Policy implications of Castle and colleagues on Colorado River Basin groundwater depletions

Ever since we saw early glimpses last spring of data from Stephanie Castle, Jay Famiglietti and colleagues about groundwater depletion in the Colorado River Basin, I’ve been puzzling over the policy implications. Their data, published today in GRL, is worth an “OMG IT’S WORSE THAN WE THOUGHT!” While we’ve been watching Lake Mead’s bathtub ring grow and the basin has lost 12 million acre feet of stored surface water in the last decade, aquifers have declined more than 40 million feet, essentially unnoticed.

In a fascinating talk in the spring at the Bureau of Reclamation, Oregon State University’s Aaron Wolf, the guru of water conflict (or lack thereof) explained how his data showed collaboration and cooperation is far more common than fighting over water. The key thing to look for is how a basin’s water users are able to respond to change. Change can be abrupt, like flooding, or drawn out, like drought or the sort of inexorably rising demand that we’re seeing in the Colorado Basin. Actually, we’ve got drought too, so check that off. At this point you look to institutions:

There’s change in the basin. All the things you look for. Scarcity, floods, droughts, economic growth, tensions. Everything that’s going on. Managing that is a certain level of institutional capacity. How good are the agreements? How good are the relationships? How well do they work together? That helps mitigate the change…. The likelihood of conflict rises as the rate of change in a basin exceeds the institutional capacity to absorb the change.

Pumping groundwater in the Colorado River Basin, March 2014

Pumping groundwater in the Colorado River Basin, March 2014

I’ve been arguing that one of the things we see in the Colorado River Basin is an institutional maturity that has thus far had the ability to absorb the changes we’ve seen to date – the 2001 Interim Surplus Guidelines that jiggered Colorado Basin allocations to rein in California’s overuse (pdf), the 2007 shortage sharing agreement, the remarkable Minute 319 deal with Mexico. Those institutions are now wrestling with the fact that none of this has been enough, which is why maybe this week we’ll see the announcement of a system conservation program deal that will be one more brick in that wall.

All of that shows the maturity, I think of basin-wide water management institutions. Only time will tell of that’s enough, but it’s the sort of thing Wolf’s research shows is a necessary precondition to success. But all of those institutions deal with surface water. What Castle and colleagues are pointing out is that groundwater drawdown is a huge unaddressed problem. And we’ve got no basin-wide institutions for that.

Maybe we simply don’t need them. One of the core principles of basin management is that we need these institutional arrangements because the Colorado River and its tributaries flow alongside or through nine different states. One of the key institutional approaches has been to set allocations at the states’ borders and leave each state to deal with its own internal allocations within its own state-by-state institutions. That’s where groundwater institutions lie. Each state has its own (and they’re widely varied – in preparing for this post I found a few reviews explaining how each state handles its own groundwater, and for now I’ll just categorize them as “tl;dr” – too long, didn’t read :-).

The short answer, then, is that if an individual state has permitted overpumping, that state is going to have to deal with the consequences. This isn’t a basin-wide problem, so much as seven individual problems, for seven individual states.

But even if that assessment is correct, it still becomes a basin-wide problem if groundwater problems at home make it more difficult for the states to make the water-sharing and curtailment decisions that are going to be needed to cobble together overarching solutions to the Colorado River Basin’s problems as a whole. Does groundwater depletion add a level of difficulty that exceeds our institutional capacity to absorb change?

What Castle and her colleagues have given us is enormously helpful, but it’s only the start. It’s gross data. To understand the implications, we really need to drill down now to a much more local level – who’s doing all this pumping, where, and what are the implications at a zillion local watershed scales?

That’s how we’ll figure out where we are on Wolf’s scale of institutional readiness to deal with a problem that’s bigger than I thought it was before I saw this data.

New data shows Colorado Basin groundwater loss greater than reservoir shrinkage

While we’ve been watching Lake Mead and Lake Powell drop during the last decade, the hidden reservoirs of groundwater beneath the Colorado River Basin have been dropping even faster, according to new research by Stephanie Castle and her colleagues at U.C. Irvine. Using gravity data from NASA’s GRACE satellite (the new tool all the cool kids are using to study groundwater), Castle and colleagues have posted up some striking data.

In the time period, they studied, 2004-13, they identified a total of 12 million acre feet of water lost as we drew down reservoirs. That’s the white bathtub ring in Lake Mead, the drop in Lake Powell, and the reductions in storage in smaller reservoirs – the stuff we can see. But they also found a whopping 40.5 million acre feet of loss in aquifer storage, as folks pumped groundwater to make up for surface water shortfalls.

This has significant policy implications, because essentially all the basin-scale policy discussions focus on surface water, while dealing with residual problems left by groundwater shortfalls left to state governance:

While the need to exploit groundwater resources to meet Basin water demands has long been recognized,  withdrawals required to meet current demands remain undocumented and are uncertain in the future. In particular, water management under drought conditions focuses on surface water  resources without a regulatory framework to manage  groundwater withdrawals outside of “river aquifer” systems. At question is  the potential impact of solely managing surface water allocations and diversions in the Basin,  without regard to groundwater loss, on meeting future water demands.

(I don’t have a link yet, will add one as soon as it’s available.)

Here’s a link to the paper, and Jay Famiglietti, one of the authors, has a writeup over at National Geographic. Also, AGU news release.

Pressure increasing to do something about a dropping Lake Mead

With a meeting coming up Friday of representatives of the seven Colorado River basin states, pressure is increasing to do something about the dropping reservoir levels in the basin, Tony Davis reports:

“How urgent it is depends on what you think the risk is,” said attorney Wade Noble, who has represented Yuma-area irrigation districts for 30 years. “If the risk is high that the water is not going to be there … then something needs to be done in the immediate future, not next year.”

The entire story is worth clicking for.


How well is California weathering the drought?

Peter Gleick runs down some of the impacts of California’s remarkable drought:

[W]ater still comes out of my tap, in unrestricted amounts and superb quality, at a reasonable price. And this is true of every resident in the state: drinking water supplies have not been affected, especially for the vast majority of the population that lives in cities of the San Francisco Bay area, Central Valley, and southern California.

But surely the agricultural sector has been hit hard?

While there will be some adverse impacts of some farmworkers and farmers, the overall agricultural sector will not have a bad year. Some farmworkers will be out of work this summer and fall, some farmers will be forced to fallow land because of the lack of water, and others will have higher costs associated with the need to replace surface water shortages with temporary groundwater pumping. But initial estimates from the University of California, Davis, the agricultural community as a whole will not see very large losses – a drop of perhaps 4% or so of normal farm revenue.

See, California, I told you that you can do this!

Brian Devine on ag-urban transfers

Ag-to-urban water transfers are one of the ways it’s easy to make the Colorado River Basin’s water math balance. Seventy percent of the water is currently consumed by agriculture. Just a fraction of that, transferred to use in cities (AMI, or “agriculture to municipal and industrial”), should make solving the problem relatively straightforward, right?

Brian Devine, who’s working on these issues at the University of Colorado, talks about why this is tricky in practice:

The question that needs answering is threefold. First, what are the impacts of a sizable AMI transfer on the community of origin, and on the purchasing municipality? This question is an empirical one with ecological, economic and sociological dimensions. Second, are these impacts, and the tradeoffs they entail, permissible? This is a theoretical question drawing on environmental ethics and a concept of justice. Third, what can be done to bring these impacts more in line with our ethical demands? This last question is a practical one that demands the promulgation of policy instruments which might have economic, political or ecological flavors. The fact that AMI transfers on a large scale are relatively new and have been the subject of rigorous examination only in the last several years, after scholars turned their attentions away from dams and pipelines, makes any definitive conclusions difficult. AMI transfers, like all water issues in the West, are highly dependent on geographic, political and historical context. Nevertheless, it seems clear that the pure form of water transfers that ignores consequences to third parties has serious harmful effects on areas of origin, effects that cannot reasonably be said to be outweighed by the gains made by receiving areas. Fortunately, there appear to be promising policy options for mitigating these consequences, protecting rural values at risk, and still allowing for the growth of urban regions.

The full piece is worth reading.

Eating the Colorado River shortage elephant, one bite at a time

This line from a paper a few years back by Edella Schlager and Tanya Heikkila may seem obvious, but in the context of current discussions over the future of Colorado River management, it bears repeating:

A water allocation rule that allocates more water than is available in a river is not well matched to its setting.

Lower Basin Water Budget, courtesy USBR

Lower Basin Water Budget, courtesy USBR

Yup. That in a nutshell is the problem highlighted by this oft-revisited Bureau of Reclamation slide demonstrating how the Lower Colorado River Basin’s water budget works. Everyone here is following the rules, living within their legal allocation, and Lake Mead keeps dropping because the Law of the River has allocated more water than is available in the river.

This is the critical thing to understand as we see the beginnings of the new “Colorado River System Conservation Program” taking shape, which would create a framework to pay farmers to leave water in the river. It’s not enough to simply save water. The way we go about it must be embedded within, and take into account, the rules governing allocation and distributions of Colorado River water.

Jim Lochhead, now head of Denver Water, wrote an excellent introduction to the problem a decade ago in a remarkably candid history of the Colorado River water wars of 1990-2003 (paywalled). In the early ’90s, it was clear that overuse was an increasing problem. The overuse then came in the form of California’s dependence on surplus that other states had not been using. Nominally, California’s allocation of Colorado River water was supposed to be 4.4 million acre feet of water per year, but since the 1950s, it had been dining on surplus left unused by other states. With Las Vegas’s growth and Arizona’s completion of the Central Arizona Project, the surplus was disappearing, and California had to figure out a way to get from the 5-plus maf it had been using most years down to 4.4 maf.

Imperial Valley ag still thrives despite drought and cutbacks in delivery of Colorado River water to California

Imperial Valley ag still thrives despite drought and cutbacks in delivery of Colorado River water to California

In the decade since that time, Southern California has been remarkably successful at making the adjustment. Coastal Southern California’s economy is still as robust as ever, even in the face of the current drought, and the Coachella-Imperial-Palo Verde farm belt is still thriving with big ag. The details of how they did it – the largest ag->urban water transfer in history – are interesting (buy my book! as soon as I finish writing it!). But a critical necessary condition, before the details of fallowing and ag efficiency were worked out, was the need for a change in the “water allocation rules” that Schlager and Heikkila are talking about. The water wars of 1990-2003 that Lochhead writes about were all about the basin states and the federal government realizing the water allocation rules were inadequate. One could argue that the “structural deficit” was in a sense even larger then, and that first round of action was about reducing it.

When you look at the size of the current structural deficit in that Bureau slide above – 1.2 million acre feet per year – the Colorado River System Conservation Program looks modest indeed. We’re likely to see more details as the agreements are finalized over the next week, but at the going price of water, the $11 million effort looks like enough money to generate 75,000 to 100,000 acre feet of water savings. It sounds like most of the initial attention will be initially be focused on the Lower Basin, with Reclamation hoping to issue “requests for proposals” this year to farmers interested in fallowing land for a price per acre foot of water yet to be determined. With Palo Verde, Coachella and Imperial all actively engaged in the already existing California ag->urban transfer programs, much of the action is likely to be on the Arizona side, especially Yuma and Wellton-Mohawk (though I’ve also heard mention of some possible interest across the river with the Bard Irrigation District, which is on the California side of the border but gets Yuma Project water).

When I asked recently whether 75k – 100k acre feet was enough, given a structural deficit an order of magnitude larger, one of the people working on the program pointed out that this is a pilot program, to learn how to do it, then added, “How do you eat an elephant? One bite at a time.”

The 1990-2003 experience suggests that the water conservation piece of this may be the easy part. As a nice new Western Resource Advocates white paper explains, we know how to conserve the water. The key piece here, and the reason the System Conservation Program is so interesting and important, is that we need to get the water allocation rules and river management policies right in order to cause those conservation savings to happen.

Border arbitrage

Guy on the move beneath the San Luis Bridge on the U.S.-Mexico border. John Fleck, March 2014

Guy on the move beneath the San Luis Bridge on the U.S.-Mexico border. John Fleck, March 2014

I’m not positive, but I’m reasonably certain the guy with the bedroll in this picture is performing arbitrage. He was sleeping out the midday sun under the San Luis Bridge on the Sonora-Baja-Arizona border before I saw him pick up and head toward those bushes on the Colorado River’s east bank. Just beyond that array of fences is an Arizona farm, where workers are paid substantially more than they are paid on this, the Mexican side of the border.

Here’s how Irving Fisher explained this in his “Elementary Principles of Economics“:

Elementary Principles of Economics

Elementary Principles of Economics

Competition, in the case of our friend with the bedroll facing the border, is not “perfect”, to borrow Fisher’s word. The fence, and the Border Patrol patrolling the no-man’s land between the fence and the Colorado River, have the effect of preventing the easy transaction of Bedroll Guy’s business. But if he can slip past the border fence defenses, he can sell his labor for a substantially higher price.

This very practical exercise in arbitrage, carried out many times over along the U.S.-Mexico border, has enormous implications.

Gatsby and the Colorado River

In December of 2002, an effort to sort out the problems of the Colorado River appeared as though it was about to blow up. California had long been living beyond its means (using more than the 4.4 million acre feet minimum guaranteed under the Law of the River). The deal would have given California time to reduce gradually – they called it a “soft landing”. But the deal had to be accompanied by a binding California commitment to really get to 4.4.

At the last minute, California tried to insert weasel clauses into the deal, known as the QSA, that would have given them the “soft landing” surplus water now but would have given them a way to wriggle out of their 4.4 maf commitments later. This had happened before – close to a deal, then a “but wait, this one more little thing” from California. On Dec. 9, Assistant Secretary of the Interior Bennett Raley wrote this in a letter to California:

The Department has no interest in a QSA that does not represent a long term Quantification of the parties’ portion of California’s apportionment of Colorado River water, lest in fifteen years we find ourselves as Gatsby did – ‘So we beat on, boats against the current, borne back ceaselessly into the past.’

I am indebted to An Upper Basin Perspective on California’s Claims to Water from the Colorado River Part II: The Development, Implementation and Collapse of California’s Plan to Live within Its Basic Apportionment, a remarkable history by Jim Lochhead.

Stuff I wrote elsewhere: moving groundwater in New Mexico

The Augustin Plains Ranch project, New Mexico’s version of a trend toward meeting urban needs in the west by pumping rural groundwater in to cities, is taking another whack at winning state approval after losing resoundingly two years ago:

A for-profit group hoping to pump New Mexico groundwater to the Rio Grande Valley and sell it to thirsty cities has asked state water managers for a new hearing on a proposal the Office of State Engineer turned down two years ago.

Augustin Plains Ranch LLC, an investment group that includes the owners of a ranch near Datil in the high country west of Socorro, is again proposing to pump groundwater and send it through a pipeline that would follow the Rio Grande north to the Albuquerque metro area.

The new proposal seems to have the same deficiency that caused the previous one to go down in flames – the failure to identify a “place and purpose of use”, something the State Engineer said was essential to state review and approval of the water transfer:

Augustin Plains Ranch repeats its prior approach – listing all possible water uses in much of a seven-county area from Socorro to Santa Fe. The ranch, in its filing with the state, said it hopes to win preliminary approval for the application before specifying who will use the water, and where.

That drew quick criticism from attorney Bruce Frederick with the New Mexico Environmental Law Center in Santa Fe, who represented a number of the 248 people who protested the group’s previous application.

“The application is a public relations piece and suffers from the same basic legal deficiency as the prior applications – the ranch has again failed to identify any actual place or purpose of use of water,” Frederick told the Journal.