How much water do we use, and for what? Water “permeates” almost every aspect of our lives (no pun intended!). Some uses of water are obvious – for example, municipal and domestic supply used for drinking, cleaning, flushing and watering. Others are less obvious, such as water used for irrigation to grow produce, grains, or feed. The water needed to raise livestock is one step further removed, since the water “used” to produce the product includes the water that must go into growing feed. Yet other uses of water are even less visible, for example for refining fuels, cooling for thermo-electric power generation, and the manufacturing of almost everything in our day-to-day lives.;
Because the types and scales of water use vary widely – from domestic wells that pump at a few gallons per minute, to allocations of major rivers in billions of gallons, the units of measurement used for water management also span an enormous range (see Units).
Globally, there is a widely varied usage of water, as a result of differing total populations and population densities, geography and climate (i.e. water availability), cultures, economies, lifestyles, and water use and reuse efficiency. This can be described both in terms of total water abstraction from surface water and groundwater sources and as per capita water withdrawal. It can also be divided to consider the end uses (for example, as percentages of the total use), or to consider the source of the water. Each of these facets of water use illuminates different aspects of the “water story”.
In many industrialized nations, the dominant water uses are for industry (including thermoelectric power generation, manufacturing, etc…) and agriculture (Figures 10-11). In contrast, domestic and municipal water use generally constitutes less than 15-30% of the total. In developing nations, this is somewhat different – total water use is smaller, less is used for industry, and the proportion used for domestic water supply is larger.
In the U.S., the average per capita use of domestic or municipal water (i.e. the most direct uses – those that would be measured by the water meter at your home) is about 215 m3 per person per year, equivalent to 156 gallons per day (as of 2002). For comparison, the total abstraction of water from surface and groundwater sources in the U.S. is about 1700 m3/person/yr, or 1230 gallons per day. The difference in these numbers represents the large proportion of water that goes to so-called “indirect” uses: food production, manufacturing, power generation, and mining, among others.
In contrast, in sub-Saharan Africa, total water use is less than 200 m3 per person per year (less than 12% of water use in the U.S.). Total abstractions in Western Europe are about 600 m3 per person per year, about 850 m3 per person per year in the Middle East; and 1150 m3 per person per year in Australia. Among those nations with the highest water use, agriculture accounts for anywhere from <40% of use (U.S.), to 67-81% (India and China), to as much as 96% (Pakistan). Industrial use (including power generation) ranges from over 80% of total water use to less 1%. In the U.S. water use for power generation is near 50%; in China, it constitutes 25% and in India about 5%. Germany, Russia, Canada, France, and much of Western Europe use around 60% of withdrawn water for power generation. Municipal and domestic water use typically constitutes about 10-20% of the total and varies little among the worlds most populous countries (Figure 9). You can explore these patterns on your own via a useful interactive plotting engine at Gapminder. [1]
It is important to note that because many products are imported or exported across state and national borders, the total abstractions of water in a given place do not necessarily map to the distribution of water “consumption” there. Consider tomatoes that are transported from California to Massachusetts. The water withdrawal from rivers and aquifers needed to grow the tomatoes would appear on California’s “water tab”, but the eventual use of that water would be elsewhere. The same goes for agricultural and industrial products exported internationally. This flow of indirectly used water, embedded in products, is termed virtual water, and is defined as the amount of water used in generation of the product, or alternatively, the amount of water that would be needed to generate the product at the site where it is ultimately used. It is “virtual” because the water use is indirect; it is required to make or grow the item but is not actually physically contained in the item or transported with it.
Another important aspect of water use is the degree to which the water is available for recycling and/or reuse (Figure 12; cf. Figure 2). For some water uses, including industrial or domestic applications, the wastewater is captured, treated, and may be reused. These are termed nonconsumptive uses. For example, water used in homes is, for the most part, recaptured for treatment and discharged to surface water or groundwater systems – or for recycling of supply. In this sense, the water is not removed from the system (i.e. not “consumed”). In other applications, the water is effectively removed from the Earth’s surface environment and is not available to be re-captured. These are consumptive uses. Examples include water used for agriculture, which is mostly transpired by plants or evaporated and thus transferred to the atmosphere, or thermoelectric power generation, in which much of the water also evaporates (think of the steam you may have seen rising from power plants – this is consumptive water use, in action!).
1. Describe the difference between consumptive and non-consumptive water use. Provide an example of each.
ANSWER: Consumptive use means that the water cannot be recovered, usually because it is lost to evaporation or transpiration, or to deep aquifers. Examples include irrigation, lawn watering, and some fraction of the water used for fracking or cooling in thermoelectric power generation. Non-consumptive use implies that the water may be recovered and treated for reuse either by the same users or by downstream users. Examples include many industrial uses and domestic use.
1. Based on Figures 10-13, what are the two largest uses of water in the U.S.?
ANSWER: Irrigation (agriculture) and thermoelectric power generation.
2. Have the dominant uses of water in the US changed much in the past 50 years? If so, how?
ANSWER: Yes, they have. Prior to around 1965, irrigation was the largest use of water in the US. From then to the present, thermo-electric power generation has overtaken it, although the amount of water used for both applications has grown.
3. Note three regions or countries where the dominant water use is for agriculture (look at Figure 12). Note three where it is for industry. Is this what you would have expected?
ANSWER: Agriculture: mainly in equatorial regions of Africa, Asia, Indonesia, and S. America (dark green in the top panel of Figure 12). Industry: mainly in N. America, Europe, and Russia (pink and red areas in the center panel).
4. How much water do you think you use per day for household or domestic activities (e.g., washing dishes, laundry, showering, cooking, drinking)?
ANSWER: Most people underestimate their use and guess 20-50 gallons per day.
The uses of water for human activity vary immensely, and as a result, water resource management covers a wide range of temporal and spatial scales. In some cases, the timescales are short and volumes relatively small (i.e. domestic pumping of several gallons per minute, over timescales of minutes or hours). At the other extreme, water allocations for states or municipalities are often considered in the context of average annual flows in the billions of gallons. Because so many different scales of measurement are used to describe water flux or discharge (volumes of water) and flow rates (the velocity of flow), it is important to have some facility with the various units of measurement and get a sense for their relative magnitudes.
As one example, the total fluxes of water through river systems – commonly used to define allocations of water for states or nations - are measured and reported in acre-feet. This is a unit of water volume equal to the amount of water that covers an area of one acre, one foot deep. One acre-foot is equivalent to 325,851 gallons (see summary of unit conversions from the U.S. Geological Survey [3]), and is often considered as the amount of water needed for a family of four for about one year.
As we’ll discuss in Module 3, over shorter timescales, river discharges are reported in units of cubic feet per second (cfs), cubic meters per second (m3/s), or gallons per minute (gpm). As one example, on average, Spring Creek carries about 50 cfs at Houserville, PA; this increases downstream to about 90-100 cfs at Axemann as the creek is fed by springs and small tributaries. Short-lived peak discharge may exceed 500 cfs after storm events. For comparison, the flow of the Mississippi River at St. Louis, MO is typically about 400,000-600,000 cfs; in major floods the discharge is over 1,000,000 cfs. The flow rates of rivers and groundwater, as we will see in Modules 3-4 and 6, are reported as a velocity - units of length per time. These measures represent the velocity of the water itself, or of an object (stick, boat, person, etc…) carried by the river or stream.
Yet other key quantities in hydrology are reported in units of an equivalent depth (or length) per time. For example, rainfall rates are described in units of inches, cm, or mm per hour (for individual storm events) or per year (i.e. annual average precipitation). Evaporation rates are reported in the same way – but of course, represent water transport in the opposite direction (up!). The total volume of water these represent depends on the area over which they occur.
It is also instructive to look in more detail at the distribution of different water uses. For example, in the U.S., industry is concentrated East of the Mississippi, mainly in the “steel belt” (also known as the “rust belt”) and in Texas and Louisiana (primarily related to oil and gas extraction) – and thus water use for industry is as well (Figure 14). It’s worth considering whether this pattern is ultimately rooted in the timing of settlement and westward expansion in the U.S., availability of fuel (i.e. coal), or availability of water sources and rivers as a means of transportation for goods and raw materials. The pattern of water withdrawal for agriculture in the US is even more dramatic (Figure 15). Large agricultural water withdrawals from surface water and groundwater are dominantly West of the Mississippi. This is evident from a state-by-state map view and shown even more clearly when plotted simply from West to East (Figure 15, bottom panel).
The source of the water we use also provides clues about where water may be most readily available, and/or where typical rainfall and snowmelt cannot meet demand. Inspect the maps below (Figure 16). Surface water withdrawals are spread more or less uniformly across the U.S., and reflect overall water use reasonably closely. This is influenced in large part by total population, energy production, and industrial and agricultural activity (i.e. CA, TX, NY, and FL are the most populous states). However, groundwater withdrawals (obtained by pumping at wells) are a good indication that surface water flows alone are not sufficient to meet demand.