Water security: from abstract

concept to meaningful metrics

An initial overview of options
Nathaniel Mason and Roger Calow

Working Paper 357

Results of ODI research presented
in preliminary form for discussion
and critical comment

shaping policy for development

 Working Paper 357

Water security: from abstract concept to

meaningful metrics
An initial overview of options

Nathaniel Mason and Roger Calow

October 2012

ODI
203 Blackfriars Road,
London
SE1 8NJ
www.odi.org.uk

* Disclaimer: The views presented in this paper are those of the author(s) and do not necessarily
represent the views of ODI or DFID
ii
 Acknowledgements
This paper has benefited from detailed review and comments from the following staff of the UK
Department for International Development: Jean-Paul Penrose, Water Resources Adviser; John
Carstensen, Head of Profession, Climate and Environment; and Guy Howard, WASH Team
Leader. Information on specific initiatives was also gratefully received from the following sector
experts: Tom Slaymaker, Deputy Head of Policy and Senior Policy Analyst, Governance, WaterAid
(on the Joint Monitoring Programme Post-2015 working groups); Dominick de Waal, Senior
Economist, World Bank Water Supply and Sanitation Program (on the African Ministers’ Council on
Water’s pan-African Monitoring and Evaluation initiative); and Mike Muller, Global Water
Partnership Technical Committee (on the World Water Assessment Program’s Expert Group on
Indicators, Monitoring and Databases, and Pilot Study on Indicators). Within ODI, several staff also
provided comments and guidance: Peter Newborne, Research Associate, Water Policy; Lindsey
Jones, Research Officer, Water Policy/ Climate Change, Environment and Forests; and Emily
Wilkinson, Research Fellow, Climate Change, Environment and Forests. Finally, thanks go to
Naomi Oates for editing.

ISBN 978 1 907288 95 1

Working Paper (Print) ISSN 1759 2909
ODI Working Papers (Online) ISSN 1759 2917

© Overseas Development Institute 2012

Readers are encouraged to quote or reproduce material from ODI Working Papers for their own publications, as long as they are not
being sold commercially. For online use, we ask readers to link to the original resource on the ODI website. As copyright holder, ODI
requests due acknowledgement and a copy of the publication.
iii
 Contents
Executive summary viii
1 Introduction 1
2 Rationale: opportunities to set the goalposts 3
2.1 The Post-2015 Development Agenda and SDGs 3
2.2 The Green Economy 5
2.3 Understanding water security at country level 6
3 Scarcity, risk and security: competing framings 8
3.1 Scarcity 8
3.2 Risk 12
3.3 Security 14
3.4 An expansive and inclusive framing of water security 18
4 Metrics and meaning: key considerations for measuring water security 20
4.1 Beyond the physical resource: capacity to access, use and manage 20
4.2 Variability and risk: spatial and temporal factors 24
4.3 Human focus: water for health and livelihoods 26
4.4 Environmental needs: biodiversity and ecosystems 28
4.5 Competition and conflict: governance arrangements 29
5 Politics and pragmatism: the architecture for data gathering and interpretation 31
5.1 Data difficulties: variability, complexity and politics 31
5.2 The existing architecture: databases and initiatives 32
5.3 Future aspirations 34
6 Recommendations 38
6.1 Introduction to proposed indicators and caveats 38
6.2 Water security indicator options 40
Annex 1: The various UN processes around the SDGs and Post-2015 Development
Agenda 52
References 53

iv
 Tables, figures and boxes
Figure 1: Areas of physical and economic water scarcity 9
Figure 2: Groundwater storage and freshwater availability in Africa 12
Figure 3: Fifteen-model mean changes in (a) precipitation (%), (b) soil moisture content (%),
(c) runoff (%), and (d) evaporation (%) for the last decade of the 21st century, relative to
the last decade of the 20th century 13
Figure 4: Water stress in Africa as percentage of the population computed with increasing
resolution 23
Figure 5: Positioning proxy indicators in relation to the challenge of increasing productivity of
irrigated agriculture: according to pragmatic considerations (y-axis) and relevance to sector
decision-making (x-axis) 37

Box 1: Water Security: the right norm for water management? 1

Box 2: Metrics terminology 2
Box 3: Efforts to measure WASH and what this tells us about water security metrics 3
Box 4: The Sustainable Energy for All proposal: a chance to integrate developmental and
environmental paradigms? 4
Box 5: A pan-African monitoring and evaluation initiative 6
Box 6: Groundwater storage: a missing piece in water security assessments? 10
Box 7: Water poverty as human insecurity 17
Box 8: Key themes for a water security metrics framework 19
Box 9: Enhancing estimates of TARWR: Pilot Study on Indicators (PSI) 21
Box 10: IWRM: resourcing and operationalising 29
Box 11: Fifteen indicators proposed by the UN Water Task Force on Indicators, Monitoring
and Reporting: data challenges 33
Box 12: Using proxies to measure progress 35

Table 1: Comparison of groundwater and surface water resources 10

v
 Acronyms
AMCOW African Ministers’ Council on Water
BIP Biodiversity Indicators Partnership
CAFOD Catholic Overseas Development Agency
CGIAR Consultative Group on International Agricultural Research
CIWA Cooperation on International Waters in Africa
CMI Climate Moisture Index
CoFR Committee on Foreign Relations (US Senate)
CUNY City University of New York
CVI Climate Vulnerability Index
DCDC Development Concepts Doctrine Centre, Ministry of Defence (UK)
DEG German Investment Corporation (Deutsche Investitions- und
Entwicklungsgesellschaft)
DFID Department for International Development (UK)
DIE German Development Institute (Deutsches Institut für
Entwicklungspolitik)
DRC Democratic Republic of Congo
DRR Disaster Risk Reduction
ECDPM European Centre for Development Policy Management
EG-IMD Expert Group on Indicators, Monitoring and Databases
FAO Food and Agriculture Organization (UN)
GAR Global Assessment Report (on Disaster Risk Reduction)
GEMI Global Environmental Management Initiative
GEMS Global Environment Monitoring System
GIS Geographic Information Systems
GLAAS Global Annual Assessment of Sanitation and Drinking Water
GTN-H Global Terrestrial Network for Hydrology
GWP Global Water Partnership
GWSP Global Water System Project
HDR Human Development Report
ICF International Climate Fund
ICOLD International Commission on Large Dams
IGRAC International Groundwater Resources Assessment Centre
IISS International Institute for Security Studies
INBO International Network of Basin Organisations
IPCC Intergovernmental Panel on Climate Change
IWMI International Water Management Institute
IWRM Integrated Water Resources Management
JMP Joint Monitoring Programme
MDG Millennium Development Goal
MEWINA Monitoring and Evaluation for Water in North Africa
MRI Mortality Risk Index
N-AMCOW North African Ministers’ Council on Water
NGO Non-Governmental Organisation
NIC National Intelligence Committee
ODI Overseas Development Institute
PSI Pilot Study on Indicators
RBO River Basin Organisation
SDG Sustainable Development Goal
SEEA(W) System of Environmental-Economic Accounting (for Water)
SSA sub-Saharan Africa
SWAR Surface Water Runoff
TARWR Total Actual Renewable Water Resources
TF-IMR Task Force on Indicators, Monitoring and Reporting
vi
UN United Nations
UNCSD United Nations Conference on Sustainable Development
UNDP United Nations Development Programme
UNEP United Nations Environment Programme
UNESCO United Nations Educational, Scientific and Cultural Organization
UNICEF United Nations Children’s Fund
UNISDR United Nations International Strategy for Disaster Reduction
UNSD United Nations Statistics Division
USACE US Army Corps of Engineers
WASH Water Supply, Sanitation and Hygiene
WBCSD World Business Council on Sustainable Development
WDPA World Database on Protected Areas
WEF World Economic Forum
WHO World Health Organization
WPI Water Poverty Index
WRI World Resources Institute
WRM Water Resource Management
WWAP World Water Assessment Programme
WWDR World Water Development Report
WWF World Wildlife Fund

vii
 Executive summary
With renewed global awareness of planetary boundaries and resource constraints, water’s vital role
in underpinning equitable, stable and productive societies, and the ecosystems on which we
depend, is undisputed. Water security has emerged as a powerful concept to encapsulate the many
competing objectives of water resource management, and is increasingly gaining traction in global
debates and the agendas of governments, businesses and NGOs. While deliberation continues
about exactly how far the scope of the term extends, the emphasis to date has been at a
theoretical, qualitative level. While this is vital, we have to be able to measure progress in more
rigorous terms if we are to translate water security from abstract concept to a meaningful tool to
guide policy and practice.

This paper is a first attempt to meet this need. It responds simultaneously to two concerns. On the
one hand a political concern, to articulate the objectives of water management in aspirational terms.
On the other hand, a technical or operational concern, to know what we are dealing with and how
much progress we are making on water management, in clear, measurable terms.

Bridging between these political and technical concerns is becoming increasingly urgent as 2015
approaches, by which point a global framework of goals, targets and indicators needs to be defined
to succeed the Millennium Development Goals (MDGs). The move to develop Sustainable
Development Goals (SDGs), potentially applying to all countries and placing greater emphasis on
the natural capital underpinning human development, could provide greater room for water than the
current MDG targets, which focus on drinking water supply and sanitation. The process to define
SDGs is only now beginning, in parallel to the UN Secretary General’s existing initiative on the
Post-2015 Development Agenda. The integration of these parallel processes, and the place of
different resources including water within them, is therefore of increasing interest. Water security is
emerging as a possible unifying concept for the different things water managers are trying to
achieve, which could therefore be relevant in thinking about how to frame global goals and targets
on development and the environment. At the same time, irrespective of global policy agendas,
developing country governments and donors continue to be faced with pressing challenges about
how best to manage and develop water resources for the benefit of people, ecosystems and
economies. There is, then, an acute need to identify appropriate water security metrics at national
level also.

The paper is written primarily from the technical perspective, with a pragmatic focus on what can be
measured: the emphasis is therefore on indicators and the availability and quality of underlying
data. At the same time it retains the political perspective, with attention to the aspirational debates
about what should be measured. The concept of water security is relatively young, and carries
different associations with longer-established concepts, notably national security and human
security, as well as food and energy security. While there are a number of definitions, overall
consensus on what water security means has not yet been reached. The related concepts of water
scarcity and water risk have also generated considerable debate. After setting out the rationale for
the research in greater detail, the paper reviews the three concepts of water scarcity, risk and
security. Building on this analysis, it identifies five key themes which are arguably encompassed by
the emerging concept of water security, and which can help structure the development of a
pragmatic, yet aspirational, metrics framework:

• Water security goes beyond immediate physical availability: water in the atmosphere,
on the surface and below ground interacts in complex ways, with different responses to
human impacts; availability in any given period or place is furthermore moderated by
the economic and social capacity to access water.
• Water security requires us to address variability and risk: while water security implies
permanence, spatial and temporal variability is inherent to water systems. As variability
amplifies, and where we do not have the capacity to adapt, it translates into water-
related risks, including flood, drought and pollution.
viii
 • Water security needs a human focus: to be real and meaningful, beyond technical and
policy circles, water security has to focus on the needs of individuals, especially the
poor and vulnerable. The water security of all matters equally, irrespective of social,
economic or political disparities.
• Water security also requires us to meet environmental needs: whether viewed as
intrinsically valuable, or valuable for the services they provide, freshwater ecosystems
require protection. Ecosystem water requirements may vary over time, and must be met
in terms of both quantity and quality.
• Water security requires management of competition and conflict: given the breadth of
human and environmental needs which must be met, there are inevitable tradeoffs,
particularly in those areas where water is intensively used, or where withdrawals are
rapidly accelerating. The institutional capacity to avoid or resolve these tradeoffs, and
mediate between the claims of competing users through rules-based systems rather
than force or coercion, is therefore essential.

Each of these themes presents different challenges in terms of measurement, which are
considered in turn. Physical availability of water is, at first sight, the fundamental concern, and
appears relatively straightforward to measure. But even the most basic indicators for availability are
fraught with conceptual and methodological difficulties, including accounting for complex
interactions of ground and surface water, and in any case omit how water security is mediated by
demand for and capacity to access water. Variability and risk are difficult to measure in and of
themselves, whether in probabilistic or more qualitative terms, as are corresponding concerns, such
as society’s capacity to adapt. Human-focused measures of water security have been
comparatively well-developed, particularly in relation to health via the drinking water and sanitation
targets of the current MDGs. But there has been less attention paid to measuring the extent, costs
and benefits of other human uses of water, particularly those associated with agriculture and
industry. Environmental requirements are highly context-dependent and likely to vary seasonally,
making generic measurement difficult. Measures of institutional capacity, required for example to
allocate across competing uses, have tended to be conceived in terms of process, which may not
lead directly to substantive outcomes.

Methodological difficulties in defining appropriate indicators are compounded by gaps in the

availability and quality of underlying data. Even for key data items like average renewable water
resources (flowing in and out of a country or falling as precipitation) internationally consolidated
data is not available for all countries, especially for key components such as groundwater, and is
rarely updated. Data quality and availability are further constrained by reluctance of countries and
other entities, such as corporations, to share information on water. Important initiatives have been
undertaken, notably by the World Water Assessment Programme and UN-Water, but overall the
architecture for water monitoring is marked by a lack of coordination and collaboration.

But while political and technical challenges around data and indicators abound, this paper does not
focus only on constraints. Consideration is given to the positive lessons of other initiatives on
metrics – for example monitoring of progress on water supply and sanitation up to and beyond
2015, and efforts to develop targets and indicators on energy. The potential role of innovative
methods is also considered – for example placing more emphasis on proxy indicators, or utilising
new technologies such as remote sensing. Above all, the paper closes by identifying a range of
indicator options in relation to each of the five identified themes, a selection of which could feasibly
be employed for recurrent monitoring of different aspects of water security at national and global
level. Looking beyond what is currently, pragmatically possible to measure, for each theme a
second, more aspirational set of indicator options is also proposed, which would require further
effort in data collection and interpretation. For each of the proposed indicators data sources,
calculation methods, technical notes and an assessment of the potential policy implications of their
use are given. The options are proposed to prompt debate about which indicators are appropriate,
in the knowledge that only a selection of the proposed indicators is required, and alternatives or
additional options may be available. This technical debate needs to evolve on a parallel, iterative
basis to political debates about how water security should be defined, to ensure pragmatism does
not limit aspiration, and aspiration does not ignore what is pragmatically possible.
ix
1

1 Introduction
This paper outlines options for the development of a set of metrics for monitoring water security,
principally at the level of countries. Multiple definitions have been proposed for water security,
reflecting the desire to articulate, in a few words, the objectives of water resource management in
general, as well as reflecting interest in other related ‘security’ agendas including food, energy,
national and human security. This paper does not propose another ‘definitive’ characterisation of
water security. Rather, it considers existing definitions of water security and related terms, notably
water scarcity and risk, to identify the themes in relation to water security that recur in ongoing
debate (Box 1). From there, workable options for measuring progress on each component are
identified, as a first step in moving from an abstract concept, to measurable policy targets.

Outside drinking water and sanitation coverage, which have become the focal indicators for the
water sector as a whole due to the prominence of Millennium Development Goal (MDG) target 7c,
there is little consensus, nor any unified international architecture, for monitoring progress on water
security, broadly conceived.

Box 1: Water security: the right norm for water management?

This paper responds to the increasing prominence of water security in policy and academic usage
(Cook and Bakker 2012). However, it should be noted that the term does not have the
endorsement of all those working on water issues. Even if understood broadly, there is an
argument that the word ‘security’ will always carry militaristic overtones, or will imply that solutions
to water problems will be achieved by force, rather than negotiation and cooperation. Proponents
of the term therefore need to monitor the way it is being interpreted by different actors – its place
as a useful and universally endorsed term is not yet assured. At the same time, the remainder of
this paper proceeds on the assumption that ‘water security’ is currently the simplest and most
widely accepted term to articulate the outcome of sound water management, and therefore the
mission of the water management community.

Combined with other challenges, such as the low economic value placed on water as a resource,
and the complexity of natural and anthropic water systems, the absence of clear indicators and
targets has militated against government and donor attention to water resource management.
Assessment of water resources to date has tended to provide either broad measures focusing on
availability (Falkenmark, Lundqvist and Widstrand 1989; Seckler et al. 1998); complex composite
indicators for human and ecological water threats (Vörösmarty et al. 2010); or has focused on
process rather than outcomes, for example the number of countries that have developed
Integrated Water Resource Management (IWRM) plans or instituted river basin organisations
(AMCOW 2012a; UNEP 2012). Routine monitoring of these measures has been constrained by
data availability and a fragmented institutional architecture.

At the same time, awareness of the importance of the resource base for human development and
the ecological systems on which we depend is growing. Global meetings such as the UN
Conference on Sustainable Development in Rio (Rio+20) have flagged the need to frame new
goals, targets and indicators beyond 2015. There is a re-emergent desire to unite, or at least not to
further polarise, the spheres of environment and development (Melamed, Scott and Mitchell 2012),
notably under a framework of Sustainable Development Goals (SDGs) that could go significantly
beyond the current MDGs in scope and application (potentially applying to all, rather than just
developing countries). A broad conception of water security should therefore allow for recognition
of our universal dependency on water as a fundamental form of natural capital, in a way that
recognises both an environmental dimension (protecting the resource for our own and future
generations) and a developmental one (providing access to sufficient water to permit all to fulfil
their capabilities).
2

The paper provides a number of options that could form part of a framework of metrics for
assessing progress on water security, at national and global level, to help focus attention and
potentially to direct finance and capacity for water security to those countries most in need. The
focus is on water security outcomes rather than intermediate outputs or processes. The paper
starts from the recognition that a metrics framework comprises a family of components (Box 2) and
requires attention to the scientific, objective and empirical, as well as the political, moral and
normative.

Box 2: Metrics terminology

The following definitions clarify the terminology around metrics used in this paper:
• Goal. A broad statement of a desired, usually longer-term, outcome of a
program/intervention.
• Target. The objective a program/intervention is working towards, expressed as a
measurable value; the desired value for an indicator at a particular point in time.
• Indicator. A quantitative or qualitative variable that provides a valid and reliable
way to measure achievement, assess performance, or reflect changes connected
to an intervention.
• Monitoring. Routine tracking and reporting of priority information about a
program/project, its inputs and intended outputs, outcomes and impacts.

The UK Department for International Development (DFID), in its guidance on monitoring for
the projects and programmes it funds, describes a ‘Results Chain’ which moves from input,
through process activities, to output, to outcome, and thence impact. Indicators or specific
deliverables are used to track progress at each link in the results chain. Impact can be
broadly associated with Goal in the sense outlined above and is described by DFID as ‘a
higher-level situation which the project will contribute towards achieving’. Although Goal is
used throughout this paper for consistency with the hierarchy of goals, targets and indicators
established with the MDGs, the DFID thinking around a ‘results chain’ may also be useful.

Source: For general definitions, UNAIDS (2010), for DFID-specific terminology, DFID (2011)

The paper is structured as follows: The rationale for re-examining water security and related
metrics is established in Section 2, with reference to key applications in policy and practice, for
example around the SDGs, the green economy and the continued need to provide guidance to
policy makers and practitioners at different levels on whether water resource development and
management is moving in the right direction. These potential applications are likely to influence the
perceived need for, and shape of, any water security metrics framework. Section 3 further sets the
scene, providing a simplified overview of the main framings of water security, current among
different constituencies, as well as the related terms water scarcity and water risk. The section
closes by identifying five constituent themes and a working definition, which provide a reference
point for subsequent analysis. Section 4 examines how different aspects of water security are
being measured by existing indicators and approaches. In the light of the preceding analysis,
Section 5 considers the current architecture for coordinating and undertaking monitoring, with close
attention to the serious challenges of data quality and availability. Section 6 concludes with a
number of recommendations for potential indicators to underpin a metrics framework. Indicator
options are presented according to two levels of aspiration: a scenario in which limited further
resources are made available and existing data sources and monitoring systems must be used,
versus a more aspirational scenario under which a coordinated international effort is made to
strengthen the gathering and interpretation of water-related data.
3

2 Rationale: opportunities to set the goalposts

This paper is intended to inform a number of emerging, and evolving agendas. Given this aim, the
following section is dedicated exclusively to unpacking those agendas and identifying the potential
opportunities for engagement around water security and its measurement.

2.1 The Post-2015 Development Agenda and SDGs

Water’s place in any post-2015 framework of goals and targets has, to date, been most extensively
explored in relation to water supply, sanitation and hygiene (WASH), thanks largely to the efforts of
the Joint Monitoring Programme (JMP) of UNICEF and the World Health Organization (WHO) to
devise relevant and feasible goal, target and indicator options (Box 3) Meanwhile, the SDGs were
a key feature of Rio+20 (20-22 June 2012), with one of the few substantive results from the
conference being a commitment to define and agree them. The process on SDGs is initially to be
led by a 30 strong panel nominated by member states with inputs from the UN Secretary General,
though no deadline has been specified for their agreement (UN, 2012). The Rio+20 outcome
document reflects various themes around water, including WASH and others such as ‘floods,
droughts and water scarcity’, and ‘the role of ecosystems in maintaining water quantity and quality’
(UN 2012: 23-24).

The proposal to develop SDGs came with the UN Secretary General already having initiated a
process on the ‘Post-2015 Development Agenda’, which has a less explicit emphasis on the
environmental dimensions of development. There are now two processes running in parrallel, and
it is not yet clear as to how they will be reconciled.

Box 3: Efforts to measure WASH and what this tells us about water security metrics
Several working groups of invited experts are convening between 2011 and 2013 to consider how
progress on WASH should be measured, once the MDG deadline is reached in 2015. The JMP is
able to derive estimates of what kinds of water sources or sanitation facilities are used by what
proportion of the population. But attempts to measure other important considerations (for example
water quality, the extent of treatment of wastewater, or the sustainability and affordability of
services) have faced difficulties, notably around data availability and accessibility. Identifying a
window of opportunity in the run-up to 2015, the working groups are tasked with developing options
for extending the scope of WASH indicators and targets. There is also a wider public consultation
process.

Concerted and well-coordinated efforts around WASH metrics provide an instructive example for
the broader water security agenda. In the first place, globally agreed goals and targets for water
supply and sanitation have helped direct resources, not only to the subsectors themselves, but
also to the monitoring architecture – exemplified by the initiation of the JMP, and now the post-
2015 working groups. Simplicity has been a key hallmark of the MDG indicators on water supply
and sanitation, which has aided their uptake as policy and communications tools. At the same time
the working groups, drawing on the expertise of many different agencies and interests, show that
agreement on difficult decisions about what matters, and how to measure it, needs to be obtained
as part of a systematic and consultative process. Finally, the JMP and working groups are
circumspect about their ability to dictate the place of WASH in any global post-2015 architecture,
describing the goal as being to ‘pave the path for the menu of options to be offered for
consideration by the UN-member States in their deliberations on post-2015 goals and targets’
(JMP, 2012). A similar degree of circumspection may be required by proponents of water security,
for example in considering synergies with other resource agendas, notably around the water-food-
energy nexus.

Source: JMP (2012)

4

At root, this is a debate about whether social and economic objectives (a developmental paradigm)
can be combined with sustainable management of our environment (an environmental paradigm),
especially if the latter is perceived as growth-limiting (Melamed, Scott and Mitchell 2012). There
are other, related uncertainties, for example around whether the new goals should apply
universally or only to developing countries, or should be aspirational or binding (Evans and Steven
2012).

As elaborated in this paper an expansive and inclusive framing of water security has the potential
to synthesise developmental and environmental objectives – for example equitable access to
productive uses of water, and protection from water-related disasters, while safeguarding minimum
flows to protect ecosystem services. But in reality those objectives can often appear to be in
competition, presenting tradeoffs rather than opportunities for synthesis: for example conventional
approaches to flood protection may involve heavy engineering (e.g. dams and dikes) which can
disrupt natural freshwater flows on which ecosystems depend. This implies the need for a
measurement framework that can permit meaningful assessment and comparison of progress
against potentially competing objectives. The proposals which have developed from the
Sustainable Energy for All Initiative offer an example from another issue at the bridge (or faultline)
between environment and development (Box 4).

Box 4: The Sustainable Energy for All Initiative: a chance to integrate developmental and
environmental paradigms?
The UN Secretary-General’s proposal on sustainable energy for all comprises three objectives
(Sustainable Energy for All Initiative 2012). One of these speaks primarily to the developmental
paradigm (universal access) while the other two have greater kinship with the environmental
paradigm (energy efficiency and renewable energy):

• Ensuring universal access to modern energy services.

• Doubling the global rate of improvement in energy efficiency.
• Doubling the share of renewable energy in the global energy mix.

Importantly, the underlying goals do not necessitate any major trade-offs, in and of themselves.
Could the Sustainable Energy for All proposal provide lessons for the identification of metrics for
water security or water resource management?

Energy differs from water, in that renewable forms are, to some extent, unlimited (whereas water is
theoretically renewable but limited in absolute availability). By rapidly deploying renewable
technologies there is potential to increase access, with attendant social and economic benefits,
without necessarily exerting pressure on the environment, at least in terms of safe levels of
greenhouse gas emissions. This so-called ‘triple-win’ is particularly open to developing countries
that have yet to put in place long-lived, expensive energy generation and distribution capital (ODI,
DIE and ECDPM 2012). There are certainly potential trade-offs, for example the water (and land)
requirements of renewable technologies such as large scale solar, biofuels and hydropower, or the
risk of a rebound effect from increasing efficiency, but these depend on how the objectives are
achieved, not on the objectives themselves. The triple-win is less apparent in the case of water,
since on first impression every increase in access without an attendant increase in efficiency would
appear to create greater pressure on the finite resource. But this would be to frame the challenge
in a way that privileges the global limit over local realities. As explored in Section 3, patterns of
access and availability are locally heterogeneous. In most countries there is more than enough
water to meet basic needs and fulfil the human right to water, i.e. personal and domestic uses (UN
n.d.), while highly consumptive uses, such as irrigation, in already water-scarce catchments are of
course less likely to be sustainable. In this context increasing water productivity, e.g. ‘crop per
drop’, also opens up the space for a ‘triple win’ (social, environmental and economic) in water
management. However, all objectives are only as good as the metrics that underpin them –
Sections 4 to 6 consider what is, and what might be, available to this end.
5

In reflecting on potential directions for engagement with the SDG and Post-2015 Development
Agenda processes, caution should be sounded about when and how to promote issue-specific
agendas, such as water, within the overarching negotiations of a new agreement on global goals.
There are concerns that premature arguments around the specific content of a new set of goals
and targets could distract from the immediate task of building consensus on fundamental questions
(e.g. scope and applicability). Melamed (2012) points to the long gestation period for the MDGs, in
an era in which there was broad support for multilateralism and long-run prosperity among nations
belonging to the Organisation for Economic Co-operation and Development (OECD). It can be
argued that the current context is very different, and less auspicious – especially given the limited
progress in international negotiations around some of our most pressing environmental challenges,
notably climate change (Melamed, Scott and Mitchell 2012). In this context, while work needs to be
done now to define options on issue-specific indicators, goals and targets, careful consideration is
needed as to how and when to enter the fray with issue-specific agendas. It is worth considering
the profusion of parallel processes and events which will help define the fundamental architecture.
Annex 1 provides a visual guide to these processes, elaborated by CAFOD.

2.2 The Green Economy

The other theme for Rio+20, the ‘green economy’, also presents a window to consider how to
measure the effectiveness of water resource management and progress towards water security,
particularly given a need for monitoring and decision-making tools that can integrate environmental
and growth/ poverty reduction objectives (Melamed, Scott and Mitchell 2012). This need is implicit
in most working definitions of the green economy, including UNEP’s, which describes the green
economy ‘as one that results in improved human well-being and social equity, while significantly
reducing environmental risks and ecological scarcities’ (UNEP 2011).

The concept of the green economy has emerged from a long tradition of thinking on sustainable
development, which can be traced back at least to the 70s (Runnals 2011). The economic crisis
and renewed realisation of environmental degradation has catalysed a desire to articulate
workable, alternative economic paradigms. The green economy featured as one of the themes for
Rio+20, alongside the institutional framework for sustainable development. The Rio+20 outcome
document included the green economy as ‘one of the important tools available for achieving
sustainable development’, but following concerns from developing and emerging economies that
the concept is intrinsically growth-limiting, numerous caveats were added, for example ‘each
country’s national sovereignty over their natural resources taking into account its national
circumstances, objectives… and policy space’ (UN 2012: 9-10).

Though the green economy can be viewed as qualitatively different to previous articulations of
sustainable development, it faces the same problem of being difficult to operationalise. A key
realisation has been that we currently lack the monitoring and accounting frameworks to allow
policy-makers to evaluate trade-offs between economic, social and, particularly, environmental
objectives. With respect to the latter, there has been energetic work on the subject of natural
capital accounting, which allows countries to compare their economic progress against their stock
of natural capital, for example freshwater or agricultural land. Joseph Stiglitz has pointed out that,
to date, countries make their economic decisions on the basis of an income statement (Gross
Domestic Product) and, unlike companies, do not have a balance sheet against which to set their
economic progress.

Natural capital accounting is more evolved for material resources (e.g. timber and fisheries) than it
is for the more fundamental forms of natural capital (e.g. water and land) and ecosystem services
(e.g. flood protection, water filtration) which form the basis for material resources. The UN
Statistical Commission has recently adopted the System for Environmental and Economic
Accounts (SEEA) which includes material resources. But while many countries also wish to apply
accounting to ecosystem services, there is little agreement on the methodology for doing so (World
Bank 2012a). Nevertheless, although more work needs to be done, the green economy agenda
6

presents an opportunity to frame workable ways to account for society’s impact on water – whether
in terms of the raw resource (i.e. water quantity and quality), the goods produced from it (e.g.
fisheries, or rainfed and irrigated crops) or the ecosystem services dependent on it (e.g. flow
regulating capacity of wetlands).

2.3 Understanding water security at country level

Beyond the global debates highlighted above, individual country governments are grappling with
the challenge of water resource management, with varying degrees of success – above all
attempting to allocate a spatially and temporally variable resource across multiple uses, each with
differing social, economic and environmental costs and benefits. Similarly, with a resurgent interest
in natural resource management generally, and the prominent place of water in climate change
impacts, donors are seeking to understand how they can direct support for water resource
management (WRM) to the areas most in need, as well as to assess ‘results’ and account for the
cost-effectiveness of their investments to their own citizens.

For this task reliable metrics are critical, but quality and availability of data in many countries is
severely lacking at basin, national and global level in turn (WWAP, 2012). This points to the
interconnectedness of metrics frameworks at different spatial scales and for different policy
purposes. The issue of what data can reliably be used to frame goals and targets at international
level, for global comparison as per the MDGs (and SDGs), is often bound up with what is available
at national and sub-national level.

Box 5: A pan-African monitoring and evaluation initiative

The initiative, led by AMCOW, is intended to assist in assessing progress made on the Sharm-El-
Sheikh commitments on water and sanitation. As such, the pan-African Monitoring and Evaluation
initiative is to some degree focused on policy goals defined at supra-national level. Nonetheless,
the Sharm-El-Sheikh commitments were agreed by heads of state and member governments of
the African Union and the initiative thus has scope both to respond to, and to potentially help
strengthen, the monitoring capacity of African governments facing water management challenges
in their own countries.

The initiative guidelines issued to country governments do not specifically refer to water security,
but pick up on numerous themes encompassed by a broad understanding of the term (elaborated
in this report in Section 3.4), including productive use of water, good management of different
water resources (groundwater, rainwater and transboundary resources), improving access to
WASH, and institutional aspects. According to the reporting guidelines, the selection of data and
indicators have undergone extensive consultation with different institutions working on water-
related monitoring, ‘taking into consideration the unique situation of the opportunities and
challenges in Africa’s water sector, especially with regard to data acquisition and analysis’
(AMCOW 2012b). Although in its early stages, the ability of the initiative to generate information
across 25 performance categories, drawing on 15 discrete indicators, will be a key test of what is
currently feasible in terms of water-related monitoring in African countries, whether or not this is
framed under the label of water security. This will in turn provide valuable lessons on where
countries need support, with a view to monitoring for both domestic and transnational purposes.

Source: AMCOW (2012b)

While intensive use and degradation of water resources has been much debated in the Asia
context, accelerating investment in water development in Africa is receiving growing attention,
‘with almost all countries lacking the human, economic and institutional capacities to effectively
develop and manage their water resources sustainably’ (WWAP 2012: 177). As part of its
response to this challenge, the African Ministers’ Council on Water (AMCOW) has launched an
initiative to consolidate numerous indicators across different aspects of water resource
7

management under a ‘pan-African Monitoring and Evaluation and Reporting Format’ (Box 5). An
equivalent initiative is also underway from the North African Ministers’ Council on Water - the
Monitoring and Evaluation for Water in North Africa (MEWINA) project (N-AMCOW 2012).
Although these are regional initiatives they have a strong focus on understanding, and potentially
contributing to, monitoring and evaluation capacity at country level and below.

In general donors are inclined to support country governments on water related monitoring and
evaluation capacity as an intrinsic part of their efforts to enhance water resource management, for
example the initiative described in Box 5 is funded by the German government. But they may also
be concerned to enhance WRM metrics with a view to their own programming – identifying
priorities and assessing value for money. As an example, DFID estimates that its recent approval
and pipleline WRM-related spending totals over £85m, not including contributions to the
Cooperation in International Waters in Africa (CIWA) initiative and Nile Basin Initiative. The
initiation of the International Climate Fund (ICF), the primary channel for UK climate change
finance intended for climate change adaptation (50%), low-carbon growth and tackling
deforestation in developing countries, has opened up a further important funding stream for water
resources management and water security interventions. The ICF Implementation Plan 2011/12-
2014/15 Technical Paper identifies water resources management as one of seven priority sectors
(DFID n.d.), and notes that:

The evidence base to inform investment decisions is of variable quality, and the results chains
to demonstrate impact and value for money are still limited. Building a more robust evidence
base will be a priority for ICF spend during the Spending Review period. (DFID n.d.: 4)

A number of illustrative indicators for measuring impact and results are presented in the plan, but
none of these relate specifically to water security outcomes. Like other donors, DFID have few
tools at their disposal to weigh up the costs and benefits of different programmes, and thence to
guide spending and assess the effectiveness of that spend.

Any assessment of metrics for water security should therefore also be acutely aware of country-
level needs, as well as restrictions, in terms of monitoring and data. How water security is defined
and measured at community level is also an emerging concern (WaterAid, 2012), with strong links
to national and global water security agendas, though detailed review at this level is beyond the
scope of this paper.
8

3 Scarcity, risk and security: competing framings

Before turning to a detailed discussion of metrics it is necessary to set out what we mean by water
security. As stated, with a number of definitions for water security already proposed, and others
forthcoming, another attempt to definitively encapsulate the concept in a few lines is probably not
required. But it will be impossible to discuss metrics without an idea of its normative meaning, or
the substantive aspects which are of concern in terms of designing a metrics framework. Hence,
this section briefly reviews how water security, and the related concepts of water scarcity and risk,
are being framed, before identifying 5 key aspects or themes which could be understood to fall
under an expansive, inclusive framing of water security. In so doing, a brief ‘working definition’ is
offered, for the purposes of this paper, to provide a reference point in the subsequent, detailed
discussion of appropriate metrics (Section 4).

3.1 Scarcity
Across a number of resources, including potentially renewable stocks such as freshwater, limited
absolute availability is being set against continuing population growth, changing patterns of
consumption and changes in global environmental systems, notably the climate (ODI, DIE and
ECDPM 2012). This arithmetic of scarcity - the difference between growing demand and finite (and
spatially and temporally shifting) supply - is at first sight compelling, and has captured the attention
of the mainstream media, governments and corporations. Certain quantifications of limits or the
supply-demand gap have been particularly influential (Rockström et al. 2009; Dobbs et al. 2011).
These studies have had the positive effect of galvanising attention around significant global
challenges. On the other hand, they tend to privilege a conceptualisation of scarcity as natural fact,
rather than as a construct of political, social and economic inequities. But despite resurgent interest
in resource scarcity being driven, in many cases, by perceived physical limits, in the case of water
divergent views have circulated for some time.

The 2006 Human Development Report (HDR) concluded that ‘The scarcity at the heart of the
global water crisis is rooted in power, poverty and inequality, not in physical availability’ (UNDP
2006: v). Another forceful argument cautions against privileging scarcity - construed at national,
regional or global scale and divorced from relational concepts such as need, want and access -
over and above scarcities – the multiplicity of realities experienced by local people in local contexts
(Mehta 2011) of which physical availability is only one component. A further key intervention in this
spirit was made by the International Water Management Institute (IWMI), which characterised the
essential distinction as being between areas of the world facing physical water scarcity and those
facing what it called ‘economic water scarcity’. In this case economic water scarcity is defined as
affecting those areas with abundant physical resources relative to current use (25% of available
water from rivers withdrawn) but where malnutrition exists, while physical scarcity affects those
areas where more than 75% of river flows are withdrawn, accounting for return flows. Despite the
simplifications – for example in taking malnutrition as a proxy for insufficient development of supply
relative to need, and the focus on surface water only – IWMI’s assessment and the resulting map
(Figure 1) have done much to underscore that water scarcity needs to be considered in more
nuanced terms than physical availability alone.
9

Figure 1: Areas of physical and economic water scarcity

Source: Molden (2007)

The core finding from IWMI’s research has been confirmed by the Challenge Programme on Water
and Food’s analysis of ten major river basins including the Limpopo, Niger, Nile, Volta, Ganges,
Mekong and Yellow rivers, which are home to half the world’s poorest people. The analysis
highlights that inefficient and inequitable use of water (including rainfed agriculture) is a more
widespread problem than physical scarcity (CGIAR 2011) and that the nature of the water scarcity
challenge tends to vary according to the nature of countries’ economies, from agricultural, to
transitional, to industrial. Other forms of water scarcity have been proposed, for example by Molle
and Mollinga (2003), including managerial, institutional and political water scarcity. Therefore the
term ‘economic water scarcity’ should not be understood as a reduction of all water management
failures to failures of the market. Instead, these other socially generated forms of scarcity should
be understood to be implicit.

Framing the scarcity problematic in this way implies that a range of responses are appropriate,
including improved management systems and helping the poorest to access water in the first
place. It would be a simplification to say that all of the recent interest in the physical dimension of
scarcity ignores other forms, or excludes a broad range of solutions, in favour of unilateral action
with a focus on developing or securing supplies. However, it is worth considering how far
embedded political and economic norms underpin each analysis of the scarcity problem, and
responses to it. For example, the management consulting firm McKinsey and Company advocates
increased resource-efficiency, spurred by market signals and enlightened, proactive public policy
(Dobbs et al. 2011), while the Centre for a New American Security advocates better integration of
natural resource issues into the politico-military space (Parthemore and Rogers 2010).
At the same time an emphasis on physical scarcity is not confined to the corporate or national
security spheres. The UN’s World Water Development Report 2012 argues that ‘The world is
transitioning to a new era where finite water constraints are starting to limit future economic growth
and development. It is becoming clear that even renewable water resources cannot supply enough
water if not managed carefully’ (WWAP 2012: 124). Although this does not preclude the continued
threat of other forms of scarcity, it may yet mark a departure in UN thinking (insofar as this is ever
homogenous) from the position adopted in the 2006 HDR.

One final caveat should be added around the way that the physical resource is often
conceptualised and, in particular, the frequently implicit assumption that the key flows in time and
space (which determine what is available, when and where) are between surface water and the
atmosphere. But groundwater and soil moisture are also critical components of the cycle and
respond at different rates, both to physical water phenomena on the surface and atmosphere, and
to human interventions. Table 1 highlights some of the key differences between groundwater and
10

surface water systems. Of critical importance is groundwater storage, which may be many times
greater than annual renewable freshwater resources, and which provides a vital buffer against
rainfall variability (Box 6 and Figure 2).

Table 1: Comparison of groundwater and surface water resources

Key features and Groundwater resources and Surface water resources and
characteristics aquifers reservoirs
Storage volumes Comparatively large Small to moderate

Resource area Widespread Restricted to water bodies

Flow velocities Low Moderate to high

Residence times Often decades/centuries Generally weeks/months

Evaporation losses Low and localised High for reservoirs

Resource evaluation High cost, significant uncertainty Lower cost, less uncertainty

Resource monitoring, data Very limited, especially rural More comprehensive

availability

Public perception, awareness Very limited More visible

Development cost, risk Often modest Frequently high

Style of development Mixed public and private Largely public

Source: adapted from Tuinhof et al. (2002)

Box 6: Groundwater storage: a missing piece in water security assessments?

Groundwater accounts for roughly one-third of the world’s total freshwater, and the vast majority
(96%) of all freshwater not bound up in ice (Shiklomanov 1998). It also plays a fundamental role in
supporting human and environmental systems: groundwater abstraction probably accounts for
about one-quarter of total water withdrawals; between 1.5 and 2.8 billion people (nearly half the
world’s population) rely on groundwater as their primary source of domestic supply; and large-
scale groundwater use has brought massive benefits to millions of impoverished farmers,
particularly in South Asia and the North China Plain (Morris et al. 2003; Shah 2007; Giordano
2009).

In view of its significance, we would expect to see groundwater figure prominently in assessments
of global, regional and national water availability, and in the growing literature on water security.
Surprisingly, it does not. As Taylor (2009) and Gleeson et al. (2012) note, most assessments of
global resources have focussed on surface water only, or have failed to differentiate between the
fraction of freshwater that is well distributed as groundwater with long residence times (years to
decades, or longer), or that which is relatively ephemeral and concentrated in river channels.
Crucially, this means that while groundwater may implicitly be included in freshwater assessments
through its contribution to surface water baseflow (see e.g. Vörösmarty et al. 2010), the
significance of groundwater storage is overlooked. Yet as MacDonald et al. (2012) highlight in the
African continent, estimated groundwater storage represents a water resource of a different
magnitude to all other freshwater sources and many countries designated as ‘water scarce’ in
terms of annual flows have significant groundwater reserves (see Figure 2).
11

Why the consistent omission or under-emphasis? Giordano (2009) refers to hydro- schizophrenia:
the inappropriate differentiation of the natural connection between surface and groundwater, and
the creation of separate surface and groundwater governance, policy and bureaucracies. On the
supply side, this relates to the fact that groundwater is ‘out of sight and out of mind’; a hidden
resource whose location, quantity and function in natural and human systems is poorly understood.
On the demand side, the rapid acceleration in groundwater exploitation over the last five decades
or so has been described as a silent revolution, with massive increases in groundwater
withdrawals, particularly in south Asia, occurring largely outside the public realm, and self-financed
by millions of farmers (Shah et al. 2003; Shah 2007). In Sub-Saharan Africa (SSA), the imperative
to extend drinking water access to the 344 million without safe water will depend overwhelmingly
on groundwater (MacDonald et al. 2009, 2011; Calow et al. 2010). However, in most
circumstances data on groundwater availability are patchy at best, and data on use are even less
reliable (Giordano 2009).

The particular characteristics of groundwater, beyond its relative ubiquity and use, are also poorly
appreciated (Table 1). Of particular significance in a discussion of water security is groundwater
storage highlighted above - specifically the large storage volume per unit of inflow - as this makes
groundwater less sensitive to annual and inter-annual rainfall variation (and longer-term climate
change) than surface water. The buffering or stabilisation effect this confers is hugely valuable, for
example allowing groundwater sources to provide reliable dry season or drought supply for rural
communities in Africa, and supplementary or full irrigation for farmers in the Indo-Gangetic plain
and semi-arid northern China (Shah 2007; MacDonald et al. 2009; Calow et al. 2010). Moreover,
unlike most surface water, self-supplied groundwater can be delivered precisely when needed, with
the result that groundwater irrigation is generally more productive than its surface water equivalent
(Burke, Sauveplane and Moench 1999; Shah 2007). Storage potential varies significantly between
different hydrogeological environments, and the quality of groundwater can also vary. As Moench
(2000) notes, in some locations the Ganges basin contains over 20,000 feet of saturated sediment
(of variable quality) and while pumping may become uneconomic if water levels continue to fall, the
resource is not about to dry up. In contrast, the basement aquifers underlying much of SSA store
much less water. While they cannot support the kind of water-intensive Green Revolution
witnessed in south Asia, they can still provide reliable supplies for domestic and small-scale
productive uses (Calow and MacDonald 2009; MacDonald et al. 2012).

Whether withdrawals are sustainable depends, in physical terms, on the relationship between
abstraction and recharge (from rainfall) over a period of time. The terms sustainable yield, safe
yield, overdraft and over-exploitation are commonly used to describe this relationship, with any
decline in water levels (and/or quality) frequently labelled ‘unsustainable’. A key issue here is that
some very large aquifers (e.g. in North Africa – see Figure 2) do not receive any contemporary
recharge from rainfall, so any exploitation of non-renewable or ‘fossil’ groundwater is, by definition,
unsustainable in these terms. However, taking a broader line on sustainability, exploitation can be
justified where there are clear benefits in use and parallel investment in long-term substitutes -
what Foster et al. (2003) describe as ‘planned depletion’. From a water management perspective,
overdraft or over-exploitation of shallow renewable aquifers can also be justified in circumstances
where it makes storage available for wet season recharge, reducing flood risk and providing water
supply in the dry season, as it does in Bangladesh (Morris et al. 2003). Moreover, some authors
(e.g. COMMAN 2005; Moench 2007) argue that ‘over-exploitation’ of renewable groundwater can
also be justified on social transition grounds, for example where it allows farmers to re-invest in
less water-intensive and more sustainable livelihoods in the rural non-farm and urban economies.
This, in turn, may make pricing or regulatory control over large aquifers easier by reducing the
number of resource users and increasing the stake each remaining user has in positive resource
outcomes.
12

Figure 2: Groundwater storage and freshwater availability in Africa

British Geological Survey © NERC 2012.

Source: MacDonald et al. (2012). Note: areas of largest groundwater storage are in sedimentary basins – both renewable and non-
renewable. In North Africa, for example, water is stored in extensive ‘fossil’ aquifers (e.g. the Nubian sandstone aquifer beneath Chad
and Egypt, with roughly 150,000 km3 of reserves) that receive no contemporary recharge but offer significant development potential.
Aquifers with least storage occur in thin basement rocks across much of SSA, but even these store water from several decades and can
support domestic use and minor irrigation. The graph on the right compares groundwater storage with annual renewable freshwater
availability (FAO data).

3.2 Risk
Like water security, the concept of water risk is more expansive than even a multivalent
interpretation of water scarcity, in that it extends to cover challenges of over-abundance as well as
insufficiency. Another key feature of risk, as a general concept, is that it can help us think
systematically about uncertainty. ‘Risk analysis encourages us to think about a whole range of
possible future conditions, from the everyday to the extremely unlikely. That’s an important feature
in aquatic systems, which are inherently variable.’ (Hall 2012)

The concept of water risk has been used extensively in initiatives coming from or intended for the
private sector. For example, the latest ‘Water Stewardship’ report from Coca-Cola refers frequently
to water risk and does not use the term water security at all (The Coca-Cola Company 2012). A
wide range of water tools developed for strategic and operational managers within corporations, as
well as external investors, refer to risk prominently in their titles or straplines (WRI 2012; Batton et
al. 2011; WWF and DEG 2011; GEMI n.d.a, n.d.b, n.d.c; WBCSD 2011). This proliferation of water
risk assessment and management tools appears to respond to demand – business representatives
positioned ‘water supply crises’ second in terms of impact and fourth in terms of likelihood in the
World Economic Forum’s 2012 survey of 50 global risks (WEF 2012).

In responding to water risk, private sector actors can seek to address their internal operations, for
example by reducing water use in industrial processes. This is often harder than it first appears, for
example the hidden disincentive of needing to guarantee a minimum flow for effective functioning
of most wastewater treatment technologies. A second area for engagement is in the supply chain,
13

where large corporations may have considerable contractual leverage. In a third emerging but
important development, corporations appear increasingly keen to mitigate ‘external’ risks arising
from the wider environment, hence interest on the part of some corporations in convening
stakeholders and engaging in broader water resources management - a role traditionally reserved
for public agencies (Newborne 2012; 2030 Water Resources Group 2012).

Figure 3: Fifteen-model mean changes in (a) precipitation (%), (b) soil moisture content (%), (c)
runoff (%), and (d) evaporation (%) for the last decade of the 21st century, relative to the last
decade of the 20th century

Source: Bates et al. (2008)

The World Water Development Report (WWDR) 2012 Managing Water under Uncertainty and Risk
(WWAP 2012) offers a number of important insights as to how risk can be usefully employed to
identify, assess and respond to water resource management challenges. The report distinguishes
several sources of uncertainty in relation to water systems and their management, including
inadequate or unreliable data, and disagreement or ignorance about natural, physical and human
processes which underpin hydrological cycles and our relationship to them. Climate change
increases uncertainty, as illustrated by the Intergovernmental Panel on Climate Change (IPCC)
maps of expected future change in, respectively, precipitation, soil moisture, runoff and
evaporation. In Figure 3, it is only in the stippled areas that more than 80% of models agree even
as to the overall direction of future change. A significant improvement is required in modelling of
precipitation as well as better integration of climatic and hydrological models. Recently the EU
funded Water and Global Change (WATCH) project has sought to develop a multi-model approach
to assess impacts of climate change on the water cycle, bringing together the hydrological, water
resource and climate research communities (Harding and Warnaars 2011). However, the authors
of the WWDR 2012 also point out that even if we can improve our understanding of these
processes and the quality of data that describe them, perception of risk – moderated by a number
of factors, including likelihood of harm, magnitude of harmful effects, ability to moderate those
effects and, critically, trust in the source of information – will determine how far individuals and
society as a whole are willing to respond.

Together, these features of water risk test the capacity of all decision makers, in both public and
private sectors, to respond effectively. Uncertainty and risk have long presented a challenge to
14

water managers, but the fundamental paradigm has been to calculate future variability on the basis
of statistical analysis of historic data. Climate change and other complex manifestations of
society’s interaction with physical and biological processes have led to a fundamental re-evaluation
of this paradigm, with recognition of the non-stationarity of hydrological systems (Milly et al. 2008)
and the fact that the past may not alert us to emergent future change, particularly if there is a risk
of trespassing tipping points.

In response to the challenging new water risk paradigm, the WWDR 2012 recommends a number
of strategies. One option is to plan in an adaptive manner – avoiding commitment to infrastructure
or decision pathways that may be irrevocable. The aim here is to enhance resilience, ‘the ability to
adapt to changes and recover from disturbances, while providing options for future development’
(WWAP 2012: 240). An alternative where adaptation is difficult, for example with major capital
infrastructure on the scale of reservoirs and flood control structures, is to aim for robustness, or
‘how well a system performs over a range of possible input scenarios pertaining to what is
uncertain’ (Ibid.). This means taking account of an expanded range of possible scenarios, beyond
what the historical data suggests. Options to operationalise these approaches include scenario
development (for example ‘back-casting’ to better account for radical changes than conventional,
incremental approaches to envisioning the future) and the increased use of the natural adaptive
capacity of ecosystems (for example wetlands) to buffer against change.

3.3 Security
With water’s significant social and cultural importance, its intersection with the already loaded term
‘security’ results in some alarmist responses. For example, the World Economic Forum describes
water security as an emerging ‘headline geopolitical issue’ that may ‘tear into various parts of the
global economic system’ (WEF 2011). Like water risk, the concept of water security implies
mitigating the effects of overabundance as well as scarcity. Indeed, one widely quoted definition of
water security embraces the concept of water risk as one side of the coin – the other being
availability:

The availability of an acceptable quantity and quality of water for health, livelihoods,
ecosystems and production, coupled with an acceptable level of water-related risks to people,
environments and economies.
Grey and Sadoff (2007: 547f)

Compressing this definition further, Professor David Grey has referred to water security as
‘tolerable water-related risk to society’ (Hope 2012). But while this definition has the advantage of
brevity, it is ambiguous – one must still ask what risks matter, and to whom. ‘Society’ means
different things to different people, and may leave room for the privileging of some interests over
others.

While the concept of water security is not new, the term appears to have gained greater profile
recently, judging from a range of reports and conferences that have considered water security in
isolation or in relation to the security of other resources, notably energy and food/land (WEF 2011;
NIC 2012; Martin-Nagle et al. 2012; Oxford University Water Security Network 2012).

Compared to water scarcity, there has been more limited problematisation of water security
discourses, notwithstanding some important interventions (Tarlock and Wouters 2009; Wouters
2010; Cook and Bakker 2012) and debates about the significance of the term have had less time to
evolve and polarise. The definition of water security quoted above privileges availability of the
resource. To some degree this underplays issues of access and allocation and aligns more with
the concept of physical water scarcity than with other manifestations. A definition giving greater
emphasis to these issues was in fact offered in the Ministerial Declaration of the Second World
Water Forum in the Hague in 2000, whereby providing ‘water security in the 21st century means’:
15

Ensuring that freshwater, coastal and related ecosystems are protected and improved; that
sustainable development and political stability are promoted, that every person has access to
enough safe water at an affordable cost to lead a healthy and productive life and that the
vulnerable are protected from the risks of water-related hazards.
Ministerial Declaration of the Second World Water Forum, The Hague.

But while the discourses on water security have not been extensively interrogated in their own
right, there has been considerable thinking around longer-established security concepts with which
water security is inevitably associated – notably national and human security.

In relation to national security, concepts of climate security and resource security in general have
featured more prominently in foreign policy and defence communities’ portfolios than water security
per se (DCDC MoD 2010; CNA Corporation 2007; IISS 2011). However, there has been
longstanding consideration of water’s potential role in conflict, often with reference to water scarcity
(CoFR, US Senate, 2011). Clear examples of international conflicts with water as a central causal
factor, or as a weapon of war, are in fact rare (Yoffe, Wolf and Giordano 2001: 64). Nonetheless,
there are well-documented instances of water playing a part in more localised unrest, terrorism and
political oppression into recent history (Pacific Institute 2011), and commentators reflect that this is
likely to be an ongoing and intensifying phenomenon (IISS 2011). The World Bank has attempted
to identify potential water conflict hotspots based on physical risk and ability, at least on paper, to
manage that risk, matching projected change in hydrological variability against the presence of
relevant institutions, notably treaties and river basin organisations, for different transboundary river
basins (de Stefano et al. 2010).

In any case, water security may increasingly be referred to in articulating water’s role in national
and international peace and stability due to water’s strategic significance as both a ‘fugitive
resource’ that often traverses borders (UNDP 2006: vi) and, in its ‘embedded’ or ‘virtual’ form, a
globally traded commodity (see Section 4.3). A report on Global Water Security was recently
commissioned by US Secretary of State Hillary Clinton from the US National Intelligence Council
(NIC). The report considers the implications of water (in)security, understood in terms of national
security, for US interests – identifying not only threats but also opportunities, for example in relation
to the US’s status as a major global food exporter (NIC 2012), picking up on the emerging theme of
the interconnections, or ‘nexus’, between water and food security (as well as, elsewhere, energy
security). The report concludes that within 10 years water insecurity could be a contributing factor
to state failure, and increasingly feature as a mechanism for contestation and leverage between
states. Beyond 10 years, the report has high confidence that water is more likely to be used as a
weapon by states or terrorists.

Observing governmental concern with the global water crisis, and a narrow interpretation of water
security aligned closely with national security, some commentators express unease that
responses are more likely to be unilateral and backstopped by the threat of force, rather than
multilateral and based on cooperative legal forms and management regimes (Tarlock and Wouters
2009). But it can equally be noted that in announcing the NIC commission in 2011, Clinton chose
to counterbalance ‘the potential for unrest, conflicts, and instability’ with the rejoinder that ‘the
water crisis can bring people together… on water issues, cooperation, not conflict, is and can be
the rule’ (US Department of State 2011). The report itself includes a further headline conclusion
that improved water management (including pricing, allocations and virtual water trade) and
investments ‘afford the best solutions for water problems’ (NIC 2012: 6). This goes some way to
temper concerns that the defence and foreign policy communities will necessarily co-opt the
concept of water security in support of unilateral military responses.

Beyond transboundary water resources, the national security implications of water extend also to
how a country manages its own internal water resources for its economic development and
stability. The economic significance of water is clear, in spite of the fact that the resource itself is
often under-priced or not priced at all. One influential paper suggests that economic growth is
much more closely correlated to an even temporal and spatial distribution of water (i.e. low rainfall
16

variability) than it is to high physical availability overall, and that many agricultural low-income
countries are particularly vulnerable to intra-annual variability (Brown and Lall 2006). However,
this may underplay the importance of groundwater storage and its potential to provide a buffer
against shorter-term variability, especially as groundwater replenishment is unlikely to correlate
directly with precipitation (Box 6). Growth is unlikely to be so sensitive to fluctuating rainfall in
groundwater-dependent economies, for example in parts of South Asia and the North China Plain
where agricultural yields have increased largely on the back of small-scale, farmer-financed
irrigation from boreholes (Shah 2007). A further argument around the issue of hydrological
variability proposes that vulnerability is exacerbated where countries lack a minimum platform of
hydraulic infrastructure (for conveyance and storage), leaving them ‘hostage to hydrology’ (Grey
and Sadoff 2007) – a predicament that applies to some extent in the case of groundwater also, in
terms of infrastructure to access, if not store, the resource.

There are clearly both politico-military and economic imperatives at the intersection of water
security and national security, which may yet influence future paradigms for WRM. The current
dominant paradigm IWRM has been extensively promoted (if not applied) at national level, but as a
recent report highlights, more work needs to be done to make its goals relevant in a transboundary
context where national security discourses tend to play out, and where more heterogeneous legal
and institutional regimes and greater disparities of power and interest may be at play (GWP and
INBO 2012).

Human security is the second of the existing major security concepts which is likely to influence
interpretations of water security. Since its origins, human security has been conceptually opposed
to a narrow, conventional interpretation of national security. The 1994 HDR, which brought the
term to popular attention, recognised that the scale and nature of many threats to peace and
sustainable development cannot be tackled solely through a territorial paradigm of the nation-state
backed by force of arms (UNDP 1994). The 1994 HDR represents a landmark in a narrative which
continues through to the water-focussed 2006 HDR and beyond, whereby security is conceived in
multidimensional terms, rooted in individual rights and cognisant of insidious disparities in power
and resources between individuals and groups. Water security is not mentioned by the 1994 HDR
as a category in its own right, but aspects are subsumed within the categories of health and
environmental security. Furthermore, its conception of food security requires that ‘all people at all
times have both physical and economic access to basic food’ (Ibid.: 27) foreshadowing some of the
thinking which has developed on different forms of water scarcity (and by extension, water
security).

Because it places the emphasis on individuals, the concept of human security aligns most naturally
with human-centred interpretations of the water crisis, and principle among these is the concept of
‘water poverty’. In turn, water poverty tends to be a concept most often deployed in relation to
drinking water hygiene and sanitation (WASH). For example, WASH is the focus of the
international End Water Poverty coalition of 180 civil society organisations and networks. Water
poverty also chimes naturally with the MDGs – not only with Target 7c on drinking water and
sanitation, but also with Goal 1 ‘Eradicate extreme poverty and hunger’. The association of water
security with national security agendas may make water poverty a more palatable option for rights-
oriented organisations, though UNICEF refers to ‘household water security’ as a synonym for water
supply, specifically (UNICEF 2010).
A WASH-focussed interpretation of water poverty or indeed of ‘household’ or ‘human’ water
security directs attention to some of the most pressing water challenges. Despite the achievement
of the MDG target for water at a global level in 2012, huge geographical and social disparities
remain, especially when the many non-functional water points are discounted – as many as 40% of
the total in rural Liberia, for example (Hirn 2011). But an exclusive emphasis on WASH may risk
overlooking the other ways in which water interlinks with people’s livelihoods. In fact, earlier
definitions of water poverty do not necessarily restrict themselves to WASH. Black and Hall’s
categorisation of the water poor (Box 7) puts the headline emphasis on broader relations between
17

water and the ‘livelihood base’ (Black and Hall 2004: 24), including water for cultivation – though
water for other productive purposes, such as small scale manufacturing and industry, is not
mentioned directly. Another important feature of water poverty as articulated by Black and Hall is
its clear emphasis on gender dimensions. The greater impacts of inadequate water on women has
been well-articulated around WASH, for example in terms of caring for children and others
affected by water-related diseases, and the time costs and personal security risks involved in
collecting water (UNICEF 2003). But gender dimensions of wider water use and management
should also be considered, for example the tendency for water access for irrigation and livestock to
be dependent on land rights – which are limited for many rural poor people, but especially women
(IFAD 2007). While Black and Hall propose a number of quantitative and qualitative thresholds in
their definition, it should be noted that applying these collectively would result in double counting in
many contexts. The thresholds would therefore need some attention if they were to be applied as
part of a water-related target.

Box 7: Water poverty as human insecurity

Black and Hall define the water poor as:
• those whose livelihood base is persistently threatened by severe drought or flood
• those whose livelihood depends on cultivation of food and natural products, and whose
water source is not dependable
• those whose livelihood base is subject to erosion, degradation, or confiscation (e.g. for
construction of major infrastructure) without due compensation
• those living far (e.g. >1 km) from a year-round supply of safe drinking water
• those obliged to spend a high (e.g. >5%) percentage of household income on water;
slum dwellers obliged to pay for water at well above market rates
• those whose water supply is contaminated bacteriologically or chemically, and who
cannot afford to use, or have no access to, an alternative source
• women and girls who spend hours a day collecting water, and whose security,
education, productivity, and nutritional status is thereby put at risk
• those living in areas with high levels of water-associated disease (bilharzia, malaria,
trachoma, cholera, typhoid, etc.) without any means of protection.
Source: Black and Hall (2004: 11)

But even if availability of, and access to, water for various needs and productive purposes is
included in a human-oriented framing of water security there is a risk of overlooking non-human
users. This returns us to the debates about integrating environmental and developmental agendas
discussed in Section 2. The original framing of human security in the 1994 HDR acknowledged
‘environmental security’ as one of seven key categories, with threats to water highlighted as ‘one of
the greatest environmental threats’ in developing countries. However, partly due to the subsequent
alignment of the human security agenda with the poverty and social development focused MDGs,
environmental issues have tended to be somewhat squeezed out. There is a natural concern that
meeting ecosystem needs will inevitably necessitate trade-offs with human needs. The concept of
ecosystem services as fundamental building blocks for human development has helped to reframe
apparent tensions between human and environmental water security, as necessary synergies
(Carpenter 2005; UNEP 2009).

On the other hand, others refer to water security in a way which treats the resource base as an
intrinsic good, separately from any status as an instrumental good which provides a variety of
services to humans. For the Global Water Partnership (GWP), ‘the essence of water security is
that concern for the resource base itself is coupled with concern that services which exploit the
resource base for human survival and well-being, as well as for agriculture and other economic
enterprise, should be developed and managed in an equitable, efficient, and integrated manner’
(GWP 2009: 1). Elsewhere, GWP refer to a water secure world as integrating ‘a concern for the
intrinsic value of water with a concern for its use for human survival and well-being’ (GWP 2012).
18

3.4 An expansive and inclusive framing of water security

Taking account of the above discussion, key aspects or themes which could be included under an
expansive and inclusive characterisation of are identified in Box 8. Water security differs from both
water scarcity and water risk in that it is inherently a positive ideal for water resources
management whereas scarcity and risk, as things to be averted or avoided, need to be reframed in
positive terms. Any attempt to characterise water security in a few words could therefore be read
as an attempt to define a goal, in the manner of the MDGs or SDGs. But there are already a
number of definitions and proposing another in an already crowded space is unlikely to add much.
Rather, the purpose of identifying five key themes for water security is to frame subsequent
discussion of metrics. Understanding of metrics is, in turn, critical for discussion of goals and
targets, in order to provide a technical basis to inform discussions at the normative or political level
(as these proceed through multilateral fora such as the UN General Assembly Working Groups on
the SDGs – see Annex 1). Integrating the technical and political will need to done on an iterative
basis, to retain a measure of ambition while retaining a goal, target and indicator set for which
sufficient data, of sufficient quality, can be obtained.
This said, a working definition or framing of water security is needed for the purposes of this paper,
to provide a reference point and encapsulate the five themes. This can be proposed in two parts –
one relating to the physical dimensions of the resource, the other pertaining to issues of economic,
social and political capacity, drawing on IWMI’s distinction between physical and economic water
scarcity, and adapting existing definitions of water security including Grey and Sadoff (2007) and
ODI, DIE and ECDPM (2012). Thus for the purposes of this paper:

Water security means having sufficient water, in quantity and quality, for the needs of humans
(health, livelihoods and productive economic activities) and ecosystems, matched by the
capacity to access and use it, resolve trade-offs, and manage water-related risks, including
flood, drought and pollution.
19

Box 8: Key themes for a water security metrics framework

From analysis of the concepts of water scarcity, water risk and water security it is evident that an
expansive and inclusive framing of water security should:

• Look beyond immediate physical availability. While physical availability is likely to

assume increasing localised importance under demographic, socio-economic and
environmental drivers, there is a risk of oversimplification. First, conventional measures
of physical availability have a number of problems, including limited attention to
groundwater and soil moisture, which can be key parts of the resource base but which
often react to climate, hydrology and human intervention in complex ways. Second,
such measures overlook the economic, institutional, managerial and political inequities
which mediate access to the physical resource. Such inequities can also give rise to
water insecurity by reducing the ability to manage water-related risk (see below). In this
regard the idea of capacity to access the resource, or manage risks associated with it,
may be useful.
• Address variability and risk. Being ‘water secure’ intrinsically implies a state of being
that will continue in perpetuity. But water resources, and the ways they are used by
society (e.g. for agriculture) are subject to wide temporal and spatial variations. The
implications of this variability and uncertainty therefore need to be articulated explicitly.
Water-related risks, both from hydrological hazards (associated with insufficiency and
overabundance of water) and anthropic hazards (e.g. pollution or over-exploitation)
should also be addressed for water security – this implies reducing exposure and
vulnerability, the two factors which determine our ability to manage or adapt to hazards
(IPCC 2012).
• Have a human focus. An inclusive framing of water security should, at root, be human-
focused, with an emphasis on individual livelihoods, especially for the poorest and most
vulnerable, including women. Framings that are concerned only with higher levels (for
example the nation-state) may make it easier to privilege some peoples’ water security
over others’, both within and outside a country’s borders. Paramount are the
fundamental needs associated with water, sanitation and hygiene, which in turn
underpin health and opportunity, though water security also implies access for
productive purposes (in terms of individual livelihoods and broader economic activities).
• Acknowledge environmental needs. Though water security is a human-centred
concept, it must be framed in such a way that the dependence of human society on the
wider natural environment, through ecosystem services, is acknowledged – even if this
implies viewing environmental needs as instrumental to human needs.
• Manage competition and conflict. There are justified concerns that ‘securitisation’ of
water will privilege competitive management strategies over cooperative ones, or even
the use of force rather than negotiation. However, it is important to acknowledge that,
given the ultimately finite nature of the resource and the many inequities in terms of
access, use, and protection from harmful effects, trade-offs are likely to arise and
tensions between competing needs will need to be managed. An expansive and
inclusive framing of water security must therefore also emphasise the ability to manage
such tensions at different levels through appropriate governance, negotiation and the
rule of law, rather than force (Wouters 2010).
20

4 Metrics and meaning: key considerations for

measuring water security
The following subsections take the requisite ‘themes’ of water security identified in Box 8 in turn,
analysing existing datasets and indicators which touch on each. This provides an overview of the
water metrics landscape, while critically interrogating current approaches to measuring different
aspects.

As intimated in the following subsections, there are important pragmatic considerations of data
availability and quality which will inform what is realistic and relevant to measure. There are also
important political considerations – reduction to numerical indices may be administratively and
politically useful, but also means political and administrative agendas invariably shape the
selection of data and use indicators as interpretative tools (Molle and Mollinga 2003; Kaczan and
Ward 2011). Moreover, pragmatic and political considerations are not mutually exclusive in that
data availability can be enhanced with greater political priority and resourcing, as Box 3 (above)
illustrates in an example from the WASH subsector. In the case of this paper, the identified
underlying themes of water security (Box 8) are an attempt to be explicit about what is being
promoted as important. While such issues are touched on in this section, they are considered in
greater detail in Section 5.

4.1 Beyond the physical resource: capacity to access, use

and manage
Water availability and our capacity to access, use and manage it are central to the conception of
water security, and have been a major focus of water resource metrics developed to date. Total
Actual Renewable Water Resources (TARWR) is a key measure of the physical availability of
water resources, estimated by a number of institutions, notably the FAO as a national-level long-
term average, from data collected over 15 to 25 years. There are significant challenges associated
with calculating and interpreting TARWR, as well as the many other indicators for which it is a
component, which are discussed extensively in this and subsequent sections. However, the cost
of obtaining data at greater spatial or temporal resolution has meant that the FAO estimates of
TARWR remain their ‘best estimate’ and are in widespread use.

TARWR is often compared to population statistics to give an idea of water competition (m3 per
person). On the basis of TARWR per person per year, Falkenmark et al. (1989) proposed
thresholds for water stress (countries with less than 1700 m3 /person/year), water scarcity (<1000
m3/p/y) and severe water scarcity (below 500 m3/p/y). These thresholds, known as the
Falkenmark Index, are widely accepted as crude but straightforward measures of physical
availability relative to human demand – as exemplified by their prominence in UN-Water’s new Key
Water Indicator Portal.

There are a number of concerns with the TARWR indicator and Falkenmark index, however. It
tends to obscure in-country variability, and estimates are derived from data spanning a 25-30 year
period which may not account for climatic or anthropic changes that could affect the water cycle.
While the per-capita measure of TARWR is revised to show trends over time, changes tend to
reflect population growth rather than changes in the underlying resource (the FAO updates the
water availability estimates only when new data becomes available from countries). Furthermore,
TARWR statistics are computed from a number of component indicators, including water flowing
into the country from outside, the subsurface component of which (transboundary groundwater
flows) is difficult to estimate. The FAO points out additional issues: dependence on government
estimates of differing accuracy and reliability; no distinction on water quality (particularly brackish
or saline water); crude adjustment for interaction of ground- and surface-water; and omission of
‘green water’ stored as soil moisture and critical for rainfed agriculture (prevalent in Sub-Saharan
Africa), as well as non-conventional sources such as desalinated water and water re-used through
21

drainage of agricultural water to rivers and seepage to groundwater (Margat, Frenken and Faures
2005). Box 9 explains how TARWR is calculated, and highlights recent efforts to address some of
the above issues with an enhanced methodology for its estimation.

Box 9: Enhancing estimates of TARWR: Pilot Study on Indicators (PSI)

‘The World Water Assessment Programme (WWAP) pilot initiative is a collaboration with the City
University of New York (CUNY) and the Global Water System Project (GWSP) to produce a
dynamic estimate of the basic data item, TARWR. TARWR is the fundamental measure of water
resource availability (in a country, river basin or region) and is used in many indicators. It is defined
as the maximum theoretical amount of water actually available for the country (or other unit),
calculated from:

• Sources of water within a country itself

• Water flowing into a country
• Water flowing out of a country (treaty commitments)

Availability, defined as the surface and groundwater resource volume renewed each year in each
country, means the amount of water theoretically available for use on a sustainable basis. In more
specific terms, TARWR is the sum of:

• External water resources entering the country

• Surface water runoff (SWAR) volumes generated in the country
• Groundwater recharge (GAR) taking place in the country

Less:

• The volume in the country of the total resource effectively shared as it interacts and
flows through both the groundwater and surface water systems - not to subtract this
volume would result in its being counted twice (it is also referred to as ‘overlap’)
• The volume that flows to downstream countries based on formal or informal
agreements or treaties.

The WWAP Pilot Study on Indicators (PSI) is being undertaken at the CUNY by Charles
Vörösmarty in partnership with the Global Terrestrial Network for Hydrology (GTN-H) and
GEO/IGWCO (Water Community of Practice), with support from the US Army Corps of Engineers
(USACE). The group has developed an innovative methodology for estimating country-level
TARWR. This approach is based on (but not limited to) a combination of hydro-meteorological and
high-resolution (6 minute river network and ESRI country boundaries) surface elevation data,
which will allow the identification of TARWR trends (e.g. if certain countries are getting wetter or
dryer) and variability (e.g. variation of water supply from one year to the next).

This ‘dynamic TARWR’ is used to produce an alternative set of countries’ per capita water
availability. This data item will be further developed. Given its observational basis and dynamic
nature, it is hoped that it will eventually become the primary point of reference, as it enables
longer-term variations in water availability to be tracked over time. This will overcome some of the
current constraints imposed by the assumption of stationary hydrology, which is considered to be
inappropriate in the face of climate and related challenges.’
Source: verbatim from WWAP (2012: 171)
22

Perhaps most critically, the FAO point out that the theoretical measure of physical availability
inherent in the TARWR indicator does not correspond to what is actually exploitable. Their
proposed concept of exploitable water resources depends on technical and economic feasibility of
development and exploitation, as well as reservation of any amount for environmental needs –
calculation methods vary by country and, to date, FAO's database on water, AQUASTAT (FAO
2012a), only provides estimates for a limited number of countries, principally OECD members
(Margat, Frenken and Faures 2005). Although comprehensive indices of exploitable water
resources by country are not available, the concept adumbrates the wide range of issues relating
to capacity to access, use and manage water, which take us beyond the physical resource. Also
worth highlighting here is that any metric of exploitable resources should also account for water
held in storage, and changes in that storage over time. As Taylor (2009) argues, the continued
exclusion of water storage (groundwater, soil moisture, glaciers) from indicators and projections
‘…critically undermines their ability to adequately represent water scarcity and profoundly
constrains scientists’ understanding of the global water crisis’. For example, our understanding of
available water resources may in some cases need to be extended to include non-renewable
(‘fossil’) groundwater reserves which provide significant sources of water (and water security) in
some areas, and where there is an economic logic for their managed depletion assuming time and
economic growth will permit development of alternatives (see Box 6).

By setting per capita thresholds for water scarcity, the index proposed by Falkenmark et al.
introduces relative human needs to the otherwise exclusively hydrological TARWR indicator. But
water stress and scarcity thresholds based on TARWR per capita implicitly assume that all people
require the same quantity of theoretically available water, regardless of their geographical location
or socio-economic circumstance, and that the required volume can be accessed by those people,
either directly (in the case of domestic needs) or indirectly (in the case of the water required for the
production of goods and services).

Successive attempts have been made to improve on the Falkenmark index, particularly by
comparing availability to withdrawals on a country-by-country basis. A major UN assessment set
the threshold for water scarcity where withdrawals are greater than 20% of available water, and
severe water stress where withdrawals exceed 40% of availability (Raskin et al. 1997). Subsequent
efforts to improve on this include work by the University of New Hampshire Water Systems
Analysis Group, using geospatial data to map (surface)water supply and demand across grid cells
(6 minute grid cells for Africa; 30 minute grid cells globally). This allows a much more spatially
disaggregated comparison of supply (employing data on locally generated runoff as well as the
effects of rivers in transporting water from wet to dry areas) and demand (based on irrigated area,
population and domestic and industrial water demand). The approach also relies on data derived
globally from Earth System Science (GIS, modelling, weather prediction, remote sensing) rather
than hydrographic estimates obtained from government (Vörösmarty et al. 2005). A key finding of
the more spatially disaggregated approach is that the proportion of the population found to be living
in areas subject to water stress tends to increase at greater resolution (Vörösmarty et al. 2000; see
Figure 4), which is apparently a function of the fact that populations tend to be concentrated in
areas of lower water availability.

However, there is no final agreement on how to calculate a comparison of availability and

withdrawals. The UN-Water Task Force on Indicators, Monitoring and Reporting (TF-IMR)
proposed using what it called the MDG water indicator, calculated from national figures for TARWR
and withdrawals data derived from AQUASTAT, and setting the water scarcity threshold at 75% of
TARWR abstracted, rather than 40%, but accounting for return flows (UN-Water TF-IMR 2009).
The latest WWDR (2012) focuses on a Relative Water Stress Index derived from the University of
New Hampshire geospatial approach, which retains the 40% threshold for severe water stress
(UNESCO 2012a).
23

Figure 4: Water stress in Africa as percentage of the population computed with increasing
resolution

Source: WWF and DEG (2011)

While comparison of withdrawals with availability is useful for highlighting that scarcity is a function
of society’s interaction with the resource, rather than physical availability alone, it does not account
for water insecurity that arises from under-exploitation (from weak economic, technical or
management capacity, and an inability to convey and store water to address spatial and temporal
variability) rather than over-exploitation. The Democratic Republic of the Congo has physically
abundant water resources (over 24,000m3/person/year) that appear even more abundant when set
against very low withdrawals (0.05% of available water) and yet water insecurity, particularly in
terms of clean, safe water for human consumption, is a reality for many Congolese citizens, due to
inadequate capacity to abstract and utilize the resource. Seckler et al. (1998) attempted to
account for this by projecting demand (based on e.g. increased irrigation for food production and
increased population) and supply (infrastructure, irrigation efficiency) to 2025, on a country-by-
country basis. On this basis, to satisfy ‘reasonable future requirements’ by 2025 (1998: vi), DRC
would need to develop supply almost four-fold compared to 1990 levels.

Other attempts to account for water insecurity deriving from low capacity and the under-
development of water resources have involved introducing other parameters, for example the
Comprehensive Assessment’s use of data on malnutrition (see Section 3.1). Similarly, Ohlsson
(1998) used the composite Human Development Index (HDI) to weight the Falkenmark index. The
calculation was undertaken by simply dividing each country’s score on a ‘hydrological water
stress/scarcity index’ (essentially the inverse of available water per capita, so high scores indicate
greater physical stress/ scarcity) by the HDI score. On this measure, countries such as the UK and
Belgium are no longer classed as water stressed, while Niger, Afghanistan and Burkina Faso,
which at national level have sufficient water resources relative to population, are re-classified as
water stressed.

The Water Poverty Index (WPI) is another composite index designed to be applied at household
level. The WPI was calculated most comprehensively by Sullivan et al. (2003) incorporating
measures of water quantity, quality and variability, access, uses, capacity for water management
and ecosystem requirements. Critics argue that, for all their subtlety, such complex composite
indicators are generally ‘black boxes’ which prevent interrogation of underlying realities
(Rijsberman 2006). Vörösmarty et al. (2010) developed the geospatial approaches described
above to produce a human water security index, made up of weighted indices for a wide range of
‘drivers of stress’ including dam density, nitrogen loading and a ratio of population to (river)
discharge (a ‘localised’ version of TARWR per capita). They then adjusted the index to take
account of the potential for technology and investment to mitigate the risks: ‘areas with substantial
technology investments have effectively limited exposure to threat whereas regions with little or no
24

investment become the most vulnerable in a global context’ (Vörösmarty et al. 2010: 558). The
calculations showed that South East Asia and Sub-Saharan Africa face much higher threats to
human water security once the inadequacy of technology and investment is accounted for, while
countries in the Global North (notably the US, Western Europe and South-East Australia) are, for
the time being, mitigating the threat. But while the Vörösmarty et al. study assimilated a staggering
array of data, the application of their index as a regular monitoring tool to track the performance of
national efforts to improve water security may be limited, in the near term, by the complexity of
calculation and the limits to regularly updated data. Moreover, although the study marshals an
array of indicators for quantifying water security, it is completely silent on the contribution of
groundwater.

4.2 Variability and risk: spatial and temporal factors

Variability has always been a feature of hydrological systems. Temporal variability in water
resources is a concern in terms of both ‘ordinary’ oscillations and extremes. As mentioned, an
influential study suggests that a high degree of intra-annual rainfall variability is a more reliable
explicator of GDP growth in agricultural economies than average physical water availability (Brown
and Lall 2006), although other research has underscored the complexity of relations between
development trends and hydrological variability (Conway and Schipper 2011). The WWDR 2012
identifies the coefficient of variation for the climate moisture index (CMI) - a statistical measure of
variability in plant water demand to precipitation - as an appropriate indicator for areas with
variable climates that may be subject to periodic water stress and scarcity (UNESCO 2012b;
WWAP 2012). The global gridded data set (30 minute grid cell resolution) is compiled by the
Environmental Crossroads Initiative at University of New York, computed as a 40-year average
(1971-2000). The co-efficient of variation for CMI can therefore give a crude indication of the
overall degree of climatic variability although it should be noted, particularly for agriculture, that
how variability plays out is a location-specific function of hydro-meteorological factors, water
management responses (e.g. storage, groundwater pumping) and crop-water demand at different
moments of plant development (Woodhouse 2012). Meanwhile, in spatial terms water is distributed
unevenly but also moves with the hydrological cycle. Vörösmarty et al.’s (2000) calculation of water
stress at higher resolution on a geospatial grid, and mapping of flows along digitized river corridors,
constitute attempts to grapple with the challenge of spatial variability, albeit through a surface
water lens only.

Indicators like the coefficient of variation for CMI rely on past experience to give a guide to the
lower and upper bounds of uncertainty, particularly uncertainty associated with temporal variability.
However, the concern is that with climate change, past experience will no longer be a reliable
guide to those bounds of uncertainty (Section 3.2) and the likelihood and magnitude of extreme
events. The recent IPCC SREX report finds that ‘A changing climate leads to changes in the
frequency, intensity, spatial extent, duration and timing of extreme weather and climate events, and
can result in unprecedented extreme weather and climate events’ (IPCC 2012: 5). This will further
complicate the prediction and management of disaster risk presented by hydrological extremes –
principally drought and flood. Even without the additional uncertainty presented by climate change,
disaster risk is acknowledged as a complex concept that is difficult to measure, whereby the
physical hazard is moderated by exposure (essentially, being in the wrong place at the wrong time)
and vulnerability (capacity to anticipate, cope with, resist and recover from adverse events).

Among global indices of disaster risk, the Mortality Risk Index (MRI) developed for the 2009 Global
Assessment Report on Disaster Risk Reduction (GAR) and updated for the 2011 report, is notable.
This uses GIS data on hazards, vulnerability (computed from statistical analysis of historical
events) and modelled population exposure, for a number of different disaster types including
floods. The authors argue that the high spatial resolution (1x1km) and potential for comparing risks
indicators between countries and over time, makes the MRI particularly useful (Peduzzi et al.
2010). However, a major challenge arises when applying the MRI to drought-risk due to lack of
sufficient and suitable datasets, which results in African countries’ overall MRI being under-
25

estimated. Additional problems are the lack of clear correlation between population exposed to
drought and mortality, due in part to the tendency for impacts to be bound up with other problems
such as civil unrest and conflict, rather than purely meteorological conditions (UNISDR 2011a),
though drought risks can be computed on a more localized basis where data is available.

It also appears that the MRI, designed to reflect evolving risks in the short term, is not ideally suited
to taking account of long-term changes in hazards associated with climate change. This is
understandable given the difficulty of projecting climate change impacts, especially at localised
scales, due both to uncertainty around future emissions and the complexity of the climate system.
The development of scenarios, for example the IPCC scenarios, can be used to help structure
thinking around various plausible futures and reduce the number of variables considered in each.
This is one premise behind the Climate Vulnerability Index (CVI), a composite indicator to assess
local vulnerability to water related risk, drawing on ‘Global Impact Factors’ which include
geographical and topographical data (e.g. slopes, proximity to coast), as well as data for water
availability, the capacity of people and institutions, ecological impact, access (including property
rights) and the economic efficiency of water use (Sullivan and Huntingford 2009). The need for
stakeholder-determined weightings of each impact factor, detailed underlying data and (ideally)
locally relevant scenarios, mean it may be more relevant to application at local scale – though it
has so far been applied to 148 countries (UNESCO 2012c).

A simpler approach to measuring adaptive capacity to hydrological variability is proposed by the

TF-IMR, using water storage and conveyance capacity as a proxy, reflecting the idea that a
minimum platform of hydraulic infrastructure is required to avoid being held ‘hostage to hydrology’
(Grey and Sadoff 2007). This would primarily involve compiling data from the ICOLD database on
large dams (ICOLD 2012), supplemented with information on irrigated infrastructure from the FAO
(FAO 2011). To some degree data on irrigated extent would help capture those areas which utilise
groundwater and small dams to supply irrigation, although the proxy indicator, as proposed, would
still give a possibly undue emphasis to large dams. Furthermore, while the FAO has recently
included irrigated area in its annual land-use questionnaire sent to governments, data is not yet
available for many countries from the FAOSTAT database. The TF-IMR also proposes that
adaptive capacity as measured by storage should also take account of natural reservoirs and
aquifers, acknowledging in particular the role groundwater plays in buffering hydrological variability
across seasons and between years (MacDonald et al. 2009; Calow et al. 2010). Estimates of
groundwater resources at country level are provided by the International Groundwater Resources
Assessment Centre (IGRAC 2012) and AQUASTAT. However, data availability and reliability for
groundwater are in general limited (WWAP, 2012). Local and regional studies, such as those
supported by DFID in Africa (MacDonald et al. 2012: 5) and now South Asia, are contributing to
improved knowledge in this regard.

As noted (Section 3.2) water risk is an attractive concept to the private sector and a number of
tools have emerged to characterise risks which may damage reputation, impede license to
operate, or create threats to core operations by affecting the quantity, quality and/or reliability of
available supply (SABMiller, GTZ and WWF 2010; WRI n.d.; WWF and DEG 2011). The World
Business Council on Sustainable Development (WBCSD) Global Water Tool, and the equivalent
Local Water Tool developed in partnership with the Global Environmental Management Initiative,
allow strategic and operational managers to identify water risk at the level of the entire company
and specific operational facilities, respectively. Such tools nonetheless tend to provide a
customised interface for existing, publicly available datasets at watershed and country level. An
exception is the WRI Aqueduct tool, which draws on previously proprietary data on water risk
compiled by Coca-Cola to generate maps of global water stress, including projection of water
stress over timeframes of more than 80 years using IPCC scenarios to 2095 (Jenkinson, 2011).
Given private companies’ interest in water risk and the commensurate resources large companies
can bring to gathering and analysing data, initiatives to make proprietary data publicly available are
welcome (though the Coca-Cola data in its raw form does not appear to be fully open access).
26

Neither the principle private sector-oriented tools in the public domain, nor the indices described
above in relation to climate change, take a formal probabilistic approach to risk assessment,
especially for medium to long-term risks. Probabilistic risk assessment (PRA) approaches are
widely used by industries dealing with engineered technologies (such as nuclear power plants)
where the causal chains for potential risk events are arguably clearer (USNRC 2007). PRA
calculates risk as a factor of the magnitude of an event and its likelihood, with both expressed in
quantitative terms (e.g. 1 in 100 year probability of a water-related disaster incurring costs to the
business of US$1m). The MRI is one of several water-related indicators with a probabilistic
component, in that it quantifies hazards according to their probable frequency, as well as
magnitude and spatial extent (Peduzzi et al. 2010). However, it does not seek to project how this
frequency may vary long-term with climate change.

But even in its application to engineered technologies, PRA has been criticised for failing to take
account of features of complex systems such as nonlinearity, feedback and emergence, as well as
for the simple psychological difficulty we face in conceiving of as-yet unprecedented catastrophic
events (Ramana 2011). Climatic and hydrological systems, which comprise natural and anthropic
interactions, are still more complex. The difficulty of attributing probabilities to long-range climate
forecasts is a familiar challenge (Morello and Climatewire 2010). Nonetheless, attempts are being
made (Michel-Kerjan et al. 2012), with the insurance industry prominent in efforts to develop
probabilistic modelling of climate change impacts, including those relating to water. Synthetic
stochastic catalogues of extreme events have been developed to underpin catastrophe models,
and are widely used by the insurance industry. These overcome limited historical experience to
some degree but still require sound input data, which is generally more limited in developing
countries (Ranger, Muir-Wood and Priya 2009).

4.3 Human focus: water for health and livelihoods

Although there is still much progress to be made in relation to the MDG targets for water supply
and sanitation, these targets and the indicators against which they are tracked have drawn
international attention to the importance of water for basic human requirements, which in turn
underpins health, productivity and opportunities including education. The service coverage
indicators, compiled by the JMP, are based on household survey data (primarily census,
Demographic and Health Surveys and Multi-Indicator Cluster Surveys). Compared to many water
resource indicators, data availability is comparatively good, derived from broadly standardised
survey questions and updated with reasonable regularity even in many low-income countries. The
targets’ (relative) simplicity has also contributed to their success. However, as the debate around
what should succeed the water supply and sanitation MDGs makes clear, there are still many
concerns around data quality, and the JMP’s definition (and measurement) of ‘sustainable access
to safe drinking-water and basic sanitation’, which tends to overlook problems with service quality
and functionality and, in the case of sanitation, the environmental impacts of inadequate faecal
sludge and wastewater management (WHO and UNICEF 2011).

Arguably due to having had an agreed target and monitoring framework for several years around
which to focus debate, the discourse on WASH monitoring is more advanced than that for broader
water resource management or water security. The work being undertaken by the JMP and the
post-2015 working groups, for which regular updates are available (JMP 2012), provides many
lessons for the water resource management community (see Box 3).

Other human water requirements, notably food production and other productive uses, are not
within the ambit of the MDG target on water supply or the JMP’s post-2015 working groups. The
aforementioned Water Poverty Index integrates measures of water use for irrigation, livestock and
industry, but the data must be gathered using dedicated household surveys, limiting the WPI’s
utility as a global monitoring indicator.
27

Indicators to track the relationship between water and industry and agriculture are available at less
refined spatial scales. The agricultural sector is a core concern because globally withdrawals for
agriculture (primarily irrigation) make up a more than 70% of the total. Various indicators attempt to
measure agricultural demand to give an idea of the agricultural sector’s vulnerability to water stress
and the potential for competition with other uses – based for example on irrigated land compared
to total cropland, or agricultural withdrawals as a share of total withdrawals (WWAP 2012). In both
cases, data are available from FAO AQUASTAT (in most countries only for a few years) though
there are issues with data reliability. In particular, estimating agricultural withdrawals is difficult
where many rural users self-supply groundwater, and where the partition between rainfed and
irrigated agriculture varies from year to year. In the case of estimates of irrigated area relative to
overall cropland, while increasing data is expected via the FAOSTAT annual questionnaires sent to
government, remote sensing methods for land-cover and evapotranspiration offer a supplementary
estimation approach (WWAP 2012).

Other relevant indicators include those for water productivity. The World Bank maintains an index
of water withdrawals against GDP, which provides some indication of economy-wide water
productivity (World Bank 2012b). The TF-IMR proposed two sectoral measures of water
productivity, for agriculture and industry, respectively. In the case of agriculture, the measure
proposed compares agricultural water withdrawals to value added by agriculture. However, the
estimates of agriculture’s value added (World Bank data) conflate the contribution of irrigated and
rainfed production, leading the Task Force to propose calculating a measure of ‘water use intensity
in irrigated agriculture’ only in those countries where the substantial majority of cropland is
irrigated. Cai et al. (2011) point out that agricultural water productivity could also be measured in
other terms, for example altering the ‘numerator’ to a non-monetary value, such as calories or
kilograms of agricultural product, or the ‘denominator’ from water withdrawals to, for example,
evapotranspiration or water diverted to irrigation. In the case of industrial water use, an equivalent
measure of water withdrawals for industry against industrial value added can be derived. However
for the industrial sector the AQUASTAT data on withdrawals is also patchy. In terms of productive
use of water, it may also be instructive to compare estimates of current hydropower capacity with
potential (International Journal on Hydropower and Dams, 2011), though an increase in this
indicator needs to be interpreted in the context of the social and environmental costs of
hydropower development. Numerous guidelines have been developed by both independent entities
(World Commission on Dams 2000) and the hydropower industry itself (International Hydropower
Association 2010), the lessons of which need to be heeded in future.

Virtual water is also a potentially useful indicator in thinking about water’s role in productive
processes. Countries with apparently low levels of water demand (e.g. on the ratio of withdrawals
to TARWR – see Section 4.1) may be reducing their water consumption by importing virtual or
embedded water from other countries, effectively displacing their water use or water polluting
activities onto other countries (Hoekstra and Mekonnen 2012). This is an extremely complex area,
with methodological challenges including tracing water footprints more than one step back in the
value chain, estimating grey water footprints associated with freshwater pollution, and the familiar
difficulty of deriving time series estimates (insufficient data). However, there is a rapidly growing
body of work on virtual water assessment (Liu and Yang 2010; Hanasaki et al. 2010; Fader et al.
2011). The normative implications of national water footprints are also not straightforward, though
evidence suggests that they play a strong but rarely acknowledged role in international trade and
countries’ ability to manage their water and land resources (Fader et al. 2011). Hoekstra and
Mekonnen propose that high external water footprints (imports of virtual water) should be explicitly
taken into account by countries when considering their foreign and trade policies. Detailed data on
water footprints per capita and per unit of product (e.g. kg of beef) can also be used to inform
considerations about where to target effort around water productivity. Data on per capita water
footprints (blue, green and grey water) of consumption and production across agricultural,
industrial and domestic sectors, as well as for international virtual water flows by product and
country, are available from the Water Footprint Network (Water Footprint Network 2012). At
country level, water footprint data may therefore be useful to cross-reference with other indicators,
for example on water productivity, withdrawals or availability.
28

4.4 Environmental needs: biodiversity and ecosystems

Our economies and societies are fundamentally dependent on ecosystems, including aquatic
ecosystems, and the services they provide (Carpenter 2005). Human water use can destabilise the
functioning of such ecosystems by reducing water quantity or discharge of emissions. This implies
a need for data on environmental flow requirements (quantity, timing and quality), human
interventions which may damage the aquatic environment (pollution, flow disruption), and the
status of freshwater ecosystems (WWAP and UNSD 2011).

Ecosystem water needs (environmental flows) have often been overlooked, even by metrics and
definitions which attempt to ground availability within its social, political and economic context.
Smakhtin et al. (2004) attempted to estimate the volumes of water required to maintain freshwater
ecosystems and the services they provide across the world’s river basins, making an important
distinction between low-flow requirement (the minimum for fish and other species through the year)
and high-flow requirement (important as a stimulus for migration and spawning, for wetland
flooding and other critical processes). The essential aim of the study was to adjust a standard ratio
of withdrawals to availability (in their case, mean annual runoff rather than TARWR) with an
estimate of the ‘Environmental Water Requirement’. Their calculations showed that basins such as
the Yellow River in China, and the Orange in Southern Africa, have been developed to the extent
that environmental flows are severely disrupted and depleted, and would face much greater water
stress/scarcity were this requirement properly accounted for. Smakhtin et al. (ibid) computed the
Environmental Water Requirement for river basins rather than individual countries, since
ecosystems are not organised along administrative boundaries. Further work would therefore need
to be done to calculate the implications for riparian countries of the Environmental Water
Requirement of transboundary river basins.

Besides withdrawals, human impacts on the aquatic environment include both emissions and flow
disruption. Of the former, phosphorous and nitrogen loading are a particular concern, causing
eutrophication and the formation of toxic nitrate, ammonia and cyano-bacteria (Vörösmarty et al.
2010). The nitrogen and phosphorous cycles have together been characterised as one of nine
critical planetary boundaries by Rockström et al. (2009). Sampled data and statistics on
phosphorous, ammonia, nitrate and nitrite (as well as metals, nutrients, organic matter and
physical-chemical characteristics) are available from the UNEP Global Environment Monitoring
System (GEMS) Water Programme. The data (including trends where available) are searchable by
river basin, major lakes and regions, as well as individual in-country monitoring stations for those
countries that have shared their data (UNEP n.d.). However, there are concerns that there are
significant temporal and spatial gaps in the GEMS data (UNESCO 2012d). Emissions sources
themselves could be monitored but data is limited. The population connected to wastewater
treatment, which is only readily available for OECD countries, is one possible proxy indicator
highlighted by the TF-IMR (UN-Water TF-IMR 2009), though in terms of nitrogen and phosphorous
this would omit the substantial contributions from other processes, notably non-point source
pollution primarily associated with fertilised agriculture.

A principal cause of flow disruption is impounding dams. While the potential importance of such
dams, and the reservoirs they store, for coping with variability and climate change (Section 4.2)
and for energy generation (Section 4.3) has been highlighted, the negative environmental
implications of their development also require attention. The fragmentation and flow regulation
effects of impounding dams can damage ecosystems through nutrient and sediment retention,
prevention of animal migration and invasion by lentic bacteria (UNESCO 2012e). Vörösmarty et al.
(2010) modelled dam density and impacts on ecosystem health on a geospatial basis (0.5 degree
resolution) using geo-referenced data on large dams from the Global Water Systems Partnership
(GWSP-GRanD) and spatial averaging of country-by-country data from ICOLD. The GRanD
database makes an important distinction between dams and reservoirs, on the basis that several
dams can be associated with a single reservoir, and run-of-the river schemes may not hold
29

reservoirs at all (Lehner et al. 2011). Nilsson et al. (2005) derived their own index of dam impacts
for the Biodiversity Indicators Partnership (BIP), accounting for fragmentation and flow regulation
effects. Severely affected river systems are characterized as those where less than one quarter of
the main channel is dam-free, the largest tributary has at least one dam, and where reservoirs
retain a considerable portion of a year’s flow (Ibid.). The indicator was prepared at basin and
regional level, and a pilot application at national level was undertaken for Sweden, but no further
developments are planned by the BIP at this stage (BIP 2011).

The third of the aforementioned data areas is the state of aquatic ecosystems themselves. Data on
freshwater species is available in time series from the Freshwater Living Planet Index, though this
is calculated at a global and regional level only (WWF 2012). In the absence of a database of
aquatic biodiversity by country, proxies are necessary. FAO time-series data on fish catch and
aquaculture could be one option (FAO 2012b; Vörösmarty et al. 2010), though this would require
some estimate of sustainable yields, which is difficult to achieve.

4.5 Competition and conflict: governance arrangements

Conflict over water can take many forms. Violent conflict between individuals or even countries is
the most visible, but far from the most likely, manifestation (see Section 3.3). In response,
governance regimes must be capable of mitigating the risks proactively rather than reactively. This
means going beyond dispute resolution mechanisms to the whole range of technical, economic,
administrative, legal, institutional and social/participatory measures (Plummer and Slaymaker
2007) that underpin water resource management, including equitable and efficient allocation
between users, enhanced access for the poorest, and protection of the environment. As such,
governance is a domain ‘fundamentally different’ from those of water availability, uses, access or
environmental degradation (UN-Water TF-IMR 2009: 18). It must often be measured in more
qualitative, process-oriented terms - what the TF-IMR referred to as ‘water governance means’
(Ibid.). Effective governance is also a pre-requisite for, or encompasses, the aspects of water
security discussed above (Sections 4.1 to 4.4), for example protection of biodiversity or allocation
between productive sectors. Furthermore, given the problems highlighted around data availability
and reliability, a further key component of governance is effective monitoring, in and of itself.

Efforts to track the evolution of water governance processes have consolidated around the IWRM
paradigm, stemming in large part from the target ‘To develop integrated water resources
management and water efficiency plans by 2005’ adopted at the Johannesburg Summit for
Sustainable Development in 2002. There is a vast body of literature on IWRM. Box 10 summarises
some of the key recent developments in trying to measure progress.

Box 10: IWRM: resourcing and operationalising

The Johannesburg Summit in 2002 gave IWRM additional international sanction and a clear target,
but one which focused primarily on putting processes in place, rather than outcomes or the
resources needed to achieve those outcomes. This has set the tone for much reporting on IWRM
progress, though there is increasing realisation that plans and policies constitute means, rather
than ends.

Work following Johannesburg, led by GWP and UN-Water, proposed consolidating reporting
around common indicators and extending the focus to management instruments and the
institutional framework, as well as the enabling environment of laws, plans and budgets. Increasing
emphasis on operationalising IWRM (not least through investment plans and vehicles), as well as
change outcomes (e.g. treatment of human waste waters), rather than just putting plans and
roadmaps in place, is evident in the GWP and UN-Water reporting framework (UN-Water and
GWP 2008; UN-Water 2008).

The latest report in this family was released in June 2012 (UNEP 2012). It remains difficult to
interrogate the quality or effectiveness of IWRM institutional or management changes, as reporting
30

is primarily achieved through a questionnaire for countries to self-assess their own progress, with
limited independent external assessment. However certain questions, for example on the trajectory
of water resource development spending as a proportion of national budgets (53% of countries
report an increasing trend in the last 20 years), show that the search continues for pragmatic
indicators which assess action rather than words.

The IWRM progress reporting led by UN-Water currently offers the best available assessment of
the highly qualitative domain of water resource management capacity at the global level. In future
the pathway approach developed for the AMCOW Country Status Overviews on Water Supply and
Sanitation might be adapted to the WRM sphere, combining sequential assessment of different
kinds of capacity (enabling, developing and sustaining services) with analysis of resourcing relative
to need, and how these translate into substantive outcomes (de Waal, Hirn and Mason 2011).

The TF-IMR point out that global and regional agreements and conventions provide their own
architecture for monitoring progress, for example wetland areas protected under the Ramsar
convention. The World Database on Protected Areas is a useful resource for tracking the progress
on this and other global, regional or national initiatives (WDPA n.d.). FAO’s legal databases on
water (WATERLEX and information on international water treaties) also offer a potential source of
data on governance (FAO 2012c), though again it is much harder to establish how far agreements,
laws and conventions are being adhered to in practice.
31

5 Politics and pragmatism: the architecture for data

gathering and interpretation
As intimated throughout Section 4, there are significant issues of underlying data availability and
quality for almost all of the indicators described. In an ideal world, development of a coherent water
security metrics framework would start with what is important, rather than what is opportune or
feasible. The paper has sought to explore what is important in proposing five normative themes
which might be encompassed by the term water security. However, indicators are only ‘fit for
purpose’ if there is reasonable confidence that they are a reliable guide to phenomena in the real-
world. The quality and availability of data are thus inevitable constraints to what is feasible. The
following subsection (5.1) gives an overview of these constraints, while subsection 5.2 indicates
the current architecture for gathering data on water resources and their management, and 5.3
outlines prospects for enhancement.

5.1 Data difficulties: variability, complexity and politics

As noted, hydrological variability presents a particular challenge for determining relevant and
reliable metrics (Section 4.2), and indicators are usually only relevant at a particular spatial or
temporal scale. At the same time, increasing spatial and temporal resolution for a given indicator
invariably increases the time and financial costs of data gathering, analysis and quality assurance.
Data for many indicators for describing and tracking important water considerations are simply
unavailable at meaningful scale.

Furthermore, the complexity of interactions between human and natural systems mean that even
before seeking to improve how water-related data is gathered, important decisions have to be
made about what information is relevant. Data on biodiversity is a case in point. While it is possible
to devise various biodiversity indicators across the common causal framework for society-
environment interactions (driver, pressure, state, impact and response - DPSIR), there is continued
uncertainty as to what are the most important concerns and how they relate to each other. For
example, the UN-Water Expert Group on Indicators, Monitoring and Data Bases cautions that the
Living Planet Index of species biodiversity ‘does not capture the most important species indicators
for water-related purposes … [and] it is difficult to link the trend with causes’ (UN-Water EG-IMD
2009: 21-22).

The political economy of data and knowledge management around water can also be a barrier, for
example when a country is reluctant to share its water resource data internationally because of
sensitivity around transboundary agreements, or when a company maintains data confidentiality for
legal or competitive reasons (WWAP 2012).

As already implied, the way data is gathered and presented can inform political priorities, and vice-
versa. Despite acknowledged shortcomings with TARWR per capita and the Falkenmark index
(Section 4.1) they are widely used and promoted by sources such as the UN-Water Key Water
Indicator Portal and the FAO AQUASTAT database. Common usage arguably lends ‘a kind of
universal and unquestioned validity’ to the scarcity thresholds, and thence to an implicit definition
of water scarcity which privileges physical availability of renewable supplies (Molle and Mollinga
2003: 542). At the same time, data availability should not be viewed as an absolute binding
constraint on what it makes sense to try to measure. Presenting a clear case for the importance of
certain parameters may increase the likelihood that required data could be obtained in future,
either by initiating new monitoring or by tapping into existing but currently unexploited sources of
information. Global WASH monitoring by the JMP is an example of the latter, since it has been
undertaken on the back of existing general household survey initiatives.
32

5.2 The existing architecture: databases and initiatives

There is, as yet, no central resource for water security data to cover the broad range of concerns
implied by this paper’s working definition of the term, or the five proposed themes. Currently most
indicators are compiled by different agencies and initiatives on specific water issues and
subsectors, for example WASH (the JMP and GLAAS reports), utility performance (IWA and
IBNET), water foot printing (the Water Footprinting Network), groundwater (IGRAC) and
biodiversity (BIP). Hitherto, FAO has been the agency with the most prominent role in compiling or
hosting water-related data across various aspects of water resources, use, development,
management and quality (most prominently in its AQUASTAT database). However, across the
water monitoring spectrum there remains a high degree of fragmentation leading to gaps,
duplication and an inevitable reliance on second or third-hand data which disguises underlying
quality problems where it is not made explicit. For the most part the above agencies and initiatives
ultimately depend on data derived from national sources, though discrepancies often arise
between international estimates and those of government. For example there may be differences
between the JMP’s estimates of water supply and sanitation access based on household surveys,
and government estimates that are derived from service providers. As a result there is a need for
vertical reconciliation (between international and national level) as much as horizontal
reconciliation between different parts of the international water monitoring system (UNSD 2007).

In response to this fragmentation, UN-Water and the World Water Assessment Programme
(WWAP) have taken a lead role in identifying relevant data and key water indicators across
different agencies and databases, and initiating a programme to mobilise information. UN-Water’s
new Key Water Indicator Portal (UN-Water 2012) suggests a desire to take on a role as central
repository for data, although to date only TARWR per capita, dam capacity per capita, % of
TARWR withdrawn, sectoral withdrawals (all via FAO AQUASTAT) and water supply and
sanitation coverage (via JMP) are available.

The Portal is the latest manifestation of an ongoing programme of work for the triennial WWDRs,
motivated by continued difficulties in establishing the state of the world’s water resources and
water management challenges. While 160 indicators were outlined in the first WWDR (published
2003), for lack of new data only 62 indicators were presented in the second report (2006) and only
30 in the third report (2003).

In 2009 UN-Water, UNESCO and the WWAP published the findings of the Expert Group on
Indicators, Monitoring and Data Bases (EG-IMD) and the Task Force on Indicators, Monitoring and
Reporting (TF-IMR). While the mandate of these initiatives was not framed in terms of monitoring
for water security specifically, they touched on numerous issues which could be embraced by the
term. The EG-IMD was convened to ‘initiate a process to identify the key dimensions and
indicators of water resources and their management as well as the work required to be able to
produce such indicators on an ongoing basis’ (UN-Water EG-IMD 2009: 2). Certain
recommendations have had traction to date, for example the programme to enhance estimates of
TARWR (Box 9) and collaboration with the UN Statistical Office around environmental accounting
for water (see below). However, there has been little progress on the majority of recommendations
made by the EG-IMD.

The TF-IMR, meanwhile, developed 15 indicators using the SMART (specific, measurable,
achievable, relevant and time-bound) criteria i.e. with specific attention to pragmatic concerns
including data availability and quality. But as indicated in Box 11, there were serious constraints
when the task-force published their findings. Negligible progress on the task-force’s ‘medium term
(within three years)’ ambitions, for example to have reliable sub-national breakdowns on most
indicators, underscores the challenge. More generally, the TF-IMR identified particular problems
with the data on water productivity, gender-related issues, water quality, wastewater production
and treatment, groundwater, biodiversity, and a widespread lack of reliable time-series data for
many indicators.
33

The WWDR 2012 included 49 indicators pertaining to water and related concerns (e.g. energy,
health), for which detailed infosheets explaining underlying definitions, computation, data
availability, scale of application and other considerations are available from the WWAP website
(UNESCO 2012f). Some of these overlap with those proposed by the TF-IMR, though not all, and
most encounter data quality and availability problems of their own.

Box 11: Fifteen indicators proposed by the UN Water Task Force on Indicators, Monitoring
and Reporting: data challenges
Fifteen key indicators were proposed by the TF-IMR for ‘assessing progress in the water sector’,
across the categories of context, function and performance. Data problems were nonetheless
significant for all at the time of publication in 2009, and have yet to improve significantly:
1. TARWR per capita (context – finite resources and population). As noted, while it is
possible to calculate this for most countries, only population data is available on a time-
series basis. FAO AQUASTAT, the main source for TARWR estimates, updates on an
ad-hoc basis when new estimates are available. Other uncertainties persist around the
data (see Section 4.1).
2. Storage capacity compared to potential (context – climate change impact and
adaptation capacity). Data is available on large and most medium dams, but is less
reliable for small dams and irrigated areas.
3. National expenditure for water supply and sanitation as percentage of total budget
(context – ability to invest for sustainable management). While budget data is available
at country level, there is as yet no central global mechanism to identify sectoral (still
less subsectoral) spend.
4. Total water withdrawals/TARWR (function – intensity of use). Problems as above for
TARWR. Withdrawals data (compiled by FAO AQUASTAT based on generic country
surveys) is generally of poor quality, with differing sector definitions, inadequate trend
data, and difficulty estimating sectoral withdrawals especially for agriculture where
there are numerous small users.
5. Use by abstraction by main sector as percentage of total withdrawals (function –
importance of consumptive uses). Problems as above for total withdrawals and sectoral
withdrawals. Accurately establishing consumptive use (taking account of e.g.
evapotranspiration and pollution) is even harder.
6. Trends in fish capture and aquaculture production (function – on-stream direct use of
freshwater services). Time series data is available (FAO FISHSTAT), though catch
data at country level does not always distinguish between marine and inland catches.
There is limited data on small-scale fisheries and few reliable estimates of overall fish
population to establish sustainable yields.
7. Share of blue, green and virtual water used to produce food in a country (function –
trade and water use). Water footprint data is improving rapidly but often requires using
a number of proxies and assumptions, e.g. deriving country level virtual water export or
import based on scaled-up average water footprints for each major product for which
trade data is available.
8. Perentage of population with access to improved water sources (performance – access
to improved water supply). Though this is arguably the area where data quality and
collection effort have been most sustained, issues persist e.g. with how to capture data
on quality, sustainability and affordability.
9. Percentage of population with access to improved sanitation (performance – access to
improved sanitation). As for water supply; tends to focus on the containment stage of
the sanitation chain and does not consider disposal, treatment and re-use.
10. Change in water productivity of irrigated agriculture, based on agriculture added value
compared to agricultural withdrawals (performance – food production). Data on
agricultural withdrawals is unreliable (see no. 4); data on agriculture added value
conflates irrigated and rainfed systems.
11. Water productivity in industrial sector, based on industrial sector added value
compared to industrial withdrawals (performance – industrial production). Data on
34

industrial withdrawals unreliable (see no. 4).

12. Change in hydropower productivity, based on production relative to potential
(performance – energy production). TF-IMR points to annually updated country
estimates of potential and installed hydropower capacity from FAO AQUASTAT and
the International Energy Agency, but does not give details. Data does not currently
appear to be freely available from either source.
13. Change in percentage freshwater of samples meeting quality standards (performance
– degradation of key renewable water resources). To be derived principally from
GEMS data, but consistency and coverage of sampling stations is uneven.
14. Urban wastewater treatment connection rates (performance – pollution mitigation
effort). Data tends only to be available for OECD countries and Europe, and even here
detail on the level of treatment is partial.
15. Threatened freshwater species (performance – risk of biodiversity loss). The EG-IMD
points out that, while the Living Planet Index estimates of freshwater species
biodiversity trends are regularly updated and could be analysed on a country basis, it
is difficult to establish causal relationships between detrimental human activities,
biodiversity trends and ecosystem services.
Indicators were also proposed around water for improved livelihoods, to be developed in the
medium term (three years) - an affordability indicator for urban areas and an indicator around
access to water for multiple use e.g. irrigation, or for rural contexts. However, no indicators of
sufficient standard were available at the time of publication.

An additional category of indicators was proposed on governance, linking to IWRM objectives.

However, specific recommendations on what to measure were not made given the nascent state of
monitoring on IWRM progress. The UN-Water reports on IWRM progress show continued effort in
this regard (UN-Water 2008; UNEP 2012).

Source: UN-Water (UN-Water TF-IMR, 2009)

5.3 Future aspirations

As noted, UN-Water and the WWAP are at the forefront of emerging initiatives, though progress on
implementing the TF-IMR and EG-IMD recommendations has been mixed. The Key Water
Indicator Portal and the PSI initiative (Box 9) are currently the most visible. However, the latter has
encountered delays in final release because of the need to collectively resolve differences in the
present estimates on TARWR and in the new dynamic estimates, before they can be rolled out.

The PSI is one instance of an attempt to harness new technologies in pursuit of more reliable data.
Remote sensing is often touted as the answer to challenges of water monitoring, including
improved spatial and temporal resolution, as well as monitoring hard-to measure aspects like
groundwater (gravimetric sensing) and quality (spectral band analysis). However, there are
concerns that this does not obviate the need to ground-truth data acquired remotely with in-situ
measurements, and thus to address the worldwide deterioration in hydrometeorological monitoring
stations (WWAP 2012). Numerous innovative proxies have been proposed to bridge data gaps,
which merit further investigation. For example mobile signal attenuation can be used to estimate
precipitation (Ibid.), and employment data can be used to estimate calculate industrial pollution
based on a widely observed correlation between number of workers and rates of BOD discharge
(UNESCO 2012g). Since the use of proxies represents a contentious, but potentially useful, way
to measure progress towards water security or other ‘sectoral’ goals, it is considered in further
detail in Box 12.

The wealth of scientific studies estimating the extent of water security challenges (for example the
work of Vörösmarty et al. 2010; 2005; 2000) suggests that it is possible to derive estimates across
35

a wide range of variables, albeit on a one-off basis and using a large number of proxies. More can
be done to ensure that the most up-to date science and technology informs policy decisions
around enhanced monitoring, underscoring the importance of concluding and releasing the results
of the WWAP PSI.

The System of Environmental-Economic Accounts for Water (SEEAW) is another key initiative
which could increase consistency and quality on water-related data. SEEAW is a subset of the
SEEA framework, which the UN Statistical Division promotes to countries in order to move beyond
the conventional System of National Accounts, to a better understanding of countries’ dependence
and impacts on their natural capital. SEEAW ‘provides a conceptual framework for organizing
hydrological and economic information in a coherent and consistent manner’, in such a way as to
consider the water flows between the hydrological system and the economy (UN DESA 2012). It
will be important to follow uptake of SEEAW, particularly in the context of the green economy
(Section 2.2). However, countries with limited capacity in conventional public financial
management and accounting are likely to face similar, if not greater, difficulties with SEEAW.

Initiatives driven by global agendas such as the SDGs and green economy must nonetheless
retain a focus on the needs of individual countries, to understand their water resource situation and
make more informed policy decisions. The regional initiative led by AMCOW to establish a pan-
African water monitoring and evaluation format is a positive example in this respect, being closely
linked to governments’ needs via the ministerial membership of AMCOW and links to the African
Union (Box 5).

Box 12: Using proxies to measure progress

Use of proxy indicators for decision making appear to run counter to the objective of evidence-
based policy making, on the logic that the evidence in question should relate directly to the issue at
stake.

But many broadly-accepted water-related indicators are proxies, especially those which describe
progress around normative concepts, for example ‘improved’ water supply or sanitation. Wherever
there is a dearth of directly related data, proxies provide a possible alternative. The different ways
in which proxy indicators can relate to the policy issue can offer other advantages. On the one
hand, proxies that are somewhat further ‘upstream’ in the causal chain, relating to processes or
inputs, are more likely to be within the direct sphere of influence of the policy community in
question. On the other hand, proxies which measure broader ‘downstream’ outcomes may actually
be more significant in a global public goods sense, though they often require simultaneous action
by others outside the immediate policy community. This latter category of proxies may be
particularly relevant where the policy space is crowded with different issues, with equal claim to
decision makers’ attention (as is the case with the processes on the Post-2015 Development
Agenda and SDGs, for example).

To take the example of a policy maker seeking to increase productivity of agricultural water use:
Data for the ‘directly relevant’ indicator, agricultural water productivity (value added per m3
withdrawn) is collated by the FAO. However, for reasons including patchy data on withdrawals, and
the difficulty of separating value added by irrigated and rainfed farming, it is not straightforward to
estimate. The policy maker may chose to focus instead on an input or process which it can be
reasonably assumed will lead to increased agricultural productivity, for example the area equipped
for irrigation (again collated by the FAO). This has advantage of being more within the influence of
policy choices, and in this case the data may be slightly better, offering a further pragmatic
incentive. On the other hand, it is one (or more) steps back in the causal chain, and there is no
guarantee that increasing the area equipped for irrigation will result in an increase of functioning, or
productive, irrigation.

An alternative would be to look further down the causal chain to broader public goods outcomes,
which may require inputs from other sectors. This can be a valid exercise in and of itself, to
36

(re)consider what’s really important: does agricultural water productivity matter in and of itself, or
because it contributes to reduced malnutrition (which depends also on e.g. access to WASH,
various infrastructure services and healthcare) or poverty reduction (which arguably depends on
action across all sectors). Obviously, choosing such an indicator removes the metric, and perhaps
the political and financing attention that come with it, from the direct influence of the sector in
question. But some would argue this is a necessary ‘sacrifice’ where there is a risk of
fragmentation across numerous issues, as is inevitably the case in devising global development
and environment goals and targets. Malnutrition is nonetheless an extreme example. For most
water managers, malnutrition would be seen as simply too far removed from their influence, with
agricultural productivity not even being the main concern in many contexts, for example where food
distribution systems are inadequate.

Proxies are widely used in the water policy and science communities. For example Vörösmarty et
al. (2010) employ numerous proxies in compiling their indices of threat to human water security
and biodiversity. It is therefore recommended that any water security metrics framework should
make considered use of proxies, taking close account of their different attributes, particularly in
relation to the two axes depicted in Figure 4.

The large amount of data collected by private corporations is a further source that could be
significantly developed. This will entail negotiating difficult terrain of corporate competition and law
to make the incentives for increased openness clear, and remove barriers. Narratives around
shared water risk (SABMiller, GTZ and WWF 2010; UN Global Compact 2011) imply an awareness
that new modes of governance, public and private, will be required to address water security
challenges. Robust engagement is needed to make clear the quid pro quo: shared risk requires
shared, transparent data. Investor- and public-relations also provide strong motivations for sharing
water data, and investing in data gathering and analysis (see for example Ceres’ Aqua Gauge and
WRI’s Aqueduct tools).

It is important to be realistic about the pace of change: technology is unlikely to yield a sudden
revolution in water monitoring; the complex incentives that discourage companies and countries
from sharing data will take time to re-orientate; a long decline in capacity and instrumentation for
water monitoring in many countries will not be reversed overnight. But beyond making
recommendations on the most relevant data and indicators with regards to water security, a take-
home message of this paper is the urgent need to improve the availability and quality of data, and
the important opportunities to do so.
37

Figure 4: Positioning proxy indicators in relation to the challenge of increasing productivity of

irrigated agriculture: according to pragmatic considerations (y-axis) and relevance to sector
decision-making (x-axis)

Source: Author
38

6 Recommendations

6.1 Introduction to proposed indicators and caveats

Five key themes encompassed by the concept of water security (Box 8) were proposed in order to
review existing indicators and measurement approaches in Section 4. Bearing in mind the caveats
relating to the practical and political difficulties to obtain sufficient data of good quality (Section 5),
this section presents options for a limited number of indicators which could form part of a
framework for measuring water security as described by the five themes, on a recurrent basis, at
global and national level.

For each of the five themes – titled resource stress, variability and risk, basic human needs and
productivity, environmental needs, and governance – two sets of indicator options are proposed in
the tables that follow: a first set that could currently be obtained from existing global datasets, and
a second, more aspirational set that would require considerable further work in terms of primary
data gathering and/or analysis. Effort is made to highlight the coverage (number of countries) of
each indicator, relevant data sources, and to identify key issues relating to their calculation, use
and interpretation, especially for defining associated policy goals and targets.

More generally, it is important to underscore the following:

• The proposed indicators follow from the underlying themes of water security and
working definition proposed in Box 8. Alternative definitions and typologies are of
course available (Grey and Sadoff 2007; ODI, DIE and ECDPM 2012; WWAP and
UNSD 2011; GWP 2012) and more are likely to emerge as wider debates evolve e.g.
around resource security and the SDGs. The definition and typology used in this paper
are therefore not proposed as definitive, but are used as conceptual devices to
structure thinking around the abstract concept of water security.
• Similarly, indicator options are not proposed as a definitive or necessarily coherent list,
in and of themselves. They are intended to promote dialogue on the technical aspects
of measuring water security, which needs to evolve on a parallel, iterative basis to
political debates about how water security should be defined. Alternative or additional
indicator options may be available. Similarly, prioritising different themes for water
security might yield different indicators. Even within the options currently proposed,
only a selection of indicators would be required – a key lesson from the MDG
monitoring framework is the importance of simplicity.
• Among the different windows of opportunity identified in Section 2, the indicator options
presented are in general most relevant to the SDG/ post-2015 debates i.e. indicators
that can be monitored at global level across most countries. Some of the indicators
could also be relevant to thinking on the green economy (e.g. agricultural and industrial
water productivity) or orienting the WRM policy and spending priorities of national
governments and their donor partners (e.g. flood mortality risk index). However, further
work is needed to match indicators to the particular priorities, country focus, and
spending time-frames of particular countries.
• In addition to general limitations on data availability, disaggregated data (for example
by gender and different socio-economic groups) is particularly weak. A general point is
that, for currently feasible and ‘aspirational’ indicators alike, enhanced disaggregation
is needed to improve their usefulness as policy tools and address the much greater
water security challenges faced by the poorest and most vulnerable in society.
• Notwithstanding the complexity of water systems and the inadequacy of information,
effort has been made to focus on simple indicators where the underlying data is readily
apparent, rather than complex, weighted composite indicators. Some of the indicators,
and especially the ‘aspirational’ options, may rely on innovative methods (modelling,
39

proxies, remote sensing) to fill data gaps, though these do not remove the need for
ground-truthing.
• Consideration should be given to how to name and articulate the indicators
themselves, recognizing that they are as much political and communications tools, as
they are analytical devices.
 6.2 Water security indicator options

Resource stress
Key data: renewable water availability (external, groundwater, surface components); withdrawals (by sector); capacity to access (proxies e.g.
malnutrition or HDI score)

Indicator name Calculation Data sources/ Countries Indicates Technical notes Potential implications of
method; units relevant studies covered1 use as country/ global
targets

Currently feasible
Renewable TARWR/ FAO AQUASTAT 183 Theoretical Subcomponents of TARWR, e.g. Gives sense of the
water population; renewable water dependence on external water environmental water
resources per m3/person/year availability per resources, may also be endowment but privileges
capita person; inverse informative; time series data physical water
(people/ represents change in pop’n not scarcity/insecurity over other
1000000m3 ) gives TARWR; masks spatial forms
sense of water variability; does not account for
crowding non-renewable sources, e.g.
non-renewable groundwater,
which may be significant and for
which there may be a rationale
for exploitation; continued
concerns around data quality
Social People/m3 FAO AQUASTAT, 183 Water crowding Explicitly incorporates non-
resource water TARWR/HDI; no UN Population (people per m3) physical forms of water
stress unit (weighted Division, UNDP; adjusted by HDI scarcity/insecurity, but HDI
composite) originally suggested score as a proxy may not be reliable proxy for
by Ohlsson (1998) for capacity to social capacity to access
access water; water; unitless composite
water abundant indicator requires expert
developing interpretation and is not
countries still show intuitive
up as socially
water stressed

1. Estimate. In most cases, figures taken from UN-Water TF-IMR (2009).

Relative water Total FAO AQUASTAT 160 Intensity of water Data on withdrawals are Explicitly considers the
stress withdrawals/ withdrawals unreliable, especially for relation between demand and
TARWR; % ratio relative to agriculture availability as a component of
renewable supply water security, but may
conceal social/economic
water scarcity/insecurity, e.g.
in countries with limited water
infrastructure for abstraction
and conveyance

Aspirational
Dynamic water Dynamic Dynamic TARWR Most Dynamic water Data not yet released (dynamic As for ‘standard’ water
resources per TARWR/ estimates from countries availability (cross- TARWR requires reconciliation resources per capita above,
capita population; WWAP; UN in theory referenced with with standard TARWR though has the advantage of
(dynamic) m3/person/year Population Division remote sensing estimates) greater accuracy and ability to
data; permits trend track trends; need for
analysis of reconciliation indicates that
TARWR over time) physical water availability may
per person be politically sensitive,
especially where
transboundary waters are
contested
Non- Geosptatially Dynamic TARWR Most Locally estimated Data on dynamic TARWR not As for relative water stress
sustainable derived water estimates from countries withdrawals yet released; grid-cell estimates (above); also note above
water use withdrawals/ WWAP; geospatially in theory relative to dynamic need to be extrapolated to caveats on reconciling
(dynamic/ dynamic TARWR; derived water estimate of country level; geospatial conventional and dynamic
locally derived unitless ratio withdrawals from renewable supply estimates of water withdrawals TARWR estimates; may
version of Centre for are derived from grid cell- discourage temporary use of
relative water Environmental specific estimates of population non-renewable sources and
stress) Systems Research/ and irrigated land density with investment in innovation to
GWSP Digital Water available water modelled develop alternatives
Atlas through digital river networks
Variability and risk
Key data: hydrometeorological/precipitation variability; water storage and conveyance capacity; population exposure and vulnerability

Indicator Calculation method; units Data sources/ Countries Indicates Technical notes Potential implications
name relevant studies covered of use as country/
global targets

Currently feasible
Water Dam storage capacity/ UN-Water Key 107 for dam Water storage Validity of storage as proxy for Underscores the
storage population*100000 [conversion Water Indicators capacity; 49 capacity as a resilience to variability depends importance of a basic
capacity factor]; AND/OR regular Portal (dam for regular proxy for ability on extent and timing of variability platform of hydraulic
renewable groundwater per storage capacity renewable to manage and complex hydrological and infrastructure, but
capita; m3/person per capita); FAO groundwater rainfall hydrometeorological interactions insensitive application
AQUASTAT variability (stream flow, recharge etc.); may encourage
(regular between definitions of large dams vary by ‘hydraulic mission’ and
renewable seasons country; small and medium heavy engineering at
groundwater) dams may be omitted, effects of the expense of other
silting not taken into account; solutions (i.e. if lessons
GRanD database (GWSP from previous large
2012a) may be more accurate dam development are
than ICOLD (georeferences and not learned)
specifications); for groundwater
availability, estimates are
unreliable, infrequently updated
and unavailable for many
countries
Flood Risk calculated as function of GAR Global risk Almost all Risk of mortality Equivalent drought indicator has Captures destructive
mortality hazards (GIS data), data platform, countries from flood as a not been calculated due to the potential of water as a
risk vulnerability (statistical GAR (UNISDR, function of difficulty of deriving accurate component of water
analysis of historical events), 2011b) hazard, mortality statistics insecurity, but better
and modelled population vulnerability ability to estimate flood
exposure; unitless risk index and exposure and other disaster risks
may have diverted
international attention
and funds from tackling
more complex drought
risk; unitless composite
indicator requires
expert interpretation
 Aspirational
Rainfall Coefficient of variability for CMI Environment Most Vulnerability to Currently calculated on grid-cell Emphasises rainfall
variability (StdDev(CMI)/Mean(CMI); CrossRoads countries in periodic water basis, would require conversion variability as a more
unitless ratio Initiative, CUNY theory stress for country-by-country estimates important predictor of
economic growth than
physical water
availability, but unitless
composite indicator
requires expert
interpretation, and
underlying data is not
intuitive
Climate Risk index based on Sullivan and Most Vulnerability to Currently calculated on grid-cell Can capture
vulnerability topographical variability, water Huntingford countries in impacts on basis, would require conversion stakeholder
resources, water access, water (2009) theory water resources for country-by-country estimate perceptions of
utilisation, human and as a function of vulnerability across
institutional capacity; unitless topography, Note: a general point for all multiple dimensions,
composite index water indicators that require estimates but differences in
availability and of climate change impacts on weightings (as well as
access, and hydrology for their calculation - data availability) across
capacity improvements in model skill and countries may limit
linkage to hydrological models usefulness of
are required international
comparisons; not a
formal probabilistic
assessment of risk
Basic human needs and productivity
Key data: access to drinking water and sanitation; agricultural/industrial withdrawals

Indicator Calculation Data sources/ Countries Indicates Technical notes Potential implications of use as
name method; units relevant covered country/ global targets
studies

Currently feasible
Access to Population using an JMP Almost all Percentage of the Ongoing debates around Existing MDG target on access to
drinking improved source of countries population using an ability to measure service drinking water has done much to
water drinking water/total improved source of quality, accessibility, galvanise political attention and
population; drinking water affordability, sustainability finance around the issue, but this
percentage etc; ‘aspirational’ equivalent may have come at the expense of
for this indicator to be wider water resource
identified based on findings management
of JMP Working groups
Access to Population using an JMP Almost all Percentage of the As for access to drinking Existing MDG target on access to
sanitation improved sanitation countries population using an water (above) sanitation has done much to
facility/ total improved sanitation galvanise political attention and
population; facility finance around the issue, but risks
percentage overlooking environmental/health
impacts of poorly managed
wastewater or faecal sludge
Irrigated Value added (value FAO/ FAO Only countries Water productivity Data on agricultural Emphasises the importance of
agricultural of output less value AQUASTAT where irrigated of irrigated withdrawals may be water as a contributor to economic
water of immediate agriculture agriculture unreliable, especially where growth, but cannot be calculated
productivity consumption) by predominates there are numerous small- for countries where rainfed
agriculture/ scale users, and is updated agriculture predominates,
agricultural water only every 5 years; can only including much of Sub-Saharan
withdrawals; be calculated for countries Africa, where agriculture is also a
US$/m3 where irrigation major economic sector
predominates as estimates
of irrigation-only value
added are not available
Industrial Value added by FAO 162 Water productivity Data on industrial Emphasises the importance of
water industry/ industrial AQUASTAT, of industry withdrawals may be water as a contributor to economic
productivity water withdrawals; World Bank unreliable and is updated growth, but does not distinguish
US$/m3 only every 5 years non-consumptive uses of water,
where water is not significantly
 polluted or evaporated, e.g.
cooling for thermal power
Aquaculture Aquaculture FAO fisheries Most countries Productivity of Data better for aquaculture Emphasises the increasing
production production/ and aquaculture of aquatic animals than for importance of aquaculture to diets,
population; tonnes aquaculture (farming of aquatic aquatic plants, which may especially in providing protein, but
per capita department organisms) relative still be important for some intensive aquaculture can have
to population countries; fisheries (aquatic significant negative environmental
organisms exploitable by the impacts, and can be difficult to
public as a common establish and enforce sustainable
property resource) production thresholds for both
production may be used in fisheries and aquaculture
addition/instead

Aspirational
Virtual water Various e.g. Water 174 Indicates e.g. Time series data is not yet Many potential policy applications
footprint dependence on Footprint dependence of available (average 1996- and implications, e.g. could be
virtual water Network; water scarce 2005) used to focus attention on the
imports; national Hoekstra and countries on potential for virtual water trade to
blue, green and Mekonen imports of virtual mitigate against localised water
grey water footprints (2012) water to balance scarcity, but thinking is relatively
of consumption water budget; or young and virtual water footprint
AND/ OR displacing of water data needs careful interpretation
consumption; m3/ use and impacts to
capita other countries
Environmental needs
Key data: key freshwater species populations; water withdrawals; water availability; dam locations and size; remote sensing spectral data; various
variables to estimate Environmental Water Requirement

Indicator name Calculation Data sources/ Countries Indicates Technical notes Potential implications
method; units relevant studies covered of use as country/
global targets

Currently feasible
Freshwater Annual change in WWF/ZSL Living Global Status and trends on Currently index is Focuses attention on the
species freshwater Planet Index (WWF, freshwater species unavailable by country, importance of
species; Unitless 2012) though LPI concept note biodiversity for provision
Living Planet implies this is possible of freshwater ecosystem
Index (ZSL and WWF 2012); services
currently focuses on
vertebrates only;
estimates are derived by
tracking a limited number
of key species, with data
compiled from numerous
different sources;
interactive open-access
data is planned

Aspirational
Water re-use Aggregate CUNY Environmental Most Level of upstream human Further work is needed to Emphasises the way
index upstream water Crossroads Initiative, countries in withdrawals relative to extrapolate grid-cell water withdrawals
demand/total according to UNESCO theory supply; increase indicates calculation to country impact downstream
water supply; % (2012h). greater likelihood that level; environmental flow through natural
index water abstraction will requirement is not hydrological flows, but
jeopardise environmental uniform; better data on making clear conceptual
flows; note that this is extent of irrigated areas distinction with relative
calculated on geospatial is required water stress index
grid-cell basis using (water re-use index
estimates of national looks at upstream use)
industrial and domestic may be difficult
water demand, population
data, irrigated land extent,
 discharge fields and digital
river networks
Environmental Water GWSP Digital Water Most Water withdrawals relative Broad consensus needed Integrates
adjusted water withdrawals/ Atlas. GWSP (2012b); countries in to available water, taking on appropriate method of environmental water
stress available water, Smakhtin et al. (2004) theory into account environmental calculation, using needs which are often
adjusted for flows (calculated based on Smakhtin et al.’s study as overlooked, but
estimated national level TARWR and a starting point (2004) estimating these is
Environmental withdrawal estimates or notoriously difficult
Water grid-cell derived estimates
Requirement; of supply and demand –
percentage ratio see ‘Index of non-
sustainable water use’
(above)
Water quality Change in Spectral data from In principle Water quality of lakes While individual studies Draws attention to
chlorophyll/ MODIS satellite, most have been undertaken health of major water
turbidity/ initially for 600+ large countries using this technique, bodies (lakes and
suspended solids; lakes monitored by US where there there is no currently reservoirs) but
various units Department of are large available global potentially overlooks
Agriculture Foreign standing assessment; UN Water threats to riverine
Agriculture Service; for surface TF-IMD estimate cost for (flowing) systems
surface elevation water bodies setting up initial analysis
change atmospheric (lakes) at US$100k-US$1m, with
correction required; annual repetitions at less
UN-Water EG-IMD than US$100k
(2009)
River Geospatial Various sources In principle River fragmentation and Overlooks small dams; Emphasises the
fragmentation estimate calculate dam density, most flow regulation based on work needed to potential negative
and flow e.g. GRanD (GWSP countries distribution of dams extrapolate country-level implications for aquatic
regulation Digital Water Atlas relative to river networks estimates from geospatial ecosystems of river
(GWSP 2012a); UMEA data; impounding dams fragmentation, but does
University (BIP 2011); and run-of-river schemes not take account of
ICOLD data on large may have very different potential economic
dams can also be impacts in terms of flow benefits of water
probabilistically regulation and river infrastructure (e.g.
distributed (Vörösmarty fragmentation dams)
et al. 2010)
Treated % of wastewater Currently data is only Currently Percentage of wastewater Even if data can be Draws attention to the
wastewater flows receiving reliably available for only OECD flows sourced for non-OECD importance of
primary/ OECD countries countries countries, national wastewater treatment
secondary/ wastewater treatment for human and
tertiary treatment; methods and standards environmental health, of
percentage ratio vary significantly, making increasing importance
comparison difficult; with urbanisation, but
needs to be discussed in potentially overlooks
context of any post-2015 non-point sources of
sanitation monitoring pollution that in some
proposal contexts (especially
developed countries)
are a more significant
problem
Governance2
Key data: data on IWRM process components (see footnote); RAMSAR wetland status data (qualitative) and remote sensing data on wetland extent
(quantitative); data on budgetary commitments to aspects of WRM and sanitation.

Indicator Calculation method; Data sources/ Countries Indicates Technical notes Potential
name units relevant studies covered implications of use
as country/ global
targets

Currently feasible
IWRM Based on survey UN-Water (2008), More than How far countries have Survey method – third party Captures the
planning questionnaire sent to UNEP (2012) and 130 countries progressed in meeting the verification undertaken for importance of
national governments AMCOW (2012a) 2002 Johannesburg few countries; represents governance
(self-assigned into 5 summit commitment ‘to only ‘national government processes and
qualitative categories); develop integrated water perspective’ (and often that systems, but does
percentage of resources management of the individual completing not give indication of
countries in each and water efficiency the survey) the quality of those
category plans’; other similar processes, or how
indicators, looking at e.g. they relate to
water quality monitoring, outcomes
information management
or early warning systems,
could be selected from
the survey results
International Proportion of de Stefano et al. Countries with Extent to which countries Treaties and institutions Underscores the
water transboundary waters (2010) transboundary are managing potential (RBOs) indicate only need for robust
governance identified as waters water disputes through capacity to manage institutional
vulnerable to current transboundary water tensions in principle; some mechanisms to
or future water stress management work needed to translate de manage tension
that are covered by a agreements, including Stefano et al.’s basin- between countries

2. Process-oriented indicators such as status of implementation of IWRM policies, plans and mechanisms are nominally outside the scope of this paper, which focuses on water
security outputs and outcomes. Existing initiatives are underway to monitor progress in putting in place different aspects of an IWRM architecture. This is principally undertaken
through a UN-Water led initiative (UNEP 2012; AMCOW 2012a) to monitor progress against various process components, including the enabling environment (policies, plans);
institutional frameworks (institutions, participation, capacity building); management instruments (information, demand management programmes, knowledge sharing); and
infrastructure and development financing (investment plans and status of development). The information is derived principally from surveys sent to government, with limited third party
verification, so there is potential for inconsistency. However, such process-oriented monitoring remains the current best available option for assessing nominal progress on water
governance. Hence in the short term, the paper recommends that a holistic approach to water security monitoring can draw on the UN-Water led initiative, while making it explicit that
elaborating a process on paper does not equal progress on the ground, in terms of IWRM outcomes. In the longer term, it is hoped that countries’ progress against specific, outcome-
oriented commitments (e.g. the RAMSAR convention) can be monitored.
 treaty (with mechanisms and oriented index into a over water
mechanisms for institutions for managing workable ratio at country resources, but
allocation, variability variation and dispute level historic evidence
management, and and expert opinion
conflict resolution) and suggests localised
one or more river unrest and conflict
basin organisations relating to water is
(RBOs); percentage more likely than
interstate water
conflict

Aspirational
Water Percentage of key UNEP (2012); the Most countries Extent of national Data reconciliation is an Underlines the
monitoring water indicators 2012 IWRM progress in principle monitoring effort and arduous, long-term but importance of data
effort reliably tracked by assessment supported need for reconciliation essential process; reliably gathering and
national government; by UN-Water included between national and establishing what countries analysis for informed
percentage an additional ‘level-2’ international water are regularly monitoring is policy decisions
assessment monitoring systems only a first, essential step
undertaken by
external consultants
for a selected subset
of countries, which
included assessing
whether countries
were regularly tracking
42 different indicators
Protection of Change in extent of UN-Water EG-IMD 161 parties to Extent to which countries Could equally be applied as Emphasises the
aquatic wetland areas (2009) the RAMSAR are meeting their an ‘Environmental needs’ importance of
environments protected as Ramsar convention environmental indicator, especially if non- wetlands as
Sites; in addition to governance Ramsar wetlands were important loci for
qualitative and some responsibilities in relation included (i.e. not just freshwater
quantitative reporting, to wetlands monitoring ‘governance’ in ecosystem services
this indicator would terms of adherence to
ideally assessed using international commitments);
remote-sensing data wetland extent would
on the actual extent of ideally be complemented
Ramsar-designated with e.g. data on water
wetlands over time, quality or Ramsar reporting
with ground-truthing on key species
Water Compliance with UN-DESA (2012) Most countries Extent to which countries Could incentivise the
accounting environmental in principle are monitoring the collation of large
accounting standards/ interaction between volumes of new,
frameworks e.g. water resources and standardised data
SEEAW; no unit, or their economies, and on water and its
percentage how far growth is based relation to the
compliance with on degradation of natural economy, but
different aspects of capital significant capacity
framework building will be
required for effective
uptake of e.g.
SEEAW in many
countries
 Annex 1: The various UN processes around the SDGs and Post-2015 Development Agenda

Source: CAFOD (2012)

53

References
2030 Water Resources Group (2012) The Water Resources Group: Background, Impact and the
Way Forward, Briefing report prepared for the World Economic Forum Annual Meeting 2012 in
Davos-Klosters, Switzerland.

AMCOW (2012a) Status report on the application of integrated approaches to water resources
management in Africa, African Ministers' Council on Water, African Union, EU Water Initiative.

AMCOW (2012b) The pan African Water and Sanitation Monitoring, Evaluation and Reporting
Format: The Guidelines, African Ministers' Council on Water.

Bates, B., Kundzewicz, Z., Wu, S. and Palutikof, J. E. (2008) Climate Change and Water,
Technical Paper of the Intergovernmental Panel on Climate Change, IPCC Secretariat, Geneva.

Batton, B., Adrio, B., Hampton, D. and Lynn, W. (2011) The Ceres Aqua Gauge: A Framework for
21st Century Water Risk Management, Ceres, Boston MA.

BIP (2011) River fragmentation and flow regulation, Retrieved June 9, 2012, from Biodiveristy
Indicators Partnership: http://www.bipindicators.net/riverfragmentation

Black, M. and Hall, A. (2004) ‘Pro-poor water governance’, In A. D. Bank, Water and poverty: the
themes (pp. 11-20), Asian Development Bank, Manila.

Brown, C. and Lall, U. (2006) ‘Water and economic development: The role of variability and a
framework for resilience’, Natural Resources Forum, 30: 306-317.

Burke, J.J., Sauveplane, C and Moench, M. (1999) ‘Groundwater management and socio-
economic responses’, Natural Resources Forum, 23: 303-313.

CAFOD (2012) Post-2015 policymaking: What is being planned, what might actually happen, and
CAFOD's current policy lines, CAFOD, London.

Cai, X., Molden, D., Mohammed, M., Sharma, B., Ahmad, M.-u.-D. and Karimi, P. (2011)
‘Producing more food with less water in a changing world: assessment of water productivity in 10
major river basins’, Water International, 36(1): 42-62.

Calow, R.C. and MacDonald, A.M. (2009) ‘What will climate change mean for groundwater supply
in Africa?’ ODI Background Note, Overseas Development Institute, London.

Calow R.C., MacDonald A.M., Nicol A.L. and Robins N.S. (2010) ‘Ground Water Security and
Drought in Africa: Linking Availability, Access and Demand’, Ground Water, 48(2): 246-256.
10.1111/j.1745-6584.2009.00558.x

Carpenter, S. (2005) Ecosystems and Human Well-being. Volume 2: Scenarios. Findings of the
scenarios working group, Millennium Ecosystem Assessment, Washington, D.C.: Island Press.

CGIAR (2011, December 20) Can we feed the world making efficient and fair use of available
water? Retrieved March 2, 2011, from http://consortium.cgiar.org/can-we-feed-the-world-making-
efficient-and-fair-use-of-available-water/

CNA Corporation (2007) National Security and the Threat of Climate Change, Alexandria, CNA
Corporation, VA.

The Coca-Cola Company (2012) The Water Stewardship and Replenish Report, Atlanta, The
Coca-Cola Company, GA.
54

CoFR, US Senate (2011) Avoiding Water Wars: Water Scarcity and Central Asia's Growing
Importance for Stability in Afghanistan and Pakistan, U.S. Government Printing Office,
Washington.

COMMAN (2005) ‘Managing groundwater resources in rural India: the community and beyond’,
British Geological Survey Commissioned Report, CR/0535N.

Conway, D. and Schipper, E. L. (2011) ‘Adaptation to climate change in Africa: Challenges and
opportunities identified from Ethiopia’, Global Environmental Change, 21: 227-237.

Cook, C. and Bakker, K. (2012) ‘Water security: Debating an emerging paradigm’, Global
Environmental Change, 22(1): 94-102.

DCDC, MoD (2010) Global Strategic Trends - Out to 2040, Development, Concepts and Doctrine
Centre, Ministry of Defence, UK Government, London.

de Stefano, L., Duncan, J., Dinar, S., Stahl, K., Strzepek, K. and Wolf, A. (2010) ‘Mapping the
Resilience of International River Basins to Future Climate Change-Induced Water Variability’,
Water Sector Board Discussion Paper Series, No. 15, World Bank, Washington DC.

de Waal, D., Hirn, M. and Mason, N. (2011) Pathways to Progress: Transitioning to Country-Led
Service Delivery Pathways to Meet Africa's Water Supply and Sanitation targets, Water and
Sanitation Program – Africa, Nairobi.

DFID (2011) How to note: Guidance on using the revised Logical Framework, UK Department for
International Development, London.

DFID (n.d.) International Climate Fund (ICF) Implementation Plan 2011/12-2014-15: Technical
Paper, Retrieved May 28, 2012, from: www.dfid.gov.uk:
http://www.dfid.gov.uk/Documents/publications1/uk-International-Climate-Fund-techncial-working-
paper.pdf

Dobbs, R., Oppenheim, J., Thompson, F., Brinkman, M. and Zornes, M. (2011) Resource
Revolution: Meeting the world's energy, materials, food and water needs, McKinsey Global
Institute, McKinsey and Company.

Evans, A. and Steven, D. (2012) ‘Beyond the Millennium Development Goals: Agreeing to a Post-
2015 Development Framework’, MGO Working Paper, Managing Global Order, Brookings, CISAC
Stanford, NYY Centre on International Cooperation, New York.

Fader, M., Gerten, D., Thammer, M., Heinke, J., Lotze-Campen, H., Lucht, W. and Cramer, W.
(2011) ‘Internal and external green-blue agricultural water footrpints of nations, and related water
and land savings through trade’, Hydrology and Earth Systems Science, 8: 483-527.

Falkenmark, M., Lundqvist, J. and Widstrand, C. (1989), ‘Macro-scale water scarcity requires
micro-scale approaches: Aspects of vulnerability in semi-arid development’, Natural Resource
Forum, 13(4): 258-267.

FAO (2011, July 21) FAOSTAT, Retrieved June 09, 2012, from Food and Agriculture Organisation
of the United Nations: http://faostat.fao.org/site/377/DesktopDefault.aspx?PageID=377#ancor

FAO (2012a) AQUASTAT, Retrieved June 9, 2012, from aquastat:

http://www.fao.org/nr/water/aquastat/dams/index.stm
55

FAO (2012b) Fisheries and Aquaculture Department, Retrieved June 9, 2012, from FishStat Plus -
Universal software for fishery statistical time series:
http://www.fao.org/fishery/statistics/software/fishstat/en

FAO (2012c) FAOLEX, Retrieved June 9, 2012, from FAO Legal Office: http://faolex.fao.org/

Foster, S., Nanni, M., Kemper, K., Garduno, H. and Tuinhof, A. (2003) Utilization of Non-
Renewable Groundwater: a socially-sustainable approach to resource management, World Bank,
Washington D.C..

GEMI (n.d. a) Connecting the Drops: Towards Creative Water Strategies: A Water Sustainability
Tool, Retrieved April 4, 2012, from Global Environmental Management Institute (GEMI):
http://www.gemi.org/water/

GEMI (n.d. b) Collecting the Drops: A Water Sustainability Planner, Retrieved April 25, 2012, from
Global Environmental Management Institute (GEMI): http://www.gemi.org/waterplanner/

GEMI (n.d. c) GEMI Local Water Tool (LWT), Retrieved March 8, 2012, from Global Environmental
Management Initiative (GEMI): http://www.gemi.org/localwatertool/

Giordano, M. (2009) ‘Global Groundwater? Issues and Solutions’, Annual Review of Environment
and Resources, 34: 153-178.

Gleeson, T., Yoshihide, W., Biekens, M.F.P. and van Beek, L.P.H. (2012) ‘Water balanace of
global aquifers revealed by groundwater footprint’, Nature 488: 197-200.

Grey, D. and Sadoff, C. W. (2007) ‘Sink or Swim? Water security for growth and development’,
Water Policy, 9(6): 545-571.

GWP and INBO (2012) The Handbook for Integrated Water Resources Management in
Transboundary Management of Rivers, Lakes and Aquifers, International Network on Basin
Organizations and Global Water Partnership, Paris.

GWP (2009) Global Water Security Submission by the Global Water Partnership to
ICE/RAE/CIWEM Report to Professor John Beddington, Chief Scientific Adviser to HM
Government, Global Water Partnership Secretariat, Stockholm.

GWP (2012) What is Water Security? Retrieved July 26, 2012, from Global Water Partnership:
http://www.gwp.org/The-Challenge/What-is-water-security/

GWSP (2012a) GRanD Database, Retrieved June 14, 2012, from GWSP Atlas:
http://atlas.gwsp.org/index.php?option=com_contentandtask=viewandid=207andItemid=68

GWSP (2012b) Environmental Water Stress Indicator, Retrieved June 14, 2012, from GWSP Atlas:
http://atlas.gwsp.org/index.php?option=com_contentandtask=viewandid=189andItemid=63

Hall, J. (2012) ‘Risk-based principles for defining and managing water security. Extended
Abstracts’, Water Security, Risk and Society: International Water Security Conference, 16-18 April
2012, St Hugh's College, Oxford, UK. (pp. 4-5 of 178), Oxford University Water Security Network,
Oxford.

Hanasaki, N., Inuzuka, T., Kanae, S. and Oki, T. (2010) ‘An estimation of global virtual water flow
and sources of water withdrawal for major crops and livestock products using a global hydrological
model’, Journal of Hydrology, 384: 232-244.

Harding, R. and Warnaars, T. (2011) Water and Global Change: The WATCH Project Outreach
Report, Centre for Ecology and Hydrology, Wallingford.
56

Hirn, M. (2011) Rural Water Point Mapping: Preliminary Analysis and Verification Brief, Liberia
WASH Consortium, Monrovia.

Hoekstra, A. and Mekonnen, M. (2012) ‘The Water Footprint of Humanity’, PNAS, 109(9): 3232-
3237.

Hope, R. (2012) Water Security, Risk and Society - Key Issues and Research Priorities for
International Development, Synthesis report submitted to UK Collaborative on Development
Sciences by Oxford University Water Security Network, June 2012.

ICOLD (2012) Register of Dams, Retrieved June 9, 2012, from International Commission on Large
Dams: http://www.icold-cigb.org/GB/World_register/world_register.asp

IFAD (2007) Gender and water, International fund for Agricultural Development, Rome.

IGRAC (2012, May) IGRAC, Retrieved June 9, 2012, from Global Overview: http://www.un-
igrac.org/publications/119

IISS (2011) The IISS Transatlantic Dialogue on Climate Change and Security - Report to the
European Commission, International Institute for Strategic Studies, London.

International Hydropower Association (2010) Hydropower Sustainability Assessment Protocol,

November 2010, International Hydropower Association, London.

International Journal on Hydropower and Dams (2011) Hydropower and Dams World Atlas, Aqua-
Media International Ltd.

IPCC (2012) ‘Summary for Policymakers’, In Field, C.B., Barros, V., Stocker, T.F., Qin, D., Dokken,
D.J., Ebi, K.L., Mastranda, M.D., Mach, K.J., Plattner, G.K., Allen, S.K., Tignor, M. and Midgley,
P.M. (edis.) Managing the Risks of Extreme Events and Disasters to Advance Climate Change
Adaptation, IPCC Special Report, (pp. 1-19), Cambridge University Press, Cambridge, UK and
New York, NY, USA.

Jenkinson, K. (2011, September 19) The Coca-Cola Company Donates Extensive Water Risk
Database to Aqueduct, Retrieved June 8, 2012, from WRIInsights:
http://insights.wri.org/aqueduct/2011/09/coca-cola-company-donates-extensive-water-risk-
database-aqueduct

JMP (2012) Targets and indicators for global monitoring after 2015, Retrieved June 6, 2012, from
WHO/UNICEF Joint Monitoring Programme (JMP) for Water Supply and Sanitation:
http://www.wssinfo.org/post-2015-monitoring/overview/

Kaczan, D. and Ward, J. (2011) ‘Water statistics and poverty statistics in Africa: do they correlate
at national scales?’ Water International, 36(3): 283-294.

Lehner, B., Reidy Liermann, C., Revenga, C., Vörösmarty, C., Fekete, B., Crouzet, P., Döll, P.,
Endejan, M., Frenken, K., Magome, J., Nilsson, C., Robertson, J.C., Rödel, R., Sindorf, N. and
Wisser, D. (2011) Global Reservoir and Dam (GRanD) database: Technical Documentation,
Version 1.1, Global Water Systems Project, Bonn.

Liu, J. and Yang, H. (2010) Spatially explicit assessment of global consumptive water uses in
cropland: Green and blue Water, Journal of Hydrology, 384: 187-197.

MacDonald A.M., Calow R.C., Macdonald D.M.J., Darling, W.G. and Ó Dochartaigh, B.É. (2009)
‘What impact will climate change have on rural water supplies in Africa?’ Hydrological Sciences
Journal, 54(4): 690-703. 10.1623/hysj.54.4.690
57

MacDonald, A., Bonsor, H., Calow, R., Taylor, R., Lapworth, D., Maurice, L., Tucker, J. and Ó
Dochartaigh, B.É. (2011) ‘Groundwater resilience to climate change in Africa’, British Geological
Survey Open Reports OR/11/031.

MacDonald, A., Bonsor, H., O Dochartaigh, B. and Taylor, R. (2012) ‘Quantitative maps of
groundwater resources in Africa’, Environmental Research Letters 7(2).

Margat, J., Frenken, K. and Faures, J.M. (2005) ‘Key Water Resources Statistics in Aquastat’,
Paper presented at IWG-Env, International Work Session on Water Statistics, Vienna, June 20-22
2005, FAO.

Martin-Nagle, R., Howard, E., Wiltse, A. and Duncan, D. (2012) Bonn2011 Conference, The Water,
Energy and Food Security Nexus: Solutions forthe Green Economy, 16-18 November 2011,
Conference Synopsis, Federal Ministry for the Environment, Nature Conservation and Nuclear
Safety (BMU), Federal Ministry for Economic Cooperation and Development (BMZ) and
OOSKAnews, Bonn.

Mehta, L., Ed. (2011) The Limits to Scarcity: Contesting the Politics of Allocation, Earthscan,
London.

Melamed, C. (2012) After 2015: Contexts, politics and processes for a post-2015 global agreement
on development, ODI Report, Overseas Development Institute, London.

Melamed, C., Scott, A. and Mitchell, T. (2012) ‘Separated at birth, reunited in Rio? A roadmap to
bring environment and development back together’, ODI Background Note, Overseas
Development Institute, London.

Michel-Kerjan, E., Hochrainer-Stigler, S., Kunreuther, H., Linnerooth-Bayer, J., Mechler, R., Muir-
Wood, R., Ranger, N., Vazirir, P. and Young, P. (2012) Catastrophe Risk Models for Evaluating
Disaster Risk Reduction Investments in Developing Countries, Wharton School, University of
Pennsylvania, Philadelphia, PA.

Milly, P.C.D., Betancourt, J., Falkenmark, M., Hirsch, R.M., Kundzewicz, Z.W., Lettenmaier, D.P.
and Stouffer, R.J. (2008) ‘Stationarity is dead: whither water management?’ Science, 319(5863):
573-574.

Moench, M. (2000) India's groundwater challenge, from: http://www.india-

seminar.com/2000/486/486%20moench.htm

Molden, D. (Ed.) (2007) Water for Food, Water for Life: A Comprehensive Assessment of Water
Management in Agriculture, Earthscan, London.

Molle, F. and Mollinga, P. (2003) ‘Water poverty inndicators: conceptual problems and policy
issues’, Water Policy, 5: 529-544.

Morello, L. and Climatewire (2010, June 24) ‘Defense Experts Press for Probabilistic Risk
Assessment of Climate Change’, Scientific America.

Morris, B.L., Lawrence, A.R.L., Chilton, P.J.C, Adams, B., Calow R.C. and Klinck, B.A. (2003)
‘Groundwater and its susceptibility to degradation: a global assessment of the problem and options
for management’, Early Warning and Assessment Report Series, RS. 03-3. United Nations
Environment Program, Nairobi, Kenya.

N-AMCOW (2012) North African Ministers' Council on Water, Retrieved September 5, 2012, from
MEWINA Project:
http://www.cedare.int/namcow/index.php?option=com_contentandview=featuredandItemid=487
58

National Statistics (2011) Statistics on International Development, 2006/07-2010/11, UK

Department for International Development, London.

Newborne, P. (2012) Roles of Companies in Water Management - Extending the Boundaries of

Private Sector Responsibility, Background paper for the European Report on Development
2011/12, ODI, DIE and ECDPM, Brussels.

NIC (2012) Global Water Security - Intelligence Community Assessment, Office of the Director of
National Intelligence, Washington, D.C.

Nilsson, C., Reidy, C., Dynesius, M. and Revenga, C. (2005) ‘Fragmentation and Flow Regulation
of the World's Large River Systems’, Science, 308(5720): 405-408.

ODI, DIE and ECDPM (2012) Confronting Scarcity: Managing water, energy and land for inclusive
and sustainable growth: European Report on Development 2011/12, European Union, Brussels.

Ohlsson, L. (1998) ‘Water and Social Resource Scarcity - An Issue Paper Commissioned by FAO/
AGLW’, Presented as a discussion paper for the 2nd FAO E-mail Conference on Managing Water
Scarcity, WATSCAR 2.

Oxford University Water Security Network (2012) Water Security, Risk and Society: International
Water Security Conference, 16-18 April 2012, St Hugh's College, Oxford, UK. Oxford: University of
Oxford.

Pacific Institute (2011, September) The World's Water, Retrieved March 26, 2012, from Pacific
Institute: http://www.worldwater.org/conflict.html

Parthemore, C. and Rogers, W. (2010) Sustaining Security: How Natural Resources Influence
National Securit, Centre for New American Security, Washington, D.C.

Peduzzi, P., Chatenoux, B., de Bono, H., de Bono, A., Deichmann, U., Guiliani, G., Herold, C.,
Kalsnes, B., Kluser, S., Løvholt, F., Lyon, B., Maskrey, A., Mouton, F., Nadim, F. and Smebye, H.
(2010) ‘The Global Risk Analysis for the 2009 Global Assessment Report on Disaster Risk
Reduction’, International Disaster and Risk Conference (IDRC), Davos 2010, 30 May, 3 June 2010,
on-line conference proceedings, IDRC, Davos.

Plummer, J. and Slaymaker, T. (2007) ‘Rethinking governance in water services’, ODI Working
Paper 284, Overseas Development Institute, London.

Ramana, M. (2011, April 19) ‘Beyond our imagination: Fukushima and the problem of assessing
risk’, Bulletin of the Atomic Sciences.

Ranger, N., Muir-Wood, R. and Priya, S. (2009) ‘Assessing Extreme Climate Hazards and Options
for Risk Mitigation and Adaptaiton in the Developing World’, Background Note, World Development
Report 2010: Development and Climate Change2-1-, World Bank, Washington, D.C.

Raskin, P., Gleick, P., Kirshen, P., Pontius, G. and Strzepek, K. (1997) Water Futures: Assessment
of long-range patterns and problems, Stockholm Environment Institute, Stockholm.

Rijsberman, F. R. (2006) ‘Water scarcity: Fact or fiction?’ Agricultural Water Management, 80 (1-3):
5-22.

Rockström, J., Steffen, W., Noone, K., Persson, A., Stuart Chapin, F., Lambin, E., Lenton, T.M.,
Scheffer, M., Folke, C., Schellnhuber, H.J., Nykvist, B., de Wit, C., Hughes, T., van der Leeuw, S.,
Rodhe, H., Sörlin, S., Snyder, P.K., Constanza, R., Svedin, U., Falkenmark, M., Karlberg, L.,
Corell, R.W., Fabry, V., Hansen, J., Walker, B., Liverman, D., Richardson, K., Crutzen, P. and
59

Foley, J.A. (2009) ‘Planetary Boundaries: A Safe Operating Space for Humanity’, Nature, 461: 472-
475.

Runnals, D. (2011) ‘Environment and Economy: Joined at the hip or just strange bedfellows’,
Surveys and Perspectives Integrating Environment and Society, 4(2).

SABMiller, GTZ and WWF (2010) Water Futures: Working Together for a Secure Water Future,
SABMiller plc and WWF-UK, Woking, UK.

Seckler, D., Amarasinghe, U., Molden, D., de Silva, R. and Baker, R. (1998) World Water Demand
and Supply, 1990 to 2025: Scenarios and Issues, International Water Management Institute,
Colombo.

Shah, T., Roy, A.D., Qureshi, A.S. and Wang, J. (2003) ‘Sustainaing Asia's groundwater boom: An
overview of issues and evidence’, Natural Resources Forum, 27: 130-141.

Shah, T. (2007) ‘The Groundwater Economy of Soth Asia: An Assessment of Size, Significance
and Socio-ecological Impacts’, Chapter 2 in Giordano, M. and Villholth, K.G. (eds.) The Agricultural
Groundwater Revolution: Opportunities and Threats for Development, CAB International 2007,
Wallingford, UK.

Shiklomanov, I.A. (1998) ‘World water resources: an appraisal for the 21st century’, International
Hydrological Programme Report, United Nations Educational, Scientific and Cultural Organisation
(UNESCO), Paris, France.

Smakhtin, V., Revenga, C. and Doll, P. (2004) Taking into Account Environmental Water
Requirements in Global-scale Water Resources Assessments, Comprehensive Assessment
Research Report 2, Comprehensive Assessment Secretariat, Colombo.

de Stefano, L., Duncan, J., Dinar, S., Stahl, K., Strzepek, K. and Wolf, A. (2010) ‘Mapping the
Resilience of International River Basins to Future Climate Change-Induced Water Variability’,
Water Sector Board Discussion Paper Series, No. 15, World Bank, Washington DC.

Sullivan, C. and Huntingford, C. (2009) ‘Water resources, Climate Change and Human
Vulnerability’, Proceedings of the MODSIM Conference, Cairns, Australia.

Sullivan, C., Meigh, J., Giacomello, A., Fediw, T., Lawrence, P., Samad, M., Mlote, S., Hutton, C.,
Allan, J.A., Schulze, R.E., Dlamini, D.J.M., Cosgrove, W., Delli Priscoli, J., Gleick, P., Smout, I.,
Cobbing, J., Calow, R., Hunt, C., Hussain, A., Acreman, M.C., King, J., Malomo, S., tate, E.L.,
O’Regan, D., Milner, S. and Stely, I. (2003) ‘The water poverty index: development and application
at the communicy scale’, Natural Resources Forum, 27: 189-199.

Sustainable Energy for All Initiative (2012) Objectives to Achieve Sustainable Energy for All,
Retrieved May 28, 2012, from Sustainable Energy for All:
http://www.sustainableenergyforall.org/objectives

Tarlock, D. and Wouters, P. (2009) ‘Reframing the Water Security Dialogue’, Water Law, 20: 53-
60.

Taylor, R. (2009) ‘Rethinking Water Scarcity: The Role of Storage’, EoS, 90 (28), 14 July 2009.

Tuinhof, A., Dumars, C., Foster, S., Kemper, K., Garduno, H and Nanni, M. (2002) ‘Groundwater
Resource Management: an introduction to its scope and practice’, GW-MATE Briefing Note 1,
World Bank, Washington D.C.

UN (2012) The Future we Want, United Nations, Rio de Janeiro.

60

UN DESA (2012) System of Environmental-Economic Accounting for Water, Statistics Division,

United Nations Department of Economic and Social Affairs, New York.

UN Global Compact (2011, September 2) CEO Water Mandate, Retrieved September 9, 2011,
from United Nations Global Compact:
http://www.unglobalcompact.org/Issues/Environment/CEO_Water_Mandate/

UN (n.d.) UN Special Rapporteur on the Human Rights to Water and Sanitation: Frequently Asked
Questions, Office of the United Nations High Commissioner for Human Rights.

UNAIDS (2010) Basic Terminology and Frameworks for Monitoring and Evaluation, UNAIDS,
Geneva.

UNDP (1994) Human Development Report 1994, New Dimensions of Human Security, United
Nations Development Programme, New York, NY.

UNDP (2006) Human Development Report 2006, Beyond Scarcity: Power Poverty and the Global
Water Crisis, United Nations Development Programme, New York, NY.

UNEP (2009) Water Security and Ecosystem Services: The Critical Connection, United Nations
Environment Programme, Nairobi.

UNEP (2011) Towards a Green Economy: Pathways to Sustainable Development, United Nations
Environment Programme, Nairobi.

UNEP (2012) 2012 UN-Water Status Report on the Application of Integrated Approaches to Water
Resource Management, United Nations Environment Programme, Nairobi.

UNEP (n.d.) Global Environment Monitoring System – Welcome, Retrieved June 9, 2012, from
GEMSTAT: http://www.gemstat.org/default.aspx

UNESCO (2012a) Relative water stress index, Retrieved June 8, 2012, from Indicators - level of
stress on the resource, World Water Assessment Programme (WWAP):
http://www.unesco.org/new/en/natural-sciences/environment/water/wwap/indicators/wwdr-
indicators/level-of-stress-on-the-resource/

UNESCO (2012b) Coefficient of variation for climate moisture index, Retrieved June 8, 2012b,
from Indicators - Level of Stress on the Resource, World Water Assessment Program (WWAP):
http://www.unesco.org/new/en/natural-sciences/environment/water/wwap/indicators/wwdr-
indicators/level-of-stress-on-the-resource/

UNESCO (2012c) Climate Vulnerability Index, Retrieved June 12, 2012, from Indicators - Risk
Assessment, World Water Assessment Programme (WWAP):
http://www.unesco.org/new/en/natural-sciences/environment/water/wwap/indicators/wwdr-
indicators/risk-assessment/

UNESCO (2012d) Dissolved nitrogen (nitrates and nitrogen dioxide), Retrieved June 9, 2012, from
Indicators – Ecosystems, Word Water Assessment Programme (WWAP):
http://www.unesco.org/new/en/natural-sciences/environment/water/wwap/indicators/wwdr-
indicators/ecosystems/

UNESCO (2012e) Fragmentation and flow regulation of rivers (Dam Density), Retrieved June 9,
2012, from Indicators – Ecosystems, World Water Assessment Programme (WWAP):
http://www.unesco.org/new/en/natural-sciences/environment/water/wwap/indicators/wwdr-
indicators/ecosystems/
61

UNESCO (2012f) Indicators, Retrieved June 6, 2012, from World Water Assesssment Programme
(WWAP): http://www.unesco.org/new/en/natural-sciences/environment/water/wwap/indicators/

UNESCO (2012g) Organic pollution emissions (biochemical oxygen demand by the industrial
sector), Retrieved June 13, 2012, from Indicators - Industry and Energy, World Water Assessment
Programme (WWAP): http://www.unesco.org/new/en/natural-
sciences/environment/water/wwap/indicators/wwdr-indicators/industry-and-energy/

UNESCO (2012h) Water Reuse Index, Retrieved June 14, 2012, from Indicators - Level of Stress
on the Resource. World Water Assessment Programme (WWAP):
http://www.unesco.org/new/en/natural-sciences/environment/water/wwap/indicators/wwdr-
indicators/level-of-stress-on-the-resource/

UNICEF (2003, April 30) WASH and Women, Retrieved 7 26, 2012, from UNICEF: Water,
Sanitation and Hygiene: http://www.unicef.org/wash/index_womenandgirls.html

UNICEF (2010, August 2) Water Supply, Retrieved March 17, 2011, from Water, Sanitation and
Hygiene: http://www.unicef.org/wash/index_watersecurity.html

UNISDR (2011a) Global Assessment Report on Disaster Risk Reduction, United Nations
International Strategy for Disaster Risk Reduction, Geneva.

UNISDR (2011b) GAR 2011, Retrieved June 13, 2012, from Global Risk Data Platform:
http://www.preventionweb.net/english/hyogo/gar/2011/en/what/rdp.html

UNSD (2007) System of Environmental-Economic Accounting for Water, United Nations Statistics
Division.

UN-Water and GWP (2008) UN-Water and Global Water Partnership (GWP) Roadmapping for
Advancing Integrated Water Resources Management (IWRM) Processes, UN-Water and Global
Water Partnership.

UN-Water (2008) Status Report on Integrated Water Resources Management and Water Efficiency
Plans for CSD16, UN-Water.

UN-Water (2012) Key Water Indicator Portal, Retrieved June 13, 2012, from UN-Water:
http://www.unwater.org/statistics_KWIP.html

UN-Water EG-IMD (2009) Final Report o the Expert Group on Indicators, Monitoring, and Data
Bases (EG-IMD), including proposed programme to mobilise information to support decision
making about water resources, United Nations World Water Assessment Programme, UN-Water,
UNESCO, Perugia, Italy.

UN-Water TF-IMR (2009) Monitoring progress in the water sector: A selected set of indicators,
Final Report, UN-Water Task Force on Indicators, Monitoring and Reporting, UNESCO, Perugia,
Italy.

US Department of State (2011, March 22) Remarks on World Water Day - Hillary Rodham Clinton,
Secretary of State, Retrieved March 26, 2012, from US Department of State:
http://www.state.gov/secretary/rm/2011/03/158833.htm

USNRC (2007) Probabilistic Risk Assessment – Factsheet, United States Nuclear Regulatory
Commission.

Vörösmarty, C., Douglas, E., Green, P. and Revenga, C. (2005) ‘Geosptatial Indicators of
Emerging Water Stress: An Application to Africa’, Ambio, A Journal of the Human Environment, 34
(3): 230-236.
62

Vörösmarty, C., Fekete, B., Meybeck, M. and Lammers, R. (2000) ‘A simulated topological network
representing the global system of rivers at 30-minute spatial resolution (STN-30)’, Global
Biogeochemical Cycles, 14: 599-621.

Vörösmarty, C., McIntyre, P., Gessner, M., Dudgeon, D., Prusevich, A., Green, P. G., Glidden, S.,
Bunn, S.E., Sullivan, C.A., Liermann, C.R. and Davies, P.M. (2010) ‘Global threats to human water
security and river biodiversity’, Nature, 467: 555-561.

de Waal, D., Hirn, M. and Mason, N. (2011) Pathways to Progress: Transitioning to Country-Led
Service Delivery Pathways to Meet Africa's Water Supply and Sanitation targets, Water and
Sanitation Program – Africa, Nairobi.

Water Footprint Network (2012) WaterStat, Retrieved June 9, 2012, from Water Footprint:
http://www.waterfootprint.org/?page=files/WaterStat

WaterAid (2012) Water Security Framework, WaterAid, London.

WBCSD (2011) Global Water Tool, Retrieved March 7, 2012, from wbcsd - business solutions for a
sustainable world: http://www.wbcsd.org/work-program/sector-projects/water/global-water-
tool.aspx

WDPA (n.d.) Statistics, Retrieved June 9, 2012, from World Database On Protected Areas:
http://www.wdpa.org/Statistics.aspx

WEF (2011) Water Security: The Water-Food-Energy-Climate Nexus, World Economic Forum
Water Initiative, Island Press, Washington D.C.

WEF (2012) Global Risks 2012 - Seventh Edition, World Economic Forum, Geneva.

WHO and UNICEF (2011) Report of the First Consultation on Post-2015 Monitoring of Drinking-
Water and Sanitation, Berlin, 3-5 May 2011, WHO and UNICEF.

Woodhouse, P. (2012) ‘Foreign agricultural land acquisition and the visibility of water resource
impacts in Sub-Saharan Africa’, Water Alternatives, 5(2): 208-222.

World Bank (2012a) ‘Natural Capital Accounting: Helping Make Better Decisions for Sustainable
Development’, World Bank Group Brief, World Bank, Washington D.C.

World Bank (2012b) Water productivity, total (constant 2000 US$ GDP per cubic meter of total
freshwater withdrawal), Retrieved June 9, 2012, from Data:
http://data.worldbank.org/indicator/ER.GDP.FWTL.M3.KD

World Commission on Dams (2000) Dams and Development: A New Framework for Decision
Making, The Report of the World Commission on Dams. Earthscan, London.

Wouters, P. (2010) Water Security - Global, regional and local challenges, Institute for Public
Policy Research (IPPR), London.

WRI (2012) Aqueduct Water Risk Atlas - Basin Maps, Summary Methodology Document, World
Resources Insittute, Washington D.C.

WRI (n.d.) Aqueduct: Measuring and Mapping Water Risk, Retrieved March 6, 2012, from WRI
Insights: http://insights.wri.org/aqueduct/welcome
63

WWAP and UNSD (2011) Monitoring Framework for Water: The System of Environmental-
Economic Accounts for Water (SEEA-Water) and the International Recommendations for Water
Statistics, UNESCO-World Water Assessment Programme and UNSD.

WWAP (2012) The United nations World Water Development Report 4: Managing Water under
Uncertainty and Risk, World Water Assessment Programme, UNESCO, Paris.

WWF (2012) Living Planet Report 2012: Biodiversity, biocapacity and better choices, Word Wildlife
Fund.

WWF and DEG (2011) Assessing Water Risk. World Wildlife Fund, Germany and Deutsche
Investitions- und Entwicklungsgesellschaft mbH, Berlin.

Yoffe, S., Wolf, A. and Giordano, M. (2001) ‘Conflict and Cooperation over International
Freshwater Resources: Indicators and Findings of the Basins at Risk Project’, in S. Yoffe (ed.)
Basins at risk: Conflict and Cooperation over International Freshwater Resources (pp. 63-120),
Oregon State University, Corvalis, OR.

ZSL and WWF (2012) The Living Planet Index - an indicator of the state of global biodiversity,
World Wildlife Fund and Zoololgical Society of London, London.
64
65
66
Overseas Development Institute
203 Blackfriars Road
London SE1 8NJ
UK
Tel: +44 (0)20 7922 0300
Fax: +44 (0)20 7922 0399
Email: [email protected]
Website: www.odi.org.uk
ISBN 978 1 907288 95 1
Working Paper (Print) ISSN 1759 2909
ODI Working Papers (Online) ISSN 1759 2917