water to the environment,
942 million m3 =year. In Scenario A, the water demand is progres- a hypothetical case in which the water saved in Scenario A is used
sively reduced from 100% to 20% relative to the baseline scenario to meet the environmental needs for the Rio Conchos subbasin is
demand. Fig. 2 shows the results for DR-005 according to Eq. (15). considered. For purposes of brevity, only the results for control
Results show that as demand is reduced, (1) reliability increases; point VMc Camargo are presented because this point has the worst
(2) vulnerability decreases; (3) resilience increases; (4) maximum performance in the baseline scenario. Fig. 3 shows the results ac-
deficit does not decrease; and (5) SI increases after a reduction cording to Eq. (17). Results show that as water demand is reduced,
to 70%. (1) reliability and resilience increase; (2) vulnerability and the
Are there any benefits in Scenario A? If so, are they immediate maximum deficit decrease; and (3) environmental sustainability
when the water demand is reduced or are they delayed? By how for this control point improves significantly.
much? The performance criteria do not allow provide answers In Scenario A, when the water demand of DR-005 is reduced
to these questions, but the SI does. For instance, for a 50% reduc- from 100% to 90% and the water savings are used for environmen-
tion in DR-005 water demand, results of reliability, resilience, vul- tal purposes, the SI for the environment grows 33%, from 24% to
nerability, maximum deficit, and SI are 83%, 40%, 51%, 98%, 57%. The SI becomes steady at 60%, meaning that Scenario A will
and 22%, respectively. In contrast, for a 40% reduction in DR-005 be effective up to a 40% reduction in DR-005 demand, after this, no
water demand, results for the same performance criteria are 78%, environmental benefits will be gained with this policy, and other
23%, 45%, 96%, and 24%, respectively. Using the performance adaptive policies should be used to further improve the environ-
criteria it is difficult to discern if the 10% water demand reduction mental conditions. Under the baseline scenario, low reliability
improved the water management; however, the SI shows an in- and resilience and high vulnerability and maximum deficit are ex-
crease of 2%. In addition, the water supply for DR-005 is not sus- pected for the environment (100% demand); thus, under the current
tainable; one of the characteristics required for its sustainability policies, environmental sustainability is threatened.
[Eq. (15)] is a reduction in the maximum deficit, and in both Scenario B evaluates the water demand reduction of the user
scenarios it is almost 100%. Although the reliability, resilience, “Water Master Sections 8–13 Agriculture A” in Texas (WMS),
and vulnerability improve, the demand reduction proposed in Sce- the largest water user group on the U.S. side with a demand of
nario A does not solve the problem of high maximum deficit. In 1;801 million m3 =year. In Scenario B, WMS demand is reduced
Scenario A, there is almost no improvement until the demand is progressively from 100% to 40%. Fig. 4 shows the results accord-
reduced to 70%; after this point, the water supply starts improving. ing to Eq. (16). As water demand is reduced, (1) reliability in-
Two water conservation measures have already been imple- creases; (2) resilience does not change until demand reaches a
mented in the Rio Grande basin, specifically in DR-005: (1) the 50% reduction, after this point it increases quickly; (3) vulnerability
permanent buy-back of water rights through the Mexican Programa and the maximum deficit decrease; and (4) sustainability improves.
de Adquisicion de Derechos de Uso del Agua (PADUA) program The SI shows that Scenario B is beneficial for WMS.
[Secretaria de Agricultura, Ganadería, Desarrollo Rural, Pesca y During the most recent drought (1994–2004), the water supply
Alimentación (SAGARPA) 2003]; and (2) improvements in the in- for the United States was compromised. The first three years of
frastructure to reduce conveyance losses and increase application the drought (1994–1996), the water supply for WMS was 78%
efficiency through the 1944 treaty, Minute 309 (IBWC 2003). The (1;400 million m3 =year) on average; for the rest of the drought
result of both programs has been a savings of 366 million m3 =year (1997–2004), the water supply was 53% (950 million m3 =year)
(39%): 10% from PADUA and 29% from Minute 309. Even though on average of the full allocation demand. This uncertainty in the
water demand has been reduced to 61% (575 million m3 =year); the water supply provoked the Texas legislature to order a study
maximum deficit problem is not solved. The risk of experiencing a (Brandes 2004) that defined the water availability and the water
high deficit (Max def ¼ 99%) is still imminent; this risk may leave use limits and vulnerabilities of the system. As a result, the current

100% 100%
Max. Deficit Reliability
Performance Criteria and

Performance Criteria and

80% 80% Sustainability Index

Sustainability (%)

Sustainability (%)

60% 60%
Vulnerability
Max. Deficit
40% Resilience 40%

20% Sustainabili 20%

ty Index Vulnerability
0% 0%
20% 30% 40% 50% 60% 70% 80% 90% 100% 20% 30% 40% 50% 60% 70% 80% 90% 100%
Water Demand DR-005 (%) Water Demand DR-005 (%)
Reliability (%) Vulnerability (%) Resilience (%) Reliability (%) Vulnerability (%) Resilience (%)
Max. Deficit (%) Sustainability (%) Max. Deficit (%) Sustainability (%)

Fig. 2. Performance criteria and sustainability index values for DR- Fig. 3. Performance criteria and sustainability index values for the en-
005, Scenario A vironmental control point VMc Camargo, Scenario A

386 / JOURNAL OF WATER RESOURCES PLANNING AND MANAGEMENT © ASCE / SEPTEMBER/OCTOBER 2011
 100%
is filled. In Scenario B, the WMS demand is progressively reduced;
Performance Criteria and therefore, less water is called for from the reservoirs, and as a result,
80% ficit
Sustainability (%) Max. De the U.S. storage capacity in the international reservoirs is filled
more frequently. Thus, the period of time the treaty obligations
60%
are met (reliability) is greater than the baseline scenario, and if
Sustainabil
a deficit happens, the system recovers faster because it is more
40% ity Index likely that the deficit can be made up with delivery from the six
tributaries or by filling the U.S. storage capacity. The SI shows that
y
Vulnerabilit
20%
Resilience the treaty obligation improves under Scenario B. These results are
important because they show that fulfilling the treaty obligations is
0%
40% 50% 60% 70% 80% 90% 100%
not only a function of the water delivered by Mexico, but also of the
water demand in the U.S.
Water Demand WMS 8-13 (%)
Reliability (%) Vulnerability (%) Resilience (%) Sustainability by Group
Max. Deficit (%) Sustainability (%)
Each water user has a unique SI that depends on the structure
Fig. 4. Performance criteria and sustainability index values for WMS, defined for the specific water management group to which it be-
Scenario B longs (United States, Mexico, environment, or treaty obligations).
Because there are thousands of water users in the basin, and thus
the same number of SIs, the SG, shown in Eq. (12), is used to fur-
ther summarize the results. Through this method it is possible to
allocation for U.S. water users other than municipal, domestic, and (1) evaluate each water user according to required performance
industrial was set at 70% of the full allocation demand (TCEQ criteria defined for the management group to which it belongs;
2007), and this has been further reduced to 62% (personal commu- (2) summarize its performance by using the SI; and (3) summarize
nication, C. Rubenstein, commissioner of TCEQ, October 2009). the performance of groups of water users by using the SG.
These decisions can be quantified by the SI: for 70% and 62% of Table 2 shows the SG for five water user groups: (1) in the
the full demand, the SIs are 34% and 40%, respectively. Thus, United States; (2) in Mexico with treaty obligations; (3) the envi-
reducing the water allocation from 70% to 62% represents a 6% ronment in the Rio Conchos; (4) treaty obligations; and (5) all
benefit in the water allocation for WMS. water users in the Rio Grande basin (including the environment
In stressed basins such as the Rio Grande, adjustments in water and treaty obligations). Two scenarios are compared, the baseline
management policy represent changes in the water allocation for scenario and Scenario C, which is a combination of Scenarios A
stakeholders. The next example analyzes the effects of Scenario and B. Scenario C considers the water demand for WMS at
B on the treaty obligations. Fig. 5 shows the results for the treaty 62% of the full allocation (current policy), for DR-005 at 61%
obligations according to Eq. (14). Reducing the WMS demand will of the full allocation (demand after buy-backs and water conserva-
result in (1) no change in the severity of the deficits (vulnerability) tion measures), and that the water savings in DR-005 are used for
and in the variability of the deliveries (standard deviation); (2) an environmental flows.
increase in the time the treaty obligations will be met (reliability); Because of the reduction in the WMS water demand, the sus-
and (3) an increase in the recovery of the system (resilience). The tainability of the treaty obligations and the U.S. group increased by
SI shows that the treaty obligations will benefit as a result of 19% and 11%, respectively. Similarly, the sustainability for Mexico
Scenario B. and the environment increased 16% and 8%, respectively, because
The 1944 treaty specifies that Mexico must deliver to the United of the reduction in the water demand of DR-005 and the delivery of
States a specified amount of water (2;159 million m3 ) during a the saved water to the environment. Overall, the sustainability for
5-year cycle; however, the cycles may expire earlier (less than the Rio Grande increased 15% with the adaptive strategies pro-
5 years) if the U.S. storage capacity in both international reservoirs posed in Scenario C.
In addition, water users have been grouped according to their
location in the basin, using Eq. (12), to identify stressed water re-
100% source areas. Fig. 6 shows the SG of the baseline scenario for 12
geographic areas, five in the United States and seven in Mexico. For
Performance Criteria and

80% Reliab the United States, the Forgotten River (US-1), Pecos (US-2), and
ility
Sustainability (%)

Standard Deviation the lower Rio Grande valley (US-5) subbasins are the areas with the
60% Sustain lowest sustainability. For Mexico, the Forgotten River (MX-1), Rio
ability I
ndex Conchos (MX-2), and Bajo Rio Bravo (MX-7) subbasins are the
40% areas with the lowest sustainability.
Vulnerability
20%
Table 2. Sustainability by Group, Baseline, and Scenario C
0% Sustainability
40% 50% 60% 70% 80% 90% 100%
Group Baseline (%) Scenario C (%) Δ (%)
Water Demand WMS 8-13 (%)
United States 30 41 þ11
Reliability (%) Vulnerability (%) Resilience (%)
Mexico 33 49 þ16
Std. Deviation (%) Sustainability (%)
Treaty obligations 51 70 þ19
Fig. 5. Performance criteria and sustainability index values for treaty Environment 62 70 þ8
obligations, Scenario B Rio Grande 32 47 þ15

JOURNAL OF WATER RESOURCES PLANNING AND MANAGEMENT © ASCE / SEPTEMBER/OCTOBER 2011 / 387
 Fig. 6. Sustainability by region, baseline scenario Fig. 7. Change in the sustainability index by subbasin

Along the border, three areas are of particular interest because Fig. 7 shows the increment in the sustainability (Δ sust.) attrib-
of their complex water management: the Forgotten River (US-1/ utable to Scenario C. In the United States, two regions benefit:
MX-1), the Big Bend area (US-3/MX-3), and the lower Rio Grande Amistad-Falcon and the lower Rio Grande valley. In Mexico, three
valley (US-5/MX-7). The Forgotten River subbasin is the most regions benefit: Rio Conchos, Amistad-Falcon, and Bajo Rio Bravo
stressed area in the basin. The growing water demand for municipal subbasins. The geographic display of results allows identification
and industrial use in El Paso–Ciudad Juarez plus the agricultural of regions at risk and regions that will benefit from an alternative
use of El Paso Water Irrigation District #1 (EPWID #1) and water management policy.
DR-009 Valle de Juarez have exhausted the water resources in
the area; the water demands are larger than the natural availability
of water in this area. These conditions are indicated in the results Conclusions
with a sustainability of 0%. For Mexican demands in this reach, the
reliability is 0%, meaning that during the simulation period there The extent to which water management policies are sustainable can
be determined using the SI proposed in this paper. The SI identifies
was never enough water to meet their water demand; demonstrating
policies that preserve or improve the desired water management
the overallocation of water rights. For U.S. demands in this reach,
characteristics of the basin in the future. The SI makes it easier
in at least 1 year they experienced a deficit of 100%, so the maxi-
to evaluate, compare, and identify adaptive policies that improve
mum deficit criterion (1-Max def) was never met, demonstrating
water management when trade-offs among performance criteria
the stress of the system. After the Forgotten River, the lower
occur. The comparison of the SI among different policies allows
Rio Grande valley is the most stressed area in the basin; water sup-
identifying (1) if a policy is working, i.e., in Scenario A, despite
ply in this region depends on the water use in the whole basin.
the efforts to improve the water supply of DR-005 by reducing its
Water management in the tributaries consumes the water that is pro-
water demand, the SI shows that its water supply is still unsustain-
duced before it reaches the Rio Grande main stream. The water able because the maximum deficit problem has not been solved; (2)
supply of the lower Rio Grande valley depends on the storage when a policy starts working, i.e., in Scenario A the policy starts
of the international reservoirs, which depend on the water from working after the water demand of DR-005 has been reduced to
the tributaries. During drought periods, almost no water flows to 70%; (3) by how much the policy improves the water management,
the Rio Grande from the tributaries, storage in both international i.e., in Scenario A the SI for VMc Camargo increases 33% when
reservoirs is greatly decreased, and the water supply for this area the water demand of DR-005 is reduced from 100% to 90%, and the
is threatened. The SIs for MX-7 and US-5 are 18% and 34%, re- savings are allocated to the environment; (4) at what point a policy
spectively. The Big Bend region is another stressed area. Even becomes useful, i.e., in Scenario A the SI for VMc Camargo be-
though the sustainability is 100%, this calculation does not con- come steady at 60%, meaning that this policy is effective up to a
sider the environmental needs for this region; the environmental 40% reduction in DR-005 demand; and (5) if it affects other water
flows for the Big Bend have not been defined yet. This result exem- users, i.e., the SI shows that Scenario B also benefits the treaty ob-
plifies a limitation of the SI and SG: when the water demand has not ligations. The SI promotes a holistic water management evaluation
been calculated, e.g., for environmental flows in the Big Bend reach because incorporates tailor-made performance criteria in the index
where water demands for other purposes are low, it is not possible structure and uses different structures in the same system. The SI is
to estimate the SI, and as a result, the SG does not consider this versatile; it was successfully applied to water users, environmental,
water demand. In addition, most of the water in the Big Bend area and system requirements.
comes from the Rio Conchos (75% on average) and is managed by The SG was successfully implemented to summarize the indi-
CONAGUA without a defined policy to deliver water from the Rio vidual SI calculated for each water user, environmental, or system
Conchos to the Rio Grande. An international team has been work- requirements. Similar to the SI, the SG make easier to evaluate,
ing to define the environmental flows along this reach (WWF 2006; compare, and identify adaptive policies that improve water man-
Sandoval-Solis and McKinney 2009) and on a policy to provide agement for groups of water users. The SG is versatile; groups
environmental flows to the Big Bend reach. of water users can be integrated according to the type of use

388 / JOURNAL OF WATER RESOURCES PLANNING AND MANAGEMENT © ASCE / SEPTEMBER/OCTOBER 2011
(agriculture, municipal, environment), jurisdiction (United States, Comisión Nacional del Agua (CONAGUA). (2004a). Ley de aguas nacio-
Mexico) or subbasin. The comparison of the SG among different nales y su reglamento, Mexico City (in Spanish).
policies allows identifying which group of water users benefits and Comisión Nacional del Agua (CONAGUA). (2004b). Registro público de
by how much, with respect to the reference scenario. By grouping derechos de agua, Mexico City (in Spanish).
Danner, C. L., McKinney, D. C., Teasley, R. L., and Sandoval-Solis, S.
water user according to their location, the SG makes it possible to
(2006). “Documentation and testing of the WEAP model for the Rio
identify regions that are at risk from unsustainable water manage-
Grande/Bravo basin.” Online Rep. 06-08, Center for Research in Water
ment policies and regions that will benefit from an alternative water Resources, Univ. of Texas, Austin, TX.
management policy. Determining weights for the SG through the Enriquez-Coyro, E. (1976). “El tratado entre México y los Estados Unidos
annual water demand is used in this paper as an alternative method de América sobre ríos internacionales.” Facultad de Ciencias Políticas
when explicit weights for water users, system requirements, and the y Sociales, Universidad Nacional Autónoma de México, Mexico City
environment are not defined. (in Spanish).
The SI and SG have been presented to decision makers in the Esty, D. C., Levy, M., Srebotnjak, T., and Sherbinin, A. (2005). 2005 envi-
basin who have recognized the practicality of the index. On one ronmental sustainability index: Benchmarking national environmental
hand, the SI synthesizes the performance criteria that otherwise stewardship, Yale Center for Environmental Law and Policy, New
are tedious to analyze. On the other hand, SG is more convenient Haven, CT.
to compare the performance of groups of water users and regions at Fanai, N., and Burn, D. H. (1997). “Reversibility as a sustainability cri-
terion for project selection.” Int. J. Sustain. Dev. World Ecol., 4(4),
a glance.
259–273.
Giorgi, F., and Mearns, L. O. (2002). “Calculation of average, uncertainty
range, and reliability of regional climate changes from AOGCM sim-
Recommendations ulations via the reliability ensemble averaging (REA) method.” J. Clim.,
15(10), 1141–1158.
The SI is not intended to replace any performance criteria (e.g., Gwartney, J., Lawson, R., and Block, W. (1996). Economic freedom of the
reliability, resilience, vulnerability); its objective is to make easier world (1975–1995), Free Market Foundation, Johannesburg, South
the quantification and identification of policies that improve water Africa, 1–46.
management when there are trade-offs among criteria. The SI can Hajkowicz, S. (2006). “Multi-attributed environmental index construction.”
be included as one of the water management goals when decisions Ecologic. Econ., 57(1), 122–139.
are being made regarding the design, planning, and operation pol- Hashimoto, T., Stedinger, J. R., and Loucks, D. P. (1982). “Reliability, resil-
icies of water resource systems. iency and vulnerability criteria for water resource system performance
The methodology proposed in this article helps identify policies evaluation.” Water Resour. Res., 18(1), 14–20.
that are more sustainable than a policy used as a reference (i.e., Hirsch, R. M. (1979). “Synthetic hydrology and water supply reliability.”
Water Resour. Res., 15(6), 1603–1615.
baseline scenario) given the performance criteria considered for
Hopkins, M. (1991). “Human development revisited: A new UNDP report.”
each water management group and the weights used in the SG.
World Dev., 19(10), 1469–1473.
One drawback of the methodology proposed is the involvement International Boundary and Water Commission (IBWC). (1906). Conven-
of subjective judgment during the selection of performance criteria tion between the United States and Mexico. Equitable distribution of
for the SI and weights for the SG. waters of the Rio Grande, El Paso, TX.
In the simulation process, further research is needed to estimate International Boundary and Water Commission (IBWC). (1944). Treaty
and evaluate water management of the basin under different hydro- between the United States and Mexico. Utilization of waters of the
logic conditions, considering the alteration of the hydrological Colorado and Tijuana Rivers and of the Rio Grande, Washington, DC.
cycle due to climate change. Also, in this research, water demands International Boundary and Water Commission (IBWC). (2003). “Volumes
are considered fixed for the hydrologic period of analysis. Further of water saved with the modernization and improved technology proj-
research is needed to estimate future demands and their evaluation ects for the irrigation districts in the Rio Conchos basin and measures
in the planning simulation model. for their conveyance to the Rio Grande.” Minute 309, El Paso, TX.
International Union for Conservation of Nature (IUCN). (1980). The world
conservation strategy: Living resource conservation for sustainable
development, International Union for Conservation of Nature, United
Acknowledgments Nations Environment Program, and World Wildlife Fund, Gland,
Switzerland.
The authors would like to thank the National Council of Science
Klemes, V., Srikanthan, R., and McMahon, T. A. (1981). “Long-memory
and Technology of Mexico (CONACYT) for the financial support flow models in reservoir analysis: What is their practical value?” Water
provided to the first author. Partial funding for this research was Resour. Res., 17(3), 737–751.
provided by the U.S. EPA, the USDA, and the Instituto Mexicano Lence, B. J., Furst, J., and Matheson, S. (1997). “Distributive fairness as a
de Tecnología del Agua. Special thanks are given to the National criterion for sustainability evaluative measures and application to
Heritage Institute and the Stockholm Environment Institute for project selection.” Int. J. Sustain. Dev. World Ecol., 4(4), 245–258.
their support. Loucks, D. P. (1997). “Quantifying trends in system sustainability.” Hydrol.
Sci. J., 42(4), 513–530.
Loucks, D. P., and van Beek, E. (2005). Water resources systems planning
References and management, United Nations Educational, Scientific and Cultural
Organization (UNESCO), Paris.
Brown, R. M., McClelland, N. I., Deininger, R. A., and O’Connor, M. F. McMahon, T. A., Adeloye, A. J., and Sen-Lin, Z. (2006). “Understanding
(1972). “A water quality index—Crashing the psychological barrier.” performance measures of reservoirs.” J. Hydrol. (Amsterdam), 324
Indicators of environmental quality, Plenum, New York. (2006) 359–382.
Cai, X., McKinney, D. C., and Lasdon, L. S. (2002). “A framework for Mendoza, V. M., Villanuave, E. E., and Adem, J. (1997). “Vulnerability of
sustainability analysis in water resources management and application basins and watersheds in Mexico to global climate change.” Clim. Res.,
to the Syr Darya basin.” Water Resour. Res., 38(6), 1085–1098. 9, 139–145.
Collado, J. (2002). Criterios de distribución del agua en la Cuenca del Milbrink, G. (1983). “An improved environmental index based on the
Rio Bravo, Instituto Mexicano de Tecnología del Agua (IMTA), relative abundance of oligochaete species.” Hydrobiologia, 102(2),
Cuernavaca, México (in Spanish). 89–97.

JOURNAL OF WATER RESOURCES PLANNING AND MANAGEMENT © ASCE / SEPTEMBER/OCTOBER 2011 / 389
Moy, W. S., Cohon, J. L., and Revelle, C. S. (1986). “A programming Simonovic, S. P. (1998). “Sustainable reservoir development and manage-
model for analysis of reliability, resilience and vulnerability of a water ment.” No. 251, International Association of Hydrological Sciences,
supply reservoir.” Water Resour. Res., 22(4), 489. Wallingford, UK.
Palmer, W. C. (1965). “Meteorological drought.” Research Paper Slottje, D. J. (1991). “Measuring the quality of life across countries.” Rev.
45, Weather Bureau, U.S. Dept. of Commerce, Washington, DC, Econ. Stat., 73(4), 684–693.
58. Texas Commission on Environmental Quality (TCEQ). (2005). “Rio
R. J. Brandes Co. (2004). “Water availability modeling for the Río Grande Grande water master program.” Austin, TX.
basin: Water availability assessment. Final report.” Texas Commission Texas Commission on Environmental Quality (TCEQ). (2006). “Operation
on Environmental Quality, Austin, TX. of the Rio Grande: Allocation and distribution of waters.” Texas Admin-
Ray, P. A., Vogel, R. M., and Watkins, D. W. (2010). “Robust optimization istrative Code Title 30: Environmental quality, Part 1, Chapter 303,
using a variety of performance indices.” Proc., World Environmental Subchapter C, Austin, TX.
and Water Resources Congress, ASCE, Reston, VA. Texas Commission on Environmental Quality (TCEQ). (2007). “Water
Reiquam, H. (1972). “Establishing priorities among environ- availability models: Rio Grande basin.” Austin, TX.
mental stresses.” Indicators of environmental quality, Plenum, Vigerstol, K. (2002). “Drought management in Mexico’s Rio Bravo basin.”
New York. M.S. thesis, Univ. of Washington, Seattle.
Sagar, A. D., and Najam, A. (1998). “The human development index: A Von Neumann, J., and Morgenstern, O. (1974). Theory of games and eco-
critical review.” Ecologic. Econ., 25(3), 249–264. nomic behavior, Princeton University Press, Princeton, NJ.
Sainz, P. (1989). “An index of social welfare.” Towards a new way to World Commission on Environment and Development (WCED). (1987).
measure development, Office of the South Commission, Caracas, “Our common future: The Brundtland report.” Oxford University Press,
Venezuela, 156–160. Oxford, UK.
Sandoval-Solis, S., and McKinney, D. C. (2009). “Hydrological feasibi- World Health Organization (WHO). (2009). Summary and policy implica-
lity of environmental flows in the Rio Grande/Bravo basin.” Proc., tions Vision 2030: The resilience of water supply and sanitation in the
World Environmental and Water Resources Congress, ASCE, Reston, face of climate change, Geneva.
VA. World Wildlife Fund (WWF). (2006). “Conclusions from the building
Sandoval-Solis, S., McKinney, D. C., Teasley, R. L., and Patino-Gomez, C. block method specialist work session Rio Conchos.” Chihuahuan
(2011). “Groundwater banking in the Rio Grande basin.” J. Water Desert Program Office, Chihuahua City, Mexico.
Resour. Plann. Manage., 137(1), 62–71 World Wildlife Fund (WWF). (2007). World’s top 10 rivers at risk, Gland,
Secretaria de Agricultura, Ganadería, Desarrollo Rural, Pesca y Alimenta- Switzerland.
ción (SAGARPA). (2003). “Reglas de operación del programa de ad- Zongxue, X., Jinno, K., Kawanura, A., Takesaki, S., and Ito, K. (1998).
quisición de derechos de uso del agua.” Diario Oficial de la Federación, “Performance risk analysis for Fukuoka water supply system.” Water
Mexico City (in Spanish). Resour. Manage., 12, 13–30.

390 / JOURNAL OF WATER RESOURCES PLANNING AND MANAGEMENT © ASCE / SEPTEMBER/OCTOBER 2011

View publication stats