23 Mar 2015

The problem is water consumption, not use

Chris Perry, a water expert whose ABCDE framework of water management I admire, thinks that it's important for everyone to know the difference between "consumptive use" and "non consumptive use" when discussing water management in times of scarcity. I disagree with him in the short run but agree with him in the long run.

First, let's define what we're talking about. The impact of our water use somewhat depends on what happens to the water afterwards. On one extreme, water is not very "used" in a swimming pool, running river, or hydroelectric project. Many people can share the same pool water, look at or fish in a river, or benefit from water flowing below a dam. These kinds of non-consumptive uses do not really deplete the quantity of water currently available in an area.[1] On the other extreme, you have irrigation with sprinklers that results in high evaporation and leaves little water behind. That kind of "consumptive use" makes it much harder to stretch a lot of water across many uses and users.

The point that Chris wants to emphasize (and I agree with this) is that it is sometimes more important to manage consumptive use more than use per se. It's in this context that people in Singapore, Orange Country, California and other places with recycled water facilities should not worry about their showers or flushed toilets. Their water use is non-consumptive in the sense that the water can be captured, cleaned and used again. The same is true (to a degree) for farmers in Palo Verde Irrigation District who use flood irrigation. They apply about 9 feet of water per acre over the year and about 4 feet run off, back into the Colorado River. Assuming their soils are well-drained, their technique does not "waste" water because the excess runs back into the river, instead of evaporating.

The trouble is that most urban utilities and irrigation districts do NOT handle their water in a non-consumptive way. On the urban level, you have lawn irrigation that is consumptive as well as a tradition (based on cost, yuck and "abundance") of discharging wastewater into rivers or oceans. On the agricultural level, you have a tradition of "use it or lose it" water rights that are generous enough to deter investments in improving water efficiency -- especially when improvements that "yield" water cannot be rewarded by selling that water.[2]

So, why or how do these differences matter? Well, they are very important when considering the definitions of water rights.[3] They also matter for water managers facing supply shortfalls. Orange Country and Singapore count on recycled water to meet demand; Las Vegas has been intensively recycling water for re-use, but the problem there is consumptive use.

Which brings me to the question of whether these definitions matter. Yes, they do in the long run because -- as Chris rightly points out -- you need to understand and manage all water flows. In the short run, however the differences do not matter.

In my post last week, I suggested short run and long run actions that California can take to reduce risk of water shortages. In the short run, I said to reduce urban irrigation and agricultural groundwater pumping (as well as raise prices, to maintain fiscal health), but I did not discuss consumptive vs non-consumptive use. Why? Because that difference is not important in the short run. Technology and infrastructure are fixed, and return flows and recycling rates are known. I was targeting "wasteful" uses on lawns (rather than drinking water in restaurants) and unsustainable (unknown) groundwater pumping because both reductions would leave water for other uses now and in the future. In my long-run actions, I got into market, regulation and water quality improvements where "consumptive differences" matter more.

Bottom Line: Water can be managed for efficiency, equity or physicality, for minimal energy demand, maximal environmental health, or targets of justice. Decide your community's priority and then manage in that direction, but don't forget to monitor impacts on your "less important" goals.

  1. I'm ignoring the fact that water in pools and behind dams evaporates, as well as the fact that dams disrupt flows and water temperatures.
  2. There some exceptions to this. San Diego paid Imperial Irrigation District $millions for water "saved" after a canal was lined (reducing seepage). The trouble with that deal (and others like it), is that some of the "lost" water was non-consumptive, i.e., it was recharging aquifers that other, Mexican farmers were using. They lost a lawsuit claiming damages from the lining because they lacked standing (=not American).
  3. Australia's rights are evolving as farmers find ways to use less water; the idea is to extract those savings from their licenses, rather than let them sell the savings, because their "savings" end up reducing the quantity necessary to keep other licenses "wet.


JBD said...

Another reason consumptive use deserves more scrutiny than use per se is that since alomost all consumptive use is evaporative (or evapotranspirative) it can pollute the water even without adding a pollutant. All water that has touched land has some dissolved solids, measured as TDS. If that water is efficiently consumed, say 80% efficient irrigation (or cooling towers), which means 4 out of 5 parts is evaporated, all the TDS remains but is concentrated by a factor of 5. For example, good 300 mg/l TDS water becomes polluted 1,500 mg/l TDS water, and there is less of it. Of course expensive desal can fix that, but the costs for that aren't typically tagged to the pollutor, since they didn't "add" any pollutants (at least thats how current policies play out).

Chris Perry said...


Thanks for bringing up the consumptive vs non-consumptive issue. I felt it was a topic that really needed to be addressed in the Econtalk piece and elsewhere in your work and that of others.

You contend that the distinction relates to short run versus long run priorities. I am not sure how you reach that conclusion, but the following example might provide food for thought. A "pass through" thermal power cooling system diverts vast quantities of water, and returns almost all to the environment. A downstream town, irrigation system, or wetland would be virtually unaffected if the station shut down (or doubled in size). If the power station converted to a recycling system, it would divert far less water, but ALL the water diverted would be evaporated. Abstraction would be much less, consumption much more, and the impact on the downstream environment in consequence higher.

Similarly even a small irrigation system will always affect downstream availability because the purpose of irrigation is to convert liquid water into water vapour. In neither case do I see a distinction between long and short run issues: increasing water availability for one purpose typically requires reducing CONSUMPTION somewhere else.

You also state that "urban utilities do not handle water in a non-consumptive way". Again, I am not clear what this phrase actually means. Urban water suppliers are well aware that their users are primarily non-consumptive. That is why (in the UK at least) utilities assume that 95% of the water delivered to a household is returned as waste water, and the utilities charge three times as much to treat the effluent as they do to deliver the original supply. Furthermore, urban utilities are concerned about leakage from pipes because they have pumped, treated and delivered the water at considerable cost, and prefer to sell it rather than let it recharge an aquifer: their concern is financial, not about sustainable resource use.

In every "water" scenario, there are exceptions and special cases. The only universal truth in my experience is that we must first understand the accounts -- where the water comes from and where it goes to -- before presenting priorities for either long- or short-run interventions. Consumptive vs non-consumptive classifications are central to this analysis.


David Zetland said...

@Chris -- (1) I agree. (2) We are talking about different aspects of systems. In your SR/LR scenario, the accounting can be changed, but not in a hurry (e.g., building a better treatment plant to increase recycling). That's why California cannot "change consumption" to fix its shortage issues in the long run. (3) Accounting is of course important for overall efficiency. (4) urban water return depends on cities (as you say) and it's MUCH lower b/c of outdoor irrigation. Thus, consumptive use is more relevant and worth acting on. (5) It's a pity that so few managers or policy makers "understand accounts" for water, money, OR value. I, like you, am doing my best :)

Anonymous said...

We can agree that we are both doing our best...

On item 2, (a) I think you mean short run, not long run, and (b) California could release water very quickly by buying up a few thousand ha of farm land and stopping irrigation on that area. It is not easy -- but also, it is not a mystery, as many would have us believe (the Pacific Institute once published a report including a section titled "the myth of evapotranspiration"... that really is mysterious). Reduce CONSUMPTION and you have more water for other uses.

You suggest that accounting is "important". Here I disagree: it is absolutely fundamental. If your bank balance is declining every month you don't just randomly start parking for less minutes, or sewing up holes in your pockets, or buying a smaller TV -- you do some serious analysis of where the money is going, and which interventions matter!

It is also not as evident to me as it apparently is to you that letting the parks dry up and turn brown is better than reducing irrigation... many people get a lot of pleasure from urban irrigation, so do the account (A), then have the debate (Bargain) with the stakeholders, then set the rules (Codify), decide who is responsible (Delegate) and build some stuff (Engineer) as necessary.

Enough probably...

David Zetland said...

@Chris -- I'm sure we agree in broad principles, but we keep approaching with different assumptions. Without trying to extend this exchange, I'll clarify that I mean -- by SR/LR -- that it's hard to move water among sectors, thus less important to clarify accounting compared to acting. urban users cannot easily, right now, get ag water. In 6-12 months, yes they could.

Accounting, I agree, is important ALWAYS, but not compared to pushing demand down asap.

I also agree that the marginal value of water-in-use would vary among sectors (e.g., brown parks vs fields), but that takes a larger market integration...

Anonymous said...

Few of us who try understand the water situation would consider 6-12 months to be "long run", which is perhaps why our views diverge.

I believe there is time to make our best efforts to prepare water accounts (google "water accounting +" to see where the science is on this topic), and hence devise rational interventions that are likely to have real impacts in the intended direction.

The alternative is the rush to "save" water with hi-tech irrigation, which typically does the reverse, or invest heavily in low flow showers and toilets that have minimal impact except in coastal cities.
ABCDE is precisely about "integrating" demands into a common accounting framework, and phrases like "water-in-use" just confuse that type of discussion. It really matters whether the water we use is consumed, and whether the fraction that returns to the system is recoverable or not. That is true in the short, medium, and long term (by which I mean 20-50 years).


Chris Perry

Post a Comment

Note: only a member of this blog may post a comment.