Interest in water and water use is growing. Popular magazines have fortold the depletion of the Ogallala Aquifer. Entrepreneurs and venture capitalist-types are looking to water use for business and investment opportunities. How can we measure the impact of new water-efficient processes and technologies — how can we find out if new ideas will work? We can’t…unless, of course, we measure how water is currently used how usage patterns will change in respond to technological or social changes.
Lucikly, researchers are on the job! Waterfootprint.org is one group developing metrics to measure human water use. I found out about them when a friend told me that a single sheet of A4 paper requires 10 liters of water to produce. Ten liters!…for a single sheet! I was incredulous. So I checked out their site, and there it was. “[One sheet of] A4 has a green water footprint of 10 litres.”
But what does it mean that a sheet of paper has a “water footprint” of ten liters? The sad part is that after spending several hours perusing this group’s web site, I suspect the answer is “not much”. I see several problems with the “water footprint” methodology, as I currently understand it.
1. Lack of additivity.
From the site’s homepage: The water footprint of an individual, community or business is defined as the total volume of freshwater that is used to produce the goods and services consumed by the individual or community or produced by the business.
As I understand it, if I opened an apple store, sold one apple to my first customer, and closed up shop, my store would have a water footprint of 70 L. But the apple buyer would apparently also has a water footprint 70 L, using the site’s definition. Maybe I am just naive, but I would think that to figure the total water footprint of a nation, we could just add water footprints for all the producers and consumers comprising that nation. Water footprints apparently do not work that way. They are not additive; if we added them, the total water consumption stemming (ha!) from apple related commerce would appear to be 140 L, two times too high.
2. Attribution error and/or failure to consider substitution effects
If you dig in deeper to the methods behind calculating “virtual water footprints” or vWFs, it turns out that the total evapotranspiration of water through crops and trees counts as water use. For example the figure for A4 sheets of paper begins with a “production” forest with an evapotranspiration of 600 mm/yr (equivalent to 6000 m3/ha/yr). I can’t wrap my head around this. First of all, if fresh water is a valuable resource, and trees require fresh water for evapotranspiration to live, it seems inevitable that getting rid of trees by cutting them down would save water. If we cut down a hectare of forest, that’s 6000 m3 of fresh water “saved” a year! But to the contrary, the water footprint methodology would have us believe that cutting down trees uses water. Another way to look at the problem is to suppose that the paper company folds up shop, and lets its “production” forest revert back to natural growth. Does the evapotranspiration of the forest change? It shouldn’t, at least not by much. Yet our water footprint has plummeted.
I have trouble understanding how these characteristics of the “water footprint” are features which render the metric valuable as a policy-making tool. Instead, I think these characteristics are bugs. I would personally favor a metric based more on perturbations to the pre-industrial water cycle, for example, on the differences in evapotranspiration between a production forest and an old-growth forest, or between a cornfield and natural grasslands. But what do the defenders of water footprint analysis say? These are just my initial thoughts on the approach of water footprint analysis. I’m sure there is a lot more to say on the subject.