A Latin lesson on urban water resilience by design

San Miguel de Allende, Mexico © ferrantraite / iStock

The high levels of urbanisation seen in Latin America illustrate the growing need for ‘resilience by design’ to address the impacts of climate change. Diego Rodriguez explains an approach developed by the World Bank in collaboration with the University of Massachusetts and other partners.


Latin America is the most urbanised of all developing regions. In 2017, the population of Latin America and the Caribbean region reached 644 million, 80% of which lived in urban areas. According to the UN DESA 2018 Revision of World Urbanization Prospects, by 2030 the total population in the region will be 718 million, with an urban concentration of 84%.

This high rate of urbanisation, coupled with the rapid growth of cities in the region, has created a host of water-related challenges. These include degraded water quality and inadequate water and sanitation infrastructure, particularly in expanding peri-urban and informal settlements. As cities continue to grow rapidly, and climate change impacts the availability and distribution of water resources, meeting the water demands of cities will become increasingly difficult and energy intensive.

Combined, these problems present a challenge for policy-makers and municipalities of how to provide services to their citizens, ensure that there are enough resources – such as food, water and energy – and protect public health, all while protecting the environment.

Careful planning promotes long-term water security and resilience to climatic and non-climatic uncertainties, and can mitigate some of the increasing challenges. But the traditional planning (‘predict then act’) and investment design approaches are not adequate when looking to address the increasing challenges, including climate change risks and public health.

From ‘predict then act’ to resilience by design

In the traditional ‘predict then act’ approach, decision-makers attempt to predict the future and select an array of interventions and investments that will produce the desired outcome for the system of the future scenario selected.

In a traditional planning exercise, decision-makers will identify a discrete number of potential future scenarios (usually three) and select one of those scenarios, while identifying a portfolio of solutions. In most cases, the portfolio of selected investments is chosen by applying criteria such as cost-effectiveness, cost-minimisation, cost-benefit ratio, or maximisation of the net present value.

The ‘predict then act’ approach can be appropriate if conditions are stable, the future can be predicted easily, and if there is consensus among decision-makers and water users about what the future may hold and what interventions must be prioritised to achieve a desired outcome. But the future is highly uncertain and cannot be predicted, and a failed attempt to do so will lead to the financing of a portfolio of inefficient investments, the financing of unnecessary projects leading to stranded assets, a high opportunity cost of financial resources, and a failure to meet the goals defined for the water system. Furthermore, ‘predict then act’ also does not explicitly assesses social equity or environmental considerations, or the impacts of climate change.

This has fuelled a paradigm shift to flexible and robust decision-making approaches. Analytical approaches have been developed to look at the intrinsic characteristics of a water system or project and describe them in terms of exposure, sensitivity, and their adaptive capacity to withstand stress.

These approaches are commonly referred to as ‘bottom-up’, as they first assess the characteristics of the existing system and its sensitivity to a wider range of climate inputs and identify the parts of the system that are most vulnerable and the specific attributes of those vulnerabilities. These approaches look to examine the response, or sensitivity, of a system to a series of climate projections or other variables that stakeholders consider may affect the expected outcomes of their plan or project.

Thus, bottom-up approaches permit analysts to include potential sources of risk, such as change in population dynamics or economic conditions. In these approaches, the level of complexity, as well as the resources needed to inform decisions, is scaled (adjusted) according to the unique characteristics of a project and other relevant factors, as well as the issues and decisions that may arise in the process that are generated by stakeholders.

Several methodologies of the bottom-up approach exist in the current scientific literature, and the World Bank has developed its approach, Confronting Climate Uncertainty in Water Resources Planning and Project Design: The Decision Tree Framework (DTF), (see Ray and Brown, 2015; see Figure 1 for a summary of the approach, and the text box for an example of an application to water supply in Mexico City).

These approaches allow diagnosis of the potential impacts of climate and other variables in a water project, plan or strategy. They allow adaptation options to be generated to reduce the effect of relevant stressors in a system. Often, these approaches permit water planners to assess the chances of success, or failure, of a specific application under current or plausible future conditions.

From here, water managers can propose modifications, allowing testing of the response of the modification to the stressors. At this stage, decision-makers and stakeholders would be able to formulate and design more risk-cost-effective, resilient and robust solutions for the application.

When resilience is a critical variable to measure

Freshwater resilience is a critical element of urban resilience. Mainstreaming resilience in the planning and prioritisation of investments is a necessary condition to mitigate the vulnerabilities to stress and shocks that urban water systems face. Resilience offers the conceptual framework to prepare for the future.

A multitude of definitions of resilience suggest that resilience refers to the ability of a system to remain in its current equilibrium, persisting under and adapting to change, as well as its ability to shift or transform to a new equilibrium, thriving in a changed configuration. Under changing climate and water futures, this approach to resilience enables a design to sustain system function and services for the long term, under whatever range of stressors may be projected.

Three capabilities characterise a resilient system: i) Persistence – its ability to maintain coherent function in response to disruption and changing conditions without altering its identity; ii) Adaptability – its ability to maintain coherent function by modifying its identity to accommodate change; and iii) Transformability – its ability to change identity and to establish a new, stable function when pushed beyond tipping points that preclude maintaining its prior state. Planning for resilience is broken into a series of questions: resilience of what, to what, for whom, and what can be done? Answering these questions requires an understanding of the system dynamics and of the values of the managers and stakeholders themselves.

A plea for a transformation

In the Latin America region and elsewhere, planning and investment prioritisation in urban water systems needs to change. Business as usual will not be sufficient to address the increasing challenges of urbanisation, population growth and climate change. Freshwater resilience is a necessary condition to ensure resilience in cities. This requires moving from a ‘predict and act’ planning approach to the use of resilience by design frameworks.

Figure 1. The Decision Tree Framework. © Patrick A Ray and Casey M Brown (2015).

Water utilities are at the centre of this transformation and, as shown in the Mexico case, the current applications have demonstrated that resilience by design approaches allow the identification of investments that build resilience and robustness to urban water systems, thereby contributing to building resilient cities.


More information

Water is a master variable: solving for resilience in the modern era. Frederick N Boltz et al. Water Security 9 (2020) 10048.

Resilience by design: a deep uncertainty approach for water systems in a changing world. Casey Brown et al. Water Security 9 (2020) 100051: www.sciencedirect.com/science/article/pii/S2468312419300070

Resilience by design in Mexico City: a participatory human-hydrologic systems approach. Sarah St. George Freeman, et al. Water Security 9 (2020) 100053: www.sciencedirect.com/science/article/pii/S2468312418300245

Beyond downscaling: a bottom-up approach to climate adaptation for water resources management. LE García et al. 2014. openknowledge.worldbank.org/handle/10986/21066

Confronting climate uncertainty in water resources planning and project design: The Decision Tree Framework. The World Bank. Patrick A Ray and Casey M Brown. 2015. openknowledge.worldbank.org/handle/10986/22544


Resilient urban water supply in Mexico City

The Cutzamala Water System (CWS), completed between the mid-1970s and the mid-1990s, carries approximately 10-15m3/s from the Balsas basin to the Valley of Mexico and Mexico City, a distance of 126 kilometres and an elevation of 1200 metres. The project was capital intensive and brought a large burden of annual energy consumption for pumping. As development continues and demand for water in the Balsas basin grows, so too does the challenge to urban-rural political cooperation.

Investment evaluation for the Cutzamala Water System © HRGUMass

An example use of the World Bank’s Decision Tree Framework included the following phases:

Phase 1: Establishment of trust was critical (between stakeholders and users). This required meetings, video calls and document exchanges, and resulted in the co-creation of an improved understanding of the physics of the system, its operation, entitlements, and precedents for emergency response.

Phase 2: As the CWS simulation model was developed and tested, the potential to increase the reliable yield through reservoir reoperation was discovered. New rule curves for reservoirs El Bosque, Valle de Bravo and Villa Victoria were derived by maximising reliable yield of the existing CWS without additional investments.

Phase 3: Very small changes in precipitation and temperature were found to cause the CWS to perform unsatisfactorily. There are very few projected climate scenarios in 2050 in which the CWS will be able to maintain acceptable delivery of water to Mexico City. This suggests the need to evaluate options to improve the system’s performance.

Phase 4: The performance of all investment combinations relative to the current performance of the CWS was evaluated. The size of the bubble in the figure indicates robustness (satisfaction of target yield across a range of climate scenarios) and the colour represents the resilience. Relative to current operations (black dot in lower left), large gains are seen in optimised operations alone. Large capital investments do not by default equal large performance payoffs. Additionally, while some investments may exhibit similar yield and costs, they may differ in their ability to improve recovery after failure and robustness to future climate conditions.

Next steps: This analysis is an entry point into a comprehensive application of the DTF to strategic regional planning for the purpose of integrated urban water resources management for Mexico City and its contributing water supply regions.

Credit: Including Climate Uncertainty in Water Resources Planning and Project Design-Decision Tree Initiative. Pilot studies of the Cutzamala Water System, Mexico – Draft Final Report. World Bank, Washington DC. September 2017.


Diego J Rodriguez is Senior Water Resources Management Specialist, Water Global Practice, The World Bank