Decoupling water use from growth: the New Mexico example

Pulling together some New Mexico water use numbers today for one of my University of New Mexico colleagues, I was reminded of a cool paper from a few years back by Peter H. Gleick and Meena Palaniappan of the Pacific Institute that contained this striking graph:

Water use and GDP

Water use and GDP

It’s two times series – U.S. gross domestic product and total water use – plotted side by side. As Gleick and Palaniappan note, the two rise “in lockstep” through most of history – as the nation grows, both in population and economic activity, so does its water use. But in the 1970s, the curves decouple. Our national economy has continued to grow, but our water use has not. Increasing population and economic activity no longer requires more water.

In the research I’m doing for my book, I see curves that look like this all the time. I’m more frequently looking at population growth rather than total economic activity, and I’m often slicing up the data to look at groundwater versus surface water withdrawals, municipal versus agricultural use, and ag water use compared to ag productivity. But looking at this in lots of different ways, I almost invariably find some sort of decoupling.

New Mexico water

Here’s today’s decoupling – municipal water use in New Mexico:


New Mexico municipal water use

This particular USGS dataset is “public supply”, which is essentially all the state’s major municipal water agencies and captures 85 percent of the state’s population. Municipal use peaked in 1995. In Albuquerque, the state’s largest metro area (and site of my own backyard rain barrels), per capita use looks like it’ll be about half this year of what it was back then. If you add in agriculture (which in New Mexico uses ten times as much water as municipalities), New Mexico water use peaked in 1980. Both groundwater pumping and surface water diversions have been declining ever since, even as our state’s population and economy has grown. In inflation-adjusted terms, New Mexico’s ag sector was about the same size in 2010 that it was in 1980 (data from BEA), but it’s using a million acre feet per year less water. That’s a big part of how “decoupling” works.

Finding these points of decoupling and looking at the hydrologic and policy drivers is one of my new hobbies (where by “hobby” I mean “book research”), because they point to examples of what solutions to our problems might look like. If I can generalize, it’s the relatively straightforward notion that when people have less water, they get clever about using less water to get stuff done. Municipalities facing scarcity get the conservation bug, and farmers are just plain smart about adapting when they have less water to work with.

Arizona water

Arizona is a particularly interesting example. Despite a reputation for groundwater management problems, its groundwater use peaked in 1975, according to the USGS, and is now barely more than half of what it was then. A big part of that is the substitution of surface water from the Colorado River, but groundwater use has dropped more than replacement use of surface water has risen. Arizona’s 1980 Groundwater Management Act has its problems, but by this important measure (and others, especially rising aquifers in key parts of the state) it seems to be working. An annual reduction of  more than 2 million acre feet per year in groundwater pumping seems like kind of a big deal. Phoenix, to cite one example, used to get nearly all of its water from groundwater. Now it gets almost none, switching from mining ancient and non-renewable groundwater to renewable surface water supplies.

Arizona groundwater


According to the USGS data, Arizona’s overall “peak water” moment (ground + surface water) came back in 1980. Arizona’s population has more than doubled since then, and it’s using 25 percent less water, even with those crazy “misters” people install at outdoor restaurants and backyard patios to try to make Phoenix summers bearable.

This “when people have less water, they use less water” thing is one of the lines of evidence that makes me optimistic about our ability to solve our region’s water problems if we can identify the characteristics that make it work and harness them in the cases where we’re still having problems.

Note on sources: The data for my graphs (and also the one in the Gleick/Palaniappan paper) is the USGS water use report series.


  1. I’ll have to read Gleick and Palaniappan, because I am wondering whether the inclusion of “virtual water” would result in such a stark decoupling. Globalized capitalism = stealing other peoples’ water and not accounting for it.

  2. Additionally, the quick nationwide decoupling seems strange. It might indicate a large nationwide change (unlikely for us to expect from policy), rapid technology diffusion (more likely), or something with international policy/economics (most likely?). I wasn’t alive in the 70s, so I can’t say, but I would expect to see some trade agreement(s) right around then that can be attributed to the decoupling.

  3. i think it’s a short term trend. we’ve shifted jobs, pollution and waste to other countries, when those countries run out of cheap resources to exploit then we’ll see the trend start to reflect reality again (you don’t get something for nothing).

    i.e. the living system which supports us needs a certain amount of water to keep functioning. we’ve broken large parts of it by shifting water into reservoirs and draining wetlands.

    at one time we had a suitable experiment going which could help us quantify how many gallons per day a system would need that could support a single person (Biosphere II), but that wasn’t run long enough (due to short-sighted funding and organizational problems).

    it may take an attempt at a colony on the moon or Mars to finally get some harder numbers, but i hope that’s not how it actually happens.

  4. John,
    If we make another graph and build it from the bottom up, what do we see as minimum water usage. I want that graph to be minimal per person usage from any source times the number of people. Then i want a ‘recycle’ graph. In this graph, I want to know how much of current water usage goes back into the ground, reservoirs, or aquifers from sewage plants, better storm drains, etc. If you feel ambitious, I would also like some graph of the amount of water lost to other states through evaporation and winds blowing east.
    Basically, I want a mass balance on water so that I can make some sense out of the decoupling. Is that possible?

  5. Interesting question/data, but need to disaggregate data to get closer to an explanation, Fourier analysis kind of thing. How much of decline due to urban conservation programs like in ABQ and SF in NM? Does GDP growth reflect something about shift to services or global trade and do those represent lower water footprint? Can’t tell and too lazy on Sunday morning to look.

  6. The ’70s takes us into the Nixon era, loss of the gold standard, the eventual to formal recognition of China, and the first signs of a “globalized” economy. One has to have a few grey hairs to remember the first really, really, cheap Chinese crap that flooded our markets. There has been a significant rise in the quality factor lately. Perhaps because China’s water use is up, along with pollution.

  7. Surprising hockey-stick chart. We reached US consumer “peak vehicle-mile” around 2006 and “peak new house square footage” around 2007, and neither of those were the result of importing commute miles or residential home space from overseas. People will just trot out their own hobbyhorse explanation for this pretty significant shift until someone does an actual detailed analysis. Why doesn’t John use his university contacts to get a bright PhD student to do a dissertation on this? It’s too big for a Masters thesis and could be the research platform for a decent academic career for someone.

  8. Pingback: The Economist gives Las Vegas points for water management - jfleck at inkstain

  9. Pingback: Stuff I wrote elsewhere: The Great Decoupling of the West's Water - jfleck at inkstain

Comments are closed.