The supply of ecosystem services depends on the capacity of ecosystems and the potential of human economies to demand those services. Assessing this relationship is, however, quite challenging because we typically lack information on both capacity and demand. Our research addresses this gap by integrating an existing high-resolution map of lakes and wetlands with a global land cover database that creates a comprehensive assessment of wetland extent across the globe and at national spatial scales (Dickinson et al 2014 ) and combines this information with an assessment of freshwater demand through a statistical model that estimates the human population’s capacity to access freshwater via downstream supply (see methods).
We created an index of freshwater provision across the globe by combining the annual freshwater runoff estimated from our global wetland database with the demand estimates computed from the human population density map. We then combined the resulting map with a map of biodiversity hotspots to identify locations of greatest freshwater provision importance and mapped the ecosystem service flow pathways connecting high-ranked biodiversity hotspots to the most important areas for freshwater provision. The resulting maps highlight areas with high potential for safeguarding freshwater provision at critical scales. These results are important because, while most spatial analyses to date have focused on capacity or supply alone, our study demonstrates the importance of including the human demand dimension of the relationship.
To identify areas of supply that could feed downstream people, we created a national map of demand by combining a map of population density with an assessment of the population’s access to freshwater from downstream supply. Specifically, we estimated the relative downstream demand for each watershed by combining a water-demand weighting index, which is a composite metric of the non-metropolitan population density, industrial and agricultural land uses, and non-rural urban settlement types, with a hydrologic connectivity factor that corrects for the fact that the population downstream from a supply area is likely to be greater than the demand for water from that area. 3d9ccd7d82