The Water Footprint Of Hydroelectric Dam Construction

Freshwater availability is limited. During many periods of the year and in many places around the world, the demand for freshwater for human use exceeds what is sustainably available, while growing populations, increasing wealth, and climate change all aggravate the already dire current status. Historically, we turned to constructing dams in order to regulate freshwater availability. Building and operating dams and reservoirs brought humanity much good, to this present day. But it also gave rise to associated costs, both in (socio)economic and ecological terms. It is therefore increasingly acknowledged that reservoirs are not mere in-stream water users, but rather large water consumers.

Why is that? Because of the water that evaporates from their surfaces, and which is, therefore, “lost’” from the system for other uses. Since this consumptive term – or footprint – adds to the pressure on (regional) water resources, it should be accounted for. But no one so far has done this accounting. How much water are we talking here? How much freshwater is consumed annually from reservoir surfaces globally? Who or what should we assign this water loss to? And is this water footprint really worsening water scarcity?

To answer these questions, we first estimated the water footprint of the world’s artificial reservoirs. The total water footprint of a reservoir is the sum of all water-using activities in its value chain, in our case, both the construction of the dam itself and its operation (i.e. ongoing reservoir evaporation). We found that water used in dam construction contributes less than 1% to the total global water footprint of reservoirs of 66,000,000,000 m3 a year. This number – after prudent extrapolation to account for reservoirs that we missed in this study – adds roughly 25% to the total water footprint of humanity, making reservoir footprints a considerable blind spot missed by earlier studies.

Moreover and paradoxically, a reservoir may increase water availability during parts of the year, but only at the cost of reducing availability over the whole year. Clearly, reservoirs are not just constructed to waste water, but to serve one or more specific beneficial purposes. It is, therefore, only fair to say these purposes share in the burden of the water that is consumed by the reservoir. We attributed the reservoir water footprint to the specific purposes the reservoir serves, based on the economic value that is generated by these purposes.

After all, water taken from a reservoir for irrigation leads to crop yields that can be sold in the market, and each kWh generated by hydropower can also be sold against the local price for electricity. We thus set out to calculate, for each reservoir globally, how much money is being generated for the purposes hydroelectricity generation, irrigation water supply, residential and industrial water supply, flood protection, fishing, and recreation and assigned each reservoir water footprint to the purposes accordingly.

This exercise taught us that economic gains from over two thousand major reservoirs included in this study amount to $265 billion each year. The lion’s share of this amount comes from residential and industrial water supply and hydroelectricity generation. Zooming in on hydropower as an example, we found that each evaporated m3 of water generates $2.26 (the revenue). The water footprint of hydropower (the cost), on the other hand, is 14.6 m3 GJ-1, meaning it requires 14.600 liters to generate one GJ of electricity through hydropower (global average).

Compare that to the much lower water footprints of both fossil-based electricity and solar or wind, and you see that hydroelectricity is a highly water-intensive form of electricity. Moreover, although contributing to a much needed reduction in carbon emissions, transitioning from fossil-based electricity to renewable hydropower may in the process create or worsen water issues instead.

Considering water consumption of electricity sources is therefore paramount in drafting energy policies that are sustainable beyond just carbon. Fortunately, our assessment revealed that the majority of reservoir water footprints (57%) is located in basins that are not water stressed (yet). A mere 1% is located in basins that already face year-round water scarcity even at present. Furthermore and finally, we found that the purpose a reservoir is predominantly used for changes when water scarcity worsens: in water-abundant regions, hydroelectricity generation is the most common purpose, while in scarcer basins they typically serve to supply residential, industrial and irrigation water.

The figure shows the total national reservoir water footprint and how it is shared among the different purposes the reservoirs serve, for twenty-five selected countries. The bracketed number indicates how many reservoirs in each country we analyzed.

The blue water footprint of the world’s artificial reservoirs for hydroelectricity, irrigation, residential and industrial water supply, flood protection, fishing and recreation, recently published in the journal Advances in Water Resources. This work was conducted by Rick J. Hogeboom and Arjen Y. Hoekstra from the University of Twente and Luuk Knook from Rijkswaterstaat.