Simply stated, water scarcity occurs when demand for freshwater exceeds supply in a
specified domain.
Water scarcity = an excess of water demand over available supply
This condition arises as consequence of a high rate of aggregate demand from all
water-using sectors compared with available supply, under the prevailing institutional
arrangements and infrastructural conditions. It is manifested by partial or no
satisfaction of expressed demand, economic competition for water quantity or quality,
disputes between users, irreversible depletion of groundwater, and negative impacts on
the environment.
Water scarcity is both a relative and dynamic concept, and can occur at any level of
supply or demand, but it is also a social construct: its causes are all related to human
interference with the water cycle. It varies over time as a result of natural hydrological
variability, but varies even more so as a function of prevailing economic policy,
planning and management approaches. Scarcity can be expected to intensify with most
forms of economic development, but, if correctly identified, many of its causes can be
predicted, avoided or mitigated.
The three main dimensions that characterize water scarcity are: a physical lack of water
availability to satisfy demand; the level of infrastructure development that controls
storage, distribution and access; and the institutional capacity to provide the necessary
water services.
DRIvIng foRCEs bEhInD WAtER sCARCIty
AnD thE RoLE of AgRICuLtuRE
Unconstrained water use has grown at global level to a rate more than twice the rate of
population increase in the 20th century, to the point where reliable water services can
no longer be delivered in many regions. Demographic pressures, the rate of economic
development, urbanization and pollution are all putting unprecedented pressure on a
renewable but finite resource, particularly in semi-arid and arid regions.
Of all economic sectors, agriculture is the sector where water scarcity has the
greatest relevance. Currently, agriculture accounts for 70 percent of global freshwater
withdrawals, and more than 90 percent of its consumptive use. Under the joint pressure
of population growth and changes in dietary habits, food consumption is increasing
in most regions of the world. It is expected that by 2050 an additional billion tonne
of cereals and 200 million tonnes of meat will need to be produced annually to satisfy
growing food demand.
But to what extent is this steady growth in water demand ‘negotiable’? There is a
general agreement that water to satisfy basic needs is not – human health requires a
minimum level of access to good quality water. Similarly, with the right to food being
increasingly recognized, since water as a critical factor in food production, a minimum
quantum for subsistence production could be considered non-negotiable. However,

domestic water withdrawal represents globally only about 10 percent of all water uses,
but has a very low consumption rate – most domestic use is returned to the environment
with little evaporative loss even if quality is degraded. By contrast, agricultural use has
direct downstream (or down-gradient) consequences since the production of biomass
requires huge quantities of water to be transpired. If the water is sourced for irrigation
and transpired, this represents a local hydrological loss that reduces availability in the
downstream domain. The purpose of this report is to assess the options and scope for
adjustment in agricultural water use as a response to water scarcity.
MEAsuRIng WAtER sCARCIty: thE hyDRoLogICAL CyCLE
A correct understanding of water scarcity hinges on an understanding of the laws of
physics that govern hydrological processes, and the means to allocate and measure use.
1. Water is a renewable resource, but patterns vary in space and time.
2. Water exists in a continuous state flux in all its phases (solid, liquid, gas) that
is driven by energy gradients applying to the physical processes of evaporation,
transpiration, condensation, precipitation, infiltration, runoff, subsurface flow,
freezing and melting. It is these flows and fluxes, rather than stocks, that should be
the focus of planning and management.
3. A water balance is governed by conservation of mass, and the rate of water
entering a specified domain is equal to the rate of water leaving the same domain
with any differences resulting in changes in storage. The linkages between surface
water, groundwater, soil moisture content and the process of evapotranspiration are
of critical importance, and still inadequately reflected in many water management
plans.
4. All land areas in a river basin are interlinked through water. Therefore actions
in one part of a hydrological system will have impacts on other parts of the
system, and for most intents and purposes water is best managed on the basis of
hydrographic units.
5. As water use intensifies, the diluting and cleaning functions of aquatic
ecosystems are stretched to their limit, resulting in accumulation of pollutants.
6. Any desire to maintain a set of aquatic ecosystem goods and services implies a
limitation in the availability of water for human use in a given domain.
7. Water accounting, i.e. the systematic organization and presentation of information
relating to the physical volumes and quality of flows (from source to sink) of water
in the environment as well as the economic aspects of water supply and use, should
therefore be the starting point of any strategy for coping with water scarcity. Water
accounting involves a comprehensive view of the water resources and supply
systems and how they are related to societal demands and actual use.
8. Water audits go one step further, and place water supply and demand in the
broader context of governance, institutions, finance, accessibility and uncertainty
These are all elements needed to design effective water scarcity coping strategies.
PoLICy AnD MAnAgEMEnt oPtIons
Options to cope with water scarcity can be divided between supply enhancement and
demand management. Supply enhancement includes increased access to conventional
water resources, re-use of drainage water and wastewater, inter-basin transfers,
desalination, and pollution control. Demand management is defined as a set of actions
controlling water demand, either by raising the overall economic efficiency of its
use as a natural resource, or by operating intra- and intersectoral re-allocation of 

water resources. Options to cope with water scarcity in agriculture can be seen as a
continuum from the source of water to the end user (the farmer), and beyond, to the
consumer of agricultural goods. These options are discussed below. However, it should
be stressed that at the level of agricultural water demand commonly observed in food
producing countries, supply enhancement and demand management measures are
often linked through the hydrological cycle.
suPPLy EnhAnCEMEnt
During the twentieth century, large multipurpose dams have served the needs of
agriculture, energy and growing cities, and helped protect populations from flood
hazards. While potential for further dam development still exists in some regions,
most of the suitable dam sites are already in use, and the development of new dams is
increasingly questioned in terms of economic, social and environmental considerations.
On-farm water conservation, particularly the adoption of agricultural practices that
reduce runoff, to increase the infiltration and storage of water in the soil in rainfed
agriculture, is the most relevant local supply enhancement option that farmers have to
increase production. On a slightly larger scale, small, decentralized water harvesting
and storage systems contribute to increasing water availability and agricultural
production at the household and community levels. However, large programmes of
small-scale water harvesting, like the watershed management programmes developed
in Andhra Pradesh and other parts of India, have shown significant impacts on the
catchment’s hydrology and downstream water availability.
Groundwater exploitation has grown exponentially in scale and intensity over recent
decades. Groundwater’s capacity to provide flexible, on-demand water in support
of irrigation has been seen as a major advantage by farmers. While intensification of
groundwater use has contributed to improved livelihoods of millions of rural people,
it has also resulted in long-term aquifer depletion, groundwater pollution and saline
intrusion into important coastal aquifers.
The adoption of re-cycling of drainage water and wastewater use in agriculture tends
to be positively correlated with water scarcity. Re-use of drainage water is a reality in
most large irrigation schemes, in particular in the large rice-based systems of Asia. Of
lesser global significance, but locally important, is the re-use of urban wastewater (it
is estimated that world-wide some 20 million hectares of agricultural land is irrigated
with wastewater). Efforts are needed to better assess re-use and its potential, and
promote safe recycling of wastewater in agriculture, in particular in water-scarce areas.
DEMAnD MAnAgEMEnt In AgRICuLtuRE
In broad terms, agriculture has three options for managing overall water demand
within the water domain:
¾ reduce water losses;
¾ increase water productivity; and
¾ water re-allocation.
The first most commonly perceived option is that of increasing the efficiency of
water use by reducing water losses in the process of production. Technically, ‘water
use efficiency’ is a dimensionless ratio that can be calculated at any scale, from
irrigation system to the point of consumption in the field. It is generally applied to 

any management that reduces the non-beneficial use of water (i.e. reducing leakage
or evaporative losses in water conveyance and application). The second option is
increasing crop productivity with respect to water. This involves producing more crop
or value per volume of water applied. The third option is to re-allocate water toward
higher value uses through intersectoral transfers (transfers to municipal supply, for
instance) or intrasectoral transfers by limiting the irrigated harvested area under a
particular crop to reduce evapotranspiration or diverting water towards higher value
crops.
Clearly there is scope for managing the demand for water in agriculture in time and
in space. But excessive emphasis is often placed on the first option, with efforts aimed
at reducing water ‘losses’ within irrigation distribution systems. Two factors limit
the scope for and impact of water loss reduction. First, only part of the water ‘lost’,
while withdrawn for beneficial use (defined as water that is diverted for purposes that
have clear and tangible benefits, such as for household purposes, irrigation, industrial
processing and cooling), can be recovered effectively at a reasonable cost. Second, part
of the water ‘lost’ between the source and final user returns to the hydrologic system,
either through percolation into the aquifers or as return flow into the river systems. The
share of water lost through non-beneficial consumption, either through evaporation or
through drainage into low quality water bodies or to the sea, varies according to local
conditions. A clear understanding of the real potential for reducing water losses is
needed to avoid designing costly and ineffective demand management strategies.
In most cases, the single most important avenue for managing water demand in
agriculture is through increasing agricultural productivity with respect to water.
Increase in crop yields (production per unit of land) is the most important source of crop
water productivity increase. Yield increases are made possible through a combination
of improved water control, improved land management and agronomic practices.
This includes the choice of genetic material, and improved soil fertility management
and plant protection. It is important to note that plant breeding and biotechnology
can help by increasing the harvestable parts of the biomass, reducing biomass losses
through increased resistance to pests and diseases, reducing soil evaporation through
vigorous early growth for fast ground cover, and reduced susceptibility to drought.
Therefore managing overall demand through a focus on water productivity rather
than concentrating on the technical efficiency of water use alone is an important
consideration.
If productivity is considered in terms of added value and not production, re-allocating
supply from lower value to higher value crops is an obvious choice for farmers
seeking to improve income levels. For this to happen, changes are required in both the
management and technology associated with irrigation to provide farmers with a much
higher level of control of water supply. In addition, shifts to higher value crops also
require access to inputs, including seeds, fertilizers and credit, as well as technology
and know-how, and reasonable conditions to operate in much more competitive
market conditions. However, in practice, very few farmers are able to make this choice
since the market for higher value crops is limited compared with the market for staples.
Beyond productivity concerns, agricultural water demand can simply be limited or
capped. This is a commonly applied measure where the volume of evapotranspiration
used in the production of a unit of agricultural output is limited by reducing the area
under irrigation.
Understanding the roles, attitudes and strategies of various stakeholders, including
relevant institutions, is a key aspect of demand management strategies. Ultimately, 

it is at the farmer level that most water will be consumed. Their behaviour and their
capacity to adapt will be driven by a carefully selected set of incentives that include
both structural and institutional changes, improved reliability and increased flexibility
of water supply. Farmers’ strategies will be driven by water saving only when water
availability becomes their main limiting factor. Policies based on systems of water
tariffs aiming to reduce agricultural water demand have proved successful in some
cases, but require very constraining conditions and are often difficult to enforce.
Approaches based on water quotas and water use (or withdrawal) rights have, in most
cases, a higher probability of success.
ACtIons bEyonD thE WAtER DoMAIn
The agricultural response to water scarcity lies, at least partially, outside of the water
domain. To this extent it is possible to recognize other measures that can help manage
water demand:
¾ reduction of losses in the post-harvest value chain;
¾ reduction in demand for irrigated production through substitution by imports
of rainfed staples; and
¾ reduction of per capita agricultural water demand.
Reduction of losses in the post-harvest value chain
Beyond agricultural production, substantial savings of water can also be obtained by
addressing the issues of waste in the food chain, diets, and the role of agricultural trade.
Losses and wastages occur all along the food chain, and have been estimated at up to
50 percent of production in developed countries. While part of these losses may be
irretrievable, it makes sense to carefully identify the major sources of losses and assess
the scope for their reduction.
Reduction of demand for irrigated production through substitution
Options include enhanced production in rainfed agriculture, and imports of food
product through international trade.
There are several reasons to consider investing in rainfed agriculture as part of a water
scarcity coping strategy, but the opportunities vary greatly from one place to another.
In places where climate is conducive to rainfed agriculture, there is high potential
to improve productivity where yields are still low, as is the case in many regions of
sub-Saharan Africa. Here, a combination of good agricultural practices, upward and
downward linkages (access to finance, inputs and markets), and weather insurance
schemes can improve agricultural productivity with little impact on water resources.
The issue of trade is particularly relevant in countries where water scarcity limits the
capacity of agriculture to satisfy all the needs for other agricultural commodities. The
concept of ‘virtual water’ was developed in the 1990s to indicate that in a reasonably
safe and interdependent world, gains in water productivity can be achieved by growing
crops in places where climate enables high water productivity at lower cost and trading
them to places with lower water productivity. Although rarely expressed in water
terms, virtual water trade is already a reality for many water-scarce countries, and is
expected to increase in the future.
Reduction of per capita water demand
Finally, increasing consumption of meat and, to a lesser extent, also dairy products
translates into increased water consumption, as their production requires large volumes 

of water. The extent to which societies are willing to modify their diets as part of a
larger effort to reduce their environmental footprint reaches far beyond water scarcity
concerns. Yet, it has implications in terms of national food security and associated
water-scarcity coping strategies.
AssEssIng AnD CoMbInIng fooD suPPLy oPtIons
thRough A Cost CuRvE APPRoACh
In order to guide decision-makers’ choices among the range of available options, these
options need to be assessed in terms of their effectiveness, cost, and technical, social
and environmental feasibility. The political dimension of their choice will also be
carefully scrutinized.
The “food supply cost curve” can help to provide insight in the way a country can
bridge its food supply gaps in a cost-effective way. The curve ranks food supply
options in terms of their cost and provides an easy way of assessing cost-effectiveness
in the achievement of food supply objectives. When used at national level, each country
will have its own curve, based on current level of intensification, availability of land
and water, and level of losses in the food chain. The cost curve provides a simple but
powerful method for identifying and ranking options for food production in conditions
of water scarcity. Much of the complexity lies in the establishment of the individual
cost curves for the different options, which requires a good understanding of the
agronomic, hydrological and socio-economic conditions under which improvements
will take place.
PRInCIPLEs foR ACtIon
The selection of the right range of options will depend on local conditions, and it is
unlikely that a single set of options can be designated as the ‘optimal’ solution. Nor
is a particular option to be seen as desirable in all contexts. The choice of ‘no action’
is not an option under scarcity; it would translate into environmental degradation,
sub-optimal use of scarce resources, inequity in access to these resources, and overall
negative impacts on the economy and societal well-being. Therefore, rather than
attempting to prescribe solutions to water scarcity, it is suggested that policy options
and related strategies should be based on a set of generic principles that are valid across
socio-economic settings. Six basic principles have been developed, and are presented
below.
Knowledge: base strategies on a clear understanding
of the causes and effects of water scarcity
Strategies should be based on the best available evidence, and not on hearsay or
intuition, and detailed accounting of water supply and demand should be carried
out from the onset. The inter-relationship between surface water and groundwater,
between upstream and downstream catchments, between quality and volumes, and
the importance of water recycling within river basins all have implications in terms
of effectiveness of proposed actions. Well intentioned but ill-informed strategies for
coping with water scarcity can have significant perverse impacts on the way water is
distributed within the river basin, without achieving expected savings.
Impact: assess the full range of benefits and costs
and use systematic and comprehensive decision criteria
It might seem obvious that cost-effectiveness should be considered along with equity 

and collective values when choosing between options. However, past experience shows
that cost-benefit analyses have often overlooked or under-estimated the potential
negative impact of water development interventions on people or the environment,
while overestimating other benefits. In particular, supply enhancement options have
often been selected beyond any reasonable analysis, leading to an over-equipped
subsector and ‘artificial’ or ‘constructed’ water scarcity. Calculating cost-effectiveness
needs to encompass several dimensions. It varies with time, as a result of change in
knowledge of social and environmental processes and values, as well as relative changes
in added value of different water use sectors. Only a careful analysis of the costeffectiveness of each option allows for better identifying the most promising sources
of gains in water demand management.
Realistic financing mechanisms are required for water initiatives to meet the full costs
of water scarcity interventions and programmes. In many cases, this involves putting
less emphasis on capital costs of construction and engineering and more emphasis on
capacity building, stakeholder-based planning, operation and maintenance, and other
long-term institutional support costs.
Capacity: ensure that the right level of water
governance and institutional capacity is in place
Disputes between users increase with water scarcity, as does the likelihood of negative
impacts on vulnerable social groups and on the environment. As demand management
takes increasing importance, much stronger institutions are needed to guarantee
equitable distribution of benefits and maintenance of environmental services. Better
definition of roles and responsibilities, empowering of local institutions, review of
policies, adaptation of laws, and the use of incentive mechanisms become increasingly
important as water scarcity progressively builds up. Efforts for a new water
management culture are needed, including public awareness campaigns, educational
programmes, capacity building and training at all levels, including water users groups.
Institutions also need to adapt to approaches where public, private and other operators
can carry out management tasks jointly.
Context-specificity: adapt response to local conditions
The response of a country to water scarcity depends on a number of conditions,
including local agro-climatic conditions, levels of water scarcity, the role agriculture
plays in national economies, and societal values. It will also depend on external
factors, including the global trade and cooperation environment, and the prospects for
climate change. Further, in view of the rapid changes in the geo-political, societal and
environmental fields, what could be considered well adapted today may no longer be
so tomorrow, and strategies must be expected to change.
Coherence: ensure policy alignment between water, agriculture and food security
Decisions outside the water domain, such as those determining energy prices, trade
agreements, agricultural subsidies and poverty reduction strategies, can all have a major
impact on water supply and demand, and therefore on water scarcity. Alignment of
the many policies, legislation and fiscal measures that influence water management,
service delivery and level of demand is crucial. Agriculture and food security policies
are strongly connected to water policies and that degree of connection needs to be
appreciated to ensure overall coherence.
Preparedness: anticipate change through robust
decision-making and adaptive management
Planning and management systems need to be flexible, adaptive and based on continuous 

social and institutional learning. Adaptive management recognizes the high level of
uncertainty associated with future situations, and places emphasis on flexible planning
that allows regular upgrading of plans and activities. Such a level of responsiveness
is only possible if information and knowledge are updated, and if monitoring and
information management systems continually provide decision-makers with reliable
information. There is always the risk that coping strategies will be derailed by external
factors, such as climate change, global financial and economic shocks, and shifting
international cooperation agreements. Scenario building, as an integral part of strategy
development, is one means of identifying and mitigating these risks, and developing
robust responses to uncertainty of future situations.