Agriculture is intrinsically dependent on water and increasingly exposed to growing water-related risks. These include water shortages, excess water and inadequate water quality, alongside systemic risks of degraded freshwater ecosystems and a destabilised water cycle. Trade can be a channel for impacts from water risks but also a mechanism to alleviate water risks. Countries may be exposed to water risks originating beyond their borders via their imports of agricultural and food commodities. At the same time, the ability to access goods on international markets reduces vulnerability to food insecurity and can relieve pressure on local water demand. Effective policy responses require a holistic understanding of water risks that accounts for both blue water resources, such as rivers, lakes and aquifers, and green water resources, including soil moisture and land-sourced rainfall that underpin crop production, while understanding the role of cross-border linkages.
Countries’ ability to anticipate and monitor water risks is a critical component of agricultural resilience and food security. Water risk management contributes to the capacity of agricultural systems to plan and prepare for adverse events, absorb shocks, recover from disruptions and adapt over time. While evidence points to an intensification of water risks for agriculture, many countries lack the tools and methods needed to assess these risks comprehensively across time horizons, spatial scales and risk types. Existing approaches often focus on individual hazards rather than systemic risks and their impacts on agricultural outcomes.
Drawing on recent literature and an expert workshop, the paper examines water risks for agriculture; outlines a typology of tools to support public authorities in anticipating monitoring and assessing these risks; and discusses when and how to integrate risk assessment tools into decision-making processes.
Given diverse water risks and decision contexts, the agricultural sector requires a suite of tools tailored to different risks, spatial scales and across short-, medium- and long-term timeframes. The relevance of specific tools depends on the nature of decisions being taken, with the highest value achieved when tools inform choices with high costs.
Advances in technology are also expanding the possibilities for improved anticipation and monitoring. Satellite-based remote sensing, in situ sensors, the Internet-of-Things, modelling approaches, and the application of Artificial Intelligence and machine learning are increasingly underpinning tools that are used to inform water-related risk assessment. For example, a digital twin of a river basin can combine data streams from smart sensors and satellites with models to monitor water status in near real-time, explore scenarios and test interventions to see the simulated consequences.
Despite this progress, important gaps and challenges persist:
Moving from hazard monitoring to meaningful assessments of impacts on agriculture remains difficult. While data on hydrological hazards are increasingly available, translating this into risk intelligence requires integration with agronomic and socio-economic data, such as crop types, planted areas, yields and exposure of farmers and communities. These datasets are often fragmented, outdated, insufficiently granular or incompatible across systems. Nevertheless, there are promising examples of tools that move towards outcome-based approaches, such as agricultural drought indicators that consider impact on crop yields, pasture growth and farm business profits.
There is a mismatch between the scale of available data and the scale at which decisions are made. Farmers often require field-level and seasonal information; public authorities typically need risk assessment information at basin, regional or national levels. Temporally, seasonal data is often key yet missing in water risk assessment for agriculture. For example, tools may rely on outdated crop maps that fail to reflect actual planting patterns, limiting their usefulness for seasonal decision making. Moreover, water risks frequently compound and cascade over time, yet most assessment tools analyse hazards in isolation rather than capturing interacting and cumulative effects.
Several critical dimensions of water risk remain difficult to measure and monitor and are, therefore, insufficiently covered by existing tools. Risks linked to freshwater ecosystem health, water quality, groundwater depletion and the recycling of moisture through terrestrial ecosystems are not yet fully integrated into commonly used assessment frameworks.
For water risk intelligence to deliver value, assessment tools must be effectively embedded in decision-making processes. Tools are most useful when applied at critical decision points where the costs of poor choices are high, such as prior to major infrastructure investments or policy reforms. At the same time, tools can be costly to develop and operate. Decision makers therefore need to ensure that investments in water risk assessment deliver a proportionate return and are matched to the decision context. In some cases, simple approaches may be sufficient; in others, data limitations and uncertainty mean that complex tools risk providing misleading signals unless uncertainty is clearly communicated. Attention is often placed on improving tools rather than understanding how decisions are made. A more effective approach may be to start from decision-making needs and work backwards to the information required. Furthermore, water risk information is only one input into decision making. While assessment tools can generate relevant evidence, policy choices are also influenced by the legal and institutional context as well as the cultural and social values attached to water. Consequently, policymakers may have valid reasons not to fully defer to the outputs of risk assessment tools.
There is no single acceptable level of water risk, highlighting the need for flexible, context-specific approaches that consider different risk thresholds rather than one-size-fits-all solutions. Governments play a central role in creating the enabling conditions for effective water risk management. This includes maintaining authoritative hydrological, meteorological and agronomic datasets, establishing clear governance and co-ordination structures, and encouraging knowledge and technology transfer. In doing so, governments can engage farmers as key partners in risk monitoring, as they are often the first to observe the impacts of water risks. Innovation ecosystems and market openness in terms of trade, private capital and competition can also be drivers of technology deployment. Finally, public authorities are also well-positioned to provide a long-term and holistic perspective on water risks. This helps counterbalance short-term incentives and support decisions that build systemic agricultural resilience rather than reactive responses to immediate water-related shocks.