JOURNAL OF THE AMERICAN WATER RESOURCES ASSOCIATION

Vol. 48, No. 2 AMERICAN WATER RESOURCES ASSOCIATION April 2012

MACRO, MESO, AND MICRO-EFFICIENCIES IN WATER RESOURCES MANAGEMENT:

A NEW FRAMEWORK USING WATER BALANCE1

Naim Haie and Andrew A. Keller2

ABSTRACT: One of the most important performance indicators for water resources systems (WRSs) manage-
ment is efficiency. Here, water balance, based on mass conservation, is utilized to systemically develop three lev-
els of composite efficiency indicators for a WRS, which are configurable based on two types of water totals: total
inflow and total consumption (outflow that effectively is not available for reuse). The indices characterize hydrol-
ogy of an area by including in their formulations the flow dynamics at three integrated levels. Furthermore, the
usefulness of water is incorporated into the indicators by defining two weights: one for quality, and the other for
beneficial attributes of water use. Usefulness Criterion is the product of quality and beneficial weights, empha-
sizing the equal significance of the two dimensions. Both of these weights depend on the system itself and the
priorities of the supervising organization, which also are shaped by the objectives and values of the given soci-
ety. These concepts lead to the definition of Macro, Meso, and Micro-Efficiencies, which form a set of integrated
indicators that explicitly promotes stakeholder involvement in evaluation and design of WRSs. Macro, Meso,
and Micro-Efficiencies should be maximized for both water totals, which is an integrated prerequisite for
sustainability and is less promoted by competing stakeholders. To demonstrate this new framework, it is applied
to published data for urban and agricultural cases and some results are explained.

(KEY TERMS: 3ME; efficiency; Usefulness Criterion; beneficial weight; quality weight; water resources manage-
ment; irrigation; water balance.)

Haie, Naim and Andrew A. Keller, 2012. Macro, Meso, and Micro-Efficiencies in Water Resources Management:
A New Framework Using Water Balance. Journal of the American Water Resources Association (JAWRA) 48(2):
235-243. DOI: 10.1111 ⁄ j.1752-1688.2011.00611.x

INTRODUCTION beneficially, such as by plant evapotranspiration

(ET). This efficiency (Israelsen, 1932) has been used
throughout the world for decades up to the present
As water resources systems (WRSs) become more day. Burt et al. (1997) in their informative paper,
complex and competition among water users effectively propose a CE type efficiency (p. 428). How-
increases, simple efficiency indicators have proven ever, Willardson et al. (1994), and later others such
inadequate in promoting an effective WRS design and as Allen et al. (2005) show that application of CE can
evaluation hence composite performance indicators, sometimes lead to error. For example, improving CE
such as the ones proposed in this paper, are in great of a WRS, such as a farm, at times does not promote
need. For example, classical efficiency (CE) is defined the objective of more real water being available to
as the percentage of the abstracted water consumed the downstream users. In the meantime, managers,
1
Paper No. JAWRA-10-0108-P of the Journal of the American Water Resources Association (JAWRA). Received July 11, 2010; accepted
August 18, 2011. ª 2011 American Water Resources Association. Discussions are open until six months from print publication.
2
Respectively, Associate Professor with Habilitation, Civil Engineering Department, University of Minho, 4800-058 Guimarães, Portugal;
and President, Keller-Bliesner Engineering, LLC, Logan, Utah (E-Mail ⁄ Haie: [email protected]).

JOURNAL OF THE AMERICAN WATER RESOURCES ASSOCIATION 235 JAWRA

 HAIE AND KELLER

practitioners, and policy makers are faced with to distinguish between beneficial and nonbeneficial
mounting problems in an environment of increasing uses of water. For example, beneficial water permits
complexity, and the need is growing for a coherent and certificates are needed in various states of the
framework that puts some light on the performance United States for using water (e.g., Marcus and
of a WRS in the context of sustainable development. Kiebzak, 2008; WRB, 2009). Of course, in water-
In response to this challenge, other efficiency indica- scarce regions in particular, proof of beneficial use
tors have been proposed, such as effective efficiency is a crucial step in being able to develop water, as
(EE), by Keller and Keller (1995) and later Haie and well as having a priority of beneficial water use
Keller (2008), as tools to promote more sustainable when severe scarcity or drought hits a region (RCM,
performance of water resources interventions. 2005). The first goal of the Strategic Planning Goals
However, any robust indicator should be based on of the Department of Water Resources (DWR,
sound principles and scientific formulations (UN, 2009a) of the State of California, United States
2006) such as water balance, but so far no such indi- (U.S.), declares that ‘‘DWR is responsible for pro-
cator for water resources efficiencies using the princi- moting California’s general welfare by ensuring ben-
ple of the conservation of mass has been defined. eficial water use and development statewide.’’ The
Even EE, much more complete than CE, lacks such a beneficial use of water and its usefulness is increas-
systemic development and consequently is an incom- ingly becoming a matter of political debate (David
plete formulation. For example, it does not consider Molden, 2009, International Water Management
all water flow paths that influence the efficiency of a Institute (IWMI), personal communication), and effi-
WRS. It also does not include the Usefulness Crite- ciency of water use can be seen as a key policy or
rion, in both of its dimensions as described here, in strategy in resource management in order to
an explicit and complete manner. This lack of sys- ‘‘achieve a beneficial outcome’’ (DWR, 2009b).
temic approach resulted in the definition of two levels The purpose of this paper is to develop composite
of analysis in EE instead of three, as proposed in this efficiency indicators for water use that are systemic
paper. On the other hand, it should be noted that according to the general law of the conservation of
water accounting approaches described in the litera- mass (water balance), keeping in mind the two cru-
ture are based on water quantity and does not cial water totals of a system: total inflow and total
include water quality (or Usefulness Criterion in consumption. An indicator should reflect the dynam-
general). These schemes are explained using water ics of different scales from, for example, a farm and a
balance and are descriptive in nature, while efficiency city to a basin. It should also incorporate in its for-
expressions developed in this paper are performance mulation the important concepts of beneficial ⁄ non-
composite indicators, which are different from water beneficial, and water and system quality. The next
accounting. two sections of this paper outline the development of
Molden and Sakthivadivel (1999) refer to Macro, various basic concepts that are brought together in
Meso, and Micro-levels as three spatial hierarchies the two subsequent sections, which present the new
without using them or defining them explicitly, that composite indicators.
is, through specific expressions. Hussain et al. (2007)
also use them as ‘‘micro ⁄ local, meso ⁄ regional and
macro ⁄ national levels.’’ These three prefixes are used
extensively in economics; however, this paper is a WATER RESOURCES SYSTEMS AND
water resources study that takes into account flow WATER BALANCE
paths in the hydrology of an area and its policy and
management preferences and does not explicitly
include economic considerations, although they can In order to develop an indicator based on water
be utilized indirectly as will be explained later. balance, first, a typical WRS (e.g., a city, farm, collec-
Besides water quantity, another important aspect tion of them or even a subbasin) is defined as shown
of water use is its usefulness (ASCE, 2004; Marcus in Figure 1, having in mind that its boundaries are
and Kiebzak, 2008), which incorporates quality, such known to the manager and the analyst.
as the level of salt accumulation, and whether a par- The variables in the figure are defined in Table 1.
ticular use of water is beneficial or not (Vance, 2005). These must have consistent units such as volume,
Molden and Sakthivadivel (1999) use process and depth, percentage or fraction.
nonprocess concepts to indicate whether it is useful It should be noted that in Table 1 and throughout
for the purpose of a project or activity and then sepa- this paper the word ‘‘consume’’ (‘‘consumptive’’ or any
rate this into beneficial and nonbeneficial. other derivative) has a special significance in water
An important aspect of managing a WRS (e.g., a resources and irrigation management. It means the
city, farm, collection of them or even a subbasin) is portion of water use that effectively is not available

JAWRA 236 JOURNAL OF THE AMERICAN WATER RESOURCES ASSOCIATION

 MACRO, MESO, AND MICRO-EFFICIENCIES IN WATER RESOURCES MANAGEMENT: A NEW FRAMEWORK USING WATER BALANCE

River a region, it is recommended that NRWR should be

analyzed separately.
PP OS ET
VU VA
NR Types of Water Totals

VD RF Table 1 gives the individual variables needed to

RP develop performance indicators such as efficiency;
however, they must be set against a ‘‘total’’ value.
Two important ‘‘totals’’ are utilized in this study; for
example, the first part of Equation (1) is a ‘‘total’’
FIGURE 1. Diagram of a Water Resources System.
value. However, Equation (1) can be written in an
alternative way by rearranging its variables as fol-
lows:
TABLE 1. Water Balance Variables in a Water Resources System
and Their Definitions.
ðV1 þ OS þ PP  V2  RPÞ  ðET þ NRÞ ¼ 0: ð2Þ
Variable Description
Note that while (V1 + OS + PP) of Equation (1) is
ET Evapotranspiration total inflow, denoted by index i,
NR Nonreusable, water consumption (V1 + OS + PP ) V2 ) RP) of Equation (2) is total
OS Water from other sources water consumption of the system, referred to by index
PP Total precipitation
RF Return flows
c. In order to incorporate these two types of ‘‘totals’’
RP Potential return (does not return to the main source) into one expression, an alternative form of water bal-
VA Abstracted water from the main source ance equation can be written by adding Equations (1)
VD Volume of water downstream after RF in the and (2) and using the two binary indices (i, c) with
main source values 0 or 1 and i + c = 1 in order to distinguish,
VU Volume of water upstream before abstraction in the
main source
respectively, between total inflow and total consump-
V1 Volume of water at section 1 (VU or VA) tion based indicators:
V2 Volume of water at section 2 (VD or RF)

½ðV1 þ OS þ PPÞ  cðV2 þ RPÞ

ð3Þ
½ðET þ NRÞ þ iðV2 þ RPÞ ¼ 0:
for further reuse (EEA, 2009). ET, very saline or pol-
luted return flows, and water that becomes part of To clarify, Equation (3) is the condensed form of
products are examples of consumptive water use, two different equations. For example, i = 1 and c = 0
which is a special kind of outflow from a WRS. gives Equation (1), and the indicators developed
The WRS depicted in Figure 1 has two parts: (1) using this equation are based on total inflow.
total inflow = V1 + OS + PP and (2) total out- The reason that total water consumption was
flow = ET + V2 + RP + NR. As such, Water Balance chosen is because of its vital importance in WRS design
(total inflow ) total outflow = D) for any WRS is writ- for real-water saving. Reducing unproductive evapora-
ten as follows for the purposes of this paper: tion and transpiration, growing lower-water consum-
ing crops or using plants at lower-water consuming
ðV1 þ OS þ PPÞ  ðET þ RP þ V2 þ NRÞ ¼ 0: ð1Þ times and places, preventing (high) water pollution, or
reducing water flow to saline or highly polluted sinks,
This universal mass conservation equation is and increasing effective use of rainfall are the four
defined for a time period in which the change in main paths to true water saving (Clemmens et al.,
water storage of a WRS, D, can be assumed 0 (Burt 2008), that is, consumptive saving (SC). According to
et al., 1997, p. 428) and, consequently, does not Haie and Keller (2008), SC is a reduction in the water
appear at the end of Equation (1). This time period permanently removed from the river basin, and
can be years, a season, or even an event such as irri- abstraction saving (SA) is a reduction in water diver-
gation, assuming that the soil moisture content of the sion (abstraction or withdrawal) from its source such
root zone before two consecutive irrigations is the as a river. Reducing total inflow of water through SC or
same. It should be mentioned that OS can come from SA mechanisms is crucial in developing a sustainable
RP of a previous WRS, from nonrenewable water system (the Applications section will make this point
resources (NRWR), or other sources. However clearer). Properly implemented both of these two types
because of the sustainability of the water resources of of saving – parallel to the two types of water total –

JOURNAL OF THE AMERICAN WATER RESOURCES ASSOCIATION 237 JAWRA

 HAIE AND KELLER

may promote 1 or more of the 11 categories of benefits Xq ¼ WqX  X

enumerated in Haie and Keller (2008), which are from
Xb ¼ WbX  X ð4Þ
environmental protection to reducing costs to more
equity. It is also of interest to note that Hussain et al. Xs ¼ WqX  WbX  X:
(2007) propose indicators for valuing agricultural
water by stating, to a limited extent, the same ideas. Quality weights depend on water constituents and
For example, they use in one of their indicators the there are various methods in specifying them as
‘‘volume of water depleted by agricultural use (ET), weights, having in mind that water quality is one of
flows into sinks and pollution or volume of water sup- the components that defines the quality weight (WbX),
plied (rain plus irrigation water).’’ as explained above. It is not the purpose of this study
to discuss these methods but examples are mentioned
here for the better understanding of the concept. One
can refer to the ‘‘water status’’ and its sub-status des-
DIMENSIONS OF USEFULNESS ignations such as ‘‘surface water chemical status’’
defined in the European Water Framework Directive
(EU, 2000). Another example is to define weights for
In order to define performance indicators, such as the removal of salt, which should not be allowed
the efficiency of a system, the usefulness of variables to accumulate in agriculture and ecosystems, an
needs to become explicit. Two dimensions are consid- increasingly important issue. To do this and to keep
ered for making a variable useful: beneficial dimen- salinity within acceptable levels, leaching fractions
sion, b, and quality dimension, q. These two (LF) are calculated (Ayers and Westcot, 1994). LF is
correspond, to some extent, to what Vance (2005) to maintain an acceptable salt balance according to
explains in relation to the role of science in water specified management criteria like 90% crop yield,
resources decision making. Having defined these hence, most of the time in irrigation management,
dimensions, then the useful part of a variable, desig- WqX = 1 ) LFX. Of course, this goes beyond agricul-
nated with a subscript ‘‘s’’ and called Usefulness Crite- ture, and the weights can direct general policies in
rion, is the product of both dimensions, indicating the such a way that Venot et al. (2008) call ‘‘for a formal
equally high significance attached to both of these allocation of water to be set aside for the purpose of
aspects. It is crucial to note that these are influenced environmental preservation.’’ On the other hand, the
by the management of a WRS as well as its physical setting of these weights can be linked to the costs
attributes. For example, quality dimension is about and benefits of different levels of pollution, or can be
the quality of water and the system that water flows related to water quality trading (Garcı́a et al., 2011;
through but also the level of toleration for a design Horan and Shortle, 2011). Relative to this last point,
quality, which is a management decision. Although and unlike CO2 trading that has no flow path differ-
the economics involved in applying these dimensions entiation, trading in water quality and pollution
are not explicitly discussed in this paper, any final con- depends on flow paths. For example, the same
sideration as to the reasonable use of water should be amount of pollution going to a river might be much
based on these two dimensions. Also, it is possible that more damaging (from the stakeholders’ point of view)
the words ‘‘beneficial’’ and ‘‘useful’’ have different than going to say groundwater.
meaning in different localities, but what is crucial, are A beneficial use of water is defined according to
the concepts and the exact definitions in a coherent the purposes of the project and is also influenced by
framework rather than the labels and the names. the stakeholders and the society, as explained previ-
In order to find the useful value of any variable, it is ously. Table 2 lists some of the possibilities adapted
multiplied by the beneficial and quality weights from Allen et al. (2005) and Vance (2005).
(between 0 and 1 with 1 being its ‘‘best’’ state). A qual- Having described the concept of Usefulness Crite-
ity weight (WqX, X being any of the variables shown in rion and its two dimensions, and for the purposes of
Table 1) gives the effective value of X because water is this study, the following weights are all set to ‘‘1’’ in
polluted and gets more degraded as it goes through a the absence of better data:
system. On the other hand, some of the variables in
Table 1 have both a beneficial and a nonbeneficial com- • Quality weights: WqET, WqPP, WqNR.
ponent. The beneficial weight WbX gives the beneficial • Beneficial weights: WbV1, WbVU, WbVA, WbOS,
component of X, which is defined by the managers, WbV2, WbVD, WbRF, WbRP.
consultants, and decision makers and depends on the These unitary values convey that, for example, only
intent of an intervention and societal priorities. the effective part of PP (in this study called beneficial
Having in mind these definitions, the following are the part, i.e., PPb) should be utilized because it is the part
general equations for any variable X: that satisfies water requirements of a system, that is,

JAWRA 238 JOURNAL OF THE AMERICAN WATER RESOURCES ASSOCIATION

 MACRO, MESO, AND MICRO-EFFICIENCIES IN WATER RESOURCES MANAGEMENT: A NEW FRAMEWORK USING WATER BALANCE

TABLE 2. Beneficial and Nonbeneficial Uses of Water in a Water Resources System.

ET NR RF + RP

Beneficial ET (crop, landscape) Nonreusable deep percolation for Reusable deep percolation for salt
Evaporation for climate salt control (leaching) control (leaching)
control Virtual water
Nonbeneficial ET (Phreatophyte) Nonreusable excess deep percolation Reusable excess deep percolation
Evaporation (sprinkler, Runoff or spills to salt sinks Reusable runoff
reservoir, excess wet soil) Reusable canal spills

Note: ET, evapotranspiration; NR, nonreusable; RF, return flows; RP, potential return.

total inflow. Or, it is assumed that all of V1 and OS are flow have the same unit of measure. This is conve-
potentially beneficial and that they are only to be nient because the water balance expression, Equa-
reduced according to their quality in order to give the tion (1), and its generalized form, Equation (5), have
total useful inflow of water into a WRS. It should be two parts. It is Equation (5) that is of interest here
noted that the rest of this paper only uses the subscript because it utilizes the Usefulness Criterion, making
‘‘s,’’ which represents the useful part of a variable. the efficiency terms more general, and, hence, the
At this point, we can incorporate the Usefulness Cri- equation enables more possible uses in regards to the
terion of inflows and outflows of a WRS into the water formation of versatile policies and practices, as will
balance Equation (3) as shown in the following equation: be described subsequently.
One of the important tasks of a manager is to pro-
f½ðV1 þ OS þ PPÞ  cðV2 þ RPÞ
ð5Þ mote Usefulness Criterion of the water that is being
 ½ðET þ NRÞ þ iðV2 þ RPÞgs ¼ K: utilized. This is done, as mentioned above, by trying
to minimize K in Equation (5), that is, minimizing
The right hand side of Equation (5) is explained as water losses and wastefulness. Putting it in another
follows. Rifkin (1980) states that ‘‘pollution is just form, the objective function for efficient and produc-
another name for entropy,’’ because systems like WRSs tive WRSs becomes [Min (K)], or
degrade water, which effectively results in water
‘‘loss.’’ Also, Hussain et al. (2007) explain the negative
impacts of ‘‘head enders’ overuse ⁄ pollution of water on Minff½ðV1 þ OS þ PPÞ  cðV2 þ RPÞ
ð6Þ
availability ⁄ quality of water for tail enders.’’ Nonbene-  ½ðET þ NRÞ þ iðV2 þ RPÞgs g:
ficial water of a WRS can also be considered as the
‘‘water loss’’ for the system under study. Thus, after Minimizing the preceding expression is equivalent
incorporating the Usefulness Criterion ‘‘s’’ as seen in to maximizing the ratio of the two positive parts
Equation (5), a nonnegative loss factor Lambda, K, is (Appendix), as shown in the following equation:
inserted into the equation in order to maintain the
water balance. As such, K can be defined as the total  
ET þ NR þ iðV2 þ RPÞ
effective water loss from a WRS, which results from Max : ð7Þ
V1 þ OS þ PP  cðV2 þ RPÞ s
considering both beneficial and quality weights. Conse-
quently, it is clear that a fundamental aim of WRS Per the definition given at the beginning of this
management should be to minimize K. Finally, it section, the expression to be maximized in Equa-
should be noted that Equation (3) is a special case of tion (7) is actually the ‘‘universal’’ efficiency indicator
Equation (5), with WqX = WbX = 1 (no Usefulness or (E) of a system based on water balance:
unitary Usefulness Criterion) for all the variables, and
K = 0, which implies that there are no effective system  
losses due to the Usefulness Criterion. ET þ NR þ iðV2 þ RPÞ
E¼ : ð8Þ
V1 þ OS þ PP  cðV2 þ RPÞ s

Equation (8) is a composite performance indicator

EFFICIENCIES based on water balance that calculates efficiency of a
WRS. In general, (i, c) = (0, 1) gives the percentage of
total consumption that is useful consumption. While
At this point, we are ready to define one of the (i, c) = (1, 0) gives the percentage of total useful
more important performance indicators: efficiency. It inflow that is useful outflow. However, sustainability
is defined as the ratio of useful outflow to its corre- of a system depends on the level of analysis, which is
sponding total flow; note that both outflow and total made explicit in the next section.

JOURNAL OF THE AMERICAN WATER RESOURCES ASSOCIATION 239 JAWRA

 HAIE AND KELLER

Macro, Meso, and Micro-Efficiencies greater than or equal to MicroE, because downstream
impacts of a WRS are not considered in MicroE,
When considering three levels of analysis in rela- which is an efficiency relative to the amount of water
tion to water flows and areal scales, the following effi- abstracted from a source without any consideration
ciencies can be defined, keeping in mind the for water reuse. Considering the unitary values of
definition of efficiency provided in the last section: the beneficial weights (WbX) mentioned in the section
Dimensions of Usefulness, cMacroEb =
• Macro-Efficiency (MacroE): This is the relation- cMesoEb (cMacroE, e.g., denotes a Macro-Efficiency
ship between useful outflow and total flow as that utilizes total consumption, i.e., i = 0 and c = 1).
related to a river basin. MacroE is used to indi- In addition, if NR = PP = OS = 0, then MicroEb = CE,
cate the impact of a WRS on a basin, for example which means that, assuming 0 for the three flows,
the major river where water was abstracted. Micro-Efficiency without any quality consideration
• Meso-Efficiency (MesoE): This is the relationship results in CE. It is readily seen in Equations (9) that
between useful outflow and total flow as related iMicroE is always equal to cMicroE (any of the 3ME
to a situation between micro and macro-levels. without any subscript denotes its validity with ‘‘s’’
MesoE is used to indicate, for example, the and with ‘‘b’’), meaning that Micro-Efficiency is the
impact of return flows generated by a WRS. same relative to total consumption of a WRS or total
• Micro-Efficiency (MicroE): This is the relationship inflow. It should be also noted that for a single WRS,
between useful outflow and total flow within a a high MicroE would produce a high MacroE but the
WRS. MicroE is used to indicate the useful out- opposite (high MacroE producing high MicroE) is
flow generated by a WRS for itself. uncertain as will be seen in the Applications section.
As related to the stakeholders, while Macro-Effi-
It should be noted that inflow and outflow are dif- ciency is important for river basin and regional insti-
ferent for these three levels (see Figure 1): tutions and associations, Micro-Efficiency is
important to each single user (such as a farm or a
• For MacroE: V1 = VU and V2 = VD city) in addition to institutions and associations.
• For MesoE: V1 = VA and V2 = RF Meso-Efficiency is important for institutions, associa-
• For MicroE: V1 = VA, V2 = RF, and tions, owners, etc., that have interest at the meso
(RF + RP)S = 0 (because, according to the defini- level.
tion of MicroE, WbRF = WbRP = 0, meaning that Understanding how to determine the efficiencies
relative to the WRS itself, its return flows have associated with water resources management
zero beneficial use) enhances our ability to understand the dynamics of
their interactions, which in turn is crucial when decid-
Combining the preceding equalities with Equa- ing on proper design of the components of a WRS for
tion (8), the following new efficiency expressions real water savings and allocating water resources. In
(together called ‘‘3ME’’ [Macro, Meso, and Micro-Effi- evaluating systems or alternative interventions, 3ME
ciencies]) result: (Equations 9) must be analyzed together. Doing so
  may reveal opposing tendencies; for example, using
ETþNRþiðVDþRPÞ the Full Efficiency Models, four of them might
MacroEs ¼ ;
VUþOSþPPcðVD þRPÞ s increase while two decrease. These analyses may
i;c ¼ 0 or 1; iþc ¼ 1 expose inconsistencies in data, including weights, or
  reveal situations that require tradeoffs that a decision
ETþNRþiðRFþRPÞ ð9Þ
MesoEs ¼ maker should consciously determine, mostly in a polit-
VA þOSþPPcðRFþRPÞ s ical process including public participation.
 
ETþNR This is interesting because stakeholders are differ-
MicroEs ¼ :
VA þOSþPP s ent for various 3ME and inline with Integrated
Water Resources Management (IWRM), 3ME can be
The ‘‘s’’ subscript is used in the Full Efficiency important instruments to guide people in participa-
Models, that is, including both beneficial and quality tory decision making. For example, consider using
aspects of variables resulting in six models (three for freshwater to clean or dilute pollution such as salt
i and three for c). A ‘‘b’’ subscript is used in quantity- around a desired plant root zone. Whether this is a
only models, where WqX values for all the variables useful application of water or not would depend on
are set to 1. Note that the ‘‘b’’ subscript refers to two the stakeholders and the purposes of the allocation
concepts (beneficial weight and quantity model) but scheme being developed. Multi-objective planning and
the meaning is very close and did not warrant a fur- management would depend on various preferences,
ther index. It is apparent that MesoE is always for example, pollution clean-up has different weights

JAWRA 240 JOURNAL OF THE AMERICAN WATER RESOURCES ASSOCIATION

 MACRO, MESO, AND MICRO-EFFICIENCIES IN WATER RESOURCES MANAGEMENT: A NEW FRAMEWORK USING WATER BALANCE

for different stakeholders. And, it is of importance TABLE 3. The Basic Data for the Grand Valley (GV),
that 3ME and particularly the weights can contribute Nile Valley (NV), City, and H Cases.
‘‘with scenarios to make outcome of current policy GV NV City H
paths visible and understandable,’’ while promoting
‘‘integrated approaches to water quantity and qual- Inflow
ity’’ (Falkenmark, 2005). Anyhow the treatment of Water, VU (ha-m ⁄ year) 486.000 5,320.000
WqVU 0.864 0.945 0.9 1
weights in a multi-objective multi-sector environment
Abstraction (%VU), VA 12.1 67.3 26.4 90
is beyond the scope of this paper, which basically Consumptive use
tries to introduce an innovative framework. Total (%VU), ET 3.7 28.3 12 75
WbET 0.865 0.926 0.8 0.95
Return flow (RF)
Water (%VU) 8.4 39 14.4 5
WqRF 0.626 0.897 0.83 0.72
APPLICATIONS WbRF 1 1 1 0.90
Potential return (RP)
Water (%VU) 0 0 0 10
We are at the beginning stages of developing these WqRP 0.79
WbRP 0.87
composite indicators, and not all of the complex rela-
Outflow, VD
tionships between them are known yet. For example, Water (%VU) 96.3 71.7 88 15
only MicroE has five flow variables and some five WqVD 0.844 0.918 0.88 0.91
weights, which influence its complex behavior. Thus, WbVD 1 1 1 0.97
the cases presented here are used for illustrative pur-
Note: VU, volume of water upstream before abstraction in the main
poses and are not examples of their thorough analyses, source; VA, abstracted water from the main source; ET, evapo-
particularly in the context of IWRM. The cases are: transpiration; VD, volume of water upstream before abstraction in
the Grand Valley (GV) in the U.S., the Nile Valley the main source.
(NV) in Egypt, and the ‘‘City 1 2020 with saving’’ case
from the U.S., all taken from Haie and Keller (2008).
Also an H (Hypothetical) case is included, which is TABLE 4. The Results (in %) of Applying Macro, Meso,
adapted from a real situation. Table 3 shows the basic and Micro-Efficiencies to the Grand Valley (GV), Nile Valley
(NV), City, and H Cases.
data as various fractions. The variables that are not
shown are taken to be 0. To determine the weights GV NV City H
that are not given in the following tables, refer to
the defaults provided in the section Dimensions of i=1
MacroEs 97.8 97.4 96.7 91.4
Usefulness.
MesoEs 80.9 96.2 90.7 90.4
Table 4 shows the results of applying the concepts MicroEs 30.6 41.2 40.4 79.2
described in this paper to these cases, and the subse- MacroEb 99.5 97.9 97.6 94.5
quent paragraphs make some general comments MesoEb 95.9 96.9 90.9 93.8
about these results. MicroEb 26.5 38.9 36.4 79.2
c=1
These differing 3ME values in Table 4 have vari-
MacroEs 62.5 91.4 76.4 89.2
ous interpretations. The general relationships men- MesoEs 61.6 91.6 81.3 89.2
tioned in the section Efficiencies are validated with MicroEs 30.6 41.2 40.4 79.2
these cases, which are as follows: cMacroEb = cMe- MacroEb 86.5 92.6 80.0 92.8
soEb, MesoE ‡ MicroE, iMicroE = cMicroE, and MesoEb 86.5 92.6 80.0 92.8
MicroEb 26.5 38.9 36.4 79.2
MicroEb = CE.
Nile Valley values demonstrate that all the differ-
ent efficiencies are almost the same and portray the
same idea, that is, the system at the higher level The case for the GV shows the real complexities
(Macro and Meso) is functioning at its peak efficiency, involved in some systems. For example, there are large
though the local, Micro-efficiencies are low. This differences between i and c efficiencies, for example, it
means that at the basin level, there are practically no shows a difference of some 35pp between MacroEs
possibilities of saving real water, that is, there are no values, and also high differences between cMacroEb
opportunities for consumptive saving (SC). However, and cMacroEs. These are indicative of high impacts of
a challenge of sustainability for this case is to pollution (in this case salt) in increasing total water
increase the Micro-efficiencies while maintaining consumption of the system, which in turn decreases
high Macro-efficiencies. This achievement is of partic- efficiencies. For example, for these two last Macro-effi-
ular importance because of benefits associated with ciencies, the total consumption increases about 39%
water abstraction savings, as is mentioned earlier. (from 3.7% to 5.1%) for GV, which indicates that salt is

JOURNAL OF THE AMERICAN WATER RESOURCES ASSOCIATION 241 JAWRA

 HAIE AND KELLER

a major problem, but a difference of only 1.3% (from when conducting any systems analysis and design. But
28.3% to 28.7%) for NV indicates a minor salt problem. in order to expand the utility of the indicators so that
The iMacroEs and cMacroEs of the City case show decision makers can use them, this paper goes beyond
a rather large difference of more than 20pp. This is the pure hydrology of a WRS and employs an important
probably due to the reason that the conservation concept called ‘‘Usefulness Criterion,’’ which is incorpo-
practices envisioned for 2020 (LFTs: Low Flush Toi- rated into the water balance through a logical approach.
lets) are abstraction type savings and not consump- In this process, two dimensions are identified: one deals
tive savings (SA and not SC). As mentioned in the with the quality and the other with the beneficial use of
above section on the Types of Water Totals, SA has water. Both of these are determined through weights
advantages but does not promote real water savings given by the manager based on physical characteristics
for downstream uses by 2020. On the other hand, low of used and reused water, system quality, and manage-
MicroE values suggest that the City itself is not effi- ment goals. Management goals are particularly relevant
cient in using its water resources and perhaps it because, for example, a decision maker or a consultant
would be better to try to increase MicroE than SA, should decide on the acceptable salt tolerance for an irri-
although if both of them are increased the system gation system design or acceptable pollution for the
becomes more sustainable. This is an interesting downstream ecosystem. In this regard it should be noted
example of how different stakeholders can have dif- that societies and political processes are getting involved
ferent solutions to water scarcity. Some advocate SA, in making judgments concerning the use of water, that
others SC, and most of the time with very costly is, what constitutes a reasonable beneficial use, which
interventions, but the real and sustainable solution then should be incorporated into the weights.
lies in integrating the two (through 3ME). After these preliminary definitions, 3ME are
For the H case, high levels of water abstraction, described for a WRS based on two total flow types for
(useful) consumption, and MicroE values are dominant water balance that incorporates different scales, flow
factors. A minimum value of flow in the river (VD) is paths, and Usefulness Criterion. MacroE links the
required by law, while recharge of the aquifer through system to the main water body of a river basin such
RP is highly desired for downstream wells. As a WRS as a river, while MicroE is of prime importance to the
reaches such a balanced high performance at the three subsystems, such as an urban or agricultural area.
levels, increasing 3ME gets harder. But scenarios of The MesoE represents an analysis of the situations
different water allocation priorities can lead to better that are happening between MacroE and MicroE and
solutions even economically. The Usefulness Criterion the flows that occur between the two. Consequently,
‘‘s’’ for RP and VD are 0.687 and 0.827, respectively, these three levels are of interest to different stake-
meaning that aquifer recharge is less useful than the holders within a river basin. And it can readily be
minimum flow requirement of the river. So which one shown that EE and CE are simplified models of 3ME.
should be increased: 79% MicroE (the owner of the sys- The framework is applied for demonstrative pur-
tem feels the better efficiency), 89% MesoE (probably poses to four different cases. Some preliminary dis-
promoting more aquifer recharge), about 90% MacroE cussions of these illustrative cases show the utility of
(more water for the river), or a combination of the 3ME. This paper presents a new framework for man-
three? Besides the hydrology of the area, this clearly aging water resources, and the complex dynamics of
needs a compromise between conflicting stakeholders. 3ME will be the focus of future studies, particularly
in line with IWRM analysis.

CONCLUSIONS
APPENDIX: EQUIVALENCY OF MINIMIZING A
DIFFERENCE AND MAXIMIZING ITS RATIO
For a long time, CE has been used to evaluate and
design WRSs and it is shown that if it is used alone,
it can lead to error. Other efficiency models are also Let Y be the minimum of the difference (F ) H) as
available, but none have been developed systemically follows:
using fundamental laws of physics such as water bal-
ance, and as such are not complete. • Y = min (F ) H), with F and H positive
In this paper, performance composite indicators are
developed in order to determine efficiency using water The solution (i.e., real values of F and H) is the
balance. Total flow used in this law of conservation of minimum value of F and the maximum value of H.
mass can be total inflow into a WRS or total consumption These two values of F and H yield the minimum for
out of it. It is important to use both of these two types (F ) H), or Y = min (F) ) max (H).

JAWRA 242 JOURNAL OF THE AMERICAN WATER RESOURCES ASSOCIATION

 MACRO, MESO, AND MICRO-EFFICIENCIES IN WATER RESOURCES MANAGEMENT: A NEW FRAMEWORK USING WATER BALANCE

Now let Z be the maximum of the ratio (H ⁄ F) as DWR (Department of Water Resources), 2009b. California Water
follows: Plan: Update 2009. Volume 2 – Resource Management Strate-
gies, Chapters 1 & 2. DWR (Department of Water Resources),
Sacramento, California.
• Z = max (H ⁄ F), with F and H positive EEA (European Environment Agency), 2009. EEA Multilingual
Environmental Glossary. http://glossary.eea.europa.eu/terminology/
The solution is the minimum value of F and the concept_html?term=consumptive use of water, accessed May 13.
maximum value of H. Therefore, although Y and Z EU (European Union), Directive 2000 ⁄ 60 ⁄ EC of the Parliament
and of the Council, 2000. Establishing a Framework for
are different, the solutions (i.e., the values for F and Community Action in the Field of Water Policy. Water Frame-
H in both cases) are the same. work Directive (WFD). Official Journal of the European Com-
munity L327:1-72.
Falkenmark, M., 2005. Water Usability Degradation. Water Inter-
national 30(2):136-146.
Garcı́a, J.H., M.T. Heberling, and H.W. Thurston, 2011. Optimal
SUPPORTING INFORMATION Pollution Trading Without Pollution Reductions: A Note.
Journal of the American Water Resources Association 47(1):52-
58.
Additional Supporting Information may be found Haie, N. and A.A. Keller, 2008. Effective Efficiency as a Tool for
in the online version of this article: Sustainable Water Resources Management. Journal of the
American Water Resources Association 44(4):961-968.
Data S1. Easy to use spreadsheet that calculates Horan, R.D. and J.S. Shortle, 2011. Economic and Ecological Rules
the efficiencies, both for ‘‘s’’ and ‘‘b,’’ developed in this for Water Quality Trading. Journal of the American Water
paper, that is, 3MEs based on Equations (9). Resources Association 47(1):59-69.
Please note: Neither AWRA nor Wiley-Blackwell is Hussain, I., H. Turral, D. Molden, and M.U.D. Ahmad, 2007. Mea-
responsible for the content or functionality of any suring and Enhancing the Value of Agricultural Water in Irri-
gated River Basins. Irrigation Science 25(3):263-282.
supporting materials supplied by the authors. Any Israelsen, O.W., 1932. Irrigation Principles and Practices. John
queries (other than missing material) should be direc- Wiley and Sons, New York.
ted to the corresponding author for the article. Keller, A.A. and J. Keller, 1995. Effective Efficiency: A Water Use
Efficiency Concept for Allocating Freshwater Resources. Discus-
sion Paper 22 (January), Center for Economic Policy Studies,
Winrock International, Arlington, Virginia.
ACKNOWLEDGMENTS Marcus, R.R. and S. Kiebzak, 2008. The Role of Water Doctrines in
Enhancing Opportunities for Sustainable Agriculture in Ala-
Naim Haie thanks the Portuguese Foundation for Science and bama. Journal of the American Water Resources Association
Technology (FCT) for their partial financial support that made this 44(6):1578-1590.
project possible. Both authors appreciate the review and the com- Molden, D. and R. Sakthivadivel, 1999. Water Accounting to Assess
ments made by Dr. David Molden of International Water Manage- Use and Productivity of Water. Water Resources Development
ment Institute (IWMI). Finally, the authors would like to thank 15(1 ⁄ 2):55-71.
the five anonymous referees and the Associate Editor for their RCM (Resolução do Conselho de Ministros), 2005. Programa de
helpful comments, and convey our special appreciation to the Acompanhamento e Mitigação dos Efeitos da Seca 2005.
Editor for his understanding and scientific objectivity. Resolução do Conselho de Ministros No. 83 ⁄ 2005. RCM, Lisbon,
Portugal (in Portuguese).
Rifkin, J., 1980. Entropy: A New World View. New York. Also
available at: The Entropy Law. http://miller.public.iastate.edu/
LITERATURE CITED civics/Rifkin-The%20Entropy%20Law.pdf, accessed February
2009.
Allen, R.G., A.J. Clemmens, and L.S. Willardson, 2005. Agro- UN (United Nations), 2006. Water: A Shared Responsibility. World
Hydrology and Irrigation Efficiency. http://www.kimberly.uidaho. Water Development Report 2. UNESCO, Paris, France.
edu/water/papers/converse_use_def/Allen_2005_ICID_Agro_ Vance, W.A., 2005. Role of Science and Engineering in Decision-Mak-
Hydrology_and_efficiency2b.pdf, accessed October 10, 2011. ing Within the State and Regional Water Boards. State Water
ASCE, 2004. Regulated Riparian Model Water Code. EWRI ⁄ ASCE Resources Control Board (SWRCB), Sacramento, California.
40-03. American Society of Civil Engineers, Virginia. Venot, J.P., B.R. Sharma, and K.V.G.K. Rao, 2008. The Lower
Ayers, R.S. and D.W. Westcot, 1994. Water Quality for Agriculture. Krishna Basin Trajectory: Relationships Between Basin Devel-
FAO Irrigation and Drainage Paper 29 Rev 1. FAO, Rome, Italy. opment and Downstream Environmental Degradation. IWMI
Burt, C.M., A.J. Clemmens, T.S. Strelkoff, K.H. Solomon, R.D. Research Report 125, 36 pp., International Water Management
Bliesner, L.A. Hardy, T.A. Howell, and D.E. Eisenhauer, Institute, Colombo, Sri Lanka.
1997. Irrigation Performance Measures: Efficiency and Unifor- Willardson, L.S., R.G. Allen, and H.D. Frederiksen, 1994. Elimina-
mity. Journal of Irrigation and Drainage Engineering, ASCE tion of Irrigation Efficiencies. 13th Technical Conference (Octo-
123(6):423-442. ber). USCID, Denver, Colorado.
Clemmens, A.J., R.G. Allen, and C.M. Burt, 2008. Technical Con- WRB (Water Rights Bureau), 2009. Application for Beneficial
cepts Related to Conservation of Irrigation and Rainwater in Water Use Permit. The Water Resources Division of the State of
Agricultural Systems. Water Resources Research 44:W00E03. Montana. http://dnrc.mt.gov/wrd/water_rts/wr_general_info/wrforms/
DWR (Department of Water Resources), 2009a. Strategic Planning 600-b.pdf, accessed July 2009.
Goals. State of California, USA. http://www.water.ca.gov/about/
mission.cfm, accessed July 2009.

JOURNAL OF THE AMERICAN WATER RESOURCES ASSOCIATION 243 JAWRA