In the next 40 y, the number of urban dwellers in the developing world is forecast to grow by nearly 3 billion (1). While this urban demographic transformation is unfolding, climate change is expected to affect the global hydrologic cycle. Anthropogenic emissions of greenhouse gases will likely raise average global temperatures, with temperature changes expected to be greater near the poles than the equator. Climate change will also likely alter precipitation patterns, with some areas becoming wetter and others becoming drier (2). For some regions, climate and demographic trends will present a fundamental challenge: how will water be provided on a sustainable basis for all those new urbanites?
Freshwater provision to urban residents has three components: water availability (is there enough water nearby?), water quality (how much treatment is needed before it is clean enough to use?), and delivery (are systems in place to bring water to users?) (3). This article examines only the water availability component, recognizing that for many cities challenges of water quality and delivery are paramount. Throughout this article, “water availability” and “water shortage” refer solely to the amount of water physically available, not accounting for issues of water quality and delivery. In a sense, our estimates of water shortage are conservative: we assume cities can use all nearby water and map where problems of water shortage are likely to remain.
This article models how population growth and climate change will affect water availability for all cities in developing countries with >100,000 people. These cities had 1.2 billion residents in 2000, 60% of the urban population of developing countries and, according to our demographic projections, will account for 74% of all urban growth globally from 2005 to 2050 (1). We describe the magnitude and general patterns of the global challenge of water availability for urban residents, recognizing that such a global approach cannot account for each city's particular circumstances.
We used data on population distribution (30 arc-second resolution) from the Global Rural–Urban Mapping Project (GRUMP) (4), as well as demographic forecasts for our study cities (5). Two demographic scenarios are explored: the “Basic Demographic” scenario, which predicts a city's population growth according to its size and national-level urban fertility and mortality trends; and the “Ecological Factors” scenario, which in addition allows cities in specific biomes (e.g., arid regions) to have different rates of population growth, all else being equal. Hydrologic data (6-min resolution) on monthly sustainable surface and groundwater flows are taken from the Water Balance Model (6–10). Four scenarios of climate and land use change, driven by consistent scenarios of global economic development, are based on the Millennium Ecosystem Assessment (11) scenarios as implemented by Fekete et al. (12): Adaptive Management; Global Orchestration; Order from Strength; and Technogarden.
We first calculate whether a city has sufficient sustainable surface or groundwater within the urban extent delineated by GRUMP. The GRUMP urban extents used to spatially delineate a city are relatively large and include many suburban and exurban locations surrounding a given city center. Water shortage is defined as 100 L per person per day, a rough measure of the amount an urban resident needs to live comfortably long-term, including water for drinking, bathing, cleaning, and sanitation including flush toilets (13–16). If an urban area does not meet this minimum standard, nearby areas are evaluated using a series of buffers (Fig. S1), out to 100-km distance. The underlying assumption is that cities with a water shortage within the 100-km buffer will have to obtain water by other means, such as long-distance transport, extracting groundwater faster than aquifer recharge, or desalinization.